Physiology of Human and Animal Disease Vectors

Advances in Insect Physiology Volume 37 Advances in Insect Physiology edited by Stephen J. Simpson School of Biologi...

Author: Stephen Simpson | Jerome Casas

134 downloads 3241 Views 4MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Advances in Insect Physiology

Volume 37

Advances in Insect Physiology edited by Stephen J. Simpson School of Biological Sciences, The University of Sydney, Sydney, Australia

Je´roˆme Casas Universite´ de Tours, Institut de Recherche en Biologie de l’Insecte UMR, CNRS, Tours, France

Volume 37

Academic Press is an imprint of Elsevier 32 Jamestown Road, London NW1 7BY, UK Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands Linacre House, Jordan Hill, Oxford OX2 8DP, UK 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 525 B Street, Suite 1900, San Diego, CA 92101-4495, USA First edition 2009 Copyright # 2009 Elsevier Ltd. All rights reserved No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (þ44) (0) 1865 843830; fax (þ44) (0) 1865 853333; email: [email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://www.elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material

Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made

ISBN: 978-0-12-374829-4 ISSN: 0065-2806 For information on all Academic Press publications visit our website at elsevierdirect.com

Printed and bound in United Kingdom 09 10 11 12 10 9 8 7 6 5 4 3 2 1

Contents Contributors

vii

Preface

ix

Orientation Towards Hosts in Haematophagous Insects: An Integrative Perspective CLAUDIO R. LAZZARI

1

From Sialomes to the Sialoverse: An Insight into Salivary Potion of Blood-Feeding Insects ` JOSE´ M. C. RIBEIRO, BRUNO ARCA

59

The Enemy Within: Interactions Between Tsetse, Trypanosomes and Symbionts DEIRDRE P. WALSHE, CHER PHENG OOI, MICHAEL J. LEHANE, LEE R. HAINES

119

Interactions of Trypanosomatids and Triatomines ¨ NTER A. SCHAUB GU

177

Lyme Disease Spirochete–Tick–Host Interactions KATHARINE R. TYSON, JOSEPH PIESMAN

243

Epidemiological Consequences of the Ecological Physiology of Ticks SARAH E. RANDOLPH

297

Index

341

Preface How do insect vectors of disease find their animal hosts? Once a host is located, how do insects deploy their intricate mouthparts and the extraordinary complexities of salivary chemistry to secure a blood meal, and in so doing cause transmission of disease organisms? What are the critical molecular components that mediate the interactions between insect vectors and disease organisms? How do insect physiology and life history interact with environmental conditions to shape patterns of disease incidence? Given the enormous health and socioeconomic impacts of insect-borne diseases, the study of insect physiology acquires special significance when applied to questions such as these. In addition to reviewing the progress made towards answering these questions, the papers in this special issue of Advances in Insect Physiology give rise to key themes for the future. Hence, the field has benefited enormously from recent advances in molecular biology and protein biochemistry, yet in order to be fully effective, these tools need to be used within the context of a deep understanding of vector physiology, behaviour and ecology. In turn, an understanding of ecophysiology, behaviour and life-history is necessary for explaining and predicting the biogeography and epidemiology of insect vector-borne diseases - especially in a world experiencing global warming, population growth and changing patterns of land use. STEPHEN J. SIMPSON JE´ROˆME CASAS

Contributors Bruno Arca` Department of Structural and Functional Biology, University ‘‘Federico II’’, Naples, Italy; and Parasitology Section, Department of Public Health, University ‘‘La Sapienza’’, Rome, Italy

Lee R. Haines Liverpool School of Tropical Medicine, Pembroke Place, Liverpool L3 5QA, United Kingdom

Claudio R. Lazzari Institut de Recherche sur la Biologie de l’Insecte, UMR 6035 CNRS – Universite´ François Rabelais, Faculte´ des Sciences et Techniques, Av. Monge, Parc Grandmont, 37200 Tours, France

Michael J. Lehane Liverpool School of Tropical Medicine, Pembroke Place, Liverpool L3 5QA, United Kingdom

Cher Pheng Ooi Liverpool School of Tropical Medicine, Pembroke Place, Liverpool L3 5QA, United Kingdom

Joseph Piesman Centers for Disease Control and Prevention, Coordinating Center for Infectious Diseases, National Center for Zoonotic, Vector-Borne and Enteric Diseases, Division of Vector-Borne Infectious Diseases, Fort Collins, Colorado

Sarah E. Randolph Department of Zoology, University of Oxford, Oxford OX1 3PS, United Kingdom

Jose´ M. C. Ribeiro Laboratory of Malaria and Vector Research, National Institute of Allergy and Infectious Diseases, 12735 Twinbrook Parkway room 2E-32D, Rockville, Maryland 20852, USA

Gu¨nter A. Schaub Zoology/Parasitology Group, Ruhr-Universita¨t Bochum, 44780 Bochum, Germany

viii

Contributors

Katharine R. Tyson Department of Microbiology and Immunology, University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA

Deirdre P. Walshe Liverpool School of Tropical Medicine, Pembroke Place, Liverpool L3 5QA, United Kingdom

Orientation Towards Hosts in Haematophagous Insects: An Integrative Perspective Claudio R. Lazzari Institut de Recherche sur la Biologie de l’Insecte, UMR 6035 CNRS – Universite´ François Rabelais, Faculte´ des Sciences et Techniques, Av. Monge, Parc Grandmont, 37200 Tours, France

1 Introduction 2 2 Functional neuroanatomy 3 3 A brief history of haematophagy 5 3.1 The relationship between insects and vertebrate hosts 6 3.2 Feeding on blood 6 4 The host signals 7 4.1 Odours 7 4.2 Heat 8 4.3 Water vapour 10 4.4 Visual cues 12 5 Looking for food 14 5.1 Activation 15 5.2 Appetitive search 15 5.3 Host detection 16 5.4 Host finding 16 5.5 Host contact 17 5.6 Host biting 17 5.7 Food recognition and feeding 18 5.8 Leaving the host 20 6 Stimulus propagation and sensory reception 20 7 Orientation mechanisms 23 8 Thermal sensing in kissing bugs 26 9 Sensory parsimony 34 9.1 Parsimonious use of information in blood-sucking insects 34 9.2 Practical consequences 35 10 State-dependency of host-seeking behaviour 36 10.1 The temporal modulation of the response to odours 37 10.2 Maturation and responsiveness 38 10.3 The modulation of host-seeking activity by reproduction 39 10.4 Feeding conditions and host searching 40 11 Why some people are bitten more than others? 41 12 Learning and memory 42 ADVANCES IN INSECT PHYSIOLOGY VOL. 37 ISBN 978-0-12-374829-4 DOI: 10.1016/S0065-2806(09)37001-0

Copyright # 2009 by Elsevier Ltd All rights of reproduction in any form reserved

2

CLAUDIO R. LAZZARI

13 Repellents, how they work 42 14 Conclusions and perspectives 43 Acknowledgements 48 References 48

1

Introduction

Neuroethology (‘neuro’ Greek; related to nerve cells, ‘ethos’ Greek; habit or custom) addresses the neural basis of animal behaviour, through an evolutionary and comparative approach. Its main focus is understanding how the central nervous system translates biologically relevant stimuli into behavioural activity (Ewert, 1980). Various notions about the origins and goals of neuroethology exist (Ewert, 1980; Hoyle, 1984; Bullock, 1990; Pfluger and Menzel, 1999). However, the main questions addressed in this area of study, by means of experimental exploration, are as follows (Ewert, 1980): (1) Which sensory processes are responsible for distinguishing between behaviourally relevant and irrelevant stimuli? (2) How are signals localized in space and time? (3) How is information acquired, stored and recalled? (4) What is the neurophysiological basis for the motivation of a behavioural pattern? (5) How is behaviour coordinated and controlled by the central nervous system? (6) How is behaviour ontogeny related to neuronal mechanisms? Insect neuroethology is a well-developed field, thanks to advances in the development of several model systems. Studies on some of the major topics, such as wind-triggered escape, the recognition of acoustic signals, learning and memory and others, have provided considerable insight into how the nervous system controls adaptive behavioural responses in insects. The particular species chosen for detailed analysis in such studies include honeybees, cockroaches, crickets, flies and a few others. However, no blood-sucking insects are included in this ‘‘select’’ group, even though some of them are classical models in insect physiology (e.g. Rhodnius prolixus) or the subject of intense study due to their impact on human health (mosquitoes). This does not mean that we lack information on their behaviour and neurobiology. On the contrary, important aspects of their behaviour, sensory physiology and functional neuroanatomy have been extensively studied. Nevertheless, very few studies have analysed this aspects with an integrative view in haematophagous insects.

ORIENTATION TOWARDS HOSTS IN HAEMATOPHAGOUS INSECTS

3

In this chapter, I summarize the current understanding of the physiological mechanisms that underlie host-seeking in blood-sucking behaviour, with an integrative view, to elucidate some of the issues described above.

2

Functional neuroanatomy

To understand the neurobiological basis of behaviour, we firstly need to understand the organization of the nervous system. Functional neuroanatomy provides the basis for focusing physiological studies on the neural elements associated to a particular behaviour. A series of detailed studies have been published over the past few years on the functional neuroanatomy of mosquitoes, particularly Aedes aegypti and Anopheles gambiae. These studies have revealed the neural architecture of the olfactory brain and the organization of sensory pathways from the antennae, maxillary palps and labium (Anton, 1996; Anton et al., 2003; Anton and Rospars, 2004; Ignell and Hansson, 2005; Ignell et al., 2005; Ghaninia et al., 2007a,b; Siju et al., 2008). Some of the most significant findings of these studies concerning these malaria and yellow fever mosquitoes (A. gambiae and A. aegypti, respectively) are described below: In A. gambiae, antibody labelling and subsequent three-dimensional reconstructions of the antennal lobes showed that males have 61 glomerular neuropils and females have 60. The size of the antennal lobe and of individual glomeruli was also tested for sexual dimorphism (Ghaninia et al., 2007b). In A. aegypti, sexual dimorphism has been demonstrated both for the number of total glomeruli (49 in males and 50 in females) and size of certain glomeruli (Ignell et al., 2005). Maxillary palp projections in A. aegypti are restricted to two posteromedial glomeruli, which do not receive antennal afferents. These include nerve projections from carbon dioxide receptors, which project to a single glomerulus (Anton, 1996; Anton et al., 2003). Five non-overlapping projection zones were identified within the antennal lobe of A. gambiae, with one zone receiving input exclusively from maxillary palp sensilla and two zones each receiving input exclusively from trichoid or grooved-peg antennal sensilla (Anton and Rospars, 2004). Extensive serotonergic neurohemal plexi have been observed in the peripheral chemosensory organs of A. aegypti and A. gambiae, that is in the antenna, the maxillary palp and the labium, suggesting a potential role of serotonin as a neuromodulator in the chemosensory system (Siju et al., 2008).

4

CLAUDIO R. LAZZARI

The central projections of the contact chemoreceptors in the labium and cibarium of A. gambiae and A. aegypti have been described in detail (Ignell and Hansson, 2005). Notably, despite the differences in feeding habits between male and female mosquitoes, sexual dimorphism in the olfactory brain is not very marked, at least in terms of anatomical differences. Given the specificity and sensitivity of sensory organs dedicated to perceiving specific host-associated cues in females but not in males, one would expect a greater divergence between the sexes, but this does not seem to be the case. This could be related to the fact that both sexes feed on nectar, with females additionally sucking on blood. Even assuming that both males and females detect the same plant-associated volatiles and that certain chemical cues are common to both plants and animals (Syed and Leal, 2007), one would still expect a greater difference between the olfactory brains of males and females. It is possible that the glomeruli responsible for processing host-specific signals in females are, in males, involved in processing signals specific to this sex. This would thus allow for physiological dimorphism that it is not anatomically visible. Neuroanatomical studies of Hemiptera have focused on bedbugs and kissing bugs. Whereas studies on bedbugs have been mostly limited to the general anatomy of the central nervous system (Singh et al., 1996), more complete sets of data are available on the neuroanatomy of kissing bugs. Indeed, following on from the classical studies of V.B. Wigglesworth on the histology of the peripheral and central nervous systems in R. prolixus (Wigglesworth, 1953, 1959a,b), further studies described the general anatomy of their central nervous system, the organization of antennal lobes and the ocellar and mechanosensory pathways (Barth, 1952, 1976; Insausti, 1994; Insausti and Lazzari, 1996, 2000a; Barrozo et al., 2009). Recent studies on triatomines suggest that there is no sexual dimorphism in the organization of the antennal lobes (Barrozo et al., 2009). This is not surprising, given that, in contrast to mosquitoes, both sexes feed on vertebrate blood, and no marked difference has been found in the number and type of sensory organs between males and females. Another interesting finding, meriting further analysis, is the fact that antennal inputs do not only project into the deutocerebrum, but descend further along the ventral nervous chain, synapsing nervous elements in the suboesophageal, porthoracic and posterior (mesoþmethathoracicþabdominals) ganglia (Barrozo et al., 2009). These kind of direct connections had been previously described for ocellar interneurons and cephalic mechanoreceptors in kissing bugs (Insausti and Lazzari, 1996, 2000a), thought to be associated with the direct control of motor patterns. These neuroanatomical findings provide a good basis for further analyses of the central processing of sensory signals. Such analyses should address haematophagy-related traits, such as the representation of host odours in the olfactory


5

brain of blood-sucking insects or the potential involvement of specific glumeruli in host recognition. Haematophagous insects may be good models for the investigation of more general aspects, such as the mechanisms allowing multimodality to be represented in the insect brain. The central processing of odours is a major aspect of insect neurobiology and is studied intensively in non-haematophagous experimental models (e.g. honeybees, moths, locusts and Drosophila). Such studies require the use of electrophysiological and optophysiological (most frequently calcium imaging) methods that need to be supported by accurate neuroanatomical data (Galizia et al., 1997, 1999a,b; Galizia and Menzel, 2000; Zube et al., 2008). The brain neuroanatomy of mosquitoes is the best known of the bloodsucking insects, making them the most suitable model for studying the central processing of sensory information. However, in vivo optical recordings could be technically difficult due to their small size. Kissing bugs may thus be a good alternative model because of their larger size, robustness and easy access to their brain and rest of the nervous system. Even though mosquitoes and bugs differ considerably in terms of phylogeny, they exploit the same source of food (vertebrates) and respond to similar spectra of signals.

3

A brief history of haematophagy

Haematophagy is believed to have evolved independently many times in insects (Lehane, 2005). Blood provides a rich and usually sterile source of many nutrients needed for survival and reproduction. The circulation of blood within vessels hidden under the skin of moving animals, however, makes it difficult to obtain. Additionally, hosts exhibit defensive behaviours to avoid being bitten or may even play the dual role of prey and predator at the same time. The ability to feed on blood has thus evolved under the control of selective pressures, with blood-sucking insects able to detect and locate a potential host, locate in space, identify the best site to land on, locate the blood vessels, pierce the skin and find the blood. These processes must all be accomplished minimizing the risk of being detected and killed by the antiparasitic behaviour of the host or of being predated on by the host itself. With more than 14,000 species and 400 genera of arthropods feeding on blood, haematophagy would appear to confer certain advantages (Ribeiro, 1995; Lehane, 2005). The ability to feed on the blood of vertebrate hosts may have developed through one of two possible routes. The first possibility involves a gradually closer association with vertebrates, first to their nest, and later to their body, leading to morphological changes to allow the insects to feed on their hosts’ blood. The second possibility is based on pre-adaptation, whereby insects were able to pierce animal or vegetal tissues and feed on fluids before developing the ability to suck blood (Lehane, 2005).

6

3.1

CLAUDIO R. LAZZARI THE RELATIONSHIP BETWEEN INSECTS AND VERTEBRATE HOSTS

Blood-sucking arthropods and their vertebrate hosts exhibit various degrees of association. In some cases, the association between the host and haematophagous insect is limited to the nutritional aspect, with the arthropod contacting the vertebrate just briefly, for the time necessary to feed on its blood. Each partner thus lives independently, even exploiting different habitats. This is seen, for instance, in the relationship between haematophagous flies and vertebrate hosts, and for many mosquito species. Some insect–host pairs share the same habitat, usually the burrow or nest of the vertebrate, or the house of human hosts. Blood-sucking Hemiptera (kissing bugs, bed bugs) are typical examples of insects that occupy the habitat of the host, exploiting the habitat’s protective aspect, its microclimate and the ready access to food sources. The closest relationships between insects and their hosts are found when the insect becomes a true ectoparasite, living on the host’s body. This is observed for fleas, lice and ticks, all three displaying different extents of association with their host. Fleas leave their host to reproduce and their larvae obtain blood from vertebrate only indirectly, by eating the excrement of adults; lice reproduce on their hosts, but may move from one host to another; and ticks only leave their hosts for moulting, remaining fixed to the host skin by their mouthpieces the rest of the time. The sensory machinery used to gather information required to identify a host, and to locate it in space and time, is determined by the nature of the insect–host relationship. The sensory organs of blood-sucking insects involved in receiving signal from their hosts exhibit patterns of organization typically seen in most insects. The structure of haematophagous sensilla does not seem to particularly differ between the species, but the number present does appear to be related to the nature of the relationship of each species with their host: the closer the relationship with their host, the fewer sensory organs are present (Chapman, 1982; Lehane, 2005). Thus, insects closely associated to the host have a smaller number of sensory organs than those that search for a potential host across a large area: lice have 10 sensory organs; sensilla, 20; fleas, about 50; bedbugs, 56; kissing bugs, 2900; and stable flies, nearly 5000 (Lehane, 2005). 3.2

FEEDING ON BLOOD

A blood-based regime offers a number of advantages. Blood is rich in nutrients, sources of blood can be found almost everywhere and, except for the incidental presence of parasites, it is otherwise sterile. The exploitation of vertebrate blood as food has led to many specific morphological, physiological and behavioural adaptations. To feed on blood, a haematophagous insect needs to be able to identify a vertebrate source, approach it at the right moment and locate a site on


7

the host skin where the blood supply is assured; it then needs to cut or pierce the skin, locate a vessel hidden under the skin, identify the fluid, suck up a sufficient amount of blood as quickly as possible, avoiding the coagulation of blood within the vessel or the mouthparts, and then escape, carrying the extra weight. These actions all require the input of multimodal signals from the host, environment and within the arthropod itself.

4

The host signals

Vertebrates, particularly endothermic vertebrates, emit a large number of informational cues of different types. Blood-sucking insects can then exploit these signals, using their various sensory systems. The human skin emits around 350 volatile compounds potentially exploitable by haematophagous insects (Bernier et al., 2000); many other compounds are released with breathing. Vertebrates able to physiologically or behaviourally regulate the temperature of their bodies (e.g. by basking) also act as sources of heat. When active, their movements and contrast against the background may be detected by insects’ visual systems, and the vibrations produced may be detected by the mechanoreceptive organs of insects. Given that many of these signals (e.g. heat and CO2) are emitted by most, if not all, warm-blooded vertebrates, we would expect them to be general cues that are perceived by most haematophagous insects. The way they use this information, however, may differ from one species to another. As observed for other insects, the different types of chemo-, mechano-, thermoand hygrosensitive sensilla are mainly concentrated on the antennae. Antennae are not the only bearers of sensilla (sensilla are present in other regions of the body as well), nor is their function limited to providing support. The antennae have an important role in the active gathering of sensory information. The visual organs are also involved in the detection of moving hosts by some insects. Compound eyes are the main organs dedicated to detailed analysis of the contrast, form and colour of objects in the visual field. Simple eyes or ocelli are probably not involved at all, but cannot be excluded as potential inputs of host-related information given the wide variety of functions that these organs are involved in, in other species. 4.1

ODOURS

Virtually every haematophagous insect modifies its behaviour in the presence of carbon dioxide, either increasing their activity or responsiveness to other stimuli or orienting itself towards the source (Guerenstein and Hildebrand, 2008). Carbon dioxide is produced intermittently (animal breath) and, as with other odours, travels through the air in discrete packets interspersed with clean air,

8

CLAUDIO R. LAZZARI

being detected as intermittent stimuli. Thus, mosquitoes only orientate towards a source of CO2 that emits an intermittent (pulsed) signal (Geier et al., 1999a). In contrast, Triatoma infestans is attracted by a CO2 source only under continuous or low-rate pulsed stimulation, and is repelled by CO2 signals delivered at high frequency (Barrozo and Lazzari, 2006). This difference between two haematophagous species exploiting the same cue and the same CO2 sources may be explained by differently adapted strategies to approach their hosts. For a flying insect, such as the mosquito, the frequency of encountering CO2 packets can be relatively high, with an increased rate with proximity to the emitting source. Triatomines, in contrast, search for food by walking in a relatively closed habitat, in which air currents have a smaller effect on the propagation of CO2 than in open spaces. Additionally, they feed mostly during the night on sleeping hosts, which do not generate air turbulence by movement. Thus, odours may disperse more homogeneously and the rate of release is more stable in such environments. Despite being sufficient to attract most blood-sucking insects, CO2 is not an indispensable stimulus for triatomines (Nuń˜ez, 1982; Taneja and Guerin, 1995; Barrozo and Lazzari, 2004a), or for some mosquito species (Gibson and Torr, 1999; Bosch et al., 2000; Bernier et al., 2003; Smallegange et al., 2005). Previous studies on tsetse flies have demonstrated that CO2, due to its high diffusion rate, is only useful for insects when they are relatively close to the host (Torr, 1990; Leak, 1999). As well as responding to CO2, different haematophagous insects appear to respond to similar or the same host odorants. Such odorants include nonanal (triatomines: Guerenstein and Guerin, 2001; mosquitoes: Du and Millar, 1999), lactic acid (triatomines: Barrozo and Lazzari, 2004b; mosquitoes: e.g. Geier et al., 1999b), ammonia (triatomines: Taneja and Guerin, 1997; mosquitoes: e.g. Meijerink et al., 2001), 1-octen-3-ol (triatomines: Barrozo and Lazzari, 2004b; mosquitoes: Takken and Knols, 1999), short-chain carboxylic acids such as butyric acid (triatomines: Guerenstein and Guerin, 2001; Barrozo and Lazzari, 2004a; mosquitoes: e.g. Pappenberger et al., 1996), C4–5 aliphatic amines such as isopentylamine (triatomines: Diehl et al., 2003; mosquitoes: Pappenberger et al., 1996) and terpenes including a(þ) pinene (triatomines: Guerenstein, 1999; mosquitoes: Bowen, 1992). 4.2

HEAT

It is well established that blood-sucking insects respond to heat emanated from the body of warm-blooded vertebrates. The mechanisms underlying their use of thermal information to find food have received much less attention than the response to odours. In-depth studies have been carried out investigating the morphology and physiology of thermosensilla in insects (Altner et al., 1981; McIver and Siemicki, 1985; Altner and Loftus, 1985; Gingl and Tichy, 2001; Tichy et al., 2008), including those of haematophagous insects (McIver and


9

Siemicki, 1985; Gingl et al., 2005). However, the use of thermal information by blood-sucking insects has been analysed only partially or not at all in most species. The transfer of heat between two bodies at different temperatures may occur through three different physical processes, conduction, convection and radiation (Fig. 1). When heat is conducted, atoms are energized, increasing their vibration rate. This causes neighbouring atoms to vibrate more vigorously, and the signal thus spreads through the material. In our case, the material is the air, and the result is the formation of a temperature gradient around the body of a host. This gradient may then be used as an orientation cue by the insects. Convection concerns heat exchange with a moving fluid. When the fluid (air) is heated by conduction from the body, its temperature increases and it becomes less dense. It then starts to ascend away from the heat source. When its temperature drops back, its density increases and it descends. For some haematophagous insects, convection is not only a matter of heat transfer, but also of the production of ascending airstreams that transport host odours. Mosquitoes, for example, seem to make use of convection currents to approach a host (Lehane, 2005). Convection currents are only useful as a cue when approaching from just above a host.

Conduction Heat source

Convection

Heat source

Heat source Radiation

FIG. 1 Diagrams representing the three mechanisms for heat exchange. Conduction: atoms are energized and vibration increases; vibration spreads through the material. Convection: air is heated and rises, and descends when cool. The air movement from a host transports odorants that can be detected by haematophagous insects. Radiation: any object at a temperature higher than 0 K emits infrared radiation of a wavelength dependent on its temperature. Radiation spreads in all directions and is not affected by air movements.

10

CLAUDIO R. LAZZARI

The third mechanism of heat transfer, radiation, involves the emission and absorption of radiant heat at wavelengths corresponding to the infrared region of the electromagnetic spectrum. This exchange does not require a conducting material or the movement of fluid. Any object at a temperature above absolute zero (0 K or 273 C) emits infrared radiation (IR) of a wavelength corresponding to its temperature. The perception of radiant heat has only been observed for a few groups of animals and in insects belonging to the orders Coleoptera and Hemiptera. For a blood-sucking insect to detect radiation, it needs to be able to assess, from any relative position, the heat emitted by a potential host, without the effects of wind, which can disrupt conduction gradients and convective currents. Experimental evidence has been obtained over many years for the role of heat in the orientation of blood-sucking insects, for example in mosquitoes (Peterson and Brown, 1951; Khan et al., 1966), bed bugs (Usinger, 1966; Reinhardt and Siva-Jothy, 2007), kissing bugs (Wigglesworth and Gillet, 1934a,b; Nicolle and Mathis, 1941) and lice (Wigglesworth, 1941). However, since the initial discovery of a sensory basis for host orientation, relatively few studies have been carried out on this topic. The responses to radiant heat and convection currents were first studied more than 50 years ago in mosquitoes (Peterson and Brown, 1951). When an infrared transparent filter was placed between the heat source and the insects, the response of the insects was weaker than in the absence of the filter. It was thus concluded that convection currents, but not radiant heat, constitute the main thermal cue for mosquitoes. Further work reinforced the notion of ascending convection currents acting as carriers of water vapour and odours (Eiras and Jepson, 1991, 1994). Thus, mosquitoes flying over a host perceive ascending air currents that are warm, humid and charged with host odours. Heat plays a major role in the orientation of kissing bugs (Heteroptera: Reduviidae). The first experiments on the responses of these bugs were conducted on R. prolixus by Wigglesworth and Gillet (1934a,b). They showed that bugs orientate themselves towards a heat source by a mechanism of telotaxis. When the bugs reach a distance of 1–2 cm from the source, they extend their proboscis and switch to tropotaxis (Fig. 2). 4.3

WATER VAPOUR

Water vapour has been suggested to play a role in the location of vertebrate hosts by blood-sucking insects (Bernard, 1974; Altner and Loftus, 1985). However, most of the studies were restricted to examining the electrophysiological or morphological characteristics of sensory organs. Given the small number of behavioural studies on this topic, the role of water vapour in shortand long-range orientation remains unclear in most cases. The mosquito A. aegypti moves upwind towards humid and warm airstreams in conditions

ORIENTATION TOWARDS HOSTS IN HAEMATOPHAGOUS INSECTS A

5 cm C

5 cm

11

B

5 cm D

5 cm

FIG. 2 Trajectories of unilaterally antennectomized and control insects towards a thermal source in the absence of sensory cues other than heat. (A) An intact insect in a big size arena and (B) in a small one. (C) Right antennectomized insect in a big size arena and (D) left antennectomized bug in a small one. Note that the insects approached the source using a direct path, but when close to the source, antennectomized bugs deviated in the direction of the intact antennae and missed the goal. Two orientation mechanisms seem to be involved: telotaxis up to 1.5–2 cm from the source and tropotaxis when in close proximity, at the same time as the proboscis is extended (not shown). The switch from telo- to tropotaxis occurred before any physical contact with the heat source. This and the fact that insects were not exposed to either visual or chemical information suggest that thermal cues provided enough information for both the intact and unilaterally antennectomized bugs to estimate their position relative to the source (modified from Flores and Lazzari, 1996).

of water deprivation; Culex quinquefasciatus also seems to use water vapour as a cue to orientate itself towards human hosts (Mboera et al., 1998). In kissing bugs, the presence of a humid source alone is sufficient to elicit short-range, but not long-range, orientation in T. infestans. The humidity of an airstream has no visible effect on the orientation behaviour of this species, regardless of the physiological state of the bugs (Barrozo et al., 2003). Interestingly, the response to thermal sources seems to be enhanced by moisture, leading to stronger responses at greater distances (Barrozo et al., 2003). This enhancement is probably due to a combination of factors: (1) an increased amount of heat reaching the insect thermoreceptors, the enthalpy of moist air being greater than that of dry air; (2) bimodal convergence of peripheral inputs

12

CLAUDIO R. LAZZARI

onto thermosensitive cells, the activity of which varies not only as a function of air temperature, but also relative humidity (Altner and Loftus, 1985); (3) the potential bimodal convergence of thermosensitive and hygrosensitive inputs in the central nervous system (Horn, 1985). 4.4

VISUAL CUES

Insects have a highly developed visual system, composed of simple and compound eyes (Warrant and Nilsson, 2006). Two types of simple eye, stemmata and ocelli, are found in larvae of endopterygota and most adult insects, respectively. Compound eyes are present in adults of all groups and in the immature instars of exopterygota. Each type of eye fulfils different functions at different phases of the insect’s life. Stemmata lie laterally in the head of larvae. The number of stemmata varies between species. The capacity of these structures to gather visual information – determining changes in light intensity, shape detection and spectral discrimination – is also variable (Gilbert, 1994). Ocelli are not able to form images; indeed the light entering their optical apparatus is focused beyond the retina, eliminating any structural detail of the visual panorama. Each ocellar interneuron gathers information from a large number of photoreceptors, that is from large areas of the visual field. In some insects, ocellar photoreceptors are sensitive to either green or UV light. Ocellar inputs are transmitted through giant descending neurons, which form synapses at different nervous centres in the brain and in the ventral nervous chain. In most insects, they function as horizon detectors. The characteristics of this system are well suited to its role in horizon detection (the green earth/UV sky boundary not interrupted by structural details), with the rapid conduction of nervous signals directly to motor centres allowing rapid flight stabilization of the insect (Mizunami, 1995). Compound eyes constitute the most developed and sophisticated sensory system in insects. They are composed of hundreds or thousands of visual units (ommatidia) and are capable of transmitting detailed spatial information, object shapes and, in many cases, colour and polarization vision (Stavenga and Hardie, 1989). Whereas other sensory systems exhibit particular characteristics adapted to haematophagy – for example numbers of thermoreceptors or the specificity of chemoreceptors – the visual system of blood-sucking insects does not seem to be related to their particular lifestyle. Tsetse flies, for instance, detect and follow potential hosts visually (Brady, 1972a; Gibson, 1992; Leak, 1999; Lehane, 2005), but their compound eyes do not display any particular specialized features. The ocelli and compound eyes of kissing bugs are both involved in negative phototaxis (Lazzari et al., 1998), but this seems to be related more closely to the central processing of information than to structural or physiological adaptations in the periphery. Vision is directly involved in host location in many diurnal haematophagous insects and plays an indirect role in nocturnal insects. Compound eyes are characterized by their high sensitivity to the movement of objects (Stavenga


13

and Hardie, 1989). Tsetse flies exploit this capacity to detect and react to the moving elements in their visual field, exhibiting a stronger response to movement with increasing starvation time (Brady, 1972b; Torr, 1988). The detection of movement also plays an important role in the orientation of haematophagous insects when they follow an odour plume, notably during flight. As with other insects in flight, their small mass allows them to be easily displaced from their intended path by the wind (Fig. 3). Additionally, being submerged in a moving medium and without any contact with the substrate, flying insects do not extract reliable information about their displacement from mechanoreceptors, but from visual inputs. Retinal image motion is elicited when a moving object crosses the visual field (object motion); even if the outside world is stationary, there is a continuous image flow on the retina when the insect moves. The optic flow is a key source of information about the three-dimensional layout of the environment as well as the path and speed of displacement (Egelhaaf, 2006). It is thus the optic flow of stationary objects

Active flight direction Actual flight path

Translational flow field

Wind Rotational flow field

FIG. 3 Flight direction of insects in the presence of wind blowing from the side. The actual path depends on both the forward impulse of the flight motor system and the wind force. Insects make use of the movement of objects over the retina to extract information about the real displacement. During linear translation, objects move differently depending on their position on the retina, relative to the direction of displacement. Objects in the regions of the visual field perpendicular to the translation direction move faster than those in the direction of flight. If the insect turns, objects move uniformly over the whole retina (i.e. at the same speed). By distinguishing translational from rotational components in the flow field, they may compensate for rotational movements (i.e. optomotor reaction) to correct for deviations in their flight path. In the example (right bottom), the animal turned clockwise and the retinal image moved in the opposite direction. Thick arrows over the insect heads indicate the direction of the movement, and thin arrows indicate the apparent movement of the objects in the visual field.

14

CLAUDIO R. LAZZARI

through their visual field which allows the insect to determine flight direction and speed, and to avoid collision. When insects fly towards a source of odour, the motor responses to visual and olfactory stimuli are superimposed. Olfactorymediated changes in muscle mechanics, wing kinematics and aerodynamics are somehow gated by visual feedback in a context-dependent manner to maintain a constant heading (Frye, 2007). The overall concept of such a system may also explain, for instance, how mechanosensory, chemosensory and visual information combine to allow tsetse flies and other haematophagous insects to track an odour source (Colvin et al., 1989; Hardie et al., 2001; see also Lehane, 2005). Compound eyes undergo dynamic changes in the distribution of pigment and in the position of photosensitive regions of the visual unit, the rhabdom. These changes allow the photon flux reaching the rhabdom to be controlled, enabling the eye to operate across a wide range of light intensities, particularly in crepuscular and nocturnal insects. Nocturnal mosquitoes and kissing bugs possess compound eyes capable of adjusting both their absolute sensitivity and spatial resolution (Land et al., 1999; Reisenman et al., 2002). During the day, their sensitivity to light is lower, but spatial resolution higher, than during the night. During the dark period, the eye becomes more sensitive, but sacrifices spatial resolution. These changes are linked to daily variation in the behavioural sensitivity to light, and, in kissing bugs at least, are under the dual control of circadian clocks and environmental light intensity (Reisenman et al., 1998, 2002). Several species of blood-sucking insects respond to the colour of objects by approaching or landing on them. This behaviour has been interpreted as evidence of colour vision and is exploited to attract them to targets impregnated with insecticides (Lehane, 2005). The range of wavelengths to which insects are sensitive varies with species. A common belief, originating from early classical studies of honeybees, is that insects are blind to red, or more precisely, that they are not sensitive to wavelengths of around 700 nm. This is true for honeybees and certain other insects, but not for all insects (Briscoe and Chittka, 2001). Indeed, spectral sensitivity, reminiscent of the capacity to distinguish between colours, has been modelled by the selective pressures acting on each species. The spectral sensitivity of the haematophagous insect Aedes aegypty ranges from 327 (UV) to 621 nm (orange) (Muir et al., 1992), whereas that of the kissing bug T. infestans ranges from 357 (UV) to 695 nm (red) (Reisenman et al., 2000; Reisenman and Lazzari, 2006).

5

Looking for food

As described above, various types of information are used by haematophagous insects to detect, localize and recognize a potential host. Each step of the feeding behaviour of blood-sucking insects may be associated with a different strategy used to acquire the relevant information and with specific mechanisms


15

allowing the insect to reach its food source. The strategies used to acquire information consist of particular behavioural patterns guiding the activity and displacements of the insect and are under the temporal control of endogenous clocks. The mechanisms involved determine which parameter of a given stimulus the insect responds to and underlie the way insect behaviour is then modified. Different phases involved in location of a host by haematophagous insects have been described in several studies. Most of the phases identified were defined according to the particular behavioural characteristics of the species studied. However, the same or similar general components can be recognized across different groups of blood-sucking insects, particularly in Diptera. I will describe these phases below, based on a modified version of the model proposed by Lehane (2005), taking into account recent findings from different groups of haematophagous insects. These different phases are thus referred to as: activation, appetitive search, host detection, host finding and host contact. 5.1

ACTIVATION

As in other animals, the daily life of blood-sucking insects is organized as a function of time and is controlled by circadian clocks (Barrozo et al., 2004b). When hosts are less active or the environmental conditions become adequate for food search, the insects start moving spontaneously. This activity may be restricted to just one part of the day or split into two periods (bimodal daily activity) (Brady, 1972c; Lazzari, 1992; Barrozo et al., 2004b). 5.2

APPETITIVE SEARCH

As the spontaneous activity increases and the insect starts walking or flying spontaneously, it leaves its initial refuge and moves over a progressively larger area. The non-oriented movements subside, to give way to movements that enhance the probability of the insect encountering signs that indicate the proximity of a host. This active behaviour may last for the whole active period (or periods, if bimodal activity), as observed for tsetse flies (Barrozo et al., 2004b), or during only a part of the daily activity period. Kissing bugs, for example, only devote their first period of bimodal activity to food search, whereas the second activity period is concerned with the return to their refuge (Lazzari, 1992; Lorenzo and Lazzari, 1998; Barrozo et al., 2004b). Bugs experience their maximal motivation to feed and exhibit their strongest responses to host-associated stimuli at the start of the night (Barrozo et al., 2004a; Bodin et al., 2008). In the presence of air currents, insects orientate their movements according the direction of the wind to optimize their search efforts (Fig. 4). If the wind direction is fairly constant (not differing by more than 30 in either direction),

16

CLAUDIO R. LAZZARI 30°

Upwind

Crosswind

Downwind

FIG. 4 During the exploratory phase of appetitive behaviour, insects need to adjust the direction of their displacement relative to the wind to increase their chance of encountering host-derived cues transported by air currents. For wind coming always from the same direction or shifting within a range of less than 30 , the most efficient way to increase the probability of encountering a stimulus is moving crosswind. When fluctuations in wind direction exceed 30 on both sides, the optimal strategy involves moving either upwind or downwind (Sabelis and Schippers, 1984; Dusenbery, 1990).

the most effective strategy is to move crosswind; however, for conditions in which changes in wind direction are greater than 30 , insects are more likely to encounter a chemical cue transported by the wind by moving either upwind or downwind (Sabelis and Schippers, 1984; Dusenbery, 1990). 5.3

HOST DETECTION

The detection of signals revealing the presence of a host may not be informative enough to elucidate the direction of the source. In such cases, the signal may exert an effect on the insect’s movements, enhancing the insect’s general locomotion activity or making the trajectory more or less direct. By moving faster and scanning a wider area, the insect eventually encounters a stimulus or combination of stimuli able to direct it towards the source and to determine whether the sensory signals detected belong to an adequate host (host selection or recognition). 5.4

HOST FINDING

As mentioned above, orientation towards a host is guided by direct and indirect cues (i.e. not associated to the host) and is mediated by a combination of kinetic and directional mechanisms. Orientation towards a distant host mostly relies on the combination of multimodal information. When host-associated odours are dispersed by the wind, they may reach blood-sucking insects looking for food and, if intense enough, may stimulate specific olfactory receptors. The chemical gradient established between the host and the insect is usually neither strong nor


17

stable enough to guide the insect as far as the odour source itself. Nevertheless, air currents loaded with host odours may provide insects with the information they need to locate a potential food source. Indeed, haematophagous insects, as observed for other insects, follow odour-laden airstreams moving upwind (odour-triggered anemotaxis). The orientation of insects following odour plumes has been analysed in depth, for moths tracking pheromones (e.g. Carde´, 1996). Available data on blood-sucking insects mostly concern tsetse flies, which perform odour-triggered anemotaxis (Warnes, 1990; Lehane, 2005). This orienting mechanism depends on wind direction and insect behaviour is subject to modulation by chemical stimulation. Even though airstreams provide directional information, neither wind nor odours provide signals precise enough for the insect to maintain a straight trajectory to the odour source. Integration of a third system is thus required, that is vision. The visual flow field helps by compensating for any rotational movement during displacements of the insect, allowing only translational components (Fig. 3). This is particularly important for flying insects, which lack any contact with a solid substrate and need to keep their path suspended in the air, at the same time being subjected to turbulence that causes them to deviate from their intended trajectories. The interplay of all these components, together with modulation of the tendency to turn (klinotaxis) with the odour concentration within an odour plume and upon intermittent contact with pulsed odours in turbulent air, gives rise to characteristic zigzagging trajectories during displacement towards the origin of the odour (Mafra-Neto and Carde´, 1994; Carde´, 1996). 5.5

HOST CONTACT

During the final approach to the host, odours gradually become more concentrated and chemical gradients are more easily perceived. These cues thus become more useful for directional orientation. At the same time, air currents and anemotaxis becomes less important and other signals, such as visual contrast, colour or movement, and heat and water vapour, can be perceived, bringing the insect in to land or into contact with the host body. Some haematophagous species exhibit characteristic preferences for biting at particular sites on the host body (Lehane, 2005), and orientate themselves using specific kairomones produced locally by the animal itself or by micro-organisms living in the preferred part of the body (Knols et al., 1994; De Jong and Knols, 1995). Arrival at these sites is the final step in the approach to the host, usually taking place before any physical contact is established. 5.6

HOST BITING

Most studies on food searching in haematophagous insects have focused on how the insects detect and follow host-emitted signals to find their food source. However, this is only part of the challenge faced by insects feeding on blood.

18

CLAUDIO R. LAZZARI

These insects confront another, largely neglected, problem: the localization of blood vessels. Once in contact with a host, the insect must find its food circulating in vessels below the skin surface. Regardless of the feeding strategy used (solenophagy or telmophagy), insects need to obtain blood as quickly as possible to avoid detection by the host. Two possible approaches can be envisaged: biting repeatedly at random until a vessel is located by chance or confining bites to the most irrigated areas. Biting at random sites does not require any particular ability to detect vessels. However, the insect runs a higher risk of being detected by the host if the host is repeatedly bitten. This is particularly relevant for telmophagous insects, which cause skin lacerations and pain, but also applies in the case of solenophagous insects, which cannulate blood vessels with their piercing mouthparts. A study by Ferreira et al. (2007) suggested that changes in temperature over the host skin, due to variation in irrigation, may help blood-sucking insects to locate blood vessels. The authors analysed the distribution of bites over the inner face of a rabbit ear. Most bites occurred over or close to blood vessels. The authors then tested the biting activity of bugs over a surface maintained at a given temperature, crossed by a linear heat source, which was slightly warmer than the background temperature. In this experiment, most bites were directed to the linear heat source (Ferreira et al., 2007). In both the living host and artificial model, the insects extended their proboscis or rostrum directly to the warmest place. This suggests that the signal used by the insects for directing bites is perceived and evaluated before any physical contact between the rostrum and the host skin. The crucial role of the antennae in the detection of heat emanated by the host suggests that antennal thermoreceptors, rather than rostral thermoreceptors, are involved in this process. Bugs with one antenna removed systematically miss the warmest area, directing their bites to the side of the remaining antenna. This is a typical behaviour, indicative of the tropotactic use of sensory information. When both antennae were removed, no response could be evoked, suggesting that only antennal thermoreceptors are involved in guiding biting (Ferreira et al., 2007). 5.7

FOOD RECOGNITION AND FEEDING

Once the skin is pierced, haematophagous insects start a probing phase, in which stylets are moved inside the skin to pierce a vessel, usually a venule. Having located the circulating fluid, the insect starts sucking it through the alimentary channel. Sensory receptors located near the pharynx recognize the blood through the detection of its key properties (Friend and Smith, 1977). Some haematophagous insects are sensitive to ATP and other adenine nucleotides and related compounds (e.g. GTP and GDP). These factors differ in their phagostimulant power (Galun and Margalit, 1970; Friend and Smith, 1975; Friend and Smith, 1977; Galun et al., 1988). For the sandfly Lutzomyia longipalpis and certain anopheline mosquitoes, the most important feature seems to be the tonicity of the solution, with no effect of ATP on feeding in these insects


19

(Lehane, 2005). This has also been observed for T. infestans, which is particularly sensitive to food osmolarity (Guerenstein and Nuń˜ez, 1994). The use of phagostimulant compounds or osmolarity to trigger gorging seems to be dependent on thresholds in some species. For example, R. prolixus shows an all-ornothing gorging response to phagostimulants, which switches after a long starvation period to a graduated response to food osmolarity, similar to that seen in T. infestans (Guerenstein and Nuń˜ez, 1994). The comparison of the strategies adopted by these two closely related triatomines is particularly interesting because their differences in feeding behaviour may be related to the type of habitat and host exploited by each species. Wild R. prolixus associates with sylvatic hosts, whereas T. infestans almost exclusively targets human dwellings. Access to a potential host is much less predictable in wild habitats than in anthropic habitats. Thus, less selective strategies and partial meals may be advantageous for survival in areas of permanent proximity to potential food sources (e.g. T. infestans) or during prolonged periods of food deprivation (e.g. the long starvation periods of R. prolixus); however, when food sources are less predictable, it may be more advantageous to be selective and to have large meals (e.g. the all-or-nothing R. prolixus response). The sensory basis for the perception of phagostimulants and osmorality seems to differ between different blood-sucking species. In mosquitoes, phagostimulants activate receptors located in the mouthparts (Werner-Reiss et al., 1999a,b). In triatomines, although both the maxillae and mandibles are innervated, their sensory nerve endings appear to be only mechanoreceptive, with their response to phagostimulants mediated by sensilla within the food canal, probably by the eight peglike sensilla located in the epipharynx (Bernard, 1974; Friend and Smith, 1977). For some haematophagous insects the contact with host skin triggers another process associated with blood ingestion, the change in the mechanical properties of the abdominal cuticle. This ‘‘plasticization’’ process is similar to the process allowing extension of the new cuticle in the period between ecdysis and the sclerotization of the exoskeleton. In triatomine bugs, at least, plasticization occurs each time a larva has a blood meal. Plasticization is triggered by the simple contact of the proboscis with a warm object, feeding not being necessary (Ianowski et al., 1998). During feeding, serotonin is released from neurohaemal organs located at the abdominal wall. The released serotonin acts on many targets throughout the body to coordinate the physiological changes associated with feeding, probably including plasticization (Orchard et al., 1988; Lange et al., 1989; Barrett and Orchard, 1990; Orchard, 2006). Changes in the molecular structure of the endocuticle (disappearance of low-energy bonds between protein chains) allow the endocuticle to extend during the feeding period. These changes are then reversed, with the original stiffness recovered within about 1 h (Melcon et al., 2005). If a bug cannot complete a meal, plasticization may occur repeatedly during the same larval instar once the cuticle has recovered its normal stiffness (Melcon et al., 2005). The sensory mechanism responsible for triggering this process is only partially understood.

20

CLAUDIO R. LAZZARI

As mentioned above, mechanical contact with a warm object is sufficient for evoking a response. However, given that heat is necessary to evoke contact in the first place, it is difficult to separately examine the effects of the individual stimuli. It is possible that mechanoreception alone is linked to plasticization, with heat evoking proboscis extension. Indeed, thermoreceptors have not been conclusively shown to exist in the proboscis, but mechanosensory sensilla are abundant (Bernard, 1974; Catala, 1996). Despite the key role of heat in the various processes of host localization, biting and plasticization, blood temperature is not involved in food recognition by triatomines. Indeed, blood as cold as 3 C can be ingested by haematophagous bugs if their antennae are sufficiently stimulated by heat for the bugs to extend their proboscis and to bite (Lazzari and Nuń˜ez, 1989b). 5.8

LEAVING THE HOST

The end of the gorging process is marked by irregular activity of the ingestion pump, which decreases rapidly to eventually stop once the abdomen is full of blood. Two different mechanisms may cause feeding to stop, one based on the increase of mechanical resistance to blood pumped into the intestine (BennetClark, 1963) and another based on a neural pathway involving stretch receptors in the abdominal wall (Anwyl, 1972; Nijhout and Sheffield, 1979; Belzer, 1979; Chiang and Davey, 1988). These mechanisms are not necessarily exclusive and the process may involve a combination of factors. Once gorged, blood-sucking insects leave their host and look for a protected place. Many haematophagous insects ingest large meals, meaning that they then have to displace an extra weight that may be many times their own body weight. Triatomines are particularly vulnerable to aggressive hosts, due to the plasticization of their abdominal cuticle. The most effective behavioural response is to move away from the host as quickly as possible to escape from any dangerous antiparasitic reaction or from predation by their food source. Theoretically, the same cues serving to approach a potential host can be used to move away from it. Host-associated volatile compounds and heat can guide the insects just by inverting the direction of the orientation setting mechanism (Jander, 1963). As I will described below, a switch in the response to carbon dioxide from attraction to repulsion occurs in triatomine bugs some time after feeding (Bodin et al., 2009b); it remains unclear whether this is the case just after gorging.

6

Stimulus propagation and sensory reception

The location in space of the source of a stimulus depends on several factors, including: (1) the specific nature of the signal and the way it is propagated, (2) the ability of sensory organs to gather directional information about the


21

source and (3) the way in which the animal uses the sensory information or the orientation mechanism (Fig. 5). Insects, as observed in other animals, are also able to use indirect cues to orientate themselves to a particular site. However, haematophagous insects do not seem to use indirect cues to locate a host. Eyes are able to provide precise spatial information, thanks to both the discrete nature of light and the arrangement of photoreceptors in the retina. Other sensory systems, such as the wind-sensitive cercal system of Orthoptera and cockroaches, which uses directional information from the deflection of hairs, are able to determine the direction that the stimulus came from, but not the exact location of its source in space. In other sensory systems, the sensory information must be actively gathered. In some cases, the location of the source of a stimulus in space requires successive comparisons of sensory information. This is usually accompanied by active movement, either of the whole individual, or of the organs bearing sensory receptors, that is the antennae, to detect the stimulus intensity gradient. Movements concerning the whole individual are characterized by turning motion, associated with klinotaxic orientation.

A

B

C

D

FIG. 5 Examples of factors determining the acquisition of sensory information. (A) An animal with an unpaired sensory organ can locate the source of a stimulus diffusing in the form of a gradient (e.g. chemical volatiles, water vapour) by comparing the intensity of the stimulation at different moments and spatial positions. (B) When sensory organs are distributed bilaterally, as in insect antennae, directional information about the same type of source can be obtained by comparing stimulation both simultaneously (between bilateral inputs) and successively, as a function of displacement. (C) Stimuli propagating radially (e.g. light) provide more precise directional information, which can be identified by animals by successive comparison of stimulation of unpaired sense organs or by comparing stimulation of bilateral organs; in the case of spatial vision, the pattern of stimulation of different regions of the retinal lattice of one eye is sufficient. (D) Some radially propagating stimuli, such as radiant heat, could be perceived using the stimulation of specific sensory organs distributed over the antennae (e.g. kissing bugs) or other regions of the body (e.g. pyrophilic beetles).

22

CLAUDIO R. LAZZARI

The antennae can also be moved actively to collect information. Antennal movements occur when insects are confronted with a source of stimuli. For example, kissing bugs, T. infestans and R. prolixus display typical antennal movements when exposed to a heat source (Wigglesworth and Gillet, 1934a; Flores and Lazzari, 1996). These movements differ depending on whether the insect is walking or standing during its approach to the source. During the standing phase, antennae move synchronously and in a saccadic fashion, across wide angles; when walking, both antennae are kept at a constant angle in the horizontal plane, moving smoothly up and down in the vertical plane (Flores and Lazzari, 1996). Saccadic and smooth antennal movements can be interpreted as the ‘‘scanning’’ (of the environment) and ‘‘fixation’’ (of the source) phases. Thus, insects stop walking to scan wide regions of the environment with antennal saccades and, when the direction to the thermal source is established, the insect starts walking, fixing the source with the antennae (Flores and Lazzari, 1996; Flores, 2001). We currently have a relatively good understanding of the physiology, transduction mechanisms and molecules involved in reception of the signal, the first level of sensory information processing. This concerns all the sensory systems, vision (e.g. Warrant and Nilsson, 2006), olfaction (e.g. Qiu et al., 2006; Xia et al., 2008) and thermoreception (e.g. Gingl et al., 2005). Considerable efforts have been made to try to understand how haematophagous insects respond to sensory cues (e.g. Takken and Knols, 1999; Zwiebel and Takken, 2004; Barrozo and Lazzari, 2004a). Relatively little is known, however, about how sensory information is acquired and how different systems are integrated to locate a host. Even though it is widely accepted that a given behavioural input results from the interplay of many external and internal factors, the roles of multimodality, state-dependency and active gathering of information remain unclear in most cases. In this section, we will discuss some of the processes that occur in an intermediate stage, between sensory reception and behaviour. We will firstly define the three levels of information processing, determining how animals acquire and make use of information to locate resources. The first level of processing is ‘‘reception,’’ which refers to peripheral events in the sensory organs, occurring as a consequence of a change in the physical or chemical properties of the environment. In physiological terms, this phase determines the type of variables the sensory system of a given animal is potentially able to detect. Reception is usually studied using molecular tools and electrophysiological methods to analyse the transduction of external signals as nervous activity. The main output from such studies is obtained in the form of spectra showing the type and intensity of stimuli received and transduced into nervous signals by a given sensory organ. The second level is ‘‘perception,’’ which we use here to refer to the central processing of sensory information, that is filtering and multimodal integration of sensory inputs and endogenous signals. The analysis of these processes requires several approaches, including electrophysiological


23

and optophysiological techniques to study central nervous activity and behavioural methods (black-box analysis of inputs and outputs). The final level involves behavioural output, revealing how the animal makes use of the information gathered from the environment. The distinction between these three levels helps us to determine the type of information we expect to obtain from the different types of experiments and to interpret the results obtained. As an example, the fact that a given chemical compound stimulates a particular olfactory receptor neuron does not necessarily mean that such a compound will modify the behaviour of the insect. This input may require integration with other inputs from the same or different systems and with endogenous signals defining a particular physiological state (determined by circadian clocks, nutritional and reproductive status, and other physiological processes), to cause a behavioural effect. Similarly, the fact that an insect possesses two or more different types of photoreceptor sensitive to different wavelengths does not mean that it is able to use colour vision to solve different visual tasks (object recognition, movement detection etc.). Conversely, the preference for landing on particular colours of targets cannot be considered as evidence of colour vision either; indeed, other possibilities such as intensity-guided choice cannot be excluded (Kelber et al., 2003; Kelber, 2006). In summary, the behavioural output shows us how the information is actually used by the insect. However, behavioural tests can be time and effort consuming, and a large number of individuals usually need to be tested to compensate for individual variability and obtain sufficient statistical power. Studies of reception and perception processes usually require expensive equipment. However, the independence of data and rigorous statistical analysis are not always required for publication, because of the limited number of experimental units that can be obtained in some cases (e.g. the same neuron in different individuals). The combination of different approaches may be the most effective strategy. For example, in studies aimed at identifying biologically relevant compounds in a complex mixture, initial screening using an electroantennogram (EAG) may exclude compounds that do not evoke sensory activity.

7

Orientation mechanisms

The way in which sensory information is made available and becomes exploitable for an animal depends on the nature of three main factors: signal propagation, the morphology and function of sensory tools and processing of the information by the central nervous system. Figure 5 illustrates the effects of signal propagation and sensory tool morphology and function on the orientation of animals. Two types of stimulus are represented. The first type includes stimuli that disperse through the medium, giving rise to intensity gradients emanating from the source. Dispersion of these signals is dependent on turbulence and flow and the stability of the signal is variable, depending on

24

CLAUDIO R. LAZZARI

movements of the air. Examples include the dispersion of chemicals or water vapour and warm air gradients generated through conduction. The second type includes discrete stimuli that disperse radially. They are not affected by movements of the medium and thus provide accurate directional information. This group includes light and infrared radiation. The sensory organs detecting these stimuli may or may not able to discriminate the potential spatial information provided. Eyes, for instance, generate an image on each retina allowing the location of objects to be determined, even using just one organ. The lattice of receptive units on each eye provides sufficient directional information to locate a source. Other sensory organs are organized differently, requiring comparisons of stimulation intensity between bilateral inputs to elucidate the direction the stimuli are coming from. Insect ears are classical examples, in which the comparison of bilateral inputs allows insects to locate the source of a sound. The animal turns around and, when the bilateral inputs equalize, the insect heads towards the source. To reach a source, information from the environment can be used in different ways, requiring different orientation mechanisms. The nature of the stimuli and the properties of the sensory organs are two major factors determining the orientation mechanism used. A third factor is how the sensory inputs are integrated in the central nervous system. Indeed, the orientation in space of an animal using the same information can be controlled by different mechanisms (Fraenkel and Gunn, 1961; Jander, 1963; Dusenbery, 1992). Some mechanisms modulate activity as a function of the intensity of a given stimulus, independent of the direction the stimulus is coming from kinesis. In other words, the individual can reach the source just by modulating the velocity of its displacement (orthokinesis) or by modulating its tendency to turn (kinokinesis), without using directional information at all. In other animals, directional information is taken into account and the animal moves as a function of the spatial position of the stimulus source (taxis). These mechanisms are not mutually exclusive. Haematophagous insects use these mechanisms in combination when approaching a host. The traditional classification of orientation mechanisms (Fraenkel and Gunn, 1961; Jander, 1963; Dusenbery, 1992) may be considered out of date or too restrictive. However, it is not only very useful, but essential, for designing experiments, interpreting results and understanding how information is acquired and used to find resources. According to the classical definition (Fraenkel and Gunn, 1961; Dusenbery, 1992), stimuli may affect behaviour in two ways, affecting either kinetics or direction. In the first case, called ‘‘kinesis,’’ the intensity of the stimulus (and not the direction that it comes from) modulates either locomotion speed (orthokinesis) or the turning tendency of the animal (klinokinesis). In the second case, the animal extracts directional information from the signal (taxis). The direction of the source may be evaluated by successive comparisons of stimulation received during a programmed pattern of alternate left and right movements (klinotaxis) or by the simultaneous comparison of bilateral inputs (tropotaxis) or


25

of receptive units of a single organ (telotaxis); in cases involving receptive units of a single organ, each component of a bilateral pair of sensory organs (eyes, antennae) provides enough information to determine the direction of the source. In most cases, both elements, stimulus structure and spatial resolution of sensory organs, are linked by the fact that sensory organs devoted to detecting stimuli providing poor directional information, such as chemicals, are not able to extract directional information unilaterally, so that telotaxis is not possible. It should be noted, however, that the relationship between the signal, sensory organs and orientation mechanism varies, even for the same individual. Kissing bugs approaching a source of heat orientate by telotaxis until they are about 2 cm from the source, switching to tropotaxis to bite the host at the site of a blood vessel, for instance (Wigglesworth and Gillet, 1934a; Flores and Lazzari, 1996; Ferreira et al., 2007). Thus, insects with just one antenna (i.e. unilaterally antennectomized) are able to approach a heat source, but miss the goal when they try to extend their proboscis towards it. The orientation mechanism associated with the response of a particular insect to a particular cue may provide important information about how sensory information is acquired and employed. Taking the example of kissing bugs, they approach heat sources by telotaxis from relatively long distances (Flores and Lazzari, 1996), walking in the same horizontal plane as the source or below it. Ascending convection currents are therefore rarely involved, being effective only when the insects approach from above, as shown in mosquitoes by Eiras and Jepson (1994). Given that animals with just one antennae approach the source by walking in a straight path (Wigglesworth and Gillet, 1934a; Flores and Lazzari, 1996), they are using neither bilateral simultaneous comparison (if this were the case, the path would deviate towards the intact antennae, Jander, 1963), nor successive comparisons (otherwise, the path would meander, Dusenbery, 1992). The ability to approach a source using directional information from just one sensory organ depends on two criteria being met: first, that the stimulus provides reliable directional information over the required distance, and second, that the sensory organs provide spatial information unilaterally. As described above, both of these criteria are met in the normal functioning of the eyes, the classical example of telotaxis, by forming images on the retina; but this is not the case for insect antennae. We can now evaluate the effects of each of the three mechanisms of heat exchange – convection, conduction and radiation – on the potential use of thermal telotaxis by an insect walking towards a source of heat. Convection currents propagate upwards and cannot be perceived by an insect located tens of centimetres away from, and at the same level or below, the source, and thus are not likely mediators of thermal telotaxis. Warm air gradients extend in all directions and could reach the insect. However, not only are gradients of air temperature disrupted by externally induced turbulence, but they themselves create turbulence in the form of ascending convention currents. Thus, the resulting signal may not be sufficiently stable to provide precise directional

26

CLAUDIO R. LAZZARI

information. Convection currents and warm air gradients may still provide some directional information; however, the signal is in a non-discrete form and does not provide a sufficient level of directional information to promote thermal telotaxis. The third mechanism, heat radiation, being independent of air turbulence and propagated in the form of waves in all directions, seems to be the only signal that provides information in an appropriate form and at a sufficient level for thermal telotaxis to take place. Kissing bugs are the only group of haematophagous insects for which IR perception has been demonstrated (Lazzari and Nuń˜ez, 1989a; Schmitz et al., 2000), although IR perception has been observed for other heteropterans displaying phytophagous and pyrophilous activity, as well as in some Coleoptera (Schmitz and Bleckmann, 1997; Schmitz et al., 2008; Takaćs et al., 2009). It remains less clear how insects with just one antenna are able to determine the direction of a heat source from a distance, with sufficient precision to maintain a straight path (i.e. without klinotaxic turns). One possibility is that the insect recognizes the spatial pattern of thermoreceptor stimulation along the length of the antenna or even the body. The insect does not need to create a thermal ‘‘image’’; rather, it must simply recognize which side of the antenna (or body) is receiving more stimulation. This example thus demonstrates the importance of taking into account the orientation mechanism in determining the nature of the biological signal detected and how the sensory information is used.

8

Thermal sensing in kissing bugs

As mentioned above, triatomine bugs rely on various sensory cues to search for a host. These cues are mainly olfactory and thermal. The thermal cue is the only one capable to induce biting by the insect (Wigglesworth and Gillet, 1934a; Flores and Lazzari, 1996). Additionally, triatomine bugs are the only haematophagous insects that have been demonstrated to perceive radiant heat (Lazzari and Nuń˜ez, 1989a; Schmitz et al., 2000). When confronted with heat sources of different sizes and located at various distances from the source, triatomine bugs respond only to objects with a temperature corresponding approximately to that of a warm-blooded vertebrate host (Nicolle and Mathis, 1941; Fujita and Kloetzel, 1976; Lazzari and Nuń˜ez, 1989a). Generally, a distant (or small) burning source and a close (or big) tepid source are easily identified by these bugs. The amount of heat reaching the thermoreceptors depends not only on the temperature of a heat source, but also on the distance between the insect and the object, the surface area of the emitting object and the environmental temperature. Thermal sensory organs alone therefore do not provide enough information to determine the temperature of the source. To achieve this, the insects need to be able to determine the temperature and distance from a heat source, independently of the surface area of the source, using only thermal information (Flores and Lazzari, 1996).


27

This capacity may be related to their ability to perceive radiant heat (Lazzari and Nuń˜ez, 1989a). Radiation is a more precise cue than heat conduction or convection; consequently, the distance to a thermal source may be estimated using information derived from the different levels of stimulation at different sites on the antennae. In particular, the angle of incidence, surface area and temperature, would provide useful information. The importance of heat in triatomines is highlighted not only by the high level of sensitivity of their thermal sensing and by their ability to evaluate the thermal properties of a distant source, but also by its various the biological effects. Indeed, heat is the only host-associated signal, both necessary and sufficient to induce the proboscis extension response (PER) in these bugs and to induce them to bite an object. As mentioned, odours are important cues in the orientation of the bugs. However, an object emitting odours at room temperature is never bitten. Only objects at temperatures between a few degrees above ambient levels and about 47 C are considered to be potential hosts (Lazzari and Nuń˜ez, 1989a). The overall behaviour of these insects changes dramatically in the presence of a thermal source. Insects in a state of akinesis (immobility) ‘‘wake up’’ (Wigglesworth and Gillet, 1934a) and display characteristic movements of their antennae, sweeping the air within the horizontal and vertical planes at particular angles (Wigglesworth and Gillet, 1934a; Flores and Lazzari, 1996; Flores, 2001). When active, insects typically alternate between standing and bouts of walking, displaying two different patterns of antennal movements (Figs. 6 and 7). While standing, the movements are jerky and the antennae sweep at wide angles, but while walking, the movements become smooth, and the angles between the antennae remain largely constant, as described earlier. The insect’s standing phases can be interpreted as phases during which the insect scans for odours and heat, whereas during phases of walking, the antennal positions may function as a possible means of detecting the edges of a warm object (Flores and Lazzari, 1996). These different antennal movements are observed in particular when the insect is approaching a thermal source. These observations suggest that the antennae, which bear thermoreceptors (Bernard, 1974), play a central role in the insect’s search for blood. However, how the information conveyed is then used by the insect remains unclear. The basic laws of thermodynamics predict that the information received by the insect upon delivery of an amount of heat is ambiguous. A large, hot, distant object could produce the same local warming effect on the insect’s antennae as a small tepid object placed at a shorter distance. Thus, other characteristics of a thermal source may provide additional and useful information about the source. To recognize a host using temperature (between 32 and 42 C at the body surface of most warm-blooded vertebrates), the insect must integrate information based on the amount of heat reaching the antennae with information concerning the distance of the source and its size. Together with temperature, the source and size of the source directly affect the amount of heat reaching the

28

CLAUDIO R. LAZZARI

Angle (degrees)

A

B 90

90

60

60

30

30

0

0

−30

−30

−60

−60

−90 0.0

0.5

1.0

1.5

2.0

Angle (degrees)

C

−90 0.0

0.5

1.0

1.5

2.0

2.5

3.0

D 90

90

60

60

30

30

0

0

−30

−30

−60

−60

−90

0

2

4

6

8 10 Time (s)

12

14

16

−90

0

2

4

6

8 10 Time (s)

12

14

16

FIG. 6 Sample records of antennal movements performed by blood-sucking bugs during their approach to a heat source. The insects alternated between walking and standing, performing characteristic antennal movements in both horizontal and vertical planes (A) horizontal and (B) vertical antennal movements during walking; (C) horizontal and (D) vertical sample records of saccadic antennal movements in a standing insect. Positive values in (A) and (C) correspond to the left antenna and negative values correspond to the right antenna. In (B) and (D), positive values correspond to angles above the horizontal plane and negative values correspond to angles below the horizontal plane. During both phases, antennae are moved in coordination (modified from Flores and Lazzari, 1996). −20

−20

Right antenna

Walking

Stop

−30

−30

−40

−40

−50 −60

−50

−70

r2 = 0.627

−80 −90

P < 0.001 N = 62 0

10

20

30 40 50 Left antenna

60

70

r2 = 0.944 P < 0.001 N = 486

−60 −70 30

35

40

45 50 55 Left antenna

60

65

70

FIG. 7 Coordination of the movement of both antennae during stopping and walking periods of a haematophagous bug approaching a thermal source. Movements are coordinated and the positions of the right and left antennae are significantly correlated in both cases (Flores and Lazzari, unpublished).


29

insect’s thermoreceptors. Thus, several possible combinations of these three variables may give rise to the same amount of heat. T. infestans is able to estimate the distance to a thermal source in the absence of visual information or physical contact with the object, relying only on thermal information (Lazzari and Nuń˜ez, 1989a; Flores and Lazzari, 1996). The lack of one antenna causes errors in targeting the goal, but does not affect the distance at which the insects try to bite (Flores and Lazzari, 1996). So far, there are no data available showing the ability of any insect to estimate the size of an object using only thermal information. This possibility was investigated in kissing bugs (Flores and Lazzari, 1996; Lazzari and Flores, unpublished). Insects were placed under open-loop conditions (i.e. at a fixed position). Heated aluminium square plates (34 C) of three different sizes were placed at a fixed distance from the insect, or plates of a given size were placed at three different distances (Fig. 8). Each assay was carried out using a single plate placed at a particular distance. Plate distances and sizes were chosen to give three different subtended angles (i.e. 14 , 21 and 28 ) at the insect position. The angle formed by the antennae during locomotion (i.e. tendency to approach the source) was recorded during each experiment. In both experimental series, the mean angle formed by the antennae significantly increased with the size of the source (ANOVA, p < 0.0001 in both cases). It was also observed a direct relationship between the antennal angles and the angles subtended by the source at the position of the insect, regardless of the size or distance of the source used in each experimental series (Fig. 9) (Flores, 1996; Lazzari and Flores, unpublished). We compared data sets from the two series using a paired t-test. We did not observe any difference in the mean antennal angle between groups of insects exposed to objects displaying the same subtended angle, regardless of the actual distance or size of the source.

14⬚

40 cm

21⬚

27 cm

28⬚

20 cm

FIG. 8 Experimental conditions for testing the relationships between the position of the antennae (antennal angle) and the size and distance of the thermal source. In a first series of experiments, a square (10 10 cm) plate heated to 34 C was placed at 20, 27 or 40 cm, at solid angles of 14 , 21 and 28 subtended by the insect position (open-loop conditions). In a second series, the same subtended angles were obtained by placing plates of different sizes (5 5, 7.5 7.5 and 10 10 cm) at 20 cm of the insect. The position of the antennae during the walking phases of independent groups of insects was recorded for a period of 2 min and the mean antennal angle was determined for each individual.

30

CLAUDIO R. LAZZARI 80

Antennal angle (degrees)

n.s.

75

27 cm 20 cm n.s.

70

20 cm 40 cm 20 cm

65

n.s. Constant distance, variable size Constant size, variable distance

60 14

21 Solid angle subtended by the source (degrees)

28

FIG. 9 Antennal angles in insects exposed to sources subtending various solid angles. Significant differences in the position of the antennal angle as a function of the apparent size of the source were found for both experimental series (i.e. varying distance or actual size, ANOVA p < 0.0001) However, no differences were observed between assays for plates differing in size or distance to the insect, but subtending the same solid angle. Black, grey and white rectangles represent the differently sized plates that were placed at different distances from the insect to obtain similar subtended angles.

These findings reveal several important features of thermal sensing in haematophagous bugs. First, the insects can actively control the position of their antennae according to the apparent size of a thermal source. Second, the insects can integrate thermal information from both antennae in such a way that enables them to maintain a constant angle. This finding confirms previous findings from orientation experiments (Wigglesworth and Gillet, 1934a; Flores and Lazzari, 1996). Third, these insects integrate proprioceptive inputs (probably from mechanoreceptors located at the joints of the antennae) with thermal information to produce the appropriate motor output to control antennal movements. According to the laws of thermodynamics, the amount of heat reaching a thermoreceptor depends on three variables: the distance to the thermal source, the surface area of the source and its temperature (more precisely, the difference between the temperature of the source and that of the environment). The processes allowing triatomine insects to estimate distance remain unclear, but may involve simple mechanisms. Previous experiments on the orientation of bugs towards thermal sources suggest that the insects integrate bilateral inputs


31

from the antennae through triangulation, i.e. by comparing the amount of heat received simultaneously by two thermoreceptors located at different sites on the body (e.g. on each antenna) (Flores and Lazzari, 1996). The insects may also be comparing the increases in heat energy as they approach the thermal source using internal representations of these increases. Studies on the response to heat in unilaterally antennectomized insects (Wigglesworth and Gillet, 1934a; Flores and Lazzari, 1996), together with the fact that heat increases at a constant rate with decreasing distance for objects of different temperatures and sizes (Fig. 10), are consistent with a second possibility. In this case, the insects compare the amount of heat received by the same thermoreceptor at successive time points during their approach to the thermal source or use the temperature gradient along the length of the antennae and the rest of the body (Fig. 11).

0.06 30 °C 35°C 40 °C 45 °C 50 °C

0 05

Energy (watt)

0.04 x4 0.03

0.02

0.01

x4 x2.25

x1.78

x1.56

0 0

0.5

1

1.5 2 2.5 3 3.5 Distance from the source (cm)

4

4.5

5

FIG. 10 A possible mechanism used to estimate the distance to a heat source, based on the gradient of heat energy decrement around a heat source. The figure depicts the gradient of thermal energy formed around a heat source of 1 cm2 area in an environment at 25 C. Each curve represents a different source temperature. The relative rate of decrement is independent of the temperature of the source. Similar curves can be obtained for sources differing in size, but sharing the same temperature. During the approach, the insect experiences heat energy increasing at different rates, as a function of the insect’s relative position. Successive comparisons of thermal stimulation during the approach or simultaneous comparisons between two regions of the body could allow bugs to estimate the distance to the source. These mechanisms could underlie thermotelotaxis in triatomines, a process that requires evaluation of the direction of a source even if bilateral integration is abolished by atennectomy. They may also determine how triatomines estimate their position relative to a heat source when only thermal information is available (Wigglesworth and Gillett, 1934a; Flores and Lazzari, 1996).

32

CLAUDIO R. LAZZARI 32.2 ⬚C

33.0 ⬚C

31.5 ⬚C

32.2 ⬚C

30.8 ⬚C

31.4 ⬚C

30.1 ⬚C

30.6 ⬚C

29.4 ⬚C

29.8 ⬚C

28.7 ⬚C

29.0 ⬚C

28.0 ⬚C

28.2 ⬚C

27.4 ⬚C

27.4 ⬚C

26.7 ⬚C

26.5 ⬚C

26.0 ⬚C

25.7 ⬚C

25.3 ⬚C

24.9 ⬚C

24.6 ⬚C

24.1 ⬚C

23.9 ⬚C

23.3 ⬚C

FIG. 11 Thermographic images of a haematophagous bug exposed to a heat source (35 C; diameter 2.5 cm). A temperature gradient is established along the antennae. The form and intensity of the gradient over each antenna varies as a function of the exact position of the organs during active movement.

Preliminary experiments suggest that the way in which distance is estimated differs depending on whether the insect itself approaches a heat source or whether the source appears suddenly at a fixed position. In the second case, the insect reacts extending its proboscis when the source is at a much shorter distance. These results seem to suggest that these insects use successive rather than simultaneous comparisons. The exact mechanism, however, remains to be confirmed, since the thermal information received in both cases theoretically provides sufficient information to determine the distance to the source. Based on our findings, previous experimental evidence and basic thermodynamics, a simple model of the multimodal integration of thermal and proprioceptive information can be proposed as a hypothesis to explain how triatomines recognize a host using thermal cues. The insects may elucidate the actual size of a thermal object by integrating information on distance to the thermal source and the subtended angle (the apparent size of the source). The temperature of the object could then be estimated using information on the amount of heat energy reaching the thermoreceptors and the distance to and size of the source. This would allow a host to be recognized by its temperature alone, allowing the insect to react accordingly. This model (Fig. 12) is based on the multimodal integration of proprioceptive and thermoreceptive inputs, the comparison of thermal stimuli and basic thermodynamics. The heat emanating from the source stimulates thermoreceptors. When the insect moves its antennae under proprioceptive control, the position of the thermoreceptors relative to the source changes, causing a change in the intensity of thermal stimulation. By integrating these changes with proprioceptive inputs to determine the position of the antennae, the insect could detect the edges of the source and thus estimate its apparent size. The insects may also determine the distance to the source based on the inverse square relationship between heat energy and distance. This can be achieved either by simply

ORIENTATION TOWARDS HOSTS IN HAEMATOPHAGOUS INSECTS Heat

Antennal mov.

Thermoreceptors

Proprioceptors

33

Multimodal integration

Apparent area

Bilateral integration

Energy amount

Actual size

Distance Comparison

E1 ≈ (Tsource4-Tenv4) A d2

central integration

Temperature

Reject

No

Host?

Yes

Approach

FIG. 12 Model of integration of thermal and proprioceptive information. Heat emanating from the source stimulates thermoreceptors. The intensity of the thermal stimuli and the position of the thermoreceptors change when the insect moves its antennae under proprioceptive control. The insect may estimate the apparent size of the source through integration of information on both the intensity of the stimuli and the position of the antennae. Alternatively, the insect may use the inverse square relationship between energy and distance to determine the distance to the source, either through simultaneous comparisons of the heat energy reaching different parts of its body or through successive comparisons of the heat energy reaching a certain part of its body during its approach to the thermal source. By integrating the distance with the apparent size of the source, the insect could easily determine the actual size. Thus, the insect would possess all the required variables, as described in the Stephan–Bolzmann law (inset), to estimate the temperature of a distant source, that is the heat energy reaching thermoreceptors, the surface area and the distance of the source, and the environmental temperature directly measured by the thermoreceptors. The insect could then use this information to ‘‘decide’’ whether a warmth object qualifies as a host, allowing it to react accordingly (Lazzari and Flores, unpublished).

comparing the heat energy reaching different parts of its body simultaneously or through successive comparisons of the heat energy reaching a certain part of its body as it approaches the thermal source. The insect could then easily compute the actual size of the source by integrating information on the distance with the information on the apparent size of the source. Thus, according to the Stefan– Bolzmann law, the insect would have all the information needed to estimate the

34

CLAUDIO R. LAZZARI

temperature of a distant source, that is the heat energy reaching thermoreceptors, the surface area of the emitting source, the distance to the source and the environmental temperature measured directly by the thermoreceptors. The insect could then use this information to determine whether a warm object qualifies as a potential host and whether or not to approach it. Triatomines display a very high level of behavioural sensitivity to heat, reacting to very low amounts of heat energy that reach them. In the experiments by Lazzari and Nuń˜ez (1989a), strong responses were obtained to a blackpainted circular object, with a 2 mm diameter and a temperature of 30 C (room temperature 20 C). These responses remained statistically significant when three sheets (250 mm each) were placed between the source and the insects, preventing detection of any warm air gradients or convection current and reducing the amount of IR radiation reaching the insects by 64%. Under these conditions, given that the amount of thermal energy decays with the square of the distance, the Stefan–Bolzmann law gives a value of about 3.3 mWatt/cm2 for the amount of heat energy that should have reached the insect in these experiments (Lazzari and Nuń˜ez, 1989a; Lorenzo et al., 1999). This would imply that the insects can perceive the heat emitted from a host located several meters away. This value is lower than the thresholds computed for other animals that respond to radiant heat, such as snakes (10 mWatt/cm2), vampire bats (50 mWatt/cm2) and Melanophila beetles (60 mWatt/cm2) (Campbell et al., 2002). If these results are confirmed, triatomines would possess one of the most sensitive thermal sensing systems in animals.

9

Sensory parsimony

Despite the fact that host-related information can be relatively specific and that insects undergo a process of selection of both their hosts and the sites on the vertebrate body at which they prefer to bite, specific sensory cues are not necessarily involved. Many haematophagous insects seem to make use of the same cues to locate different resources, through parsimonious use of sensory information. In such cases, interpreting a particular type of such information does not depend on specific cues, but is dependent on the biological context. 9.1

PARSIMONIOUS USE OF INFORMATION IN BLOOD-SUCKING INSECTS

Tsetse flies spent much of their time resting on plants, which provide refuge and protection from excessive heat (Leak, 1999). Some of these plants release chemicals that attract the insects. The combined study of the volatile substances released by the invasive plant species in Africa, Lantana camara, and of the sensory and behavioural responses of tsetse flies, led to the identification of the compounds involved in this biological process (Syed and Guerin, 2004). The study showed that, in addition to plant-specific compounds, Lantana


35

releases 1-octen-3-ol (octenol). Octenol is part of the mixture of volatile compounds released in the breath of vertebrate hosts and is a powerful attractant for tsetse flies and other haematophagous insects (Hall et al., 1984; Vale and Hall, 1985; Takken and Kline, 1989; Barrozo and Lazzari, 2004b). Octenol can thus be used as a chemical lure for trapping tsetse flies in the field (Leak, 1999). Mosquitoes, as observed for many other haematophagous insect, use carbon dioxide to locate potential hosts (Gillies and Wilkes, 1969; Gillies, 1980; Takken and Kline, 1989). The sensitivity of several species to CO2 and the synergic interaction between CO2 and other volatile compounds have been characterized (Acree et al., 1968; Geier et al., 1999b). Carbon dioxide is thus widely used as an attractant in different kinds of devices to capture mosquitoes (Burkett et al., 2001). Carbon dioxide is also released by opening flowers, where it is possibly used by insects sensitive to CO2 to find nectar (Guerenstein and Hildebrand, 2008). In mosquito maxillary palps the same sensilla house both olfactory receptor neurons (ORN) sensitive to CO2 and ORNs sensitive to 1-octen-3-ol and to plant odours (Syed and Leal, 2007). With a dual diet of blood and nectar, female mosquitoes may use the same sensory receptors to locate both food sources. Some haematophagous arthropods seem to use volatile compounds that are also released by their hosts for intraspecific communication. Thus, the same compound acts as a host-associated kairomone, possibly as part of a pheromonal blend. Several volatile compounds released by the skin act in this way and have been identified in alarm and sexual blends of kissing bugs (Bernier et al., 2000; Manrique et al., 2006; Pontes et al., 2008). These findings demonstrate that at least some haematophagous insects seem to make parsimonious use of their sensory system and the information provided. This has several implications. The first and most obvious is the efficient use of sensory organs, for which sensitivity to the same cues may help to identify and locate different resources. The second is that a given cue (e.g. a particular volatile compound) does not have an unequivocal biological meaning for an individual; rather, the information it provides is ambiguous until it is put into context via integration with cues targeting the same or different sensory systems. The third major implication is that a given cue involved in different behaviours may be associated with quite different responses; for instance, signals that incite attraction when the source is identified as a potential host may cause a repellent response if integrated as part of an alarm signal. So, the function of at least some cues is determined by the context in which they are detected, rather than being associated with one particular biological response. 9.2

PRACTICAL CONSEQUENCES

Sensory parsimony has important practical consequences for attracting and capturing haematophagous insects, given that attractants do not act in an additive fashion. Thus, combining different attractive compounds does not

36

CLAUDIO R. LAZZARI

necessarily increase the chances of finding a better lure. Indeed, different elements need to be combined in specific ways to make a bait biologically effective, in terms of not only the components used but also their specific proportions in the mixture (e.g. Barrozo and Lazzari, 2004a). Mixing hostassociated cues with assembling or sexual pheromones may therefore not have the expected effect, simply because the combination does not resemble a particular source of interest for the insect, despite individual components being potential attractants. Similarly, capture devices may prove ineffective if based on exploiting a behavioural characteristic that is linked to a specific context. This point was well illustrated by a trap designed to exploit the tendency of kissing bugs to fall down onto their hosts from walls or ceilings (Guerenstein et al., 1995). The trap was associated to a culture of baker yeast as lure. The bugs only jumped inside the trap (they could not climb away) when odours came from below. If the odours just diffused freely (i.e. not reaching the insect from below), the insects were attracted, but the jumping down behaviour was not evoked. This same effect – insects being attracted but not displaying jumping behaviour – was observed when assembly pheromones from faeces were used as lure. This variable jumping down behaviour was interpreted as being specifically associated to host-seeking when the insect walks on the roof and perceives events below it, but not when walking on the ground or when the attractant is perceived in a different context. Aggregation pheromones also include some volatile components emitted by vertebrate hosts, but are mostly associated with insect refuges, and jumping down into places of refuge is not a usual behaviour (Guerenstein et al., 1995; Lorenzo and Lazzari, 1996)

10

State-dependency of host-seeking behaviour

The response of blood-sucking insects does not only depend on external signals and their interactions outside (i.e. physical interaction) and inside the insect (multimodal convergence). Various signals originating from within the insect also play major roles in determining if, how and when a given behaviour will be evoked in the presence of environmental stimuli. These internal signals may exist in different forms (i.e. nervous or endocrine signals) and involve different physiological processes, which modulate the behaviour of the insect. The most widely recognized physiological determinants affecting the response to host signals include nutritional state (or feeding condition), moulting, reproductive state (mating and oviposition) and the circadian cycle. These factors not only determine whether or not insects respond to a given signal, but also how they respond. As we will discuss below, the identical stimulus may be attractive our repellent depending on the physiological state of the insect. Before analysing in detail how endogenous factors modulate the response to the presence of a host, we will briefly discuss the biological function or adaptive


37

value of such modulation. These factors minimize the costs and risks associated with feeding on blood, by making the insect feed only when necessary. Ideally, the insect should feed quickly, on an inactive host and only when the energetic expenditure for contacting the host is at a minimum. The insect therefore needs to be mature enough (or their mouthparts sufficiently developed) to pierce the host skin and feed during the host’s resting period at a time when all mature eggs have been laid and no extra weight due to stored food is being carried. The insect thereby minimizes the number of contacts with the host and the associated risks. 10.1

THE TEMPORAL MODULATION OF THE RESPONSE TO ODOURS

Biological rhythms are expressed ubiquitously in almost every organism as temporal oscillations in the course of biochemical, physiological and behavioural processes, from the cellular to the population level and even in multitrophic interactions. The host–vector–parasite interaction offers an excellent example of the adaptive value of biological rhythms (Pittendrigh, 1974; Aschoff, 1989; Barrozo et al., 2004b). In particular, most haematophagous insects synchronize their daily activity to seek food during host resting period, thus during a period when the host is less able to defend itself (Barrozo et al., 2004b). Haematophagous insects do not necessarily search for food for the duration of their activity period every day. Actively searching for something does of course require some activity, but the inverse is not true. Even in insects in which their physiological state promotes feeding, activity is devoted to a variety of behaviours, not only to obtaining food. For example, in insects with extended or bimodal activity periods every day, food-searching activity may be limited to just a narrow window of time. Many examples demonstrate the temporal control of insect responsiveness to external stimuli. Examples include the daily modulation of olfactory responses (Saunders et al., 2002) and the modulation of chemoreceptor sensitivity by the circadian system (van der Goes van Naters et al., 1998; Krishnan et al., 1999; Page and Koelling, 2003; Zhou et al., 2004; Merlin et al., 2007; Saifullah and Page, 2009). Paradoxically, in some cases, olfactory sensitivity seems to be maximal during the insect resting period. Various explanations have been proposed, but further studies are needed to fully understand the biological relevance of this phenomenon. Blood-sucking insects are no exception, both their sensory and behavioural responsiveness to odours being under the control of the circadian system (Van der Goes van Naters et al., 1998; Barrozo et al., 2004a). It remains unclear how the responsiveness to specific odours in particular behavioural contexts is controlled within certain temporal windows. The general hypothesis, based on studies carried out in Drosophila and cockroaches, suggests that the overall sensitivity of the antennae is modulated for all chemical stimuli at the same time

38

CLAUDIO R. LAZZARI

(Krishnan et al., 1999; Page and Koelling, 2003; Zhou et al., 2005; Merlin et al., 2007). However, this method of modulation may not be the most beneficial for insects that exhibit bimodal activity and perform different tasks during different temporal windows, requiring different sensitivities for the different activity periods of the day. This has been addressed recently in triatomine bugs, which display bimodal daily activity, looking for a host at dusk and for a refuge at dawn, both activities guided by different chemical cues (Lazzari, 1992; Lorenzo and Lazzari, 1998). Experimental analysis of the temporal modulation of the bugs’ responsiveness to carbon dioxide, a host-associated cue, and to refugeassociated chemical cues in aggregation pheromones showed that these insects respond to carbon dioxide only during the first hours of the night and are attracted by the pheromones only at sunrise (Barrozo et al., 2004a; Bodin et al., 2008). Further analysis of the chronobiological basis of this differential modulation showed that an endogenous circadian system controls responsiveness to carbon dioxide, whereas the responsiveness to aggregation pheromones depends on environmental signals (Bodin et al., 2008). This shows that endogenous clocks and exogenous cycles may act together to adaptively modulate sensory responses. 10.2

MATURATION AND RESPONSIVENESS

The moulting process involves profound changes in the insect’s body, which persist beyond ecdysis and the appearance of a new instar. Indeed, completion of the changes in the insect morphology, physiology and behaviour of the insect takes a certain amount of time. In many teneral insects, the sclerotization of their exoskeleton may remain incomplete for a time following ecdysis. In other insects, the development of certain sensory organs continues for a few days after the ecdysis (Insausti and Lazzari, 2000b). This post-ecdysis delay in completing the development of certain organs is usually associated with a period of maturation of behaviours linked to the use of these structures. Haematophagous insects (both holometabolous and hemimetabolous groups) show a delay before they start responding to host-associated signals. In holometabolous blood-sucking insects, moulting and other activities occurring after the emergence of the adult, such as reproduction, affect feeding behaviour. Blood meals taken before a certain degree of ovarian maturation do not increase the reproductive success of these insects, but visiting the host increases the insect’s chances of being damaged or killed. In mosquitoes, there is period after adult emergence during which insects do not seek for a host. Blood-feeding behaviour begins between 24 and 96 h after a female mosquito emerges (Seaton and Lumsden, 1941; Bishop and Gilchrist, 1946; Laarman, 1955; Bowen and Davis, 1989). A similar period of maturation appears to be required before the peripheral sensory organs are fully responsive (Davis, 1984a). The activation of lactic acid-sensitive neurons in newly emerged virgin female A. aegypti is dependent on age and correlated to the


39

development of host-seeking behaviour (Davis, 1984a). The receptors for lactic acid develop faster (12 h) in Aedes atropalpus. The absence of attractive behavioural response to host components between the time of emergence and the end of the first gonotrophic cycle in this species cannot be attributed to the delay in the development of the peripheral sensory system only (Bowen et al., 1994a,b). The hemimetabolous insect R. prolixus is an obligate blood feeder throughout its lifetime. This insect has thus been used to follow post-ecdysis development of the response to host-associated cues in larvae, that is without the effects of physiological processes associated with reproduction (Bodin et al., 2009a). The development of the insect responses to carbon dioxide and heat, and their motivation to feed, is age-dependent; all three types of response reach a plateau more or less at the same time, but show different development patterns and rates. The response to CO2 does not increase gradually with age, but is an all-ornothing response. In other words, insects remain indifferent to this cue during the first week post-ecdysis and are highly attracted afterwards. In contrast, the response to heat progresses gradually, with the responsiveness to heat increasing steadily for about 10 days before reaching a maximum level. These observations suggest that each of the three responses – attraction to CO2 or heat, and feeding – follows its own pattern of maturation. This is probably related to the roles of different physiological processes, such as the maturation of sensory organs, sclerotization of the mouthparts, modulation of the biochemical machinery associated with food intake, handling and digestion, and others that still need to be elucidated. Before the behavioural response to CO2 in R. prolixus becomes properly established, bugs display a bimodal response, either approaching the odour source or walking away from it. This phenomenon has also been observed under similar stimulation conditions, when insects were exposed to CO2 at a time of the day when they normally search for a refuge, rather than for food (Bodin et al., 2008). So, three different and successive phases have been observed for the post-ecdysis development of the response to CO2 in R. prolixus: indifference, bimodal response and attraction. This particular pattern of development suggests that maturation processes also occur at the central level. During the second phase responses, the insects seem to perceive the signal, but their behaviour is ambiguous. In fed insects, a new period of repulsion to CO2 appears 2 days before completing ecdysis (Bodin et al., 2009b). 10.3

THE MODULATION OF HOST-SEEKING ACTIVITY BY REPRODUCTION

The relationship between reproduction and feeding in mosquitoes, underlying the generation of gonotrophic cycles, is a classical topic in the study of insect physiology. It has been mostly analysed in female mosquitoes, in which oogenesis depends on a blood meal, the relationship between blood feeding and reproduction, and endocrine factors (Klowden, 1997). Once the gonadotrophic

40

CLAUDIO R. LAZZARI

cycle has started, host-seeking activity is inhibited until the oocytes have matured or oviposition has occurred, depending on the species (Klowden and Briegel, 1994). Two processes mediate this inhibition. The first depends on the stimulation of abdominal mechanoreceptors due to abdominal distension during feeding (Klowden and Lea, 1978, 1979a; Klowden, 1990). The second process occurs after the first one and is induced by a humoral response through a complex mechanism involving signals from the ovaries, fat body and neurosecretory cells of the insect (Klowden and Lea, 1979b; Klowden, 1981; Klowden et al., 1987; Brown et al., 1994; Takken et al., 2001). The absence of host-seeking behaviour in fed mosquitoes may be due to the inhibition of the response of ORN sensitive to lactic acid by humoral factors. This inhibitory effect would thus act on peripheral sensory neurons, rather acting centrally (Klowden and Lea, 1979a,b; Davis, 1984b; Davis et al., 1987), resulting in a 10-fold reduction in the sensitivity of olfactory neurons to lactic acid (Davis, 1984b). Host-seeking behaviour remains inhibited in A. aegypti until oviposition (Klowden, 1981, 1990) and, in A. gambiae, until oocyte maturation (Takken et al., 2001). Gonotrophic cycles have not been defined for triatomine bugs. However, after feeding, modulation of the responsiveness to host cues differs between males and females. Whereas males remain indifferent to CO2 for at least 20 days after feeding, females experience a phase of repulsion to the same stimulus (Bodin et al., 2009a). Thus, even though neither sex shows an attraction response, females do not remain indifferent to the stimulus, suggesting that the mechanism does not operate at the periphery, but acts centrally (Bodin et al., 2009b). It also suggests that egg production in females is probably involved in modulating the response to host-associated cues in these insects. 10.4

FEEDING CONDITIONS AND HOST SEARCHING

As discussed in the previous section, ovarian activity is involved in the modulation of haematophagous insect responses to host-associated cues. However, the link between feeding and reproduction prevents independent analyses of these two types of process. In a recent study on R. prolixus (Bodin et al., 2009b) the responses to CO2 and to heat and the motivation to feed were analysed in male and female larvae, the use of larvae allowing separation of these responses from the effects of reproductive processes. R. prolixus larvae do not respond to carbon dioxide for the first 2 days that follow feeding. On the third day, they start responding to the direction of the CO2 laden air-current, but by orientating themselves downwind. This tendency to be repelled by CO2 is gradually reduced, and insects become indifferent again for several days, with another period of repulsion appearing before ecdysis (Bodin et al., 2009b). Host-associate odours exert allomonal effects on mosquitoes (Mukabana et al., 2004; Logan et al., 2008). Repulsion to an otherwise attractive odour only seems to have been observed for kissing bugs.


41

Additionally, whereas the allomonal effects observed in mosquitoes are associated with specific volatile compounds, it is the physiological state of the insect that seems to determine the type of response observed in bugs. Carbon dioxide exerts only kairomonal effects, such as repulsion to CO2 when the animal is full of undigested blood, constituting an adaptive response that benefits the insect and not its host. The neural mechanism underlying this modification of the response to host stimuli takes place centrally in R. prolixus, but involves the modification of peripheral sensitivity in mosquitoes (Davis, 1984b). The fact that the insects do not stop responding, but rather modify their response (Bodin et al., 2009b), suggests that they perceive the stimulus normally, with the central integration of exogenous and endogenous information determining the direction of oriented behaviour in a state-dependent way. It should be noted that the available data on both mosquitoes and bloodsucking bugs only concern responses to single chemical cues. We thus cannot assume these observations to be true for the general modulation of any hostassociated cue or conclude that modulation takes place at the periphery for one group of insects and in the central nervous system for another group of insects. Both mechanisms probably occur together in both groups of insects for a better control of host-seeking behaviour. In both mosquitoes and bugs, humoral factors, rather than nervous signals, seem to be involved (Klowden, 1995; Bodin et al., 2009b). Indeed, the transfer of haemolymph from recently fed individuals to starved individuals inhibits the response to host-associated signals in the receivers.

11

Why some people are bitten more than others?

This is a central question in the epidemiology of diseases transmitted by arthropods. It has been addressed using various approaches, including evolutionary and ecological methods and physiological studies of insect sensory systems (Kelly and Thompson, 2000; Kelly, 2001; Mukabana et al., 2002, 2004). Many inductive studies, based on statistical surveys, have also been carried out. The differences in host attractiveness have been related to age, sex, pregnancy, hygiene, blood group, physiological state and others. However, the major factors and mechanisms involved remain unclear: predictions have not been tested and experimental studies have mostly been based on small numbers of human subjects. Indeed, the insects are generally considered as the experimental unit, despite the fact that the question concerns human subjects. So, independency is kept for the insects, but not for the source the odours, introducing pseudoreplicative bias (Hurlbert, 1984; Ramirez et al., 2000). However, despite these limitations, these studies have introduced the idea that differential repulsion rather than attractiveness may underlie the differences observed. Recently, Logan et al. (2008)

42

CLAUDIO R. LAZZARI

identified 33 physiological active people from a group of volunteers exhibiting variable degrees of attractiveness to A. aegypti. Chemical analyses, combined with electrophysiological and behavioural tests, revealed allomonal effects of some volatile compounds. This is consistent with the notion that unattractiveness of individuals may result from a repellent, or attractant ‘‘masking’’ mechanism.

12

Learning and memory

Some insect species exhibit highly complex cognitive abilities. Honey bees, for instance, demonstrate complex non-elementary forms of learning, being able to learn concepts such as ‘‘symmetry,’’ ‘‘asymmetry,’’ ‘‘sameness’’ or ‘‘difference’’ (Giurfa et al., 1996, 2001; Giurfa, 2003). Their evolutionary history and their exploitation of the environment have led them to acquire capacities that were until only recently believed to be exclusive to superior vertebrates. Bloodsucking insects exhibit various behaviours that have also been subject to particular selective pressures, notably associated with obtaining blood from vertebrate hosts, a food source much more reactive and unpredictable than flowers. Given that many insects modify their behaviour adaptively through experience, and given the potential impact that vector learning and memory could have on the transmission of parasites, the cognitive abilities of blood-sucking insects are of particular interest (McCall et al., 2001; McCall and Eaton, 2001; McCall and Kelly, 2002; Bouyer et al., 2007). Currently available data on the learning capacity of these insects, however, are inconclusive for most haematophagous insects. The title of a relatively recent article by Alonso et al. (2003) seems illustrates the lack of clarity on this topic: ‘‘Are vectors able to learn about their hosts?’’ The chapter describes a series of unsuccessful attempts at inducing olfactory conditioning in mosquitoes. More recently, a critical analysis of the available evidence on learning and memory in mosquitoes demonstrated the lack of conclusive findings in most of the published work analysed, with a few exceptions (Alonso and Schuck-Paim, 2006, including Mwandawiro et al., 2000; McCall and Eaton, 2001). In most of these studies, the experiments were not designed to test learning and thus lacked adequate controls; in other studies, alternative mechanisms appear to be more likely explanations than learning.

13

Repellents, how they work

A repellent has been defined as a chemical that, in the vapour phase, prevents an insect from reaching a target to which it would otherwise be attracted (Browne, 1977). Among them, DEET ([N-N]-diethyl-mtoluamide) is widely used around the world as a repellent for mosquitoes and other biting insects. A number of other compounds with a similar activity have been identified, but DEET


43

remains the gold standard. However, and despite a large number of studies (Debboun et al., 2007), a plausible and evidence-based mechanism for DEET’s action remains to be elucidated (Pickett et al., 2008). Two recent studies have shed some light on how DEET works (Syed and Leal, 2008; Ditzen et al., 2008), demonstrating two major effects of DEET. Firstly, DEET seems to act as a ‘‘fixative’’ of molecules in the main mosquito attractant, 1-octen-3-ol, reducing the stimulation of ORN and giving the impression of repellency. Secondly, it stimulates olfactory neurons that normally detect linalool and other plant odours, inducing true behavioural repellency to chemical and physical hostassociated cues (Syed and Leal, 2008; Pickett et al., 2008) (Fig. 13). These findings give rise to additional questions, such as how DEET repels insects in the absence of chemical attractants, for example kissing bugs, which have no biological relationship with plants (Sfara et al., 2006, 2008).

14

Conclusions and perspectives

The study of how blood-sucking insects locate their host is largely biased towards chemoreception-related aspects and the identification of biologically significant compounds. Other aspects, such as the neurobiological and molecular bases of olfaction, are also being intensively studied. This bias is justified for a number of reasons: (1) odours are able to attract insects from relatively long distances; (2) chemoreception is a well studied subject in insect sensory physiology, with data from Drosophila, moths, honeybees and other classical models being easily transferred to understand odour reception in haematophagous;

1-octen-3ol ONR

DEET/terpenoids ONR

1-octen-3-ol

Activation

No-activation

Attraction

1-octen-3-ol + DEET

No-activation

Activation

Repellency

DEET + 1-octen-3-ol

Activation

Activation

Repellency

Behaviour

FIG. 13 Sensory and behavioural effects of DEET on mosquitoes. DEET interacts with some attractants, reducing or abolishing their effect on specific olfactory neurons (ORNs). The mechanism underlying such interactions remains unknown. Furthermore, DEET specifically stimulates ORNs sensitive to plant-derived terpenoids. The combination of these two effects abolishes attraction and causes repellency (Syed and Leal, 2008; Pickett et al., 2008; Ditzen et al., 2008).

44

CLAUDIO R. LAZZARI

(3) mosquitoes, the most important group of disease vector insects, seem to rely mainly on host-associated odours to find food; (4) the potential manipulation of the behaviour of disease vectors using chemical baits or by blocking odour perception provides a clear objective, despite the difficulties in applying this approach at levels beyond the individual. How sensory systems other than olfaction are used to locate a food source is poorly understood. As described above, olfaction plays a key role in the feeding behaviour of mosquitoes, whereas visual and thermal cues are relatively less important in host-seeking. For many haematophagous insects, however, other sensory signals play a major role, for example vision in tsetse flies or heat for kissing bugs, bed bugs and lice (Wigglesworth, 1941; Brady, 1972b; Flores and Lazzari, 1996; Anderson et al., 2009). A large amount of data has been, and continues to be, collected on the individual roles of attractive stimuli. Hosts provide an abundant source of a complex mixture of stimuli. The analysis of the individual components of these mixtures seems indispensable in understanding how host-insect interactions work. However, in natural conditions, stimuli are presented to their receivers as part of a multimodal complex and are subjected to various interactions. Indeed, the integration of these signals via unimodal, multimodal and exogenous/endogenous mechanisms is involved in modifying not only response thresholds, but also the way an insect reacts at a given moment. Kissing bugs demonstrate the use of all these different forms of integration. Compounds such as lactic acid and short-chain fatty acids have no effect on the behaviour of these bugs when tested alone, but their combination in the presence of sub-threshold amounts of carbon dioxide show the same level of attractiveness as a live host (Barrozo and Lazzari, 2004a). Multimodal mechanisms are used to integrate information from heat and water vapour (Barrozo et al., 2003), from mechanical and chemical cues (Barrozo and Lazzari, 2006) and from thermal and proprioceptive inputs (Flores, 2001; this chapter). Endogenous circadian clocks and exogenous temporal cues modulate responsiveness to odours associated with specific biological contexts (Barrozo et al., 2004a; Bodin et al., 2008). Finally, physiological state-dependency (i.e. moulting, reproduction and nutritional state) modulates their response to chemical and thermal stimuli, as well as their motivation to have a blood meal (Bodin et al., 2009a,b). Given such modifications of the bugs’ responses, the separate analysis of individual components, such as particular volatile compounds or signals, is only the first step in understanding their biological role. It should be noted that these multilevel interactions (i.e. unimodal, multimodal and endogenous modulation) need to be considered not only to understand how blends of volatile compounds or multimodal cues work together or how their biological effectiveness is modulated. Their consideration is also crucial in the identification of individual components in the laboratory, particularly in behavioural bio-essays. The classical example is the bimodal interaction of mechanical and chemical cues in odour-triggered anemotaxis, with neither


45

the wind nor the odour evoking a response when presented separately. In other cases, the inverse can be true, for example when the presence of wind disrupts the chemical gradient created by an odour, preventing chemotaxis. In both cases, the same modalities are involved but opposite outcomes can be achieved depending on the experimental design used in the bio-essay. The choice of experimental design may be associated with a specific objective, for example testing whether a given volatile compound attracts insects located some distance away. In this case, if the candidate compound is found to be ineffective, the result will be negative in relative terms, concerning the original hypothesis tested. However, to exclude any behavioural response to a given cue, it needs to be tested under different contexts. If such experiments are not carried out, or the biological role of a particular signal is not the main focus of the study, caution should be taken in the interpretation of negative results. False-positive results mainly arise due to experimental problems and can be controlled with adequate control experiments and by avoiding pseudoreplication (Hurlbert, 1984; Ramirez et al., 2000). Similar considerations need to be taken into account when interpreting statedependency or endogenous effects. A recent analysis of a representative number of papers on the chemical ecology of blood-sucking insects showed that temporal effects (i.e. the time of the day when the experiments were conducted) were rarely taken into account (Lazzari et al., 2004). The reproducibility of such experiments is thus compromised, as are the validity of negative results or the reliability of response thresholds measured. The role of circadian clocks, exogenous temporal signals and the state-dependency of behavioural responses seem to be receiving increasing interest in studies. The same rigour applied to the statistical treatment of data should be applied to the standardization of the conditions of the subjects tested (Takken, 2005). This does not only require the synchronization of individuals and their physiological state: the relevant temporal and physiological context for the response studied, or signal tested must also be taken into account. Indeed, synchronization of the insects studied is not helpful if the experiments are carried out during a daytime period that does not correspond to the temporal window normally associated with the biological signal. This is also true for the physiological state of the insects. Thus, rather than homogeneity, it is the general context of the experiment that determines whether meaningful results may be obtained. The study of the cognitive abilities of haematophagous insects has important fundamental and epidemiological implications. It is difficult to conduct experiments on learning in a natural context due to limitations in the adequate control of many of the variables. Experimental conditions can be better controlled in the laboratory, but some common protocols, such as those used to establish associations between stimuli and positive reinforcements are not as easily tested in haematophagous insects as they are in other insects. In particular, in contrast to insects that can be rewarded with a drop of sugar solution, haematophagous insects need to bite to obtain blood. Artificial feeders may be used in cases

46

CLAUDIO R. LAZZARI

where the insects pierce a membrane to obtain blood (e.g. Nuń˜ez and Lazzari, 1990). These devices can be relatively simple, but the reward offered needs to be carefully controlled, to ensure that a known amount of reward is given and, more importantly, to control the motivational state of insects subjected to several consecutive trial sessions. Insects should thus be allowed to take only small volumes of blood, without abruptly interrupting feeding, which could otherwise cause an aversive effect. The study of non-associative forms of learning, such as habituation, does not require rewards. The insect is repeatedly exposed to a stimulus (unconditional stimulus, UCS) able to evoke an automatic response (unconditional response, UCR). The repeated stimulation causes the response to decrease gradually to no response. Such experiments require an in-depth knowledge of the behaviour of the species being studied and an appropriate set of UCS–UCRs. For instance, kissing bugs extend their proboscis when exposed to an object with a temperature close to that of a potential host. Thus, the PER is the UCR, which is associated to thermal stimulation, the UCS (Fig. 14). The PER can be habituated or rapidly extinguished by associating with negative reinforcement (Lazzari et al., unpublished). Thus, most efforts are devoted to the study of learning and memory in mosquitoes, which is perfectly justified by their paramount importance in health. However, the study of other haematophagous insects allows the use of training protocols that have been mostly validated in classical models, such as Drosophila or honeybees. More attention should be paid to similarities between most haematophagous insects and the most suitable model species, in terms of factors such as size, particular behaviour or type of locomotion, should be used for each analysis. As summarized in Fig. 15, host-seeking by blood-sucking insects is a complex behaviour, based on the exploitation of multimodal sensory cues emitted

25 °C

35 °C

FIG. 14 Proboscis extension response (PER) in blood-sucking bugs. Triatomine bugs perform PER when exposed to an object with a temperature similar to that of a warmblooded vertebrate host. PER is an unconditional response evoked by heat, an unconditional stimulus. The response habituates and can be inhibited by aversive conditioning (Lazzari et al., 2009).


47

Mechanical contact aggregation signals Resting Visual cues Odours CO2 Heat Water vapour

Host leaving

Activation Stretch receptors

Circadian clocks Appetitive search

Feeding Feeding status reproduction moult

Phagostimulants Biting

Wind Host detection

Experience Heat Host finding

Visual cues Odours CO2 Heat Water vapour

FIG. 15 Factors affecting the haematophagous act in a generalized blood-sucking insect. Outside the circle of events, factors exogenous to the insect are represented, and inside it, the influences of signals originating inside the insect’s body. The insect activates at a particular moment under the influence of internal clocks. If its physiological state is the appropriate (starved, no reproductive activity) it starts searching for appetitive signals using the wind direction as reference. The presence of host-associated directional cues allows the insect detecting a host and to locate it. The sensitivity and responsiveness to this signals is modulated by endogenous factors (state-dependency). The individual experience may affect the final approach mediating host selection if more than one is present. Once on the host, the insect should chose the most appropriate place to bite, guided probably by thermal gradients over the skin indicating the degree of irrigation and proximity of blood vessels. Once a potential food is contacted, chemoreceptors located in the alimentary channel sense to presence of phagostimulants. The insect feeds until mechanical proprioreceptors indicate that the digestive tube is full of blood. Mouthparts detach then from the host skin and the gorged insect goes moves away from its host guided probably by the same cues involved in the approach. Finally, it returns to a refuge or a protected place for resting thanks to specific cues, assembling pheromones and physical contact with the substrate (thigmotaxis).

by vertebrate hosts. Additionally, the behavioural response of haematophagous insects to host signals is dependent on the motivational state of the insects, which results from the interplay of diverse physiological signals. Specific links between particular external cues and the physiological state of the insects may depend on individual experience (e.g. Pompilio et al., 2006), adding a subjective element. Further studies should now address how these elements interact, to gain a better understanding of the real abilities of haematophagous insects and to identify novel targets for controlling them.

48

CLAUDIO R. LAZZARI

Acknowledgements The author wish to thank Stephen Simpson and Je´roˆme Casas for their kind invitation to contribute this chapter and for their valuable comments on the manuscript, and the University of Tours, the CNRS and the ANR (France) for their support. References Acree, F. Jr, Turner, R. B., Gouck, H. K., Beroza, M. and Smith, N. (1968). L-Lactic acid: a mosquito attractant isolated from humans. Science 161, 1346–1347. Alonso, W. J. and Schuck-Paim, C. (2006). The ‘ghosts’ that pester studies on learning in mosquitoes: guidelines to chase them off. Med. Vet. Entomol. 20, 157–165. Alonso, W. J., Wyatt, T. D. and Kelly, D. W. (2003). Are vectors able to learn about their hosts? A case study with Aedes aegypti mosquitoes. Mem. Inst. Oswaldo Cruz 98, 665–672. Altner, H. and Loftus, R. (1985). Ultrastructure and function of insect thermoreceptors and hygroreceptors. Annu. Rev. Entomol. 30, 273–295. Altner, H., Routil, C. and Loftus, R. (1981). The structure of bimodal chemo-, thermo-, and hygroreceptive sensilla on the antenna of Locusta migratoria. Cell Tissue Res. 215, 289–308. Anderson, J. F., Ferrandino, F. J., McKnight, S., Nolen, J. and Miller, J. (2009). A carbon dioxide, heat and chemical lure trap for the bedbug, Cimex lectularius. Med. Vet. Entomol. 23, 99–105. Anton, S. (1996). Central olfactory pathways in mosquitoes and other insects. Ciba Found. Symp. 200, 184–192. Anton, S. and Rospars, J. P. (2004). Quantitative analysis of olfactory receptor neuron projections in the antennal lobe of the malaria mosquito, Anopheles gambiae. J. Comp. Neurol. 475, 315–326. Anton, S., van Loon, J. J. A., Meijerink, J., Smid, H. M., Takken, W. and Rospars, J. P. (2003). Central projections of olfactory receptor neurons from single antennal and palpal sensilla in mosquitoes. Arthropod Struct. Dev. 32, 319–327. Anwyl, R. (1972). The structure and properties of an abdominal stretch receptor in Rhodnius prolixus. J. Insect Physiol. 18, 2143–2145. Aschoff, J. (1989). Temporal orientation: circadian clocks in animals and humans. Anim. Behav. 37, 881–896. Barrett, M. and Orchard, I. (1990). Serotonin-induced elevation of cAMP levels in the epidermis of the bloodsucking bug, Rhodnius prolixus. J. Insect. Physiol. 36(625– 627), 629–633. Barrozo, R. B. and Lazzari, C. R. (2004a). Orientation behaviour of the blood-sucking bug Triatoma infestans to short-chain fatty acids: synergistic effect of L-lactic acid and carbon dioxide. Chem. Senses 29, 833–841. Barrozo, R. B. and Lazzari, C. R. (2004b). The response of the blood-sucking bug Triatoma infestans to carbon dioxide and other host odours. Chem. Senses 29, 319–329. Barrozo, R. B. and Lazzari, C. R. (2006). Orientation response of haematophagous bugs to CO2: the effect of the temporal structure of the stimulus. J. Comp. Physiol. A 192, 827–831. Barrozo, R. B., Manrique, G. and Lazzari, C. R. (2003). The role of water vapour in the orientation behaviour of the blood-sucking bug Triatoma infestans (Hemiptera, Reduviidae). J. Insect Physiol. 49, 315–321.


49

Barrozo, R. B., Minoli, S. A. and Lazzari, C. R. (2004a). Circadian rhythm of behavioural responsiveness to carbon dioxide in the blood-sucking bug Triatoma infestans (Heteroptera: Reduviidae). J. Insect Physiol. 50, 249–254. Barrozo, R. B., Schilman, P. E., Minoli, S. A. and Lazzari, C. R. (2004b). Daily rhythms in disease-vector insects. Biol. Rhythm Res. 35, 79–92. Barrozo, R. B., Couton, L., Lazzari, C. R., Insausti, T. C., Minoli, S. A., Fresquet, N., Rospars, J. P. and Anton, S. (2009). Antennal pathways in the central nervous system of a blood-sucking bug, Rhodnius prolixus. Arthropod Struct. Dev. 38, 101–110. Barth, R. (1952). Estudos anatoˆmicos e histolo´gicos sobre a subfamı´lia Triatominae (Heteroptera, Reduviidae) I. A cabeça do Triatoma infestans. Mem. Inst. Oswaldo Cruz 50, 156–196. Barth, R. (1976). Estudos anatoˆmicos e histolo´gicos sobre a subfamı´lia Triatominae (Heteroptera, Reduviidae) Ce´rebro e seus nervos de Triatoma infestans. Mem. Inst. Oswaldo Cruz 74, 153–176. Belzer, W. R. (1979). Abdominal stretch receptors in the regulation of protein ingestion by the black blowfly, Phormia regina. Physiol. Entomol. 4, 7–14. Bennet-Clark, H. C. (1963). The control of the meal size in the blood sucking bug, Rhodnius prolixus. J. Exp. Biol. 40, 741–750. Bernard, J. (1974). E´tude e´lectrophysiologique des rećepteurs implique´s dans l’orientation vers l’hoˆte et dans l’acte he´matophage chez un He´mipte`re: Triatoma infestans. Ph.D. Thesis, Universite´ de Rennes, pp. 285. Bernier, U. R., Kline, D. L., Barnard, D. R., Schreck, C. E. and Yost, R. A. (2000). Analysis of human skin emanations by gas chromatography/mass spectrometry. 2. Identification of volatile compounds that are candidate attractants for the yellow fever mosquito (Aedes aegypti). Anal. Chem. 72, 747–756. Bernier, U. R., Kline, D. L., Posey, K. H., Booth, M. M., Yost, R. A. and Barnard, D. R. (2003). Synergistic attraction of Aedes aegypti (L.) to binary blends of L-lactic acid and acetone, dichloromethane, or dimethyl disulfide. J. Med. Entomol. 40, 653–656. Bishop, A. and Gilchrist, B. M. (1946). Experiments on the feeding of Aedes aegypti through animal membranes with a view to applying this method to the chemotherapy of malaria. Parasitology 37, 85–100. Bodin, A., Barrozo, R. B., Couton, L. and Lazzari, C. R. (2008). Temporal modulation and adaptive control of the behavioural response to odours in Rhodnius prolixus. J. Insect Physiol. 54, 1343–1348. Bodin, A., Vinauger, C. and Lazzari, C. R. (2009a). State-dependency of host-seeking in Rhodnius prolixus: The post-ecdysis time. J. Insect Physiol. 55, 574–579. Bodin, A., Vinauger, C. and Lazzari, C. R. (2009b). Behavioural and physiological state dependency of host seeking in the blood-sucking insect Rhodnius prolixus. J. Exp. Biol. 212, 2386–2393. Bosch, O. J., Geier, M. and Boeckh, J. (2000). Contribution of fatty acids to olfactory host finding of female Aedes aegypti. Chem. Senses 25, 323–330. Bouyer, J., Pruvot, M., Bengaly, Z., Guerin, P. M. and Lancelot, R. (2007). Learning influences host choice in tsetse. Biology Letters 3, 113–116. Bowen, M. F. (1992). Patterns of sugar feeding in diapausing and nondiapausing Culex pipiens (Diptera: Culicidae) females. J. Med. Entomol. 29, 843–849. Bowen, M. F. and Davis, E. E. (1989). The effects of allatectomy and juvenile hormone replacement on the development of host-seeking behaviour and lactic acid receptor sensitivity in the mosquito Aedes aegypti. Med. Vet. Entomol. 3, 53–60. Bowen, M. F., Davis, E. E., Romo, J. and Haggart, D. (1994a). Lactic-acid sensitive receptors in the autogenous mosquito Aedes atropalpus. J. Insect Physiol. 40, 611–615.

50

CLAUDIO R. LAZZARI

Bowen, M. F., Davis, E. E., Haggart, D. and Romo, J. (1994b). Host-seeking behavior in the autogenous mosquito Aedes atropalpus. J. Insect Physiol. 40, 511–517. Brady, J. (1972a). Visual stimulus velocity and ‘‘following’’ response of tsetse flies. Trans. R. Soc. Trop. Med. Hyg. 66, 313. Brady, J. (1972b). Visual responsiveness of tsetse fly Glossina morsitans Westw. (Glossinidae) to moving objects: the effects of hunger, sex, host odor and stimulus characteristics. Bull. Entomol. Res. 62, 257–279. Brady, J. (1972c). Spontaneous, circadian components of tsetse fly activity. J. Insect Physiol. 18, 471–484. Briscoe, A. D. and Chittka, L. (2001). The evolution of color vision in insects. Annu. Rev. Entomol. 46, 471–510. Brown, M. R., Klowden, M. J., Crim, J. W., Young, L., Shrouder, L. A. and Lea, A. O. (1994). Endogenous regulation of mosquito host-seeking behavior by a neuropeptide. J. Insect Physiol. 40, 399–406. Browne, L. B. (1977). Host-related responses and their supression: some behavioural considerations. In: Chemical control of insect behavior: theory and application, (eds Shorey, H. H. and McKelvey, J. J. Jr.), pp. 117–127. John Wiley & Sons, New York. Bullock, T. H. (1990). Goals of neuroethology. Bioscience 40, 244–248. Burkett, D. A., Lee, W. J., Lee, K. W., Kim, H. C., Lee, H. I., Lee, J. S., Shin, E. H., Wirtz, R. A., Cho, H. W., Claborn, D. M., Coleman, R. E. and Klein, T. A. (2001). Light, carbon dioxide, and octenol-baited mosquito trap and host-seeking activity evaluations for mosquitoes in a malarious area of the Republic of Korea. J. Am. Mosq. Control Assoc. 17, 196–205. Campbell, A. L., Naik, R. R., Sowards, L. and Stone, M. O. (2002). Biological infrared imaging and sensing. Micron 33, 211–225. Carde´, R. T. (1996). Odour plumes and odour-mediated flight in insects. Ciba Found. Symp. 200, 54–66. Catala, S. (1996). Sensilla associated with the rostrum of eight species of triatominae. J. Morphol. 228, 195–201. Chapman, R. F. (1982). Chemoreception: the significance of receptor numbers. Adv. Insect Physiol. 16, 247–356. Chiang, R. G. and Davey, K. G. (1988). A novel receptor capable of monitoring applied pressure in the abdomen of an insect. Science 241, 1665–1667. Colvin, J., Brady, J. and Gibson, G. (1989). Visually-guided, upwind turning behavior of free-flying tsetse flies in odor-laden wind: a wind-tunnel study. Physiol. Entomol. 14, 31–39. Davis, E. E. (1984a). Development of lactic-acid receptor sensitivity and host-seeking behavior in newly emerged female Aedes aegypti mosquitos. J. Insect Physiol. 30, 211–215. Davis, E. E. (1984b). Regulation of sensitivity in the peripheral chemoreceptor systems for host-seeking behavior by a hemolymph-borne factor in Aedes aegypti. J. Insect Physiol. 30, 179–183. Davis, E. E., Haggart, D. A. and Bowen, M. F. (1987). Receptors mediating host-seeking behavior in mosquitos and their regulation by endogenous hormones. Insect Sci. Appl. 8, 637–641. De Jong, R. and Knols, B. G. (1995). Selection of biting sites on man by two malaria mosquito species. Experientia 51, 80–84. Debboun, M., Frances, S. P. and Strickman, D. (2007). Insect repellents: principles, methods, and uses. CRC Press, Boca Raton, FL. Diehl, P. A., Vlimant, M., Guerenstein, P. and Guerin, P. M. (2003). Ultrastructure and receptor cell responses of the antennal grooved peg sensilla of Triatoma infestans (Hemiptera: Reduviidae). Arthropod Struct. Dev. 31, 271–285.


51

Ditzen, M., Pellegrino, M. and Vosshall, L. B. (2008). Insect odorant receptors are molecular targets of the insect repellent DEET. Science 319, 1838–1842. Du, Y. J. and Millar, J. G. (1999). Electroantennogram and oviposition bioassay responses of Culex quinquefasciatus and Culex tarsalis (Diptera: Culicidae) to chemicals in odors from Bermuda grass infusions. J. Med. Entomol. 36, 158–166. Dusenbery, D. B. (1990). Upwind searching for an odor plume is sometimes optimal. J. Chem. Ecol. 16, 1971–1976. Dusenbery, D. B. (1992). Sensory Ecology. How Organisms Acquire and Respond to Information. W.H. Freeman, New York. Egelhaaf, M. (2006). The neural computation of visual motion information. In: Invertebrate Vision, (eds Warrant, E. J. and Nilsson, D. E.), pp. 399–461. Cambridge University Press, Cambridge. Eiras, A. E. and Jepson, P. C. (1991). Host location by Aedes aegypti (Diptera, Culicidae) – A wind-tunnel study of chemical cues. Bull. Entomol. Res. 81, 151–160. Eiras, A. E. and Jepson, P. C. (1994). Responses of female Aedes aegypti (Diptera, Culicidae) to host odors and convection currents using an olfactometer bioassay. Bull. Entomol. Res. 84, 207–211. Ewert, J.-P. (1980). Neuro-ethology. Springer-Verlag, Berlin. Ferreira, R. A., Lazzari, C. R., Lorenzo, M. G. and Pereira, M. H. (2007). Do haematophagous bugs assess skin surface temperature to detect blood vessels? PLoS ONE 2, e932. Flores, G. B. (2001). Las antenas y el sentido te´rmico de la vinchuca Triatoma infestans (Heteroptera: Reduviidae). Ph.D. Thesis, University of Buenos Aires, pp. 157. Flores, G. B. and Lazzari, C. R. (1996). The role of the antennae in Triatoma infestans: Orientation towards thermal sources. J. Insect Physiol. 42, 433–440. Fraenkel, G. S. and Gunn, D. L. (1961). The orientation of animals: kineses, taxes and compass reactions. Oxford University Press, London. Friend, W. G. and Smith, J. J. (1975). Feeding in Rhodnius prolixus: increasing sensitivity to ATP during prolonged food deprivation. J. Insect Physiol. 21, 1081–1084. Friend, W. G. and Smith, J. J. (1977). Factors affecting feeding by bloodsucking insects. Annu. Rev. Entomol. 22, 309–331. Frye, M. A. (2007). The neuromechanics of fly flight control. In: Invertebrate Neurobiology, (eds North, G. and Greenspan, R. J.), pp. 209–227. Cold Spring Harbor Laboratory Press, New York. Fujita, T. and Kloetzel, K. (1976). A resposta de Rhodnius prolixus a estimulos te´rmicos. Rev. Soc. Bras. Med. Trop. 10, 119–125. Galizia, C. G. and Menzel, R. (2000). Probing the olfactory code. Nat. Neurosci. 3, 853–854. Galizia, C. G., Joerges, J., Kuttner, A., Faber, T. and Menzel, R. (1997). A semi-in-vivo preparation for optical recording of the insect brain. J. Neurosci. Methods 76, 61–69. Galizia, C. G., Menzel, R. and Holldobler, B. (1999a). Optical imaging of odor-evoked glomerular activity patterns in the antennal lobes of the ant Camponotus rufipes. Naturwissenschaften 86, 533–537. Galizia, C. G., Sachse, S., Rappert, A. and Menzel, R. (1999b). The glomerular code for odor representation is species specific in the honeybee Apis mellifera. Nat. Neurosci. 2, 473–478. Galun, R. and Margalit, J. (1970). Some properties of the ATP receptors of Glossina austeni. Trans. R. Soc. Trop. Med. Hyg. 64, 171–174. Galun, R., Friend, W. G. and Nudelman, S. (1988). Purinergic reception by culicine mosquitoes. J. Comp. Physiol. A 163, 665–670. Geier, M., Bosch, O. J. and Boeckh, J. (1999a). Influence of odour plume structure on upwind flight of mosquitoes towards hosts. J. Exp. Biol. 202, 1639–1648.

52

CLAUDIO R. LAZZARI

Geier, M., Bosch, O. J. and Boeckh, J. (1999b). Ammonia as an attractive component of host odour for the yellow fever mosquito, Aedes aegypti. Chem. Senses 24, 647–653. Ghaninia, M., Ignell, R. and Hansson, B. S. (2007a). Functional classification and central nervous projections of olfactory receptor neurons housed in antennal trichoid sensilla of female yellow fever mosquitoes, Aedes aegypti. Eur. J. Neurosci. 26, 1611–1623. Ghaninia, M., Hansson, B. S. and Ignell, R. (2007b). The antennal lobe of the African malaria mosquito, Anopheles gambiae – innervation and three-dimensional reconstruction. Arthropod Struct. Dev. 36, 23–39. Gibson, G. (1992). Do tsetse-flies see zebras? A field-study of the visual response of tsetse to striped targets. Physiol. Entomol. 17, 141–147. Gibson, G. and Torr, S. J. (1999). Visual and olfactory responses of haematophagous Diptera to host stimuli. Med. Vet. Entomol. 13, 2–23. Gilbert, C. (1994). Form and function of stemmata in larvae of holometabolous insects. Annu. Rev. Entomol. 39, 323–349. Gillies, M. T. (1980). The role of carbon dioxide in host-finding by mosquitoes (Diptera: Culicidae): a review. Bull. Entomol. Res. 70, 525–532. Gillies, M. T. and Wilkes, T. J. (1969). A comparison of the range of attraction of animal baits and of carbon dioxide for some West African mosquitoes. Bull. Entomol. Res. 59, 441–456. Gingl, E. and Tichy, H. (2001). Infrared sensitivity of thermoreceptors. J. Comp. Physiol. A 187, 467–475. Gingl, E., Hinterwirth, A. and Tichy, H. (2005). Sensory representation of temperature in mosquito warm and cold cells. J. Neurophysiol. 94, 176–185. Giurfa, M. (2003). Cognitive neuroethology: dissecting non-elemental learning in a honeybee brain. Curr. Opin. Neurobiol. 13, 726–735. Giurfa, M., Eichmann, B. and Menzel, R. (1996). Symmetry perception in an insect. Nature 382, 458–461. Giurfa, M., Zhang, S., Jenett, A., Menzel, R. and Srinivasan, M. V. (2001). The concepts of ‘sameness’ and ‘difference’ in an insect. Nature 410, 930–933. Guerenstein, P. G. (1999). Sensory and behavioural responses of Triatoma infestans to host and conspecific odours. Ph.D. Thesis, University of Neuchaˆtel, 137 pp.. Guerenstein, P. G. and Guerin, P. M. (2001). Olfactory and behavioural responses of the blood-sucking bug Triatoma infestans to odours of vertebrate hosts. J. Exp. Biol. 204, 585–597. Guerenstein, P. G. and Hildebrand, J. G. (2008). Roles and effects of environmental carbon dioxide in insect life. Annu. Rev. Entomol. 53, 161–178. Guerenstein, P. G. and Nuń˜ez, J. A. (1994). Feeding response of the hematophagous bugs Rhodnius prolixus and Triatoma infestans to saline solutions. A comparative-study. J. Insect Physiol. 40, 747–752. Guerenstein, P. G., Lorenzo, M. G., Nuń˜ez, J. A. and Lazzari, C. R. (1995). Baker’s yeast, an attractant for baiting traps for Chagas’ disease vectors. Experientia 51, 834–837. Hall, D. R., Beevor, P. S., Cork, A., Nesbitt, B. F. and Vale, G. A. (1984). 1-Octen-3-ol. A potent olfactory stimulant and attractant for tsetse isolated from cattle odors. Insect Sci. Appl. 5, 335–339. Hardie, J., Gibson, G. and Wyatt, T. D. (2001). Insect behaviours associated with resource finding. In: Insect Movement: Mechanisms and Consequences, (eds Woiwod, I. P., Reynolds, D. R., and Thomas, C. D.), pp. 87–109. CABI Publishing, Wallingford, UK. Horn, E. (1985). Multimodal convergences. In: Comprehensive Insect Physiology, Biochemistry and Pharmacology, Vol. VI, (eds Kerkut, G. A. and Gilbert, L. I.), pp. 653–669. Pergamon Press, Oxford.


53

Hoyle, G. (1984). The scope of neuroethology. Behav. Brain Res. 7, 367–381. Hurlbert, S. H. (1984). Pseudoreplication and the design of ecological field experiments. Ecol. Monogr. 54, 187–211. Ianowski, J. P., Manrique, G., Nuń˜ez, J. A. and Lazzari, C. R. (1998). Feeding is not necessary for triggering plasticization of the abdominal cuticle in haematophagous bugs. J. Insect Physiol. 44, 379–384. Ignell, R. and Hansson, B. S. (2005). Projection patterns of gustatory neurons in the suboesophageal ganglion and tritocerebrum of mosquitoes. J. Comp. Neurol. 492, 214–233. Ignell, R., Dekker, T., Ghaninia, M. and Hansson, B. S. (2005). Neuronal architecture of the mosquito deutocerebrum. J. Comp. Neurol. 493, 207–240. Insausti, T. C. (1994). Nervous system of Triatoma infestans. J. Morphol. 221, 343–359. Insausti, T. C. and Lazzari, C. R. (1996). Central projections of first-order ocellar interneurons in the bug Triatoma infestans (Heteroptera: Reduviidae). J. Morphol. 229, 161–169. Insausti, T. C. and Lazzari, C. R. (2000a). The central projection of cephalic mechanosensory axons in the haematophagous bug Triatoma infestans. Mem. Inst. Oswaldo Cruz 95, 381–388. Insausti, T. C. and Lazzari, C. R. (2000b). An ocellar ‘‘pupil’’ that does not change with light intensity, but with the insect age in Triatoma infestans. Mem. Inst. Oswaldo Cruz 95, 743–746. Jander, R. (1963). Insect orientation. Annu. Rev. Entomol. 8, 95–114. Kelber, A. (2006). Invertebrate colour vision. In: Invertebrate Vision, (eds Warrant, E. J. and Nilsson, D. E.), pp. 250–290. Cambridge University Press, Cambridge. Kelber, A., Vorobyev, M. and Osorio, D. (2003). Animal colour vision – behavioural tests and physiological concepts. Biol. Rev. 78, 81–118. Kelly, D. W. (2001). Why are some people bitten more than others? Trends Parasitol. 17, 578–581. Kelly, D. W. and Thompson, C. E. (2000). Epidemiology and optimal foraging: modelling the ideal free distribution of insect vectors. Parasitology 120, 319–327. Khan, A. A., Maibach, H. I., Strauss, W. G. and Fenley, W. R. (1966). Quantitation of effect of several stimuli on the approach of Aedes aegypti. J. Econ. Entomol. 59, 690–694. Klowden, M. J. (1981). Initiation and termination of host-seeking inhibition in Aedes aegypti during oocyte maturation. J. Insect Physiol. 27, 799–803. Klowden, M. J. (1990). The endogenous regulation of mosquito reproductive behavior. Experientia 46, 660–670. Klowden, M. J. (1995). Blood, sex, and the mosquito: the mechanisms that control mosquito blood-feeding behavior. Bioscience 45, 326–331. Klowden, M. J. (1997). Endocrine aspects of mosquito reproduction. Arch. Insect Biochem. Physiol. 35, 491–512. Klowden, M. J. and Briegel, H. (1994). Mosquito gonotrophic cycle and multiple feeding potential: contrasts between Anopheles and Aedes (Diptera, Culicidae). J. Med. Entomol. 31, 618–622. Klowden, M. J. and Lea, A. O. (1978). Blood meal size as a factor affecting continued host-seeking by Aedes aegypti (L.). Am. J. Trop. Med. Hyg. 27, 827–831. Klowden, M. J. and Lea, A. O. (1979a). Abdominal distention terminates subsequent host-seeking behaviour of Aedes aegypti following a blood meal. J. Insect Physiol. 25, 583–585. Klowden, M. J. and Lea, A. O. (1979b). Humoral inhibition of host-seeking in Aedes aegypti during oocyte maturation. J. Insect Physiol. 25, 231–235.

54

CLAUDIO R. LAZZARI

Klowden, M. J., Davis, E. E. and Bowen, M. F. (1987). Role of the fat-body in the regulation of host-seeking behavior in the mosquito, Aedes aegypti. J. Insect Physiol. 33, 643–646. Knols, B. G., Takken, W. and De Jong, R. (1994). Influence of human breath on selection of biting sites by Anopheles albimanus. J. Am. Mosq. Control Assoc. 10, 423–426. Krishnan, B., Dryer, S. E. and Hardin, P. E. (1999). Circadian rhythms in olfactory responses of Drosophila melanogaster. Nature 400, 375–378. Laarman, J. J. (1955). The host-seeking behaviour of the malaria mosquito Anopheles maculipennis atroparvus. Acta Leiden 25, 1–144. Land, M. F., Gibson, G., Horwood, J. and Zeil, J. (1999). Fundamental differences in the optical structure of the eyes of nocturnal and diurnal mosquitoes. J. Comp. Physiol. A 185, 91–103. Lange, A. B., Orchard, I. and Barrett, F. M. (1989). Changes in hemolymph serotonin levels associated with feeding in the bloodsucking bug, Rhodnius prolixus. J. Insect Physiol. 35, 393–399. Lazzari, C. R. (1992). Circadian organization of locomotion activity in the hematophagous bug Triatoma infestans. J. Insect Physiol. 38, 895–903. Lazzari, C. R. and Nuń˜ez, J. A. (1989a). The response to radiant heat and the estimation of the temperature of distant sources in Triatoma infestans. J. Insect Physiol. 35, 525–529. Lazzari, C. R. and Nuń˜ez, J. A. (1989b). Blood temperature and feeding behavior in Triatoma infestans (Heteroptera, Reduviidae). Entomol. Gener. 14, 183–188. Lazzari, C. R., Reiseman, C. E. and Insausti, T. C. (1998). The role of the ocelli in the phototactic behaviour of the haematophagous bug Triatoma infestans. J. Insect Physiol. 44, 1159–1162. Lazzari, C. R., Minoli, S. A. and Barrozo, R. B. (2004). Chemical ecology of insect vectors: the neglected temporal dimension. Trends Parasitol. 20, 506–507. Lazzari, C. R., Lallement, H., Defrize, J., Vinauger, C. and Minoli, S.A. (2009). Learning and Memory in Chagas disease vectors. Proceedings of the International Symposium on the Centennial of the Discovery of Chagas Disease. p. 1258. Leak, S. G. A. (1999). Tsetse Biology and Ecology: Their Role in the Epidemiology and Control of Trypanosmosis. CABI Publishing, Wallingford, UK. Lehane, M. J. (2005). The biology of blood-sucking in insects. Cambridge University Press, New York. Logan, J. G., Birkett, M. A., Clark, S. J., Powers, S., Seal, N. J., Wadhams, L. J., Mordue Luntz, A. J. and Pickett, J. A. (2008). Identification of human-derived volatile chemicals that interfere with attraction of Aedes aegypti mosquitoes. J. Chem. Ecol. 34, 308–322. Lorenzo, M. G. and Lazzari, C. R. (1996). The spatial pattern of defaecation in Triatoma infestans and the role of faeces as a chemical mark of the refuge. J. Insect Physiol. 42, 903–907. Lorenzo, M. G. and Lazzari, C. R. (1998). Activity pattern in relation to refuge exploitation and feeding in Triatoma infestans (Hemiptera: Reduviidae). Acta Trop. 70, 163–170. Lorenzo, M. G., Flores, G. B., Lazzari, C. R. and Reisenman, C. E. (1999). Sensory Ecology. A: Orientation. In: Atlas of Chagas’ disease vectors in America, vol. III, (eds. Carcavallo, R. et al.) pp. 1071–1087. Editora Fiocruz, Rio de Janeiro. Mafra-Neto, A. and Carde´, R. T. (1994). Fine-scale structure of pheromone plumes modulates upwind orientation of flying moths. Nature 369, 142–144. Manrique, G., Vitta, A. C., Ferreira, R. A., Zani, C. L., Unelius, C. R., Lazzari, C. R., Diotaiuti, L. and Lorenzo, M. G. (2006). Chemical communication in Chagas disease vectors. Source, identity, and potential function of volatiles released by the


55

metasternal and Brindley’s glands of Triatoma infestans adults. J. Chem. Ecol. 32, 2035–2052. Mboera, L. E., Knols, B. G., Takken, W. and Huisman, P. W. (1998). Olfactory responses of female Culex quinquefasciatus Say (Diptera: Culicidae) in a dual-choice olfactometer. J. Vector Ecol. 23, 107–113. McCall, P. J. and Eaton, G. (2001). Olfactory memory in the mosquito Culex quinquefasciatus. Med. Vet. Entomol. 15, 197–203. McCall, P. J. and Kelly, D. W. (2002). Learning and memory in disease vectors. Trends Parasitol. 18, 429–433. McCall, P. J., Mosha, F. W., Njunwa, K. J. and Sherlock, K. (2001). Evidence for memorized site-fidelity in Anopheles arabiensis. Trans. R. Soc. Trop. Med. Hyg. 95, 587–590. McIver, S. and Siemicki, R. (1985). Fine-structure of antennal putative thermohygrosensilla of adult Rhodnius prolixus Stal (Hemiptera, Reduviidae). J. Morphol. 183, 15–23. Meijerink, J., Braks, M. A. and van Loon, J. J. (2001). Olfactory receptors on the antennae of the malaria mosquito Anopheles gambiae are sensitive to ammonia and other sweat-borne components. J. Insect Physiol. 47, 455–464. Melcon, M. L., Lazzari, C. R. and Manrique, G. (2005). Repeated plasticization and recovery of cuticular stiffness in the blood-sucking bug Triatoma infestans in the feeding context. J. Insect Physiol. 51, 989–993. Merlin, C., Lucas, P., Rochat, D., Francois, M. C., Maibeche-Coisne, M. and JacquinJoly, E. (2007). An antennal circadian clock and circadian rhythms in peripheral pheromone reception in the moth Spodoptera littoralis. J. Biol. Rhythms 22, 502–514. Mizunami, M. (1995). Functional diversity of neural organization in insect ocellar systems. Vision Res. 35, 443–452. Muir, L. E., Thorne, M. J. and Kay, B. H. (1992). Aedes aegypti (Diptera, Culicidae) vision. Spectral sensitivity and other perceptual parameters of the female eye. J. Med. Entomol. 29, 278–281. Mukabana, W. R., Takken, W., Coe, R. and Knols, B. G. (2002). Host-specific cues cause differential attractiveness of Kenyan men to the African malaria vector Anopheles gambiae. Malar. J. 1, 17. Mukabana, W. R., Takken, W., Killeen, G. F. and Knols, B. G. (2004). Allomonal effect of breath contributes to differential attractiveness of humans to the African malaria vector Anopheles gambiae. Malar. J. 3, 1. Mwandawiro, C., Boots, M., Tuno, N., Suwonkerd, W., Tsuda, Y. and Takagi, M. (2000). Heterogeneity in the host preference of Japanese encephalitis vectors in Chiang Mai, northern Thailand. Trans. R. Soc. Trop. Med. Hyg. 94, 238–242. Nicolle, P. and Mathis, M. (1941). Le termotropisme, facteur de´terminant primordial pour la piqure des Re´duvide´s he`matophages. C. R. Seances Soc. Biol. Fil. 135, 25–27. Nijhout, H. F. and Sheffield, H. G. (1979). Antennal hair erection in male mosquitoes: a new mechanical effector in insects. Science 206, 595–596. Nuń˜ez, J. A. (1982). Food source orientation and activity in Rhodnius prolixus Stal (Hemiptera, Reduviidae). Bull. Entomol. Res. 72, 253–262. Nuń˜ez, J. A. and Lazzari, C. R. (1990). Rearing of Triatoma infestans Klug (Het., Reduviidae) in the absence of a live host. 1. Some factors affecting the artificial feeding. J. Appl. Entomol. 109, 87–92. Orchard, I. (2006). Serotonin: a coordinator of feeding-related physiological events in the blood-gorging bug, Rhodnius prolixus. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 144, 316–324. Orchard, I., Lange, A. B. and Barrett, F. M. (1988). Serotonergic supply to the epidermis of Rhodnius prolixus: evidence for serotonin as the plasticising factor. J. Insect Physiol. 34, 873–879.

56

CLAUDIO R. LAZZARI

Pappenberger, B., Geier, M. and Boeckh, J. (1996). Responses of antennal olfactory receptors in the yellow fever mosquito Aedes aegypti to human body odours. Ciba Found. Symp. 200, 254–263. Page, T. L. and Koelling, E. (2003). Circadian rhythm in olfactory response in the antennae controlled by the optic lobe in the cockroach. Journal of Insect Physiology 49, 697–707. Peterson, D. G. and Brown, A. W. A. (1951). Studies of the responses of the female Aedes mosquito. 3. The response of Aedes aegypti (L) to a warm body and its radiation. Bull. Entomol. Res. 42, 535–541. Pfluger, H. J. and Menzel, R. (1999). Neuroethology, its roots and future. J. Comp. Physiol. A 185, 389–392. Pittendrigh, C. S. (1974). Circadian oscillations in cells and the circadian organization of multicellular systems. In: The Neurosciences: third study program, (eds. Schmitt, F. O. and Worden, F. G.), pp. 437–458. Cambridge, Mass., MIT Press. Pickett, J. A., Birkett, M. A. and Logan, J. G. (2008). DEET repels ORNery mosquitoes. Proc. Natl. Acad. Sci. USA 105, 13195–13196. Pompilio, L., Kacelnik, A. and Behmer, S. T. (2006). State-dependent learned valuation drives choice in an invertebrate. Science 311, 1613–1615. Pontes, G. B., Bohman, B., Unelius, C. R. and Lorenzo, M. G. (2008). Metasternal gland volatiles and sexual communication in the triatomine bug, Rhodnius prolixus. J. Chem. Ecol. 34, 450–457. Qiu, Y. T., van Loon, J. J., Takken, W., Meijerink, J. and Smid, H. M. (2006). Olfactory coding in antennal neurons of the malaria mosquito, Anopheles gambiae. Chem. Senses 31, 845–863. Ramirez, C. C., Fuentes-Contreras, E., Rodriguez, L. C. and Niemeyer, H. M. (2000). Pseudoreplication and its frequency in olfactometric laboratory studies. J. Chem. Ecol. 26, 1423–1431. Reinhardt, K. and Siva-Jothy, M. T. (2007). Biology of the bed bugs (Cimicidae). Annu. Rev. Entomol. 52, 351–374. Reisenman, C. E. and Lazzari, C. (2006). Spectral sensitivity of the photonegative reaction of the blood-sucking bug Triatoma infestans (Heteroptera: Reduviidae). J. Comp. Physiol. A 192, 39–44. Reisenman, C. E., Lazzari, C. R. and Giurfa, M. (1998). Circadian control of photonegative sensitivity in the haematophagous bug Triatoma infestans. J. Comp. Physiol. A 183, 533–541. Reisenman, C. E., Lorenzo Figueiras, A. N., Giurfa, M. and Lazzari, C. R. (2000). Interaction of visual and olfactory cues in the aggregation behaviour of the haematophagous bug Triatoma infestans. J. Comp. Physiol. A 186, 961–968. Reisenman, C. E., Insausti, T. C. and Lazzari, C. R. (2002). Light-induced and circadian changes in the compound eye of the haematophagous bug Triatoma infestans (Hemiptera: Reduviidae). J. Exp. Biol. 205, 201–210. Ribeiro, J. M. (1995). Blood-feeding arthropods: live syringes or invertebrate pharmacologists? Infect. Agents Dis. 4, 143–152. Sabelis, M. W. and Schippers, P. (1984). Variable wind directions and anemotactic strategies of searching for an odor plume. Oecologia 63, 225–228. Saifullah, A. S. M. and Page, T. L. (2009). Circadian regulation of olfactory receptor neurons in the cockroach antenna. J. Biol. Rhythms 24, 144–152. Saunders, D. S., Steel, C. G. H., Vafopoulou, X. and Lewis, R. D. (2002). Insect Clocks. Elsevier Science, Amsterdam. Schmitz, H. and Bleckmann, H. (1997). Fine structure and physiology of the infrared receptor of beetles of the genus Melanophila (Coleoptera: Buprestidae). Int. J. Insect Morphol. Embryol. 26, 205–215.


57

Schmitz, H., Trenner, S., Hofmann, M. H. and Bleckmann, H. (2000). The ability of Rhodnius prolixus (Hemiptera; Reduviidae) to approach a thermal source solely by its infrared radiation. J. Insect Physiol. 46, 745–751. Schmitz, A., Gebhardt, M. and Schmitz, H. (2008). Microfluidic photomechanic infrared receptors in a pyrophilous flat bug. Naturwissenschaften 95, 455–460. Seaton, D. R. and Lumsden, W. H. R. (1941). Observations on the effects of age, fertilization and light on biting by Aedes aegypti (L.) in a controller microclimate. Ann. Trop. Med. Parasitol. 35, 23–36. Sfara, V., Zerba, E. N. and Alzogaray, R. A. (2006). Toxicity of pyrethroids and repellency of diethyltoluamide in two deltamethrin-resistant colonies of Triatoma infestans Klug, 1834 (Hemiptera: Reduviidae). Mem. Inst. Oswaldo Cruz 101, 89–94. Sfara, V., Zerba, E. N. and Alzogaray, R. A. (2008). Decrease in DEET repellency caused by nitric oxide in Rhodnius prolixus. Arch. Insect Biochem. Physiol. 67, 1–8. Siju, K. P., Hansson, B. S. and Ignell, R. (2008). Immunocytochemical localization of serotonin in the central and peripheral chemosensory system of mosquitoes. Arthropod Struct. Dev. 37, 248–259. Singh, R. N., Singh, K., Prakash, S., Mendki, M. J. and Rao, K. M. (1996). Sensory organs on the body parts of the bed-bug Cimex hemipterus fabricius (Hemiptera: Cimicidae) and the anatomy of its central nervous system. Int. J. Insect Morphol. Embryol. 25, 183–204. Smallegange, R. C., Qiu, Y. T., van Loon, J. J. and Takken, W. (2005). Synergism between ammonia, lactic acid and carboxylic acids as kairomones in the host-seeking behaviour of the malaria mosquito Anopheles gambiae sensu stricto (Diptera: Culicidae). Chem. Senses 30, 145–152. Stavenga, D. G. and Hardie, R. C. (eds.) (1989). In: Facets of Vision, Springer-Verlag, Berlin. Syed, Z. and Guerin, P. M. (2004). Tsetse flies are attracted to the invasive plant Lantana camara. J. Insect Physiol. 50, 43–50. Syed, Z. and Leal, W. S. (2007). Maxillary palps are broad spectrum odorant detectors in Culex quinquefasciatus. Chem. Senses 32, 727–738. Syed, Z. and Leal, W. S. (2008). Mosquitoes smell and avoid the insect repellent DEET. Proc. Natl. Acad. Sci. USA 105, 13598–13603. Takaćs, S., Bottomley, H., Andreller, I., Zaradnik, T., Schwarz, J., Bennett, R., Strong, W. and Gries, G. (2009). Infrared radiation from hot cones on cool conifers attracts seed-feeding insects. Proc. R. Soc. B Biol. Sci. 276, 649–655. Takken, W. (2005). Chemical ecology of insect vectors: temporal, environmental and physiological aspects. Trends Parasitol. 21, 57. Takken, W. and Kline, D. L. (1989). Carbon dioxide and 1-octen-3-ol as mosquito attractants. J. Am. Mosq. Control Assoc. 5, 311–316. Takken, W. and Knols, B. G. (1999). Odor-mediated behavior of Afrotropical malaria mosquitoes. Annu. Rev. Entomol. 44, 131–157. Takken, W., van Loon, J. J. and Adam, W. (2001). Inhibition of host-seeking response and olfactory responsiveness in Anopheles gambiae following blood feeding. J. Insect Physiol. 47, 303–310. Taneja, J. and Guerin, P. M. (1995). Oriented responses of the triatomine bugs Rhodnius prolixus and Triatoma infestans to vertebrate odors on a servosphere. J. Comp. Physiol. A 176, 455–464. Taneja, J. and Guerin, P. M. (1997). Ammonia attracts the haematophagous bug Triatoma infestans: Behavioural and neurophysiological data on nymphs. J. Comp. Physiol. A 181, 21–34.

58

CLAUDIO R. LAZZARI

Tichy, H., Fischer, H. and Gingl, E. (2008). Adaptation as a mechanism for gain control in an insect thermoreceptor. J. Neurophysiol. 100, 2137–2144. Torr, S. J. (1988). The activation of resting tsetse flies (Glossina) in response to visual and olfactory stimuli in the field. Physiol. Entomol. 13, 315–325. Torr, S. J. (1990). Dose responses of tsetse flies (Glossina) to carbon dioxide, acetone and octenol in the field. Physiol. Entomol. 15, 93–103. Usinger, R. L. (1966). Monograph of Cimicidae (Hemiptera- Heteroptera). Entomological Society of America, College Park, ML. Vale, G. A. and Hall, D. R. (1985). The role of 1-octen-3-ol, acetone and carbon dioxide in the attraction of tsetse flies, Glossina spp (Diptera, Glossinidae), to ox odour. Bull. Entomol. Res. 75, 209–217. Van der Goes van Naters, W. M., Den Otter, C. J. and Maes, F. W. (1998). Olfactory sensitivity in tsetse flies: a daily rhythm. Chem. Senses 23, 351–357. Warnes, M. L. (1990). The effect of host odor and carbon dioxide on the flight of tsetse flies (Glossina spp) in the laboratory. J. Insect Physiol. 36, 607–611. Warrant, E. J. and Nilsson, D. E. (eds.) (2006). In: Invertebrate Vision, Cambridge University Press, Cambridge. Werner-Reiss, U., Galun, R., Crnjar, R. and Liscia, A. (1999a). Sensitivity of the mosquito Aedes aegypti (Culicidae) labral apical chemoreceptors to phagostimulants. J. Insect Physiol. 45, 629–636. Werner-Reiss, U., Galun, R., Crnjar, R. and Liscia, A. (1999b). Factors modulating the blood feeding behavior and the electrophysiological responses of labral apical chemoreceptors to adenine nucleotides in the mosquito Aedes aegypti (Culicidae). J. Insect Physiol. 45, 801–808. Wigglesworth, V. B. (1941). The sensory physiology of the human louse Pediculus humanus corporis DeGeer (Anoplura). Parasitology 33, 67–109. Wigglesworth, V. B. (1953). The origin of sensory neurones in an insect, Rhodnius prolixus (Hemiptera). Q. J. Microsc. Sci. 94, 93–112. Wigglesworth, V. B. (1959a). The histology of the nervous system of an insect, Rhodnius prolixus (Hemiptera). 1. The peripheral nervous system. Q. J. Microsc. Sci. 100, 285–298. Wigglesworth, V. B. (1959b). The histology of the nervous system of an insect, Rhodnius prolixus (Hemiptera). 2. The central ganglia. Q. J. Microsc. Sci. 100, 299–314. Wigglesworth, V. B. and Gillet, J. D. (1934a). The function of the antennae in Rhodnius prolixus (Hemiptera) and the mechanism of orientation to the host. J. Exp. Biol. 11, 120–138. Wigglesworth, V. B. and Gillet, J. D. (1934b). The function of the antennae in Rhodnius prolixus: confirmatory experiments. J. Exp. Biol. 11, 408. Xia, Y., Wang, G., Buscariollo, D., Pitts, R. J., Wenger, H. and Zwiebel, L. J. (2008). The molecular and cellular basis of olfactory-driven behavior in Anopheles gambiae larvae. Proc. Natl. Acad. Sci. USA 105, 6433–6438. Zhou, G., Pennington, J. E. and Wells, M. A. (2004). Utilization of pre-existing energy stores of female Aedes aegypti mosquitoes during the first gonotrophic cycle. Insect Biochem. Mol. Biol. 34, 919–925. Zhou, X. J., Yuan, C. Y. and Guo, A. K. (2005). Drosophila olfactory response rhythms require clock genes but not pigment dispersing factor or lateral neurons. J. Biol. Rhythms 20, 237–244. Zube, C., Kleineidam, C. J., Kirschner, S., Neef, J. and Rossler, W. (2008). Organization of the olfactory pathway and odor processing in the antennal lobe of the ant Camponotus floridanus. J. Comp. Neurol. 506, 425–441. Zwiebel, L. J. and Takken, W. (2004). Olfactory regulation of mosquito-host interactions. Insect Biochem. Mol. Biol. 34, 645–652.

From Sialomes to the Sialoverse: An Insight into Salivary Potion of Blood-Feeding Insects Jose´ M. C. Ribeiro* and Bruno Arca`†,‡ *Laboratory

of Malaria and Vector Research, National Institute of Allergy and Infectious Diseases, 12735 Twinbrook Parkway room 2E-32D, Rockville, Maryland 20852, USA † Department of Structural and Functional Biology, University ‘‘Federico II’’, Naples, Italy ‡ Parasitology Section, Department of Public Health, University ‘‘La Sapienza’’, Rome, Italy

1 Insects discover blood as food 60 2 Blood feeders like fast food: A historical perspective 63 3 Problems faced by arthropods when taking blood 64 3.1 Haemostasis 64 3.2 Inflammation 66 3.3 Annoying itching 67 3.4 . . . and pain . . . 70 3.5 The attacked endothelium fights back 70 3.6 Microbiological concerns 70 4 Toward a longitudinal definition of the salivary components of blood-feeding insects 70 4.1 Enzymes 71 4.2 Receptor antagonism and platelet aggregation inhibitors 75 4.3 Physiological antagonists, primarily vasodilators 76 4.4 Kratagonists 77 4.5 Protease inhibitors 80 4.6 Anaesthetics 82 4.7 Antigen (Ag5) family members 87 4.8 Immunity-related products 87 4.9 The unexpected 88 5 Salivary diversity 88 6 The evolutionary scramble 90 7 On the odd, the paradoxical, the bizarre and the bias 94 8 Measuring the size of our ignorance 95 8.1 A forecast of the costs and time required for acquiring sialome wisdom 99 9 Salivary antigens: Epidemiological tools? 99 Acknowledgements 100 References 100

ADVANCES IN INSECT PHYSIOLOGY VOL. 37 ISBN 978-0-12-374829-4 DOI: 10.1016/S0065-2806(09)37002-2

60

1

JOSE´ M. C. RIBEIRO AND BRUNO ARCA`

Insects discover blood as food

The habit of blood feeding evolved several times within the Insecta, including several instances within the Hemiptera and Diptera, as well as in the Anoplura, Siphonaptera and even in the Lepidoptera (Table 1) (Ribeiro, 1995). Several scenarios can be envisioned to understand the evolution of such diet. Insects may have been nest parasites and initially predators on other insects, as proposed for triatomine bugs (Cobben, 1979; Schofield, 1979; Sweet, 1979), and still extant on the Blephariceridae, Nematocera midges the adults of which can feed on insect haemolymph (Grimaldi and Engel, 2005), or they may have approached vertebrate eyes or other mucosal membranes to feed on their secretions, as eye gnats and some moth species do. This habit may well have been the stepping stone for individuals to obtain further nourishment by piercing the skin and acquiring blood. Indeed this may have happened with all or most blood-feeding Diptera, as well as with the few blood-feeding moth species, which include close relatives that do not blood feed but lick the eyes of vertebrates (Ba¨nziger, 1970, 1975, 1979; Hilgartner et al., 2007; Zaspel et al., 2007). Many lice species, even today, do not blood feed, but live on their host’ skin feeding on dander. Obviously, insects that were already equipped with either piercing or cutting mouthparts were provided with an important preadaptation to remove blood from vertebrates and were favoured by natural selection. The insect orders feeding on blood include both holometabolous (those having full metamorphosis, with immature instars that are very different and occupy a different niche as compared to adults) and hemimetabolous organisms (where immature forms and adults inhabit the same niche and are not very different from each other). Accordingly, all haematophagous hemimetabolous orders feed exclusively on blood and nothing else. In the case of holometabolous orders (all flies and fleas), only the adults take blood meals, except for the Congo floor maggot. Within the Nematocera, it is also common that only the adult female will feed on blood whereas both adult sexes will feed also on sugar solutions. The adults of haematophagous Brachycera flies feed exclusively on blood. Blood is not an ideal meal because it is heavily unbalanced towards proteins, produce dangerous amounts of pro-oxidating haem (Dansa-Petreski et al., 1995), and lacks or has little amount of some vitamins. However, it can provide, in the form of proteins, the large amounts of amino acids needed for egg development. In mosquitoes and sand flies, only adult females are haematophagous and blood feeding is promptly followed by egg development. Those insects that feed solely on blood, such as the bugs and tsetse, require the presence of bacterial endosymbionts to survive. Killing the symbionts with antibiotics prevents bugs from developing to adults and suppresses tsetse reproduction and immunity, a deleterious effect that can be reversed with B vitamin supplements (Wigglesworth, 1936; Baines, 1956; Hill et al., 1973, 1976; Pais et al., 2008).

TABLE 1 List of BFI Families Order Diptera

Sub-order

Common name

Genera

Species

References

Nematocera

Sand flies Mosquitoes Black flies Biting midges Frog-biting midges Tsetse Horse flies Stable flies Keds Bat flies Blowfly Bed bugs Bat bugs Kissing bugs Lice Fleas Moth 17

6 36 24 4 1 1 5 3 19 31 1 23 5 14 42 239 1 455

70 3450 1571 1000 97 23 4400 50 130 520 5 91 32 118 490 2500 8 14,555

Lane and Crosskey (1993) Lane and Crosskey (1993) Lane and Crosskey (1993) Lane and Crosskey (1993) Borkent (2008) Lane and Crosskey (1993) Lane and Crosskey (1993) Lane and Crosskey (1993) Maa (1964) Maa (1965) Lane and Crosskey (1993) Lane and Crosskey (1993) Lane and Crosskey (1993) Lane and Crosskey (1993) Lane and Crosskey (1993) Lane and Crosskey (1993) Zaspel et al. (2007)

Brachycera

Hemiptera

Anoplura Siphonaptera Lepidoptera Total

62


The surface availability of blood varies considerably among vertebrates. There, blood not only serves a nourishing tissue function but also contributes to the organism’s thermoregulation. Fresh water turtles, for example, have a venous plexus on the basis of their carapace that serves a heat exchange function with the environment. On the other extreme, humans can use their whole body skin as a thermoregulatory organ. Surface capillary loops are attended by a plexus of arterioles and venules and blood flow to these capillaries is controlled by the arteriolar precapillary sphincter. There are 20 times more vessels in the skin than needed for nourishment of its cells, so only 5% of these capillaries need to be with active blood flow, but they can all be turned on when the organism needs to lose heat. The normally closed state of the precapillary sphincter is affected by a sympathetic tonus, through the neural release of the smooth muscle contracting agent norepinephrine. Arteriolar-to-venous anastomosis serves as bypass to conduct blood directly from arterioles to venules in the case there is no flow to the capillaries (Fig. 1) (Abramson, 1989; Braverman, 1997). Other mammalians use specialized skin regions for thermoregulation: this is the case for rodents, whose tail and ears can be more vascularized as compared to other skin regions, as those in their furry backs or abdomens. Accordingly, the blood sources that can be tapped by haematophagous insects can vary from surface capillaries to deeper venules and arterioles. The accessibility to the skin surface and their vessels can also be turned more difficult to blood suckers by the presence of scales, feathers and fur, which can completely prevent access to some blood-feeding insects (BFI). The mechanics of blood feeding by insects have been described in classical studies using histology of fixed insects while feeding on the skin of their hosts (Short and Swaminath, 1928), or directly by live observations using frog feet or the transilluminated ear of the mouse (Gordon and Lumsden, 1939; Gordon and Crewe, 1948; O’Rourke, 1956; Dickerson and Lavoipierre, 1959; Lavoipierre et al., 1959). From these studies two modes of feeding became apparent. Feeding can either occur from haemorrhagic pools or haematomas that accumulate in the tissues following skin lacerations (pool feeders or telmophagous insects), or directly from a cannulated venule or arteriole (vessel feeders or solenophagous insects) (Lavoipierre, 1964). In some cases, both modes of feeding can be observed in the same species (O’Rourke, 1956). Some insects, such as the sand flies, have very short mouthparts that penetrate no more than 0.5 mm into their host skin (Adler and Theodor, 1926; Lewis, 1975). Accordingly, they can only feed from superficial pools of blood (haematomas) formed from haemorrhages following the laceration of capillaries. Other insects like the bugs and mosquitoes can penetrate their host skin for several millimetres and cannulate the arterioles and venules deeper in the skin circulatory plexus. The tip of the mouthparts of triatomine bugs have only 10 mm in diameter (Lavoipierre et al., 1959); accordingly, only one erythrocyte can pass at a time, showing the limits achieved by evolution in developing a fine blood drawing needle. On the other hand, Tabanids have veritable cutting

FROM SIALOMES TO THE SIALOVERSE

63

scissors that can lacerate several vessels simultaneously, potentially forming a large haematoma, as in cattle feeding horse flies, or tabanid species that specialize in feeding on the venous plexus of turtles (DeGiusti et al., 1973). It is important to notice that blood feeding does not initiate immediately after the insect pierces their host skin. Sometimes blood ingestion may start in a few seconds, but most commonly it may take 1 min or more for the insect to find blood, and then several more minutes to feed to repletion. The period of time from the initial penetration of the skin to the initial ingestion of blood is known as the exploratory phase or probing time (Gillett, 1967; Ribeiro et al., 1985a). Saliva is continuously ejected during this exploratory phase of probing, while the insect mouthparts move actively inside their host skin (Friend and Smith, 1971; Soares et al., 2006). It is during this phase that saliva interacts with host tissues. Saliva is also continuously discharged during the blood ingestion phase, which ends up mostly in the insect gut, reingested with the food (Ribeiro and Garcia, 1980).

2

Blood feeders like fast food: A historical perspective

Although saliva of blood-sucking arthropods (BFA) was known to contain anti-clotting activities near a century ago (Cornwall and Patton, 1914), its role was postulated to have a post-meal effect by preventing blood clotting in the insect mouthparts and gut (Lester and Lloyd, 1928), or to have no positive function at all, since allergic reactions to insect bite were obviously nonadaptive (De Meillon, 1949). The classical method to study the role of any gland in animal physiology was to remove it, or prevent its juices from reaching their target, and observe its effect on the impaired animal (Pavlov, 1902). Such protocols applied to insects led to the conclusion that saliva played no important role in blood feeding because insects (tsetse and mosquitoes) could still feed after such treatment (Lester and Lloyd, 1928; Hudson et al., 1960; Hudson, 1964; Rossignol and Spielman, 1982). Indeed, the last sentence of a 1982 paper studying salivation in a mosquito was ‘‘We conclude that saliva is not prerequisite to blood feeding’’ (Rossignol and Spielman, 1982). These studies, however, allowed a long contact time between the surgically treated insects and their hosts, and did not measure the feeding kinetics. When similar experiments were repeated with mosquitoes and triatomine bugs, but at this time measuring the effect of salivation on the speed of blood meal acquisition, a significant role of saliva in feeding was discovered (Mellink and Van Den Boven Kamp, 1981; Ribeiro and Garcia, 1981; Ribeiro et al., 1984a). Notably, salivation significantly reduced the time from the initial mouthpart penetration of the skin to the first observed ingestion of blood, the so-called probing time (Gillett, 1967; Mellink and Van Den Boven Kamp, 1981; Ribeiro et al., 1985a; Ribeiro, 1988). Speed of blood meal acquisition is important to insects because they are at a great risk of predation during their stealing act. The role of saliva was


64

then postulated to function by antagonizing host haemostasis, allowing for haematomas to form at the feeding site, which could themselves be the source of the blood meal in pool-feeding insects, or, when the haematomas were rapidly sucked, their walls collapsed and drove the mouthparts towards the ruptured vessel, in the case of vessel-feeding insects (Ribeiro, 1987, 1988, 1995). In the past 25 years anti-platelet and vasodilatory activities, in addition to anti-coagulants, were described in several blood-sucking insects (Champagne, 2004; Valenzuela, 2004); vector saliva was also discovered to enhance pathogen transmission and thus became an attractive target for developing vaccines against vector borne diseases (Mejia et al., 2006; Oliveira et al., 2006; Gomes et al., 2008). At less than a decade ago, the omics revolution kicked in, in the form of whole genomes, salivary gland transcriptomes and tissue proteomes, generating the sialomes (from the Greek, sialo ¼ saliva) which changed the direction we now do science: pre-revolutionary times started with a bioactivity identified in an organ homogenate and ended with the characterization of the molecule responsible for the biological effect, but now starts with a recombinant molecule identified in a transcriptome, often without any clue regarding its function, and ends with the determination of the molecule’s function and verification of its presence in saliva (Ribeiro and Francischetti, 2003). This chapter will concentrate on the past 5 years advances on the role of haematophagous insect saliva in antagonizing their hosts’ haemostasis and inflammation, specifically the processes that occur within 15 min of skin injury. An attempt will also be made to classify the modes of action of the various antagonists so far characterized, and of the genomic and evolutionary mechanism that allowed the concoction of the haematophagous ‘‘magic potion’’. Immunity-related products and the potential to use salivary proteins as markers of exposure to bites of insect disease vectors will be mentioned only briefly. The reader is directed to other reviews related to the physiological role of saliva in ticks (Bowman and Sauer, 2004; Steen et al., 2006; Hovius et al., 2008; Francischetti et al., 2009), specifically those related to host immunity (Brossard and Wikel, 2004), and to the investigations related to vaccine development using salivary proteins as antigens to disrupt tick feeding or modify transmission of vector borne diseases, such as leishmaniasis (Valenzuela, 2004; Mejia et al., 2006; Nuttall et al., 2006; Titus et al., 2006).

3 3.1

Problems faced by arthropods when taking blood HAEMOSTASIS

The idea that saliva of BFA counteracted their hosts’ blood clotting, platelet aggregation and vasoconstriction drove much of the early research on this area. Earlier reviews focused on the components of this haemostasis tripod (Ribeiro, 1987; Law et al., 1992) to develop experiments leading to the discovery first of


65

the biological activities of saliva or salivary gland homogenates, followed then by the isolation of the responsible molecules. The recognition of the physiological redundancy of haemostasis led to the consequent idea that a ‘‘magic bullet’’ is virtually impossible to function against such a system, and that BFA opted instead for a ‘‘magic potion’’ to accomplish such a pharmacological task (Ribeiro, 1995). This salivary anti-haemostasis paradigm was enlarged to include the contributions of inflammation, tissue repair or angiogenesis and immunity (Ribeiro and Francischetti, 2003; Francischetti et al., 2009), mostly to account for the effects of tick saliva. Results from the last 7 years, mostly driven by unexpected findings deriving from sialome discovery projects, are pointing out that fast BFI have a plethora of antagonists targeting neutrophils or their products, as well as directed against other arms of inflammation that have been previously neglected. Some of them also have enzymes that attack the extracellular matrix (ECM), which would be natural for ticks, but unexpected in fast-feeding insects. Within this framework, we will attempt in this section to provide an expanded (but simplified) description of the haemostasis process to include the role of inflammation, including the participation of neutrophils, mast cells and complement, as they may affect those critical 15 min taken by the majority of haematophagous insects to blood feed. It is not our goal in this chapter to produce a detailed account of these processes, which are the subject of textbooks, but rather to describe their general outlines in the context of the vertebrate barriers to blood feeders, and to direct the reader to more detailed literature. Haemostasis is the natural response to vascular injury that serves the purpose of arresting bleeding and it is based on the triad of platelet aggregation, blood clotting and vasoconstriction (Colman et al., 2005). Inflammation comprises the larger spectrum of physiological responses following tissue (not only vascular) injury, and it was classically described by Cornelius Celsus in the first century (Collier, 2008) as rubor (redness) and tumour (swelling) with calor (heat) and dolour (pain). Notice that redness and heat both result from vasodilation: actually, while at the site of vascular injury there is vasoconstriction due to haemostasis, injured tissue in the vicinity of non-injured vascular tissue brings about vascular changes that are of a vasodilatory nature. Accordingly, a puncture wound shows vasoconstriction at the site of the injury due to haemostasis reactions, and vasodilatation in the nearby areas due to inflammation. We will proceed to describe the events occurring at the site of tissue injury, such as one being created by a hungry kissing bug (Fig. 2). To our purposes, the immediate relevant events following vascular and tissue injury are the mixing of blood components with extravascular tissues, and the release of ATP and ADP by broken cells to the extracellular milieu. These nucleotides have cytoplasmic concentrations at the millimolar level and only sub-micromolar concentrations extracellularly. Within seconds of exposure to micromolar concentrations of ADP, platelets expose surface receptors to fibrinogen, a multi-domain protein that clamps platelets together. The mixing of extra


66 EP D

PCS A V

FIG. 1 Diagram of human skin showing the epidermis (EP), dermis (D), arterioles (A), venules (V) and precapillary sphincter (PCS). Based on Abramson (1989).

and intravascular compartment components allows platelets to interact with collagen, which by a specific signalling pathway also induce fibrinogen receptors to be expressed in the platelet surface. Exposure of plasma components to membrane bound tissue factor, present in most extravascular cells, promotes activation of the extrinsic pathway of blood clotting, with activation of prothrombin to thrombin, which cleaves fibrinogen into fibrin, leading to blood clotting. Thrombin activates the platelet protease-activated receptor (PAR) which independently of the ADP or collagen receptors, also activate platelets to expose their fibrinogen receptors. The ATP released by broken cells also activates neutrophils to aggregate and to adhere to the endothelium (Colman et al., 2005). Accordingly, within the first minute of vascular injury, platelets had acquired the signalling of three different receptors that induce platelet aggregation, and neutrophils are also being warned. 3.2

INFLAMMATION

But not only platelet aggregation occurs during this first minute. Platelets under those three strong stimuli (ADP, collagen and thrombin) release their granule contents, which include 5-hydroxytriptamine (5-HT, serotonin), epinephrine and norepinephrine (NE), ADP, inorganic polyphosphate and the chemokine precursors platelet factor 4 (PF-4) and b-thromboglobulin (von Hundelshausen et al., 2007). ADP, collagen and thrombin, and also less effectively 5-HT and NE, activate platelet membrane phospholipases that release arachidonic acid (AA) from membrane phospholipids. AA is converted by platelet cyclooxygenase to prostaglandin H2 (PGH2) which is then transformed by platelets into thromboxane A2 (TXA2) and by endothelial and smooth muscle cells to prostacyclin (PGI2). During the granule exocytosis process, the platelet membrane inverts its membrane polarity, exposing negatively charged phospholipids in its outer surface. The polyphosphates and the negatively charged membrane phospholipid surface serve as scaffolds for the assembly and activation of the Xase and prothrombinase complexes of the blood coagulation cascade, thus connecting platelets to blood clotting. TXA2, NE and 5-HT are potent vasoconstrictors, thus decreasing the


67

blood flow to the site of injury. Aggregated platelets also expose the P-selectin protein on their outer surface. P-selectin promotes neutrophil binding to platelets and neutrophil activation. The platelet-derived chemokine precursors are processed within minutes by neutrophil proteases, cathepsin G (CG) in particular, which further activate neutrophils. Neutrophil CG can also activate platelets in a similar way as thrombin does, by activation of PAR. Activated neutrophils expose additional adhesive surface molecules, release their granules, which include ATP and proteases, and produce the lipidic mediators PAF (platelet-activating factor), and leukotriene B4 (LTB4). PAF is a potent activator of platelet aggregation and neutrophil activation. LTB4 is a potent neutrophil activator. It is to be noted also that activated neutrophils release antimicrobial peptides (AMP) and induce a respiratory burst producing superoxide radicals, toxic to microbes. Taking it all together, platelets and neutrophils are very actively at work within the first minute of vascular injury. Within 1 min, a platelet plug containing adhered neutrophils is formed in the vicinity of the injury, this platelet plug is consolidated by a fibrin network that gives more rigidity to the plug; 5-HT, NE and TXA2 contracts the arterioles’ and venules’ smooth muscle (capillaries have no smooth muscle). Within 1 min, the full haemostasis triad is in place. (For a review on inflammation, see Chapter 2 of Kumar et al., 2004.) It is interesting to point out that the most immediate triggers of haemostasis, ADP and ATP, are themselves products of degranulating platelets and neutrophils, representing a positive feed back system (shown by the double arrows in Fig. 2). Notice also that the agonists TXA2, NE, 5-HT, PAF and LTB4, produced by platelets and neutrophils, are also self-activating agonists, representing another group of molecules in the positive feed back loop of platelet and neutrophil activation. Neutrophil proteases also represent a positive feed back loop because they activate platelet-derived chemokine precursors. With such potent feed back loops, how come a needle puncture does not lead to clotting and platelet aggregation of the whole vascular system? First, the ADP and ATP concentrations produced by injured cells decrease as one moves away from the injury site. Secondly, endothelial cells and smooth muscle cells convert platelet produced PGH2 to PGI2 which is a very potent platelet aggregation inhibitor and vasodilator. Third, the activated clotting enzymes all have a relatively brief life, as there are many endogenous inhibitors (such as anti-thrombin III, a2-macroglobulin, thrombomodulin, etc.) that regulate their activity, so clotting will stay active only in the vicinity of a strong promoter. This demonstrates the criticality of the initial mix of agonists and haemostasis mediators in the propagation of the haemostatic reaction, these critical agonists being targets of salivary components of haematophagous insects. 3.3

ANNOYING ITCHING

A vast literature exists on the annoying allergic effects of insect saliva (Peng and Simons, 2007; Hoffman, 2008). The salivary potion injected by BFA into the vertebrate host skin can induce cutaneous reactions that are immunological in


68 A

ATP

TF Collagen ADP

Neutrophil activation

CG Platelet aggregation Thrombin PPi Blood clotting

Salivary antigens Tryptase

PF-4 β-TG

PAF

NE TXA2 5HT ULvWF Vasoconstriction

LTB4 PSelec

Complement Mast cell activation degranulation C5a

CysLT H

Endothelium activation

Bk Edema B

H + ATP + 5-HT + BK + LTB4 =

Itch, pain

FIG. 2 (A) Simplified diagram of vertebrate haemostasis and inflammation and (B) agonists of pain. Abbreviations: 5-HT, serotonin; ADP, adenosine diphosphate; ATP, adenosine triphosphate; BK, bradykinin; b-TG, beta-thromboglobulin; C5a, anaphylatoxin; CG, cathepsin G; CysLT, cysteinyl leukotrienes; H, histamine; LTB4, leukotriene B4; NE, norepinephrine; PAF, platelet-activating factor; PF-4, platelet factor 4; PPi, inorganic pyrophosphate; PSelec, P-selectin; TF, tissue factor; TXA2, thromboxane A2; ULvWF, ultra-large von Willebrand factor. (Photo of Rhodnius prolixus feeding on the arm of Andrew Spielman taken by Philippe Rossignol).

nature and may vary, in different individuals, from no reaction to severe hypersensitivity allergic responses. Several factors, including host genetic background, immunological profile, age and history of exposure can contribute to type and intensity of the response. Most typically, the reaction to insect bites of humans and experimental animals evolves with repeated exposure and can be described as a five stage sequence of skin reactivity (Mellanby, 1946; Feingold, 1968; Lehane, 2005). Initially, after exposure to bites of naı¨ve hosts, no reaction is observed (stage I). This is followed by a delayed-type reaction that appears 12–24 h after the bite and consists of erythema, often associated with papules and pruritus (stage II). As exposure continues an immediate response causing erythema and itching appears within 15–20 min, and it is followed by the delayed response (stage III). This delayed reaction tends then to disappear leaving only an immediate response lasting for a couple of hours (stage IV). Finally, after repeated long-term exposure, the host may develop desensitization and no skin reaction is observed anymore (stage V). Desensitization to mosquito bites has been observed to occur in experimental conditions in rabbits and


69

humans (Peng and Simons, 1998). In natural conditions desensitization is also acquired, but it may take long time or never occur, and the underlying immune mechanisms are still not completely understood. These cutaneous reactions are associated, in humans and other vertebrates, to both humoral and cellular responses that result in circulating anti-saliva IgG and IgE antibodies. Understanding the mechanisms involved in these immune responses may be relevant both for the development of vaccines based on salivary antigens (Titus et al., 2006) and for the diagnosis and immunotherapy of severe allergic reactions (Peng and Simons, 2007). Within the limits of our chapter, aimed at the problems insects face while feeding, immediate-type hypersensitivities are of importance because they can add more mediators to the haemostasis reaction, thus drying up the source of the meal, and perhaps more importantly, by promoting host behavioural responses that can minimally disrupt feeding and maximally leading to the insect’s death. It is indeed quite amazing that some plump triatomine bug species inhabit the nests of insectivorous rodents, or bats, or lizards, from which they obtain their sole source of nourishment. Figure 2, above, lists two host immune-mediated mechanisms producing effects within minutes of antigen contact; these are of humoral (complement) and cellular (mast cell) natures. The complement system comprises a complex cascade of proteases that may be activated by four different pathways, the so-called classic (antibody dependent), alternative (involving C3 recognition of carbohydrate surfaces), the colectin pathway (also involving carbohydrates, such as mannan), and the recently discovered activation of C3 by thrombin, linking the clotting and the complement pathways (Paul, 2008). The final result of complement activation is the formation of a lytic complex on the surface of invading organisms, but it can also occur in pure soluble form, without such substrates, as for example with antigen–antibody complexes (dependent on the IgG subclass) or soluble polysaccharides and plasma. It is of our concern that during the proteolytic activation of complement the C3 and C5 components are hydrolysed producing the relatively small peptidic fragments C3a and C5a, which are potent activators of endothelial cells as well as mast-cell degranulators. Skin mast cells appear to have evolved with the sole purpose to make the life of a blood sucker miserable and dangerous, and perhaps to provide a vain attempt to protect their owners from infection of the various lethal or disabling vector borne pathogens. When specific IgE or other subclasses of IgG on the surface of mast cells meet a divalent antigen that can crosslink two Ig receptors, a signalling cascade starts that can degranulate the whole contents of the cell (Paul, 2008). This is quite an amazing amplification, as a single molecule can trigger an event that can cause a macroscopic effect. A single degranulating mast cell can be felt by itching and its resulting small urticaria is quite visible to the naked eye. In addition to histamine (H) mast cells also release proteases, including tryptase that can activate PAR on platelets and endothelium, and they also synthesize the arachidonate-derived cysteinyl leukotriene LTC4, which is metabolized to the


70

still active LTD4 and LTE4 leukotrienes. These leukotrienes are potent vasodilators, and additionally activate the endothelium, which is also redundantly activated by various other agonists described above, such as histamine, serotonin, bradykinin (BK), thrombin, cathepsin G, tryptase and anaphylatoxins. 3.4

. . . AND PAIN . . .

Pain and itching can be produced by direct stimulation of certain nerves, as well as by the action of several inflammation agonists which initiate a nerve impulse interpreted as pain or itch by the central nervous system. Among these agonists are BK, ATP, 5-HT, H and LTB4 all of which described above as mediators of haemostasis and inflammation (Fig. 2B) (Millan, 1999). 3.5

THE ATTACKED ENDOTHELIUM FIGHTS BACK

Activated endothelial cells expose on their surfaces P-selectin, and E-selectin, that increase the adhesion of neutrophils (Kumar et al., 2004), and also secrete ultra-high molecular weight von Willebrand factor (vWF), that can form heavy complexes functioning as platelet activators. Importantly, in the activated endothelium the connections between neighbouring endothelial cells are modified, allowing larger spaces between them. Fluid then flows from intra- to extravascular space creating oedema. Neutrophils go with the flow, following the chemical gradient created by the anaphylatoxins, LTC4 and LTB4 which are potent chemotactic substances. Neutrophil and platelet products can also activate resident macrophages to produce PGE2, a potent skin vasodilator. This process of endothelium activation and oedema formation takes a few minutes to occur, but is within the time frame of most insect blood feeders. 3.6

MICROBIOLOGICAL CONCERNS

To obtain blood, insects necessarily have to break their host’s skin, which is a barrier to pathogen infection. The opportunity arises for skin surface microbes to contaminate their host circulation, and the ingested blood meal. The insect digestive system may also be previously contaminated by microbes that might develop with the blood meal. Perhaps for this reason, a common finding in sialotranscriptomes of BFA is a plethora of proteins known to have antimicrobial activity.

4

Toward a longitudinal definition of the salivary components of blood-feeding insects

From the analysis of the several spitomes done so far (Table 3), some generalizations can be made regarding the composition of the salivary potion of BFI. These can be summarized in the following arbitrary categories: (a) enzymes;


71

(b) receptor antagonists; (c) physiological antagonists, with an important subcategory of vasodilators; (d) kratagonists; (e) protease inhibitors; (f ) antigen 5 proteins; (g) immunity-related products, including antimicrobial peptides; (h) anaesthetics and (i) the unexpected. 4.1

ENZYMES

Salivary enzymes can help haematophagy by destroying agonists of haemostasis and inflammation, by destroying the final products of clotting, such as fibrin, by activating their host plasminogen, by enlarging the feeding cavity in telmophagous insects and helping the spread of salivary components into the skin matrix, and by potentially interfering the signalling pathway of their hosts. 4.1.1

Apyrase

All mammalian BFI studied so far have large amounts of salivary apyrase (ATP diphosphohydrolase) activity. The apyrase reaction degrades both ATP and ADP to AMP and orthophosphate, thus inhibiting platelet aggregation. This activity was associated with the first papers describing platelet aggregation inhibitors in the saliva of Rhodnius and tsetse (Ribeiro and Garcia, 1980; Smith et al., 1980; Mant and Parker, 1981) and later was described in several other BFA (Ribeiro et al., 1984b, 1985b, 1986, 1989, 1990, 1991; Kerlin and Hughes, 1992; Cupp et al., 1993, 1995; Marinotti et al., 1996; Valenzuela et al., 1996; Cheeseman, 1998). Interestingly, lizard (Ribeiro et al., 1989) or birdfeeding (Ribeiro, 2000) species have very small amounts of the enzyme activity, which is in accordance with platelets being a mammalian invention that uses ADP as a main agonist; accordingly those insects selecting mammals on their menu were better off if they found ways of destroying it. Male mosquitoes, which do not blood feed, and mosquito species that lost their blood-feeding ability also have very little amount of salivary apyrase activity (Ribeiro et al., 1985b; Calvo et al., 2008). Mosquitoes and triatomine bugs of the Triatoma genus (but probably not Rhodnius genus) elected the 50 -nucleotidase family to take care of host ADP and ATP (Champagne et al., 1995b; Faudry et al., 2004; Sun et al., 2006). This appears also to be the case with tabanids, where a collagen-induced platelet aggregation inhibitor named chrysoptin is a member of the 50 -nucleotidase family, although the authors reported this protein, possibly erroneously, as a specific collagen receptor inhibitor (Reddy et al., 2000). Bed bugs and sand flies (Valenzuela et al., 1998, 2001b) selected the Cimex type of apyrase, then a novel protein family that was found later to exist ubiquitously in eukaryotes (Failer et al., 2002). Fleas appear to have elected the CD-39 family of nucleotidases (Andersen et al., 2007), although the evidence is so far circumstantial. Salivary apyrase are so abundant on these organisms that, unusually for enzymes, they are displayed as a strong protein band in SDS–PAGE experiments (Francischetti et al., 2002c;


72

Valenzuela et al., 2002b, 2004). The presence of different protein families to serve the apyrase function in BFA is an example of convergent evolution that occurred when unrelated BFI had to adapt to the menu change that occurred with the extinction of dinosaurs and irradiation of mammals, at 60 millions years ago (MYA). 4.1.2

Additional nucleotidases

In addition to apyrase, Lutzomyia longipalpis sand flies have been found to contain secreted 50 -nucleotidases (Charlab et al., 1999; Ribeiro et al., 2000b), thus further degrading AMP, the final product of the apyrase reaction, into adenosine. Also mosquito saliva carries the product of a second 50 -nucleotidase family member (Arca` et al., 1999; Lombardo et al., 2000; Ribeiro et al., 2004b, 2007) that may function either as an alternative apyrase or like in L. longipalpis, as secreted salivary 50 -nucleotidase. Adenosine is a potent vasodilator (Collis, 1989) and platelet aggregation inhibitor (Dionisotti et al., 1992). However, adenosine is also a powerful degranulator of mast cells and basophils (Lohse et al., 1987; Tilley et al., 2000), and perhaps for this reason, salivary adenosine deaminase has also been found in sialotranscriptomes of sand flies (Charlab et al., 1999; Valenzuela et al., 2004; Anderson et al., 2006; Kato et al., 2006), mosquitoes (Ribeiro et al., 2001; Valenzuela et al., 2002b) and tsetse (Li and Aksoy, 2000; Li et al., 2001). In one case, the recombinant enzyme from Phlebotomus duboscqi was produced and its kinetics characterized (Kato et al., 2007). The function of this enzyme may be related to conversion of adenosine to inosine, which is a weaker agonist of the adenosine receptor (Tilley et al., 2000). Perhaps because inosine is still a mast-cell agonist, Aedes aegypti contains large amounts of a salivary purine hydrolase, which hydrolyses inosine to hypoxanthine and ribose (Ribeiro and Valenzuela, 2003), thus completely destroying any remaining purinergic activity. No such activity was found in Culex quinquefasciatus, Anopheles gambiae or Lutzomyia longipalpis sand flies. 4.1.3

Peroxidase

So far a true vasodilator agonist was not yet found in anopheline mosquitoes. However, their saliva inhibits norepinephrine-induced aortic ring contractions with a peroxidase enzyme that destroys catecholamines (Ribeiro and Nussenzveig, 1993; Ribeiro et al., 1994; Ribeiro and Valenzuela, 1999). Norepinephrine is responsible to keep more than 95% of human skin capillaries normally closed, thus destruction of norepinephrine may bring about vasodilatation. In the presence of hydrogen peroxide, this peroxidase can also destroy 5-HT but it is not clear whether this is physiologically relevant, since the concentration of H2O2 is low in tissues, but could be high near activated neutrophils. This enzyme is not found in Aedes or in Culex sialomes.


4.1.4

73

PAF hydrolases and phospholipases

Two insects so far have salivary enzymes that hydrolyse PAF. A phospholipase C activity-inhibiting PAF-induced platelet aggregation was found in saliva of C. quinquefasciatus (but absent of Ae. aegypti and An. gambiae) (Ribeiro and Francischetti, 2001), and a PAF acetylhydrolase was found in cat flea salivary gland homogenates (Cheeseman et al., 2001). This flea enzyme also displays an esterase activity. None of the enzyme activities have been molecularly characterized. In Ae. aegypti, an esterase was demonstrated to be secreted in saliva from both males and female mosquitoes (Argentine and James, 1995), suggesting it is not related to obtaining a blood meal. Its biological substrate remains unknown. cDNAs coding for diverse phospholipases with signal peptide indicative of secretion have been found in insect sialotranscriptomes, but none have been functionally characterized so far. Among these phospholipases are included phospholipase A2, which is known to have pro-inflammatory effects in the skin, with the production of free arachidonic acid (AA) and lysophosphatidylcholine (LPC). Most tissues have cyclooxygenases that can convert AA to PGH2, but the final product of the unstable PGH2 depends on the tissue, as indicated above. Mammalian skin has a PGD2 synthase that converts PGH2 to PGD2, a potent vasodilatory and platelet aggregation inhibitor (Ujihara et al., 1988; Braun and Schror, 1992; Morrow et al., 1992; Warren et al., 1994). Phospholipases can also have haemolytic activity. In ticks, a salivary phospholipase A2 activity has been identified and proposed to be important in the lysis of erythrocytes (Bowman et al., 1997; Zhu et al., 1997, 1998). Interestingly, saliva of Rhodnius prolixus contains LPC which is proposed to have an antihaemostatic function (Golodne et al., 2003), but LPC is also a haemolysin, as its name testifies. 4.1.5

Inositol phosphatase

The sialotranscriptomes of the kissing bugs R. prolixus, T. infestans and T. brasiliensis revealed transcripts coding for a protein with high similarity to inositol phosphatases. This finding is so far unique among kissing bugs. We initially thought that the enzyme could represent the salivary apyrase of Rhodnius but ADPase activity of the recombinant protein was not found, but rather the predicted inositol phosphatase activity (Andersen and Ribeiro, 2006). Vertebrate homologues of this enzyme are important regulators of the inositol triphosphate signalling cascade, because the enzyme activity hydrolyses phosphoinositolphosphate(3,4,5)P3 (PIP3). Indeed the enzyme is an important regulator of the activity of cytokines and immune cells (Kalesnikoff et al., 2003; Kashiwada et al., 2007; Sly et al., 2007; Harris et al., 2008; Leung et al., 2008). There is one problem only, the salivary enzyme is on the wrong side, PIP3 is an intracellular mediator. We here raise the speculation that saliva of kissing bugs,


74

by its content of LPC may actually transfect cells with some of their salivary components, for example Rhodnius LPC may induce the salivary inositol phosphatase to become intracellular and inactivate platelets, mast cells and other immune cells. Indeed Triatoma infestans saliva is very soapy in nature, foaming prolifically if air is flowed in a solution containing it (unpublished observations). 4.1.6

Other hydrolases

Transcripts coding for hydrolases, in particular serine proteases, have been regularly found in sialotranscriptomes of BFA. These enzymes could have a role in degrading fibrin, or in activating plasminogen, a vertebrate protein that is proteolytically activated to an enzyme that has high specificity for fibrin, and is part of the self-regulation of the clotting process. Tabfiglysin from Tabanus yao is one of such BFI salivary enzymes shown to degrade fibrinogen, and possibly fibrin (Xu et al., 2008), an activity similar to that of tick metalloproteases (Francischetti et al., 2003). Bioinformatic analysis of many mosquito and Triatoma salivary serine proteases revealed enzymes similar to chymotrypsin, suggesting they could have a role as an elastase, or in degrading other components of the ECM. Support for a salivary function associated with ECM degradation is also found in the form of hyaluronidases that have been described in C. quinquefasciatus mosquitoes (but absent in Aedes or Anopheles), in sand flies, black flies, Culicoides and tabanids (Charlab et al., 1999; Ribeiro et al., 2000a; Cerna et al., 2002; Campbell et al., 2005; Volfova et al., 2008; Wilson et al., 2008; Xu et al., 2008). Except for Culex, which may be a vessel feeder, all other insects are strict pool feeders, indicating that telmophagy may promote evolution of enzymes that digest ECM constituents, with the benefit of creating an enlarged feeding cavity. Hyaluronidase decreases the viscosity of the ECM, thus favouring the spread or diffusion of salivary agonists in the host skin and, for example, favouring diffusion of vasodilators from the surface of the skin to the deeper precapillary sphincter in the arteriolar plexus. C. quinquefasciatus also possess a salivary endonuclease that was functionally characterized (Calvo and Ribeiro, 2006), and found to be abundantly secreted in saliva. Endonucleases may help reduce skin viscosity caused by DNA released from broken cells, and may additionally produce pharmacologically active DNA products. Transcripts coding for endonucleases have also been found in sand fly and tsetse sialotranscriptomes, but their recombinant forms have not yet been characterized. Ticks have a salivary dipeptidyl peptidase that is quite active on hydrolysing bradykinin (Ribeiro and Mather, 1998). The same enzyme or a different one is also capable of completely inactivating the anaphylatoxins (Ribeiro and Spielman, 1986). It is possible that similar enzymes may be found in insect saliva in the future. Sand fly sialotranscriptomes have revealed the presence of transcripts coding for pyrophosphatases with a signal secretion, but the enzymatic activity was


75

never reported. This enzyme could have a possible role in degrading the polyphosphates secreted by platelets, which activate blood clotting (Ruiz et al., 2004). Since this is quite a novel discovery in platelet physiology, the discovery that a sand fly cares about it would be a confirmation of an ancient role of platelet polyphosphates in mammalian haemostasis. The reader should be cautioned, though, that not all serine proteases found in sialotranscriptomes should have a blood-feeding role. They could represent proximal serine proteases associated with innate immunity, either the prophenoloxidase or the TEP cascades. This is reinforced in mosquitoes and other nematocera by the finding of other proximal members of this cascade, such as pattern recognition molecules of the ficolin and C-lectin families, for example (although some lectins of both types were found to be selectively expressed in adult female salivary glands suggesting a role in blood feeding) (Arca` et al., 2005, 2007; Ribeiro et al., 2007). Non-blood-feeding male mosquitoes and mosquitoes of the genus Toxorhynchites also have many of these salivary enzymes, but lack those associated with blood feeding such as the D7 proteins (Calvo et al., 2006b, 2008). 4.2

RECEPTOR ANTAGONISM AND PLATELET AGGREGATION INHIBITORS

Although receptor antagonists constitute an important chapter of pharmacological antagonists designed by humans, only one example was found so far in blood-sucking insects. The tripeptide Arg-Gly-Asp (RGD), flanked by cysteines, binds to integrin receptors on platelets that would otherwise bind fibrinogen and produce platelet aggregation. In the presence of RGD peptides, platelets cannot aggregate by any agonist, because fibrinogen cannot crosslink them. Although this type of domain is very prevalent in tick salivary proteins (Mans et al., 2008; Francischetti et al., 2009) and snake venom (Lu et al., 1993; Suehiro et al., 1996), in insects it has been recognized solely in tabanids, where a member of the CAP/antigen 5 protein family (tabinhibitin) acquired such a domain, possibly by exon shuffling. This RGD-containing antigen 5 protein is a potent platelet inhibitor, as expected (Xu et al., 2008). No such domain has been found so far in fleas, sand flies or bugs, although recently it has been found by transcriptome analysis of a New World anopheline mosquito (Anopheles darlingi) (Calvo et al., 2009b), within an ubiquitous mosquito protein family (30-kDa antigen/Aegyptin) that has not such domain in at least six other species studied in three genera. However, confirmation of An. darlingi anti-platelet activity of this protein is lacking. It is interesting to note that while plants produce a multitude of receptor antagonists in the form of alkaloids (small, basic, nitrogen-containing organic compounds), this is quite rare in insects. We can here speculate that plants are primarily ‘‘organic chemists’’, while insects, or animals, are ‘‘biochemists’’ in their pharmacological ‘‘schools’’. Indeed, plant genomes reveal a greater number of enzymes than found in animals, largely those associated with intermediary metabolism leading to a

76


‘‘combinatorial organic synthesis’’ function, of adaptive value in their biological warfare with ruminants, for instance, which would digest any potent peptidic protein produced by plants, a problem not relevant for BFI that inject proteins directly into the skin. 4.3

PHYSIOLOGICAL ANTAGONISTS, PRIMARILY VASODILATORS

Physiological antagonism is the process where one agonist is antagonized by another agonist of a different receptor that triggers a contrary or neutralizing signal transduction cascade. For example, serotonin and TXA2 contract vascular smooth muscle while PGI2 is a potent vasodilator. Accordingly, PGI2 is a physiological antagonist of serotonin (and vice versa). Of course all these physiological antagonists are themselves agonists of different receptors. Vasodilatory agonists are commonly found in the saliva of BFI. The variety of these compounds is enormous. From the small to the big, it starts with nitric oxide (NO) produced by bed bugs and Rhodnius (but not Triatoma), carried by different haem proteins that transport and stabilize the NO, named nitrophorins (NP). In Rhodnius, the NP is a member of the lipocalin family and in Cimex it is a truncated member of the inositol phosphatase family of enzymes. Continuing with small molecules, adenosine was identified as the main vasodilator in the Old World sand fly Phlebotomus papatasi and Phlebotomus perniciosus (Ribeiro et al., 1999; Katz et al., 2000; Ribeiro and Modi, 2001), but not in P. duboscqi (a West African species) or in the New World sand flies. Members of the New World sand fly genus Lutzomyia possess a unique protein, maxadilan, the most potent vasodilator known and an agonist of the PACAP receptor (Lerner et al., 1991; Lerner and Shoemaker, 1992; Moro and Lerner, 1997). Mosquitoes of the genus Aedes have a salivary vasodilator member of the tachykinin family, named sialokinin (Champagne and Ribeiro, 1994; Beerntsen et al., 1999), which act by stimulating the endothelial cells to produce NO. These peptides are not found in mosquitoes of the genus Culex or Anopheles. Tabanids (deer flies and horse flies) have a protein member of the Kazal family acting as a vasodilator (Takaćˇ et al., 2006; Xu et al., 2008), but its receptor is still undefined. Normally Kazal domain proteins are protease inhibitors, but the domain was hijacked in tabanids to serve another function. The black fly Simulium vittatum has a peptidic vasodilator named SVEP, which is of unique sequence. It activates ADP-dependent Kþ channels, possibly using the PACAP receptor as well (Cupp et al., 1998). Ticks produce abundant amounts of prostaglandins, specifically PGE2 and PGF2a in their saliva, potent skin vasodilating substances (Dickinson et al., 1976; Higgs et al., 1976; Ribeiro et al., 1992; Inokuma et al., 1994; Bowman et al., 1996). It is quite possible that insects also produce salivary PGs, but perhaps because collection of insect saliva is not always as easily done as with ticks, it remains undiscovered. Scrambling convergent evolution appears here to counteract the vasoconstrictor component of haemostasis using diverse vasodilatory compounds, acting on a variety of receptors.


77

Although we mentioned above that there can be no magic bullet to prevent haemostasis and inflammation, many vasodilators described in this section are closer to this ideal design, because many of them can also prevent platelet aggregation, when platelets have the same type, or similar receptors as the smooth muscle cell. This is the case with NO, for example, that not only is a powerful vasodilator but is also a platelet aggregation inhibitor, in both cases by activation of cytosolic guanylate cyclase, existing in both type of cells. Similarly, adenosine leads platelets to increase cytosolic cyclic AMP that prevents also platelets from aggregating. The same is true for PACAP agonists. These molecules that are simultaneously vasodilators and platelet aggregation inhibitors must be of a high benefit/cost in terms of adaptive value to their bearers. It will not be surprising if agonists targeting the receptors of the atrial natriuretic factor (ANF), vasoactive intestinal peptide (VIP), calcitonin generelated peptide (CGRP) and adrenomedullin (ADM) are discovered in the future, as these are potent vasodilators and platelet inhibitors. The pleiotropic effects of these vasodilators are not limited to smooth muscle cells and platelets. Immune cells are also susceptible to many of these agonists. The vasodilator maxadilan is known to be a potent suppressor of macrophage function (Soares et al., 1998a), as is PGE2 for dendritic cells (Sa´-Nunes et al., 2007), for example. 4.4

KRATAGONISTS

No textbook of pharmacology could forecast the unique type of agonist inhibition that we are finding over and over again in insects (and ticks). We are discovering salivary proteins that have very high affinity (usually on the low nM range) for 5-HT, H, NE, ADP, TXA2, LTB4 and cysteinyl leukotrienes. Notice that these agonists are all represented by double arrows in Fig. 2, providing an important role in the propagation of the haemostasis response. The first kratagonist was discovered accidentally. It is known for over 25 years that saliva of R. prolixus has anti-histaminic activity (Ribeiro, 1982). Twelve years later, having discovered the salivary nitrophorins of the same insect (in the plural because at the time we have purified four similar proteins from the saliva, having similar NO-carrying properties) (Champagne et al., 1995a), we became interested to verify their interactions with histamine, an imidazolic compound known to interact with haem proteins. To our surprise, not only histamine binding to the haem moiety dislodged the NO from the nitrophorin, but this binding was of very high affinity, larger than any imidazolic compound so far investigated with haem proteins (Ribeiro and Walker, 1994). Crystallization of NP-1 with histamine revealed very specific interactions of histamine not only with the haem iron, but with specific amino acids strategically located to clamp H in place (Andersen et al., 1998). Moreover, this histamine mopping effect accounted for all the anti-histaminic activity of Rhodnius saliva. At first sight, this type of inhibition appears to be of very ‘‘stupid design’’, because a

78


19,000-Da molecule is being synthesized to scavenge a 100-Da agonist. It does not make reasonable sense. However, the math works. Histamine can accumulate to 1 mM amounts in inflamed tissues, a concentration that saturates the H1 receptor. If we take in consideration that a fifth instar nymph of Rhodnius has 50 mg of salivary nitrophorins, and can inject 90% of it in a blood meal that amounts to 300 ml, the final concentration of NPs in the blood meal are 8 mM, more than sufficient to chelate inflammatory concentrations of histamine. The stupidity of the design is thus compensated by brute force, or a large mass. In the absence of a name for this kind of inhibition, we propose the name kratagonist to characterize it, from the Greek word representing to arrest or to seize. In addition to the nitrophorins exquisite histamine kratagonist function, other abundant lipocalins in Rhodnius bind 5-HT and catecholamines (Andersen et al., 2003), explaining also earlier work demonstrating anti-5-HT activity in Rhodnius saliva (Ribeiro, 1982). Yet another Rhodnius family of lipocalins, named RPAI, initially described as inhibitors of collagen-induced platelet aggregation was shown to be ADP kratagonists (Francischetti et al., 2000, 2002a). Their inhibition of collagen-induced aggregation happens because after exposure to collagen, platelets secrete ADP, and the combination of the collagen signal plus the ADP signal triggers aggregation. Removal of ADP aborts platelet aggregation by collagen. It appears that Rhodnius co-opted the lipocalin family of proteins (Andersen et al., 2005), which are known carriers of small compounds in metazoa (Flower et al., 2000; Schlehuber and Skerra, 2005), to exercise various kratagonist functions. Salivary lipocalins were also discovered and used by ticks as kratagonists of histamine, serotonin, NE, TXA2 and leukotrienes (Paesen et al., 1999, 2000; Sangamnatdej et al., 2002; Mans and Ribeiro, 2008a,b), so this represents another example of convergent evolution, as ticks and insects are over 400 MYA from a common ancestor, and do not share a common blood-feeding ancestor. In mosquitoes, D7 proteins are among the most abundant salivary components and the first member of the family was identified in the culicine mosquito Ae. aegypti (James et al., 1991). It was later shown that the D7 is a multi-gene family in An. gambiae (Arca` et al., 1999, 2002), and that it is widely spread in blood-feeding Nematocera (Valenzuela et al., 2002a). The D7 are distantly related to the insect odorant-binding protein (OBP) family that, similarly to lipocalins, functions in carrying small organic, usually hydrophobic substances (Hekmat-Scafe et al., 2000). However, the function of the D7 family was elusive, until their role as kratagonists was identified. The D7 family comes in two sizes, long and short, the long having two similar domains, possibly representing an exon duplication event. An. gambiae has five genes coding for short and three genes coding for long D7 proteins, the most abundantly expressed being members 1–4 of the short family, member 5 perhaps being a pseudogene (Arca` et al., 2005). The first two proteins of the long family are expressed at relatively low levels, the third hardly so. Recombinant forms of the abundant short D7 proteins of An. gambiae were shown to bind the biogenic


79

amines H, 5-HT and NE (Calvo et al., 2006a). Although most can bind all amines with high affinity, each has its preference for a particular agonist, a typical process of divergence of function following gene duplication. Interestingly, in Aedes the expression pattern of the family is opposite to that seen in Anopheles: the large D7 proteins are highly expressed, while the short ones are poorly expressed (Ribeiro et al., 2007). The carboxy domain of the large D7 protein of Aedes was shown to bind biogenic amines similarly to the short D7 proteins of Anopheles (Calvo et al., 2006a). The crystal structures of one long D7 protein of Aedes (Calvo et al., 2009a) and one short protein of Anopheles (Mans et al., 2007) have been obtained, showing that the D7 domain has two additional helices as compared to the OBP family. Solution of the crystal structure and their binding pockets revealed that the amino-terminal domain of the long D7 protein of Aedes could bind a hydrophobic compound. After testing several potential agonists using isothermal calorimetry, it was discovered that the N-terminal domain of the long Aedes D7 binds cysteinyl leukotrienes, making this protein actually to look intelligent in targeting mastcell agonists (Calvo et al., 2009a). In Anopheles, both domains of the long D7 proteins appeared also to be suitable to high affinity interaction with hydrophobic ligands. These discoveries also can explain the relative abundance of the proteins in the salivary glands: H and 5-HT accumulate and saturate their receptors at one order of magnitude above the levels required for TXA2 and LTs. According to the brute force needed with the kratagonist design, one needs 10 times less force, or protein mass, to be effective with TXA2 or LTs as compared to biogenic amines. Possibly for this reason, the long D7 of Aedes are in high concentrations, and the small D7 of Anopheles are in higher concentrations, these being the D7 that bind 5-HT and H. We still do not know the ligands of the short D7 of Aedes, but it will be probably lipidic ligands, possibly TXA2. Another ubiquitous protein family found in mosquitoes, as well as in other blood-sucking Nematocera, comprises the 30-kDa antigen/Aegyptin family. It is as abundant as D7 protein members, and in some mosquito species it is the most abundant salivary protein. It was first recognized as an important salivary antigen in Aedes (Simons and Peng, 2001), but its function was elusive. More recently, members of this family have been shown to be potent inhibitors of platelet aggregation by collagen (Calvo et al., 2007b; Yoshida et al., 2008). This protein interacts with collagen at high affinity, masking its sites that are recognized by platelet receptors, and also by vWF. Aegyptin family members thus illustrate that kratagonists are not limited to binding small agonists, but proteins far larger than the kratagonist itself. Following the discovery that kratagonists for biogenic amines may be widespread in BFA, we can hypothesize that a kratagonist function for 5-HT and H, or perhaps collagen should be found among the most abundant salivary proteins of arthropods, because of the ‘‘brute force’’ needed against these agonists. Unpublished results from Dr. Valenzuela’s laboratory appear to support the validity of this law: indeed, a sand fly salivary member of the Yellow family of


80

proteins, which constitutes the most intense protein band in SDS–PAGE analysis of salivary extracts, is a 5-HT kratagonist. As mentioned above salivary kratagonists against LT and TXA2 should be present at 1/10 or less of the concentration of kratagonists for H or 5-HT, because these agonists saturate receptors at 1/10 of the concentrations needed for 5-HT or H. A corollary of the idea of kratagonists is that this mechanism should be found abundantly within animals, perhaps related to immunity or neurobiology, from where the prototype genes must have been recruited to serve a salivary function. An approximation of the mechanism in immunity is the idea of ‘‘soluble receptors’’ for cytokines, which function as scavengers of cytokines. Soluble receptors appear following the proteolytically cleavage of cytokine receptors on the surface of cells that, when released to the environment, chelates and neutralizes cytokines that could otherwise interact with intact receptors and thus trigger the signalling cascade (Horuk et al., 1993; Tsimanis et al., 2005). Soluble receptors, accordingly, are fragmented receptors and not products of a different gene that evolved with the function of modulating the effect of a particular agonist. We propose that many regulators of cytokines and neuromediators may be promoted by specialized proteins adapted to be their endogenous kratagonists, and these may have not been discovered yet due to the ‘‘stupidity’’ of its mechanism, and scientist bias. In entomology, for instance, many of the juvenile hormone (JH)-binding proteins that were discovered, falsely, as the elusive JH receptor may actually be JH kratagonists that modulate JH availability in the vicinity of particular cells. 4.5

PROTEASE INHIBITORS

The clotting and complement cascades of vertebrates, as the name implies, represent a system of proteases with, in one end, a proximal protease acting on intermediate inactive protease precursors and, at the other end, a terminal protease that finally produces the effector molecule(s), thrombin in the case of clotting or the lytic complex in the case of complement. Because one enzyme produces another enzyme that produces another enzyme, with a single activated protein in the beginning of the process, millions can be produced at the end, thus the apt name of an amplification cascade. In these amplification steps there are many non-enzymatic proteins that serve as modulators of the system, either by activating it or by tuning it down. These multiple steps of control allow the system to go forward in certain states, or to be deactivated, preventing their unregulated and noxious spread throughout the body. Examples of nonenzymatic activators are factor VIII of the clotting cascade and factor B of the complement pathway, while endogenous inhibitors are, for example, antithrombin III of the clotting cascade and factor H of the complement system. Proteins of the serpin and Kunitz family serve as endogenous inhibitors of activation of these proteolytic cascades. Insects have their proteolytic cascade equivalent in the prophenoloxidase activation cascade, in their own clotting


81

cascade, and in the activation of thioester proteins (TEP) on pathogen surfaces, which is analogous to the colectin pathway of complement activation in vertebrates (Thiel, 2007). Insects also use regulators of these proteases, the serpin and Kunitz families of proteins being abundant in insect genomes, and known to play a role in regulation of insect proteolytic cascades (Kanost et al., 2004). For a review on blood clotting and complement cascades, the reader is directed to recent reviews and books (Colman et al., 2005; Paul, 2008). We add also to this section the proteases from neutrophils and mast cells as possible targets of disruption for inhibitors to be found in the saliva of BFI. The various proteases thus involved in clotting, complement activation and inflammation provide a variety of targets for BFI. Toward this goal, insects recruited traditional genes as well-adapted novel protease inhibitor families to be expressed in their saliva. While clotting inhibitors are ubiquitously described in the salivary gland homogenates of BFI for almost one century (Cornwall and Patton, 1914; Yorkee and Macfie, 1924; Metcalf, 1945), only a few have been so far characterized molecularly. Belonging to traditional families are the anti-factor Xa serpin of the mosquito Ae. aegypti (Stark and James, 1995, 1998), and tabkunin, the Kunitz anti-thrombin of T. yao (Xu et al., 2008). Transcripts coding for members of the Kunitz family of peptides have also been described in sialotranscriptomes of Culicoides sonorensis and the black fly Simulium vittatum (Table 3), but no recombinant protein has been characterized so far. Ticks have also co-opted the Kunitz family to serve as anti-clotting agents (Francischetti et al., 2002b, 2004; Lai et al., 2004). The remaining molecularly characterized salivary clotting inhibitors from insects are of non-traditional protease inhibitor families, or belong to completely novel families. Triabin and prolixin-S are anti-thrombin and anti-factor VIII/IXa lipocalins from Triatoma pallidipennis (Noeske-Jungblut et al., 1995; Fuentes-Prior et al., 1997) and Rhodnius prolixus (Hellmann and Hawkins, 1965; Ribeiro et al., 1995; Sun et al., 1996; Zhang et al., 1998; Isawa et al., 2000; Gudderra et al., 2005), respectively. These anti-clotting roles of lipocalins are unique, and may have evolved by accretion of domains (exon shuffling) or acquired by classical evolution of a particular existing protein domain. For example, in the case of R. prolixus the lipocalin NP-2 has a unique anti-clotting activity, but it is also a NO-carrying protein, and an H kratagonist. It is reasonable to assume that the ancient function of NP-2, similarly to the other nitrophorin paralogues, was to carry NO and bind histamine. These functions require a specific conformation inside the lipocalin barrel where the haem, NO and histamine are bound. Amino acid side chains outside the loops are free to evolve neutrally or perhaps fast to avoid host immunity, and eventually may acquire a new function. While serpins serve as an anti-clotting in Aedes mosquitoes (transcripts for serpins also being found in sialotranscriptomes of Aedes albopictus, the close related Ochlerotatus triseriatus (submitted) and in Culex quinquefasciatus), the salivary anti-clotting of Anopheles albimanus is a very negatively charged peptide uniquely found in anophelines, named anophelin. It functions by tightly binding to


82

thrombin (Francischetti et al., 1999; Valenzuela et al., 1999). Sialotranscriptomes of An. darlingi, An. gambiae, An. funestus and An. stephensi reveal abundant transcription of homologues (Arca` et al., 1999, 2005; Valenzuela et al., 2003; Calvo et al., 2004, 2007a). A novel and relatively short peptide ( 3 kDa) also serves as an anti-thrombin in tsetse (Cappello et al., 1998), and yet another, named simulidin in black flies (Abebe et al., 1995; Cupp and Cupp, 2000), although this may be a distant and truncated member of the D7 protein family. The muscid Haematobia irritans (the face fly) has a unique peptidic anti-thrombin, named thrombostasin (Zhang et al., 2002). The related stable fly Stomoxys calcitrans has a homologue, albeit its primary structure is only 41% identical. Activated factor XII of clotting activates pre-kallikrein to kallikrein, and this in turn acts on blood kininogens to produce BK, an inflammatory and painproducing peptide. Because the most important physiological activator of FXII is thrombin, inhibition of clotting should also decrease BK production. Nonetheless, saliva of anopheline mosquitoes possess two inhibitors of the FXII–kallikrein axis, one promoted by one member the short D7 protein from An. stephensi, which was named hamadarin (Isawa et al., 2002), and the other by a protein found specifically in anopheline mosquitoes, named anophensin (Isawa et al., 2007). Molecularly characterized salivary anti-complement activities have been described in soft and hard ticks (Valenzuela et al., 2000; Nunn et al., 2005); moreover, anticomplement activity has been reported in salivary homogenates of sand flies and triatomine bugs (Cavalcante et al., 2003) although their molecular nature has been so far elusive. Soon to be published results from Dr. Valenzuela’s lab will present some novel protein families with such a function on sand flies (J. G. Valenzuela, personal communication). Similarly, although ticks have Kunitz proteins inhibitory of tryptase (Paesen et al., 2007), no such activity has been so far discovered in insect saliva, although transcriptome analysis has revealed such protein families in biting midges and sand flies. No inhibitors of cathepsin G have been so far described in either insect of tick saliva (Table 2). From the above account, it can be observed that the nature of insect salivary anti-clotting is very diverse, even within insects sharing a common haematophagous ancestor, as is the case with Triatoma and Rhodnius or culicines and anopheline mosquitoes. We can also observe that we lack the knowledge of the anti-clotting nature for whole orders or families of BFA, such as the fleas, bed bugs, biting midges, lice, etc. Actually, some of these orders/families do not have a single functionally characterized protein. When considering that considerable diversity is found at the genus level, with 455 genera accounted for in Table 1, the extent of our ignorance on this subject becomes quite apparent. 4.6

ANAESTHETICS

Following nerve activation by pain agonists (ATP, H, 5-HT, NE and BK), the nervous signal is conduced from the periphery to the spinal chord by special neurons using a specific Naþ channel. Traditional anaesthetics block nerve

TABLE 2 Molecularly characterized salivary proteins of blood-feeding insects (excluding immunity related) Name Apyrase Cimex nitrophorin

Sialokinin

Apyrase

Biological action

Activity

Molecular family

Type

Naturea

Characterizationb

Organism

Family or order

References

Degrades ADP and ATP Carriers of NO

Anti-platelet

Cimex type

Enzyme

1

1, 2, 3, 4, 5

Cimex lectularius

Cimicidae

Vasodilator, anti-platelet

Kratagonist

5

1, 2, 3, 4, 5

Cimex lectularius

Cimicidae

Endotheliumdependent vasodilator Degrades ADP and ATP

Vasodilator

Inositolphosphate phosphatase, truncated Tachykinins

Agonist

4

1, 2, 3, 4, 5

Aedes aegypti

Culicidae

Anti-platelet

50 -nucleotidase

Enzyme

1

1, 2, 3, 4, 5

Aedes aegypti

Culicidae

Purine nucleosidase

Enzyme

1

1, 2, 3

Aedes aegypti

Culicidae

Ribeiro and Valenzuela (2003)

Serpin

Protease inhibitor

2

1, 2, 3, 4, 5

Aedes aegypti

Culicidae

Stark and James (1998) Calvo et al. (2007b) and Yoshida et al. (2008) Ribeiro and Valenzuela (1999)

Anti-Xa serpin

Factor Xa inhibitor

Destroys mastcell degranulating adenine and inosine Anti-clotting

Aegyptin

Collagen inhibitor

Anti-platelet

Unique

Kratagonist

2

1, 2, 3, 5

Aedes aegypti, Anopheles stephensi

Culicidae

Peroxidase

Degrades catecholamines

Vasodilatory

Peroxidase

Enzyme

1

1, 2, 3, 4, 5

Anopheles albimanus

Culicidae

Purine Degrades adenine nucleosidase to hypoxanthine

Notes

Valenzuela et al. (1998) Valenzuela and Ribeiro (1998) Champagne and Ribeiro (1994) Champagne et al. (1995b)

So far found only in Aedes aegypti Several other mosquito species have been enzymatically characterized to have salivary apyrase activity

Found in other culicines, not anophelines Family pervasive in blood-sucking Nematocera Found only in anophelines, not culicines

(continues)

TABLE 2 (Continued) Name

Biological action

Activity

Molecular family

Type

Nature

a

Characterization

b

Organism

Family or order

References

Anophelin

Thrombin inhibitor

Anti-clotting

Unique

Protease inhibitor

2

1, 2, 3, 4, 5

Anopheles albimanus

Culicidae

D7 large

Sequesters leukotrienes

Antiinflammatory

D7–OBP superfamily

Kratagonist

5

1, 2, 3, 5

Anopheles gambiae

Culicidae

D7 small

Sequesters biogenic amines

Vasodilatory, antiinflammatory

D7–OBP superfamily

Kratagonist

5

1, 2, 3, 5

Anopheles gambiae

Culicidae

Hamadarin

Factor XII inhibitor Factor XII inhibitor Degrades doublestranded DNA

Anti-bradykinin

Kratagonist

2

1, 2, 3, 5

Anopheles stephensi

Culicidae

Anti-bradykinin

D7–OBP superfamily gSG7

2

1, 2, 3, 5

Anopheles stephensi

Culicidae

Spreading factor

Endonuclease

Protease inhibitor Enzyme

1

1, 2, 3, 5

Culex quinquefasciatus

Culicidae

Hyaluronidase

Degrades skin matrix components

Spreading factor

Hyaluronidase

Enzyme

1

1, 2, 3, 4, 5, 6

Tabanus yao

Diptera

Xu et al. (2008)

Adenosine deaminase

Converts adenosine to inosine

Destroys mastcell degranulating adenosine Anti-clotting

Adenosine deaminase

Enzyme

1

1, 2, 3

Glossina morsitans

Glossinidae

Li and Aksoy (2000)

Unique

2

1, 2, 3, 4, 5

Glossina morsitans

Glossinidae

Anti-clotting

Thrombostasin

3

1, 3, 5

Haematobia irritans

Muscidae

Binds Ig

Antigen 5

Protease inhibitor Protease inhibitor Kratagonist?

1, 2, 3, 4

Stomoxys calcitrans

Muscidae

Cappello et al. (1998) Zhang et al. (2002) Ameri et al. (2008)

Anophensin Endonuclease

Tsetse thrombin Thrombin inhibitor inhibitor Thrombostasin Anti-thrombin Stomoxys Ag5

Unknown target (complement activation?)

Francischetti et al. (1999) and Valenzuela et al. (1999) Calvo et al. (2009a) Calvo et al. (2006a) and Mans et al. (2007) Isawa et al. (2002) Isawa et al. (2007) Calvo and Ribeiro (2006)

Notes Found only in anophelines, not culicines

Family pervasive in blood-sucking Nematocera Family pervasive in blood-sucking Nematocera

Not found in culicines Not found in Aedes or Anopheles genera Activity also found in black flies, sand flies, tabanids, Culex quinquefasciatus

Charlab et al. (2000), Ribeiro et al. (2001) and Kato et al. (2007) Lerner et al. (1991)

Adenosine deaminase

Converts adenosine to inosine

Destroys mast-cell degranulating adenosine

Adenosine deaminase

Enzyme

1

1, 2, 3

Aedes aegypti, Phlebotomus duboscqi

Nematocera

Maxadilan

PACAP receptor agonist

Vasodilator

Unique

Agonist

4

1, 2, 3, 4, 5

Lutzomyia longipalpis

Psychodidae

50 -nucleotidase

Degrades AMP to adenosine

Produces vasodilatory and antiplatelet adenosine Anti-platelet

50 -nucleotidase

Enzyme

1

1, 2

Lutzomyia longipalpis

Psychodidae

Ribeiro et al. (2000b)

Cimex type

Enzyme

1

1, 2, 3, 5

Psychodidae


Nucleotides

Agonist

4

1, 3, 4, 5

Phlebotomus papatasi Phlebotomus papatasi, P. argentipes

Valenzuela et al. (2001b) Ribeiro et al. (1999)

LysophospInhibits platelet hatidylcholine aggregation Rhodnius Carriers of NO Nitrophorins

Anti-platelet

Lipid

Agonist

4

1, 3, 4, 5

Rhodnius prolixus

Reduviidae


Kratagonist

5

1, 2, 3, 4, 5

Rhodnius prolixus

Reduviidae

Inositol Degrades inositol phosphatase phosphates

Unknown

Lipocalin of the nitrophorin family SHIP

Enzyme

1

1, 2, 3, 5

Rhodnius prolixus

Reduviidae

RPAI

Sequesters ADP

Anti-platelet

Kratagonist

3

1, 2, 3, 5

Rhodnius prolixus

Reduviidae

BABP

Anti-serotonin, anti-adrenergic Inhibits factor VIII

Vasodilatory

Lipocalin of the triabin superfamily Lipocalin

Kratagonist

5

1, 2, 3, 4, 5

Rhodnius prolixus

Reduviidae

Protease inhibitor

2

1, 2, 3, 4, 5

Rhodnius prolixus

Reduviidae

Reduces nerve sodium current Degrades sialic acid

Anesthetic

Lipocalin of the nitrophorin family Unknown

?

1

Triatoma infestans

Reduviidae

Dan et al. (1999)

Anti-neutrophil?

Unknown

Channel blocker Enzyme

1

1

Triatoma infestans

Reduviidae

Amino et al. (1998)

Apyrase

Degrades ADP and ATP Adenosine, AMP Agonist of purinergic receptors

Prolixin-S

Triatoma anesthetic Sialidase

Anti-clotting

Psychodidae

Golodne et al. (2003) Andersen et al. (2005) Andersen and Ribeiro (2006) Francischetti et al. (2000, 2002a) Andersen et al. (2003) Ribeiro et al. (1995)

Protein not found in Old World sand fly genera

Not found in Lutzomyia, not found in P. duboscqi

Multi-gene family

At least two members in Rhodnius

Also carries NO and binds histamine

(continues)

TABLE 2 (Continued) Name Apyrase Triapsin

Triplatin

Pallidipin

Biological action Degrades ADP and ATP Serine protease

Possible GPVI antagonist or ADP binder Possible ADP binding

Activity

Molecular family

Type

Naturea

Characterizationb

Organism

Family or order

References

Anti-platelet

50 -nucleotidase

Enzyme

1

1, 2, 3, 4, 5

Triatoma infestans

Reduviidae

Unknown – matrix degradation? Anti-platelet

Trypsin

Enzyme

1

1, 2, 3, 4, 5

Triatoma infestans

Reduviidae

Lipocalin of the triabin superfamily Lipocalin of the triabin superfamily Lipocalin of the triabin superfamily Ricin lectin family

Kratagonist

3

1, 2, 3, 5

Triatoma infestans

Reduviidae

Morita et al. (2006)

Kratagonist

3

1, 2, 3, 5

Triatoma pallidipennis

Reduviidae

Noeske-Jungblut et al. (1994)

Protease inhibitor

3

1, 2, 3, 5

Triatoma pallidipennis

Reduviidae

Noeske-Jungblut et al. (1995)

Agonist

4

1, 2, 3, 4, 5

Simulium vittatum

Simuliidae

Cupp et al. (1998)

Protease inhibitor

2

1, 2, 3, 4, 5

Simulium vittatum

Simuliidae

Abebe et al. (1995)

Enzyme

1

1, 2, 3

Xenopsylla cheopis

Siphonaptera

Anti-platelet

Faudry et al. (2004) Amino et al. (2001)

Triabin

Thrombin inhibitor

Anti-clotting

SVEP

Possible PACAP receptor agonist Anti-thrombin

Vasodilator

Degrades ADP and ATP Degrades ADP and ATP Possible ion channel effect

Anti-platelet

Possible D7 distant member CD-39

Anti-platelet

50 -nucleotidase

Enzyme

1

1, 3, 4

Chrysops spp.

Tabanidae

Vasodilator

Kazal

4?

1, 2, 3, 4, 5

Tabanidae

2

1, 2, 3, 4, 5, 6

Hybomitra bimaculata, Tabanus yao Tabanus yao

Tabanidae

Andersen et al. (2007) Reddy et al. (2000) Takaćˇ et al. (2006) and Xu et al. (2008) Xu et al. (2008)

3

1, 2, 3, 4, 5, 6

Tabanus yao

Tabanidae

Xu et al. (2008)

1

1, 2, 3, 4, 5, 6

Tabanus yao

Tabanidae

Xu et al. (2008)

1

1, 2, 3, 4, 5, 6

Tabanus yao

Tabanidae

Xu et al. (2008)

Simullidin

Apyrase Chrysoptin

Anti-clotting

Vasotab, Vasotab TY Tabkunin

Anti-thrombin

Anti-clotting

Kunitz

Tabinhibitin

RGD disintegrin

Anti-platelet

Antigen 5

Tabserin

Serine protease

Trypsin

Tabfiglysin

Serine protease

Anti-clotting, target unknown Fibrinogenolytic

Agonist, receptor unknown Protease inhibitor Receptor antagonist Enzyme

Trypsin

Enzyme

a b

Notes

Not found in Rhodnius

1, enzymatic; 2, enzyme or target inhibitor; 3, receptor antagonist; 4, receptor agonist; 5, kratagonist. 1, enzymatic activity or bioassay; 2, candidate found in sialotranscriptome; 3, proteomic or mass confirmation; 4, purified compound confirms activity; 5, recombinant/synthetic compound confirms activity.


87

conduction by preventing nerve impulse conduction, and thus the pain signal from reaching the brain. Saliva of T. infestans blocks nerve conduction in the sciatic nerve and the activity of Naþ channels in cell cultures (Dan et al., 1999), but it is still molecularly uncharacterized. No other aesthetic of BFA has been reported so far. Notice that the efficient removal of pain agonists with enzymes, protease inhibitors and kratagonists may prevent pain from being elicited, making an aesthetic not necessary. 4.7

ANTIGEN (AG5) FAMILY MEMBERS

All BFA sialotranscriptomes analysed so far, including those from ticks, indicate the expression of one or more members of the antigen 5 protein family, which was originally described in the venom of ants and wasps (King and Spangfort, 2000). Insect genomes encode several divergent members of this widely spread protein family that is part of the larger superfamily of cysteine-rich extracellular proteins ubiquitously found in animals and plants (Schreiber et al., 1997; Megraw et al., 1998). The Drosophila melanogaster genome encodes 25 different Ag5 genes (Kovalick and Griffin, 2005) and 19 family members sharing 32–51% amino acid identity can be identified in the An. gambiae genome. At least one member of the Ag5 family was recruited by haematophagous insects to play some essential blood-feeding role as suggested by the abundant and tissue-enriched expression in salivary glands (Arca` et al., 2005, 2007; Ribeiro et al., 2007). Members of this family in snake venoms are associated with ion channel inhibitors and toxins (Yamazaki et al., 2003; Yamazaki and Morita, 2004). In Conus snails, an Ag5 protein was found to have a specific protease activity (Milne et al., 2003), and as seen above, a protein of this family acquired a RGD domain and acts as a salivary inhibitor of platelet aggregation (Xu et al., 2008). Recently, a recombinant Ag5 protein expressed in the salivary glands of the stable fly S. calcitrans was found to bind immunoglobulins strongly, particularly their Fc fragment (Ameri et al., 2008). The action of this protein on the activation of the classical pathway of complement was not described. Besides the Tabanus and Stomoxys proteins, no other salivary Ag5 member from BFA has been so far functionally characterized. It is difficult to predict the function of these proteins in the saliva of BFA because of the very diverse functions found for members of this protein family. 4.8

IMMUNITY-RELATED PRODUCTS

A common feature of insect sialotranscriptomes is the presence of AMP such as defensins, cecropins, gambicin and lysozyme, as well as pattern recognition molecules (Gram-negative binding proteins, C-type lectins and ficolins) and serine proteases that may act as proximal activators of the prophenoloxidase or TEP cascades. Many AMP are lytic to microbes, and some can also act as eukaryotic cytolysins. The salivary trialysin of T. infestans is of a unique family that can


88

lyse bacteria and protozoa (Amino et al., 2002; Martins et al., 2006). Recently, novel AMP were isolated from the saliva of T. yao (Xu et al., 2008). Saliva of BFA must have a large number of still unknown antimicrobial families, an area virtually untapped by the scientific community. 4.9

THE UNEXPECTED

As will be discussed in more detail below, the sialome studies from different genera reveals many proteins that are unique to the genus, and this includes whole families of related proteins that are exclusively found expressed in the salivary glands of BFA. Most of these protein families have no known function, but may function as members of several generic categories above, such as physiological antagonists (agonists), kratagonists, enzymes, antimicrobials or protease inhibitors.

5

Salivary diversity

It has become apparent within the last decade that molecular evolution of genes coding for salivary proteins in BFA is at a very fast pace as compared to housekeeping genes. For example, when comparing several anopheline mosquitoes of the subgenus Cellia with the African mosquito An. gambiae, for which there is available a genome draft that allows reasonable identification of the orthologs, it is verified that orthologous salivary proteins have much less identity than housekeeping ones. For the Asiatic An. stephensi the % identity ( SE) is 62.4 15 and 93.1 6 (Valenzuela et al., 2003) and for the African An. funestus the values are 66.7 1.9 and 96 0.84 for salivary gland and housekeeping proteins, respectively (Calvo et al., 2007a). These differences are highly significant in both cases. In another example, species within the Rhodnius genus are very similar and differentiation is morphologically feasible only by looking at the male genitalia characters (Lent and Wygodzinsky, 1979), but the electrophoresis pattern of the salivary gland proteins are very distinct among these morphologically very similar species (Soares et al., 1998b, 2000). Few studies have tackled the issue of polymorphism of the salivary gland genes from wild populations, but in one such instance the vasodilatory maxadilan from L. longipalpis was found to be polymorphic (Lanzaro et al., 1999). Interestingly, all forms studied were equally powerful vasodilators but had different antigenic properties. Antibodies against a particular form neutralized the vasodilatory effect of that form, but not of all forms (Milleron et al., 2004), suggesting that the gene variation may represent a form of antigenic variation, and that variation could be maintained in the fly population by balanced polymorphism through frequency-dependent selection. However, it should be pointed out that high degree of intra-specific conservancy of salivary proteins was found when Phlebotomine sand flies from two different geographic locations were


89

compared (Kato et al., 2006), suggesting that intra-specific variation (both within and between populations) deserves a deeper evaluation with extension of the analysis to other insect species. It should be noted that the invariance in the salivary proteins of P. papatasi may be related to the exceptionally strong DTH response following this fly’s bite, a response that facilitates feeding, thus making advantageous to the fly to conserve its antigenicity (Belkaid et al., 2000). When more distantly related species are compared, such as mosquitoes belonging to different subfamilies, qualitative, not quantitative genetic differences are manifested. As indicated above, the anti-clotting function in Ae. aegypti saliva is provided by a member of the serpin family, and serpins have also been found in sialotranscriptomes of Ae. albopictus and C. quinquefasciatus, both mosquitoes belonging to the Culicine subfamily. However, in the anopheline subfamily a completely novel polypeptide, named anophelin, serves this function. Why is this the case? While anophelines and culicines share many unique protein families, such as the 30-kDa antigen/Aegyptin and D7 proteins, suggesting the ancestral mosquito ( 150 MYA) already possessed these salivary proteins, could it be that this ancestor did not have an anti-clotting and its acquisition occurred only after the split of the lineages? While this could be the case, another possibility is that in the early anophelines the anti-clotting may have become ineffective, creating the conditions for the evolution of an alternative gene to serve the lost function. In an analogous manner to the anti-maxadilan antibodies, host immunity may have neutralized the ancestral anopheline anti-clotting peptide. This is supported by the author’s unpublished observation that salivary gland homogenates of Ae. aegypti show no anti-clotting activity if the assay is done with the plasma of previously exposed Guinea pigs or humans, but not when plasma of naive animals are used. The vertebrate host immune pressure may thus be a powerful driving force for the fast evolution of the salivary potion, not only by conventional, point mutational modes (as seems to be occurring with maxadilan), but also by promoting complete gene loss of function, the opportunity arising for a completely new gene recruitment to occupy the vacuum (as may have been the case with anopheline’s anti-clotting proteins). It is apparent that the inflammatory, haemostatic and immune pressure from hosts are driving the evolution of the salivary magic potion of BFA. Adaptive host immunity may prevent reaching a ‘‘perfect’’, immutable potion, creating the engine for the fast, perpetual evolution of these molecules, or at best reaching a balanced polymorphism state where genes are under frequency-dependent selection (the gene with the lower frequency has the best fitness, because it codes for a rare epitope). Short lived hosts, may well favour a balanced polymorphism, while long lived hosts, such as humans, may well lead to BFA gene function loss. For example, mice and other small rodents live on average less than 1 year in the wild, equivalent to less than 10 generations of sand flies. Because newborn mice do not inherit the memory of sand fly antigens circulating in their parent’s generation, a gene driven to low frequency in the end of the previous year may become a better gene when a new cohort of mice arises.


90

Conversely, humans and other mammals and birds can accumulate their immunological memory over decades, representing hundreds of sand fly or mosquito generations, which could lead to local extinction of alleles in a finite population. In this case a gene could lose its function, and the genome will have to scramble for a new solution. The still unanswered question is to what extent the BFA salivary proteins are driving the evolution of vertebrate inflammatory, haemostatic and immune systems? If this is the case, we have a true scenario of coevolution of vertebrates and BFA. Coevolution is not the mere fact of evolving together, but rather that characters of one organism influence another organism to change, and this in turn changes the characters of the first organism, as is the coevolution of pollinators and flowers. The case of the evolution of mammalian skin mast cells appears to be one such case in point. In the absence of BFA and the burden of the diseases that they transmit, mast cells appear to be of negative fitness value, as they are effectors of many disabling dysfunctions. Extrapolating from our scope on insects, the evolution of basophils, which are very similar to mast cells, appear to have been the response of some mammals to prevent ticks from feeding (Allen, 1973), by their ability to congregate in much larger numbers than mast cells and thus release large amounts of their mediators. One of the characteristics of coevolution and interspecies arms race is phenotypic exaggeration, such as the elaborate flowers and bird beak adaptation, for example (Thompson, 2005; Hanifin et al., 2008). The load of histamine/ serotonin granules in a mast cell and the large amounts of H/5-HT kratagonists in insects are indicative of these coevolutionary exaggerations. Under this light, the evolution of the BFA salivary magic potion and vertebrate skin is represented by an arms race scenario resulting from coevolving organisms.

6

The evolutionary scramble

Gene duplication is responsible for more than 50% of the genes in eukaryote genomes (Holland, 1999; Ohno, 1999; Sankoff, 2001; Mazet and Shimeld, 2002; Nei and Rooney, 2005), and it has been proposed as an important mechanism for the evolution of salivary genes in ticks (Mans et al., 2002). Sialotranscriptome analysis of >20 species (Table 3) so far studied indicates that this mechanism is surely occurring in BFI, not only with genes belonging to ubiquitous families, but also within genes uniquely found associated with particular BFI families or subfamilies, such examples being found in all papers associated with Table 3. As mentioned above, the D7 protein family has eight genes in Anopheles, and is polygenic in all Nematocera so far studied. There are two salivary serpins in culicines, one of which is a factor Xa inhibitor, as indicated above, the other being now identified as a tryptase inhibitor (M. Kotsyfakis, personal communication). Perusal of the papers cited in Table 3 will demonstrate many other examples, including several families so

TABLE 3 Sialotranscriptomes of blood-feeding insects Order

Sub-order

Family

Tribe

Subgenus

Species

References Francischetti et al. (2002c), Arca` et al. (2005) and Calvo et al. (2006b) Valenzuela et al. (2003) Calvo et al. (2007a) Calvo et al. (2004, 2009b) Valenzuela et al. (2002b) and Ribeiro et al. (2007) Arca` et al. (2007) In process Ribeiro et al. (2004b)

Diptera

Nematocera

Culicidae

Anophelini

Cellia

Anopheles gambiae

Diptera

Nematocera

Culicidae

Anophelini

Cellia

Diptera Diptera

Nematocera Nematocera

Culicidae Culicidae

Anophelini Anophelini

Cellia Nyssorhynchus

Anopheles stephensi Anopheles funestus Anopheles darlingi

Diptera

Nematocera

Culicidae

Culicini

Stegomyia

Aedes aegypti

Diptera Diptera Diptera

Nematocera Nematocera Nematocera

Culicidae Culicidae Culicidae

Culicini Culicini Culicini

Stegomyia Culex Culex

Diptera

Nematocera

Culicidae

Toxorhynchitini

Diptera

Nematocera

Psychodidae

Diptera

Nematocera

Psychodidae

Aedes albopictus Culex tarsalis Culex quinquefasciatus Toxorhynchites amboinensis Lutzomyia longipalpis Phlebotomus ariasi

Calvo et al. (2008) Charlab et al. (1999) Oliveira et al. (2006) (continues)

TABLE 3 (Continued) Order

Sub-order

Family

Tribe

Diptera

Nematocera

Psychodidae

Diptera

Nematocera

Psychodidae

Diptera

Nematocera

Psychodidae

Diptera

Nematocera

Psychodidae

Diptera

Nematocera

Ceratopogonidae

Culicoidini

Diptera Hemiptera Hemiptera

Nematocera

Simuliidae Reduviidae Reduviidae

Rhodnini Triatomini

Reduviidae

Triatomini

Hemiptera Siphonaptera

Subgenus

Culicoides

Species Phlebotomus papatasi Phlebotomus duboscqi Phlebotomus argentipes Phlebotomus perniciosus Culicoides sonorensis Simulium vittatum Rhodnius prolixus Triatoma infestans Triatoma brasiliensis Xenopsylla cheopis

References Valenzuela et al. (2001a) Kato et al. (2006) Anderson et al. (2006) Anderson et al. (2006) Campbell et al. (2005) Andersen et al. (2009) Ribeiro et al. (2004a) Assumpçaõ et al. (2008) Santos et al. (2007) Andersen et al. (2007)


93

far not mentioned because their function is still unknown. One of the immediate outcomes of gene duplication is an increased dosage of the gene product, because there are more templates available for translation. Following this ‘‘saltatory’’ evolutionary event, traditional mutations can both change the promoter region of the gene, changing tissue expression levels of the message, as well as allowing functional divergence, as exemplified above by the specialization of the anopheline short D7 proteins regarding their different affinities to biogenic amines (Calvo et al., 2006a). While the normal process of gene duplication is through segmental chromosome duplication creating tandem duplications (Eichler and Sankoff, 2003), a more unusual process is by genomic insertion of reverse transcribed mRNA by the machinery of transposable elements. While the vast majority of eukaryotic genes contain two or more exons separated by introns, these novel retrotransposed genes are uniexonic. Although this mechanism of gene duplication is rare, analysis of the sialogenomes (the organization of the salivary genes in the organism’s genome) of the mosquitoes An. gambiae (Arca` et al., 2005), Ae. aegypti (Ribeiro et al., 2007) and C. quinquefasciatus (unpublished observations) indicates the presence of many uniexonic genes coding for salivary proteins, including novel protein families larger than 60 kDa, a rare occurrence of single exon genes in metazoa. This is the case with the following families: SGS/Wolbachia-like genes and 56-kDa family found in the three mosquitoes, and the 62- and 34-kDa families of Ae. aegypti (not found in Anopheles). In Culex, the 16.8-kDa family (Ribeiro et al., 2004b) has more than 30 genes, most of which are single exonic (unpublished observations), suggesting its genomic expansion by retrotransposition, and perhaps later gene duplication of uniexonic copies. Horizontal gene transfer (HGT), or lateral gene transfer, is an unusual mechanism of novel gene acquisition in eukaryotes (Andersson, 2005; Keeling and Palmer, 2008), although relatively common in prokaryotes (Jain et al., 2002; Gogarten and Townsend, 2005). Interestingly, horizontal transfer in eukaryotes is more common in organisms that have intracellular endosymbionts, or organisms associated with particular ecological adaptations (Keeling and Palmer, 2008), the two conditions being true for most blood-sucking insects and ticks (Wigglesworth, 1936; Hill et al., 1973; Nogge, 1978; Shaw and Moloo, 1991; Hypsa, 1993; Ricci et al., 2002; Sinkins, 2004; Aksoy and Rio, 2005; Mattila et al., 2007; Pais et al., 2008). Remarkably, the mosquito salivary families SGS/ Wolbachia and the 56-kDa family (found in Aedes, Culex and Anopheles), as well as the 62- and 34-kDa families (found only in culicines) and the exclusive 16.8kDa family of Culex represent uniexonic protein families matching bacterial proteins when compared to the GenBank non-redundant (NR) protein database using the tool Psiblast (Altschul et al., 1997), suggesting their acquisition by HGT into the mosquito genomes. We have mentioned above several instances of salivary genes coding for novel proteins. The label ‘‘novel’’ is imposed because comparison of the deduced primary sequence of these proteins to available databases, such as the

94


NR database of GenBank (now amounting to over 6 million proteins), fails to retrieve matches with convincing statistical values of significance. It is possible that these ‘‘novel’’ genes originated from an ancestor of a recognizable family, but that their fast evolution led to obscuring any residual primary sequence similarity. This is the case, for instance with the salivary nitrophorins from Rhodnius, which give only limited sequence similarity to other salivary triatomine proteins when their primary structure is used to find similar proteins. However, their crystal structure revealed unequivocally that nitrophorins belonged to the lipocalin superfamily (Andersen et al., 1997, 1998; Weichsel et al., 1998), which is a well known and very diverse protein family (Flower et al., 2000). Similarly, the D7 proteins produce strong primary similarities only to other BFI proteins, but have motifs identifiable by the tool Rpsblast (Marchler-Bauer et al., 2002) as belonging to the OBP family. Crystal structure of a D7 protein allowed identification of two extra helices in D7 proteins, in addition to the normal six helices of canonical OBPs (Mans et al., 2007), confirming its origin as an OBP, possibly evolving by exon duplication. More recently, the crystal structure of SVEP, the salivary vasodilator of S. vittatum, revealed that this protein belongs to a classical trefoil structure found in lectins, such as ricin and many other bacterial proteins (J.F. Andersen, personal communication). The primary structure of SVEP does not reveal any relatives when compared to the NR database. The solution of the crystal structure of these novel proteins will certainly help to classify many of these ‘‘novel’’ proteins into already known families, or help to confirm then as real novel structures.

7

On the odd, the paradoxical, the bizarre and the bias

Although many BFA studied to date have salivary anti-histaminic activity, there are reports of the existence of histamine in the salivary glands and saliva of black flies (Wirtz, 1988, 1990). Although confirmation by mass spectrometry would be desirable, it is possible that some BFI actually use this inflammatory amine to assist haematophagy. Histamine is known to have both vasoconstrictory and vasodilatory properties, depending on the nature of the vascular bed (Carroll and Neering, 1976; Kelm et al., 1993). In conjunction with other salivary components, histamine may constitute an odd ingredient for these flies magic potion. Histamine may also be recruited to the feeding site by active mast-cell degranulation, as shown to occur with saliva of the mosquito An. stephensi in non-immune mice (Demeure et al., 2005). Another group of flies that might exploit host inflammatory and immune reactions are sand flies of the Phlebotomus genus. Their saliva can rapidly induce strong delayed-type hypersensitivity in mammals. Flies feeding in these DTH skin sites feed much faster, in accordance with their higher blood flow, as measured by a laser Doppler apparatus (Belkaid et al., 2000). As mentioned above, P. papatasi is also odd in its recruitment of AMP and adenosine to their salivary glands, in


95

opposition to the reverse strategy of many other BFI that specifically destroy adenosine. Because the mediators of haemostasis, inflammation and immunity are many, and the possible manner to antagonize them is still higher (there are many ways of antagonizing each agonist), the product resulting from many independent and bifurcating evolutionary paths for the creation of the salivary potion may attain odd and apparently paradoxical compositions. While transcriptome studies point to putative secreted proteins largely by comparison of their primary sequence with known databases, and by the existence of a leader sequence indicative of secretion (Nielsen et al., 1997), proteomics studies done with saliva (not salivary gland homogenates) of Ae. aegypti and An. gambiae (Orlandi-Pradines et al., 2007) indicated the presence of salivary peptides deriving from the cleavage of membrane proteins, specifically fragments from a mosquito-exclusive protein family postulated to be a salivary gland receptor for Plasmodium parasites (Korochkina et al., 2006). It is possible that these proteins are expressed in the membrane of secretory vacuoles from where they are processed and make their way to the secreted saliva. Perhaps due to the power of the ‘‘omics’’ revolution, this chapter is biased to the description and findings of bioactive polypeptides as components of the salivary potion of BFI. We have mentioned, however, several examples of nonpeptidic components such as NO, adenosine, AMP and LPC that are pharmacologically active compounds found in the saliva of bugs and flies. Advances in mass spectrometry, reaching fentomol levels of detection, make it ripe right now to investigate the presence of small agonists in the saliva of BFA. Because many of these compounds, such as prostaglandins or other eicosanoids, are produced during salivation and thus not previously stored, experimental collection of insect saliva should be used instead of salivary gland homogenates.

8

Measuring the size of our ignorance

Biology has reached an interesting point in its history that makes it quite different from other natural sciences. Genomics and transcriptomic analysis can give us a good measure of what is there to learn. Imagine if in physics there was an experimental way to measure the number of atomic sub-particles that remain undiscovered! Physicists are thus unaware of their ignorance’s extent but biologists can now have a measure of it, or at least part of it, with some precision. For example, the genome of An. gambiae has approximately 16,000 genes coding for proteins, and humans have less than 10,000 more. Within these 15,000–30,000 genes found in eukaryotes, there is a substantial proportion coding for proteins that fit in the class of conserved hypothetical proteins (Galperin and Koonin, 2004), representing a measure of our ignorance. Even when we have an idea of the biochemical function for a protein, indicated by its sequence signatures, such as those for serine protease, or a G-coupled receptor, we may still not know when and where the protein is expressed and its role in


96

the organism’s physiology, thus our physiological ignorance can be considered greater than that of the biochemist. Sialotranscriptomic analyses done so far are giving us a measure of our ignorance, or how much more we need to learn (Tables 3 and 4). Of the 21 sialotranscriptomes listed in Table 3, nine derive from mosquitoes, including two subfamilies, four genera and five subgenera. Additionally, there are six transcriptomes from sand flies (two genera) and one transcriptome each from a biting midge and a black fly. Mosquitoes, biting midges, black flies and sand flies belong to the Diptera sub-order Nematocera, which, perhaps except for sand flies, may have had a common blood-feeding ancestor (Grimaldi and Engel, 2005). Three transcriptomes are from triatomine bugs, including two different tribes (Rhodnini and Triatomini). A single flea transcriptome has been done and one more is on the pipeline, from two different genera. With this limited data set, we have found (references on Table 3) that insects have from 70 to 150 putative salivary secreted proteins, as indicated by polypeptides containing a signal secretory leader sequence (Nielsen et al., 1997), and excluding those of obvious endoplasmic reticulum function. Of these nearly 100 secretory products, we can assign the function of less than 20 proteins per transcriptome. Most transcriptomes listed in Table 3 have been done by randomly sequencing 1000–2000 clones from salivary gland cDNA libraries. It has been our experience that, perhaps due to the relatively low complexity of the salivary gland proteome (when compared to organs like the mammalian liver or brain, for example), 1000–2000 sequenced clones are enough to display the majority of the sialome. Indeed, 2000 clones sequenced from an Ae. aegypti salivary gland cDNA library (Ribeiro et al., 2007) discovered virtually all those found in another effort that obtained 20,000 sequences (Nene et al., 2007). A similar situation was encountered with the An. gambiae sialome, where 4000 sequences identified basically all those in a large sequencing effort, also 20,000 sequences done by the Pasteur Institute (Arca` et al., 2005). Sequencing of normalized libraries, or more intensive sequencing of existing libraries, perhaps with newer generation of pyrosequencing methodology, may increase the number of salivary secreted peptides, but possibly to no more than 10% of TABLE 4 Relationship between number of known blood-feeding insect species and sialotranscriptome knowledge

Orders Family Genera Species

From Table 1

From Table 3

Percentage analysed

5 17 455 14,555

3 6 11 21

60.00 35.29 2.42 0.14


97

the number found with random sequencing of 2000 clones. It should be indicated, however, that these additional low abundance transcripts may account for pharmacologically potent peptides. If we use the rough estimate of 100 polypeptides as the ballpark number for the constitution of the BFI’ salivary potion, how much of these vary from one organism to another, and what is the extent of the universe of the BFI’ salivary proteins, or using the word Ben Mans coined, the sialoverse? First of all, there is virtually no molecular overlap between the magic potions from one order of insects when compared to another order. The sialome of Rhodnius is completely different from that of mosquitoes (as indicated above, even their apyrase activities derive from different enzyme families), or any other non-triatomine listed in Table 1, so we can assign 200 for the number of unique proteins in the sialoverse of triatomine bugs and mosquitoes. Table 1 lists 17 BFI families, five of which are from Nematocera that may have evolved to haematophagy from a common ancestor, sand flies excepted (Grimaldi and Engel, 2005). From these considerations we can number 14 families that evolved independently to haematophagy, and thus a rough maximum number of 1400 unique polypeptides. This number may be somewhat reduced, but not to less than one-half. For example, if we compare sand flies with mosquitoes and black flies, the majority of the sand fly protein families are still unique, or if related, they are very divergent. As mentioned above, sand flies have members of the D7 protein family, which are quite distinct from those of mosquitoes (Valenzuela et al., 2002a) and black flies (Andersen et al., 2009). Sand flies also recruited proteins from the Yellow family to their sialome, not found in any other blood-feeding Nematocera studied so far. But they recruited the Cimex type of apyrase, in complete distinction of their Nematoceran relatives. For the possible overlaps, we can lower our estimate for the family level sialoverse to 1400/2 ¼ 700 unique polypeptides. What is the sialome overlap between same family organisms belonging to different genera, or even at subgenus level? Mosquito sialotranscriptomes of three different genera (excluding the non-blood-feeding Toxorhynchites amboinensis) (Table 3) indicate the presence of several unique protein families belonging to each genus. For example, in Anopheles we mentioned earlier the unique anopheline proteins anophelin, anophensin and the salivary peroxidase, but we here may add the unique families SG1 (a protein family with four members) and SG6, all with unknown function (Arca` et al., 2005). Aedes and Culex share several protein families unique to their culicine subfamily level, namely the 62, 41.9, 34, 27, 23.4, 8.9, basic 3.8 and basic 7.6-kDa families (Ribeiro et al., 2004b, 2007; Arca` et al., 2007), thus eight families of unknown function not found in anophelines, or any other organism. C. quinquefasciatus sialome, in its turn, has one large unique tryptophan-rich protein family, with over 30 genes, and 14 additional putative secreted polypeptides with no known function (Ribeiro et al., 2004b). Similarly, comparison between Ae. albopictus and Ae. aegypti both from the same subgenus Stegomyia reveals eight novel

98


peptides in Ae. albopictus, including four that are of a novel multi-gene family (Arca` et al., 2007). Comparison of Rhodnius and Triatoma sialomes, insects from different tribes, reveals various non-overlapping protein families, including the unique nitrophorin family of haem-binding lipocalins carriers of NO, found only in Rhodnius, a multi-gene family coding for at least 20 different proteins (Ribeiro et al., 1998, 2004a). Although members of the triabin lipocalin superfamily are expressed in both Rhodnius and Triatoma, they are very divergent, and many are functionally distinct, as exemplified above with the T. pallidipennis salivary anti-thrombin named triabin that gave its name for this lipocalin family (Noeske-Jungblut et al., 1995; Fuentes-Prior et al., 1997; Glusa et al., 1997). Rhodnius does not have a salivary anti-thrombin, and instead has a lipocalin of the nitrophorin family that blocks the Xase complex of clotting (Gudderra et al., 2005). These considerations indicate that sialome novelty can be reasonably expected between two different genera of the same blood-feeding insect family, or even within different species of the same genus. It is not unreasonable at this point to extrapolate that we should find at least two novel proteins per genus of BFI, or, according to Tables 1 and 4, another 890 novel proteins from 445 genera, reaching a limit of 890 þ 650(700) ¼ 1500 proteins within the BFI sialoverse. On the other hand, if we presume there is a single novel salivary protein in every species of BFI, we should expect the sialoverse to contain near 15,000 proteins. We have to be careful, though, with the extrapolation above regarding the number of different genera and species. First, we have many knowledge gaps at high levels, including nearly one-half of the orders and families still unexplored (Table 4). Only 2.4% of the genera listed in Table 1 have been studied (Table 4). It is clear that two mosquito genera that have a common ancestor at >150 MYA, as are Aedes and Anopheles (Krzywinski et al., 2006) should be quite divergent, as their sialomes bifurcated at the time dinosaurs were still irradiating, and nearly 90 million years before mammal irradiation. Two more closely related genera, such as Ochlerotatus and Aedes (Shepard et al., 2006) should provide more overlapping sialomes, as they probably share a common ancestor within the time frame of mammalian irradiation. Taxonomists are classified into lumpers and splitters, lumpers being those that aggregate many species into a single genus. Observation of Table 1 may lead to the conclusion that flea taxonomists, Miriam Rothschild in particular, may be accused of being a splitter (Rothschild, 1975). One-half of all genera listed in Table 1 indeed derive from the Siphonaptera. However, the Siphonaptera is the youngest of the haematophagous orders, and are thought to have irradiated with the mammals and birds to which they are closely associated. In other words, if a rat flea should be in the same genus as a rabbit flea, rats and rabbits should be, nonsensically, promoted to the same genus. What may be more indicative of the true level of salivary diversity, but the data are not there yet, is to determine how many genera are at least 50 million years apart, a time estimate that would lead as much time to evolve the sialome as it did to evolve mammals.


8.1

99

A FORECAST OF THE COSTS AND TIME REQUIRED FOR ACQUIRING SIALOME WISDOM

If we consider a low estimate number for the sialoverse as consisting of 1000 proteins, and that for each protein it takes the effort of one scientist for the period of 2 years, including its recombinant expression and functional evaluation, an estimate of 2000 scientist years are to be expected to do this job. A group of 10 scientists accordingly would take 200 years and 100 scientists would take 20 years to this effort. These numbers are clearly an underestimate of the work ahead. Our experience in these novel proteins indicate that they end up generating many more publications than the single one used in the calculation. For example, papers with maxadilan in their title now number 34, nitrophorin number 35, five have D7 in their title and relate to mosquitoes, ixolaris appears in four titles and triabin in three. It is thus possible that 1 in 10 salivary proteins become the target of further pharmacological/vaccine/epidemiological investigations, leading each selected case to the additional production of dozens of additional papers, thus at least doubling the amount of research time needed, to over 40 years for a group of 100 scientists. Prioritization of the type of research to be pursued will probably follow the haphazard motions of scientist initiated research, unless organized departments or institutes decide to attack the subject in a coordinated way, as has been done, for example, with the study of snake venoms. Many generations of scientists will be necessary to decipher the magic potion of BFA, and to collect its fruits.

9

Salivary antigens: Epidemiological tools?

While we may speculate on the size of the task ahead to decipher the magic potion of BFI, a more immediate use of this knowledge is becoming available to help evaluate parameters associated with the transmission of vector borne diseases. Indeed the host immune response to arthropod saliva may also be used as an epidemiological marker of exposure to vector bites. This may represent a very useful serological indicator of risk of disease transmission by a given vector and a convenient tool for surveillance and for evaluation of vector control interventions. After the first evidence that anti-tick saliva antibodies could be a useful epidemiological marker of tick exposure and Lyme disease risk (Schwartz et al., 1990, 1991), several additional reports described the possible use of saliva or salivary extracts as potential markers of exposure. This is the case for sand fly vectors of Leishmania (Barral et al., 2000), triatomine vectors of Chagas disease (Nascimento et al., 2001), tsetse vectors of African trypanosomiasis (Van Den Abbeele et al., 2007) and for mosquito vectors (Trevejo and Reeves, 2005; Trevejo et al., 2005; Remoue et al., 2006; Waitayakul et al., 2006; Orlandi-Pradines et al., 2007). These studies established the interesting principle that the host immune response to BFA saliva

100


may represent a reliable indicator of vector density and, therefore, an indirect measure of disease risk, not only at population but also at individual level. However, as we just learned, the salivary potion of BFI is a complex cocktail consisting both of species-shared and of species-specific components and therefore cross-reactivity may significantly contribute to the observed response. This potential cross-reactivity of salivary antigens from different BFI may represent an advantage, for example for vaccine development (Mejia et al., 2006) or for potential use in immunotherapy of hypersensitivity reactions (Peng and Simons, 1997; Peng et al., 1998); however, on the contrary, it may be misleading if the response to salivary antigens is to be used for epidemiological studies. Proof of principle that individual recombinant salivary antigens may be efficient substitutes for saliva or salivary extracts has been obtained for different disease vectors such as ticks (Sanders et al., 1998, 1999), sand flies (Barral et al., 2000), tsetse (Caljon et al., 2006) and mosquitoes (Poinsignon et al., 2008; Lombardo et al., 2009). In this respect, the BFI sialotranscriptomes available so far (Table 3) started revealing the complexity of the sialoverse, providing detailed information on the salivary repertoires of important disease vectors, as mosquitoes and sand flies, and offering the opportunity to carefully select species- or genus-specific antigens for recombinant expression and for the development of novel epidemiological tools for improved vector and disease control (Billingsley et al., 2006). Acknowledgements This work was supported by the Intramural Research Program of the Division of Intramural Research, National Institute of Allergy and Infectious Diseases, National Institutes of Health. We are grateful to Drs. Michalis Kotsyfakis and Babis Savakis for independently suggesting the name kratagonist. Because JMCR is a government employee and this is a government work, the work is in the public domain in the United States. Notwithstanding any other agreements, the NIH reserves the right to provide the work to PubMedCentral for display and use by the public, and PubMedCentral may tag or modify the work consistent with its customary practices. You can establish rights outside of the US subject to a government use license. References Abebe, M., Cupp, M. S., Champagne, D. and Cupp, E. W. (1995). Simulidin: a black fly (Simulium vittatum) salivary gland protein with anti-thrombin activity. J. Insect Physiol. 41, 1001–1006. Abramson, D. I. (1989). Dermal blood vessels and Lymphatics. In: Handbook of Experimental Pharmacology: The Pharmacology of the Skin, Vol. 1 (eds Greaves, M. W. and Shuster, S.), pp. 89–116. Springer-Verlag, Berlin.


101

Adler, S. and Theodor, O. (1926). The mouthparts, alimentary tract and salivary apparatus of the female Phlebotomus papatasi. Ann. Trop. Med. Parasitol. 20, 109–142. Aksoy, S. and Rio, R. V. (2005). Interactions among multiple genomes: tsetse, its symbionts and trypanosomes. Insect Biochem. Mol. Biol. 35, 691–698. Allen, J. R. (1973). Tick resistance: basophil in skin reactions of resistant guinea pigs. Int. J. Parasitol. 3, 195–200. Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W. and Lipman, D. J. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402. Ameri, M., Wang, X., Wilkerson, M. J., Kanost, M. R. and Broce, A. B. (2008). An immunoglobulin binding protein (antigen 5) of the stable fly (Diptera: Muscidae) salivary gland stimulates bovine immune responses. J. Med. Entomol. 45, 94–101. Amino, R., Porto, R. M., Chammas, R., Egami, M. I. and Schenkman, S. (1998). Identification and characterization of a sialidase released by the salivary gland of the hematophagous insect Triatoma infestans. J. Biol. Chem. 273, 24575–24582. Amino, R., Tanaka, A. S. and Schenkman, S. (2001). Triapsin, an unusual activatable serine protease from the saliva of the hematophagous vector of Chagas’ disease Triatoma infestans (Hemiptera: Reduviidae). Insect Biochem. Mol. Biol. 31, 465–472. Amino, R., Martins, R. M., Procopio, J., Hirata, I. Y., Juliano, M. A. and Schenkman, S. (2002). Trialysin, a novel pore-forming protein from Saliva of haematophagous insects activated by limited proteolysis. J. Biol. Chem. 277, 6207–6213. Andersen, J. F. and Ribeiro, J. M. C. (2006). A secreted salivary inositol polyphosphate 5-phosphatase from a blood-feeding insect: allosteric activation by soluble phosphoinositides and phosphatidylserine. Biochemistry 45, 5450–5457. Andersen, J. F., Champagne, D. E., Weichsel, A., Ribeiro, J. M. C., Balfour, C. A., Dress, V. and Montfort, W. R. (1997). Nitric oxide binding and crystallization of recombinant nitrophorin I, a nitric oxide transport protein from the blood-sucking bug Rhodnius prolixus. Biochemistry 36, 4423–4428. Andersen, J. F., Weichsel, A., Balfour, C. A., Champagne, D. E. and Montfort, W. R. ˚ resolution: transport of nitric (1998). The crystal structure of nitrophorin 4 at 1.5 A oxide by a lipocalin-based heme protein. Structure 6, 1315–1327. Andersen, J. F., Francischetti, I. M. B., Valenzuela, J. G., Schuck, P. and Ribeiro, J. M. C. (2003). Inhibition of hemostasis by a high affinity biogenic amine-binding protein from the saliva of a blood-feeding insect. J. Biol. Chem. 278, 4611–4617. Andersen, J. F., Gudderra, N. P., Francischetti, I. M. B. and Ribeiro, J. M. C. (2005). The role of salivary lipocalins in blood feeding by Rhodnius prolixus. Arch. Insect Biochem. Physiol. 58, 97–105. Andersen, J. F., Hinnebusch, B. J., Lucas, D. A., Conrads, T. P., Veenstra, T. D., Pham, V. M. and Ribeiro, J. M. C. (2007). An insight into the sialome of the oriental rat flea, Xenopsylla cheopis (Rots). BMC Genomics 8, 102. Andersen, J. F., Pham, V. M., Meng, Z., Champagne, D. E. and Ribeiro, J. M. C. (2009). Insight into the sialome of the black fly, Simulium vittatum. J. Proteome Res. 8, 1474–1488. Anderson, J. M., Oliveira, F., Kamhawi, S., Mans, B. J., Reynoso, D., Seitz, A. E., Lawyer, P., Garfield, M., Pham, M. and Valenzuela, J. G. (2006). Comparative salivary gland transcriptomics of sandfly vectors of visceral leishmaniasis. BMC Genomics 7, 52. Andersson, J. O. (2005). Lateral gene transfer in eukaryotes. Cell. Mol. Life Sci. 62, 1182–1197. Arca`, B., Lombardo, F., de Lara Capurro, M., della Torre, A., Dimopoulos, G., James, A. A. and Coluzzi, M. (1999). Trapping cDNAs encoding secreted proteins

102


from the salivary glands of the malaria vector Anopheles gambiae. Proc. Natl. Acad. Sci. USA 96, 1516–1521. Arca`, B., Lombardo, F., Lanfrancotti, A., Spanos, L., Veneri, M., Louis, C. and Coluzzi, M. (2002). A cluster of four D7-related genes is expressed in the salivary glands of the African malaria vector Anopheles gambiae. Insect Mol. Biol. 11, 47–55. Arca`, B., Lombardo, F., Valenzuela, J. G., Francischetti, I. M. B., Marinotti, O., Coluzzi, M. and Ribeiro, J. M. C. (2005). An updated catalogue of salivary gland transcripts in the adult female mosquito, Anopheles gambiae. J. Exp. Biol. 208, 3971–3986. Arca`, B., Lombardo, F., Francischetti, I. M. B., Pham, V. M., Mestres-Simon, M., Andersen, J. F. and Ribeiro, J. M. C. (2007). An insight into the sialome of the adult female mosquito Aedes albopictus. Insect Biochem. Mol. Biol. 37, 107–127. Argentine, J. A. and James, A. A. (1995). Characterization of a salivary gland-specific esterase in the vector mosquito, Aedes aegypti. Insect Biochem. Mol. Biol. 25, 621–630. Assumpçaõ, T. C., Francischetti, I. M. B., Andersen, J. F., Schwarz, A., Santana, J. M. and Ribeiro, J. M. C. (2008). An insight into the sialome of the blood-sucking bug Triatoma infestans, a vector of Chagas’ disease. Insect Biochem. Mol. Biol. 38, 213–232. Baines, S. (1956). The role of the symbiotic bacteria in the nutrition of Rhodnius prolixus. J. Exp. Biol. 33, 533–541. Ba¨nziger, H. (1970). The piercing mechanism of the fruit-piercing moth Calpe [Calyptra] thalictri Bkh. (noctuidae) with reference to the skin-piercing bloodsucking moth C. eustrigata Hmps. Acta Trop. 27, 54–88. Ba¨nziger, H. (1975). Skin-piercing blood-sucking moths I: ecological and ethological studies on Calpe eustrigata (Lepid., noctuidae). Acta Trop. 32, 125–144. Ba¨nziger, H. (1979). Skin-piercing blood-sucking moths II: studies on a further 3 adult Calyptra [Calpe] sp. (Lepid., Noctuidae). Acta Trop. 36, 23–37. Barral, A., Honda, E., Caldas, A., Costa, J., Vinhas, V., Rowton, E. D., Valenzuela, J. G., Charlab, R., Barral-Netto, M. and Ribeiro, J. M. C. (2000). Human immune response to sand fly salivary gland antigens: a useful epidemiological marker? Am. J. Trop. Med. Hyg. 62, 740–745. Beerntsen, B. T., Champagne, D. E., Coleman, J. L., Campos, Y. A. and James, A. A. (1999). Characterization of the Sialokinin I gene encoding the salivary vasodilator of the yellow fever mosquito, Aedes aegypti. Insect Mol. Biol. 8, 459–467. Belkaid, Y., Valenzuela, J. G., Kamhawi, S., Rowton, E., Sacks, D. L. and Ribeiro, J. M. C. (2000). Delayed-type hypersensitivity to Phlebotomus papatasi sand fly bite: an adaptive response induced by the fly? Proc. Natl. Acad. Sci. USA 97, 6704–6709. Billingsley, P. F., Baird, J., Mitchell, J. A. and Drakeley, C. (2006). Immune interactions between mosquitoes and their hosts. Parasite Immunol. 28, 143–153. Borkent, A. (2008). The frog-biting midges of the World (Corethrellidae: Diptera). Zootaxa 1804, 1–456. Bowman, A. S. and Sauer, J. R. (2004). Tick salivary glands: function, physiology and future. Parasitology 129, S67–S81. Bowman, A. S., Dillwith, J. W. and Sauer, J. R. (1996). Tick salivary prostaglandins: presence, origin and significance. Parasitol. Today 12, 388–396. Bowman, A. S., Gengler, C. L., Surdick, M. R., Zhu, K., Essenberg, R. C., Sauer, J. R. and Dillwith, J. W. (1997). A novel phospholipase A2 activity in saliva of the lone star tick, Amblyomma americanum (L.). Exp. Parasitol. 87, 121–132. Braun, M. and Schror, K. (1992). Prostaglandin D2 relaxes bovine coronary arteries by endothelium-dependent nitric oxide-mediated cGMP formation. Circ. Res. 71, 1305–1313.


103

Braverman, I. M. (1997). The cutaneous microcirculation: ultrastructure and microanatomical organization. Microcirculation 4, 329–340. Brossard, M. and Wikel, S. K. (2004). Tick immunobiology. Parasitology 129, S161–S176. Caljon, G., Van Den Abbeele, J., Stijlemans, B., Coosemans, M., De Baetselier, P. and Magez, S. (2006). Tsetse fly saliva accelerates the onset of Trypanosoma brucei infection in a mouse model associated with a reduced host inflammatory response. Infect. Immun. 74, 6324–6330. Calvo, E. and Ribeiro, J. M. C. (2006). A novel secreted endonuclease from Culex quinquefasciatus salivary glands. J. Exp. Biol. 209, 2651–2659. Calvo, E., Andersen, J., Francischetti, I. M. B., de Lara Capurro, M., de Bianchi, A. G., James, A. A., Ribeiro, J. M. C. and Marinotti, O. (2004). The transcriptome of adult female Anopheles darlingi salivary glands. Insect Mol. Biol. 13, 73–88. Calvo, E., Mans, B. J., Andersen, J. F. and Ribeiro, J. M. C. (2006a). Function and evolution of a mosquito salivary protein family. J. Biol. Chem. 281, 1935–1942. Calvo, E., Pham, V. M., Lombardo, F., Arca`, B. and Ribeiro, J. M. C. (2006b). The sialotranscriptome of adult male Anopheles gambiae mosquitoes. Insect Biochem. Mol. Biol. 36, 570–575. Calvo, E., Dao, A., Pham, V. M. and Ribeiro, J. M. C. (2007a). An insight into the sialome of Anopheles funestus reveals an emerging pattern in anopheline salivary protein families. Insect Biochem. Mol. Biol. 37, 164–175. Calvo, E., Tokumasu, F., Marinotti, O., Villeval, J. L., Ribeiro, J. M. C. and Francischetti, I. M. B. (2007b). Aegyptin, a novel mosquito salivary gland protein, specifically binds to collagen and prevents its interaction with platelet glycoprotein VI, integrin a2b1, and von Willebrand factor. J. Biol. Chem. 282, 26928–26938. Calvo, E., Pham, V. M. and Ribeiro, J. M. C. (2008). An insight into the sialotranscriptome of the non-blood feeding Toxorhynchites amboinensis mosquito. Insect Biochem. Mol. Biol. 38, 499–507. Calvo, E., Mans, B. J., Ribeiro, J. M. C. and Andersen, J. F. (2009a). Multifunctionality and mechanism of ligand binding in a mosquito antiinflammatory protein. Proc. Natl. Acad. Sci. USA 106, 3728–3733. Calvo, E., Pham, V. M., Marinotti, O., Andersen, J. F. and Ribeiro, J. M. C. (2009b). The salivary gland transcriptome of the neotropical malaria vector Anopheles darlingi reveals accelerated evolution of genes relevant to hematophagy. BMC Genomics 10, 57. Campbell, C. L., Vandyke, K. A., Letchworth, G. J., Drolet, B. S., Hanekamp, T. and Wilson, W. C. (2005). Midgut and salivary gland transcriptomes of the arbovirus vector Culicoides sonorensis (Diptera: Ceratopogonidae). Insect Mol. Biol. 14, 121–136. Cappello, M., Li, S., Chen, X., Li, C. B., Harrison, L., Narashimhan, S., Beard, C. B. and Aksoy, S. (1998). Tsetse thrombin inhibitor: bloodmeal-induced expression of an anticoagulant in salivary glands and gut tissue of Glossina morsitans morsitans. Proc. Natl. Acad. Sci. USA 95, 14290–14295. Carroll, P. R. and Neering, I. R. (1976). Relationship between vasoconstriction and histamine induced vasodilation. Aust. J. Exp. Biol. Med. Sci. 54, 197–202. Cavalcante, R. R., Pereira, M. H. and Gontijo, N. F. (2003). Anti-complement activity in the saliva of phlebotomine sand flies and other haematophagous insects. Parasitology 127, 87–93. Cerna, P., Mikes, L. and Volf, P. (2002). Salivary gland hyaluronidase in various species of phlebotomine sand flies (Diptera: Psychodidae). Insect Biochem. Mol. Biol. 32, 1691–1697.

104


Champagne, D. E. (2004). Antihemostatic strategies of blood-feeding arthropods. Curr. Drug Targets Cardiovasc. Haematol. Disord. 4, 375–396. Champagne, D. E. and Ribeiro, J. M. C. (1994). Sialokinins I and II: two salivary tachykinins from the Yellow Fever mosquito, Aedes aegypti. Proc. Natl. Acad. Sci. USA 91, 138–142. Champagne, D. E., Nussenzveig, R. H. and Ribeiro, J. M. C. (1995a). Purification, characterization, and cloning of nitric oxide-carrying heme proteins (nitrophorins) from salivary glands of the blood sucking insect Rhodnius prolixus. J. Biol. Chem. 270, 8691–8695. Champagne, D. E., Smartt, C. T., Ribeiro, J. M. C. and James, A. A. (1995b). The salivary gland-specific apyrase of the mosquito Aedes aegypti is a member of the 50 -nucleotidase family. Proc. Natl. Acad. Sci. USA 92, 694–698. Charlab, R., Valenzuela, J. G., Rowton, E. D. and Ribeiro, J. M. C. (1999). Toward an understanding of the biochemical and pharmacological complexity of the saliva of a hematophagous sand fly Lutzomyia longipalpis. Proc. Natl. Acad. Sci. USA 96, 15155–15160. Charlab, R., Rowton, E. D. and Ribeiro, J. M. C. (2000). The salivary adenosine deaminase from the sand fly Lutzomyia longipalpis. Exp. Parasitol. 95, 45–53. Cheeseman, M. T. (1998). Characterization of apyrase activity from the salivary glands of the cat flea Ctenocephalides felis. Insect Biochem. Mol. Biol. 28, 1025–1030. Cheeseman, M. T., Bates, P. A. and Crampton, J. M. (2001). Preliminary characterisation of esterase and platelet-activating factor (PAF)-acetylhydrolase activities from cat flea (Ctenocephalides felis) salivary glands. Insect Biochem. Mol. Biol. 31, 157–164. Cobben, R. H. (1979). On the original feeding habits of the Hemiptera (Insecta): a reply to Merrill Sweet. Ann. Entomol. Soc. Am. 72, 711–715. Collier, G. F. (ed.) (2008). In: A Translation of the Eight Books of Aulus Cornelius Celsus on Medicine (1831), Kessinger Publishing, LLC, Whitefish, MT. Collis, M. G. (1989). The vasodilator role of adenosine. Pharmacol. Ther. 41, 143–162. Colman, R. W., Marder, V. J., Clowes, A. W., George, J. N. and Goldhaber, S. Z. (eds.) (2005). In: Hemostasis and Thrombosis: Basic Principles and Clinical Practice, 5th ed. Lippincott Williams & Wilkins, Philadelphia, PA. Cornwall, J. W. and Patton, W. S. (1914). Some observations on the salivary secretion of the common blood-sucking insects and ticks. Indian J. Med. Res. 2, 569–593. Cupp, M. S. and Cupp, E. W. (2000). Antithrombin protein and DNA sequences from black fly. US Patent Application 10/973, 046. Cupp, M. S., Cupp, E. W. and Ramberg, F. B. (1993). Salivary gland apyrase in black flies (Simulium vittatum). J. Insect Physiol. 39, 817–821. Cupp, M. S., Cupp, E. W., Ochoa, A. J. and Moulton, J. K. (1995). Salivary apyrase in New World blackflies (Diptera: Simuliidae) and its relationship to onchocerciasis vector status. Med. Vet. Entomol. 9, 325–330. Cupp, M. S., Ribeiro, J. M. C., Champagne, D. E. and Cupp, E. W. (1998). Analyses of cDNA and recombinant protein for a potent vasoactive protein in saliva of a bloodfeeding black fly, Simulium vittatum. J. Exp. Biol. 201, 1553–1561. Dan, A., Pereira, M. H., Pesquero, J. L., Diotaiuti, L. and Beiraõ, P. S. (1999). Action of the saliva of Triatoma infestans (Heteroptera: Reduviidae) on sodium channels. J. Med. Entomol. 36, 875–879. Dansa-Petreski, M., Ribeiro, J. M. C., Atella, G. C., Masuda, H. and Oliveira, P. L. (1995). Antioxidant role of Rhodnius prolixus heme-binding proteins. J. Biol. Chem. 270, 10893–10896. De Meillon, B. (1949). The relationship between ectoparasites and host. IV. Host reaction to bites of arthropods. The Leech 19, 43–46.


105

DeGiusti, D. L., Sterling, C. R. and Dobrzechowski, D. (1973). Transmission of the chelonian haemoproteid Haemoproteus metchnikovi by a Tabanid fly Chrysops callidus. Nature 242, 50–51. Demeure, C. E., Brahimi, K., Hacini, F., Marchand, F., Pe´ronet, R., Huerre, M., St.-Mezard, P., Nicolas, J.-F., Brey, P., Delespesse, G. and Mećheri, S. (2005). Anopheles mosquito bites activate cutaneous mast cells leading to a local inflammatory response and lymph node hyperplasia. J. Immunol. 174, 3932–3940. Dickerson, G. and Lavoipierre, M. M. J. (1959). Studies on the methods of feeding of blood sucking arthropods. II. The method of feeding adopted by the bedbug (Cimex lectularius) when obtaining a blood meal from the mammalian host. Ann. Trop. Med. Parasitol. 53, 347–357. Dickinson, R. G., O’Hagan, J. E., Shotz, M., Binnington, K. C. and Hegarty, M. P. (1976). Prostaglandin in the saliva of the cattle tick Boophilus microplus. Aust. J. Exp. Biol. Med. Sci. 54, 475–486. Dionisotti, S., Zocchi, C., Varani, K., Borea, P. A. and Ongini, E. (1992). Effects of adenosine derivatives on human and rabbit platelet aggregation. Correlation of adenosine receptor affinities and antiaggregatory activity. Naun. Schm. Arch. Pharmacol. 346, 673–676. Eichler, E. E. and Sankoff, D. (2003). Structural dynamics of eukaryotic chromosome evolution. Science 301, 793–797. Failer, B. U., Braun, N. and Zimmermann, H. (2002). Cloning, expression, and functional characterization of a Ca2þ-dependent endoplasmic reticulum nucleoside diphosphatase. J. Biol. Chem. 277, 36978–36986. Faudry, E., Lozzi, S. P., Santana, J. M., D’Souza-Ault, M., Kieffer, S., Felix, C. R., Ricart, C. A. O., Sousa, M. V., Vernet, T. and Teixeira, A. R. L. (2004). Triatoma infestans apyrases belong to the 50 -nucleotidase family. J. Biol. Chem. 279, 19607–19613. Feingold, B. (1968). The allergic responses to insect bites. Annu. Rev. Entomol. 13, 137–158. Flower, D. R., North, A. C. and Sansom, C. E. (2000). The lipocalin protein family: structural and sequence overview. Biochim. Biophys. Acta 1482, 9–24. Francischetti, I. M. B., Valenzuela, J. G. and Ribeiro, J. M. C. (1999). Anophelin: kinetics and mechanism of thrombin inhibition. Biochemistry 38, 16678–16685. Francischetti, I. M. B., Ribeiro, J. M. C., Champagne, D. E. and Andersen, J. (2000). Purification, cloning, expression, and mechanism of action of a novel platelet aggregation inhibitor from the salivary gland of the blood-sucking bug, Rhodnius prolixus. J. Biol. Chem. 275, 12639–12650. Francischetti, I. M. B., Andersen, J. F. and Ribeiro, J. M. C. (2002a). Biochemical and functional characterization of recombinant Rhodnius prolixus platelet aggregation inhibitor 1 as a novel lipocalin with high affinity for adenosine diphosphate and other adenine nucleotides. Biochemistry 41, 3810–3818. Francischetti, I. M. B., Valenzuela, J. G., Andersen, J. F., Mather, T. N. and Ribeiro, J. M. C. (2002b). Ixolaris, a novel recombinant tissue factor pathway inhibitor (TFPI) from the salivary gland of the tick, Ixodes scapularis: identification of factor X and factor Xa as scaffolds for the inhibition of factor VIIa/tissue factor complex. Blood 99, 3602–3612. Francischetti, I. M. B., Valenzuela, J. G., Pham, V. M., Garfield, M. K. and Ribeiro, J. M. C. (2002c). Toward a catalog for the transcripts and proteins (sialome) from the salivary gland of the malaria vector Anopheles gambiae. J. Exp. Biol. 205, 2429–2451. Francischetti, I. M. B., Mather, T. N. and Ribeiro, J. M. C. (2003). Cloning of a salivary gland metalloprotease and characterization of gelatinase and fibrin(ogen)lytic

106


activities in the saliva of the Lyme disease tick vector Ixodes scapularis. Biochem. Biophys. Res. Commun. 305, 869–875. Francischetti, I. M. B., Mather, T. N. and Ribeiro, J. M. C. (2004). Penthalaris, a novel recombinant five-Kunitz tissue factor pathway inhibitor (TFPI) from the salivary gland of the tick vector of Lyme disease, Ixodes scapularis. Thromb. Haemost. 91, 886–898. Francischetti, I. M. B., Sa´-Nunes, A., Mans, B. J., Santos, I. M. and Ribeiro, J. M. C. (2009). The role of saliva in tick feeding. Front. Biosci. 14, 2051–2088. Friend, W. G. and Smith, J. J. B. (1971). Feeding in Rhodnius prolixus: mouthpart activity and salivation, and their correlation with changes of electrical resistance. J. Insect Physiol. 17, 233–243. Fuentes-Prior, P., Noeske-Jungblut, C., Donner, P., Schleuning, W. D., Huber, R. and Bode, W. (1997). Structure of the thrombin complex with triabin, a lipocalin-like exosite-binding inhibitor derived from a triatomine bug. Proc. Natl. Acad. Sci. USA 94, 11845–11850. Galperin, M. Y. and Koonin, E. V. (2004). ‘Conserved hypothetical’ proteins: prioritization of targets for experimental study. Nucleic Acids Res. 32, 5452–5463. Gillett, J. D. (1967). Natural selection and feeding speed in a blood sucking insect. Proc. R. Soc. B 167, 316–329. Glusa, E., Bretschneider, E., Daum, J. and Noeske-Jungblut, C. (1997). Inhibition of thrombin-mediated cellular effects by triabin, a highly potent anion-binding exosite thrombin inhibitor. Thromb. Haemost. 77, 1196–1200. Gogarten, J. P. and Townsend, J. P. (2005). Horizontal gene transfer, genome innovation and evolution. Nat. Rev. Microbiol. 3, 679–687. Golodne, D. M., Monteiro, R. Q., Graça-Souza, A. V., Silva-Neto, M. A. C. and Atella, G. C. (2003). Lysophosphatidylcholine acts as an anti-hemostatic molecule in the saliva of the blood-sucking bug Rhodnius prolixus. J. Biol. Chem. 278, 27766–27771. Gomes, R., Teixeira, C., Teixeira, M. J., Oliveira, F., Menezes, M. J., Silva, C., de Oliveira, C. I., Miranda, J. C., Elnaiem, D.-E., Kamhawi, S., Valenzuela, J. G. and Brodskyn, C. I. (2008). Immunity to a salivary protein of a sand fly vector protects against the fatal outcome of visceral leishmaniasis in a hamster model. Proc. Natl. Acad. Sci. USA 105, 7845–7850. Gordon, R. M. and Crewe, W. (1948). The mechanism by which mosquitoes and tsetse flies obtain their blood meal, the histology of the lesions produced, and the subsequent reactions of the mammalian host, together with some observations on the feeding of Chrysops and Cimex. Ann. Trop. Med. Parasitol. 42, 334–356. Gordon, R. M. and Lumsden, W. H. R. (1939). A study of the behaviour of the mouthparts of mosquitoes when taking up blood from living tissue; together with some observations on the ingestion of microfilariae. Ann. Trop. Med. Parasitol. 33, 259–278. Grimaldi, D. and Engel, M. (2005). Evolution of the Insects. Cambridge University Press, New York, NY. Gudderra, N. P., Ribeiro, J. M. C. and Andersen, J. F. (2005). Structural determinants of factor IX(a) binding in nitrophorin 2, a lipocalin inhibitor of the intrinsic coagulation pathway. J. Biol. Chem. 280, 25022–25028. Hanifin, C. T., Brodie, E. D. and Brodie, E. D. (2008). Phenotypic mismatches reveal escape from arms-race coevolution. PLoS Biol. 6, 471–482. Harris, S. J., Parry, R. V., Westwick, J. and Ward, S. G. (2008). Phosphoinositide lipid phosphatases: natural regulators of phosphoinositide 3-kinase signaling in T lymphocytes. J. Biol. Chem. 283, 2465–2469.


107

Hekmat-Scafe, D. S., Dorit, R. L. and Carlson, J. R. (2000). Molecular evolution of odorant-binding protein genes OS-E and OS-F in Drosophila. Genetics 155, 117–127. Hellmann, K. and Hawkins, R. I. (1965). Prolixin-S and Prolixin-G: two anticoagulants from Rhodnius prolixus Stahl. Nature 207, 265–267. Higgs, G. A., Vane, J. R., Hart, R. J., Porter, C. and Wilson, R. G. (1976). Prostaglandins in the saliva of the cattle tick, Boophilus microplus (Canestrini) (Acarina, Ixodidae). Bull. Entomol. Res. 66, 665–670. Hilgartner, R., Raoilison, M., Buttiker, W., Lees, D. C. and Krenn, H. W. (2007). Malagasy birds as hosts for eye-frequenting moths. Biol. Lett. 3, 117–120. Hill, P., Saunders, D. S. and Campbell, J. A. (1973). Letter: the production of ‘‘symbiontfree’’ Glossina morsitans and an associated loss of female fertility. Trans. R. Soc. Trop. Med. Hyg. 67, 727–728. Hill, P., Campbell, J. A. and Petrie, I. A. (1976). Rhodnius prolixus and its symbiotic actinomycete: a microbiological, physiological and behavioral study. Proc. R. Soc. Lond. B 194, 501–525. Hoffman, D. R. (2008). Biting insect allergens. Clin. Allergy Immunol. 21, 251–260. Holland, P. W. (1999). Gene duplication: past, present and future. Semin. Cell. Dev. Biol. 10, 541–547. Horuk, R., Colby, T. J., Darbonne, W. C., Schall, T. J. and Neote, K. (1993). The human erythrocyte inflammatory peptide (chemokine) receptor. Biochemical characterization, solubilization, and development of a binding assay for the soluble receptor. Biochemistry 32, 5733–5738. Hovius, J. W., Levi, M. and Fikrig, E. (2008). Salivating for knowledge: potential pharmacological agents in tick saliva. PLoS Med. 5, 202–208. Hudson, A. (1964). Some functions of the salivary glands of mosquitoes and other bloodsucking arthropods. Can. J. Zool. 42, 113–120. Hudson, A., Bowman, L. and Orr, C. W. M. (1960). Effects of absence of saliva on blood feeding by mosquitoes. Science 131, 1730–1731. Hypsa, V. (1993). Endosymbionts of Triatoma infestans: distribution and transmission. J. Invert. Pathol. 61, 32–38. Inokuma, H., Kemp, D. H. and Willadsen, P. (1994). Prostaglandin E2 production by the cattle tick (Boophilus microplus) into feeding sites and its effect on the response of bovine mononuclear cells to mitogen. Vet. Parasitol. 53, 293–299. Isawa, H., Yuda, M., Yoneda, K. and Chinzei, Y. (2000). The insect salivary protein, prolixin-S, inhibits factor IXa generation and Xase complex formation in the blood coagulation pathway. J. Biol. Chem. 275, 6636–6641. Isawa, H., Yuda, M., Orito, Y. and Chinzei, Y. (2002). A mosquito salivary protein inhibits activation of the plasma contact system by binding to factor XII and high molecular weight kininogen. J. Biol. Chem. 277, 27651–27658. Isawa, H., Orito, Y., Iwanaga, S., Jingushi, N., Morita, A., Chinzei, Y. and Yuda, M. (2007). Identification and characterization of a new kallikrein–kinin system inhibitor from the salivary glands of the malaria vector mosquito Anopheles stephensi. Insect Biochem. Mol. Biol. 37, 466–477. Jain, R., Rivera, M. C., Moore, J. E. and Lake, J. A. (2002). Horizontal gene transfer in microbial genome evolution. Theor. Popul. Biol. 61, 489–495. James, A. A., Blackmer, K., Marinotti, O., Ghosn, C. R. and Racioppi, J. V. (1991). Isolation and characterization of the gene expressing the major salivary gland protein of the female mosquito, Aedes aegypti. Mol. Biochem. Parasitol. 44, 245–254. Kalesnikoff, J., Sly, L. M., Hughes, M. R., Buchse, T., Rauh, M. J., Cao, L. P., Lam, V., Mui, A., Huber, M. and Krystal, G. (2003). The role of SHIP in cytokine-induced signaling. Rev. Physiol. Biochem. Pharmacol. 149, 87–103.

108


Kanost, M. R., Jiang, H. and Yu, X. Q. (2004). Innate immune responses of a lepidopteran insect, Manduca sexta. Immunol. Rev. 198, 97–105. Kashiwada, M., Lu, P. and Rothman, P. B. (2007). PIP3 pathway in regulatory T cells and autoimmunity. Immunol. Res. 39, 194–224. Kato, H., Anderson, J. M., Kamhawi, S., Oliveira, F., Lawyer, P. G., Pham, V. M., Sangare, C. S., Samake, S., Sissoko, I., Garfield, M., Sigutova, L. Volf, P., et al. (2006). High degree of conservancy among secreted salivary gland proteins from two geographically distant Phlebotomus duboscqi sandflies populations (Mali and Kenya). BMC Genomics 7, 226. Kato, H., Jochim, R. C., Lawyer, P. G. and Valenzuela, J. G. (2007). Identification and characterization of a salivary adenosine deaminase from the sand fly Phlebotomus duboscqi, the vector of Leishmania major in sub-Saharan Africa. J. Exp. Biol. 210, 733–740. Katz, O., Waitumbi, J. N., Zer, R. and Warburg, A. (2000). Adenosine, AMP, and protein phosphatase activity in sandfly saliva. Am. J. Trop. Med. Hyg. 62, 145–150. Keeling, P. J. and Palmer, J. D. (2008). Horizontal gene transfer in eukaryotic evolution. Nat. Rev. Genet. 9, 605–618. Kelm, M., Feelisch, M., Krebber, T., Motz, W. and Strauer, B. E. (1993). Mechanisms of histamine-induced coronary vasodilatation: H1-receptor-mediated release of endothelium-derived nitric oxide. J. Vasc. Res. 30, 132–138. Kerlin, R. L. and Hughes, S. (1992). Enzymes in saliva from four parasitic arthropods. Med. Vet. Entomol. 6, 121–126. King, T. P. and Spangfort, M. D. (2000). Structure and biology of stinging insect venom allergens. Int. Arch. Allergy Immunol. 123, 99–106. Korochkina, S., Barreau, C., Pradel, G., Jeffery, E., Li, J., Natarajan, R., Shabanowitz, J., Hunt, D., Frevert, U. and Vernick, K. D. (2006). A mosquito-specific protein family includes candidate receptors for malaria sporozoite invasion of salivary glands. Cell. Microbiol. 8, 163–175. Kovalick, G. E. and Griffin, D. L. (2005). Characterization of the SCP/TAPS gene family in Drosophila melanogaster. Insect Biochem. Mol. Biol. 35, 825–835. Krzywinski, J., Grushko, O. G. and Besansky, N. J. (2006). Analysis of the complete mitochondrial DNA from Anopheles funestus: an improved dipteran mitochondrial genome annotation and a temporal dimension of mosquito evolution. Mol. Phylogenet. Evol. 39, 417–423. Kumar, V., Abbas, A. K., Fausto, N. and Mitchell, R. (2004). Pathologic Basis of Disease. Saunders, Philadelphia, PA. Lai, R., Takeuchi, H., Jonczy, J., Rees, H. H. and Turner, P. C. (2004). A thrombin inhibitor from the ixodid tick, Amblyomma hebraeum. Gene 342, 243–249. Lane, R. P. and Crosskey, R. W. (1993). Medical Insects and Arachnids. Chapman and Hall, New York, NY. Lanzaro, G. C., Lopes, A. H., Ribeiro, J. M. C., Shoemaker, C. B., Warburg, A., Soares, M. and Titus, R. G. (1999). Variation in the salivary peptide, maxadilan, from species in the Lutzomyia longipalpis complex. Insect Mol. Biol. 8, 267–275. Lavoipierre, M. M. J. (1964). Feeding mechanisms of blood-sucking arthropods. Nature 208, 302–303. Lavoipierre, M. M. J., Dickerson, G. and Gordon, R. M. (1959). Studies on the methods of feeding of blood sucking arthropods. I. The manner in which triatomine bugs obtain their blood meal, as observed in the tissues of the living rodent, with some remarks on the effects of the bite on human volunteers. Ann. Trop. Med. Parasitol. 53, 235–250. Law, J., Ribeiro, J. M. C. and Wells, M. (1992). Biochemical insights derived from diversity in insects. Annu. Rev. Biochem. 61, 87–112.


109

Lehane, M. J. (2005). The Biology of Blood-Sucking in Insects. Cambridge University Press, Cambridge. Lent, H. and Wygodzinsky, P. (1979). Revision of the triatominae (Hemiptera, Reduviidae), and their significance as vectors of Chagas’ disease. Bull. Am. Mus. Nat. Hist. 63, 127–520. Lerner, E. A. and Shoemaker, C. B. (1992). Maxadilan: cloning and functional expression of the gene encoding this potent vasodilator peptide. J. Biol. Chem. 267, 1062–1066. Lerner, E. A., Ribeiro, J. M. C., Nelson, R. J. and Lerner, M. R. (1991). Isolation of maxadilan, a potent vasodilatory peptide from the salivary glands of the sand fly Lutzomyia longipalpis. J. Biol. Chem. 266, 11234–11236. Lester, H. M. O. and Lloyd, L. (1928). Notes on the process of digestion in tsetse flies. Bull. Entomol. Res. 19, 39–60. Leung, W. H., Tarasenko, T. and Bolland, S. (2008). Differential roles for the inositol phosphatase SHIP in the regulation of macrophages and lymphocytes. Immunol. Res. 43, 243–251. Lewis, D. J. (1975). Functional morphology of the mouthparts in New World phlebotomine sandflies (Diptera: Psychodidae). Trans. R. Entomol. Soc. Lond. 126, 497–532. Li, S. and Aksoy, S. (2000). A family of genes with growth factor and adenosine deaminase similarity are preferentially expressed in the salivary glands of Glossina m. morsitans. Gene 252, 83–93. Li, S., Kwon, J. and Aksoy, S. (2001). Characterization of genes expressed in the salivary glands of the tsetse fly, Glossina morsitans morsitans. Insect Mol. Biol. 10, 69–76. Lohse, M. J., Maurer, K., Gensheimer, H. P. and Schwabe, U. (1987). Dual actions of adenosine on rat peritoneal mast cells. Naun. Schm. Arch. Pharmacol. 335, 555–560. Lombardo, F., Di Cristina, M., Spanos, L., Louis, C., Coluzzi, M. and Arca`, B. (2000). Promoter sequences of the putative Anopheles gambiae apyrase confer salivary gland expression in Drosophila melanogaster. J. Biol. Chem. 275, 23861–23868. Lombardo, F., Ronca, R., Rizzo, C., Mestres-Simo`n, M., Lanfrancotti, A., Curra`, C., Fiorentino, G., Bourgouin, C., Ribeiro, J. M., Petrarca, V., Ponzi, M. Coluzzi, M., et al. (2009). The Anopheles gambiae salivary protein gSG6: an anopheline-specific protein with a blood-feeding role. Insect Biochem. Mol. Biol. 39, 457–466. Lu, X., Deadman, J. J., Williams, J. A., Kakkar, V. V. and Rahman, S. (1993). Synthetic RGD peptides derived from the adhesive domains of snake-venom proteins: evaluation as inhibitors of platelet aggregation. Biochem. J. 296, 21–24. Maa, T. C. (1964). Genera and species of Hippoboscidae (Diptera): types, synonymy, habits and natural groupings. J. Med. Entomol. 1, 4–41. Maa, T. C. (1965). An interim world list of batflies. J. Med. Entomol. 1, 377–386. Mans, B. J. and Ribeiro, J. M. C. (2008a). Function, mechanism and evolution of the moubatin-clade of soft tick lipocalins. Insect Biochem. Mol. Biol. 38, 841–852. Mans, B. J. and Ribeiro, J. M. C. (2008b). A novel clade of cysteinyl leukotriene scavengers in soft ticks. Insect Biochem. Mol. Biol. 38, 862–870. Mans, B. J., Louw, A. I. and Neitz, A. W. (2002). Evolution of hematophagy in ticks: common origins for blood coagulation and platelet aggregation inhibitors from soft ticks of the genus Ornithodoros. Mol. Biol. Evol. 19, 1695–1705. Mans, B. J., Calvo, E., Ribeiro, J. M. C. and Andersen, J. F. (2007). The crystal structure of D7r4, a salivary biogenic amine-binding protein from the malaria mosquito Anopheles gambiae. J. Biol. Chem. 282, 36626–36633. Mans, B. J., Andersen, J. F., Schwan, T. G. and Ribeiro, J. M. C. (2008). Characterization of anti-hemostatic factors in the argasid, Argas monolakensis: implications for the evolution of blood-feeding in the soft tick family. Insect Biochem. Mol. Biol. 38, 22–41.

110


Mant, M. J. and Parker, K. R. (1981). Two platelet aggregation inhibitors in tsetse (Glossina) saliva with studies of roles of thrombin and citrate in in vitro platelet aggregation. Br. J. Haematol. 48, 601–608. Marchler-Bauer, A., Panchenko, A. R., Shoemaker, B. A., Thiessen, P. A., Geer, L. Y. and Bryant, S. H. (2002). CDD: a database of conserved domain alignments with links to domain three-dimensional structure. Nucleic Acids Res. 30, 281–283. Marinotti, O., de Brito, M. and Moreira, C. K. (1996). Apyrase and alpha-glucosidase in the salivary glands of Aedes albopictus. Comp. Biochem. Physiol. B 113, 675–679. Martins, R. M., Sforca, M. L., Amino, R., Juliano, M. A., Oyama, S. Jr., Juliano, L., Pertinhez, T. A., Spisni, A. and Schenkman, S. (2006). Lytic activity and structural differences of amphipathic peptides derived from trialysin. Biochemistry 45, 1765–1774. Mattila, J. T., Munderloh, U. G. and Kurtti, T. J. (2007). Rickettsia peacockii, an endosymbiont of Dermacentor andersoni, does not elicit or inhibit humoral immune responses from immunocompetent D. andersoni or Ixodes scapularis cell lines. Dev. Comp. Immunol. 31, 1095–1106. Mazet, F. and Shimeld, S. M. (2002). Gene duplication and divergence in the early evolution of vertebrates. Curr. Opin. Genet. Dev. 12, 393–396. Megraw, T., Kaufman, T. C. and Kovalick, G. E. (1998). Sequence and expression of Drosophila Antigen 5-related 2, a new member of the CAP gene family. Gene 222, 297–304. Mejia, J. S., Bishop, J. V. and Titus, R. G. (2006). Is it possible to develop pan-arthropod vaccines? Trends Parasitol. 22, 367–370. Mellanby, K. (1946). Man’s reaction to mosquito bites. Nature 158, 554. Mellink, J. J. and Van Den Boven Kamp, W. (1981). Functional aspects of mosquito salivation in blood feeding in Aedes aegypti. Mosq. News 41, 115–119. Metcalf, R. L. (1945). The physiology of the salivary glands of Anopheles quadrimaculatus. J. Nat. Malar. Soc. 4, 271–278. Millan, M. J. (1999). The induction of pain: an integrative review. Prog. Neurobiol. 57, 1–164. Milleron, R. S., Ribeiro, J. M. C., Elnaime, D., Soong, L. and Lanzaro, G. C. (2004). Negative effect of antibodies against maxadilan on the fitness of the sand fly vector of American visceral leishmaniasis. Am. J. Trop. Med. Hyg. 70, 278–285. Milne, T. J., Abbenante, G., Tyndall, J. D., Halliday, J. and Lewis, R. J. (2003). Isolation and characterization of a cone snail protease with homology to CRISP proteins of the pathogenesis-related protein superfamily. J. Biol. Chem. 278, 31105–31110. Morita, A., Isawa, H., Orito, Y., Iwanaga, S., Chinzei, Y. and Yuda, M. (2006). Identification and characterization of a collagen-induced platelet aggregation inhibitor, triplatin, from salivary glands of the assassin bug, Triatoma infestans. FEBS J. 273, 2955–2962. Moro, O. and Lerner, E. A. (1997). Maxadilan, the vasodilator from sand flies, is a specific pituitary adenylate cyclase activating peptide type I receptor agonist. J. Biol. Chem. 272, 966–970. Morrow, J. D., Awad, J. A., Oates, J. A. and Roberts, L. J. I. I. (1992). Identification of skin as a major site of prostaglandin D2 release following oral administration of niacin in humans. J. Invest. Dermatol. 98, 812–815. Nascimento, R. J., Santana, J. M., Lozzi, S. P., Araujo, C. N. and Teixeira, A. R. (2001). Human IgG1 and IgG4: the main antibodies against Triatoma infestans (Hemiptera: Reduviidae) salivary gland proteins. Am. J. Trop. Med. Hyg. 65, 219–226. Nei, M. and Rooney, A. P. (2005). Concerted and birth-and-death evolution of multigene families. Annu. Rev. Genet. 39, 121–152.


111

Nene, V., Wortman, J. R., Lawson, D., Haas, B., Kodira, C., Tu, Z. J., Loftus, B., Xi, Z., Megy, K., Grabherr, M., Ren, K. and Zbodonov, E. M. (2007). Genome sequence of Aedes aegypti, a major arbovirus vector. Science 316, 1718–1723. Nielsen, H., Engelbrecht, J., Brunak, S. and von Heijne, G. (1997). Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Eng. 10, 1–6. Noeske-Jungblut, C., Kratzschmar, J., Haendler, B., Alagon, A., Possani, L., Verhallen, P., Donner, P. and Schleuning, W. D. (1994). An inhibitor of collageninduced platelet aggregation from the saliva of Triatoma pallidipennis. J. Biol. Chem. 269, 5050–5053. Noeske-Jungblut, C., Haendler, B., Donner, P., Alagon, A., Possani, L. and Schleuning, W. D. (1995). Triabin, a highly potent exosite inhibitor of thrombin. J. Biol. Chem. 270, 28629–28634. Nogge, G. (1978). Aposymbiotic tsetse flies, Glossina morsitans morsitans obtained by feeding on rabbits immunized specifically with symbionts. J. Insect Physiol. 24, 299–304. Nunn, M. A., Sharma, A., Paesen, G. C., Adamson, S., Lissina, O., Willis, A. C. and Nuttall, P. A. (2005). Complement inhibitor of C5 activation from the soft tick Ornithodoros moubata. J. Immunol. 174, 2084–2091. Nuttall, P. A., Trimnell, A. R., Kazimirova, M. and Labuda, M. (2006). Exposed and concealed antigens as vaccine targets for controlling ticks and tick-borne diseases. Parasite Immunol. 28, 155–163. O’Rourke, F. J. (1956). Observations on pool and capillary feeding in Aedes aegypti (L.). Nature 177, 1087–1088. Ohno, S. (1999). Gene duplication and the uniqueness of vertebrate genomes circa 1970–1999. Semin. Cell. Dev. Biol. 10, 517–522. Oliveira, F., Kamhawi, S., Seitz, A. E., Pham, V. M., Guigal, P. M., Fischer, L., Ward, J. and Valenzuela, J. G. (2006). From transcriptome to immunome: identification of DTH inducing proteins from a Phlebotomus ariasi salivary gland cDNA library. Vaccine 24, 374–390. Orlandi-Pradines, E., Almeras, L., Denis de Senneville, L., Barbe, S., Remoue, F., Villard, C., Cornelie, S., Penhoat, K., Pascual, A. Bourgouin, C., et al. (2007). Antibody response against saliva antigens of Anopheles gambiae and Aedes aegypti in travellers in tropical Africa. Microbes Infect. 9, 1454–1462. Paesen, G. C., Adams, P. L., Harlos, K., Nuttall, P. A. and Stuart, D. I. (1999). Tick histamine-binding proteins: isolation, cloning, and three-dimensional structure. Mol. Cell 3, 661–671. Paesen, G. C., Adams, P. L., Nuttall, P. A. and Stuart, D. L. (2000). Tick histaminebinding proteins: lipocalins with a second binding cavity. Biochim. Biophys. Acta 1482, 92–101. Paesen, G. C., Siebold, C., Harlos, K., Peacey, M. F., Nuttall, P. A. and Stuart, D. I. (2007). A tick protein with a modified Kunitz fold inhibits human tryptase. J. Mol. Biol. 368, 1172–1186. Pais, R., Lohs, C., Wu, Y., Wang, J. and Aksoy, S. (2008). The obligate mutualist Wigglesworthia glossinidia influences reproduction, digestion, and immunity processes of its host, the tsetse fly. Appl. Environ. Microbiol. 74, 5965–5974. Paul, W. E. (2008). Fundamental Immunology. Lippincott Williams & Wilkins, Philadelphia, PA. Pavlov, I. P. (1902). The Work of the Digestive Glands. Charles Griffin & Co., London. Peng, Z. and Simons, E. R. (1997). Cross-reactivity of skin and serum specific IgE responses and allergen analysis for three mosquito species with worldwide distribution. J. Allergy Clin. Immunol. 100, 192–198.

112


Peng, Z. and Simons, F. E. (1998). A prospective study of naturally acquired sensitization and subsequent desensitization to mosquito bites and concurrent antibody responses. J. Allergy Clin. Immunol. 101, 284–286. Peng, Z. and Simons, F. E. (2007). Advances in mosquito allergy. Curr. Opin. Allergy Clin. Immunol. 7, 350–354. Peng, Z., Li, H. and Simons, F. E. (1998). Immunoblot analysis of salivary allergens in 10 mosquito species with worldwide distribution and the human IgE responses to these allergens. J. Allergy Clin. Immunol. 101, 498–505. Poinsignon, A., Cornelie, S., Mestres-Simon, M., Lanfrancotti, A., Rossignol, M., Boulanger, D., Cisse, B., Sokhna, C., Arca`, B. Simondon, F., et al. (2008). Novel peptide marker corresponding to salivary protein gSG6 potentially identifies exposure to Anopheles bites. PLoS ONE 3, 1–7. Reddy, V. B., Kounga, K., Mariano, F. and Lerner, E. A. (2000). Chrysoptin is a potent glycoprotein IIb/IIIa fibrinogen receptor antagonist present in salivary gland extracts of the deerfly. J. Biol. Chem. 275, 15861–15867. Remoue, F., Cisse, B., Ba, F., Sokhna, C., Herve, J. P., Boulanger, D. and Simondon, F. (2006). Evaluation of the antibody response to Anopheles salivary antigens as a potential marker of risk of malaria. Trans. R. Soc. Trop. Med. Hyg. 100, 363–370. Ribeiro, J. M. C. (1982). The antiserotonin and antihistamine activities of salivary secretion of Rhodnius prolixus. J. Insect Physiol. 28, 69–75. Ribeiro, J. M. C. (1987). Role of arthropod saliva in blood feeding. Annu. Rev. Entomol. 32, 463–478. Ribeiro, J. M. C. (1988). How mosquitoes find blood. Misc. Publ. Entomol. Soc. Am. 68, 18–24. Ribeiro, J. M. C. (1995). Blood-feeding arthropods: live syringes or invertebrate pharmacologists? Infect. Agents Dis. 4, 143–152. Ribeiro, J. M. C. (2000). Blood-feeding in mosquitoes: probing time and salivary gland anti-haemostatic activities in representatives of three genera (Aedes, Anopheles, Culex). Med. Vet. Entomol. 14, 142–148. Ribeiro, J. M. C. and Francischetti, I. M. B. (2001). Platelet-activating-factorhydrolyzing phospholipase C in the salivary glands and saliva of the mosquito Culex quinquefasciatus. J. Exp. Biol. 204, 3887–3894. Ribeiro, J. M. C. and Francischetti, I. M. B. (2003). Role of arthropod saliva in blood feeding: sialome and post-sialome perspectives. Annu. Rev. Entomol. 48, 73–88. Ribeiro, J. M. C. and Garcia, E. S. (1980). The salivary and crop apyrase activity of Rhodnius prolixus. J. Insect Physiol. 26, 303–307. Ribeiro, J. M. C. and Garcia, E. S. (1981). The role of saliva in feeding in Rhodnius prolixus. J. Exp. Biol. 94, 219–230. Ribeiro, J. M. C. and Mather, T. N. (1998). Ixodes scapularis: salivary kininase activity is a metallo dipeptidyl carboxypeptidase. Exp. Parasitol. 89, 213–221. Ribeiro, J. M. C. and Modi, G. (2001). The salivary adenosine/AMP content of Phlebotomus argentipes Annandale and Brunetti, the main vector of human kala-azar. J. Parasitol. 87, 915–917. Ribeiro, J. M. C. and Nussenzveig, R. H. (1993). The salivary catechol oxidase/peroxidase activities of the mosquito, Anopheles albimanus. J. Exp. Biol. 179, 273–287. Ribeiro, J. M. C. and Spielman, A. (1986). Ixodes dammini: salivary anaphylatoxin inactivating activity. Exp. Parasitol. 62, 292–297. Ribeiro, J. M. C. and Valenzuela, J. G. (1999). Purification and cloning of the salivary peroxidase/catechol oxidase of the mosquito Anopheles albimanus. J. Exp. Biol. 202, 809–816. Ribeiro, J. M. C. and Valenzuela, J. G. (2003). The salivary purine nucleosidase of the mosquito, Aedes aegypti. Insect Biochem. Mol. Biol. 33, 13–22.


113

Ribeiro, J. M. C. and Walker, F. A. (1994). High affinity histamine-binding and anti-histaminic activity of the salivary NO-carrying heme protein (Nitrophorin) of Rhodnius prolixus. J. Exp. Med. 180, 2251–2257. Ribeiro, J. M. C., Rossignol, P. A. and Spielman, A. (1984a). Role of mosquito saliva in blood vessel location. J. Exp. Biol. 108, 1–7. Ribeiro, J. M. C., Sarkis, J. J. F., Rossignol, P. A. and Spielman, A. (1984b). Salivary apyrase of Aedes aegypti: characterization and secretory fate. Comp. Biochem. Physiol. B 79, 81–86. Ribeiro, J. M. C., Rossignol, P. A. and Spielman, A. (1985a). Aedes aegypti: model for blood finding behavior and prediction of parasite manipulation. Exp. Parasitol. 60, 118–132. Ribeiro, J. M. C., Rossignol, P. A. and Spielman, A. (1985b). Salivary gland apyrase determines probing time in anopheline mosquitoes. J. Insect Physiol. 9, 551–560. Ribeiro, J. M. C., Rossignol, P. A. and Spielman, A. (1986). Blood finding strategy of a capillary feeding sandfly, Lutzomyia longipalpis. Comp. Biochem. Physiol. A 83, 683–686. Ribeiro, J. M. C., Modi, G. B. and Resh, R. B. (1989). Salivary apyrase activity of some old world phlebotomine sand flies. Insect Biochem. 19, 409–412. Ribeiro, J. M. C., Vaughan, J. A. and Azad, A. F. (1990). Characterization of the salivary apyrase activity of three rodent flea species. Comp. Biochem. Physiol. B 95, 215–218. Ribeiro, J. M. C., Endris, T. M. and Endris, R. (1991). Saliva of the soft tick, Ornithodoros moubata, contains anti-platelet and apyrase activities. Comp. Biochem. Physiol. A 100, 109–112. Ribeiro, J. M. C., Evans, P. M., MacSwain, J. L. and Sauer, J. (1992). Amblyomma americanum: characterization of salivary prostaglandins E2 and F2a by RP-HPLC/ bioassay and gas chromatography-mass spectrometry. Exp. Parasitol. 74, 112–116. Ribeiro, J. M. C., Nussenzveig, R. H. and Tortorella, G. (1994). Salivary vasodilators of Aedes triseriatus and Anopheles gambiae (Diptera: Culicidae). J. Med. Entomol. 31, 747–753. Ribeiro, J. M. C., Schneider, M. and Guimara˜es, J. A. (1995). Purification and characterization of prolixin S (nitrophorin 2), the salivary anticoagulant of the blood sucking bug, Rhodnius prolixus. Biochem. J. 308, 243–249. Ribeiro, J. M. C., Schneider, M., Isaias, T., Jurberg, J., Galvaõ, C. and Guimara˜es, J. A. (1998). Role of salivary antihemostatic components in blood feeding by triatomine bugs (Heteroptera). J. Med. Entomol. 35, 599–610. Ribeiro, J. M. C., Katz, O., Pannell, L. K., Waitumbi, J. and Warburg, A. (1999). Salivary glands of the sand fly Phlebotomus papatasi contain pharmacologically active amounts of adenosine and 50 -AMP. J. Exp. Biol. 202, 1551–1559. Ribeiro, J. M. C., Charlab, R., Rowton, E. D. and Cupp, E. W. (2000a). Simulium vittatum (Diptera: Simuliidae) and Lutzomyia longipalpis (Diptera: Psychodidae) salivary gland hyaluronidase activity. J. Med. Entomol. 37, 743–747. Ribeiro, J. M. C., Rowton, E. D. and Charlab, R. (2000b). The salivary 50 -nucleotidase/ phosphodiesterase of the hematophagus sand fly, Lutzomyia longipalpis [corrected]. Insect Biochem. Mol. Biol. 30, 279–285. Ribeiro, J. M. C., Charlab, R. and Valenzuela, J. G. (2001). The salivary adenosine deaminase activity of the mosquitoes Culex quinquefasciatus and Aedes aegypti. J. Exp. Biol. 204, 2001–2010. Ribeiro, J. M. C., Andersen, J., Silva-Neto, M. A., Pham, V. M., Garfield, M. K. and Valenzuela, J. G. (2004a). Exploring the sialome of the blood-sucking bug Rhodnius prolixus. Insect Biochem. Mol. Biol. 34, 61–79.

114


Ribeiro, J. M. C., Charlab, R., Pham, V. M., Garfield, M. and Valenzuela, J. G. (2004b). An insight into the salivary transcriptome and proteome of the adult female mosquito Culex pipiens quinquefasciatus. Insect Biochem. Mol. Biol. 34, 543–563. Ribeiro, J. M. C., Arca`, B., Lombardo, F., Calvo, E., Phan, V. M., Chandra, P. K. and Wikel, S. K. (2007). An annotated catalogue of salivary gland transcripts in the adult female mosquito, Aedes aegypti. BMC Genomics 8, 6. Ricci, I., Cancrini, G., Gabrielli, S., D’Amelio, S. and Favi, G. (2002). Searching for Wolbachia (Rickettsiales: Rickettsiaceae) in mosquitoes (Diptera: Culicidae): large polymerase chain reaction survey and new identifications. J. Med. Entomol. 39, 562–567. Rossignol, P. A. and Spielman, A. (1982). Fluid transport across the ducts of the salivary glands of a mosquito. J. Insect Physiol. 28, 579–583. Rothschild, M. (1975). Recent advances in our knowledge of the order Siphonaptera. Annu. Rev. Entomol. 20, 241–259. Ruiz, F. A., Lea, C. R., Oldfield, E. and Docampo, R. (2004). Human platelet dense granules contain polyphosphate and are similar to acidocalcisomes of bacteria and unicellular eukaryotes. J. Biol. Chem. 279, 44250–44257. Sa´-Nunes, A., Bafica, A., Lucas, D. A., Conrads, T. P., Veenstra, T. D., Andersen, J. F., Mather, T. N., Ribeiro, J. M. C. and Francischetti, I. M. B. (2007). Prostaglandin E2 is a major inhibitor of dendritic cell maturation and function in Ixodes scapularis saliva. J. Immunol. 179, 1497–1505. Sanders, M. L., Jaworski, D. C., Sanchez, J. L., DeFraites, R. F., Glass, G. E., Scott, A. L., Raha, S., Ritchie, B. C., Needham, G. R. and Schwartz, B. S. (1998). Antibody to a cDNA-derived calreticulin protein from Amblyomma americanum as a biomarker of tick exposure in humans. Am. J. Trop. Med. Hyg. 59, 279–285. Sanders, M. L., Glass, G. E., Nadelman, R. B., Wormser, G. P., Scott, A. L., Raha, S., Ritchie, B. C., Jaworski, D. C. and Schwartz, B. S. (1999). Antibody levels to recombinant tick calreticulin increase in humans after exposure to Ixodes scapularis (Say) and are correlated with tick engorgement indices. Am. J. Epidemiol. 149, 777–784. Sangamnatdej, S., Paesen, G. C., Slovak, M. and Nuttall, P. A. (2002). A high affinity serotonin- and histamine-binding lipocalin from tick saliva. Insect Mol. Biol. 11, 79–86. Sankoff, D. (2001). Gene and genome duplication. Curr. Opin. Genet. Dev. 11, 681–684. Santos, A., Ribeiro, J. M. C., Lehane, M. J., Gontijo, N. F., Veloso, A. B., Sant’Anna, M. R., Nascimento Araujo, R., Grisard, E. C. and Pereira, M. H. (2007). The sialotranscriptome of the blood-sucking bug Triatoma brasiliensis (Hemiptera, Triatominae). Insect Biochem. Mol. Biol. 37, 702–712. Schlehuber, S. and Skerra, A. (2005). Lipocalins in drug discovery: from natural ligandbinding proteins to ‘‘anticalins’’ Drug Discov. Today 10, 23–33. Schofield, C. J. (1979). The behavior of Triatominae: a review. Bull. Entomol. Res. 69, 363–379. Schreiber, M. C., Karlo, J. C. and Kovalick, G. E. (1997). A novel cDNA from Drosophila encoding a protein with similarity to mammalian cysteine-rich secretory proteins, wasp venom antigen 5, and plant group 1 pathogenesis-related proteins. Gene 191, 135–141. Schwartz, B. S., Ribeiro, J. M. C. and Goldstein, M. D. (1990). Anti-tick antibodies: an epidemiologic tool in Lyme disease research. Am. J. Epidemiol. 132, 58–66. Schwartz, B. S., Ford, D. P., Childs, J. E. and Thomas, R. J. (1991). Anti-tick saliva antibody: a biologic marker of tick exposure that is a risk factor for Lyme disease seropositivity. Am. J. Epidemiol. 134, 86–95.


115

Shaw, M. K. and Moloo, S. K. (1991). Comparative study on Rickettsia-like organisms in the midgut epithelial cells of different Glossina species. Parasitology 102, 193–199. Shepard, J. J., Andreadis, T. G. and Vossbrinck, C. R. (2006). Molecular phylogeny and evolutionary relationships among mosquitoes (Diptera: Culicidae) from the northeastern United States based on small subunit ribosomal DNA (18S rDNA) sequences. J. Med. Entomol. 43, 443–454. Short, H. E. and Swaminath, C. S. (1928). The method of feeding of Phlebotomus argentipes with relation to its bearing on the transmission of Kalazar. Indian J. Med. Res. 15, 827–836. Simons, F. E. and Peng, Z. (2001). Mosquito allergy: recombinant mosquito salivary antigens for new diagnostic tests. Int. Arch. Allergy Immunol. 124, 403–405. Sinkins, S. P. (2004). Wolbachia and cytoplasmic incompatibility in mosquitoes. Insect Biochem. Mol. Biol. 34, 723–729. Sly, L. M., Ho, V., Antignano, F., Ruschmann, J., Hamilton, M., Lam, V., Rauh, M. J. and Krystal, G. (2007). The role of SHIP in macrophages. Front. Biosci. 12, 2836–2848. Smith, J. J. B., Cornish, R. A. and Wilkes, J. (1980). Properties of a calcium-dependent apyrase in the saliva of the blood-feeding bug, Rhodnius prolixus. Experientia 36, 898–900. Soares, M. B. P., Titus, R. G., Shoemaker, C. B., David, J. R. and Bozza, M. (1998a). The vasoactive peptide maxadilan from sand fly saliva inhibits TNF-a and induces IL-6 by mouse macrophages through interaction with the pituitary adenylate cyclase-activating polypeptide (PACAP) receptor. J. Immunol. 160, 1811–1816. Soares, R. P. P., Gontijo, N. F., Romanha, A. J., Diotaiuti, L. and Pereira, M. H. (1998b). Salivary heme proteins distinguish Rhodnius prolixus from Rhodnius robustus (Hemiptera: Reduviidae: Triatominae). Acta Trop. 71, 285–291. Soares, R. P. P., Sant’Anna, M. R. V., Gontijo, N. F., Romanha, A. J., Diotaiuti, L. and Pereira, M. H. (2000). Identification of morphologically similar Rhodnius species (Hemiptera: Reduviidae: Triatominae) by electrophoresis of salivary heme proteins. Am. J. Trop. Med. Hyg. 62, 157–161. Soares, A. C., Carvalho-Tavares, J., Gontijo Nde, F., dos Santos, V. C., Teixeira, M. M. and Pereira, M. H. (2006). Salivation pattern of Rhodnius prolixus (Reduviidae; Triatominae) in mouse skin. J. Insect Physiol. 52, 468–472. Stark, K. R. and James, A. A. (1995). A factor Xa-directed anticoagulant from the salivary glands of the yellow fever mosquito Aedes aegypti. Exp. Parasitol. 81, 321–331. Stark, K. R. and James, A. A. (1998). Isolation and characterization of the gene encoding a novel factor Xa-directed anticoagulant from the yellow fever mosquito, Aedes aegypti. J. Biol. Chem. 273, 20802–20809. Steen, N. A., Barker, S. C. and Alewood, P. F. (2006). Proteins in the saliva of the Ixodida (ticks): pharmacological features and biological significance. Toxicon 47, 1–20. Suehiro, K., Smith, J. W. and Plow, E. F. (1996). The ligand recognition specificity of b3 integrins. J. Biol. Chem. 271, 10365–10371. Sun, J., Yamaguchi, M., Yuda, M., Miura, K., Takeya, H., Hirai, M., Matsuoka, H., Ando, K., Watanabe, T., Suzuki, K. and Chinzei, Y. (1996). Purification, characterization and cDNA cloning of a novel anticoagulant of the intrinsic pathway, (prolixin-S) from salivary glands of the blood sucking bug, Rhodnius prolixus. Thromb. Haemost. 75, 573–577. Sun, D., McNicol, A., James, A. A. and Peng, Z. (2006). Expression of functional recombinant mosquito salivary apyrase: a potential therapeutic platelet aggregation inhibitor. Platelets 17, 178–184.

116


Sweet, M. H. (1979). On the original feeding habits of the Hemiptera (Insect). Ann. Entomol. Soc. Am. 72, 575–579. Takaćˇ, P., Nunn, M. A., Me´sza´ros, J., Pechańˇova´, O., Vrbjar, N., Vlasa´kova´, P., Kozańek, M., Kazimı´rova´, M., Hart, G., Nuttall, P. A. and Labuda, M. (2006). Vasotab, a vasoactive peptide from horse fly Hybomitra bimaculata (Diptera, Tabanidae) salivary glands. J. Exp. Biol. 209, 343–352. Thiel, S. (2007). Complement activating soluble pattern recognition molecules with collagen-like regions, mannan-binding lectin, ficolins and associated proteins. Mol. Immunol. 44, 3875–3888. Thompson, J. N. (2005). Coevolution: the geographic mosaic of coevolutionary arms races. Curr. Biol. 15, R992–R994. Tilley, S. L., Wagoner, V. A., Salvatore, C. A., Jacobson, M. A. and Koller, B. H. (2000). Adenosine and inosine increase cutaneous vasopermeability by activating A3 receptors on mast cells. J. Clin. Invest. 105, 361–367. Titus, R. G., Bishop, J. V. and Mejia, J. S. (2006). The immunomodulatory factors of arthropod saliva and the potential for these factors to serve as vaccine targets to prevent pathogen transmission. Parasite Immunol. 28, 131–141. Trevejo, R. T. and Reeves, W. C. (2005). Antibody response to Culex tarsalis salivary gland antigens among sentinel chickens in California. Am. J. Trop. Med. Hyg. 72, 481–487. Trevejo, R. T., Reisen, W. K., Yoshimura, G. and Reeves, W. C. (2005). Detection of chicken antibodies to mosquito salivary gland antigens by enzyme immunoassay. J. Am. Mosq. Control Assoc. 21, 39–48. Tsimanis, A., Kalinkovich, A. and Bentwich, Z. (2005). Soluble chemokine CCR5 receptor is present in human plasma. Immunol. Lett. 96, 55–61. Ujihara, M., Horiguchi, Y., Ikai, K. and Urade, Y. (1988). Characterization and distribution of prostaglandin D synthase in rat skin. J. Invest. Dermatol. 90, 448–451. Valenzuela, J. G. (2004). Exploring tick saliva: from Biochemistry to ‘sialomes’ and functional genomics. Parasitology 129, S83–S94. Valenzuela, J. G. and Ribeiro, J. M. C. (1998). Purification and cloning of the salivary nitrophorin from the hemipteran Cimex lectularius. J. Exp. Biol. 201, 2659–2664. Valenzuela, J. G., Chuffe, O. M. and Ribeiro, J. M. C. (1996). Apyrase and anti-platelet activities from the salivary glands of the bed bug Cimex lectularius. Insect Biochem. Mol. Biol. 21, 557–562. Valenzuela, J. G., Charlab, R., Galperin, M. Y. and Ribeiro, J. M. C. (1998). Purification, cloning, and expression of an apyrase from the bed bug Cimex lectularius. A new type of nucleotide-binding enzyme. J. Biol. Chem. 273, 30583–30590. Valenzuela, J. G., Francischetti, I. M. B. and Ribeiro, J. M. C. (1999). Purification, cloning, and synthesis of a novel salivary anti-thrombin from the mosquito Anopheles albimanus. Biochemistry 38, 11209–11215. Valenzuela, J. G., Charlab, R., Mather, T. N. and Ribeiro, J. M. C. (2000). Purification, cloning, and expression of a novel salivary anticomplement protein from the tick, Ixodes scapularis. J. Biol. Chem. 275, 18717–18723. Valenzuela, J. G., Belkaid, Y., Garfield, M. K., Mendez, S., Kamhawi, S., Rowton, E. D., Sacks, D. L. and Ribeiro, J. M. C. (2001a). Toward a defined anti-Leishmania vaccine targeting vector antigens: characterization of a protective salivary protein. J. Exp. Med. 194, 331–342. Valenzuela, J. G., Belkaid, Y., Rowton, E. and Ribeiro, J. M. C. (2001b). The salivary apyrase of the blood-sucking sand fly Phlebotomus papatasi belongs to the novel Cimex family of apyrases. J. Exp. Biol. 204, 229–237. Valenzuela, J. G., Charlab, R., Gonzalez, E. C., de Miranda-Santos, I. K., Marinotti, O., Francischetti, I. M. B. and Ribeiro, J. M. C. (2002a). The D7 family of salivary proteins in blood sucking diptera. Insect Mol. Biol. 11, 149–155.


117

Valenzuela, J. G., Pham, V. M., Garfield, M. K., Francischetti, I. M. B. and Ribeiro, J. M. C. (2002b). Toward a description of the sialome of the adult female mosquito Aedes aegypti. Insect Biochem. Mol. Biol. 32, 1101–1122. Valenzuela, J. G., Francischetti, I. M. B., Pham, V. M., Garfield, M. K. and Ribeiro, J. M. C. (2003). Exploring the salivary gland transcriptome and proteome of the Anopheles stephensi mosquito. Insect Biochem. Mol. Biol. 33, 717–732. Valenzuela, J. G., Garfield, M., Rowton, E. D. and Pham, V. M. (2004). Identification of the most abundant secreted proteins from the salivary glands of the sand fly Lutzomyia longipalpis, vector of Leishmania chagasi. J. Exp. Biol. 207, 3717–3729. Van Den Abbeele, J., Caljon, G., Dierick, J. F., Moens, L., De Ridder, K. and Coosemans, M. (2007). The Glossina morsitans tsetse fly saliva: general characteristics and identification of novel salivary proteins. Insect Biochem. Mol. Biol. 37, 1075–1085. Volfova, V., Hostomska, J., Cerny, M., Votypka, J. and Volf, P. (2008). Hyaluronidase of bloodsucking insects and its enhancing effect on leishmania infection in mice. PLoS Negl. Trop. Dis. 2, 1–8. von Hundelshausen, P., Petersen, F. and Brandt, E. (2007). Platelet-derived chemokines in vascular biology. Thromb. Haemost. 97, 704–713. Waitayakul, A., Somsri, S., Sattabongkot, J., Looareesuwan, S., Cui, L. and Udomsangpetch, R. (2006). Natural human humoral response to salivary gland proteins of Anopheles mosquitoes in Thailand. Acta Trop. 98, 66–73. Warren, J. B., Loi, R. K. and Wilson, A. J. (1994). PGD2 is an intermediate in agoniststimulated nitric oxide release in rabbit skin microcirculation. Am. J. Physiol. 266, H1846–H1853. Weichsel, A., Andersen, J. F., Champagne, D. E., Walker, F. A. and Montfort, W. R. (1998). Crystal structures of a nitric oxide transport protein from a blood-sucking insect. Nat. Struct. Biol. 5, 304–309. Wigglesworth, V. B. (1936). Symbiotic bacteria in a blood sucking insect, Rhodnius prolixus Stal. Parasitology 28, 284–289. Wilson, A. D., Heesom, K. J., Mawby, W. J., Mellor, P. S. and Russell, C. L. (2008). Identification of abundant proteins and potential allergens in Culicoides nubeculosus salivary glands. Vet. Immunol. Immunopathol. 122, 94–103. Wirtz, H. P. (1988). Quantitating histamine in the saliva and salivary glands of two Palaearctic blackfly species (Diptera: Simuliidae). Trop. Med. Parasitol. 39, 309–312. Wirtz, H. P. (1990). Bioamines and proteins in the saliva and salivary glands of Palaearctic blackflies (Diptera: Simuliidae). Trop. Med. Parasitol. 41, 59–64. Xu, X., Yang, H., Ma, D., Wu, J., Wang, Y., Song, Y., Wang, X., Lu, Y., Yang, J. and Lai, R. (2008). Toward an understanding of the molecular mechanism for successful blood feeding by coupling proteomics analysis with pharmacological testing of horsefly salivary glands. Mol. Cell. Proteomics 7, 582–590. Yamazaki, Y. and Morita, T. (2004). Structure and function of snake venom cysteine-rich secretory proteins. Toxicon 44, 227–231. Yamazaki, Y., Hyodo, F. and Morita, T. (2003). Wide distribution of cysteine-rich secretory proteins in snake venoms: isolation and cloning of novel snake venom cysteine-rich secretory proteins. Arch. Biochem. Biophys. 412, 133–141. Yorkee, W. and Macfie, J. W. S. (1924). The action of the salivary secretion of mosquitoes and Glossina tachinoides on human blood. Ann. Trop. Med. Parasitol. 18, 103–108. Yoshida, S., Sudo, T., Niimi, M., Tao, L., Sun, B., Kambayashi, J., Watanabe, H., Luo, E. and Matsuoka, H. (2008). Inhibition of collagen-induced platelet aggregation by anopheline antiplatelet protein, a saliva protein from a malaria vector mosquito. Blood 111, 2007–2014.

118


Zaspel, J. M., Kononenko, V. S. and Goldstein, P. Z. (2007). Another blood feeder? Experimental feeding of a fruit-piercing moth species on human blood in the Primorye Territory of Far Eastern Russia (Lepidoptera: Noctuidae: Calpinae). J. Insect Behav. 20, 437–451. Zhang, Y., Ribeiro, J. M. C., Guimara˜es, J. A. and Walsh, P. N. (1998). Nitrophorin-2: a novel mixed-type reversible specific inhibitor of the intrinsic factor-X activating complex. Biochemistry 37, 10681–10690. Zhang, D., Cupp, M. S. and Cupp, E. W. (2002). Thrombostasin: purification, molecular cloning and expression of a novel anti-thrombin protein from horn fly saliva. Insect Biochem. Mol. Biol. 32, 321–330. Zhu, K., Dillwith, J. W., Bowman, A. S. and Sauer, J. R. (1997). Identification of hemolytic activity in saliva of the lone star tick (Acari: Ixodidae). J. Med. Entomol. 34, 160–166. Zhu, K., Bowman, A. S., Dillwith, J. W. and Sauer, J. R. (1998). Phospholipase A2 activity in salivary glands and saliva of the lone star tick (Acari: Ixodidae) during tick feeding. J. Med. Entomol. 35, 500–504.

The Enemy Within: Interactions Between Tsetse, Trypanosomes and Symbionts Deirdre P. Walshe, Cher Pheng Ooi, Michael J. Lehane and Lee R. Haines Liverpool School of Tropical Medicine, Pembroke Place, Liverpool L3 5QA, United Kingdom

1 Background 120 1.1 Human and animal trypanosomiases 120 1.2 Trypanosome species 121 1.3 Tsetse identification and distribution 122 1.4 Tsetse life cycle and physiology 123 1.5 Trypanosome (T. brucei sspp.) life cycle: Development and differentiation 126 2 Tsetse–trypanosome interactions 129 2.1 Parasite surface coat 130 2.2 Host blood factors 131 2.3 Tsetse midgut environment and signals for differentiation 133 2.4 Trypanosomes and tsetse digestive enzymes 134 2.5 Tsetse immune system 135 2.6 Effects of trypanosome infection on tsetse physiology 148 2.7 Fly sex, age and starvation and trypanosome transmission 149 2.8 Environmental temperature and trypanosome transmission 151 3 Symbiont–tsetse–trypanosome interactions 152 3.1 Wigglesworthia glossinidius 152 3.2 Wolbachia pipientis 153 3.3 Sodalis glossinidius 153 4 Towards new methods of disease control 156 4.1 Gene knockdown in Glossina 156 4.2 Paratransgenesis 158 5 Conclusion 159 Acknowledgements 160 References 160

ADVANCES IN INSECT PHYSIOLOGY VOL. 37 ISBN 978-0-12-374829-4 DOI: 10.1016/S0065-2806(09)37003-4


120

1 1.1

DEIRDRE P. WALSHE ET AL.

Background HUMAN AND ANIMAL TRYPANOSOMIASES

African trypanosomiasis refers to a set of diseases of humans and their domesticated animals, which have devastating consequences for Africa. Tsetse flies (Diptera: Glossinidae) are the sole insect vectors responsible for cyclical transmission of African trypanosomes, the protozoan parasites responsible for Human African Trypanosomiasis (HAT ¼ sleeping sickness) and African Animal Trypanosomiasis (AAT ¼ nagana). In the early part of the last century several HAT epidemics occurred on the African continent, but by the early 1960s the disease was controlled and had almost disappeared (Steverding, 2008). However, relaxation of surveillance and control measures led to resurgence in HAT, peaking in 1997 at an estimated 450,000 cases (Barrett, 2006) with an estimated 60 million people at risk in 37 countries of sub-Saharan Africa (corresponding to one third of Africa’s total land area) (WHO, 2000). Since that date, case detection and treatment have been increased, and by 2007, the reported number of new HAT cases had dropped to 10,769 (WHO, 2007), which probably equates to 50–70,000 total human cases. HAT can take two forms depending on the parasite involved. Trypanosoma brucei rhodesiense and T. b. gambiense are the causative agents of HAT in East/Southern Africa and Central/West Africa, respectively. Typically, the T. b. rhodesiense transmission cycle involves wild and domestic animals, but intensified human to human transmission may occur during epidemics. The T. b. gambiense transmission cycle is mostly from human to human, involving animals to a much lesser extent. In humans, T. b. rhodesiense infections are acute, lasting from a few weeks to several months, while T. b. gambiense infections are chronic, generally lasting for several years, often without any major signs or symptoms. There are no prophylactic drugs or vaccines available to prevent HAT. In both cases, without proper diagnosis and treatment, the outcome is fatal. Therefore, the earlier the disease is detected the better the chance of survival. However, all four drugs currently used to treat HAT exhibit toxicity and, in many countries, drug resistance is beginning to emerge (Legros et al., 2002; Delespaux and de Koning, 2007; Balasegaram et al., 2009). Other parasite species and sub-species of the Trypanosoma genus are pathogenic to many wild and domestic animal species. In particular, T. b. brucei, T. congolense and T. vivax are major causes of the animal form of trypanosomiasis. Disease severity is dependent on both the pathogenicity of the parasite strain and the genetics of the mammalian host (Courtin et al., 2008). While most African wildlife is tolerant of the parasites, domesticated livestock are highly susceptible to disease, particularly if they originate from European stock. Despite implementation of many tsetse control strategies,

THE ENEMY WITHIN

121

nagana still has a large impact on agricultural and livestock production systems and land use. The disease has an estimated annual economic cost of approximately US $4.5 billion to the African economy due to losses in milk, meat and wool yields through adult mortality, calf mortality and subsequent depressed herd growth (Kristjanson et al., 1999). Furthermore, nagana is a major restriction to the development of arable agricultural in sub-Saharan Africa, limiting the use of draught and pack animals and preventing the development of mixed agricultural practices (Jordan, 1986). Currently, nagana is managed predominantly by use of trypanotolerant breeds of cattle and use of chemoprophylactic and trypanocidal drugs (Miruk et al., 2008).

1.2

TRYPANOSOME SPECIES

Two distinct groups of insect-transmitted trypanosomes are generally recognized. The Stercocaria (subgenus Megatrypanum, Schizotrypanum and Herpetosoma) and the Salivaria (subgenus Nannomonas, Duttonella and Trypanozoon). Stercorarian trypanosomes develop in the hindgut of the insect and are transmitted in the faeces. The predominant vectors of stercorarian trypanosomes are tabanids, triatomines, leeches and ticks. It is the salivarian trypanosomes which cause nagana and African sleeping sickness. The salivarian trypanosomes develop in the anterior part of the tsetse fly alimentary canal and are transmitted via the mouthparts. Tsetse flies are the major vectors of T. brucei trypanosomes but mechanical transmission of several salivarian trypanosome species, by tabanids and Stomoxys vectors, also occurs. Only the salivarian trypanosomes exhibit antigenic variation, a unique form of parasite immune evasion within the mammalian host (Cross, 1996) and it is several of these species which are responsible for the complex of diseases known as African trypanosomiasis. T. vivax (subgenus Duttonella), believed to be the most ancient of the salivarian trypanosomes, produce a high incidence of infection in the tsetse proboscis (Haag et al., 1998) and are serious pathogens of cattle. The Nannomonas subgenus is comprised of three species: T. congolense (Broden, 1904), T. simiae (Bruce et al., 1912) and T. godfreyi (McNamara et al., 1994). Of these, T. congolense is the most economically important, with a broad host range and wide geographical distribution. T. simiae and T. godfreyi are primarily associated with suid infections; the former being extremely pathogenic (acute) and the latter producing a chronic, sometimes fatal, infection in pigs. The Trypanozoon subgenus contains the trypanosomes causing HAT (T. b. gambiense and T. b. rhodesiense) and T. b. brucei which is one of the parasites causing nagana. T. b. brucei is unable to infect humans as it is sensitive to a trypanolytic factor in human serum (Oli et al., 2006) and is thus restricted to development in domestic animals and many species of wildlife.

122


Of the African trypanosomes, T. congolense, T. b. brucei, T. b. gambiense and T. vivax have been selected for partial or complete genome sequencing by a consortium of institutes: The Institute for Genome Research (TIGR), now the J. Craig Venter Institute (Rockville, MD), and the Wellcome Trust Sanger Institute (Hinxton, UK). All genomes are in various states of completion. At the time of writing, the nuclear genome of T. brucei (TREU927 GUTat 10.1), the model species most often used for studying trypanosome biology, had been completed and the T. b. gambiense (MHOM/CI/86/DAL972) genome is in the last stages of finishing, with 8 coverage. Also, the partial nuclear genome sequence (5 coverage) of T. congolense (IL3000) and T. vivax are being assembled (http://www.sanger.ac.uk/Projects/Protozoa/).

1.3

TSETSE IDENTIFICATION AND DISTRIBUTION

The name tsetse (pronounced tsee–tsee) is derived from the noise the fly creates when it raises its body temperature by contraction of flight muscles decoupled from the wings, prior to energetically demanding events (e.g. birth of the larva, flight, rapid dehydration of the bloodmeal). Interestingly, tsetse means ‘‘fly’’ in the Tswana language and in Sechuana it is interpreted as ‘‘fly destructive to cattle’’. Tsetse flies are easily distinguishable from other insects (Fig. 1). They are light brown to black in colour and, dependant on the species, are roughly twice the size of a housefly. Characteristic aristae are present on the third antennal segment (Fig. 1A). In addition, the unique ‘‘hatchet’’ wing cell is found in the centre of each wing between the fourth and fifth veins (Fig. 1B). Also, tsetse flies adopt a characteristic resting attitude with their single pair of wings folded scissor-like over the dorsal surface of the abdomen. A

B

FIG. 1 Characteristic anatomical features of Glossina sspp. (A) Side view of an engorged female Glossina morsitans morsitans resting during diuresis. An anal droplet has started to form within minutes of feeding. The characteristic arista (arrow), a thin structure bearing a single unidirectional row of branched setae, is used for species identification. Photo: R. Wilson http://www.raywilsonbirdphotography.co.uk. (B) The middle of the wing contains a unique venation pattern resembling a hatchet, which is also characteristic of tsetse flies. Photo: L. Rafuse Haines.

THE ENEMY WITHIN

123

Tsetse flies fall into a single genus, Glossina, and are restricted to subSaharan Africa except for two localities in the Arabian peninsula. Twentythree species and eight sub-species of tsetse fly are currently recognized (Leak, 1999; Krafsur, 2009). These are divided into three distinct clades: Morsitans, Palpalis and Fusca, which are named after the best known species in each subgenus. Commonly these groups are described by the ecological niches they occupy; savanna (Morsitans), riverine (Palpalis) or forest (Fusca) groups. In Central and West Africa, the riverine species (Palpalis group) tend to feed predominantly on reptiles and ungulates. Humans regularly encounter these flies, particularly when visiting water sources, and these species are important vectors of human sleeping sickness. The savannah-woodlands species (Morsitans group) are the most economically important, as they preferentially feed on livestock and wildlife and are the major vectors of nagana. Both the Palpalis and Morsitans groups are vectors of T. brucei sspp. Most tsetse from the third clade (Fusca group) inhabit the damp, evergreen forests. The exception is G. brevipalpis, which is more regularly found in association with livestock. With the exception of G. brevipalpis, flies in the Fusca group are not considered to be medically or agriculturally important. Whether this will change when further pressure on land use drives them into more regular encounters with humans and their domesticated animals remains to be seen.

1.4

TSETSE LIFE CYCLE AND PHYSIOLOGY

Tsetse are unique among insect disease vectors in that they possess a viviparous lifestyle. Consequently, flies have a very low rate of reproduction, typically producing 8–10 offspring in their lifetime in optimal laboratory conditions (Leak, 1999; Attardo et al., 2006). The tsetse reproductive tract possesses extensive modifications to permit intrauterine larval development, which include a reduced number of ovarioles per ovary (two), a highly tracheated and muscular uterus and a modified uterine accessory gland (milk gland) to supply nutrients to the developing larvae (Leak, 1999; Attardo et al., 2006). Oogenesis begins before tsetse eclosion. A single oocyte develops at a time, starting with one of the two ovarioles in the right ovary, and takes 6–7 days to complete. Oogenesis appears to be regulated by the presence of a developing embryo or larvae in the uterus. In most Glossina species, the female is sexually mature 48–72 h posteclosion (i.e. emergence of the adult insect from the puparium), while males become fertile several days after eclosion. Female flies generally mate only once and can store sperm for the duration of their life. Upon completion of oogenesis, an oocyte is synchronously ovulated and fertilized in the uterus where it undergoes embryonic and larval development. The larva is solely nourished by a milk-like secretion rich in proteins and lipids produced by a pair of milk glands. A fully developed third instar larva is deposited by the

124


female approximately 16 days after fertilization in the case of the first offspring, and every 9 days thereafter. The larva burrows into the ground where rapid pupariation takes place, and the adult fly emerges from the puparium approximately 4 weeks later. In the tsetse field, the term teneral is used to describe the period of time following emergence of the adult fly from the puparium until it takes its first bloodmeal. Both male and female adult tsetse flies are obligate haematophages capable of transmitting trypanosomes. Tsetse flies are pool feeders and the repeated penetration of mammalian host tissue by the tsetse proboscis results in the formation of a sub-surface blood pool. Saliva is expressed into the wound and trypanosomes are transmitted to the mammalian host at this stage. The bloodmeal is sucked up through the proboscis and oesophagus and propelled into the rest of the alimentary canal by the rhythmic pumping of the cibarial pump aided by the contraction of circular muscles that encompass the oesophagus. The proventriculus (¼cardia) (Fig. 2C) lies at the junction of the oesophagus, midgut and crop duct. Blood may pass directly into the midgut or into the extension of the foregut known as the crop (Fig. 2A), before regurgitation into the midgut (Moloo and Kutuza, 1970). The proventriculus acts as a valve regulating the directional flow of blood and is also the organ responsible for producing the peritrophic matrix (PM) (see Section 2.5.1). The tsetse midgut is a simple tube, lacking diverticula, running from the proventriculus to the junction with the hindgut, which is marked by the entrance of the Malpighian tubules into the alimentary canal (Fig. 2, top panel and Fig. 2J). Although more complex divisions exist (Bo¨hringer-Schweizer, 1977), the midgut can be crudely separated into three functional regions: the anterior midgut, bacteriome (¼mycetome) and posterior midgut. The first part of the anterior midgut is a linear tube running though the thorax. Once it enters the abdomen, the anterior midgut becomes distended and here the blood is stored and dehydrated prior to digestion. Epithelial cells of the anterior midgut possess extensive infoldings of the basal plasma membrane with associated mitochondria, enabling the fly to achieve this rapid dehydration (Bo¨hringerSchweizer, 1977). The anterior midgut is interrupted approximately in its middle section by a region of cells called the bacteriome (Fig. 2E). The bacteriome houses the intracellular symbiotic bacteria Wigglesworthia glossinidius. Haemolysis and bloodmeal digestion commence at the very beginning of the posterior midgut, where haemolytic agents and digestive enzymes are produced. Virtually all proteolytic enzymes are restricted to the posterior midgut (Gooding, 1974a). The junction of the anterior and posterior midguts (Fig. 2F) is obvious and tightly delineated in fed flies, as the bloodmeal changes in colour from red to brown/black as haemolysis and digestion progesses. Cells of the proximal part of the posterior midgut possess extensive rough endoplasmic reticulum, large numbers of Golgi bodies and secretory vesicles allowing efficient production, storage and secretion of proteins (Bo¨hringer-Schweizer, 1977).

THE ENEMY WITHIN

125

A Posterior Midgut

Anterior Midgut

F B

Hindgut

I

C

J K

D E

L

G H

A

B

C

D

E

F

G

H

I

J

K

L

FIG. 2 Anatomy of the alimentary canal of Glossina morsitans morsitans. Top panel represents a dissected alimentary canal aligned anterior (left) to posterior (right). The letters (A–L) correspond to the lower panels, which are magnified regions of the digestive system visualized using light microscopy; (A) the crop, (B) salivary glands, (C) the proventriculus capped by a modified fat body crown (oenocytes), (C–D) the most proximal part of the anterior midgut that lies in the thorax, (D) the point where the anterior midgut enters the abdomen, (E) the bacteriome that contains the obligate symbiont Wigglesworthia glossinidia, (F–G) the most distal part of the anterior midgut and the junction where the anterior and posterior midgut meet, demarked by a bloodmeal colour change (from red to black), (H) fat body attached to the wall of the posterior midgut, (I) the distinctive posterior midgut epithelial cells (luminal surface), (J) the bases of the Malpighian tubules leave the alimentary canal at the junction of the midgut and hindgut. The two Malpighian tubules each branch near their base to form a total of four tubules, (K) ileum (severed from the MT and midgut) and the junction with the colon, (L) colon ending in the bulbous rectum; note peritrophic matrix (*) exuding from a breach in the hindgut wall. Photo: L. Rafuse Haines.

Cells of the distal part of the posterior midgut are involved in absorption of digested products. They possess extensive smooth endoplasmic reticulum and also lipid droplets and glycogen at various times of the digestive cycle (Bo¨hringer-Schweizer, 1977).

126

1.5

DEIRDRE P. WALSHE ET AL. TRYPANOSOME (T. BRUCEI SSPP.) LIFE CYCLE: DEVELOPMENT AND DIFFERENTIATION

T. brucei has the most complex, but perhaps the best characterized, life cycle of all African trypanosome species. The trypanosome life cycle was first described in detail by Muriel Robertson who described the successive stages of parasite establishment and maturation within the insect and mammalian hosts, demonstrated the migration of parasites through the fly midgut and proved that only salivary gland forms were capable of producing a mammalian infection (Robertson, 1913). Since then, a more complete understanding of trypanosome development has been achieved, with an agreed parasite nomenclature adopted (Roditi and Clayton, 1999) and a consensus achieved on many of the barriers present in the fly that the trypanosome must overcome to survive and develop in order to complete cyclical transmission. Within the vertebrate bloodstream at least two different major forms of trypanosomes are found; a long slender form, which replicates by asexual division, and a short stumpy, non-replicating form (Fig. 3(1)). These extracellular parasites are covered with an immunogenic surface coat composed of approximately 107 identical variant surface glycoprotein (VSG) molecules (Vickerman, 1969; Barry and McCulloch, 2001; Barry et al., 2005). The VSG coat physically shields underlying membrane proteins from host immune responses and is central to antigenic variation and survival in the mammalian host. The consecutive, but mainly unpredictable, expression of a large repertoire of VSG genes permits expansion of antigenically distinct trypanosome populations within the host. After activation of host immune responses (in reaction to high parasitaemia), the majority of the parasite population is destroyed. A small number of trypanosomes survive because they express an antigenically distinct VSG coat, and proceed to expand in numbers. The continuous cycles of trypanosome replication and destruction result in waves of fluctuating parasitaemia. The differentiation of the long slender bloodstream form (BSF) into the non-dividing stumpy BSF occurs in high density populations of long slender BSFs (Vassella et al., 1997; Seed and Wenck, 2003). The switch to stumpy BSF involves changes in metabolism within the trypanosome, but the molecular signals involved are not yet known. Short stumpy BSFs are believed to be pre-adapted for survival within the insect midgut due to the presence of a functional mitochondrion. In the vertebrate bloodstream, trypanosomes utilize glucose as an energy source. However, in the fly midgut, glucose is limiting and a more efficient utilization of glucose and amino acids occurs via the Krebs cycle and oxidative phosphorylation in the mitochondrion. To study the early events of trypanosome establishment in the tsetse midgut, T. b. brucei trypanosomes expressing green fluorescent protein (GFP) under the control of a procyclin promoter have been created (Gibson and Bailey, 2003). It is evident that differentiation of the BSF to the procyclic form (i.e. the insect

THE ENEMY WITHIN

127 5. Asymmetric division

4. Proventriculus

LE

6. Salivary glands M

SE

ADT

SE Ms

LT

ADT

Peritrophic matrix Epithelium SS Ms

P

LS

1. Bloodmeal

3. Ectoperitrophic space P

SS

2. Endoperitrophic space

FIG. 3 Diagram of the life cycle stages of T. b. brucei within the tsetse fly. Trypanosome morphology was based on observations by Van den Abbeele et al. (1999) and Sharma et al. (2008). Stages of the trypanosome life cycle within the tsetse are indicated within circles joined by the red arrow; the grey arrow represents transit between insect and mammalian host. 1. Short stumpy (SS) and long slender (LS) bloodstream forms (BSFs) ingested with the infective bloodmeal; 2. Transformation of short stumpy (SS) BSF into procyclics (P) within the midgut endoperitrophic space; 3. Procyclics differentiate into mesocyclics (Ms) within the ectoperitrophic space; 4. Transformation of mesocyclics (Ms) into long trypomastigotes (LT), which give rise to asymmetrically dividing trypomastigotes (ADT); 5. Trypomastigotes (ADT) divide into long epimastigotes (LE) and short epimastigotes (SE). These three stages are present in both the proventriculus and salivary glands; 6. Short epimastigotes (SE) interdigitate within the salivary gland epithelium at their anterior ends, and subsequently mature into the nondividing, mammalian infective metacyclics (M). Note the gradual migration of the kinetoplast in the fly stages of the parasite, culminating in a position anterior to the nucleus in the trypomastigote and epimastigote forms. The reversion of this position to the posterior end of the parasite is a signature of mature metacyclics.

midgut-adapted form), which involves replacement of surface VSGs by procyclins, occurs rapidly after bloodmeal ingestion by the fly (Fig. 3(2)) (Vassella et al., 2001; Acosta-Serrano et al., 2001; Gibson and Bailey, 2003). Most flies successfully kill all invading trypanosomes in a process termed self-cure.

128


For the first 3 days, trypanosomes are mostly contained within the bloodmeal as it is being digested. The critical events in parasite establishment appear to occur approximately 3 days after infection, when the relatively small proportion of surviving trypanosomes ( 10%) either die or rapidly multiply in number (Gibson and Bailey, 2003). Typically, from 8 days after the infected bloodmeal, dissected flies can be confidently divided into two groups; the first in which most flies will have self-cured (having completed the clearing of ingested trypanosomes from their midguts) and the second which have established midgut infections. Trypanosomes in an established infection migrate to the ectoperitrophic space 3–5 days post-infection (Gibson and Bailey, 2003) (Fig. 3(3)). It is believed that this occurs by direct penetration through the PM (Ellis and Evans, 1977; Gibson and Bailey, 2003) although an alternative but less likely, suggestion is that it occurs by circumnavigation around the open, posterior end of the PM in the hindgut (see Section 2.5.1). Typically the midgut population in an established infection reaches approximately 5 105 trypanosomes (Van den Abbeele et al., 1999; Gibson and Bailey, 2003). From 6 to 8 days post-infection, large numbers of trypanosomes congregate within the proventriculus (Van den Abbeele et al., 1999; Gibson and Bailey, 2003; Sharma et al., 2008) (Fig. 3(4)). Here they appear to cease division, elongate to mesocyclic forms and later differentiate into long trypomastigotes (Fig. 3(4)) (Van den Abbeele et al., 1999). Trypanosomes then migrate back into the endoperitrophic space by actively penetrating the PM and move anteriorly in the lumen of the foregut to the opening of the hypopharynx at the tip of the proboscis. An alternative theory of migration involves the direct penetration of the tsetse salivary glands after trypanosomes have traversed the fly haemolymph (Mshelbwala, 1972). It is generally accepted that this is unlikely, as trypanocidal factors known to be present in the haemolymph (Croft et al., 1982) would act as a major barrier for trypanosomes attempting to traverse it. Early positioning of trypanosomes in the anterior midgut and proventriculus should also favour passage along the foregut to the salivary glands (Peacock et al., 2007). Asymmetric division of the proventricular epimastigote form generates both long and short parasites (Fig. 3(5)) and it is either the asymmetrically dividing trypanosome or the short epimastigote that arrives at the salivary gland (Sharma et al., 2008). Each tsetse fly has two salivary glands. Evidence suggests that each gland is invaded and colonized separately, with few epimastigotes constituting the founder populations (Peacock et al., 2007). The short epimastigote forms are believed to attach to the salivary gland epithelium by interdigitation of their membranes (Fig. 3(6)). Upon binding, the non-infective epimastigotes complete several rounds of replication and differentiate into the metacyclic form. Differentiation (metacyclogenesis) includes the appearance of a VSG surface coat. Metacyclic VSGs display a specific VSG repertoire subset and their expression is regulated differently to bloodstream VSGs (Barry et al., 1998; Graham et al., 1999). Mitochondrial changes also occur, including loss of mitochondrial cristae and Krebs cycle enzymes. The biochemical changes

THE ENEMY WITHIN

129

accompany the posterior migration of the kinetoplast before the parasite detaches into the lumen as a mature, free-form, infective metacyclic trypomastigote. At this point, each mature metacyclic parasite has undergone the transformations necessary for survival in a mammalian host. The above is an account of the most complex trypanosome life cycle completed in the tsetse fly and it should be noted that there are distinct differences between the life cycle of T. brucei sp. trypanosomes and other African trypanosome species. T. vivax completes its entire life cycle in the proboscis of the fly while T. congolense infections develop in the fly midgut and mature in the hypopharynx. The number of flies that develop a mature infection and the length of time required for an infection to establish and mature into a transmissible form can vary depending on several factors, including fly species (Welburn et al., 1989, 1994), fly sex (Distelmans et al., 1982; Dale et al., 1995) (see Section 2.7) and parasite strain (Dale et al., 1995). Recent evidence suggests migration of trypanosomes from the proventriculus to the salivary glands may occur continuously (Peacock et al., 2007; Sharma et al., 2008), albeit at low numbers, as opposed to being restricted to a limited time ‘‘window’’ (Van den Abbeele et al., 1999). Once a fly is infected it will produce infective metacyclics for the duration of its life, which can be 150 or more days for females and about half that for males (Lehane and Mail, 1985; Msangi et al., 1998). Thus, there is potential for infective parasites to be transmitted every time a fly feeds on a new host. In addition, flies are capable of harbouring mixed infections of two or more species of trypanosome (Lehane et al., 2000; Van den Bossche et al., 2004; Kubi et al., 2005; Peacock et al., 2007). Sexual reproduction has been reported in T. b. brucei in the salivary glands and may exist in other trypanosome species (Jenni et al., 1986; Peacock et al., 2007; Gibson et al., 2008). However, mating is not an obligatory part of the trypanosome life cycle and occurs in only a proportion of flies co-infected with two different strains of T. b. brucei (Jenni et al., 1986; Schweizer et al., 1988). It is suggested that the unattached epimastigote is the mating stage (Gibson et al., 2008) (Fig. 3) but the mechanism of genetic exchange is not yet known. Under field conditions mating may be a rare event (Koffi et al., 2009).

2

Tsetse–trypanosome interactions

It appears from the established literature that all tsetse species are susceptible, to some degree at least, to trypanosome infections. In general, tsetse in the Palpalis group species tend to be poor vectors of congolense-type trypanosomes compared to the Morsitans group flies (Harley and Wilson, 1968; Moloo and Kutuza, 1988a; Ndegwa et al., 1992). Conversely, tsetse of the Morsitans group are poorer vectors of T. b. gambiense than the Palpalis group (Richner

130


et al., 1988). Care needs to be taken with much of the data on susceptibility, as often fly and trypanosome strains used in experiments are from widely divergent geographical origins. Many factors influence fly susceptibility to trypanosome infection. Our understanding of these factors and their underlying mechanisms is still rudimentary. In this review, we concentrate on the physiological factors influencing tsetse fly susceptibility to trypanosomes. 2.1

PARASITE SURFACE COAT

Throughout their life cycle in the tsetse fly, trypanosomes are covered by different surface molecules. Successful differentiation of BSF to procyclic forms involves the shedding of surface VSG and replacement with a set of new surface molecules, the best known being the procyclins (Roditi and Pearson, 1990; Beecroft et al., 1993). The procyclins are major GPI-anchored proteins possessing extensive C-terminal glu-pro (EP) or gly-pro-glu-glu-thr (GPEET) repeats, and comprise a surface coat of approximately three million procyclin molecules per cell (Roditi et al., 1998; Pays and Nolan, 1998; Roditi and Lehane, 2008). Three isoforms of EP procyclin, EP1, EP2 and EP3, are known, which differ in the length of their repeats, the sequence of their N-terminal domains and the presence or absence of N-glycosylation sites. Analysis of procyclin expression by mass spectrometry indicated that trypanosomes exhibit distinct procyclin expression profiles in vivo. All four procyclins are present at similar levels within a few hours of differentiation to procyclics (Vassella et al., 2001). However, 3 days post-infection, the procyclic coat consisted primarily of GPEET with lower amounts of EP forms (AcostaSerrano et al., 2001). This GPEET was 11 residues shorter than that of in vitro cultured procyclic culture forms (PCFs). From 7 days post-infection, GPEET procyclin disappeared and expression switched to the glycosylated isoforms EP1 and EP3. EP1 and EP3 were also truncated in comparison to these isoforms expressed by in vitro cultured PCF. Transcripts of EP2 are quite abundant in fly-derived procyclic forms but EP2 proteins have not yet been confirmed in vivo (Urwyler et al., 2005). While tsetse procyclins exhibit complex expression profiles (suggesting a selective advantage in their production) their function remain unknown. Proteolysis in the fly removes the N-terminal domains of all procyclins, but the acidic amino acid repeat sequences are largely resistant to proteolysis (Acosta-Serrano et al., 2001). Thus procyclins may serve a protective role, aid parasite development and/or influence ligand-associated parasite–vector signalling (Roditi and Pearson, 1990; Ruepp et al., 1997). Interestingly, parasites expressing procyclins with truncated N termini can still establish midgut infections to similar levels as wild-type parasites under laboratory conditions and form mature metacyclics within the salivary glands (Liniger et al., 2004). Thus, procyclins are not crucial for migration of procyclic forms from the tsetse midgut to the salivary glands (Vassella et al., 2009). However, in co-infection experiments,

THE ENEMY WITHIN

131

procyclin null mutants were rapidly outgrown in the midgut by wild-type parasites. This suggests that under field conditions, where mixed infection in flies is common (Lehane et al., 2000), surface procyclins probably play a significant role in trypanosome fitness (Vassella et al., 2009). As trypanosome epimastigotes within the tsetse salivary glands typically lack a procyclin coat (Urwyler et al., 2005), the function of the procyclin coat molecules probably occurs earlier in trypanosome development in the fly. T. congolense PCFs express several major stage-specific surface molecules different to those of T. brucei. In the early stage of T. congolense midgut infection, a protease-resistant surface molecule (PRS) is expressed, while later, a heptapeptide-repeat containing molecule (Butikofer et al., 2002) and a glutamic acid–alanine-rich protein (GARP) are expressed (Beecroft et al., 1993; Bayne et al., 1993; Utz et al., 2006). Interestingly, T. congolense procyclins bear similarity to T. brucei procyclins in being acidic and having repeat sequences rich in glutamic acid, glycine, threonine and proline (Utz et al., 2006). Also, genes related to GARP have been identified in both T. simiae and T. godfreyi. Epimastigote forms of T. congolense and T. brucei, found in the proboscis and salivary glands respectively, express related surface glycoproteins named congolense epimastigote specific protein (CESP) (Inoue et al., 2000) and brucei alanine-rich proteins (BARP) (Urwyler et al., 2005). The function of these different parasite surface molecules is still unknown. In the salivary glands nascent and mature metacyclics (re)acquire a VSG coat in preparation for transfer to the next mammalian host (Tetley and Vickerman, 1985). 2.2

HOST BLOOD FACTORS

Different trypanosome species are capable of infecting different mammalian hosts, which implies that host blood factors can affect trypanosome survival (Vickerman, 1985; Masaninga and Mihok, 1999). An intriguing question is whether or not the influence of these host blood factors extends to trypanosome establishment in the tsetse midgut. T. brucei developmental forms are present in the midgut for at least 8 days, before commencing migration and maturation in the foregut and salivary glands. As tsetse flies ingest a new bloodmeal every 24–48 h, there is ample opportunity for an ingested bloodmeal to have an impact on trypanosome survival. The mammalian host blood in which the trypanosome infection is acquired by tsetse affects both differentiation (Nguu et al., 1996) and multiplication of parasites in the fly midgut (Olubayo et al., 1994; Mihok et al., 1995). Buffalo and eland blood support establishment of T. congolense infections in G. m. morsitans less well than goat blood (Olubayo et al., 1994). This was apparent from 3 days after the infective bloodmeal and occurred regardless of whether flies were maintained on rabbit blood or the blood of the respective mammalian hosts in which the infective meal was given (Olubayo et al., 1994).

132


This suggests that the trypanocidal event in the fly midgut occurs early in the infection process. Serum components of the blood may be either conducive or detrimental to survival of trypanosomes in the fly (Reduth et al., 1994; Mihok et al., 1995; Muranjan et al., 1997). The serum of the Cape buffalo, for example, is trypanocidal to most species of trypanosomes (Reduth et al., 1994). Using T. b. brucei BSF, the primary cause of parasite death was inhibition of glycolysis by host hydrogen peroxide (H2O2) generated by xanthine oxidase (Muranjan et al., 1997). Whether or not this effect holds true for procyclic forms if a recently infected fly subsequently feeds on Cape buffalo merits a more detailed study. However, it is unlikely the same trypanocidal mechanism would occur in vivo since procyclic trypanosomes metabolize proline, rather than glucose, as a major energy source (Muranjan et al., 1997). While xanthine oxidase is present in the serum of other mammals, it does not cause death of different trypanosome species in these hosts (Cruz et al., 1983; Reduth et al., 1994). Therefore, it is possible that generation of H2O2 to trypanocidal levels by xanthine oxidase in Cape buffalo may be part of a wider metabolic cascade, including purine catabolizing enzymes (Muranjan et al., 1997). Thus, experiments using the serum of different mammalian species, which show seemingly little correlation between serum concentration and BSF trypanosome survival, may simply differ in the molecular constituents of different mammalian host serums (Black et al., 1999). Host serum complement (an immunological cascade present in mammalian serum, which, upon activation, induces lysis of target microbes) may also be involved in causing trypanosome mortality (Ferrante and Allison, 1983; Black et al., 1999). A BSF trypanocidal factor, which is heat labile (inactivated following 30 min at 56 C), has been documented in the serum of mammalian species lacking any xanthine oxidase activity (Black et al., 1999). Also, T. congolense and T. b. brucei PCFs are killed by another heat sensitive factor in human serum (Ferrante and Allison, 1983). Similar PCF lytic activity was observed in human serum deficient in complement C2 (part of the classical complement pathway), as well as in ethylene glycol tetra-acetic acid (a preferential chelator of calcium ions, a crucial substrate for initiation of the classical pathway). Consequently, the trypanocidal factor may rely on the alternative complement pathway for activation (Ferrante and Allison, 1983). The infective bloodmeal origin also appears to play a key role in determining the virulence and infectivity of the resulting salivary gland metacyclics (Masaninga and Mihok, 1999). In vitro grown T. congolense BSFs were mixed with either goat, eland or zebra blood and fed to G. m. centralis flies. The resulting metacyclics used to infect BALB/c mice varied in their prepatent periods and resulted in differences in mouse mortality rates (Masaninga and Mihok, 1999). A strong selection pressure appears to be exerted by host blood, most likely on the procyclic parasite stage exposed to incoming bloodmeals, which may lead to differential killing of parasites.

THE ENEMY WITHIN

2.3

133

TSETSE MIDGUT ENVIRONMENT AND SIGNALS FOR DIFFERENTIATION

Efforts to determine the triggers responsible for trypanosome switching from bloodstream to procyclic forms have primarily been conducted in vitro and have occasionally ignored the realities of tsetse fly biology. A variety of methods have been used to define differentiation including morphological markers (Hunt et al., 1994), the release of VSG proteins and concomitant detection of procyclins, or the detection of DNA synthesis (Matthews and Gull, 1997). Other novel approaches include the use of transgenic trypanosomes carrying the E. coli glucuronidase (GUS) gene under the control of procyclin expression elements (Sbicego et al., 1999) and the use of GFP-expressing trypanosomes under the regulation of a procyclin-linked promoter (Sheader et al., 2004). Reduction of temperature alone from 37 to 27 C can initiate the differentiation of BSFs to PCFs (Brown et al., 1973; Bienen et al., 1980; Overath et al., 1986) and is conducive to growth and proliferation of a transformed trypanosome population. Using the GUS system, Sbicego et al. (1999) found that both citrate and cis-aconitate (Krebs cycle intermediates) could induce trypanosome differentiation to PCF in vitro, in a manner independent of the commonly used temperature shift (from 37 to 27 C). This phenomenon appears to be highly specific as other Krebs cycle intermediates, even when present at relatively high concentrations in vitro (10 mM), did not trigger parasite differentiation (Sbicego et al., 1999). Also, compounds closely resembling citrate and cis-aconitate, for example trans-aconitate and 5-fluoro-citrate, were trypanocidal (Hunt et al., 1994). To date, the citrate concentration of the fly midgut lumen has not been reported. However, if similar to the low citrate concentration estimated within the fly haemolymph (15.9 mM), this may be insufficient to trigger BSF differentiation (Hunt et al., 1994). However, BSFs exhibit hypersensitivity towards cis-aconitate when they are exposed to a cold shock (Engstler and Boshart, 2004). Thus lower concentrations of citrate and cis-aconitate may be sufficient to trigger differentiation of BSFs in vivo when used in combination with a temperature drop (Hunt et al., 1994; Rolin et al., 1998; Fenn and Matthews, 2007). Trypanosome differentiation may also be initiated in vitro by growth of BSFs in low glucose medium (Milne et al., 1998). This may reflect what occurs in vivo, as changes in glucose concentration occur as the trypanosomes transfer from a high glucose environment (mammalian bloodstream) to a lower glucose environment (fly midgut). Furthermore, glucose concentration may be a factor involved in controlling switches in procyclin expression at later stages of trypanosome development (Morris et al., 2002). Incubation of BSFs under mildly acidic conditions (pH 5.5) was shown to cause differentiation to PCFs in vitro (with corresponding shedding of the VSG coat and expression of procyclins) at 27 C, even in the absence of cis-aconitate (Rolin et al., 1998). However, this in vitro work seems to have ignored conditions in the tsetse fly, where the best available data suggest that the pH of the midgut is highly alkaline. The use of microelectrodes to measure the midgut pH

134


in unfed flies and fed flies 48 and 72 h post-feed has shown that the tsetse digestive tract average pH lies between 9 and 9.5, with the lowest pH value (pH 7.9) occurring in the posterior midgut 72 h post-feeding (Liniger et al., 2003). The enzymes trypsin (see Section 2.4) and thermolysin stimulate trypanosome differentiation in vitro. In the case of trypsin, at least, this occurs independently of any slender BSF attrition effects (Sbicego et al., 1999). Pronase was also shown to be an efficient trypanosome differentiation trigger, with up to 95% differentiation to PCFs achieved (Hunt et al., 1994). However, as far as we are aware, of the proteases above only trypsin is found in the tsetse midgut (Gooding and Rolseth, 1976). Another molecular trigger native to the tsetse midgut, the Glossina proteolytic lectin serine protease (Gpl) isolated from Glossina fuscipes fuscipes, has been identified as having a role in trypanosome differentiation in vitro (Abubakar et al., 2006). To date, while several triggers of differentiation from BSF to PCFs have been identified, the signals that cause differentiation to epimastigote and metacyclic forms in vivo are unknown. 2.4

TRYPANOSOMES AND TSETSE DIGESTIVE ENZYMES

Tsetse flies are obligate haematophages and thus are completely specialized for digesting blood. The majority of their nutritional resources are obtained from the protein content of the bloodmeal (Moloo, 1976; Kabayo and Langley, 1985). Consequently, the fly has a range of proteolytic digestive enzymes in the posterior, digestive portion of the midgut (Gooding and Rolseth, 1976; Cheeseman and Gooding, 1985). The anterior portion is virtually free of proteolytic activity and indeed the epithelium of the anterior midgut actively secretes proteinase inhibitors into the gut lumen (Gooding, 1974b; Houseman, 1980; Stiles et al., 1991). It has previously been suggested that midgut proteases are a major barrier to trypanosome establishment within the tsetse midgut (Imbuga et al., 1992). Given the above, while this may be the case, it probably only refers to trypanosomes entering the posterior midgut. In addition, fly midgut trypsin levels do not differ in infected and refractory flies, suggesting that digestive enzymes are not part of a protective response to trypanosome infection. Neither feeding excess trypsin nor inhibition of trypsin results in any difference in infection phenotype within the fly (Welburn and Maudlin, 1999). Studies using tsetse midgut extracts have found no correlation between tsetse gut protease activity and trypanosome infection rates (Mihok et al., 1994). Collectively these data indicate that it is questionable whether tsetse proteases have any major bearing on trypanosome infections within the fly gut. Curiously, an increase in mRNA transcript levels of certain digestion related enzymes, namely the tsetse cathepsin B, zinc carboxypeptidase and zinc metallo-protease, have been reported when a trypanosome infection is present (Yan et al., 2002). It should be noted however that protein levels, particularly in the intestine of insects taking occasional meals, may not be proportional to transcript levels.

THE ENEMY WITHIN

2.5

135

TSETSE IMMUNE SYSTEM

Despite their obvious efficiency in maintaining large burdens of trypanosomebased disease in Africa, tsetse flies exhibit a considerable level of refractoriness to trypanosome infection. Even under optimal laboratory conditions, where flies are fed at regular intervals, only a proportion of flies will establish midgut infections and the number decreases dramatically after the adult fly has taken three to four bloodmeals (Fig. 4) (Distelmans et al., 1982; Welburn and Maudlin, 1992; Kubi et al., 2006). Furthermore, less than half of the infections that become established in the midgut will mature (Van den Abbeele et al., 1999; Gibson and Bailey, 2003; Peacock et al., 2006). A key factor in this refractoriness is the fly immune system (Hao et al., 2001). Immune stimulation, by injection of live E. coli or lipopolysaccharide (LPS) into the haemocoel of the fly prior to feeding an infective bloodmeal, leads to a statistically significant decrease in trypanosome midgut infection rates (Hao et al., 2001). Identification of the tsetse immune molecule(s) responsible for conferring resistance to trypanosome infection has been hampered by the lack of an annotated Glossina genome (scheduled for completion in 2011). However, the sequencing and annotation of EST libraries from several tissue sources,

100

Trypanosome prevalence (%)

90 80 70 60 50 40 30 20 10 0

1st BM n = 307 r = 11

2nd BM n = 165 r=5

3rd BM n = 333 r = 12

4th BM n = 117 r=4

5th BM n = 100 r=2

8th BM n = 52 r=2

10th BM n = 35 r=1

FIG. 4 G. m. morsitans teneral phenomenon. The timing of the infective bloodmeal (BM) affects the prevalence of establishment of trypanosomes in the tsetse midgut. The infective BM containing T. b. brucei (TSW196) BSF given at one of the stated bloodmeals (x-axis) results in the indicated prevalence of infection (mean SE). N ¼ cumulative sample size number (male), R ¼ number of replicates (Lehane laboratory, unpublished results).

136


including the major immunoresponsive tissues of midgut (Lehane et al., 2003) and fat body (Attardo et al., 2006), has provided the foundation for more extensive studies of the Glossina innate immune system at a molecular level (http://www.genedb.org/genedb/glossina/index.jsp). Based largely on work on Drosophila melanogaster, it is known that insects possess a complex, interacting, innate immune system. This system is comprised of physical barriers (such as the cuticle and the PM), cellular responses (such as encapsulation and phagocytosis), and humoral responses, such as the generation of host defence peptides (HDP, previously called antimicrobial peptides), reactive oxygen species (ROS) and melanization by the phenoloxidase pathway (Lemaitre and Hoffmann, 2007). The immune response depends not only on the nature of the immune stimulus, but also its route of delivery, with quite distinct epithelial and systemic immune response profiles generated to the same pathogen (Hao et al., 2001). Clearly, in tsetse–trypanosome interactions, it is the epithelial immune responses of the alimentary canal and salivary gland tissues that are likely to be of major importance, as trypanosomes involved in the natural life cycle are exposed only to epithelial surfaces throughout the parasite life cycle. Trypanosomes have been reported in the tsetse haemocoel (Mshelbwala, 1972; Otieno et al., 1976), but these are almost certainly not important to the completion of the normal life cycle. Those trypanosomes that do traverse the midgut epithelium are rapidly killed by an unidentified systemic immune response (Croft et al., 1982), which effectively confines the trypanosomes to the lumen of the alimentary canal.

2.5.1

Peritrophic matrix

The insect midgut epithelium is physically protected from abrasion by food (and the pathogens it may contain) by the peritrophic matrix (PM, previously known as the peritrophic membrane) (Lehane, 1997). The PM is composed of a highly organized, glycosaminoglycan-rich layer reinforced with chitin (Tellam et al., 1999). In tsetse, the PM is constitutively expressed, forming a protective sleeve along the entire length of the midgut. Type II PM, such as that occurring in tsetse flies, physically separates the midgut lumen into two compartments. The endoperitrophic space is where the food lies and the ectoperitrophic space is the region between the PM and the midgut epithelium (see inset, Fig. 3). Clearly, by separating the food from the midgut epithelium, the PM must act as a molecular sieve through which digestive enzymes and nutrients destined to be absorbed must pass. The molecular sieving properties of the PM vary with changes in the surrounding ionic environment (Miller and Lehane, 1993). Typically, the tsetse PM has a pore size of 9 nm, making the PM permeable only to globular molecules less than 150 kDa (Miller and Lehane, 1990). Consequently, the tsetse PM presents a barrier to procyclic trypanosomes that need to get into the ectoperitrophic space to continue their development.

THE ENEMY WITHIN

137

Whether trypanosomes penetrate the PM, or circumnavigate around its broken ends in the hindgut to get to the ectoperitrophic space, is still the subject of speculation. There is little dispute that trypanosomes returning to the endoperitrophic space, from 6 to 8 days post-infection, actively penetrate the PM. Given the data obtained mainly by transmission electron microscopy (Ellis and Evans, 1977; Gibson and Bailey, 2003), it seems probable that trypanosomes also actively penetrate the PM when entering the ectoperitrophic space 3–5 days post-infection. Penetration is likely to be preceded by a specific attachment event, so it is interesting to note that GFP-expressing T. b. brucei PCFs tend to line up parallel with the PM and become associated with tsetse gut fragments upon dissection (Gibson and Bailey, 2003). How trypanosomes achieve PM penetration is unknown, and no chitinase gene has been identified from the trypanosome genome. However, chitin can be a relatively minor component of dipteran PM (Tellam et al., 1999) and other enzymes capable of dealing with glycosaminoglycans may be more important in the penetration event. In support of penetration rather than circumnavigation via the hindgut, only small numbers of procyclics were observed in the posterior region of the midgut when GFP-expressing parasites were used. These parasites were considered to be an indication of a failed infection cycle in its terminal stages (Gibson and Bailey, 2003).

2.5.2

Immune signalling

Genetic studies have demonstrated that Drosophila uses two main immune signalling pathways to control immune responses to pathogen challenge. The Toll and Imd (immunodeficiency) pathways respond to different classes of microbes and are used differently in various fly tissues (De Gregorio et al., 2002). The Imd pathway is predominantly involved in regulating epithelial immune responses and is most strongly utilized following Gram-negative bacterial challenge. The Toll pathway is most heavily involved in systemic immune responses and is activated most by fungal and Gram-positive bacterial infections (Fig. 5). These pathways are highly conserved among insect species and bear similarity to the vertebrate Toll-like receptor (TLR) and tumour necrosis factor (TNF) pathways. The Drosophila immune system is also highly regulated as over- or underexpression of immune responses is detrimental to host health (Lemaitre and Hoffmann, 2007; Ryu et al., 2008; Aggarwal and Silverman, 2008). The immune response may be divided into three key stages: recognition of pathogens, activation of the signal transduction pathway(s) and production of immune effector and regulator molecules. Orthologous intracellular and extracellular members of both pathways have been identified from Glossina EST databases (Table 1), although very little experimentation to gather functional evidence for their role in tsetse flies has been undertaken.

138


Toll

Imd

Yeast

Fungi

Gram-positive bacteria

Gram-negative bacteria

GNBP3 Lysine-type peptidoglycan

DAP-type peptidoglycan PGRP-LE

Haemolymph

Necrotic GNBP1

PGRP-SD

Monomeric peptidoglycan

SPE

Persephone

Polymeric peptidoglycan

PGRP-SA Pro-Spätzle PGRP-LCx PGRP-LCa PGRP-LCx PGRP-LCx

Spätzle

PGRP-LF Imd

Toll

MYD88

dIAP2 Tube

Pelle

Caspar

dFADD

Cytoplasm

dTAB2 dTAK1

Dredd Dnr1

Cactus WntD

Dorsal

DIF

JNK pathway

Nucleus

DaPKC Drosomycin + other HDPs

Relish Kenny Ird5 (IKKγ) (IKKβ) Caudal Diptericin + other HDPs

FIG. 5 Schematic representation of the Toll and Imd signalling pathways. The function of the Toll and Imd signalling pathways has been intensively investigated in Drosophila melanogaster and has been reviewed in detail by Lemaitre and Hoffmann (2007) and Aggarwal and Silverman (2008). This figure is modified from these references. The omniBLAST search tool was used to identify putative G. m. morsitans orthologs (at 1e 20 or less) of all identified Drosophila Toll and Imd pathway genes in all library clustered EST translated sequences of G. m. morsitans (http://www.genedb.org). Orthologs of all Toll and Imd pathway members depicted were identified (except WntD) and are summarized in Table 1.

2.5.3

Recognition and activation

Differential regulation of immune effector HDP indicates that the tsetse immune system can distinguish between virulent and avirulent strains of bacteria (Weiss et al., 2008), between trypanosome and bacterial infections and also between different trypanosome life stages, that is procyclic and bloodstream forms (Hao et al., 2001). In Drosophila, pathogen recognition is achieved by pattern recognition receptors (PRRs), which each recognize a certain class of pathogen-associated molecular pattern (PAMP). Two families of PRRs, the peptidoglycan recognition proteins (PGRPs) and Gram-negative binding proteins (GNBPs), are known. Interestingly, PRRs are proteins originally derived from enzymes known to degrade microbial cell wall components, thus evolving

TABLE 1 Identification of Toll and Imd immune pathway orthologs in G. m. morsitans Toll Drosophila Molecule

Imd a

b

Ortholog

Drosophila Molecule

GNBP1 (Gram-negative FBgn0040323 binding protein 1) GNBP3 (Gram-negative FBgn0040321 binding protein 3) Persephone FBgn0030926

cn16500

Necrotic

FBgn0002930

cn8210

SPE (Spa¨tzle processing FBgn0039102 enzyme) Spa¨tzle FBgn0003495 Toll FBgn0003717

cn8446

Pelle

FBgn0010441

cn1855

Tube

FBgn0003882

Gmsg-10360

MyD88

FBgn0033402

cn12871

PGRP-LC (peptidoglycan recognition protein LC) PGRP-LE (peptidoglycan recognition protein LE) PGRP-LF (peptidoglycan recognition protein LF) PGRP-SA (peptidoglycan recognition protein SA) PGRP-SD (peptidoglycan recognition protein SD) Imd (immune deficiency) dTAK1 (TGF-beta activated kinase 1) Kenny (IKKg) (IkappaB kinase) Dredd (Death released ced-3/NeddZ-like protein) Dnr1 (defence repressor 1)

FlyBase ID

cn16500 cn353

Gmsg-10321 cn14933

FlyBase IDa

Orthologb

FBgn0035976

cn1360

FBgn0030695

Gmm-3156

FBgn0035977

cn1360

FBgn0030310

Tse122g03

FBgn0035806

cn1360

FBgn0013983 FBgn0026323

cn3932 Gmsg-7119

FBgn0041205

Gmsg-7350

FBgn0020381

cn13356

FBgn0260866

Tse104g08.qlc (continues)

TABLE 1

(Continued)

Toll Drosophila Molecule

FlyBase IDa

Imd Orthologb

Cactus

FBgn0000250

Gmsg-2191

Dorsal

FBgn0260632

cn12921

Dif (dorsal-related immunity factor) DaPKC (Drosophila atypical protein kinase C) WntD (Wnt inhibitor of Dorsal)

FBgn0011274

Gmsg-2191

FBgn0022131

cn10314

FBgn0038134

Not identified

a b

Drosophila Molecule DIAP2 (inhibitor of apoptosis-2) Ird5 (IKKb) (Immune response deficient 5) dFADD

FlyBase IDa

Orthologb

FBgn0015247

cn3040

FBgn0024222

cn5893

FBgn0038928

cn6552

dTAB2 (TAK1-associated binding protein 2)

FBgn0086358

Gmsg-6058

Caspar

FBgn0034068

GLAEJ30TV

Relish Caudal

FBgn0014018 FBgn0000251

cn7095 cn9066

Drosophila melanogaster accession numbers from genome data assembled by Flybase http://flybase.org. Glossina orthologs obtained from G. m. morsitans EST libraries clustered by GeneDB http://www.genedb.org/gendb/glossina.

THE ENEMY WITHIN

141

from a microbicidal to a recognition role (Lemaitre and Hoffmann, 2007). A role for PGRP-LB, a known negative regulator of the Imd pathway in Drosophila, has been noted in downregulating tsetse immune responses to virulent but not avirulent bacteria (Weiss et al., 2008). Notably, no PRR has yet been identified for African trypanosomes, although it is clear that there is a tsetse immune response to their presence (Hao et al., 2001). Binding of PRRs to a pathogen leads to activation of Toll/Imd downstream signalling cascades. Signalling culminates in the activation and nuclear translocation of NFkB transcription factors Dorsal/Dif (Toll pathway) or Relish (Imd pathway). This in turn causes transcription of immune effector genes such as HDPs. Importantly, molecular crosstalk may occur between the Toll and Imd pathways, thus giving rise to a wide spectrum of possible immune responses to pathogen invasion (Tanji and Ip, 2005; Tanji et al., 2007). In Drosophila, control of induced epithelial immune responses is associated with the Imd signalling pathway (Tzou et al., 2000). Gene knockdown of the Imd pathway transcription factor Relish in tsetse flies has been demonstrated to cause an increase in midgut and salivary gland infection rates with trypanosomes (Hu and Aksoy, 2006). This evidence implicates the Imd pathway in regulating fly susceptibility to trypanosome infection. Furthermore, knockdown of Relish leads to downregulation of attacin, a HDP associated with tsetse– trypanosome interactions (Hao et al., 2001; Hu and Aksoy, 2005, 2006; Nayduch and Aksoy, 2007).

2.5.4

Effector molecules

HDPs are evolutionarily conserved effector molecules of the humoral defence system and are found among all classes of life. They exhibit a broad spectrum of activity against bacteria, fungi, viruses and transformed cells (Lemaitre and Hoffmann, 2007). In addition, the anti-parasite activity of HDPs has been illustrated in several vector–parasite systems (Durvasula et al., 1997; Shahabuddin et al., 1998; Boulanger et al., 2002b) including a tsetse– trypanosome system (Hu and Aksoy, 2005; Hu and Aksoy, 2006). Many HDPs target pathogens by disturbing the pathogen membrane potential or by disrupting internal cell functioning leading to cell death by apoptosis or necrosis. Early research into these immune mediators in G. m. morsitans identified four HDPs: an attacin (AttA1), a cecropin, a defensin and a diptericin (Hao et al., 2001; Boulanger et al., 2002a). More recently, characterization of the G. m. morsitans attacin loci has recognized that attacin genes are organized in three clusters encoding three different attacins: attA, attB and attD. The amino acid sequences of AttA and AttB are almost identical while AttD is only 69% identical to the AttA/B form. These genes are differentially regulated (Wang et al., 2008). Interestingly, two additional HDPs, one with anti-Gram-negative activity and the other with anti-Gram-positive activity, have also identified

142


following trypanosome challenge (Boulanger et al., 2002a). These HDPs remain uncharacterized. In vivo analysis of HDP transcript expression during trypanosome infection has indicated that these peptides are differentially regulated in the haemolymph and in the major immunoresponsive tissues of midgut, fat body and proventiculus (Hao et al., 2001, 2003; Hu and Aksoy, 2006). Early in the infection process, the presence of trypanosomes in the midgut or haemolymph does not lead to activation and increased transcription of midgut or fat body HDP genes (Hao et al., 2001). However, by day 6, as parasite numbers increase, attacin (AttA/B and AttD) and defensin transcript expression is high in the fat body (Hao et al., 2001, 2003; Wang et al., 2008). In selfcured flies HDP transcript expression levels fall, but in flies with established midgut infections expression levels remain high in the fat body and proventriculus (Hao et al., 2001, 2003). This does not appear to affect the viability of the parasite population within the midgut, although it remains to be seen if individual parasites are affected, thus resulting in a change in the nature of the parasite population. It is possible that trypanosomes exhibit a stage-specific sensitivity to particular immune molecules, with procyclics exhibiting higher resistance to the trypanocidal activity of HDPs than BSF trypanosomes as was observed in vitro by Haines et al. (2003). Alternatively fat body synthesized peptides circulating in the fly haemolymph may fail to reach parasites located in the midgut environment. The trypanocidal activity of HDPs has been recognized. Stomoxyn, isolated from the facultative hematophagous fly Stomoxys calcitrans, exhibits trypanocidal activity (Boulanger et al., 2002b). In addition, there is direct evidence of trypanosome killing by Glossina HDPs themselves (Hu and Aksoy, 2005; Hu and Aksoy, 2006; Nayduch and Aksoy, 2007). Recombinant attacin (AttA1) inhibited both BSF and PCF growth in vitro (Hu and Aksoy, 2005). In vivo, gene knockdown of the AttA1 peptide, or its transcriptional regulator Relish, led to a statistically significant increase in midgut and mature salivary gland trypanosome infection rates (Hu and Aksoy, 2006). Relish also regulates cecropin expression in Glossina, but whether cecropin possesses trypanocidal properties is yet to be directly investigated. More recently, differential expression of AttA1 transcripts in a trypanosome susceptible species (G. m. morsitans) and two comparatively trypanosome-refractory species (G. pallidipes and G. p. palpalis) has been reported (Nayduch and Aksoy, 2007). Refractory species showed higher attacin transcript expression in fat body (systemic) and proventiculus/ midgut (local) tissues in comparison to susceptible flies in both teneral and blood-fed states. Knockdown of attacin expression in G. pallidipes led to increased trypanosome susceptibility, although the possible confounding effects of high mortality rate, starvation, wounding and low sample number in this study should be noted (Nayduch and Aksoy, 2007). Nevertheless, the evidence suggests that attacin may be an important regulator of tsetse–trypanosome interactions.

THE ENEMY WITHIN

143

Notably, several investigations of HDP function in tsetse–trypanosome interactions bypass a natural trypanosome developmental stage by using PCF rather than BSF trypanosomes to infect the flies (Hao et al., 2001, 2003). It is too early to say how this experimental approach to infection affects the tsetse– trypanosome interaction but it clearly runs the risk of bypassing part of the natural fly immune response. In some cases, the authors do not state the trypanosome life cycle stage used (Nayduch and Aksoy, 2007). Furthermore, supplementation of bloodmeals with glucosamine (Hao et al., 2001, 2003) may disrupt the natural infection process and confound the interpretation of results (Peacock et al., 2006). The HDP diptericin is constitutively expressed in the proventriculus and fat body of tsetse (Hao et al., 2001). This HDP expression profile has been attributed to the presence of symbionts, in particular to the presence of Sodalis in the gut and haemolymph. Interestingly, Sodalis is resistant to the Gramnegative bactericidal activity of mature synthetic diptericin in vitro (Hao et al., 2001), whereas the non-native bacterium E. coli is not. Sodalis is also more resistant to recombinant attacin (recGmAttA1) than E. coli (Hu and Aksoy, 2005) and to a battery of other Gram-negative and Gram-positive bacteriolytic HDPs (Haines et al., 2003). Thus, it is possible Sodalis may have evolved HDP resistance traits permitting survival in the hostile tsetse midgut environment while invading pathogens are eliminated. 2.5.5

Antioxidants

In addition to the NFkB pathway-mediated defense systems, the NFkB-independent production of microbicidal ROS is a key component of insect epithelial immune responses (Ha et al., 2005a; Lemaitre and Hoffmann, 2007). In Drosophila, natural gut infections with bacteria are associated with rapid synthesis of ROS. Studies have demonstrated that it is Drosophila dual oxidase (dDuox), and not NADPH oxidase (dNox), that provides the main source of ROS that limits bacterial proliferation in the midgut (Ha et al., 2005a,b). The infection-inducible nature of intestinal dDuox suggests the transcriptional role of dDuox may play a pivotal role in fly midgut protection against invading pathogens. dDuox is capable of generating H2O2, and the highly microbicidal HOCl is derived from H2O2 by neutrophil-derived myeloperoxidase (MPO) type activity. Physiological regulation of ROS levels in the midgut is achieved via dDuox-dependant ROS generation and immune-regulated catalase (IRC)dependant ROS removal (Ha et al., 2005a,b). A fine redox balance is critical, as knockdown of either dDuox or IRC leads to higher mortality rates due to either insufficient or prolonged oxidative stress responses respectively. It is known that ROS activates a cell death pathway in procyclic T. b. brucei trypanosomes (Ridgley et al., 1999). In tsetse, increased ROS and nitric oxide (NO) transcript expression have been reported in the proventriculus in response to trypanosome challenge (Hao et al., 2003). Additionally, upregulation of

144


oxidative stress genes has been demonstrated in the midgut transcriptome of infected flies (Lehane et al., 2003; Munks et al., 2005) and self-cured flies (Lehane et al., 2008). Hao et al. have hypothesized that reactive intermediates such as NO or H2O2 may function as chemical signals to mediate communication between different immunoresponsive tissues (Hao et al., 2001, 2003). Indirect evidence suggests a role for antioxidants in signalling during parasite development in the fly and protecting tsetse flies against trypanosomes (Hao et al., 2001, 2003; Macleod et al., 2007a,b). Recently, Macleod et al. (2007b) demonstrated the feeding of different antioxidants (glutathione, cysteine, N-acetyl-cysteine, ascorbic acid or uric acid) in the infective bloodmeal led to significant increases in midgut infection rates. Also, glucosamine, which has routinely been used to boost trypanosome midgut infection rates, has been shown to be an antioxidant molecule (Xin et al., 2006). The original interpretation of increased infection rates in flies fed glucosamine was that this sugar neutralized specific trypanocidal lectins in the midgut, permitting higher infection rates to occur (Maudlin and Welburn, 1987; Welburn et al., 1989; Murphy and Welburn, 1997) (see Section 2.5.6). However, with this more recent demonstration that N-acetyl glucosamine can scavenge ROS, superoxide and hydroxyl ions (Xin et al., 2006), an effect on trypanosome infection rates through reduction of ROS challenge is a strong alternative hypothesis. Oxidative stress is believed to be involved in control of parasites in other vector–parasite systems, including the Plasmodium–mosquito and Trypanosoma–Rhodnius systems. Augmented production of NO has been noted in Anopheles stephensi, An. gambiae and An. pseudopunctinpennis following P. berghei infection (Luckhart et al., 1998; Herrera-Ortiz et al., 2004). The R. prolixus NO system responds to T. rangeli and T. cruzi, implicating involvement in trypanosome infection regulation (Whitten et al., 2001, 2007). Furthermore, the stable suppression of trypanosome parasitaemia in Cape buffalo is associated with increased levels of serum ROS (Wang et al., 2002). A role for transferrin in immune signalling and the upregulation of NO in vertebrates has been suggested (Stafford and Belosevic, 2003). Recently, knockdown of transferrin in G. m. morsitans was demonstrated to have a statistically significant impact on trypanosome prevalence, resulting in almost a doubling of trypanosome midgut infection rate (Lehane et al., 2008). The mechanism of involvement of tsetse transferrin in mediating tsetse– trypanosome interactions is still unknown, but there is the intriguing possibility that it may be analogous to the function in vertebrates. 2.5.6

Lectins and programmed cell death

Susceptibility, defined as midgut establishment of T. congolense or T. b. brucei in G. m. morsitans infected at the first bloodmeal, can be selected for with results apparent in the F1 generation (Maudlin, 1982). This susceptibility is an extra-chromosomal trait inherited through the female line (Maudlin and

THE ENEMY WITHIN

145

Dukes, 1985). G. m. centralis infected with T. congolense show similar susceptibility phenotypes. It appears the effect operates in both the midgut and foregut, because selected lines showed similar phenotypes to T. vivax infection (Moloo and Kutuza, 1988a). The lectin hypothesis was proposed as the mechanism underpinning this phenotype by Maudlin and Welburn (1987). They argued that midgut-expressed lectins were the predominant factor killing trypanosomes in vivo in the tsetse fly, with lectin levels modulated by the changing number of the symbiont Sodalis glossinidius in the fly midgut (Welburn and Maudlin, 1999). This argument was based primarily on evidence that feeding of sugars capable of inhibiting lectins increased trypanosome midgut infection rates (Maudlin and Welburn, 1987; Ingram and Molyneux, 1988; Welburn et al., 1994). Additionally, based on the effects of the plant lectin concanavalin A (Con A) on trypanosomes in vitro, lectin-mediated killing of trypanosomes was believed to occur by a process termed proto-apoptosis (Welburn and Maudlin, 1999; Pearson et al., 2000) (see below). A lectin specific for D-glucosamine and with lesser affinity for N-acetyl-D-glucosamine has been tentatively identified in tsetse midgut (Ibrahim et al., 1984; Ingram and Molyneux, 1988). Most of this lectin is found attached to the PM (Lehane and Msangi, 1991). Inhibition of midgut lectins was hypothesized to occur naturally to the greatest extent in newly emerged flies, where lectins were neutralized by the build-up of inhibitory sugars released from chitin by the endochitinase of the symbiont Sodalis during the pupal period (Welburn and Maudlin, 1999). The increased potential of older starved flies to be infected when compared with age-matched non-starved flies (Kubi et al., 2006) is in disagreement with the lectin hypothesis, as both are reported to have similar midgut lectin levels (Lehane and Msangi, 1991). More recently, the multiple effects of N-acetyl-Dglucosamine or D-glucosamine inclusion in the bloodmeal upon trypanosome growth and tsetse physiology have been reported (Peacock et al., 2006; Ebikeme et al., 2008). Both sugars slowed bloodmeal movement along the midgut. Glucosamine significantly increased the size of bloodmeal taken, as well as increasing the fly mortality rate. Interestingly, N-acetyl-D-glucosamine stimulated trypanosome growth in the midgut and in vitro in the absence of any fly-derived factors. The ability of N-acetyl-D-glucosamine to enhance trypanosome PCF growth in vitro does not seem to involve any direct effect of N-acetyl-D-glucosamine upon metabolic processes within the trypanosome (Ebikeme et al., 2008). In vitro experiments have instead shown that N-acetylD-glucosamine indirectly enhances the rate of L-proline metabolism by inhibiting the trypanosome hexose transporter, thus depriving the parasite of D-glucose (Ebikeme et al., 2008). However, PCFs cultured in D-glucose free medium, and thus possessing a metabolism primed for L-proline use, do not appear to have a higher infection rate when used to infect tsetse flies. This suggests that the metabolic impact of N-acetyl-D-glucosamine on trypanosomes has no bearing on the success of trypanosome establishment within the tsetse midgut (Ebikeme et al., 2008). These off-target effects of feeding glucosamine complicate the

146


interpretation of experiments on the trypanosome prevalence phenotype observed (Peacock et al., 2006). Above all, the demonstration that glucosamine (which has routinely been included in infected bloodmeals to artificially boost trypanosome midgut infection rates) is an antioxidant molecule (Xin et al., 2006) suggests a simpler explanation (see Section 2.5.5) for the effect of this sugar on tsetse–trypanosome interactions. In addition, the lectin hypothesis contends that higher symbiont densities lead to increased susceptibility. However, a correlation between symbiont density and susceptibility does not occur in all instances (Shaw and Moloo, 1991; Weiss et al., 2006). Successfully established midgut infections show an average parasite density of approximately 5 105 cells per midgut (Van den Abbeele et al., 1999; Gibson and Bailey, 2003), but this number can vary with clone and species of trypanosome. Procyclic trypanosomes share a common energy source (proline) with their tsetse hosts (Welburn and Maudlin, 1999), which may become a limiting factor if trypanosome numbers expand too far. It is possible that trypanosomes self-regulate their numbers and that quorum sensing may be involved in the process (Acosta-Serrano et al., 2001; Roditi and Lehane, 2008). More specifically, it has been suggested that the cleaved N-terminal fragments of trypanosome procyclins might be involved in a density-sensing mechanism, which the trypanosomes use to control their population densities (Acosta-Serrano et al., 2001). However, this mechanism may not be essential as procyclin knockout mutants are still fly-transmissible (Vassella et al., 2009). A novel form of cell death is believed to occur in trypanosomes, known as proto-apoptosis, and it is suggested that this is part of the mechanism by which trypanosomes regulate numbers within the midgut (Welburn et al., 1996; Murphy and Welburn, 1997; Welburn and Maudlin, 1997). In support of this idea, incubation of PCFs of T. b. brucei, T. b. rhodesiense or T. congolense trypanosomes (but not their corresponding BSFs) with the plant lectin Con A in vitro induces growth arrest and death, with many of the characteristics of programmed cell death (PCD) (Welburn and Maudlin, 1997; Murphy and Welburn, 1997). In tsetse, with some trypanosome infections, cyst-like forms have been observed embedded in the PM (Robertson, 1913; Evans and Ellis, 1983; Gibson and Bailey, 2003). Some cysts contain highly motile forms while others appear to consist of aggregates of rounded up trypanosomes. Whether these cysts represent an unknown life cycle stage or degenerating trypanosome forms is unknown. Gibson and Bailey (2003) noted that these forms resembled the morphologically altered PCF trypanosomes treated with Con A in vitro to induce cell death. Proteins upregulated during the process of PCD include the trypanosome protein kinase C receptor, which is similarly upregulated in the short stumpy BSF. This suggests that stumpy form BSFs are removed from the trypanosome population within the mammalian hosts (should ingestion by a tsetse host not occur). This would avoid triggering a wider immune response within the

THE ENEMY WITHIN

147

mammalian host, which could jeopardize the survival of the long slender BSF population (Murphy and Welburn, 1997). 2.5.7

Other potential effector molecules

Melanization plays an important role in insect defence reactions such as wound healing, sequestration of microorganisms, encapsulation and the production of intermediates toxic to invading pathogens (Lemaitre and Hoffmann, 2007). Prophenoloxidases (ProPOs) are oxidoreductases related to hemocyanins that mediate melanization. These molecules have been implicated in tsetse– trypanosome interactions as inhibition of phenoloxidase (PO) activity with phenylthiourea (PTU) increases midgut trypanosome infection rates (Nigam et al., 1997). Furthermore, T. b. rhodesiense has been shown to significantly activate haemolymph proPO of female G. m. morsitans (Nigam et al., 1997). Such a complex biochemical cascade as occurs in the ProPO system is almost certainly highly reliant on the controlled environment of the haemolymph for its successful operation, and it is hard to see how it could operate effectively in the midgut lumen, with the possible exception of the immediate surface of the epithelium. However, it is possible that proPO may be a means of killing those parasites which do enter the haemocoel of the fly. TsetseEP protein may also play a role in tsetse–trypanosome interactions. This immunoresponsive protein is expressed strongly in the midgut of Glossina (Chandra et al., 2004) and is upregulated in response to immune stimulation with Gram-negative bacteria (Haines et al., 2005). This protein exhibits a high sequence identity to the carboxy terminal region of the EP form of procyclins, which is found on the surface of procyclic T. b. brucei. The glutamic acid– proline (EP) repeats found on both molecules are extremely unusual and it seems unlikely that they coincide in the tsetse midgut solely by chance. The potential role of tsetseEP in mediating fly responses to trypanosome infection merits further investigation. 2.5.8

Adaptive immunity

The existence of immunological memory has long been known in vertebrates, with humans utilizing vaccination before key mechanisms of adaptive immunity were known or understood. Vertebrate adaptive immunity is characterized by long-term protection against specific antigens achieved in part by antibodies that consist of a large diversity of somatically rearranged immunoglobulin receptors. The possibility that invertebrates are capable of adaptive immunity to pathogen challenge has recently re-emerged as a possibility (Watson et al., 2005; Dong et al., 2006; Dong and Dimopoulos, 2009). For example, Dong et al. (2006) identified Down syndrome cell adhesion molecule (Dscam) as an alternatively spliced, hypervariable immunoglobulin domain containing gene responsible for generating a broad range of pathogen recognition receptors

148


(PRRs) in Anopheles gambiae mosquitoes. The gene contains 101 exons including three variable Ig exon cassettes. Alternative splicing of this gene produces a highly diverse set of over 31,000 potential splice forms. Challenge with different pathogens leads to production of specific AgDscam splice-form repertoires. Interestingly, an ortholog of Dscam has been identified from a Glossina EST library (cn9121, e-value 6.0 10 21). Whether Glossina uses this gene in immune responses to trypanosomes or bacteria remains to be elucidated. 2.5.9

Fly immune system and microbial balance

Glossina carries a range of microorganisms (see Section 3) and hence, must manage to coexist with these symbionts while still protecting themselves from the onslaught of pathogens. The host factors that maintain the homeostatic relationship between the tsetse host and its symbionts are largely unknown, and our current insight is based primarily on studies in Drosophila. Recently, the intestinal homeobox gene Caudal (Cad) has been identified as a key factor in repressing NFkB-dependant HDP genes in Drosophila (Ryu et al., 2008). Thus, although Drosophila commensal organisms can induce a high level of local Imd-regulated NFkB activation, only a subset of target genes is activated. Regulation of the immune system is tightly controlled, as overexpression of HDP genes in Caudal knockdown flies caused a shift in the commensal population in the midgut. In particular, the dominance of Gluconobacter sp. strain EW707 eventually led to gut apoptosis and host mortality in Drosophila. The model proposed by Ryu et al. (2008) involves interplay between Caudal and the NFkB transcription factor Relish to regulate the expression of HDPs, which in turn defines the microbial community and insect health. Whether this is also the case in Glossina has not yet been investigated, although an ortholog of Caudal has been identified in the EST libraries (cn9066, e-value 1.2 10 47) making this a possibility. 2.6

EFFECTS OF TRYPANOSOME INFECTION ON TSETSE PHYSIOLOGY

Trypanosome infection may have multiple effects on the tsetse host. A suppression subtractive hybridization study (Lehane et al., 2008) identified molecules differentially expressed in established versus self-cured T. b. brucei infections in G. m. morsitans. Analysis of the gene fragments generated suggested that trypanosome infection has a marked effect upon the metabolism of the fly host. Flies self-cured of trypanosome infection displaying the signals of increased energy usage and an oxidative stress response compared to infected flies. Within the tsetse fly, procyclic trypanosomes employ the amino acid proline as a source of energy, while the fly typically uses proline reserves for flight. In infected male flies, trypanosomes infection can reduce the flight potential by 15% (Bursell, 1981). Trypanosome infection in female flies would be far more

THE ENEMY WITHIN

149

costly in terms of flight potential as females have smaller proline reserves due to the energy required for larval development. Recently, Hu et al. (2008) investigated the cost of trypanosome infection on fly reproductive fitness. Two trypanosome strains were selected that differentially activated the host immune system: a T. b. rhodesiense wild-type strain that induced expression of attacin and defensin in trypanosome infected flies and a mutant strain that did not. Only infection with the wild-type strain led to a significant increase in larval deposition periods and a decrease in milk gland protein expression. This suggests that activation of tsetse immune responses by infection with immunogenic trypanosomes delays larvigenesis via decreased expression of the milk gland protein vital for larval growth. Trypanosome infection may also impact the feeding behaviour of infected tsetse hosts. Infection of G. m. morsitans and G. austeni with T. b. brucei reportedly results in increased probing behaviour and more voracious feeding (Jenni et al., 1980). Accumulation of large numbers of trypanosomes in the salivary glands or proboscis has been observed, and may impede function of labral mechanoreceptors as detectors of blood flow rate in the gut (Molyneux and Jenni, 1981). Thus an infected fly may take several smaller meals or feed for longer, either scenario potentially leading to increased trypanosome transmission to the vertebrate host.

2.7

FLY SEX, AGE AND STARVATION AND TRYPANOSOME TRANSMISSION

Susceptibility to parasite infection is also influenced by the sex of the fly. This is evident in midgut infections with T. congolense and T. brucei, but not in infections with T. vivax. Most laboratory studies (Burtt, 1946b; Harley, 1971; Distelmans et al., 1982; Mwangelwa et al., 1987; Hide et al., 1991; Moloo, 1993; Welburn et al., 1995; Dale et al., 1995), but not all (Welburn and Maudlin, 1992; Moloo et al., 1992; Moloo, 1993), suggest that male flies are more susceptible. Female G. p. palpalis are more resistant to developing mature T. congolense infections (Distelmans et al., 1982) than males. Unexpectedly, this sex-dependent phenomenon was not observed with T. congolense infections in G. m. morsitans (Dale et al., 1995). Of interest, compared to a virgin G. m. morsitans, a mated female is twice as refractory to T. b. brucei infection (Macleod et al., 2007a). Although both sexes are capable of transmitting parasites, both male and female flies are innately resistant to parasite infection and this resistance increases rapidly with age (Fig. 4). This natural refractoriness to trypanosome maturation and/or establishment is seen in both laboratory and field populations (Distelmans et al., 1982; Leak 1999) and is termed the teneral phenomenon (Welburn and Maudlin, 1992). In the tsetse community the term teneral is used to describe a newly emerged fly that has not yet had its first bloodmeal. The fly during this time is marked by

150


a soft, often described as soapy-feeling, exoskeleton and lighter body coloration. While tsetse researchers universally use the word ‘‘teneral’’, they seem to have forgotten that the term represents a highly variable physiological state, particularly in terms of fly susceptibility. Depending on tsetse species and the nutritional status of the newly eclosed adult fly, teneral flies can survive for up to 1 week before dying of starvation (Kubi et al., 2006; Lehane lab, unpublished observations). Van Hoof et al. (1937) originally demonstrated the ‘‘teneral phenomenon’’, that is the higher susceptibility of a newly emerged fly to trypanosome infection compared to an older fed fly, using G. f. fuscipes challenged with T. b. gambiense. The highest midgut establishment rates were obtained in flies fed the infective meal 0–1 days post-emergence (p.e.). This work was later confirmed by Wijers (1958), who observed that G. p. palpalis emergents less than 30 h p.e. were more susceptible to a first infective bloodmeal infection than flies 30–78 h p.e. In recent years, the teneral definition has shifted to a nutritional focus and teneral has become synonymous with ‘‘unfed’’. Many published papers ignore the fact that parasite susceptibility during this teneral period can vary widely, and consequently flaws in experimental design may confound data interpretation. Speculation on why newly emerged flies are so susceptible to trypanosomes is profuse. Incomplete formation of the PM, variable antioxidant or lectin concentrations, molecular maturity of the midgut (i.e. proteases), immaturity of the immune response, larval meal turnover (preand post-meconium expulsion), pH gradient differences and variation in symbiont loads may, either singly or as a combination, play a role in creating the teneral phenomenon. Although further research is required to confirm which of these factors (if any) contribute to this age-related susceptibility to infection, it is crucial to standardize trypanosome infection experiments by redefining the definition for teneral as ‘‘hours post-eclosion’’ instead of simply ‘‘unfed’’. Starvation is a constant risk the tsetse fly faces throughout its lifetime. Flies deprived of a bloodmeal utilize energy reserves stored in the fat body and risk death when fat reserves drop to approximately 6% of the total dry body-mass (Rogers et al., 1994). Male flies, which use most of each bloodmeal for lipid production, are more resistant to starvation than females, which use the bloodmeal for both lipid production and growth of the larva (Randolph and Rogers, 1981). Starvation may influence feeding behaviour with the range of host choice expanding as the fly’s hunger intensifies (Bouyer et al., 2007) with obvious consequences for the epidemiology of disease. Environmental conditions, such as temperature and relative humidity, also influence fly starvation with many Glossina species becoming hungrier during the dry season (Jackson, 1933). Starvation is a key factor in tsetse–trypanosome interactions because the nutritional status of the fly at the time of the infective bloodmeal is a key determinant of fly susceptibility to trypanosome infection (Gingrich et al., 1982; Mwangelwa et al., 1987; Gooding, 1988; Kubi et al., 2006).

THE ENEMY WITHIN

151

Under laboratory conditions with bloodmeals given every 4 days, newly emerged male tsetse flies are highly susceptible to trypanosome infection at the first bloodmeal while older flies exhibit a significantly lower susceptibility to trypanosome infection (Fig. 4) (Distelmans et al., 1982; Welburn and Maudlin, 1992; Kubi et al., 2006). However, field studies of trypanosome prevalence with fly age suggest a significantly higher proportion of the older adult tsetse population than predicted develop a mature trypanosome infection (Woolhouse et al., 1993; Msangi et al., 1998; Lehane et al., 2000). Recently, laboratory-based studies by Kubi et al. (2006) demonstrated a period of starvation (3–4 days for teneral or 7 days for adult flies) increased fly susceptibility to T. b. brucei or T. congolense infection, resulting in a significantly higher number of T. congolense or T. b. brucei mature salivary gland infections. In the case of T. congolense, a significantly higher number of midgut infections were also observed. It is likely that the increased infection rates noted in older field flies compared to laboratory flies is explained by starvation events in the field. An explanation for this observation may be that under high nutritional stress the physiological barriers to trypanosome infection are suppressed because the energy cost of an immune response is high (Schmid-Hempel, 2005; Ye et al., 2009). Trypanosome infection may have a negative impact on fly longevity (Bursell, 1981). More immunogenic lines of trypanosomes adversely affect tsetse reproductive fitness (Geiger et al., 2008), which suggests that trypanosome infection places a selection pressure on tsetse flies.

2.8

ENVIRONMENTAL TEMPERATURE AND TRYPANOSOME TRANSMISSION

As puparial incubation temperature rises (within biological limits), so does the ability of the corresponding adult fly to develop a mature infection (Burtt, 1946a; Ndegwa et al., 1992). This is not attributed to an increase in the number of flies that establish a midgut infection, but rather to an increase in the proportion of these flies that go on to develop mature infections (Dipeolu and Adam, 1974). The mechanism responsible is unknown. However, this observed trend may have consequences in the field for the distribution of trypanosomiasis. There is a positive correlation between latitude relative to 7 S (the median of tsetse distribution) and infection rate (Ford and Leggate, 1961). Although the factors involved in the field are undoubtedly complex (Jordan, 1965), ambient temperature also increases towards the 7 S median, implying that it may be puparial incubation temperature that influences this trend. If that is the case, then temperature plays a central role in determining the geographical distribution and epidemiology of trypanosomiasis.

152

3 3.1


Symbiont–tsetse–trypanosome interactions WIGGLESWORTHIA GLOSSINIDIUS

All Glossina sp. harbour W. glossinidius, a primary obligate (beneficial) Gramnegative endosymbiont. Based on concordant evolutionary analyses, it is estimated the Wigglesworthia-Glossina symbiosis evolved in ancestral forms between 50 and 80 million years ago (Chen et al., 1999). This bacterium resides within the cytoplasm of specialized epithelial midgut cells, bacteriocytes, which form a horseshoe shaped organ, the bacteriome (Fig. 2E), located in the anterior midgut (Aksoy et al., 1995; Aksoy, 1995). The Wigglesworthia genome is grossly reduced in size to less than 700 kb and is predicted to encode for 617 proteins (Akman et al., 2002). This streamlined genome has undergone massive gene erosion as genes required for defense mechanisms, metabolic pathways, DNA repair mechanisms and so on, are no longer essential as this endosymbiont exists in a stable host environment. In fact, even a gene involved in chromosome replication, dnaA, which is considered essential for bacterial survival, has been lost in Wigglesworthia (Akman et al., 2002). This member of the Enterobacteriacae family, first observed over a century ago (Stuhlmann, 1907), was thought to produce metabolites to compensate for nutritional deficits in the host’s haematophagous diet and appears to be partially associated with the metabolism of B-complex vitamins essential for tsetse survival (Wigglesworth, 1929; Nogge, 1981; Akman et al., 2002). The elimination of Wigglesworthia reduces fly longevity, rate of bloodmeal digestion and fecundity by making females sterile (Nogge, 1976; Dale and Welburn, 2001; Pais et al., 2008), although supplementation of the host bloodmeal with vitamins partially reverses this infertility phenotype (Nogge, 1981). It has been suggested that the bacterium may also influence host digestive processes through either synthesis of protoheme or glycerophospholipids, fatty acids and steroids (Pais et al., 2008). Wigglesworthia is vertically transmitted from the female fly to her progeny via milk gland secretions (Denlinger and Ma, 1975). More specifically, in situ hybridization has identified Wigglesworthia exclusively associated with the milk gland lumen and the canals leading to the secretory reservoirs (Attardo et al., 2008). Recently, studies by Pais et al. (2008) have demonstrated that selective elimination of Wigglesworthia symbionts by ampicillin antibiotic treatment leads to increased susceptibility to trypanosome midgut infection in non-teneral flies. This implies that symbiont clearance may lead to a decrease in fly basal immunity, thus affecting host immune responses to trypanosome infection. However, these Wiggleworthia-cleared flies also had a compromised ability to digest their bloodmeals and increased trypanosome susceptibility was only observed in older flies. Thus, neither the effect of ‘‘starvation’’, known to cause increased parasite susceptibility (Kubi et al., 2006), nor the effects of antibiotic treatment on the fly and its other symbionts can be discounted.

THE ENEMY WITHIN

3.2

153

WOLBACHIA PIPIENTIS

A secondary tsetse symbiont, Wolbachia pipientis, is also maternally inherited. Unlike Wigglesworthia, it colonizes the female reproductive organs (not the milk gland) and is transovarially transmitted to progeny. Wolbachia are the most frequently described symbiotic association in arthropods; surveys have indicated that more than 70% of all insect species are infected with Wolbachia (Werren and Windsor, 2000). Wolbachia belong to the subdivision of Gramnegative intracellular alpha-proteobacteria and are most famous for their reproductive manipulations of the host including parthenogenesis, feminization, male-killing and cytoplasmic incompatibility (for a review of Wolbachia insect symbioses see Siozios et al., 2008). It has been reported that 100% of laboratory-reared tsetse colonies are infected with Wolbachia, while infections in wild populations vary significantly (Cheng et al., 2000). Within infected tsetse, Wolbachia colonizes several tissues, depending on the species of Glossina. Using a Wolbachia-specific PCR-based assay, Wolbachia was found in the reproductive tissues of all infected tsetse species examined. However, a wider tissue tropism was observed in the somatic tissues (gut, head, muscle, fat body, milk gland and salivary gland) in certain populations of G. austeni (Cheng et al., 2000). The role Wolbachia plays in the tsetse is unknown and whether its presence influences trypanosome infection rates by stimulating the fly immune system remains unknown. There are strong incentives to investigate whether or not Wolbachia induce cytoplasmic incompatibility in tsetse because it is one means by which selective traits can be driven through insect populations. As such, it is a highly desirable tool for promoting the success of paratransgenesis, a potential disease control strategy involving engineered symbionts and alteration of fly susceptibility to parasite infection – see Section 4.2 (Aksoy et al., 2003; Aksoy and Rio, 2005). 3.3

SODALIS GLOSSINIDIUS

The third tsetse symbiont, S. glossinidius, is a Gram-negative, microaerophilic, non-motile bacterium that forms a new bacterial taxon and species in the family Enterobacteriaceae (Dale and Maudlin 1999). Reinhardt et al. (1972) first described Sodalis as small rickettsia-like organisms (RLO) that were separated from the cytoplasm of midgut cells by a clear lytic zone. Pinnock and Hess (1974) confirmed these observations and reported the presence of this pleomorphic microbe in other tissues such as the fat body and ovaries. The development of a Sodalis-specific PCR-assay further established that this symbiont resided in the midgut, haemolymph and milk gland of teneral flies (Cheng and Aksoy, 1999; Attardo et al., 2008). Sodalis can live both intracellularly and as free-living forms in the gut lumen. This bacterium is one of the only insect symbionts that has been adapted to

154


in vitro culture (Dale and Maudlin, 1999; Matthew et al., 2005), making it a model organism particularly for research on host cell invasion, evolutionary biology (the transition from free-living to obligate endosymbiosis) and phylogenetic associations. In recent years, two new symbionts belonging to the Sodalis lineage have been identified, a bacterial isolate from the bloodsucking fly, Craterina melbae (Novakova and Hypsa, 2007) and a bacteriocyteassociated symbiont from the chewing louse, Columbicola columbae (Fukatsu et al., 2007). Unlike other bacterial symbionts, Sodalis does not produce catalase (characteristic of several microaerophiles) which, when combined with the results of additional phenotypic tests, indicates that Sodalis has a reduced biochemical profile when compared to other members of the family Enterobacteriaceae (Dale and Maudlin, 1999). Initially, 85% of the Sodalis genome was characterized by hybridization to E. coli gene macroarrays (Akman and Aksoy, 2001) and many potential gene products were identified. The use of E. coli arrays of course did not allow the identification of molecules unique to Sodalis, as exemplified by the failure to detect genes of the type III secretion system (Dale and Welburn, 2001) and genes encoding two chitinases. In 2006, the genome of Sodalis was published (Toh et al., 2006). Unlike Wigglesworthia, Sodalis has a much larger genome (4.1 Mb), but surprisingly it is predicted to encode only 2432 proteins (which is approximately 50% of what would be expected of a free-living organism). The presence of 972 pseudogenes indicates this genome is gradually eroding, although a previously designated pseudogene (carA) has now been shown to be functional and transcriptionally responsive despite its truncation (Pontes et al., 2008). Extrachromosomal DNA revealed another 160 potential ORFs located on four circular elements (Darby et al., 2005). While this would indicate a general transition to an obligate mutualistic association with the tsetse, several virulence related-factors such as haemolysin, phospholipases and invasion proteins have maintained functionality, implying that pathogenic genes are being retained (Dale and Welburn, 2001; Toh et al., 2006; Feldhaar and Gross 2009). Observations by Pinnock and Hess (1974) support this virulence concept as originally Sodalis was described as a pathogen based on electron microscopy of infected tsetse tissues; the host lytic zones surrounding Sodalis might be indicative of an adverse reaction between the tsetse host and the bacteria. In addition, the highly infected fat body in Glossina pallipides showed signs of cellular disruption and degeneration that could not be explained as procedural artefacts. This cellular disruption was also observed when Sodalis, isolated from the haemolymph of tsetse, were cultured on a layer of Aedes albopictus feeder cells (Welburn et al., 1987). Since Sodalis exists both extracellularly in the haemolymph and intracellularly within tsetse midgut epithelial cells and other tissues, the invasive form may represent a pathogenic remnant. Analysis of the Sodalis transcriptome indicates that quorum sensing may be used to alter gene expression in response to intracellular symbiont density (Pontes et al., 2008). It is further suggested that this may be a key adaptation

THE ENEMY WITHIN

155

strategy, with Sodalis and host tissues both modulating gene expression profiles and metabolic activities so that immune responses (in particular oxidative stress responses) are minimized. When the impact of different stresses on all three symbiont communities in the tsetse was assessed, the authors concluded that there was a distinct flexibility in symbiont density (with the exception of the teneral stage of the insect), which may help reduce friction between symbionts and host tissues (Rio et al., 2006). The biological contribution, if any, of Sodalis to the tsetse is unknown. A mutualistic relationship has been postulated (and generally accepted) as functional genes encoding enzymes required for vitamin synthesis (Akman and Aksoy, 2001; Akman et al., 2002) (as observed in Wigglesworthia) are present. Such a facultative association is supported by research showing that tsetse longevity is reduced in flies lacking Sodalis (Dale and Welburn, 2001). In addition, puparia that are heavily infected with Sodalis display increased survival under adverse conditions (Baker et al., 1990). It has been suggested that Sodalis may influence tsetse vector competence. Increased trypanosome midgut infections were reported in both lab-reared and wild tsetse that hosted high densities of rickettsia-like organisms (RLOs) (Maudlin and Ellis, 1985; Maudlin et al., 1990). By comparing genetically susceptible and refractory lines of G. m. morsitans, it was observed that the midguts of the susceptible line were prone to the heaviest symbiont infections. To further investigate the role of symbionts (the involvement of Sodalis is inferred) in vector competency, comparisons between refractory lines of G. m. morsitans (Maudlin, 1982) and G. m. centralis (Moloo and Kutuza, 1988a) showed that parasite susceptibility is a maternally inherited factor. In addition, puparial development at lower temperatures revealed a drastic reduction of RLO numbers in teneral flies. This symbiont-depressed tsetse population subsequently produced flies with a greater resistance to trypanosome infection (Welburn and Maudlin, 1991). Examination of the midguts of wild caught tsetse also revealed a direct correlation between trypanosome infections and the density of RLOs. Indeed, it was stated that a wild fly carrying RLOs was six times more likely to be infected with trypanosomes than a RLO-free fly (Maudlin et al., 1990). However, other populations of wild fly do not reflect such a direct relationship between bacterial densities and tsetse refractoriness or susceptibility (Moloo and Shaw, 1989; Geiger et al., 2005b). The vectorial competence between different species of Glossina is highly variable (Harley and Wilson, 1968; Moloo and Kutuza, 1988a,b; Reifenberg et al., 1997; Kazadi et al., 2000). Using this fact as a basis, Geiger et al. (2005b) demonstrated no correlation between Sodalis prevalence and maturation of a Trypanosoma congolense infection in two distinct species of tsetse. However, they could not rule out a potential role for Sodalis in the initial parasite establishment phase in the fly midgut. Further work from the same group investigated the genetic diversity between Sodalis isolated from the same two species of flies using amplified fragment length polymorphism markers (AFLP) (Geiger et al., 2005a). The data

156


(albeit weak) implied that vector competence was related to the genetic diversity of Sodalis (not merely the absence/presence of the symbiont) as the symbiont populations from each fly were distinct. The genetic diversity of Sodalis was further interrogated by screening symbionts isolated from trypanosome infected and uninfected flies (Geiger et al., 2007). The ability of a specific parasite species to establish in the insect midgut is statistically linked to the Sodalis genotype present. To address host-specificity among Sodalis, Weiss et al. (2006) transinfected two species of tsetse (previously cleared of native Sodalis with ampicillin) with reciprocal Sodalis strains. Equal symbiont densities were obtained in surrogate hosts without seriously deleterious effects on the fly. Regrettably, determining the effect of Sodalis transfection on the flies’ ability to vector trypanosomes was not reported. Although the above research does not unequivocally link the tsetse symbionts to vector competence it does suggest that symbionts are involved and forms a strong foundation for future investigations.

4

Towards new methods of disease control

Current interventions for management and control of trypanosomiasis are focused predominantly on drug chemotherapy and tsetse fly control especially for the control of nagana (Torr et al., 2005). While both these existing methods are successful, their implementation over extended time periods has not always been sustainable (Vale 1982; Aksoy et al., 2003). Additionally, concern over the development of resistance to available drugs and their toxicity is mounting, and the prospects for new drugs to treat HAT and nagana are bleak (Legros et al., 2002; Fevre et al., 2006). New means of controlling HAT are urgently required. We may find new ways of controlling trypanosomiasis by exploiting the new information becoming available from the genome projects on both tsetse and trypanosomes. To exploit these resources to their fullest we need a means of studying gene function in tsetse and means of expressing transgenic constructs in the fly. 4.1

GENE KNOCKDOWN IN GLOSSINA

The very slow reproductive rate in tsetse (one offspring every 9 days; 40 days adult to adult) and the high costs of maintaining colonies have prevented maintenance of multiple mutant fly lines. In addition, the unusual, viviparous reproductive system of tsetse flies prevents the development of germline transgenesis. Consequently, gene knockdown through RNA interference (RNAi) (Fire et al., 1998) has been of particular value in the study of tsetse biology and gene function. RNAi operates as a means of gene-specific post-transcriptional knockdown and is triggered by double-stranded RNA (dsRNA) mediated degradation of homologous mRNA sequences in the target organism. In the

THE ENEMY WITHIN

157

current model of RNAi two major steps are involved. The initiator phase involves recognition of dsRNA (that is expressed in or introduced into the cell) and cleavage into short interfering RNA (siRNA) fragments 21–23 bp in length by the enzyme Dicer (Zamore, 2000; Hammond, 2005). The subsequent effector phase involves the incorporation of cleaved dsRNA into a multi-protein complex, known as the RNA-induced silencing complex (RISC), capable of silencing homologous mRNA transcripts (Siomi and Siomi, 2009). Although several methods exist to administer dsRNA to the organism of interest, these are often very species specific. Stable transgenic expression of RNA hairpin constructs (Tavernarakis et al., 2000; Kennerdell and Carthew, 2000), and the use of recombinant viruses to deliver the dsRNA (Travanty et al., 2004) have been used in some Diptera. However, the most commonly used method for dsRNA delivery to insects is the direct injection of dsRNA into the haemocoel. This was first demonstrated in A. gambiae by Blandin et al. (2002), who showed successful gene knockdown of the HDP defensin. This technique has been successfully extended to the analysis of Glossina genes (Hu and Aksoy, 2006; Lehane et al., 2008; Attardo et al., 2008). To date three G. m. morsitans genes, a transferrin, an attacin and tsetseEP protein, have been implicated in tsetse–trypanosome interactions through use of this technique. Knockdown of these genes by dsRNA injection resulted in a statistically significant increase in trypanosome midgut prevalence (Hu and Aksoy, 2006; Lehane et al., 2008; Haines et al., 2009, submission pending). While it is universally practiced, gene knockdown by direct injection of dsRNA into the tsetse haemocoel (and the physical damage that it causes) is clearly not ideal when studying fly immunity. For example, control injections with PBS result in sustained upregulation of attacin and defensin transcripts in tsetse fat body until 18–30 h post-injection (Hao et al., 2001). Cuticular damage is known to stimulate immune responses in other insects including Bombyx mori and A. gambiae (Brey et al., 1993; Han et al., 1999). Thus any injection based RNAi studies on immunity, particularly those of tsetse–trypanosome interactions, should be interpreted with caution as it is known that the establishment of trypanosomes in the tsetse midgut is influenced by, among other factors, the fly immune system (Hao et al., 2001; Hu and Aksoy, 2006; Lehane et al., 2008). In addition, high mortality rates can occur in flies following injection (Walshe et al., 2009) and this too can complicate the interpretation of experimental results because gene knockdown may have occurred to differing extents in the surviving and killed flies. More recently, successful local gene knockdown in Glossina has been demonstrated by ingestion of dsRNA in a bloodmeal (Walshe et al., 2009). Feeding dsRNA may serve as an alternative and preferable means of delivering dsRNA as oral administration is a more natural route of delivery and less invasive than dsRNA injection. In this study, a comparative analysis of knockdown of the immunoresponsive protein tsetseEP (Haines et al., 2005), at both the transcript and protein level in the midgut, was followed after either feeding or injecting dsRNA. High knockdown efficiency was observed using either method of dsRNA delivery, but a

158


significantly lower mortality rate was observed following feeding dsRNA compared to injection. Although the midgut specific tsetseEP protein was successfully knocked down by feeding dsRNA, knockdown of the fat bodyexpressed transferrin gene failed (Walshe et al., 2009). However, this gene can be knocked down by dsRNA injection (Lehane et al., 2008). Failure of the RNAi signal to spread beyond the midgut epithelium may be due to the apparent absence of an ortholog to SID-1, the gene responsible for import of circulating RNAi silencing signals in C. elegans, from available Glossina EST databases (and from the Drosophila genome). So, it is possible that the RNAi signal is not distributed systemically in Diptera (Van Roessel et al., 2002; Winston et al., 2002; Roignant et al., 2003; Dietzl et al., 2007; Jose et al., 2009). Thus, while fed dsRNA can clearly enter midgut cells, it may be unable to cross the midgut epithelial barrier in order to cause gene knockdown in tissues beyond the midgut. If this proves to be the case, it would be a valuable tool in itself because the use of two different methods of dsRNA delivery may permit dissection of tissue-specific gene function by permitting tissue specific gene knockdown. This study was the first demonstration of successful gene knockdown in a dipteran by feeding dsRNA. As gene knockdown by feeding dsRNA has been achieved in two other blood feeding arthropods, Ixodes scapularis and Rhodnius prolixus (Soares et al., 2005; Araujo et al., 2006), it is possible this useful molecular tool may be more widely available in blood feeding insects. 4.2

PARATRANSGENESIS

Transgenesis (i.e. the introduction of foreign genes into a living organism) is currently impossible in tsetse flies due to the viviparous lifecycle. So, there is no prospect of disease control through direct transgenic means. However, one proposed approach to trypanosomiasis control involves the modulation of vector competence through paratransgenesis (Aksoy et al., 2003; Weiss et al., 2008). This involves the genetic transformation of symbiotic bacteria residing in the insect to produce anti-parasitic molecules into the midgut. This approach has proven successful in the laboratory, reducing T. cruzi transmission by R. prolixus via expression of a HDP, cecropin A, by its endosymbiont Rhodococcus rhodnii (Durvasula et al., 1997). The foreign genes expressed by the symbiotic bacterium would have trypanocidal properties that would in principal reduce or eliminate trypanosome midgut establishment, thus breaking the parasite transmission cycle (Durvasula et al., 1997; Aksoy et al., 2003). The endosymbiont Sodalis is the target expression vector in the tsetse fly. This bacterium resides in the tsetse midgut as well as other body sites. Consequently, invading brucei and congolense group trypanosomes would encounter trypanocidal secretion products before the trypanosomes could establish in the midgut. Already, recombinant Sodalis produced in vitro may be reintroduced by microinjection into the mother’s haemolymph and subsequently passed to intrauterine progeny

THE ENEMY WITHIN

159

in utero (Cheng and Aksoy, 1999). As Sodalis can be cultured in vitro (Matthew et al., 2005) and a genetic transformation system has already been developed (Beard et al., 1993), the potential for successful paratransgenesis is high. To date, several candidate anti-trypanosomal molecules have been identified which could be introduced into Sodalis to control trypanosome midgut establishment. One candidate is tsetse attacin (Hu and Aksoy, 2006). Less obvious, but potentially more powerful candidate molecules are also under investigation. For example, BMAP-18 is a truncated form of the bovine myeloid antimicrobial peptide-27 (BMAP-27). BMAP-27 (Skerlavaj et al., 1996) is expressed by bovine neutrophils and exhibits low toxicity to mammalian cells, insect cells and Sodalis, yet causes rapid death to both BSF and PCF trypanosomes (Haines et al., 2003). The truncated BMAP-18 peptide can kill a variety of kinetoplastid parasites, including trypanosomes and Leishmania, yet exhibits reduced cytotoxicity to Sodalis, mammalian and insect cell lines (Haines et al., 2009). Thus BMAP-18 is also a strong candidate for use in a paratrangenesis approach. Several technical problems will need to be overcome before this technology can be applied to disease control. A suitable biological drive system, such as use of Wolbachia (Aksoy et al., 2003), would be required to ensure replacement of the susceptible wild tsetse population with the refractory population. Additionally, this approach would only be suitable for trypanosome species (brucei and congolense groups) that establish in the fly midgut. Therefore, another tactic would be required to control trypanosome species such as T. vivax, which establish in the mouthparts and would not be affected. How rapidly trypanosome resistance would appear with such a system in place is an unknown and a concern. Therefore, it would probably be advantageous to have a selection of effector molecules available for use in a managed control strategy.

5

Conclusion

This review has come at an important time. The genome resources now available for tsetse flies, their symbionts and the trypanosomes they transmit have dramatically increased possible lines of scientific enquiry. Currently, the GeneDB and Vectorbase data repositories contain comprehensive, annotated Glossina EST libraries and completed annotation of the full tsetse genome is projected for 2011. That will mean the complete repertoire of fly, symbiont and trypanosome genomes will soon be available. There are also a range of molecular tools, including RNAi, proteomics, mutant trypanosome lines and an in vitro culture system for Sodalis, that will further our research endeavours on the enemy within. We believe that the technical resources now available mean that it is time for a more integrated approach in research on tsetse–trypanosome interactions rather than the traditional, more fragmentary, approach based on entomology, bacteriology or trypanosome biology. To this end we have used this review to try to make the pertinent entomological information more available to a wide audience across

160


the subject. We have taken an entomological viewpoint to examine the interacting biology of the three major components of the tsetse–trypanosome interaction, the fly, its symbiotic bacteria and the invading trypanosome. We have also tried to highlight the problems we believe are associated with some experimental approaches currently adopted in the entomological field. For example, routinely feeding trypanosomes in a blood meal containing glucosamine; failure to precisely define the starvation period prior to the infective bloodmeal; infecting flies with procyclic rather than blood stream form trypanosomes; the inappropriateness in some studies, particularly susceptibility studies, of using long established fly and trypanosome laboratory cultures with no view to the geographical origins of either. Continued use of these experimental approaches may well create increasing problems in interpretation of data. We believe that more field-based studies, despite their difficulties, would be a valuable addition to the understanding of the tsetse–trypanosome interaction. Also, the majority of studies on this vector–parasite system have centered on T. brucei, with T. congolense and T. vivax interactions comparatively lightly investigated. Given the economic importance of both T. vivax and T. congolense and the available genomic resources more emphasis on relations of these trypanosomes with Glossina is also called for. Finally, we are of the opinion that with major improvements in the technical resources available for experimentation of tsetse–trypanosome interactions the remaining major brake slowing progress in the tsetse–trypanosome research field continues to be the high cost and difficulty of tsetse fly colony maintenance. This difficulty has resulted in only a small number of laboratories studying this phenomenon. It is important that funding bodies are fully aware of this fundamental problem if this field of research is to expand. Acknowledgements The authors thank Alvaro Acosta-Serrano, Leyla Akman, Geoffrey Attardo, Lori Peacock and Terry Pearson for their critical reading of the manuscript and helpful comments. Any remaining errors in the manuscript are the responsibility of the authors. We are also grateful to Ray Wilson for kindly granting permission to use his image of the tsetse fly. References Abubakar, L. U., Bulimo, W. D., Mulaa, F. J. and Osir, E. O. (2006). Molecular characterization of a tsetse fly midgut proteolytic lectin that mediates differentiation of African trypanosomes. Insect Biochem. Mol. Biol. 36, 344–352. Acosta-Serrano, A., Vassella, E., Liniger, M., Renggli, C. K., Brun, R., Roditi, I. and Englund, P. T. (2001). The surface coat of procyclic Trypanosoma brucei: programmed expression and proteolytic cleavage of procyclin in the tsetse fly. Proc. Natl. Acad. Sci. USA 98, 1513–1518.

THE ENEMY WITHIN

161

Aggarwal, K. and Silverman, N. (2008). Positive and negative regulation of the Drosophila immune response. BMB Rep. 41, 267–277. Akman, L. and Aksoy, S. (2001). A novel application of gene arrays: Escherichia coli array provides insight into the biology of the obligate endosymbiont of tsetse flies. Proc. Natl. Acad. Sci. USA 98, 7546–7551. Akman, L., Yamashita, A., Watanabe, H., Oshima, K., Shiba, T., Hattori, M. and Aksoy, S. (2002). Genome sequence of the endocellular obligate symbiont of tsetse flies, Wigglesworthia glossinidia. Nat. Genet. 32, 402–407. Aksoy, S. (1995). Wigglesworthia gen. nov. and Wigglesworthia glossinidia sp. nov., taxa consisting of the mycetocyte-associated, primary endosymbionts of tsetse flies. Int. J. Syst. Bacteriol. 45, 848–851. Aksoy, S. and Rio, R. V. M. (2005). Interactions among multiple genomes: tsetse, its symbionts and trypanosomes. Insect Biochem. Mol. Biol. 35, 691–698. Aksoy, S., Pourhosseini, A. A. and Chow, A. (1995). Mycetome endosymbionts of tsetse flies constitute a distinct lineage related to Enterobacteriaceae. Insect Mol. Biol. 4, 15–22. Aksoy, S., Gibson, W. C. and Lehane, M. J. (2003). Interactions between tsetse and trypanosomes with implications for the control of trypanosomiasis. Adv. Parasitol. 53, 1–83. Araujo, R. N., Santos, A., Pinto, F. S., Gontijo, N. F., Lehane, M. J. and Pereira, M. H. (2006). RNA interference of the salivary gland nitrophorin 2 in the triatomine bug Rhodnius prolixus (Hemiptera: Reduviidae) by dsRNA ingestion or injection. Insect Biochem. Mol. Biol. 36, 683–693. Attardo, G. M., Guz, N., Strickler-Dinglasan, P. and Aksoy, S. (2006). Molecular aspects of viviparous reproductive biology of the tsetse fly (Glossina morsitans morsitans): regulation of yolk and milk gland protein synthesis. J. Insect Physiol. 52, 1128–1136. Attardo, G. M., Lohs, C., Heddi, A., Alam, U. H., Yildirim, S. and Aksoy, S. (2008). Analysis of milk gland structure and function in Glossina morsitans: milk protein production, symbiont populations and fecundity. J. Insect Physiol. 54, 1236–1242. Baker, R. D., Maudlin, I., Milligan, P. J. M., Molyneux, D. H. and Welburn, S. C. (1990). The possible role of rickettsia-like organisms in trypanosomiasis epidemiology. Parasitology 100, 209–217. Balasegaram, M., Young, H., Chappluis, F., Priotto, G., Raguenaud, M. E. and Checchi, F. (2009). Effectiveness of melarsoprol and eflornithine as first-line regimens for gambiense sleeping sickness in nine Me´decins Sans Frontie`res programmes. Trans. R. Soc. Trop. Med. Hyg. 103, 280–290. Barrett, M. P. (2006). The rise and fall of sleeping sickness. Lancet 367, 1377–1378. Barry, J. D. and McCulloch, R. (2001). Antigenic variation in trypanosomes: enhanced phenotypic variation in a eukaryotic parasite. Adv. Parasitol. 49, 1–70. Barry, J. D., Graham, S. V., Fotheringham, M., Graham, V. S., Kobryn, K. and Wymer, B. (1998). VSG gene control and infectivity strategy of metacyclic stage Trypanosoma brucei. Mol. Biochem. Parasitol. 91, 93–105. Barry, J. D., Marcello, L., Morrison, L. J., Read, A. F., Lythgoe, K., Jones, N., Carrington, M., Blandin, G., Bo¨hme, U., Caler, E., Hertz-Fowler, C. Renauld, H., et al. (2005). What the genome sequence is revealing about trypanosome antigenic variation. Biochem. Soc. Trans. 33, 986–989. Bayne, R. A. L., Kilbride, E. A., Lainson, F. A., Tetley, L. and Barry, J. D. (1993). A major surface antigen of procyclic stage Trypanosoma congolense. Mol. Biochem. Parasitol. 61, 295–310. Beard, C. B., Oneill, S. L., Tesh, R. B., Richards, F. F. and Aksoy, S. (1993). Modification of arthropod vector competence via symbiotic bacteria. Parasitol. Today 9, 179–183.

162


Beecroft, R. P., Roditi, I. and Pearson, T. W. (1993). Identification and characterization of an acidic major surface glycoprotein from procyclic stage Trypanosoma congolense. Mol. Biochem. Parasitol. 61, 285–294. Bienen, E. J., Hammadi, E. and Hill, G. C. (1980). Initiation of trypanosome tranformation from blood-stream trypomastigotes to procyclic trypomastigotes. J. Parasitol. 66, 680–682. Black, S. J., Wang, Q., Makadzange, T., Li, Y. L., Van Praagh, A., Loomis, M. and Seed, J. R. (1999). Anti-Trypanosoma brucei activity of nonprimate zoo sera. J. Parasitol. 85, 48–53. Blandin, S., Moita, L. F., Kocher, T., Wilm, M., Kafatos, F. C. and Levashina, E. A. (2002). Reverse genetics in the mosquito Anopheles gambiae: targeted disruption of the defensin gene. EMBO Rep. 3, 852–856. Bo¨hringer-Schweizer, S. (1977). Digestion in the tsetse fly: an ultrastructural analysis of structure and function of the midgut epithelium in Glossina morsitans morsitans (Machado), (Diptera: Glossinidae). Ph.D. thesis. Universita¨t Basel, Switzerland. Boulanger, N., Brun, R., Ehret-Sabatier, L., Kunz, C. and Bulet, P. (2002a). Immunopeptides in the defense reactions of Glossina morsitans to bacterial and Trypanosoma brucei brucei infections. Insect Biochem. Mol. Biol. 32, 369–375. Boulanger, N., Munks, R. J. L., Hamilton, J. V., Vovelle, F., Brun, R., Lehane, M. J. and Bulet, P. (2002b). Epithelial innate immunity. A novel antimicrobial peptide with antiparasitic activity in the blood-sucking insect Stomoxys calcitrans. J. Biol. Chem. 277, 49921–49926. Bouyer, J., Pruvot, M., Bengaly, Z., Guerin, P. M. and Lancelot, R. (2007). Learning influences host choice in tsetse. Biol. Lett. 3, 113–117. Brey, P. T., Lee, W. J., Yamakawa, M., Koizumi, Y., Perrot, S., François, M. and Ashida, M. (1993). Role of the integument in insect immunity – epicuticular abrasion and induction of cecropin synthesis in cuticular epithelial cells. Proc. Natl. Acad. Sci. USA 90, 6275–6279. Broden, A. (1904). Les infections a` trypanosomes au Congo chez l’homme et les animaux (communication pre´liminaire). Bull. Soc. E´tud. Colon. 11, 116–139. Brown, R. C., Evans, D. A. and Vickerman, K. (1973). Changes in oxidative metabolism and ultrastructure accompanying differentiation of the mitochondrion in Trypanosoma brucei. Int. J. Parasitol. 3, 691–704. Bruce, D., Harvey, D., Hamerton, A. E., Davey, J. B. and Bruce, M. (1912). The morphology of T. simiae, sp. nov. Proc. Biol. Sci. 85, 477–481. Bursell, E. (1981). Energetics of hematophagous arthropods: influence of parasites. Parasitology 82, 107–108. Burtt, E. (1946a). Incubation of tsetse pupae: increased transmission rate of Trypanosoma rhodesiense in Glossina morsitans. Ann. Trop. Med. Parasitol. 40, 18–28. Burtt, E. (1946b). The sex ratio of infected flies found in transmission-experiments with Glossina morsitans and Trypanosoma rhodesiense. Ann. Trop. Med. Parasitol. 40, 74–79. Butikofer, P., Vassella, E., Boschung, M., Renggli, C. K., Brun, R., Pearson, T. W. and Roditi, I. (2002). Glycosylphosphatidylinositol-anchored surface molecules of Trypanosoma congolense insect forms are developmentally regulated in the tsetse fly. Mol. Biochem. Parasitol. 119, 7–16. Chandra, M., Liniger, M., Tetley, L., Roditi, I. and Barry, J. D. (2004). TsetseEP, a gut protein from the tsetse Glossina morsitans, is related to a major surface glycoprotein of trypanosomes transmitted by the fly and to the products of a Drosophila gene family. Insect Biochem. Mol. Biol. 34, 1163–1173.

THE ENEMY WITHIN

163

Cheeseman, M. T. and Gooding, R. H. (1985). Proteolytic enzymes from tsetse flies, Glossina morsitans and Glossina palpalis (Diptera, Glossinidae). Insect Biochem. 15, 677–680. Chen, X. A., Li, S. and Aksoy, S. (1999). Concordant evolution of a symbiont with its host insect species: molecular phylogeny of genus Glossina and its bacteriomeassociated endosymbiont, Wigglesworthia glossinidia. J. Mol. Evol. 48, 49–58. Cheng, Q. and Aksoy, S. (1999). Tissue tropism, transmission and expression of foreign genes in vivo in midgut symbionts of tsetse flies. Insect Mol. Biol. 8, 125–132. Cheng, Q., Ruel, T. D., Zhou, W., Moloo, S. K., Majiwa, P., O’Neill, S. L. and Aksoy, S. (2000). Tissue distribution and prevalence of Wolbachia infections in tsetse flies, Glossina spp. Med. Vet. Entomol. 14, 44–50. Courtin, D., Berthier, D., Thevenon, S., Dayo, G. K., Garcia, A. and Bucheton, B. (2008). Host genetics in African trypanosomiasis. Infect. Genet. Evol. 8, 229–238. Croft, S. L., East, J. S. and Molyneux, D. H. (1982). Anti-trypanosomal factor in the hemolymph of Glossina. Acta Trop. 39, 293–302. Cross, G. A. M. (1996). Antigenic variation in trypanosomes: secrets surface slowly. Bioessays 18, 283–291. Cruz, F. S., Berens, R. L. and Marr, J. J. (1983). Xanthine oxidase in calf serum: formation of oxygen metabolites that are toxic for Trypanosoma cruzi in culture. J. Parasitol. 69, 237–239. Dale, C. and Maudlin, I. (1999). Sodalis gen. nov. and Sodalis glossinidius sp. nov., a microaerophilic secondary endosymbiont of the tsetse fly Glossina morsitans morsitans. Int. J. Syst. Bacteriol. 49, 267–275. Dale, C. and Welburn, S. C. (2001). The endosymbionts of tsetse flies: manipulating host-parasite interactions. Int. J. Parasitol. 31, 628–631. Dale, C., Welburn, S. C., Maudlin, I. and Milligan, P. J. M. (1995). The kinetics of maturation of trypanosome infections in tsetse. Parasitology 111, 187–191. Darby, A. C., Lagnel, J., Matthew, C. Z., Bourtzis, K., Maudlin, I. and Welburn, S. C. (2005). Extrachromosomal DNA of the symbiont Sodalis glossinidius. J. Bacteriol. 187, 5003–5007. De Gregorio, E., Spellman, P. T., Tzou, P., Rubin, G. M. and Lemaitre, B. (2002). The Toll and Imd pathways are the major regulators of the immune response in Drosophila. EMBO J. 21, 2568–2579. Delespaux, V. and de Koning, H. P. (2007). Drugs and drug resistance in African trypanosomiasis. Drug Resist. Updat. 10, 30–50. Denlinger, D. L. and Ma, W. C. (1975). Maternal nutritive secretions as possible channels for vertical transmission of microorganisms in insects – tsetse fly example. Ann. NY Acad. Sci. 266, 162–165. Dietzl, G., Chen, D., Schnorrer, F., Su, K. C., Barinova, Y., Fellner, M., Gasser, B., Kinsey, K., Oppel, S., Scheiblauer, S., Couto, A. Marra, V., et al. (2007). A genomewide transgenic RNAi library for conditional gene inactivation in Drosophila. Nature 448, 151–156. Dipeolu, O. O. and Adam, K. M. (1974). On the use of membrane feeding to study the development of Trypanosoma brucei in Glossina. Acta Trop. 32, 185–201. Distelmans, W., Dhaeseleer, F., Kaufman, L. and Rousseeuw, P. (1982). The susceptibility of Glossina palpalis palpalis at different ages to infection with Trypanosoma congolense. Ann. Soc. Belge Med. Trop. 62, 41–47. Dong, Y. M. and Dimopoulos, G. (2009). Anopheles fibrinogen-related proteins provide expanded pattern recognition capacity against bacteria and malaria parasites. J. Biol. Chem. 284, 9835–9844.

164


Dong, Y. M., Taylor, H. E. and Dimopoulos, G. (2006). AgDscam, a hypervariable immunoglobulin domain-containing receptor of the Anopheles gambiae innate immune system. PLoS Biol. 4, 1137–1146. Durvasula, R. V., Gumbs, A., Panackal, A., Kruglov, O., Aksoy, S., Merrifield, R. B., Richards, F. F. and Beard, C. B. (1997). Prevention of insect-borne disease: an approach using transgenic symbiotic bacteria. Proc. Natl. Acad. Sci. USA 94, 3274–3278. Ebikeme, C. E., Peacock, L., Coustou, V., Riviere, L., Bringaud, F., Gibson, W. C. and Barrett, M. P. (2008). N-acetyl D-glucosamine stimulates growth in procyclic forms of Trypanosoma brucei by inducing a metabolic shift. Parasitology 135, 585–594. Ellis, D. S. and Evans, D. A. (1977). Passage of Trypanosoma brucei rhodesiense through peritrophic membrane of Glossina morsitans morsitans. Nature 267, 834–835. Engstler, M. and Boshart, M. (2004). Cold shock and regulation of surface protein trafficking convey sensitization to inducers of stage differentiation in Trypanosoma brucei. Genes Dev. 18, 2798–2811. Evans, D. A. and Ellis, D. S. (1983). Recent observations on the behaviour of certain trypanosomes within their insect hosts. Adv. Parasitol. 22, 1–42. Feldhaar, H. and Gross, R. (2009). Insects as hosts for mutualistic bacteria. Int. J. Med. Microbiol. 299, 1–8. Fenn, K. and Matthews, K. R. (2007). The cell biology of Trypanosoma brucei differentiation. Curr. Opin. Microbiol. 10, 539–546. Ferrante, A. and Allison, A. C. (1983). Alternative pathway activation of complement by African trypanosomes lacking a glycoprotein coat. Parasite Immunol. 5, 491–498. Fevre, E. M., Picozzi, K., Jannin, J., Welburn, S. C. and Maudlin, I. (2006). Human African trypanosomiasis: epidemiology and control. Adv. Parasitol. 61, 167–221. Fire, A., Xu, S. Q., Montgomery, M. K., Kostas, S. A., Driver, S. E. and Mello, C. C. (1998). Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391, 806–811. Ford, J. and Leggate, B. M. (1961). The geographical and climatic distribution of trypanosome infection rates in G. morsitans group of tsetse flies (Glossina Wied., Diptera). Trans. R. Soc. Trop. Med. Hyg. 55, 383–397. Fukatsu, T., Koga, R., Smith, W. A., Tanaka, K., Nikoh, N., Sasaki-Fukatsu, K., Yoshizawa, K., Dale, C. and Clayton, D. H. (2007). Bacterial endosymbiont of the slender pigeon louse, Columbicola columbae, allied to endosymbionts of grain weevils and tsetse flies. Appl. Environ. Microbiol. 73, 6660–6668. Geiger, A., Cuny, G. and Frutos, R. (2005a). Two tsetse fly species, Glossina palpalis gambiensis and Glossina morsitans morsitans, carry genetically distinct populations of the secondary symbiont Sodalis glossinidius. Appl. Environ. Microbiol. 71, 8941–8943. Geiger, A., Ravel, S., Frutos, R. and Cuny, G. (2005b). Sodalis glossinidius (Enterobacteriaceae) and vectorial competence of Glossina palpalis gambiensis and Glossina morsitans morsitans for Trypanosoma congolense savannah type. Curr. Microbiol. 51, 35–40. Geiger, A., Ravel, S., Mateille, T., Janelle, J., Patrel, D., Cuny, G. and Frutos, R. (2007). Vector competence of Glossina palpalis gambiensis for Trypanosoma brucei s.l. and genetic diversity of the symbiont Sodalis glossinidius. Mol. Biol. Evol. 24, 102–109. Geiger, A., Ravel, S., Mateille, T., Janelle, J., Patrel, D., Frutos, R. and Cuny, G. (2008). The human African trypanosomiasis: interactions between the tsetse fly, its secondary symbiont Sodalis glossinidius, and the parasite. Infect. Genet. Evol. 8, S25–S26. Gibson, W. and Bailey, M. (2003). The development of Trypanosoma brucei within the tsetse fly midgut observed using green fluorescent trypanosomes. Kinetoplastid Biol. Dis. 2, 1–13.

THE ENEMY WITHIN

165

Gibson, W., Peacock, L., Ferris, V., Williams, K. and Bailey, M. (2008). The use of yellow fluorescent hybrids to indicate mating in Trypanosoma brucei. Parasit. Vectors 1, 4. Gingrich, J. B., Ward, R. A., Macken, L. M. and Esser, K. M. (1982). African sleeping sickness: new evidence that mature tsetse flies (Glossina morsitans) can become potent vectors. Trans. R. Soc. Trop. Med. Hyg. 76, 479–481. Gooding, R. H. (1974a). Digestive processes of hematophagous insects- control of trypsin secretion in Glossina morsitans. J. Insect Physiol. 20, 957–964. Gooding, R. H. (1974b). Digestive processes of hematophagous insects V. Inhibitors of trypsin from Glossina morsitans morsitans (Diptera: Glossinidae). Can. Entomol. 106, 39–44. Gooding, R. H. (1988). Infection of post-teneral tsetse flies (Glossina morsitans morsitans and Glossina morsitans centralis) with Trypanosoma brucei brucei. Can. J. Zool. 66, 1289–1292. Gooding, R. H. and Rolseth, B. M. (1976). Digestive processes of hematophagous insects XI. Partial purification and some properties of 6 proteolytic enzymes from tsetse fly Glossina morsitans morsitans Westwood (Diptera: Glossinidae). Can. J. Zool. 54, 1950–1959. Graham, S. V., Terry, S. and Barry, J. D. (1999). A structural and transcription pattern for variant surface glycoprotein gene expression sites used in metacyclic stage Trypanosoma brucei. Mol. Biochem. Parasitol. 103, 141–154. Ha, E. M., Oh, C. T., Bae, Y. S. and Lee, W. J. (2005a). A direct role for dual oxidase in Drosophila gut immunity. Science 310, 847–850. Ha, E. M., Oh, C. T., Ryu, J. H., Bae, Y. S., Kang, S. W., Jang, I. H., Brey, P. T. and Lee, W. J. (2005b). An antioxidant system required for host protection against gut infection in Drosophila. Dev. Cell 8, 125–132. Haag, J., O’hUigin, C. and Overath, P. (1998). The molecular phylogeny of trypanosomes: evidence for an early divergence of the Salivaria. Mol. Biochem. Parasitol. 91, 37–49. Haines, L. R., Hancock, R. E. W. and Pearson, T. W. (2003). Cationic antimicrobial peptide killing of African trypanosomes and Sodalis glossinidius, a bacterial symbiont of the insect vector of sleeping sickness. Vector Borne Zoonotic Dis. 3, 175–186. Haines, L. R., Jackson, A. M., Lehane, M. J., Thomas, J. M., Yamaguchi, A. Y., Haddow, J. D. and Pearson, T. W. (2005). Increased expression of unusual EP repeat-containing proteins in the midgut of the tsetse fly (Glossina) after bacterial challenge. Insect Biochem. Mol. Biol. 35, 413–423. Haines, L. R., Thomas, J. M., Jackson, A. M., Eyford, B. A., Razavi, M., Watson, C. N., Gowen, B., Hancock, R. E. W. and Pearson, T. W. (2009). Killing of trypanosomatid parasites by a modified bovine host defense peptide, BMAP-18. PLoS Negl. Trop. Dis. 3, e373. Hammond, S. M. (2005). Dicing and slicing: the core machinery of the RNA interference pathway. FEBS Lett. 579, 5822–5829. Han, Y. S., Chun, J. S., Schwartz, A., Nelson, S. and Paskewitz, S. M. (1999). Induction of mosquito hemolymph proteins in response to immune challenge and wounding. Dev. Comp. Immunol. 23, 553–562. Hao, Z. R., Kasumba, I., Lehane, M. J., Gibson, W. C., Kwon, J. and Aksoy, S. (2001). Tsetse immune responses and trypanosome transmission: implications for the development of tsetse-based strategies to reduce trypanosomiasis. Proc. Natl. Acad. Sci. USA 98, 12648–12653. Hao, Z. G., Kasumba, I. and Aksoy, S. (2003). Proventriculus (cardia) plays a crucial role in immunity in tsetse fly (Diptera: Glossinidiae). Insect Biochem. Mol. Biol. 33, 1155–1164.

166


Harley, J. M. B. (1971). Comparison of susceptibility to infection with Trypanosoma rhodesiense of Glossina pallidipes, G. morsitans, G. fuscipes and G. brevipalpis. Ann. Trop. Med. Parasitol. 65, 185–189. Harley, J. M. B. and Wilson, A. J. (1968). Comparison between Glossina morsitans morsitans, G. pallidipes and G. fuscipes as vectors of trypanosomes of Trypanosoma congolense group- proportions infected experimentally and numbers of infective organisms extruded during feeding. Ann. Trop. Med. Parasitol. 62, 178–187. Herrera-Ortiz, A., Lanz-Mendoza, H., Martinez-Barnetche, J., Hernandez-Martinez, S., Villarreal-Trevino, C., Aguilar-Marcelino, L. and Rodriguez, M. H. (2004). Plasmodium berghei ookinetes induce nitric oxide production in Anopheles pseudopunctipennis midguts cultured in vitro. Insect Biochem. Mol. Biol. 34, 893–901. Hide, G., Buchanan, N., Welburn, S., Maudlin, I., Barry, J. D. and Tait, A. (1991). Trypanosoma brucei rhodesiense: characterisation of stocks from Zambia, Kenya and Uganda using repetitive DNA probes. Exp. Parasitol. 72, 430–439. Houseman, J. G. (1980). Anterior midgut proteinase inhibitor from Glossina morsitans morsitans Westwood (Diptera: Glossinidae) and its effects upon tsetse digestive enzymes. Can. J. Zool. 58, 79–87. Hu, Y. J. and Aksoy, S. (2005). An antimicrobial peptide with trypanocidal activity characterized from Glossina morsitans morsitans. Insect Biochem. Mol. Biol. 35, 105–115. Hu, C. Y. and Aksoy, S. (2006). Innate immune responses regulate trypanosome parasite infection of the tsetse fly Glossina morsitans morsitans. Mol. Microbiol. 60, 1194–1204. Hu, C. Y., Rio, R. V. M., Medlock, J., Haines, L. R., Nayduch, D., Savage, A. F., Guz, N., Attardo, G. M., Pearson, T. W., Galvani, A. P. and Aksoy, S. (2008). Infections with immunogenic trypanosomes reduce tsetse reproductive fitness: potential impact of different parasite strains on vector population structure. PLoS Negl. Trop. Dis. 2, e192. Hunt, M., Brun, R. and Kohler, P. (1994). Studies on compounds promoting the in vitro transformation of Trypanosoma brucei from bloodstream to procyclic forms. Parasitol. Res. 80, 600–606. Ibrahim, E. A. R., Ingram, G. A. and Molyneux, D. H. (1984). Haemagglutinins and parasite agglutinins in haemolymph and gut of Glossina. Tropenmed. Parasitol. 35, 151–156. Imbuga, M. O., Osir, E. O. and Labongo, V. L. (1992). Inhibitory effect of Trypanosoma brucei brucei on Glossina morsitans midgut trypsin in vitro. Parasitol. Res. 78, 273–276. Ingram, G. A. and Molyneux, D. H. (1988). Sugar specificities of anti-human ABO(H) blood group erythrocyte agglutinins (lectins) and hemolytic activity in the hemolymph and gut extracts of 3 Glossina species. Insect Biochem. 18, 269–279. Inoue, N., Lluz, A. T., Mori, T., Nagasawa, H., Fujisaki, K. and Mikami, T. (2000). Novel species specific antigens of Trypanosoma congolense and their different localization among life-cycle stages. J. Vet. Med. Sci. 62, 1041–1045. Jackson, C. H. N. (1933). Notes on a method of marking tsetse flies. J. Anim. Ecol. 2, 289–290. Jenni, L., Molyneux, D. H., Livesey, J. L. and Galun, R. (1980). Feeding behaviour of tsetse flies infected with salivarian trypanosomes. Nature 283, 383–385. Jenni, L., Marti, S., Schweizer, J., Betschart, B., Lepage, R. W. F., Wells, J. M., Tait, A., Paindavoine, P., Pays, E. and Steinert, M. (1986). Hybrid formation between African trypanosomes during cyclical transmission. Nature 322, 173–175. Jordan, A. M. (1965). The hosts of Glossina as the main factor affecting trypanosome infection rates of tsetse flies in Nigeria. Trans. R. Soc. Trop. Med. Hyg. 59, 423–431.

THE ENEMY WITHIN

167

Jordan, A. M. (1986). Trypanosomiasis Control and African Rural Development. Longman, London. Jose, A. M., Smith, J. J. and Hunter, C. P. (2009). Export of RNA silencing from C. elegans tissues does not require the RNA channel SID-1. Proc. Natl. Acad. Sci. USA 106, 2283–2288. Kabayo, J. P. and Langley, P. A. (1985). The nutritional importance of dietary blood components for reproduction in the tsetse fly, Glossina morsitans. J. Insect Physiol. 31, 619–624. Kazadi, J. M., Losson, B. and Kageruka, P. (2000). Compe´tence vectorielle des mouches non teńe´rales de Glossina morsitans morsitans (souche Mall) infecteés par Trypanosoma (Nannomonas) congolense IL 1180. Bull. Soc. Pathol. Exot. 93, 125–128. Kennerdell, J. R. and Carthew, R. W. (2000). Heritable gene silencing in Drosophila using double-stranded RNA. Nat. Biotechnol. 18, 896–898. Koffi, M., De Meeus, T., Bucheton, B., Solano, P., Camara, M., Kaba, D., Cuny, G., Ayala, F. J. and Jamonneau, V. (2009). Population genetics of Trypanosoma brucei gambiense, the agent of sleeping sickness in Western Africa. Proc. Natl. Acad. Sci. USA 106, 209–214. Krafsur, E. S. (2009). Tsetse flies: genetics, evolution, and role as vectors. Infect. Genet. Evol. 9, 124–141. Kristjanson, P. M., Swallow, B. M., Rowlands, G. J., Kruska, R. L. and de Leeuw, P. N. (1999). Measuring the costs of African animal trypanosomosis, the potential benefits of control and returns to research. Agr. Syst. 59, 79–98. Kubi, C., Van den Abbeele, J., Dorny, P., Coosemans, M., Marcotty, T. and Van den Bossche, P. (2005). Ability of trypanosome-infected tsetse flies (Diptera: Glossinidae) to acquire an infection with a second trypanosome species. J. Med. Entomol. 42, 1035–1038. Kubi, C., Van den Abbeele, J., De Deken, R., Marcotty, T., Dorny, P. and Van den Bossche, P. (2006). The effect of starvation on the susceptibility of teneral and non-teneral tsetse flies to trypanosome infection. Med. Vet. Entomol. 20, 388–392. Leak, S. G. A. (1999). Tsetse Biology and Ecology: Their Role in the Epidemiology and Control of Trypanosomiasis. CABI Publishing, Wallingford. Legros, D., Ollivier, G., Gastellu-Etchegorry, M., Paquet, C., Burri, C., Jannin, J. and Buscher, P. (2002). Treatment of human African trypanosomiasis: present situation and needs for research and development. Lancet Infect. Dis. 2, 437–440. Lehane, M. J. (1997). Peritrophic matrix structure and function. Ann. Rev. Entomol. 42, 525–550. Lehane, M. J. and Mail, T. S. (1985). Determining the age of adult male and female Glossina morsitans morsitans using a new technique. Ecol. Entomol. 10, 219–224. Lehane, M. J. and Msangi, A. R. (1991). Lectin and peritrophic membrane development in the gut of Glossina m. morsitans and a discussion of their role in protecting the fly against trypanosome infection. Med. Vet. Entomol. 5, 495–501. Lehane, M. J., Msangi, A. R., Whitaker, C. J. and Lehane, S. M. (2000). Grouping of trypanosome species in mixed infections in Glossina pallidipes. Parasitology 120, 583–592. Lehane, M. J., Aksoy, S., Gibson, W., Kerhornou, A., Berriman, M., Hamilton, J., Soares, M. B., Bonaldo, M. F., Lehane, S. and Hall, N. (2003). Adult midgut expressed sequence tags from the tsetse fly Glossina morsitans morsitans and expression analysis of putative immune response genes. Genome Biol. 4, R63. Lehane, M. J., Gibson, W. and Lehane, S. M. (2008). Differential expression of fat body genes in Glossina morsitans morsitans following infection with Trypanosoma brucei brucei. Int. J. Parasitol. 38, 93–101.

168


Lemaitre, B. and Hoffmann, J. (2007). The host defense of Drosophila melanogaster. Ann. Rev. Immunol. 25, 697–743. Liniger, M., Acosta-Serrano, A., Van Den Abbeele, J., Renggli, C. K., Brun, R., Englund, P. T. and Roditi, I. (2003). Cleavage of trypanosome surface glycoproteins by alkaline trypsin-like enzyme(s) in the midgut of Glossina morsitans. Int. J. Parasitol. 33, 1319–1328. Liniger, M., Urwyler, S., Studer, E., Oberle, M., Renggli, C. K. and Roditi, I. (2004). Role of the N-terminal domains of EP and GPEET procyclins in membrane targeting and the establishment of midgut infections by Trypanosoma brucei. Mol. Biochem. Parasitol. 137, 247–251. Luckhart, S., Vodovotz, Y., Cui, L. W. and Rosenberg, R. (1998). The mosquito Anopheles stephensi limits malaria parasite development with inducible synthesis of nitric oxide. Proc. Natl. Acad. Sci. USA 95, 5700–5705. Macleod, E. T., Darby, A. C., Maudlin, I. and Welburn, S. C. (2007a). Factors affecting trypanosome maturation in tsetse flies. PLoS ONE 2, e239. MacLeod, E. T., Maudlin, I., Darby, A. C. and Welburn, S. C. (2007b). Antioxidants promote establishment of trypanosome infections in tsetse. Parasitology 134, 827–831. Masaninga, F. and Mihok, S. (1999). Host influence on adaptation of Trypanosoma congolense metacyclics to vertebrate hosts. Med. Vet. Entomol. 13, 330–332. Matthew, C. Z., Darby, A. C., Young, S. A., Hume, L. H. and Welburn, S. C. (2005). The rapid isolation and growth dynamics of the tsetse symbiont Sodalis glossinidius. FEMS Microbiol. Lett. 248, 69–74. Matthews, K. R. and Gull, K. (1997). Commitment to differentiation and cell cycle re-entry are coincident but separable events in the transformation of African trypanosomes from their bloodstream to their insect form. J. Cell Sci. 110, 2609–2618. Maudlin, I. (1982). Inheritance of susceptibility to Trypanosoma congolense infection in Glossina morsitans. Ann. Trop. Med. Parasitol. 76, 225–227. Maudlin, I. and Dukes, P. (1985). Extrachromosomal inheritance of susceptibility to trypanosome infection in tsetse flies 1. Selection of susceptible and refractory lines of Glossina morsitans morsitans. Ann. Trop. Med. Parasitol. 79, 317–324. Maudlin, I. and Ellis, D. S. (1985). Association between intracellular rickettsial-like infections of midgut cells and susceptibility to trypanosome infection in Glossina spp. Z. Parasitenkd. 71, 683–687. Maudlin, I. and Welburn, S. C. (1987). Lectin mediated establishment of midgut infections of Trypanosoma congolense and Trypanosoma brucei in Glossina morsitans. Trop. Med. Parasitol. 38, 167–170. Maudlin, I., Welburn, S. C. and Mehlitz, D. (1990). The relationship between Rickettsialike organisms and trypanosome infections in natural populations of tsetse in Liberia. Trop. Med. Parasitol. 41, 265–267. McNamara, J. J., Mohammed, G. and Gibson, W. C. (1994). Trypanosoma (Nannomonas) godfreyi sp. nov. from tsetse flies in the Gambia- biological and biochemical characterisation. Parasitology 109, 497–509. Mihok, S., Stiles, J. K., Mpanga, E. and Olubayo, R. O. (1994). Relationships between protease activity, host blood and infection rates in Glossina morsitans spp. infected with Trypanosoma congolense, Trypanosoma brucei and T. simiae. Med. Vet. Entomol. 8, 47–50. Mihok, S., Machika, C., Darji, N., Kangethe, E. K. and Otieno, L. H. (1995). Relationships between host blood factors and proteases in Glossina morsitans subspecies infected with Trypanosoma congolense. Med. Vet. Entomol. 9, 155–160.

THE ENEMY WITHIN

169

Miller, N. and Lehane, M. J. (1990). In vitro perfusion studies on the peritrophic membrane of the tsetse fly Glossina morsitans morsitans (Diptera, Glossinidae). J. Insect Physiol. 36, 813–818. Miller, N. and Lehane, M. J. (1993). Ionic environment and the permeability properties of the peritrophic membrane of Glossina morsitans morsitans. J. Insect Physiol. 39, 139–144. Milne, K. G., Prescott, A. R. and Ferguson, M. A. J. (1998). Transformation of monomorphic Trypanosoma brucei bloodstream form trypomastigotes into procyclic forms at 37 C by removing glucose from the culture medium. Mol. Biochem. Parasitol. 94, 99–112. Miruk, A., Hagos, A., Yacob, H. T., Asnake, F. and Basu, A. K. (2008). Prevalence of bovine trypanosomosis and trypanocidal drug sensitivity studies on Trypanosoma congolense in Wolyta and Dawero zones of southern Ethiopia. Vet. Parasitol. 152, 141–147. Moloo, S. K. (1976). Aspects of nutrition of adult female Glossina morsitans during pregnancy. J. Insect Physiol. 22, 563–567. Moloo, S. K. (1993). A comparison of susceptibility of two allopatric populations of Glossina pallidipes for stocks of Trypanosoma congolense. Med. Vet. Entomol. 7, 369–372. Moloo, S. K. and Kutuza, S. B. (1970). Feeding and crop emptying in Glossina brevipalpis Newstead. Acta Trop. 27, 356–377. Moloo, S. K. and Kutuza, S. B. (1988a). Comparative study on the infection rates of different laboratory strains of Glossina species by Trypanosoma congolense. Med. Vet. Entomol. 2, 253–257. Moloo, S. K. and Kutuza, S. B. (1988b). Comparative study on the susceptibility of different Glossina species to Trypanosoma brucei brucei infection. Trop. Med. Parasitol. 39, 211–213. Moloo, S. K. and Shaw, M. K. (1989). Rickettsial infection of midgut cells are not associated with susceptibility of Glossina morsitans centralis to Trypanosoma congolense infection. Acta Trop. 46, 223–227. Moloo, S. K., Sabwa, C. L. and Kabata, J. M. (1992). Vector competence of Glossina pallidipes and Glossina morsitans centralis for Trypanosoma vivax, Trypanosoma congolense and T. b. brucei. Acta Trop. 51, 271–280. Molyneux, D. H. and Jenni, L. (1981). Mechanoreceptors, feeding behaviour and trypanosome transmission in Glossina. Trans. R. Soc. Trop. Med. Hyg. 75, 160–163. Morris, J. C., Wang, Z. F., Drew, M. E. and Englund, P. T. (2002). Glycolysis modulates trypanosome glycoprotein expression as revealed by an RNAi library. EMBO J. 21, 4429–4438. Msangi, A. R., Whitaker, C. J. and Lehane, M. J. (1998). Factors influencing the prevalence of trypanosome infection of Glossina pallidipes on the Ruvu flood plain of Eastern Tanzania. Acta Trop. 70, 143–155. Mshelbwala, A. S. (1972). Trypanosoma brucei infection in hemocoel of tsetse flies. Trans. R. Soc. Trop. Med. Hyg. 66, 637–643. Munks, R. J. L., Sant’Anna, M. R. V., Grail, W., Gibson, W., Igglesden, T., Yoshiyama, M., Lehane, S. M. and Lehane, M. J. (2005). Antioxidant gene expression in the blood-feeding fly Glossina morsitans morsitans. Insect Mol. Biol. 14, 483–491. Muranjan, M., Wang, Q., Li, Y. L., Hamilton, E., Otieno Omondi, F. P., Wang, J., van Praagh, A., Grootenhuis, J. G. and Black, S. J. (1997). The trypanocidal Cape buffalo serum protein is xanthine oxidase. Infect. Immun. 65, 3806–3814. Murphy, N. B. and Welburn, S. C. (1997). Programmed cell death in procyclic Trypanosoma brucei rhodesiense is associated with differential expression of mRNAs. Cell Death Diff. 4, 365–370.

170


Mwangelwa, M. I., Otieno, L. H. and Reid, G. D. F. (1987). Some barriers to Trypanosoma congolense development in Glossina morsitans morsitans. Insect Sci. Applic. 8, 33–37. Nayduch, D. and Aksoy, S. (2007). Refractoriness in tsetse flies (Diptera: Glossinidae) may be a matter of timing. J. Med. Entomol. 44, 660–665. Ndegwa, P. N., Irungu, L. W. and Moloo, S. K. (1992). Effect of puparia incubation temperature- increased infection rates of Trypanosoma congolense in Glossina morsitans centralis, G. fuscipes fuscipes and G. brevipalpis. Med. Vet. Entomol. 6, 127–130. Nguu, E. K., Osir, E. O., Imbuga, M. O. and Olembo, N. K. (1996). The effect of host blood in the in vitro transformation of bloodstream trypanosomes by tsetse midgut homogenates. Med. Vet. Entomol. 10, 317–322. Nigam, Y., Maudlin, I., Welburn, S. and Ratcliffe, N. A. (1997). Detection of phenoloxidase activity in the hemolymph of tsetse flies, refractory and susceptible to infection with Trypanosoma brucei rhodesiense. J. Invert. Path. 69, 279–281. Nogge, G. (1976). Sterility in tsetse flies (Glossina morsitans Westwood) caused by loss of symbionts. Experientia 32, 995–996. Nogge, G. (1981). Significance of symbionts for the maintenance of an optimal nutritional state for successful reproduction in hematophagous arthropods. Parasitology 82, 101–104. Novakova, E. and Hypsa, V. (2007). A new Sodalis lineage from bloodsucking fly Craterina melbae (Diptera, Hippoboscoidea) originated independently of the tsetse flies symbiont Sodalis glossinidius. FEMS Microbiol. Lett. 269, 131–135. Oli, M. W., Cotlin, L. F., Shiflett, A. M. and Hajduk, S. L. (2006). Serum resistanceassociated protein blocks lysosomal targeting of trypanosome lytic factor in Trypanosoma brucei. Eukaryot. Cell 5, 132–139. Olubayo, R. O., Mihok, S., Munyoki, E. and Otieno, L. H. (1994). Dynamics of host blood effects in Glossina morsitans spp. infected with Trypanosoma congolense and Trypanosoma brucei. Parasitol. Res. 80, 177–181. Otieno, L. H., Darji, N. and Onyango, P. (1976). Development of Trypanosoma (Trypanozoon) brucei in Glossina morsitans inoculated into the tsetse haemocoel. Acta Trop. 33, 143–150. Overath, P., Czichos, J. and Haas, C. (1986). The effect of citrate cis-aconitate on oxidative metabolism during transformation of Trypanosoma brucei. Eur. J. Biochem. 160, 175–182. Pais, R., Lohs, C., Wu, Y. N., Wang, J. W. and Aksoy, S. (2008). The obligate mutualist Wigglesworthia glossinidia influences reproduction, digestion, and immunity processes of its host, the tsetse fly. Appl. Environ. Microbiol. 74, 5965–5974. Pays, E. and Nolan, D. P. (1998). Expression and function of surface proteins in Trypanosoma brucei. Mol. Biochem. Parasitol. 91, 3–36. Peacock, L., Ferris, V., Bailey, M. and Gibson, W. (2006). Multiple effects of the lectininhibitory sugars D-glucosamine and N-acetyl-glucosamine on tsetse-trypanosome interactions. Parasitol. 132, 651–658. Peacock, L., Ferris, V., Bailey, M. and Gibson, W. (2007). Dynamics of infection and competition between two strains of Trypanosoma brucei brucei in the tsetse fly observed using fluorescent markers. Kinetoplastid Biol. Dis. 6, 4. Pearson, T. W., Beecroft, R. P., Welburn, S. C., Ruepp, S., Roditi, I., Hwa, K. Y., Englund, P. T., Wells, C. W. and Murphy, N. B. (2000). The major cell surface glycoprotein procyclin is a receptor for induction of a novel form of cell death in African trypanosomes in vitro. Mol. Biochem. Parasitol. 111, 333–349. Pinnock, D. E. and Hess, R. T. (1974). The occurrence of intracellular rickettsia-like organisms in the tsetse flies Glossina morsitans, Glossina fuscipes, Glossina brevipalpis and Glossina pallidipes. Acta Trop. 31, 70–79.

THE ENEMY WITHIN

171

Pontes, M. H., Babst, M., Lochhead, R., Oakeson, K., Smith, K. and Dale, C. (2008). Quorum sensing primes the oxidative stress response in the insect endosymbiont, Sodalis glossinidius. PLoS ONE 3, e3541. Randolph, S. E. and Rogers, D. J. (1981). Physiological correlates of the availability of Glossina morsitans centralis Machado to different sampling methods. Ecol. Entomol. 6, 63–77. Reduth, D., Grootenhuis, J. G., Olubayo, R. O., Muranjan, M., Otienoomondi, F. P., Morgan, G. A., Brun, R., Williams, D. J. L. and Black, S. J. (1994). African buffalo serum contains novel trypanocidal protein. J. Eukaryot. Microbiol. 41, 95–103. Reifenberg, J. M., Cuisance, D., Frezil, J. L., Cuny, G. and Duvallet, G. (1997). Comparison of the susceptibility of different Glossina species to simple and mixed infections with Trypanosoma (Nannomonas) congolense savannah and riverine forest types. Med. Vet. Entomol. 11, 246–252. Reinhardt, C., Steiger, R. and Hecker, H. (1972). Ultrastructural study of the midgut mycetome bacteroides of the tsetse flies Glossina morsitans, Glossina fuscipes and Glossina brevipalpis (Diptera: Brachycera). Acta Trop. 29, 280–288. Richner, D., Brun, R. and Jenni, L. (1988). Production of metacyclic forms by cyclical transmission of West-African Trypanosoma brucei isolates from man and animals. Acta Trop. 45, 309–319. Ridgley, E. L., Xiong, Z. H. and Ruben, L. (1999). Reactive oxygen species activate a Ca2þ-dependent cell death pathway in the unicellular organism Trypanosoma brucei brucei. Biochem. J. 340, 33–40. Rio, R. V. M., Wu, Y. N., Filardo, G. and Aksoy, S. (2006). Dynamics of multiple symbiont density regulation during host development: tsetse fly and its microbial flora. Proc. Soc. Biol. Sci. 273, 805–814. Robertson, M. (1913). Notes on the life-history of Trypanosoma gambiense, with a brief reference to the cycles of Trypanosoma nanum and Trypanosoma percorum in Glossina palpalis. Phil. Trans. R. Soc. Lond. B 203, 161–184. Roditi, I. and Clayton, C. (1999). An unambiguous nomenclature for the major surface glycoproteins of the procyclic form of Trypanosoma brucei. Mol. Biochem. Parasitol. 103, 99–100. Roditi, I. and Lehane, M. J. (2008). Interactions between trypanosomes and tsetse flies. Curr. Opin. Microbiol. 11, 345–351. Roditi, I. and Pearson, T. W. (1990). The procyclin coat of African trypanosomes (or the not-so-naked trypanosome) Parasitol. Today 6, 79–82. Roditi, I., Furger, A., Ruepp, S., Schurch, N. and Butikofer, P. (1998). Unravelling the procyclin coat of Trypanosoma brucei. Mol. Biochem. Parasitol. 91, 117–130. Rogers, D. J., Hendrickx, G. and Slingenbergh, J. H. W. (1994). Tsetse flies and their control. Rev. Sci. Tech. 13, 1075–1124. Roignant, J. Y., Carre, C., Mugat, R., Szymczak, D., Lepesant, J. A. and Antoniewski, C. (2003). Absence of transitive and systemic pathways allows cell-specific and isoform-specific RNAi in Drosophila. RNA 9, 299–308. Rolin, S., Hanocq-Quertier, J., Paturiaux-Hanocq, F., Nolan, D. P. and Pays, E. (1998). Mild acid stress as a differentiation trigger in Trypanosoma brucei. Mol. Biochem. Parasitol. 93, 251–262. Ruepp, S., Furger, A., Kurath, U., Renggli, C. K., Hemphill, A., Brun, R. and Roditi, I. (1997). Survival of Trypanosoma brucei in the tsetse fly is enhanced by the expression of specific forms of procyclin. J. Cell Biol. 137, 1369–1379. Ryu, J. H., Kim, S. H., Lee, H. Y., Bai, J. Y., Nam, Y. D., Bae, J. W., Lee, D. G., Shin, S. C., Ha, E. M. and Lee, W. J. (2008). Innate immune homeostasis by the homeobox gene Caudal and commensal-gut mutualism in Drosophila. Science 319, 777–782.

172


Sbicego, S., Vassella, E., Kurath, U., Blum, B. and Roditi, I. (1999). The use of transgenic Trypanosoma brucei to identify compounds inducing the differentiation of bloodstream forms to procyclic forms. Mol. Biochem. Parasitol. 104, 311–322. Schmid-Hempel, P. (2005). Evolutionary ecology of insect immune defenses. Ann. Rev. Entomol. 50, 529–551. Schweizer, J., Tait, A. and Jenni, L. (1988). The timing and frequency of hybrid formation in African trypanosomes during cyclical transmission. Parasitol. Res. 75, 98–101. Seed, R. J. and Wenck, M. A. (2003). Role of the long slender to the short stumpy transition in the life cycle of the African trypanosomes. Kinetoplastid Biol. Dis. 2, 3–10. Shahabuddin, M., Fields, I., Bulet, P., Hoffmann, J. A. and Miller, L. H. (1998). Plasmodium gallinaceum: differential killing of some mosquito stages of the parasite by insect defensin. Exp. Parasitol. 89, 103–112. Sharma, R., Peacock, L., Gluenz, E., Gull, K., Gibson, W. and Carrington, M. (2008). Asymmetric cell division as a route to reduction in cell length and change in cell morphology in trypanosomes. Protist 159, 137–151. Shaw, M. K. and Moloo, S. K. (1991). Comparative study on rickettsia-like organisms in the midgut epithelial cells of different Glossina species. Parasitology 102, 193–199. Sheader, K., Vruchte, D. T. and Rudenko, G. (2004). Bloodstream form-specific upregulation of silent VSG expression sites and procyclin in Trypanosoma brucei after inhibition of DNA synthesis or DNA damage. J. Biol. Chem. 279, 13363–13374. Siomi, H. and Siomi, M. C. (2009). On the road to reading the RNA-interference code. Nature 457, 396–404. Siozios, S., Sapountzis, P., Ioannidis, P. and Bourtzis, K. (2008). Wolbachia symbiosis and insect immune response. Insect Sci. 15, 89–100. Skerlavaj, B., Gennaro, R., Bagella, L., Merluzzi, L., Risso, A. and Zanetti, M. (1996). Biological characterization of two novel cathelicidin-derived peptides and identification of structural requirements for their antimicrobial and cell lytic activities. J. Biol. Chem. 271, 28375–28381. Soares, C. A. G., Lima, C. M. R., Dolan, M. C., Piesman, J., Beard, C. B. and Zeidner, N. S. (2005). Capillary feeding of specific dsRNA induces silencing of the isac gene in nymphal Ixodes scapularis ticks. Insect Mol. Biol. 14, 443–452. Stafford, J. L. and Belosevic, M. (2003). Transferrin and the innate immune response of fish: identification of a novel mechanism of macrophage activation. Dev. Comp. Immunol. 27, 539–554. Steverding, D. (2008). The history of African trypanosomiasis. Parasit. Vectors 1, 3. Stiles, J. K., Wallbanks, K. R. and Molyneux, D. H. (1991). The use of casein substrate gels for determining trypsin-like activity in the midgut of Glossina palpalis spp. (Diptera: Glossinidae). J. Insect Physiol. 37, 247–254. Stuhlmann, F. (1907). Beitra¨ge zur Kenntnis der Tsetsefliege (Glossina fusca und G. tachinoides). Arb. Kaiserl. Gesundheits. 26, 301–383. Tanji, T. and Ip, Y. T. (2005). Regulators of the Toll and Imd pathways in the Drosophila innate immune response. Trends Immunol. 26, 193–198. Tanji, T., Hu, X. D., Weber, A. N. R. and Ip, Y. T. (2007). Toll and Imd pathways synergistically activate an innate immune response in Drosophila melanogaster. Mol. Cell. Biol. 27, 4578–4588. Tavernarakis, N., Wang, S. L., Dorovkov, M., Ryazanov, A. and Driscoll, M. (2000). Heritable and inducible genetic interference by double-stranded RNA encoded by transgenes. Nat. Genet. 24, 180–183. Tellam, R. L., Wijffels, G. and Willadsen, P. (1999). Peritrophic matrix proteins. Insect Biochem. Mol. Biol. 29, 87–101.

THE ENEMY WITHIN

173

Tetley, L. and Vickerman, K. (1985). Differentiation in Trypanosoma brucei: hostparasite cell junctions and their persistance during acquisition of the variable antigen coat. J. Cell Sci. 74, 1–19. Toh, H., Weiss, B. L., Perkin, S. A. H., Yamashita, A., Oshima, K., Hattori, M. and Aksoy, S. (2006). Massive genome erosion and functional adaptations provide insights into the symbiotic lifestyle of Sodalis glossinidius in the tsetse host. Genome Res. 16, 149–156. Torr, S. J., Hargrove, J. W. and Vale, G. A. (2005). Towards a rational policy for dealing with tsetse. Trends Parasitol. 21, 537–541. Travanty, E. A., Adelman, Z. N., Franz, A. W. E., Keene, K. M., Beaty, B. J., Blair, C. D., James, A. A. and Olson, K. E. (2004). Using RNA interference to develop dengue virus resistance in genetically modified Aedes aegypti. Insect Biochem. Mol. Biol. 34, 607–613. Tzou, P., Ohresser, S., Ferrandon, D., Capovilla, M., Reichhart, J. M., Lemaitre, B., Hoffmann, J. A. and Imler, J. L. (2000). Tissue-specific inducible expression of antimicrobial peptide genes in Drosophila surface epithelia. Immunity 13, 737–748. Urwyler, S., Vassella, E., Van Den Abbeele, J., Renggli, C. K., Blundell, P., Barry, J. D. and Roditi, I. (2005). Expression of procyclin mRNAs during cyclical transmission of Trypanosoma brucei. PLoS Pathog. 1, 225–231. Utz, S., Roditi, I., Renggli, C. K., Almeida, I. C., Acosta-Serrano, A. and Butikofer, P. (2006). Trypanosoma congolense procyclins: unmasking cryptic major surface glycoproteins in procyclic forms. Eukaryot. Cell 5, 1430–1440. Vale, G. A. (1982). The improvement of traps for tsetse flies (Diptera: Glossinidae). Bull. Entomol. Res. 72, 95–106. Van den Abbeele, J., Claes, Y., van Bockstaele, D., Le Ray, D. and Coosemans, M. (1999). Trypanosoma brucei spp. development in the tsetse fly: characterization of the post-mesocyclic stages in the foregut and proboscis. Parasitology 118, 469–478. Van den Bossche, P., De Deken, R., Brandt, J., Geerts, S., Geysen, D. and Berkvens, D. (2004). The transmission of mixed Trypanosoma brucei brucei/T. congolense infections by tsetse (Glossina morsitans morsitans). Vet. Parasitol. 119, 147–153. Van Hoof, L., Henrard, C. and Peel, E. (1937). Influences modificatrices de la transmissibilite´ cyclique du Trypanosoma gambiense par Glossina palpalis. Ann. Soc. Belge Med. Trop. 17, 249–263. Van Roessel, P., Hayward, N. M., Barros, C. S. and Brand, A. H. (2002). Two-color GFP imaging demonstrates cell-autonomy of GAL4-driven RNA interference in Drosophila. Genesis 34, 170–173. Vassella, E., Reuner, B., Yutzy, B. and Boshart, M. (1997). Differentiation of African trypanosomes is controlled by a density sensing mechanism which signals cell cycle arrest via the cAMP pathway. J. Cell Sci. 110, 2661–2671. Vassella, E., Acosta-Serrano, A., Studer, E., Lee, S. H., Englund, P. T. and Roditi, I. (2001). Multiple procyclin isoforms are expressed differentially during the development of insect forms of Trypanosoma brucei. J. Mol. Biol. 312, 597–607. Vassella, E., Oberle, M., Urwyler, S., Renggli, C. K., Studer, E., Hemphill, A., Fragoso, C., Butikofer, P., Brun, R. and Roditi, I. (2009). Major surface glycoproteins of insect forms of Trypanosoma brucei are not essential for cyclical transmission by tsetse. PLoS ONE 4, e4493. Vickerman, K. (1969). On surface coat and flagellar adhesion in trypanosomes. J. Cell Sci. 5, 163–193. Vickerman, K. (1985). Developmental cycles and biology of pathogenic trypanosomes. Brit. Med. Bull. 41, 105–114.

174


Walshe, D. P., Lehane, S. M., Lehane, M. J. and Haines, L. R. (2009). Prolonged gene knockdown in the tsetse fly Glossina by feeding double stranded RNA. Insect Mol. Biol. 18, 11–19. Wang, J., Van Praagh, A., Hamilton, E., Wang, Q., Zou, B. X., Muranjan, M., Murphy, N. B. and Black, S. J. (2002). Serum xanthine oxidase: origin, regulation, and contribution to control of trypanosome parasitemia. Antioxid. Redox Sig. 4, 161–178. Wang, J., Hu, C., Wu, Y., Stuart, A., Amemiya, C., Berriman, M., Toyoda, A., Hattori, M. and Aksoy, S. (2008). Characterization of the antimicrobial peptide attacin loci from Glossina morsitans. Insect Mol. Biol. 17, 293–302. Watson, F. L., Puttmann-Holgado, R., Thomas, F., Lamar, D. L., Hughes, M., Kondo, M., Rebel, V. I. and Schmucker, D. (2005). Extensive diversity of Ig-superfamily proteins in the immune system of insects. Science 309, 1874–1878. Weiss, B. L., Mouchotte, R., Rio, R. V. M., Wu, Y. N., Wu, Z. Y., Heddi, A. and Aksoy, S. (2006). Interspecific transfer of bacterial endosymbionts between tsetse fly species: infection establishment and effect on host fitness. Appl. Environ. Microbiol. 72, 7013–7021. Weiss, B. L., Wu, Y. N., Schwank, J. J., Tolwinski, N. S. and Aksoy, S. (2008). An insect symbiosis is influenced by bacterium-specific polymorphisms in outer-membrane protein A. Proc. Natl. Acad. Sci. USA 105, 15088–15093. Welburn, S. C. and Maudlin, I. (1991). Rickettsia-like organisms, puparial temperature and susceptibility to trypanosome infection in Glossina morsitans. Parasitology 102, 201–206. Welburn, S. C. and Maudlin, I. (1992). The nature of the teneral state in Glossina and its role in the acquisition of trypanosome infection in tsetse. Ann. Trop. Med. Parasitol. 86, 529–536. Welburn, S. C. and Maudlin, I. (1997). Control of Trypanosoma brucei brucei infections in tsetse, Glossina morsitans. Med. Vet. Entomol. 11, 286–289. Welburn, S. C. and Maudlin, I. (1999). Tsetse-typanosome interactions: rites of passage. Parasitol. Today 15, 399–403. Welburn, S. C., Maudlin, I. and Ellis, D. S. (1987). In vitro cultivation of Rickettsia-like organisms from Glossina spp. Ann. Trop. Med. Parasitol. 81, 331–335. Welburn, S. C., Maudlin, I. and Ellis, D. S. (1989). Rate of trypanosome killing by lectins in midguts of different species and strains of Glossina. Med. Vet. Entomol. 3, 77–82. Welburn, S. C., Maudlin, I. and Molyneux, D. H. (1994). Midgut lectin activity and sugar specificity in teneral and fed tsetse. Med. Vet. Entomol. 8, 81–87. Welburn, S. C., Maudlin, I. and Milligan, P. J. M. (1995). Trypanozoon: infectivity to humans is linked to reduced transmissibility in tsetse. 1. Comparison of human serum-resistant and human serum-sensitive field isolates. Exp. Parasitol. 81, 404–408. Welburn, S. C., Dale, C., Ellis, D., Beecroft, R. and Pearson, T. W. (1996). Apoptosis in procyclic Trypanosoma brucei rhodesiense in vitro. Cell Death Diff. 3, 229–236. Werren, J. H. and Windsor, D. M. (2000). Wolbachia infection frequencies in insects: evidence of a global equilibrium? Proc. Soc. Biol. Sci. 267, 1277–1285. Whitten, M. M. A., Mello, C. B., Gomes, S. A. O., Nigam, Y., Azambuja, P., Garcia, E. S. and Ratcliffe, N. A. (2001). Role of superoxide and reactive nitrogen intermediates in Rhodnius prolixus (Reduviidae)/Trypanosoma rangeli interactions. Exp. Parasitol. 98, 44–57. Whitten, M. M. A., Sun, F., Tew, I. F., Schaub, G., Soukou, C., Nappi, A. and Ratcliffe, N. A. (2007). Differential modulation of Rhodnius prolixus nitric oxide activities following challenge with Trypanosoma rangeli, T. cruzi and bacterial cell wall components. Insect Biochem. Mol. Biol. 37, 440–452.

THE ENEMY WITHIN

175

WHO (2000). WHO Report on Global Surveillance of Epidemic-prone Infectious Diseases. WHO, Geneva. WHO (2007). Report of a WHO Informal Consultation on Sustainable Control of Human African Trypanosomiasis. WHO, Geneva. Wigglesworth, V. B. (1929). Digestion in the tsetse fly: a study of structure and function. Parasitology 21, 288–321. Wijers, D. J. (1958). Factors that may influence the infection rate of Glossina palpalis with Trypanosoma gambiense. 1. The age of the fly at the time of the infective feed. Ann. Trop. Med. Parasitol. 52, 385–390. Winston, W. M., Molodowitch, C. and Hunter, C. P. (2002). Systemic RNAi in C. elegans requires the putative transmembrane protein SID-1. Science 295, 2456–2459. Woolhouse, M. E. J., Hargrove, J. W. and McNamara, J. J. (1993). Epidemiology of trypanosome infections of the tsetse fly Glossina pallidipes in the Zambezi valley. Parasitology 106, 479–485. Xin, R. G., Liu, S., Guo, Z. Y., Yu, H. H., Li, C. P., Ji, X., Feng, J. H. and Li, P. C. (2006). The antioxidant activity of glucosamine hydrochloride in vitro. Biol. Med. Chem. 14, 1706–1709. Yan, J., Cheng, Q., Li, C. B. and Aksoy, S. (2002). Molecular characterization of three gut genes from Glossina morsitans morsitans: cathepsin B, zinc-metalloprotease and zinc-carboxypeptidase. Insect Mol. Biol. 11, 57–65. Ye, Y. H., Chenoweth, S. F. and McGraw, E. A. (2009). Effective but costly, evolved mechanisms for defense against a virulent opportunistic pathogen in Drosophila melanogaster. PLoS Pathog. 5, e1000385. Zamore, P. D. (2000). RNAi: an in vitro approach to understanding mechanism. In ‘‘Regulatory Mechanisms Involving RNA Session. Biological Regulatory Mechanisms Conference’’ Gordon Research Conferences, Holderness’’, NH, July 30–August 4.

Interactions of Trypanosomatids and Triatomines Gu¨nter A. Schaub Zoology/Parasitology Group, Ruhr-Universita¨t Bochum, 44780 Bochum, Germany

1 Introduction 177 2 Triatomines 178 2.1 Distribution 178 2.2 Development 179 2.3 Intestinal tract, digestion and excretion 182 2.4 The intestinal microenvironment 183 3 The trypanosomatids 190 3.1 Distribution of species and strains 190 3.2 Developmental cycle in the triatomines 195 4 Effects of the host on trypanosomatids 198 4.1 Susceptibility and refractoriness 198 4.2 Effects of pH, osmolality and ionic composition 200 4.3 Effects of the border face 200 4.4 Effects of microorganisms and antimicrobial compounds 203 4.5 Effects of digestion, digestion products and excretion 204 4.6 Effects of other soluble factors 209 5 Effects of trypanosomatids on triatomines 210 5.1 Classification of pathogenicity and action of secondary stressors 210 5.2 Pathogenicity of Blastocrithidia triatomae and Trypanosoma rangeli 211 5.3 Subpathogenicity of trypanosomatids in triatomines 217 6 Interactions in double infections 219 7 Conclusions 220 Acknowledgements 220 References 220

1

Introduction

Investigations on the interactions of trypanosomatids and triatomines mainly consider Trypanosoma cruzi because this protozoon is the etiologic agent of Chagas disease, one of the ‘‘big six’’ which were selected in 1975/1976 by the World Health Organization for the ‘‘Special Programme for Research and Training in Tropical Diseases’’ (Schaub and Wu¨lker, 1984). Chagas disease is the only ADVANCES IN INSECT PHYSIOLOGY VOL. 37 ISBN 978-0-12-374829-4 DOI: 10.1016/S0065-2806(09)37004-6


¨ NTER A. SCHAUB GU

178

important tropical disease in which the parasite was first detected in the vector, namely triatomines (Chagas, 1909, 1922). The protozoon is mainly transmitted by these haematophagous bugs, but also through blood transfusions, organ implantations, parasite-contaminated food, etc. (Schmunis, 2004; Coura, 2007). Initially, Try. cruzi circulated exclusively between wild mammals and was transmitted by sylvatic triatomines (WHO, 2002). When man entered the natural foci and altered the ecosystem equilibrium, the domestic triatomines cycle became established (da Silva, 1986). The pathology of the disease is very variable, ranging from a sudden death within several weeks to nearly no symptoms for more than 70 years (Lana et al., 1996; Coura, 2007). Since only two compounds are available for therapy, which often results in severe side effects, the main target for control is the vector. Although house improvements can limit the colonization of the houses by triatomines (Schofield, 1994), mainly intense insecticide campaigns against the domestic populations of the most important vector, Triatoma infestans, has reduced the prevalence, according to estimates, from about 20 million chronically infected people in 1982 to about 12 million in 2007 (WHO, 1982; Dias, 2007). However, since Chagas disease is an anthropozoonosis circulating also in many wild mammals as reservoir hosts, an eradication is impossible. In addition to Try. cruzi, three other species of trypanosomatids develop in the triatomines. Besides the insect flagellate Blastocrithidia triatomae, these are species of the genus Trypanosoma: the rat trypanosomatid Try. conorhini and Try. rangeli. The latter is also regularly found in humans, but often in double infections with Try. cruzi. In single infections it is non-pathogenic, developing only a low parasitemia and persisting at detectable levels in the blood only for about 1–3 weeks (D’Alessandro, 1976; Guhl et al., 2002). The chapter does not cover bat trypanosomes which also develop in triatomines, because for these infections mainly epidemiological and systematic aspects are considered. Previous reviews focused only on interactions of Try. cruzi or Try. rangeli on triatomines (Zeledoń et al., 1977; Garcia and Azambuja, 1991; Gonzalez et al., 1998; Azambuja et al., 2005a,b; Garcia et al., 2007). Above and beyond that, a comparison of the different trypanosomatids provides an understanding of the individual adaptations and general aspects and either of factors acting on the flagellate inside the vector or of protozoon-derived factors acting on the triatomine.

2 2.1

Triatomines DISTRIBUTION

According to recent monographs, the current taxonomy of Triatominae recognizes 140 species (Galvaõ et al., 2003; Schofield and Galvaõ, 2009). However, this number will change in the near future, because molecular biological methods will identify new species and subspecies (Mas-Coma and Bargues, 2009).

INTERACTIONS OF TRYPANOSOMATIDS AND TRIATOMINES

179

Of the six tribes and 18 genera only species of the genus Linshcosteus are confined to the Indian subcontinent. Eight species forming the rubrofasciata group are associated with rats. Originating from the New World, Triatoma rubrofasciata has been distributed by sailing ships to many ports in tropical and subtropical regions especially in Southeast Asia (Haridass and Ananthakrishnan, 1980; Gorla et al., 1997). The other triatomines occur only on the American continent from latitude 42 N to 46 S, that is, between the Great Lakes of North America and Argentina (Lent and Wygodzinsky, 1979; Schofield, 1994; Gorla et al., 1997). Triatomines colonize all terrestrial habitats, but prefer to stay near the host. This is easy in caves of bats, burrows of rodents or preferred resting places of other animals, but more complicated in nests of birds which only breed once a year. Nests are often colonized by species of the tribe Rhodniini which prefer palm trees, whereas species of the genus Triatoma often prefer rocky habitats and rodent burrows and species of the genus Panstrongylus tree cavities and burrows (Gaunt and Miles, 2000; Schofield and Galvaõ, 2009). Since the dwellings of the indigenous humans of Latin America were caves or were made of material from the forest, the transition from animals to humans was not a big step. The construction of houses as a wooden frame covered with mud or adobe still offered a good habitat for triatomines since the cracks in the mud or adobe are optimal hiding places during the day for the night active bugs (Figs. 1A, B and 2A, B). In addition, the use of palm leaves for the roof provides a direct access to the house if eggs have been glued to the leaves by Rhodnius sp. According to their proclivity to approach the houses, the species of triatomines can be classified into sylvatic species, those of the peridomestic regions, mainly using farm animals near the houses as hosts, and domestic species, sucking blood of humans and animals inside the house. The most important domestic species is Tri. infestans, presumably originating from sylvatic populations in the Andean valleys of Bolivia (Noireau et al., 2005). This species has successfully displaced many other species of Triatominae (Pereira et al., 2006) and is nowadays threatened back in its distribution (see Section 1). Also, the domestic species Rhodnius prolixus is a very important vector. This species was also found on palm trees, an argument against the suggestion that it is restricted to the house and evolved from the almost morphologically indistinguishable but strictly sylvatic R. robustus (Feliciangeli et al., 2007). 2.2

DEVELOPMENT

All postembryonic stages of these hemimetabolous insects are obligate blood suckers, developing from eggs through five instars to the adults (Schaub, 2008). In each of the instars, the bugs require one full feeding engorgement – of about 6–12 times their own body weight – or several smaller ones. (Most scientists in Central Europe call the first four instars of Hemiptera ‘‘larvae’’ and the fifth a

180

FIG. 1 cover.


(A) Rural wooden framed mud covered house in Brazil. (B) Cracks in the mud

nymph while others use the term ‘‘nymph’’ for all five pre-adult instars.) The full engorgement activates via distension receptors the secretion of hormones – finally of ecdysone – and thereby the development of the new cuticle and the moult to the next instar or to the adult (Wigglesworth, 1940; Anwyl, 1972;


181

FIG. 2 (A) Rural adobe brick house in Bolivia. (B) Gaps between the adobe bricks.

Chiang and Davey, 1988). Without a full engorgement, more blood meals are necessary to obtain a specific level of reserve compounds for the new cuticle, and even with a full engorgement pathogenic flagellates may reduce the reserve level, resulting in delays of moulting (Schaub, 1988a). Therefore, the duration of the different stages is determined by the availability of blood, parasites and


182

characteristics of the respective species, but also by temperature and relative humidity (rH). The source of blood also affects the development and/or egg production, blood of mice being superior to blood of pigeons and chickens (summarized by Emmanuelle-Machado et al., 2002). In addition, development depends on symbiotic bacteria which colonize the anterior regions of the midgut and deliver essential compounds (Eichler, 1998; Eichler and Schaub, 2002) (see Section 2.4.3). The availability of hosts is a limiting factor for populations of triatomines (Ceballos et al., 2005). In wild populations many insects are starved, and only about 14% of the offspring are estimated to reach the adult stage (Schofield, 1980a,b). If no host is available, bugs can starve for long periods of time, increasingly from instar to instar up to the fourth. In Tri. infestans, at 90% rH this instar can survive 250 70 days after last feeding in the third instar, up to 14 months (Schaub and Lo¨sch, 1989). Fifth instars possess a lower, in other species a stronger starvation resistance than fourth instars, adults always a lower one (e.g. Zeledoń et al., 1970; Feliciangeli et al., 1980; Schaub and Lo¨sch, 1989; Corte´z and Gonçalves, 1998; Cabello, 2001; Almeida et al., 2003). Some days after the moult, copulation occurs, males depositing the spermatophore with the sperms in the vagina (Wigglesworth, 1972). Females lay 100–600 eggs, for most species loosely on the ground, but species inhabiting trees glue them to the leaves or feathers of birds (Schaub, 2008). Adults live 3–12 months, depending on the species, abiotic factors and availability of blood. 2.3

INTESTINAL TRACT, DIGESTION AND EXCRETION

The intestinal tract of triatomines is a relatively simple tube without diverticula (Kollien and Schaub, 2000). Of the foregut, the cibarium and pharynx are located in the head and the oesophagus with the transverse rings of the wall in the thorax (Wenk and Renz, 2003). The ducts of the salivary glands – two pairs in species of the genus Rhodnius and three pairs in those of the genera Triatoma, Dipetalogaster and Panstrongylus – end in the foregut (Barth, 1954; Lacombe, 1999). The midgut begins with the cardia, followed by the strongly distensible stomach and then the small intestine. (Other groups use the terms ‘‘crop’’ or ‘‘anterior midgut’’ for the stomach and ‘‘posterior midgut’’ for the small intestine.) The latter is subdivided by a central narrow region into the anterior, middle and posterior small intestine. The beginning of the hindgut is a very short pylorus/ileum region in which the Malpighian tubules end, followed by the big rectal sac. After feeding, several layers of membranes are shed from the microvilli, building the perimicrovillar membranes (also named extracellular membrane layers) (Billingsley, 1990; Terra, 1990). The functions of the different gut regions differ: In the foregut are located the pumps giving the saliva into the wound and blood vessels and pumping the blood into the oesophagus (Wenk et al., 2009). The short cardia seems to act as a sphincter and possesses deep enfoldings, separating the symbionts from the


183

blood flow (Kollien and Schaub, 2000). In the stomach, the blood is stored essentially undigested and concentrated, followed by a lysis of erythrocytes (Azambuja et al., 1983). Then the haemoglobin of some hosts, for example, guinea pigs, crystallizes in the stomach (Bauer, 1981). The stomach wall is not only a simple transport epithelium, pumping ions into the haemolymph and thereby passing the water to it and concentrating the blood, but also serves for the intracellular storage of lipids (Kollien and Schaub, 2000). In addition, after blood ingestion the number of symbionts increases there enormously (see Section 2.4.3). Small portions of the blood meal are passed into the small intestine. There the blood is digested, the symbionts are killed, and the nutrients are absorbed (Bauer, 1981). The remains of the blood meal are stored in the rectum for further absorption processes before being defecated. The ingestion of blood initiates a rapid release of diuretic hormones from the storage in the prothoracic and mesothoracic ganglionic masses (Maddrell, 1966; Maddrell et al., 1991). These hormones regulate the transport of fluid across the gut wall and induce a 1000-fold increase of the diuresis rate by the four Malpighian tubules, in R. prolixus and Tri. infestans to a rate as high as 3.3 ml/min/cm2 tubule (2.9 nl/min/mm) (Maddrell, 1969; Maddrell and Gardiner, 1980; Schnitker et al., 1988). Thereby, the conditions in the rectum change rapidly. The remnants of digestion are often eliminated at the end of blood ingestion, followed by clear colourless urine. Within 24 h after blood ingestion, the bugs excrete about 50–60% of the weight of the total ingested blood (76% of the imbibed fluid) (Wigglesworth, 1931). Then the yellow uric acid granules are eliminated for about 10 days, followed by a mixture with the dark remains of digestion (Wigglesworth, 1931; Hase, 1932). After the moult, increasingly less remains are defecated. 2.4 2.4.1

THE INTESTINAL MICROENVIRONMENT

The intestinal border face

According to the ectodermal origin, foregut and hindgut are lined by a cuticle, and the cells of the midgut are bordered by the microvilli. The microvilli of the epithelium of starved insects are covered by a single perimicrovillar membrane which develops to thick staples after feeding (Billingsley and Downe, 1983, 1986). This is very intensive after feeding of starved adults of R. prolixus, whereas in larvae of Tri. infestans starvation does not induce such a strong breakdown of the membranes and not such a synchronous development of the membranes after feeding (Jensen et al., 1990). The full development of the perimicrovillar membranes depends on the distension of the abdomen (see Section 2.2), blood components (especially haemoglobins) and the release of ecdysone (Albuquerque-Cunha et al., 2004). In detailed investigations especially the group of P. Azambuja and E.S. Garcia elucidated these mechanisms. The development of the membranes can be inhibited or reduced by decapitation,


184

gamma irradiation, cutting of the nerve cord transmitting the signals of the distension receptors, anti-ecdysone compounds (e.g. azadirachtin) or a saline diet without proteins (e.g. Cortez et al., 2002; Gomes et al., 2002; AlbuquerqueCunha et al., 2004). The membranes show a strong reaction of non-specific carbohydrate staining (Bauer, 1981). The hindgut is bordered by an extracellular cuticle. This cuticle has the same basic architecture as the external integument of insects (Schmidt et al., 1998). The cell membrane is covered by the procuticle. Only this region reacts with wheat-germ lectin, indicating the presence of N-acetyl-D-glucosamine of the chitin. The next layer is the epicuticle, in electron microscopy subdivided into the electron-dense inner epicuticle and the outer epicuticle built by a cuticulin layer. The latter is covered by the superficial layer, which reacts with the lipidspecific Nile red stain (Schmidt et al., 1998). At the entrance into the rectal sac, the four rectal pads – also called rectal glands – possess a much thinner cuticle than the other regions of the rectum (Wigglesworth, 1972). In other insects, the rectal glands absorb water, ions and amino acids (Wigglesworth, 1984). 2.4.2

pH, osmolality and ions

Directly after blood ingestion, the blood determines the conditions in the stomach, that is, pH 7.4 (Billker et al., 2000). According to pH determinations via pH indicators, the pH in the stomach of larvae of R. prolixus increases after feeding to pH 7.4 and within 1 week decreases slowly to pH 6.5, in the small intestine from pH 7.2 to 5.5 (J. M. C. Ribeiro and E. S. Garcia, personal communication). According to a more detailed investigation using microelectrodes, pH 6.2 in the stomach of starved fifth instars of Tri. infestans changes to pH 7.3 after blood ingestion, decreases slowly within the following 10 days to pH 5.2, and then increases to pH 5.6 at 20 days after feeding (M. Oldenburg and G. A. Schaub, unpublished data). In the small intestine, pH 6.8 in starved insects directly decreases and increases then to similar levels as in the stomach, reaching pH 6.1 at 20 days after feeding. The conditions in the rectum show a much stronger variation after feeding. Whereas the first deposited drop, mainly remains of digested blood, is acidic, pH 5.9, the pH increases to over pH 8.0 in the following four drops. For 24 h the pH of the urine is alkaline, reaching pH 8.9. Then the pH drops to pH 6.5, remaining at this level at least up to 96 h after feeding (Kollien et al., 2001). The osmolality changes from 320 mosmol/kg H2O in the first drop to 370 in the second and 410 mosmol/kg H2O in the fourth. At 1 and 3 days after feeding, the rectal contents possess an osmolality of 370 and 760 mosmol/kg H2O, respectively (Kollien et al., 2001). In determinations of the concentrations of anions and cations, sulphate and phosphate dominate in the rectal content of unfed bugs, and sodium and chloride in the clear urine; strong individual variations of the individual ions begin at 1 day after feeding (Kollien et al., 2001).


2.4.3

185

The microorganisms

Triatomines possess symbionts, mainly actinomycetes, which colonize the anterior regions of the midgut and deliver essential compounds (Eichler, 1998; Eichler and Schaub, 2002; Durvasula et al., 2008). Symbionts are transmitted within a population of bugs via coprophagy (Schaub, 1988c; Schaub et al., 1989a). Therefore, ammonia and pheromones deposited in the faeces and urine after blood ingestion and attracting other hungry bugs (Taneja and Guerin, 1997; Guerenstein and Lazzari, 2009) indicate not only the presence of a host but also of symbionts. Since dry faeces needs to be redissolved by fresh faeces (Schaub et al., 1989a), pheromones in dry faeces only mark the location of refuges (Lorenzo and Lazzari, 1996; Vitta et al., 2007), but should not attract for the transmission of symbionts. During coprophagy, also air-borne microorganisms get access to the intestinal tract. A minor risk is the contact of the mouthparts with the skin before penetration, but like all terrestrial insects, triatomines swallow air before moulting (Kollien et al., 2003). In addition, an uptake of tap water (Wiesinger, 1956) can be a source of bacteria. After the first description of a Gram-positive bacillus in the intestine of R. prolixus (Duncan, 1926), many different microorganisms were isolated from this and other species of triatomines, not only bacteria but also fungi (Figueiredo et al., 1990; Eichler et al., 1996; Vallejo et al., 2009). [The intestinal bacteria are classified as symbionts (i.e. they support the growth of the host), commensals (i.e. they develop in the intestine without positive or negative effects on the host) or as parasites if they affect the insect.] Upon incubating the intestinal contents from Tri. infestans from the field on agar plates, 16 morphologically different bacteria were isolated. These were mainly bacteria colonizing the soil or the skin (N. Reintjes and G. A. Schaub, unpublished data). In Tri. sordida from the field, a Dictyostelium-like slime mould occurred, and this bug contained no other bacteria (Eichler, 1998). Focusing on actinomycetes, seven morphologically different isolates were separated from Tri. sordida (Eichler et al., 1996). The stomach of R. prolixus contained a bacterium so far unknown from triatomines, Serratia marcescens (Azambuja et al., 2004). This Gram-negative bacterium possesses haemolytic and cytotoxic factors – at least one of them being a metalloprotease (Hertle et al., 1999; Marty et al., 2002) – and is pathogenic for immune-suppressed humans and some insects (summarized by Flyg and Xanthopoulos, 1983; Hejazi and Falkiner, 1997), but has no obvious effects on the triatomine. Also the majority of the other intestinal bacteria of triatomines can be classified as commensals of triatomines. The identification of the symbionts is mainly based on simple microbiological criteria, Gram-staining, development and morphology, with the identification of Rhodococcus rhodnii as a symbiont of many species of triatomines (summarized by Vallejo et al., 2009). However, the slow growth and the colour change of old cultures are not sufficient to differentially distinguish species of

186


actinomycetes. After sequencing the 16S rDNA and determining the symbiotic function by infections of aposymbiotic bugs, in Tri. infestans from Bolivia only a Nocardia sp. and in Panstrongylus megistus a Rhodococcus equi-like species fulfilled the criteria of a symbiont, that is, they establish and multiply in the gut, some survive the passage through the small intestine, the larvae develop without retardations or increased mortality rates, and the reproductive rates of the adults and the emergence rate of the offspring are in the normal range (Eichler et al., 1996; Vallejo et al., 2009; G. A. Schaub, unpublished data). In Tri. infestans from Argentina, a corynebacterial species fulfilled these criteria (Durvasula et al., 2008). After infection of first instars by feeding a mixture of blood and symbionts – R. rhodnii for R. prolixus and Nocardia sp. for Tri. infestans – followed by an axenic maintenance including in vitro feeding with defibrinated sheep blood, the number of colony-forming units of symbionts differs in the different regions of the intestinal tract of fifth instars and is affected by blood ingestion and digestion (Eichler and Schaub, 2002). After blood ingestion, the concentration of the respective symbiotic bacteria decreases due to the dilution of the populations by the blood. Within 10 days, the total population/bug increases 15- or 18-fold to about 0.8 109 colony-forming units in R. prolixus and 1.8 109 colony-forming units in Tri. infestans. About 95–99% of the total population of both symbionts develop in cardia and stomach. Directly after the passage from the blood-storing stomach into the digesting small intestine about 99% of the symbiont populations are killed, and only about 0.01% of the total population is present in the rectum. A lysis of bacteria in the small intestine is also visible using electron microscopy (Bauer, 1981). After blood ingestion the urine washes out the rectal population, which is thereby mainly present in the remains of blood digestion, that is, in the first drops of defecation, not in the urine (Eichler and Schaub, 2002). A first suggestion of a symbiotic function was published by Dias (1934, 1937). Experiments of Wigglesworth (1936) indicated that the symbionts support the triatomines with vitamin B. In larvae, aposymbiosis results in a retarded development and increased mortality rate and disturbances of cuticle melanization, digestion and excretion; if adults develop, their fecundity is strongly reduced (summarized by Vallejo et al., 2009). Supplying aposymbiotic larvae which show the aposymbiosis syndrome with a mixture of B vitamins or symbionts reduces the effects in the following instar. The hypothesis that the symbionts supply compounds of the vitamin B complex requires new investigations, since variable results were obtained using auxotrophic bacteria in determinations of the vitamin production by cultures of the symbionts (Geigy et al., 1953; Harington, 1960). In addition, after supplying aposymbiotic bugs with auxotrophic symbiont mutants no aposymbiosis effects occurred (Hill et al., 1976). Therefore, not supplying with B vitamins, but with sugar components might be the function of symbionts (Ribeiro and Pereira, 1984).


2.4.4

187

Antimicrobial factors

Like vertebrates, insects possess an innate immunity of cellular and humoral responses (Dillon and Dillon, 2004; Ratcliffe and Whitten, 2004; Mu¨ller et al., 2008). In triatomines the investigations focused very early on responses in the haemocoel after injection of bacteria (Azambuja and Garcia, 1987; Azambuja et al., 1989b) and rarely on the intestinal homeostasis (Ribeiro and Pereira, 1984; Kollien et al., 2004). Since triatomines require symbionts for their development, but are killed after ingestion of blood contaminated with airborne bacteria (G. A. Schaub, unpublished data), their intestinal immune system is even more important than that of the haemolymph and must distinguish between the symbionts and the pathogenic bacteria. In addition, the symbionts must resist the antibacterial activities. Antibacterial activities in the gut of triatomines were recognized very early by Duncan (1926). Later this activity was attributed for other insects to lysozyme, a widespread antibacterial compound (Mohrig and Messner, 1968). In R. prolixus, the antibacterial activity varies according to the region of the gut and the period of time after feeding. In the stomach, the activity increases after feeding and is much higher than in the small intestine (Ribeiro and Pereira, 1984). Although lysozyme has not been isolated and identified from the intestinal contents, the production in the intestinal wall is proved by the characterization of the cDNA and the determination of the expression levels in the different regions of the gut (Kollien et al., 2003; Arau´jo et al., 2006; Balczun et al., 2008; Ursic-Bedoya et al., 2008). In the cardia and stomach, the concentration of mRNA-encoding lysozymes increases after feeding, reaching a maximum before moulting, and is much higher than in the small intestine (Kollien et al., 2003). The difference is also visible in whole mount in situ hybridizations using Tri. brasiliensis and sense and antisense digoxigenin-labelled RNA (Arau´jo et al., 2006). Whereas all these lysozymes belong to the chicken-type lysozymes (c-type), invertebrate-type lysozymes (i-type) – originally found in a starfish – also occur in insects (Paskewitz et al., 2008) and have to be considered for triatomines. In addition to lysozyme, defensin appears to be present in the intestine. After injection of bacteria into the haemocoel, a rapid induction of transcription was found in the haemocoel and a delayed induction in the gut (Lopez et al., 2003). Without such an immunization, the mRNA encoding defensins is distributed uniformly throughout the cardia and stomach and to a much lower extent in the small intestine of Tri. brasiliensis (Arau´jo et al., 2006). Using quantitative realtime PCR, the expression level of the defensin gene in the stomach is found to be about 10-fold higher than that of the lysozyme gene. Comparing the different regions of the intestine, it is 500–2500 times higher in the stomach than in the cardia and very low in the small intestine. The time course of the expression is similar to that of the lysozyme gene, that is, a strong increase after feeding and a decrease between 5 and 10 days after feeding (Arau´jo et al., 2006).


188

Most recently, two additional defensin-encoding genes of Tri. brasiliensis were characterized (Waniek et al., 2009). The mRNA of defensin3 is present in the small intestine, but not in the stomach, that of defensin4 shows higher concentrations in the stomach than in the small intestine. Other antimicrobial factors of 7 and 25 kDa occur as humoral response after injection of bacteria into the haemocoel of R. prolixus (Azambuja et al., 1986, 1989b), but have not been identified nor found in the gut. According to a suppressive subtractive hybridization approach for the identification of the upregulation of immune-related genes after intracoelomic inoculation of Try. cruzi, also genes for transferrins are upregulated, which seem to be components of the humoral immune system and sequester iron away from the bacteria (Ursic-Bedoya and Lowenberger, 2007). In addition to these proteins/peptides, a low molecular weight compound, nitric oxide, is produced in the intestinal wall and may regulate the populations of microorganisms in the intestinal tract of triatomines (Whitten et al., 2001, 2007). The concentration of mRNA encoding for the nitric oxide synthase increases in the stomach within the first 24 h after feeding and then decreases. This is also evident for the small intestine in which the expression level is higher than in the stomach up to 12 h after feeding (Whitten et al., 2007). Also host-derived factors, the complement system and antibodies against components of the saliva and the Trypanosoma species have to be considered since these are ingested together with the blood of the host. However, salivary compounds of triatomines inhibit the classical pathway of complement activation (Cavalcante et al., 2003). Antibodies against the components of the saliva are produced rapidly by birds and mammals (summarized by Schwarz et al., 2009a,b). A component of the immune system of insects which is not present in the lumen of the intestine but in the cells of the intestinal wall and the haemocytes is the phenoloxidase (summarized by Ratcliffe and Whitten, 2004; Mu¨ller et al., 2008). This enzyme catalyses the oxidation of phenolic amino acids, resulting in melanin, which is used to encapsulate big parasites in the haemolymph or nodules of haemocytes that have ingested high numbers of microorganisms. Immune responses using this enzyme are regulated by juvenile hormones (Nakamura et al., 2007).

2.4.5

The enzymes and digestion products

Different enzymes originate from the intestinal tract. The saliva contains many different compounds, also serine proteases for the activation of some of these compounds (Assumpçaõ et al., 2008). Although the major digestion occurs in the small intestine, during storage in the stomach some enzymes act on the blood. A sialidase, and alkaline and acid phosphatases are found in the stomach contents, and b-acetylglucosaminidase, a-galactosidase, a-glucosidase, a-mannosidase and


189

b-mannosidase are involved in the metabolism of carbohydrates (Terra et al., 1988; Amino et al., 1995). In the small intestine, alkaline and acid phosphatases and a-glucosidase and a-mannosidase are also active, but with a higher activity than in the stomach (Terra et al., 1988). Digestion of haemoglobin starts immediately, as indicated by the colour change from red to brown. Whereas most insect groups use serine proteases – trypsins and chymotrypsins – as their principal proteases, Hemiptera depend on cysteine and/or aspartic proteases. This correlates to the pH of the insect midgut lumen, which is neutral or alkaline in the majority of insects, but acidic in Hemiptera (Lehane, 1994). The activity of cysteine proteases was attributed to cathepsin B (Terra et al., 1988; Terra, 1990), until the characterization of a cathepsin L cDNA from the intestine of R. prolixus and then from Tri. infestans (Lopez-Ordon˜ez et al., 2001; Kollien et al., 2004). Carboxypeptidases and aminopeptidases continue the digestion of the blood (Garcia and Guimara˜es, 1979; Garcia, 1987; Terra et al., 1988). Whereas cathepsin B is active in the lumen of the gut, aminopeptidases and acid phosphatases are localized on the perimicrovillar membranes (Bauer, 1981; Billingsley and Downe, 1985, 1988). Since one of the components of haemoglobin, free haem, can cause oxidative damage to the tissues, triatomines sequester it into haemozoin (Oliveira et al., 2000). Not only proteolysis, but also the digestion of lipids is restricted to the small intestine (Rimoldi et al., 1985; Canavoso et al., 2004; Grillo et al., 2007). In the lumen, the triacylglycerols are hydrolysed to free fatty acids, which are absorbed by the epithelium, used for the synthesis of lipids and stored in the fat body and the wall of the gut. Detailed investigations of the digestion products have only been performed for the rectum and the drops of faeces and urine, the latter reflecting the situation in the rectum immediately before (A. H. Kollien et al., unpublished data). Due to dilution by the urine, in the brown rectal contents of Tri. infestans, the total concentrations of free and protein-bound amino acids are higher than in the first drop of faeces, but the percentages of amino acids are similar. The first dark drop of faeces contains 4 mM free and 18 mM bound amino acids. These concentrations are reduced to 0.15 mM total free and bound amino acids in the subsequent drops of urine. The reduction in free amino acids is significantly different for nearly all amino acids. Thus, whereas in the faeces taurine predominates, followed by proline, histidine, tyrosine, alanine and valine, in the urine proline predominates, followed by alanine, glycine and valine. 2.4.6

Other soluble factors

In the stomach of the triatomines, different factors must be considered, first of all compounds from the saliva which are continuously injected into the capillaries during blood sucking and ingested together with the blood (Ribeiro and Garcia, 1980). The saliva contains compounds that, for example, counteract host pain, dilate the capillaries and inhibit clotting of platelets (Stark and James, 1996;


190

Ribeiro and Francischetti, 2003; Champagne, 2005). According to the detailed investigations of the sialome of the blood-sucking triatomines R. prolixus, Tri. infestans and Tri. brasiliensis by Ribeiro et al. at NIH (Ribeiro et al., 2004; Santos et al., 2007; Assumpçaõ et al., 2008), the most abundant secreted proteins are lipocalins, which are not only transport proteins, but act as inhibitors of collageninduced platelet aggregation and thrombin (Assumpçaõ et al., 2008). Also in the lipocalin family are the cherry red nitrophorins, which transport the vasodilatory nitric oxide (Ribeiro, 1996). This is evident for sucking blood from capillaries since a knockdown of nitrophorins by RNAi affects the feeding on the skin, but not on the relatively large tail vein (Araujo et al., 2009). Only triatomines of the tribe Rhodniini possess nitrophorins (Pereira et al., 1998; Soares et al., 1998, 2000; Ribeiro et al., 2004). Beside apyrases, which degrade ADP and thereby inhibit platelet aggregation and inflammation, saliva contains serine protease inhibitors (thrombin inhibitors), serine proteases (presumably for the activation of enzymes), defensins as antimicrobial peptides, and in Tri. infestans the pore-forming molecule trialysin (Ribeiro and Francischetti, 2003; Assumpçaõ et al., 2008). In the stomach Kazal-type inhibitors prevent the clotting of the ingested blood, for example, by reaction with thrombin (e.g. Hellmann and Hawkins, 1965; Friedrich et al., 1993; Mende et al., 1999; Campos et al., 2002; Araujo et al., 2007). The siRNA knockdown of an intestinal thrombin inhibitor or the administration of thrombin reduces the amount of blood ingested by Tri. brasiliensis (Araujo et al., 2007). In the initial phase after blood ingestion, clotting is inhibited; in a subsequent phase, erythrocytes are lysed by a haemolytic factor that has been semi-purified from R. prolixus (Azambuja et al., 1983, 1989a) (see Section 4.6). In the interaction of blood cells and the intestinal tract also agglutinins and lectins have to be considered. Different lectins seem to be present in the stomach and the small intestine of R. prolixus (Pereira et al., 1981). In a follow-up investigation, especially the lectins in the small intestine were not verified (Grego´rio and Ratcliffe, 1991a; Ratcliffe et al., 1996). However, the levels of agglutinins seem to be affected by the blood source, and the agglutinins in the small intestine might be enzymes with lectin-like properties and not a normal type of lectin (Ratcliffe et al., 1996).

3

The trypanosomatids

3.1 3.1.1

DISTRIBUTION OF SPECIES AND STRAINS

Blastocrithidia triatomae

The homoxenous flagellate B. triatomae was first described from Argentine laboratory colonies of Tri. infestans (Cerisola et al., 1971). In such colonies high infection rates can occur (Schaub, 1988d). Despite the high numbers of wild


191

triatomines dissected for determinations of the epidemiology of Chagas disease, only very low numbers of bugs were found infected in natural populations, in Brazil 0.03% and 0.08% of Panstrongylus megistus, 0.36% of Triatoma sordida and 0.02% of different triatomines (mainly Tri. sordida), in Argentina 2% of Triatoma guasayana and 1.5% of Triatoma garciabesi and in Venezuela 5–10% of Triatoma maculata (Correâ et al., 1977; da Rocha e Silva et al., 1977; de Hubsch et al., 1977; Luz and Silveira, 1984; Schijman et al., 2006). These low rates indicate that triatomines are not the natural host of B. triatomae. Since all postembryonic instars in all populations of the insectivorous reduviid bug Zelus leucogrammus were infected with a very similar B. triatomae and since flagellates from the insectivorous bug were infective for Tri. infestans (Carvalho, 1973; Carvalho and Deane, 1974), this can be the origin of the infections of triatomines. Z. leucogrammus colonizes citrus plantations all over Brazil (Carvalho, 1973). The variability of B. triatomae has not been investigated, but subspecies have been denominated (Carvalho, 1973; da Rocha e Silva et al., 1977; de Hubsch et al., 1977). However, the original description was based on morphometrics of 10 specimens (20 specimens in length measurements). Using the original strain and measuring about 45 epimastigotes, the total length was much smaller, due to the length of the free flagellum (Reduth, 1986). After the classification of hundreds of flagellates into the different stages, I emphasize that the occurrence and arrangement of vacuoles and the length of the free flagellum are very variable. Especially in the stomach, epimastigotes with a short free flagellum develop. Comparing all morphometric data (see Table 1), the separation of subspecies according to morphological criteria is doubtful and molecular biological tools should be used. 3.1.2

Trypanosoma conorhini

Try. conorhini has only been found naturally in the triatomine Triatoma rubrofasciata (Hoare, 1972). This triatomine is very closely associated with rats (Rattus rattus), and thereby is present in tropical and subtropical ports of Africa, Asia and South America (see Section 2.1). In Southeast Asia similar parasites are present in monkeys (Hoare, 1972). The flagellate has from time to time been considered in different aspects, for example, in investigations of the cultivation and ultrastructure (cf. Deane and Deane, 1961; Milder and Deane, 1967; Deane and Milder, 1972), but no recent investigations considered a variation of strains. 3.1.3

Trypanosoma cruzi

Try. cruzi has been found in many triatomines. In the United States, it mainly infects dogs, in Latin America from Mexico to Argentina it infects all mammals including bats, correlated to the distribution of triatomines. Initially two groups were separated according to isoenzyme patterns, named zymodeme

TABLE 1 Morphometric data (mm) of Blastocrithidia triatomae from different insects or in vitro cultures Tri. infestans

Length Total Body Free flagellum Width Width at nucleus Width at vacuole Post.– Kinetopl.g Kinetopl. –Nu.h Post.–Nu. (P–N)i

Tri. infestans

In vitro

T. inf./P. meg.a

P. meg.b

Tri. maculatab

Z. leucogrammus

Mean Min

Max

Mean Min

Max

Mean Min

Max

Mean

Mean

Mean Min

53.5 25.0 22.5

48.5 14.8 19.5

62.5 32.0 28.5

35.7c 26.2 9.5

– 15.0 3.3

– 39.3 19.4

32.2c 22.0 10.2

– 33.5 22.7

51.8 23.9 23.3

54.6 32.5 18.6

26.9 24.0 34.5 28.9 12.1 8.0 15.0 15.6 14.8d 5.0d 19.5d 3.3

10.5 9.1 1.4

45.6 22.8 22.8

– 2.6

– 2.2

– 2.9

2.1e –

1.6e –

3.2e –

2.1e –

–

–

– 2.8

– 3.2

1.6f –

1.0f –

2.0f –

2.1e –

1.1e –

3.7e –

3.4

2.5

3.8

–

–

–

–

–

–

–

–

–

–

–

–

–

–

14.5

10.2

18.5

28.6

13.8

22.1

–

–

–

–

–

–

–

–

5.7

4.3

7.9

3.9c

–

–

–

0.8

0.1

3

–

–

–

10.7j

10.1j

11.8j

12.9k

16.9k

11.7j

13.1j

6.4k

4k

9k

–

–

–

16.8

8.4 – 5.6k

– 21.8k

3.2c 10.5k

– 13.0 3.4 1.6e

8.4 – 6k

3.6e

Max

Mean Min

Max

Nu.–Ant. 13.6j 12.8j 14.2j 13.2k 7.4k 23.7k (N–A)l 0.60 1.50 Nuclear index 0.79 0.79 0.83 1.00 (P–N)/ (N–A) Authorsm Cerisola et al. (1971) Reduth (1986) and Del Prado (1972)n

11.7k 0.91

7.6k

18.9k

0.60 1.50

12.8j

18.4j

0.91

0.71

5.7k

3.5k

9k

1.1

0.7

1.8

da Rocha e Silva et al. de Hubsch et al. (1977) (1977)

0.9

0.5

1.8

Carvalho (1973)

T. inf. and P. meg. ¼ Triatoma infestans and Panstrongylus megistus. Triatomines from the field, otherwise laboratory colonies or cultures. Calculated from the means. d No clear statement, whether measurement of length of free or total flagellum. e Measurement at broadest point. f Position of measurement not specified. g Post.–Kinetopl. ¼ posterior end to kinetoplast. h Kinetopl.–Nu. ¼ kinetoplast to nucleus. i Post.–Nu. ¼ posterior end to nucleus. j No clear statement of the position of measurement. k Up to centre of the nucleus. l Nu.–Ant. ¼ nucleus to anterior end. m Numbers of measured flagellates: Cerisola et al. (1971): 10 stained specimens and 10 snapshots (of the latter presumably only length measurement); Carvalho (1973): no data; de Hubsch et al. (1977): no data; da Rocha e Silva et al. (1977): no data; Reduth (1986): 45/46 flagellates from Tri. infestans (width only 21) and 57 flagellates from cultures (width only 25). n Gives only data for total length and length of the free flagellum. a b c

194


1 and 2 (Z1 and Z2) (Miles et al., 1978; Ebert and Schaub, 1983). The comparison of many strains indicates a predominantly clonal genetic structure of Try. cruzi and only restricted recombinations (Tibayrenc et al., 1986; Tibayrenc and Ayala, 1988). Strains showing a pattern differing from both group criteria are classified into zymodeme 3. Similar, but not identical groupings are obtained using other methods, for example, restriction enzymes for kinetoplast DNA and molecular biology tools, and the two groups are denominated Try. cruzi I and Try. cruzi II (Anonymous, 1999). Genetic recombinations may be the cause why some strains have to be classified into a third lineage. Discrepancies even when considering the same strains may be due to the fact that under natural conditions, double infections with parasites belonging to both groups are found and after prolonged in vitro cultivation under different conditions different populations dominate (summarized by Carneiro et al., 1990; Solari et al., 1998). Changes from one zymodeme and schizodeme to another seem to occur (Carneiro et al., 1990) also if using cloned parasites (Alves et al., 1993). Recently, subpopulations of Try. cruzi I and Try. cruzi II have been classified (Brisse et al., 2000; O’Connor et al., 2007). In some regions, these haplotypes and/or lineages of Try. cruzi are associated with infections in humans and domestic vectors, humans and sylvatic vectors, and wild mammals and sylvatic vectors (Fernandes et al., 1998, 1999; Herrera et al., 2007; summarized by Vallejo et al., 2009). In Brazil, the generalizing correlation of Try. cruzi I and Try. cruzi II to the sylvatic and domestic transmission cycles, respectively, might be changed by the increasing contact of humans to sylvatic cycles, especially in the Amazon, through which originally sylvatic Try. cruzi strains are introduced into the domestic cycle and originally domestic Try. cruzi into the sylvatic cycles. However, it is possible that the introduced strains do not establish on a long-term scale since they are not only selected by the mammalian host but also by the triatomine (Arau´jo et al., 2009). Evolutionarily, Try. cruzi I seems to originate from an association with opossums and predominates north of Amazonia while Try. cruzi II arose from an association with armadillos and predominates in Southern Cone countries of South America (Yeo et al., 2005). The dominance of Try. cruzi I in human infections north of the Amazon basin is supported by several investigations (e.g. del Carmen Sańchez-Guilleń et al., 2006; Falla et al., 2009). This might be an over-simplification, as, for example, in Chile both groups of Try. cruzi occur in sylvatic areas but circulate in different species of rodents (Galuppo et al., 2009). After infection with mixed populations of strains belonging to Try. cruzi I and Try. cruzi II, also in humans and reservoirs a host-dependent selection towards one of the groups seems to occur (Steindel et al., 2008). Clone or strain specificities are not only indicated by different numbers of parasites in the gut, but also by different percentages of metacyclic trypomastigotes (e.g. Schaub, 1989a; Perlowagora-Szumlewicz et al., 1990; Perlowagora-Szumlewicz and Moreira, 1994; Alvarenga and Bronfen, 1997; de Silveira Pinto et al., 1998; Lana et al., 1998; Lima et al., 1999).


195

Using two strains from the same locality and Tri. infestans also originating from there, the number of metacyclics differed strongly (Schaub, 1989a). 3.1.4

Trypanosoma rangeli

Try. rangeli is usually restricted to species of the genus Rhodnius (Guhl and Vallejo, 2003), although it has been found in other triatomines (Cuba, 1998). Thereby, its distribution is limited from Mexico in the north to southern Brazil. Similarly to Try. cruzi, strains of Try. rangeli are classified into two groups, KP1þ and KP1, but according to the presence and absence of a specific type of kinetoplast minicircles (Vallejo et al., 2002, 2003; Marquez et al., 2007; Cabrine-Santos et al., 2009). Again similarly, there seem to be specific associations of subpopulations with specific vectors (see Section 4.1). 3.2

DEVELOPMENTAL CYCLE IN THE TRIATOMINES

The cycle of the four species of trypanosomatids in the vector is for the most part very similar, but contains species-specific peculiarities. 3.2.1

Blastocrithidia triatomae

In the homoxenous flagellate B. triatomae, which only colonizes the intestinal tract and the Malpighian tubules, the cycle starts with the ingestion of the droughtresistant cyst. Such a stage is not present in the developmental cycle of the other three trypanosomatids – although it was reported for Try. cruzi before the description of B. triatomae (Silva, 1958) – and is correlated to ultrastructural peculiarities (Schaub and Pretsch, 1981; Schaub et al., 1990a). The encystation, indicated by the outgrowth of the flagellum, does not occur in the stomach, but requires the conditions in the small intestine (Schaub and Pretsch, 1981; Reduth, 1986). Usually more flagellates colonize this region than the rectum (Kollien and Schaub, 1999, 2002, 2003). In established infections, the small intestine harbours about 7 million B. triatomae, the rectum about 3.5 million parasites, of which at most 130,000 are attached at the rectal wall (Kollien and Schaub, 1999). During the growth of the cell to the epimastigote stage, a first encystation can occur. Usually this is a development of full grown epimastigotes and is an unequal division. The resulting daughter cell remains attached to the flagellum and undergoes further divisions. The final encystation stages detach and condense to the cyst. The enormous multiplication of the population in the small intestine and rectum is accomplished by equal divisions of the epimastigotes. Spheromastigotes, their intermediates to and from epimastigotes, and ring-like forms develop in low percentages. The ring form sometimes resembles a trypomastigote with connected ends and can only be identified without doubt in living specimens (G. A. Schaub, unpublished data).


196

B. triatomae is transmitted within a population of triatomines via coprophagy, a behaviour necessary to obtain the symbionts. Usually, fresh faeces is ingested, but the drought-resistant cysts in dry faeces can also be the source, if this faeces is liquefied by fresh faeces. In addition, during cannibalism the epimastigote population from the stomach can also be the origin of a new infection (Schaub et al., 1989a). 3.2.2

Trypanosoma conorhini

The development of Try. conorhini, a parasite of rats, is mainly known through the monograph of Morishita (1938). In the vector it starts with the ingestion of blood containing blood trypomastigotes. Generally within 24 h, they transform in the stomach into long epimastigotes and multiply by equal and unequal divisions. After 5 days they colonize the small intestine (following the textbooks of his time, Morishita denominated the small intestine ‘‘hindgut’’), and after 10 days the rectum. During the early stages of infection, the prevailing forms are the epimastigotes, then a-, sphero-, trypomastigotes and all the intermediate forms appear. Metacyclic trypomastigotes originating from epimastigotes and – according to the drawings – presumably also from spheromastigotes develop in the small intestine and rectum 5 and 14 days after infection, respectively, and less frequently in the stomach. They are transmitted to the rat via the infectious faeces or if a bug is the prey of the rat. The transmission to triatomines should apparently follow the same routes as in Try. cruzi (see Section 3.2.3). The importance of cyst-like bodies in the faeces is emphasized (Hoare, 1972), but the drought resistance has not been proved. 3.2.3

Trypanosoma cruzi

Like B. triatomae, Try. cruzi can be transmitted in a population of bugs via coprophagy – but only via fresh faeces – and by cannibalism (Schaub, 1988c). Usually, the development of Try. cruzi in the vector starts with the ingestion of blood trypomastigotes. In the stomach, they immediately are aggregated (Brener, 1972, 1973; Alvarenga, 1974), shorten and develop to spheromastigotes (Brack, 1968). After aggregation, some parasites seem to fuse, enabling a genetic exchange (Brener, 1972). The flagellates are given together with the blood into the small intestine where epimastigotes develop from blood trypomastigotes and spheromastigotes. Small and long epimastigotes multiply by unequal and equal divisions. According to the flow of the intestinal contents, epi- and spheromastigotes are transported within 2–3 days to the rectum, where they also multiply intensively. Epimastigotes take up nutrients in the flagellar pocket. Beside proteases for the intracellular digestion they possess membranebound cysteine proteases and the respective inhibitors (Souto-Padroń et al., 1990; Santos et al., 2005; Sant’Anna et al., 2008).


197

Although only remains of blood are present in the rectum, in regularly fed fifth instars of Tri. infestans 1.5 million flagellates/bug develop there, about three times more parasites than in the small intestine, and about two-thirds are attached to the rectal wall (Schaub and Lo¨sch, 1988; Schaub, 1989a; Kollien and Schaub, 1998a,b). After infecting fourth instars of Tri. brasiliensis and feeding the fifth instar, the rectum even contained up to 10 times more parasites than the small intestine (Arau´jo et al., 2008). Using another parasite/vector system, the isolate mainly colonizes the small intestine (Arau´jo et al., 2007). The infection extends to the Malpighian tubules and the final enlargements, the ampullae (Schaub and Lo¨sch, 1988; Schaub et al., 1989b). In established infections and after ingestion of uninfected blood, the stomach is rarely colonized. However, if nearly all blood has passed to the small intestine, the red colour of the remaining blood changes to brown, indicating a re-flux of digestive enzymes from the small intestine into the stomach. Only then, a population of Try. cruzi re-establishes in the stomach. Although intermediate stages may develop in the small intestine, metacyclogenesis seems to be restricted to the rectum, in established infections yielding up to 50% of the rectal population (e.g. Dias, 1934; Schaub and Lo¨sch, 1988; Schaub, 1989a; Kollien and Schaub, 1997, 1998a,b; Cabral et al., 2001). Metacyclic trypomastigotes develop from spheromastigotes via drop-like forms, from epimastigotes via a translocation of the kinetoplast to the posterior end and – similarly to the encystment of B. triatomae (see Section 3.2.1) – after unequal divisions resulting in an epi- and a trypomastigote daughter cell (Brener and Alvarenga, 1976; Schaub, 1989a). They also seem to develop from spheromastigotes via a ring form that usually is an intermediate form between round forms and epimastigotes (Schaub, 1989a) and might have been classified as vacuolized spheromastigotes in stained smears (Alvarenga, 1974). During metacyclogenesis, the surface coat changes. Whereas the events of the cell cycle of epimastigotes are precisely coordinated (Elias et al., 2007), the change of the surface coat and the internal processes – the transition of the kinetoplast – is not a strictly regulated step-by-step process since the surface labelling with wheat germ agglutinin–bovine serum albumin–gold conjugates varies strongly in the early transition stages (Schaub et al., 1989b). The new surface coat contains different glycoproteins, for example, those which protect the parasite against the complement system in the blood or are involved in the penetration of the mammalian cell (summarized by Nogueira et al., 1975; Gentil et al., 2009). The level of the expression of the respective genes by Try. cruzi can also be quantified in samples from the vector (Cordero et al., 2008). 3.2.4

Trypanosoma rangeli

Usually the infection with Try. rangeli starts via the ingestion of blood trypomastigotes. In the stomach, they transform to epimastigotes and are transported with the blood to the small intestine, where they multiply. Later in the infection,


198

they cross the intestinal wall, presumably in the anterior region of the small intestine (de Oliveira and de Souza, 2001). In the haemocoel, they multiply initially as short and then as long epimastigotes, also developing amastigotes within the haemocytes and metacyclic trypomastigotes. They only take up lipophorins and no other proteins (Folly et al., 2003). Only trypanosomes in the haemolymph – not those in the midgut and salivary glands – react with the lectin of Canavalia ensiformis (Con A) (Rudin et al., 1989). Short and long epimastigotes possess different cell surface polypeptides and ecto-phosphatase activities (Gomes et al., 2006). Finally, they penetrate the epithelium of the salivary glands – presumably using a pore-forming haemolytic compound – and are transmitted via the saliva to the new mammalian host (Tobie, 1970; Cuba, 1975; D’Alessandro, 1976; Ellis et al., 1980; Mello et al., 1995; Meirelles et al., 2005). Since Try. rangeli rarely colonizes the rectum, transmission via coprophagy must be less important. A transmission between triatomines occurs via cannibalism to an uninfected bug after ingestion of parasites present in the haemolymph and/or stomach of an infected bug (An˜ez, 1982). Although not demonstrated, Try. rangeli should also be transmitted during cannibalism by an infected bug if parasites are injected with the saliva into the attacked bug.

4 4.1

Effects of the host on trypanosomatids SUSCEPTIBILITY AND REFRACTORINESS

Investigations of the interactions of trypanosomatids and triatomines have to consider the susceptibility and refractoriness of triatomines, that is, whether or not a trypanosomatid establishes in a host and develops metacyclic forms. In all triatomines given a mixture of blood and cysts of B. triatomae or the possibility to acquire the infection via coprophagy, the flagellate establishes and encysts. According to field investigations, these include Tri. infestans, Tri. maculata, Tri. guasayana, Tri. garciabesi and P. megistus (see Section 3.2.1) and in the laboratory Tri. pallidipennis, Tri. sordida, P. megistus, P. lignarius, Dipetalogaster maxima, R. prolixus, R. robustus, R. neglectus and Mepraia spinolai (Correâ et al., 1977; da Rocha e Silva et al., 1977; de Hubsch et al., 1977; Luz and Silveira, 1984; Schaub, 1988d; Schaub and Breger, 1988, Schijman et al., 2006; G. A. Schaub, unpublished data). In all triatomines, cysts develop. Try. conorhini has been found naturally only in species of the Tri. rubrofasciata complex. However, upon feeding R. prolixus, Tri. infestans, Tri. vitticeps and P. megistus on Try. conorhini-infected rats, the flagellate also develops in these triatomines (Morishita, 1938; Dias and Seabra, 1943). Over 60 species of triatomines have been found naturally infected or have been experimentally infected with Try. cruzi, indicating that probably all species are potential vectors (Schofield, 1994). However, susceptibility varies and


199

depends on various factors (Garcia and Azambuja, 1991). In selections for susceptibility and refractoriness of R. prolixus, the third generation shows differences in the intensity of infection, but the percentage of uninfected bugs remains similar (Maudlin, 1976). Using Try. cruzi and Tri. infestans from the same village, even after eight generations of selections Try. cruzi developed in all larvae of the refractory strain, but with fewer parasites than in the susceptible strain (G. A. Schaub, unpublished data). The susceptibility is very important in xenodiagnosis, in which laboratorybred triatomines ingest blood of patients suspected to be infected but with such a low parasitemia that the parasite cannot be found in microscopical blood examinations (Dias, 1940). If the blood contains Try. cruzi, the parasites multiply and can more easily be found after about 3 weeks. However, sometimes parasites do not establish and multiply. Therefore, improvements of the methodology are tested. Using different combinations of triatomine species and/or parasite strains, the prevalences vary (e.g. PerlowagoraSzumlewicz and Mu¨ller, 1982; Perlowagora-Szumlewicz et al., 1988, 1990; Sousa, 1988; de Silveira Pinto et al., 1998; Kollien et al., 1998b; Lana et al., 1998). Feeding epimastigotes of a Try. cruzi strain isolated from Panstrongylus geniculatus to three species of Rhodnius resulted in no infections (Mejıá-Jaramillo et al., 2009). Also comparisons of xenodiagnoses indicate the superiority of the local vector (Dias, 1940), especially if infection rates of naturally infected rodent hosts are determined (Campos et al., 2007). So far, the only exception is Tri. dimidiata from Costa Rica which sometimes loses the naturally acquired infection (Vargas and Zeledoń, 1985). A loss of infection also occurs after experimental infections of triatomines (e.g. Brener, 1971; Alvarenga and Bronfen, 1997), but might be caused not only by the insusceptibility of the triatomine, but also by the use of attenuated old laboratory strains of Try. cruzi. In human chronic infections, the locality of the infection is sometimes unknown. Therefore, for xenodiagnosis the largest and most aggressive triatomine, D. maxima, has been suggested (Marsden et al., 1979). However, at least a parallel use of the local vector should be recommended. The interaction of Try. cruzi and the triatomine increases the virulence, not only the passage in the vector from the same locality as the parasite, but also in triatomines foreign to the area of the respective Try. cruzi strain (Lammel et al., 1985; Magalha˜es et al., 1996). However, in such investigations the importance of Try. cruzi strain peculiarities can hardly be separated from effects of the vector (see Section 3.1.3). Try. rangeli develops nearly exclusively in species of the genus Rhodnius (see Section 3.1.4). An invasion of the haemocoel of several species of the genus Triatoma (de Stefani Marquez et al., 2006) indicates a broader range of susceptible species. However, a suboptimal supply with symbionts might have weakened the gut, enabling the invasion. Such an effect also has to be considered in invasions of the haemocoel by Try. cruzi (Lacombe and dos Santos, 1984).


200

The refractoriness of Tri. infestans for infections by Try. rangeli is based on several mechanisms, for example, the salivary glands possess compounds which lyse Try. rangeli and are absent in the salivary glands of R. prolixus (Grego´rio and Ratcliffe, 1991a). Investigations of susceptibility are more complicated with Try. rangeli, because in contrast to Try. cruzi, long in vitro cultivation often induces an attenuation. Thereby, the parasites do not penetrate the intestinal wall or do not invade the salivary glands and have to be inoculated intracoelomically. Similarly to Try. cruzi, a high adaptation between the flagellate and the local vector seems to exist, resulting in higher rates of infection and invasion of the salivary glands (Machado et al., 2001). In addition, the Try. rangeli KP1 subpopulations seem to be specifically adapted to Rhodnius pallescens, R. colombiensis and R. ecuadoriensis, those of the KP1þ subpopulations to R. prolixus and R. robustus (summarized by Vallejo et al., 2009). 4.2

EFFECTS OF PH, OSMOLALITY AND IONIC COMPOSITION

Although strong changes of these parameters occur in stomach and rectum (see Section 2.4.2), changes in the developmental stage of trypanosomatids could not be correlated to a single factor (see Section 4.5.4). 4.3 4.3.1

EFFECTS OF THE BORDER FACE

Effects of the border face of the intestine

The trypanosomatids of triatomines do not colonize the foregut, only the midgut and hindgut. In ultrastructural investigations of infected small intestines, all trypanosomatids are in intimate contact with the perimicrovillar membranes (Jensen et al., 1990; Kollien et al., 1998a; Gonzalez et al., 1999; de Oliveira and de Souza, 2001). In these interactions, no ultrastructural modifications of the flagellum of trypanosomatids are evident. A direct attachment to the cells of the midgut wall is only achieved by B. triatomae. Then the flagellum is enlarged, sometimes resembling a flagellapodium, enabling an interdigitation with the microvilli (Jensen et al., 1990). The importance of the perimicrovillar membranes for the development of Try. cruzi and Try. rangeli is shown in the detailed investigations of the group of P. Azambuja and E.S. Garcia (summarized by Garcia et al., 2007). Epimastigotes of Try. cruzi attach via glycoinositolphospholipids at their surface to the luminal surface of the small intestine of R. prolixus (Alves et al., 2007; Nogueira et al., 2007). Decapitation and giving antiserum against the membranes and midgut tissue strongly affects Try. cruzi, and after 10 days parasites are present only occasionally (Gonzalez et al., 2006; Alves et al., 2007). The flagellate develops only up to 15 days in the intestine of decapitated insects, and no metacyclics are present in the rectum. The effect on the population is partly


201

rescued by a subsequent feeding of ecdysone, but again, no metacyclogenesis occurs (Cortez et al., 2002). The effects of hormonal disturbances are also emphasized using phytochemicals, for example, lignoids and azadirachtin (Rembold and Garcia, 1989; Azambuja and Garcia, 1992; Gonzalez and Garcia, 1992; Cabral et al., 1999; Gonzalez et al., 1999; Garcia and Azambuja, 2004). In the rectum, peculiarities of colonization are evident. Try. cruzi prefers the cuticle and only about 30% of the rectal population are free in the lumen (Schaub and Lo¨sch, 1988). In infections with B. triatomae and Try. cruzi, both parasites initially colonize the four rectal pads (Zeledoń et al., 1977, 1984, 1988; Bauer, 1984; Bo¨ker and Schaub, 1984; Schaub and Bo¨ker, 1986a). In established infections, about 30–60% of the attached population of B. triatomae and about 50% of Try. cruzi are localized on the rectal pads (Schaub and Lo¨sch, 1988; Kollien and Schaub, 1997, 1998b, 1999), which cover – roughly estimated – only about 20% of the rectal surface. This preference could be due to the location directly at the entrance of the pylorus/ileum into the rectal sac, thereby offering direct access to new intestinal contents after the passage into the rectum. However, this preference is also evident in longterm starved bugs (see Section 4.5.2), in which content is rarely forwarded. Therefore, other factors seem to be responsible for this colonization. One factor might be also of nutritional nature. Although the absorption processes in the rectum of triatomines have not been investigated, the very thin epicuticle of that region and absorption processes at similar cells of other insects suggest this (see Section 2.4.1). Thereby, the flagellates would colonize a region with a very intense flow of amino acids. Another factor might be the relative stability. Whereas the cuticle of the main rectal sac is folded and unfolded due to the filling state of the rectum and defecation, the region of the four rectal pads remains relatively unchanged, since the four flat ampullae at the end of the Malpighian tubules are of similar size as the pads and are located at the haemocoelic side of the rectal wall (Lacombe, 1957). This should stabilize the rectal pads. Several mechanisms have been proposed to explain attachment of flagellates to the cuticle, for example, via lectins to chitin residues (summarized by Schmidt et al., 1998). Using wheat-germ lectin for the detection of chitin, asialofetuin for galactose-binding lectins, and heparin for heparin-binding receptors, chitin is proved only in the procuticle, which is very thin at the rectal pads and covered by the epicuticle and a superficial layer. Therefore, chitin is not accessible for Try. cruzi. In addition, no carbohydrate–lectin interaction between the rectal wall and the flagellates could be detected. Using the fluorochrome Nile red for staining of lipids and hexane for the extraction of lipids, the superficial layer at the luminal surface of the rectum was shown to be covered by a wax layer (Schmidt et al., 1998). The use of hexadecane droplets identifies a so far unrecognized small region near the tip of the flagellum of epimastigotes (Kleffmann et al., 1998). Thus, the initial


202

attachment is based on the hydrophobicity of the wax layer and is restricted to the small region of the flagellum. This was verified using different hydrophobic and hydrophilic substrates. The mechanism of attachment allows a gentle separation of trypomastigotes from epimastigotes (Kleffmann et al., 1998). Furthermore, the attachment enhances the transformation of epimastigotes to trypomastigotes in hydrophobic, wax-coated culture vessels. The importance of attachment in this crucial step of the development of Try. cruzi is also emphasized by Bonaldo et al. (1988). Strain-dependant differences in the attachment of Try. cruzi to the rectum of different triatomines seem to be caused by differences in the composition of the rectal cuticle and result in different metacyclogenesis rates. Upon incubating recta of two species of triatomines, R. neglectus and Tri. pseudomaculata, with epimastigotes of Try. cruzi strain Y and strain Berenice, epimastigotes of the latter strain adhere better to the recta from R. neglectus than to recta from Tri. pseudomaculata, corresponding to a higher metacyclogenesis rate in vivo (Carvalho-Moreira et al., 2003). The Try. cruzi strain Y also develops more metacyclic trypomastigotes in R. neglectus but shows no differences in the attachment rate to the rectum of the two triatomines in vitro. After the initial attachment to the rectal wall, the flagellum of epimastigotes of B. triatomae and Try. cruzi is modified. Located at the top of a carpet of flagellates, the flagellum of epimastigotes of B. triatomae is enormously elongated (Schaub and Bo¨ker, 1986a). At the attachment site to the rectal cuticle, epimastigotes of both species develop enlargements of the flagellum (Schaub and Bo¨ker, 1986a,b; Zimmermann et al., 1987; Kollien et al., 1998a).

4.3.2

Effects of the border face of the salivary glands

Only Try. rangeli is transmitted via the saliva and has to reach the lumen of the salivary glands. Therefore, the first border face is the basal lamina. Investigating the distribution of carbohydrate moieties on the tissues of R. prolixus and Try. rangeli and the interactions of both via incubations in different sugars, N-acetylD-glucosamine, N-acetyl-D-galactosamine and galactose show the highest inhibitory effect (Basseri et al., 2002). Thus, lectins or these sugars on the surface of the long epimastigotes of Try. rangeli might interact with the corresponding sugars or lectins of the basal lamina of the salivary glands in the adhesion before invasion. Then the epimastigotes penetrate the host cell with the flagellum (Meirelles et al., 2005), similarly to the interaction of B. triatomae with host cells (Reduth et al., 1989; Schaub et al., 1990a). The molecular interactions with the inner surface of the salivary glands have not been described in detail. After penetration, the long epimastigotes attach to the microvilli developing no flagellar adaptations. The short metacyclic trypomastigotes swim free in the saliva (Hecker et al., 1990; Meirelles et al., 2005).


4.4

203

EFFECTS OF MICROORGANISMS AND ANTIMICROBIAL COMPOUNDS

Although the intestinal tract is colonized by many different bacteria and fungi, an effect of non-symbiotic bacteria on the flagellates has only been considered for S. marcescens. Using a colony of R. prolixus which is infected with the bacterium and a subsequent infection with epimastigotes of Try. cruzi strain Y and strain Dm28c, the population density of strain Dm28c in the stomach remains unchanged, whereas the density of strain Y decreases (Azambuja et al., 2004, 2005a). In incubations of the bacterium with these Try. cruzi strains in vitro, the same differences arise. Using different strains of S. marcescens, only the strain which produces the red pigment kills Try. cruzi strain Y. Also Try. rangeli is lysed after the development of long filamentous structures that connect the bacteria with the parasites (Castro et al., 2007a,b). The effects of symbionts on the trypanosomatids are totally under-investigated. According to the only investigation of the development of Try. cruzi in aposymbiotic and symbiotic R. prolixus, the initial development of the flagellates is stronger in bugs containing symbionts, but after a longer period stronger in aposymbiotic bugs (Mu¨hlpfordt, 1959). However, the percentages of the broad and slender epimastigotes, amastigotes and metacyclic trypomastigotes remain unaffected. The importance of symbionts is also indicated using B. triatomae. In Tri. infestans, a supplementation of blood with B-group vitamins (folic acid, nicotinic acid, pantothenic acid, pyridoxine, riboflavin and thiamine) supports the initial development of B. triatomae in the small intestine of young instars, but not in the rectum (Jensen and Schaub, 1991). Possible effects of antibacterial compounds of triatomines on their respective trypanosomatid have not been investigated (Azambuja et al., 1998). However, antibacterial compounds not belonging to the repertoire of triatomines affect the development of Try. cruzi in vitro, for example, mellitin (Azambuja et al., 1989a). In addition, the production of the lepidopteran cecropin by transformed symbionts kills all Try. cruzi in the gut (Durvasula et al., 1997; Beard et al., 2002). The development of flagellates in bugs with a knockdown or overexpression of antimicrobial compounds like lysozymes or defensins or the 7- and 25-kDa compounds of the haemolymph or enzymes of the immune system like the prophenoloxidase or nitric oxide synthase remains to be considered. It is unknown whether or not compounds of the mammalian host ingested in the blood, for example, complement and antibodies, act on the flagellates during the initial transformation in the stomach after the ingestion of infectious blood. However, in old established infections, in which the stomach has been re-colonized from the small intestine (see Section 3.2.3), the epimastigotes there are killed by the complement system in the blood of chicken, but not by the weak complement system in the blood of mice (Schaub, 1988c). After such a feeding on mice, epimastigotes of the stomach population are able to bind plasminogen from the blood meal (Rojas et al., 2008). Whether or not this acquired proteolytic activity is advantageous for the parasite remains to be


204

investigated. In vitro, decomplemented sera of mice or rabbits previously immunized with homogenates of epimastigotes of Try. cruzi, agglutinate epimastigotes and induce ultrastructural damages (Fernańdez-Presas et al., 2001). 4.5 4.5.1

EFFECTS OF DIGESTION, DIGESTION PRODUCTS AND EXCRETION

Effects of digestion

In the initial development, all four trypanosomatids strongly colonize the small intestine, and the excystation of B. triatomae is restricted to this part of the intestine (see Section 3.2.1), the main region of digestion of triatomines. Thus, the trypanosomes must possess a refractory surface, a rapid shedding of attached proteases or inhibitors of digestive enzymes. An indication for the latter is the identification of chagasin, a cysteine protease inhibitor, at the surface of Try. cruzi (Monteiro et al., 2001; Ljunggren et al., 2007). A recombinant form inhibits cysteine proteinases of an insect pest of beans (Monteiro et al., 2008). Chagasin might act against cathepsins in the direct neighbourhood of the parasite or in the lumen of the gut (see Section 5.3.4). However, the developmentally regulated expression of chagasin is inversely correlated with that of papain-like cysteine proteases, cruzipains, and the inhibitor is present at lower levels in the epimastigotes, the major form in the gut, than in tissue culture trypomastigotes (Monteiro et al., 2001). In epimastigotes, it might mainly regulate the endogenous cruzipain. Inhibition of the activity of cathepsin B of R. prolixus by a supplementation of blood with pepstatin does not affect the population density or metacyclogenesis of Try. cruzi (Garcia and Gilliam, 1980), and it remains to be investigated whether or not proteases of Try. cruzi are upregulated in these triatomines. Whereas triatomines and Try. cruzi seem not to compete for major nutrients of the blood, there is a strong competition for sialic acid (summarized by Amino et al., 1995). The acquisition of sialic acid appears to be important for the survival in the mammalian host, and Try. cruzi possesses a trans-sialidase to transfer sialic acid to its surface. However, in Tri. infestans, a strong sialidase is mainly active in the stomach, rapidly desialylating the blood cells. Thereby, the epimastigotes in the midgut are poorly sialylated, an indication that sialic acid is not required for the development in the intestine (Amino et al., 1995). The reason for the high levels of the trans-sialidase in stationary phase epimastigotes in the rectum remains to be clarified. 4.5.2

Effects of starvation

Strong effects on the flagellates are induced by the concentration of digestive products due to starvation or blood ingestion. Starvation not only affects the number of parasites, but also the development of stages. In established B. triatomae infections of fifth instars of Tri. infestans, which have been totally


205

engorged in the fourth instar, a short-term starvation of 1 month induces no strong effects (Kollien and Schaub, 2002). However, after an additional 3 months, the population in the small intestine is reduced by about 45%, and 30% of the mastigote stages in the small intestine and 99% in the rectum are dead. Whereas the composition of the population in the small intestine – 90% epimastigotes, 7% cysts and 3% spheromastigotes – does not change, in the rectum starvation induces a strong encystation and a change from about 30% to nearly 90% cysts, 10% epimastigotes and some spheromastigotes (Kollien and Schaub, 2002). For species of the genus Trypanosoma, a general observation for Try. conorhini is mentioned that the number of flagellates decreases without a second feeding (Morishita, 1938). More details are only available for Try. cruzi. In the small intestine of Tri. infestans, starvation periods of 3 or 4 weeks reduce the number of Try. cruzi, and dead flagellates are present (Schaub, 1989a; Schaub and Lo¨sch, 1989). However, even in bugs which have died of starvation, some Try. cruzi are still alive (Schaub and Lo¨sch, 1989). Upon a later investigation, 2 months after the last feeding, the population in the small intestine was eliminated (Kollien and Schaub, 1998a). At that time, the rectum still contained 130,000 parasites, but 12% were dead. The population there decreased, 4 months after the last feeding to only 1% of the initial population. However, a total elimination never occurred. In Tri. dimidiata from the field, starvation induced an elimination of the infection in more insects than in the group fed regularly (Vargas and Zeledoń, 1985). A decrease of the population density of Try. cruzi is also evident in scanning electron microscopy (Schaub and Bo¨ker, 1986b): Whereas the colonization pattern throughout the first 16 weeks after feeding is similar (minimal around the entrance into the rectum, highest on the rectal pads and similar in the other regions of the rectal sac), 4 weeks later, many regions are free of flagellates. However, a residual population always remains attached to the rectal pads. In the addition, the composition of the population changes (Kollien and Schaub, 1998a). Whereas in well fed bugs only up to 2% of the population are spheromastigotes, the percentage of this stage and its intermediate forms increases to about 20% at 2 and 3 months after the last feeding (Kollien and Schaub, 1998a). An effect of starvation on the metacyclogenesis rate occurs if the bugs are fed once on an infected mammal and subsequently starved. Then, metacyclics are scarce (Piesman and Sherlock, 1985). 4.5.3

Effects of blood ingestion and excretion

Not only a reduction of the available nutrients affects the flagellates in the intestine, but also blood ingestion and the resulting excretion. In B. triatomaeinfected fifth instars of Tri. infestans, the number of parasites in the small intestine increases 16-fold within 15 days after feeding, but only 1.5-fold in the rectum (Kollien and Schaub, 2003). However, in the rectum feeding initially

206


induces a washing out of 93% of the population present there before feeding. This reflects the high population density of this flagellate in the lumen – 2 million flagellates – and the fact that only 4% of the population are attached to the rectal wall. Since the ‘‘carpet’’ of flagellates consists of about five layers (Schaub and Bo¨ker, 1986a), there seems to be no free attachment area. In Try. cruzi-infected Tri. infestans, feeding might affect the population in the stomach (see Section 4.4); in the small intestine, the population density increases (Schaub, 1989a). In bugs ingesting different amounts of blood, more epimastigotes develop in those having ingested more blood (Asin and Catala´, 1995). Direct effects of blood ingestion were only investigated for the rectal population. Similarly to B. triatomae, the majority of the population in the lumen is washed out by the urine but only a lower percentage of the attached population (Schaub and Bo¨ker, 1987; Schaub and Lo¨sch, 1988; Kollien and Schaub, 1997). However, Try. cruzi never develops such high densities as the homoxenous flagellate and 50–70% of the population are attached to the rectal wall. The phenomenon that the percentage of metacyclic trypomastigotes is low in the first drop of faeces (which however in total contains 1000–25,000 metacyclics) and that the urine often contains pure populations of metacyclics presumably is based on the inability of trypomastigotes to attach (Zeledoń et al., 1977, 1984, 1988; Bo¨ker and Schaub, 1984; Schaub and Bo¨ker, 1987; Schaub and Lo¨sch, 1988; Zeledoń, 1997). After the development of a thick surface coat also Try. brucei does not attach to the microvilli of the salivary gland of tsetse (Tetley and Vickerman, 1985). Metacyclics of Try. cruzi lying on the carpet or in the upper layers of the carpet are washed out by the urine. These metacyclics invade the skin lesions or mucous membranes of the mammalian host and initiate the infection (Schuster and Schaub, 2000). In addition to the changes of the population density, blood ingestion by the bug also strongly affects the composition of the population. The percentages of spheromastigotes and their intermediate forms is reduced from about 20% – the starvation effect – to the normal level of about 2–3% (Schaub and Lo¨sch, 1988; Kollien and Schaub, 1997). In addition, metacyclogenesis is induced (see Section 4.5.4). The reduction of the population density by the washing out of the population in the lumen also occurs after feeding of long-term starved bugs, that is, in fifth instars which have starved for 60 days (Kollien and Schaub, 1998b). More evident are the changes of the composition of the population. In the starved bugs about 30% are spheromastigotes and the respective intermediate stages, 20% epimastigotes and 30% trypomastigotes. One day after feeding, these forms represent 2%, 70% and 10%, respectively, and a new stage, giant cells, has appeared, so far rarely seen. In the following 2 days, the percentage of this form increases to 30–50% of the total population, and then it disappears nearly completely. Giant cells contain many nuclei, kinetoplasts and flagella and have been found in the initial development after the infection (Brack, 1968; Alvarenga, 1974). The strong improvement in the nutritional conditions seems to induce the generation of cytoplasm and cell organelles without cell division


207

processes. In vitro, such an uncoupling was induced by vinca alkaloids (Grellier et al., 1999). The strong decrease of the percentages of spheromastigotes in the first day after feeding without an adequate increase of the percentages of giant cells but with an increase in those of epimastigotes and the following decrease of the percentages of epimastigotes and an increase of those of giant cells indicate that the majority of giant cells or all originate from epimastigotes (Kollien and Schaub, 1998b). 4.5.4

Induction of metacyclogenesis

In the triatomines, metacyclogenesis already starts a short period of time after colonization of the rectum, about 1–2 weeks after infection, and the percentages increase with prolonged periods of infection (e.g. Schaub, 1989a). An interesting phenomenon is the induction of metacyclogenesis after blood ingestion (Schaub and Lo¨sch, 1988; Kollien and Schaub, 1997). This is important for the population of Try. cruzi. Since the urine washes out the population in the lumen, only those parasites can continue the development in the mammalian host which possess the surface of the metacyclic trypomastigote. A rapid induction of metacyclogenesis increases the chance for the continuation of development. Whereas unequal divisions and ring forms are rarely found in stained smears of the rectal content before blood ingestion, ‘‘drop-like’’ and slender forms each make up about 50% of the intermediate forms. These percentages change rapidly within the first four drops of faeces, and then only slender intermediate forms originating from epimastigotes are present (Schaub and Lo¨sch, 1988). This also occurs in vitro in incubations of the isolated complex of the rectum and the four Malpighian tubules in physiological saline and the induction of diuresis by the artificial diuretic hormone 5-hydroxytryptamine (Kollien and Schaub, 1997). Hence, the inducing factors must originate from the urine rather than from the haemolymph or small intestine. This interpretation is supported by incubations of pieces of recta with attached Try. cruzi either in saline, or with faeces or urine of triatomines. Within 4 h, metacyclogenesis is increased only in incubations with urine, not with a mixture of remains of digestion and urine deposited in the first drop after feeding (Kleffmann, 1999). The elucidation of the mechanisms in the crucial step in the development of Try. cruzi, metacyclogenesis, has been the topic of many in vitro investigations. Upon supplementing Grace medium with extracts of the small intestine or stomach of adult Tri. infestans, dissected 24–48 h after feeding, metacyclogenesis is induced between the fourth and sixth day of incubation (Isola et al., 1981). Extracts of adults fed 3 weeks before the dissections are inactive. Metacyclogenesis is also induced within 15 min after incubation in Grace medium supplemented with extracts of the rectum, but inhibited in the presence of ADP-ribosyltransferase inhibitors (Isola et al., 1986, 1987). Fraidenraich et al. (1993) identified a 10-kDa peptide, which is present in the rectum of

208


fifth instars and adults 2 days after feeding on chicken and increases the activity of the adenylate cyclase of Try. cruzi and thereby induces metacyclogenesis, being identified as a fragment from the amino terminus of chicken aD-globin. The concentration of this peptide in the rectum decreases during the subsequent days after feeding. In addition, after feeding on mice, an active compound of similar capacity is present in the rectum. The origin of the aD-globin requires further investigations since 24–48 h after feeding urine or uric acid granules are present in the rectum, but no dark remains of digestion of haemoglobin (see Section 2.3). Using synthetic peptides corresponding to different parts of the 90 amino acid residues of the aD-globin fragment, the peptide corresponding to residues 1–40 at the amino terminus possesses the highest activity at concentrations higher than 10 10 M. Pure chicken haemoglobin is inactive. A peptide with residues 41–73 is also inactive, but enhances the effect of the other peptide (Fraidenraich et al., 1993). Upon feeding blood or plasma with different concentrations of haemoglobin and the synthetic peptides to infected R. prolixus, the percentage of metacyclics increases at higher concentrations of haemoglobin (Garcia et al., 1995). However, pure blood is more efficient. The peptide with residues 41–73 which is inactive in metacyclogenesis in vitro inhibits the rate of metacyclogenesis in the bug. Peptides with the residues 30–49 and 35–73 induce a lower rate of metacyclogenesis than blood, but to the same extent if fed together. Since the aD-globin should also be present in the small intestine in which haemoglobin is digested, but where metacyclic trypomastigotes rarely develop (Schaub, 1989a), other factors must also be necessary in the rectum. One possibility is the attachment to the cuticle, since attachment is necessary for supporting metacyclogenesis in vitro (see Section 4.3). Explaining these differences in the experiments using urine of triatomines and rectal content containing aD-globin will require further investigations. Presumably urine enhances rapid metacyclogenesis in those epimastigotes which have started it. The aD-globin might be responsible for metacyclogenesis of epimastigotes after recovery of the population from the loss of a major part due to defecation. The effects of urine were mimicked in detailed investigations of the group of S. Goldenberg. Harvesting epimastigotes in the late exponential growth phase, short before an increase in the number of metacyclics, and incubating them for 2 h in ‘‘artificial urine’’ and then in this medium supplemented with glutamate, aspartate, proline and glucose, strongly increases metacyclogenesis rates (Contreras et al., 1985a,b), most intensively in a specific strain of Try. cruzi (Dm28c) (Contreras et al., 1988) and less strongly in other strains (G. A. Schaub, unpublished data). The ‘‘artificial urine’’ mimicks the nutritional stress induced in the triatomine by the urine, but has another pH, other ionic strengths and contains neither amino acids nor peptides (Kollien et al., 2001). In addition, in the in vitro assays glucose is necessary (Tyler and Engman, 2001). The presence of this carbohydrate in the rectum of a triatomine is doubtful.


209

An important difference is the period of time: urine induces metacyclogenesis within 15 min (Schaub and Lo¨sch, 1988), but the in vitro system requires an incubation for 2 h in the ‘‘artificial urine’’ and a subsequent incubation in the enriched medium. However, the reproducibility and good timing of events enables the identification of steps in this cAMP-mediated process and of genes specifically expressed during metacyclogenesis (e.g. Gonzales-Perdomo ´ vila et al., 2001). et al., 1988; Krieger and Goldenberg, 1998; A In addition to these peptides and chemically defined media, also free fatty acids induce metacyclogenesis, especially oleic acid (Wainszelbaum et al., 2003). Using concentrations similar to those found in the intestinal tract, not the usual adenylate cyclase pathway (Parsons and Ruben, 2000), but protein kinase C isoenzymes are translocated to the membrane of culture-derived epimastigotes (Belaunzarań et al., 2009). Also in these investigations, metacyclogenesis is induced in epimastigotes. Whereas several factors have been found to induce metacyclogenesis in epimastigotes, so far there is no indication which factors might induce metacyclogenesis via the other routes from spheromastigotes, ring-like forms and the unequal divisions. 4.6

EFFECTS OF OTHER SOLUBLE FACTORS

Detecting a possible effect of factors of components from the saliva or the Kazal-type inhibitors from the stomach on Try. cruzi will require investigations using knockdown or overexpressing triatomines. So far experiments using the RNAi technique have not considered infected bugs. However, some effects in the stomach are striking and perhaps are responsible for the susceptibility or refractoriness of the respective species or population of triatomines for a specific strain of Try. cruzi (see Section 4.1). Comparing the agglutination and lysis of three strains of Try. cruzi by extracts of the stomach of R. prolixus, two strains, Dm28c and Cl, are agglutinated but not lysed and establish in the bugs. Strain Y is not agglutinated but is lysed and is unable to establish (Mello et al., 1996). Since epimastigotes of the strains Cl and Y do not react with peanut agglutinin – a criterion for the classification into Try. cruzi Z1 (Schottelius, 1982) – the agglutination is not correlated to strains of one of the two major groups of Try. cruzi. A lectin of the intestinal tract is suggested to interact with a major surface glycoprotein, GP72, since monoclonal antibodies against GP72 strongly inhibit metacyclogenesis of epimastigotes (Snary, 1985). After deletion of the gp72 gene, the faeces of Tri. infestans contains less than 1% of parasites present in the faeces of bugs infected with a wild-type strain (Basombrıó et al., 2002). In infections of Rhodnius with Try. rangeli, trypanolytic factors in the haemolymph interact with the parasite (Grego´rio and Ratcliffe, 1991a,b; Mello et al., 1995). In addition, the differences of the susceptibility of different species of Rhodnius for Try. rangeli KP1þ and KP1 isolates depend on these factors, for example, haemolymph of R. prolixus lyses Try. rangeli


210

KP1 isolates but not Try. rangeli KP1þ isolates (Pulido et al., 2008). A haemolymph galactoside-binding lectin from R. prolixus affects the survival of short, but not long epimastigotes (Mello et al., 1999).

5 5.1

Effects of trypanosomatids on triatomines CLASSIFICATION OF PATHOGENICITY AND ACTION OF SECONDARY STRESSORS

The classification that ‘‘the trypanosomatids are probably all parasitic’’ (Vickerman, 1976) and the definition of parasites as organisms that affect the host which belongs to another species (Wu¨lker and Schaub, 2002) implies pathological effects of parasites on the host. However, pathological effects are evident in less than 30 trypanosomatid-insect systems (summarized by Schaub, 1992) and not even all Trypanosoma sp. are pathogenic for the respective vertebrate host, for example, Try. rangeli is classified to be a ‘‘harmless parasite of man and a variety of wild and domestic animals’’ (D’Alessandro, 1976). This discrepancy can be solved by using the term ‘‘subpathogenic’’ instead of ‘‘apathogenic’’ or ‘‘non-pathogenic’’ for those species of trypanosomatids for which no pathological effects are known. According to the definition of this term, no effects are obvious under optimal conditions and infections induce only adverse effects if a second synergistic stressor is present (Schaub, 1989b, 1992). Under optimal feeding conditions, an increase in the number of blood feeds and/ or the volume of ingested blood compensates the metabolite losses of triatomines to the parasite. Secondary stressors regularly occur in natural populations. Such populations are often subjected to adverse biotic and abiotic stress factors, for example, predators, competitors, availability of food, temperature and humidity. Effects of pathogenic trypanosomatids can be enhanced by such stressors, whereas effects of subpathogenic trypanosomatids can only be recognized under stress conditions. Synergistic actions have to be considered, if the mortality rates in groups of larvae of triatomines are >20%. Handling stress, a factor to which bugs react very sensitively, is especially important (summarized by Schaub, 1988d; Schaub and Breger, 1988). Effects of the pathogenic B. triatomae on Tri. infestans are increased by a maintenance in critical group sizes. Since triatomines often stay close together like many non-predatory Hemiptera, a maintenance isolated singly should affect them. Whereas the development of only a minor proportion of uninfected singly isolated bugs is delayed in comparison to larvae maintained in groups of 20, 30 or 40 bugs, more infected isolated bugs show a delayed development than infected bugs maintained in groups (Schaub, 1990b). Beside this isolation effect, a much stronger crowding effect on development is evident by a maintenance of groups of 50 larvae in 1-l beakers. Also the mortality rates increase by about 20–75%.


211

An enhancement of the pathogenicity also reduces the starvation capacity. After an infection in the first instar and a last feeding in the second, third or fourth instar, the mean starvation resistance period is reduced, respectively, by 51%, 55% and 32% relative to uninfected bugs (Schaub and Lo¨sch, 1989). In the most resistant stage, the fourth instar, uninfected larvae survived up to 432 days after feeding in the third instar, whereas the last bug in the B. triatomae-infected group survived 140 days. The effect on starvation resistance is the only unequivocal case of a synergistic effect of Try. cruzi and a second stressor. In the same experimental design as in the study with B. triatomae, after the last feed in the second, third and fourth instar, which allows the development to the next instar, the mean starvation resistance of third, fourth and fifth instar larvae is reduced, respectively, by 3%, 14% and 32% relative to uninfected bugs. Again more food remnants are present in the intestine of infected than in uninfected bugs. This indicates that not the availability of proteins like haemoglobin determines the starvation capacity, but the concentration of essential metabolites for which trypanosomatid and vector compete (see Section 4.4). 5.2

PATHOGENICITY OF BLASTOCRITHIDIA TRIATOMAE AND TRYPANOSOMA RANGELI

Both parasites induce a complex sickness syndrome in triatomines, but are pathogenic or apathogenic to just the opposite species of triatomines. Whereas B. triatomae strongly affects species of the genus Triatoma, Tri. rangeli is only pathogenic for species of the genus Rhodnius (Schaub, 1988d, 1992; Schaub and Breger, 1988). In the latter system, this can be explained by the limited development of Try. rangeli in species outside of the genus Rhodnius, but B. triatomae also develops well in R. prolixus without affecting it (Schaub, 1988d). (Since the effects of B. triatomae are obvious and no effect is mentioned by Carvalho, 1973, the insectivorous Hemiptera Z. leucogrammus is presumably not affected by this flagellate.) It should be emphasized that apart from the effects on feeding behaviour, the other effects of Try. rangeli have been recognized not in animals from the field, but in experimental infections. An artificial combination of strains of parasite and vector may induce effects which do not occur in field infections (summarized by Schaub, 1992). In addition, in laboratory infections handling stress may increase the pathological effects on triatomines (summarized by Schaub and Breger, 1988). 5.2.1

Pathology: Effects on fitness, development and mortality rates

For B. triatomae and Try. rangeli, both trypanosomatids affect the general fitness, resulting in sluggish movements (Grewal, 1957, 1969; Schaub and Schnitker, 1988). Both also strongly increase the period of time required until moulting especially in the older larval stages. Under the respective feeding


212

schedule and at 27 C, 50% of B. triatomae-infected first instar larvae of Tri. infestans need 150 days to moult to the adult stage, whereas uninfected same stage larvae need 130 days (Schaub, 1990a). Try. rangeli infections prolong the period of time until this final larval moult of R. prolixus by 10–40% (An˜ez et al., 1987). Also the mortality rate is strongly increased. In groups of Tri. infestans with differential exposure to coprophagic infections with B. triatomae, the larval mortality rates are correlated to the infection rates and range from 20% to 50% in groups with 10% to 50% infected bugs, respectively (Schaub and Jensen, 1990). Since mortality rates in control groups sometimes exceed 20% (summarized by Schaub, 1992) it is not surprising that no pathological effects were recognized in groups with an infection rate of 10% (da Rocha e Silva et al., 1977). The correlation of larval mortality and infection rates is also evident after experimental infections with different concentrations of cysts in the blood (Schaub et al., 1992a). After experimental infections of first instars, only 5– 39% (mean 16%) reach the adult stage (Schaub, 1990a). Especially the older instars are affected. In Try. rangeli infections of R. prolixus, the mortality rates are 18–56% higher than in uninfected controls, and not only the older instars are killed by the infections, but also the first instar (Tobie, 1965; Go´mez, 1967; An˜ez, 1984; An˜ez et al., 1987). A fivefold higher infection dose doubles the mortality rates (Grewal, 1957). B. triatomae also affects the life span of adult Tri. infestans as well as the reproduction rate. Whereas uninfected males and females have mean life spans of 35 and 30 weeks, respectively, the data for infected adults are 12 and 9 weeks, respectively (G.A. Schaub, unpublished data). The number of eggs laid per day, egg weight, hatching rate and weight of the first instars are also reduced by the infection. Together with the reduced life span, the reproduction rate is 95% lower than in uninfected Tri. infestans (G. A. Schaub, unpublished data). After infection of adult R. prolixus with Try. rangeli, the mortality rate in the following 3 months is similar to that of uninfected bugs (Tobie, 1965). In the same observation period, none of 10 uninfected, but four of 30 infected adults died (An˜ez et al., 1987). An intracoelomic inoculation of Try. rangeli, but not of Try. cruzi, reduces the egg production by 60% and the hatching rate by 27% (Watkins, 1969). 5.2.2

Pathology: Effects on behaviour

Such effects have not been considered in investigations of B. triatomae. In Try. rangeli infections of R. prolixus and R. robustus, the bugs probe more often before blood ingestion, feed less frequently, and ingest less blood and at a slower rate than uninfected bugs (D’Alessandro and Mandel, 1969; An˜ez and East, 1984; Garcia et al., 1994). This resembles effects of other parasite-infected vectors, for example, infections of fleas with Yersinia pestis, of Anopheles with


213

Plasmodium, of tsetse with salivarian African trypanosomes and of phlebotomines with Leishmania (reviewed in Schaub, 1992, 1994, 1996, 2006). The mechanisms for this parasite-advantageous change of behaviour differ, for example, a blockage or reduction of the diameter of the foregut or an attachment to mechanoreceptive sensilla in the labrum which measure the velocity of the blood flow. In Try. rangeli-infected R. prolixus, the invasion of the salivary glands by the parasites destroys salivary gland tissue – indicated by a less intensive cherry red colour – and reduces the concentration of salivary antihaemostatic components, apyrases and nitrophorins (Garcia et al., 1994). Whereas the apyrases hydrolyse ADP, released from destroyed host cells, and thereby inhibit platelet aggregation, the cherry red nitrophorins act as vasodilators and also inhibit platelet aggregation (see Section 2.4.6). Therefore, the parasitogenic changes of the behaviour can be caused by difficulties in the location of the blood vessel and/or the difficulties during blood ingestion (Garcia et al., 1994). 5.2.3

Pathology: Effects on the composition of haemolymph and intestinal contents

B. triatomae strongly affects the free amino acids in the haemolymph (Schaub et al., 1990b). The concentrations of the majority of amino acids are lower in infected bugs. Most remarkable is the decrease for phenylalanine and tyrosine, the occurrence of detectable concentrations of b-alanine and the increase in the concentration of its possible precursor aspartate. These amino acids are necessary for the sclerotization and melanization of the cuticle (see Section 5.2.4). The changes in the concentrations of the amino acids in the gut do not offer such a direct connection to a developmental process (summarized by Schaub, 1992). In strong infections of R. prolixus with Try. rangeli, the haemolymph is whitish and more copious (Grewal, 1969). After inoculation of Try. rangeli into the haemocoel of R. prolixus, concentrations of total proteins, carbohydrates and total free amino acids decreases, the latter by 27%, whereas it increases in uninfected bugs (Zeledoń and de Monge, 1966). Similarly to B. triatomae, the changes in the concentrations of individual amino acids varied, some of them increasing strongly, others being reduced (summarized by Schaub, 1992). Noteworthy is the decrease in the concentration of tyrosine (Ormerod, 1967; Watkins, 1969) since also Try. rangeli affects tanning processes (see Section 5.2.4). Interestingly, there are different effects according to the developmental stage of the flagellate. Whereas the short epimastigotes in the initial colonization of the haemolymph induce the activity of metalloproteases, such an activity is suppressed by long epimastigotes (Feder et al., 1999). At the interface with the salivary glands, short epimastigotes inhibit ecto-phosphatase activities more strongly than long epimastigotes (Gomes et al., 2008).


214

5.2.4

Pathology: Effects on cuticle and tracheal system

Both flagellates reduce the tanning and sclerotization of the cuticle and in Try. rangeli infections also the pigmentation of the eyes (reviewed by Schaub, 1992). Directly after the moult, the cuticle is soft and pink coloured. Whereas the cuticle of uninfected bugs becomes stiff and dark within 15 min, in some B. triatomae-infected bugs it changes only slightly within 1 day, and all intermediate stages to a normal tanning are present (Schaub et al., 1990b). The retarded or missing sclerotization and tanning presumably result from the competition of flagellate and insect host for the phenylic amino acids (see Section 5.2.3). In Try. rangeli infections, the pale and translucent cuticle can also be caused by the parasite’s development in the epidermal cells (Watkins, 1971a). In B. triatomae infections, effects on the tracheal system of Tri. infestans are indicated by the density of stellate cells located at the end of the tracheal system directly before the tracheoles. In bugs supplemented with the symbiont Nocardia sp. this density is reduced by 5–60%, most strongly at the rectum and least at the small intestine (Eichler and Schaub, 1998). Similar effects are evident in bugs having been infected with the symbiont of R. prolixus, R. rhodnii, and in aposymbiotic bugs. In the latter a supplementation of the blood with vitamin B reverses the effect. However, under resting conditions the oxygen consumption of infected and uninfected bugs is similar (Eichler, 1998). Therefore, an increased ventilation rate compensates the less developed tracheal system. However, this is not efficient enough during energy-dependent processes such as diuresis and moulting. Then, the oxygen consumption of infected bugs is reduced (Eichler, 1998). In Try. rangeli infections, the effect on the tracheal system is attributed to the intracellular development (Watkins, 1971a,b; Schwarzenbach, 1987), but can also be due to the same mechanisms as just mentioned (see Section 5.2.7). 5.2.5

Pathology: Effects on intestine and excretion

Infections with B. triatomae strongly affect the intestine (Jensen et al., 1990; Schaub et al., 1992b). Electron microscopy shows less developed perimicrovillar membranes and a penetration of the unprotected cells of the intestinal wall by the parasite. Finally gut cells are destroyed. Also in cultivations in vitro, which are easily possible in a co-cultivation with a triatomine cell line, the host cells are surrounded by a corona of flagellates which penetrate and destroy the cell (Reduth et al., 1989; Schaub et al., 1990a). Infected bugs often possess a dilated small intestine with red content – an indication of a missing digestion – and the red colour of the haemolymph indicates a leakage of undigested haemoglobin into the haemocoel (Schaub and Meiser, 1990). The effects of Try. rangeli can be attributed to the destruction of the cells through the intracellular development. A multiplication in nerves and muscle


215

cells of the gut affects peristalsis, and the gut may burst (Watkins, 1971a). Similar effects occur after blocking the abdominal spiracles. According to ultrastructural investigations the cells of the midgut of infected and uninfected bugs are covered by similar dense layers of perimicrovillar membranes (Schwarzenbach, 1987). Sometimes these cells also show an electron-lucent cytoplasm after penetration of the cells of the anterior small intestine by Try. rangeli (Hecker et al., 1990; de Oliveira and de Souza, 2001). Both trypanosomatids affect the excretion, indicated by a swollen abdomen even days after blood ingestion (Watkins, 1971a,b; Schnitker et al., 1988; Eichler and Schaub, 1998). During the first 24 h after feeding, B. triatomaeinfected fifth instars of Tri. infestans excrete approximately a 2.5-fold smaller volume of urine (Schnitker et al., 1988). Ultrastructural alterations are evident, but neither secretion rates of isolated tubules nor the storage and release of diuretic hormones in the prothoracic and metathoracic ganglionic mass are affected (Schaub and Schnitker, 1988; Schnitker et al., 1988). The discrepancy between the in vivo and the in vitro results might be due to the effects on the tracheal system in the bug and the good oxygen supply in the in vitro investigations (see Section 5.2.4). In Try. rangeli-infected R. prolixus, diameters and in vitro excretion rates of the Malpighian tubules are affected (Watkins, 1971a,b). The latter is caused by damage of the tissue, but also by a reduction of the concentration of diuretic hormones in the meso-metathoracic ganglia or the presence of inhibitors. 5.2.6

Pathology: Effects on the immune system

An effect on cellular immunity in the haemocoel is evident in old B. triatomae infections in which cells rarely attach to pieces of nylon thread for an encapsulation, and no melanization is evident (G. A. Schaub, unpublished data). Also the intestinal homeostasis is affected. After an infection of first instar larvae of Tri. infestans with the symbiont and B. triatomae and feeding with a mixture of blood and Candida sp., Dietzia maris, Escherichia coli or Gordonia rubropertinctus to third instar larvae, the non-symbiotic microorganisms are rarely present in the different regions of the gut of B. triatomae-uninfected fifth instars and regularly in B. triatomae-infected larvae (Eichler, 1998). Also, Try. rangeli affects the cellular immune system (Azambuja and Garcia, 2005). After an injection into the haemocoel of R. prolixus, the number of phagocytic cells increases (Zeledoń and de Monge, 1966; Mello et al., 1995). In old and heavy infections, the intracellular development results in a decrease of the number of haemocytes (Grewal, 1957; Gomes et al., 2002). After oral infection, Try. rangeli seems to inhibit the release of arachidonic acid and reduces the concentration of an insect platelet-activating factor (iPAF). Thereby, pathways of cellular immunity via eicosanoid and iPAF are depressed (Garcia et al., 2004a,b; Machado et al., 2006; Figueiredo et al., 2008). A component at the interface of cellular and humoral immunity, the


216

prophenoloxidase system, is not activated by Try. rangeli, presumably due to an immune suppression (Grego´rio and Ratcliffe, 1991a,b). However, this is only evident after an inoculation of long epimastigotes – resembling the late phase of an infection – whereas the inoculation of short epimastigotes activates the prophenoloxidase system (Gomes et al., 1999, 2003). Agglutinating and trypanolytic factors seem to be more widely distributed in the tissues of the refractile Tri. infestans, than in those of R. prolixus (Grego´rio and Ratcliffe, 1991a,b). An indication for another effect on humoral immunity is the increased lysozyme level in the haemolymph. However, the synthesis of antibacterial peptides is not induced (Mello et al., 1995). After haemocoelic inoculation of a non-attenuated strain of Try. rangeli and a strain, that had lost the capacity to multiply in the vector and to invade the salivary glands, the latter stimulates higher levels of prophenoloxidase and superoxide and nitrites than the non-attenuated strain (Whitten et al., 2001). In both strains, the long epimastigotes induce less prophenoloxidase and superoxide than short epimastigotes. According to the effects of inhibitors, NADPH oxidases and nitric oxide synthases seem to be involved in these immune reactions. Similarly to B. triatomae (see above), Try. rangeli disturbs the intestinal homeostasis, and the numbers of usually eliminated bacteria and fungi increases (Eichler, 1998). In addition, the intestinal levels of nitrite and nitrate – metabolites of nitric oxide – and the expression rate of the gene of the respective enzyme, the nitric oxide synthase, are modulated after infections with blood trypomastigotes of Try. rangeli (Whitten et al., 2007). Already in early-stage infections, Try. rangeli represses the expression of the gene encoding the nitric oxide synthase in the stomach. Then and in mid-stage infections, the expression levels in fat body and haemocytes are reduced. After invasion of the haemocoel, the concentrations of nitrite in the small intestine decrease. In late-phase infections, the expression rate is strongly induced in the rectum (Whitten et al., 2007). In the interaction of Try. rangeli and antimicrobial compounds, not only the effects of nitrogen radicals on the flagellate have to be considered, but also an effect of reactive oxygen species like hydrogen peroxide (H2O2) (CosentinoGomes et al., 2009).

5.2.7

Pathology: Effects on symbionts

In fifth instars of Tri. infestans infected with B. triatomae, the number of colony-forming units of the symbiont Nocardia sp. in the cardia, stomach, small intestine and rectum are reduced by about 90%, 15%, 80% and 80%, respectively, in comparison to uninfected larvae and those infected with Try. rangeli (Eichler and Schaub, 2002). Try. rangeli seems to act directly on the symbionts, since the supernatants of culture media of Try. rangeli inhibit the growth of the R. prolixus symbionts on


217

agar (Watkins, 1969). However, media from in vitro cultures of other trypanosomatids, for example, Try. cruzi, must be tested to verify that these effects are specific for Try. rangeli. In infected fifth instars of R. prolixus, 6–10 days after feeding the number of colony-forming units of R. rhodnii is reduced by about 45% in the cardia, 30% in the stomach, 20% in the small intestine and 25% in the rectum in comparison to uninfected larvae and larvae infected with B. triatomae (Eichler and Schaub, 2002). 5.2.8

Mechanism of pathology

The mechanisms of pathology of B. triatomae and Try. rangeli seem to be identical. Both affect the respective symbiont, B. triatomae acting on Nocardia sp. in Tri. infestans and Try. rangeli on R. rhodnii in R. prolixus (Eichler and Schaub, 2002). The sickness syndrome of both parasites is very similar to the effects of aposymbiosis (see Section 2.4.3). At least B. triatomae and Tri. infestans compete for vitamin B (see Section 4.4), and a supplementation of the diet with vitamin B reduces the pathogenicity of B. triatomae (Eichler and Schaub, 1998). However, the number of symbionts is also reduced by a blockage of the spiracles (Eichler, 1998), indicating the strong dependence of the development of symbionts on an adequate oxygen supply. Since the number of end cells of the tracheal system is also reduced in the respective triatomine infected with B. triatomae or Try. rangeli (see Section 5.2.4), the reduced oxygen supply in long-term infected larvae could induce a vicious circle in the pathology of both trypanosomatids. 5.3

SUBPATHOGENICITY OF TRYPANOSOMATIDS IN TRIATOMINES

Since the pathological effects of B. triatomae and Try. rangeli are obvious and since no pathological effects of infections with Try. conorhini were evident in infected bugs, I suggest that this species can be classified as subpathogenic like Try. cruzi. The old investigations have been described in detail previously (Schaub, 1989b), so the present chapter only covers those published afterwards or those connected to recent publications. 5.3.1

Effects on fitness, development and mortality rates

Fitness of infected bugs has not been investigated, but according to the subpathology classification, an effect should only occur after exhaustion periods. Whereas effects on the developmental rate were found only once, three investigations observed no retardation of development of Try. cruzi-infected bugs in relation to uninfected triatomines (summarized by Schaub, 1992). Also, the mortality rates are not affected by Try. cruzi if the groups are maintained under optimal conditions as indicated by having 50% in nymphs and >80% in females, just as would aid blood meal concentration. This water loss would also be facilitated by the increase in exposed surface area as the alloscutal cuticle unfolds and stretches (see above). Once engorged and detached from the host, however, both stages increased their cuticular wax nearly threefold, with a corresponding >50% drop in water loss. Interestingly, although only a modest amount of additional wax was laid down by the late pharate female before moulting, water loss decreased sharply after moulting (despite the tick losing one of its protective layers, the outer nymphal exuvium), which suggests a change in the quality, rather than quantity, of the lipid (Yoder et al., 1997), as is true of most arthropods (Hadley, 1994). An obvious question from the ecological viewpoint is to ask to what extent these dynamics are affected by differential environmental moisture stresses. Newly emerged female A. americanum held at 33% or 93% RH (both at 22–24 C) showed no difference in the quantity of their surface lipid, but those in the drier atmosphere lost 10% less lipid in the shed exuvia, and a massive 44% less lipid in their faeces (Yoder et al., 1997). Evidently the engorged nymph can detect and react to moisture stress not by committing additional wax irrevocably to the adult cuticle (which might later impede passive water loss during feeding), but by reclaiming more lipid from the exuvium and faeces. Yoder et al. (1997) suggest that this tick species relies more on behaviour than physical properties to conserve water during host location, a more flexible strategy for an organism faced with such short-term fluctuations in ambient and internal moisture conditions. Some ticks species are known to inhabit unusually dry environments, giving rise to the idea that they are especially tolerant of dehydration. In fact, one of these, the so-called kennel tick Rhipicephalus sanguineus, is thought not to differ from other species in its dehydration tolerance, capacity for water vapour absorption or free water drinking (see Section 4.1.1). Instead, it has been able to colonize human habitation, especially where dogs are present, because of its unusually low rate of water loss, at only about 44, 53 and 67% the rate shown (above) for A. americanum as unfed nymphs, adult females and adult males, respectively, under comparable conditions (Yoder et al., 2006). Indeed, R. sanguineus appears to be restricted to dry habitats because it is intolerant of moisture-rich environments due to its xerophilic (water-conserving) adaptations. The mechanisms underlying these adaptations have not been described,

EPIDEMIOLOGY FROM PHYSIOLOGY

303

A

Cuticular lipid (µg/tick)

80

60

40

20

0

B 2.0

Water loss (%/hr)

1.5

1.0

0.5

0 FF late pharate

Newly 2 days 30 days Feeding Feeding Engorged moulted post-moult post-moult (FF slow) (FF rapid) (FFpre-ovip)

FIG. 2 Observed cuticular lipid (A) and percentage water loss (B) of Amblyomma americanum adult female (solid dots), adult males (open dots) and nymphs (triangles) at various stages of development and feeding. Drawn from data given in Yoder et al. (1997).

but the phenomenon is epidemiologically highly significant. This tick has a global distribution, found in all the major bio-geographical regions of the world (Camicas et al., 1998). Its occupation of peridomestic habitats and use of dogs as primary hosts makes it a medically significant vector of many rickettsial pathogens (Parola et al., 2005, 2008), including the highly virulent Rickettsia rickettsii, causative agent of Rocky Mountain spotted fever (Demma et al., 2005) that is currently undergoing its third known recrudescence in the United States, up to a record 1514 cases in 2004 with a 5–10% case fatality rate

304

SARAH E. RANDOLPH

(Dumler and Walker, 2006) (although over-diagnosis by new sub-specific tests may be partly to blame). The transport of pets by their owners is an obvious potential means of global dispersion of this tick, with a higher than average chance of establishment after arrival within the permissive microhabitats associated with dogs. 2.3

RESPIRATION AND METABOLIC RATES

Concomitant with the adaptation of feeding at such long intervals is the ability to survive long periods of starvation. Indeed, off-host ticks can survive starvation and desiccation longer than any other arthropod, up to several years in the field (Jaworski et al., 1984; Needham and Teel, 1991). Limited mobility obviously helps, but ticks are characterized by exceptionally low metabolic rates, 12% less than those of ants, beetles or spiders (Lighton and Fielden, 1995). In non-locomotory unfed, Amblyomma hebreum and Dermacentor variabilis ticks, such low weight-specific metabolic rates and associated rates of carbon dioxide emission are maintained by intermittent opening of the spiracles and discontinuous bursts of CO2 production lasting ca. 3–5 min at ca. 20–70 min intervals (Fielden et al., 1994, 1999). Once the female D. variabilis tick starts to feed, the interval of CO2 bursts decreases to ca. 2 min by the end of the slow feeding phase (day 6), but with shorter duration and thus little change in burst volume, followed by effectively continuous respiration over the 3 days of rapid engorgement. With continuous opening of the spiracles, absolute water loss rates increases ca. 10-fold during this final phase of feeding, consistent with the tick’s state of over-hydration at this parasitic phase of its life. Water loss by combined cutaneous and respiratory avenues, however, still constituted a tiny fraction of total water loss (1% of body mass – Yoder et al., 1997) compared with that excreted back into the host via the salivary glands (ca. 100% of final engorged weight for Boophilus microplus – Tatchell, 1967). Rather, the enhanced respiratory flow rate seems to be essential for excretion of CO2 that increases exponentially in absolute terms during the feeding period, associated with blood meal digestion (Fielden et al., 1999). CO2 excretion, however, increases faster than body mass early in feeding (Fig. 1B). It would be interesting to know what the metabolic costs of egg production are and whether this involves higher respiration and water loss rates, but this would be at a time when the female is expendable, doomed to die very soon after oviposition is complete. This physiological strategy of reduced metabolic activity and associated water loss via the spiracles except during feeding, although not unique, is taken to extremes in ticks, and again has direct consequences for the epidemiology of TBDs. The pace of pathogen transmission may be very slow, retarded by the long inter-feed interval, but the extended survival of ticks in certain habitats ensures an enduring reservoir of infections. This survival period, however, is inevitably shortened by the ticks’ attempts to find a host and progress to the next stage; the choice is between sitting still to conserve energy (taken to extreme by behavioural


305

diapause – see Section 4.3.2) and moving to find a host even if that means dying of exhaustion in the attempt (see Section 4.2). It is, of course, the adaptive balance between the rates of the life-threatening and life-enhancing processes of individual ticks that results in the diverse patterns of vector seasonal population dynamics that in turn supports specific pathogen transmission to different degrees.

3

Water balance, defence and consequences for pathogen conveyance

Salivary glands are undoubtedly the workhorses of tick survival. They provide solutions to the alternating problems of over-hydration and dehydration, and also ensure that the tick can complete its long slow blood meal in the face of host defences. During feeding, salivary secretions carry a great deal of water back into the host as the principal means of osmoregulation, a great many bioactive molecules to facilitate feeding and survival during intimate cellular contact with the host, and also, inadvertently, act as a vehicle for pathogen transport. Salivary glands are therefore still the ‘‘focus of virtually all the medical and veterinary problems associated with ticks’’ (Kaufman, 1989). 3.1

OSMOREGULATION: SPITTING INTO THE HOST

Blood is not a good food. It is nutritionally imbalanced (protein and lipid – rich, but carbohydrate – poor, and lacking certain vital nutrients such as B vitamins) so requiring symbionts in the gut, is hard to digest so that much is wastefully passed undigested through the gut and is excessively dilute. About 70% of the imbibed water and ions are actively excreted directly back into the host by the salivary glands (Kaufman et al., 1980). As befits their biological importance, salivary gland morphology, morphogenesis, function and neural control during tick feeding have all been extensively and intensively investigated and reviewed, most visually by Coons and Alberti (1999) and most recently by Anderson and Valenzuela (2008). The glands are structurally complex and transiently huge, increasing 25-fold in mass and protein content to occupy 30–50% of the adult female ixodid tick’s haemocoel while she feeds (Anderson and Valenzuela, 2008). Once feeding is complete, the glands degenerate through a sequential and highly regulated physiological process of programmed cell death orchestrated by ecdysteroid (Harris and Kaufman, 1985; et sequitor). Bowman et al. (2008) describe the current model as follows: a competent ecdysteroid receptor complex in the gland responds to rising ecdysteroid levels on completion of the blood meal that is somehow stimulated by a protein factor from the male gonad (Weiss and Kaufman, 2004). This protein appears to be of the same material as a factor, named voraxin, that is passed from males to females during copulation and stimulates engorgement, which is one of many naturally occurring substances seen to hold promise as an anti-tick vaccine (Weiss and Kaufman, 2004).

306

SARAH E. RANDOLPH

While the fundamental energetics of tick survival and reproduction clearly depend on blood meal concentration and maintenance of ionic balance, the more specific epidemiological interest arises from the way in which pathogens have exploited this aspect of tick physiology to their advantage, as follows. It hinges on the sheer volume and timescale of fluid secretion, and also the specific constituents. 3.2 3.2.1

IMMUNOMODULATION: SALIVARY PHARMACOLOGY

Battling for a meal

Because of their very large, prolonged meals, ixodid ticks face particularly acute problems in dealing with host defences against haematophagy, including haemostasis, inflammation and immune responses. Early experimental evidence that the host may acquire protective immune and allergic responses after a certain degree of exposure to tick antigens (Trager, 1939) led to the idea that acquired resistance was a feature of a tick association with non-natural host species (Ribeiro, 1989). This was explicitly tested and supported by showing that laboratory mice did, but woodmice (Apodemus sylvaticus) did not acquire resistance and reject feeding ticks (Randolph, 1979). Later, however, differences between species of natural rodent hosts were demonstrated for Ixodes trianguliceps (Randolph, 1994) and I. ricinus (Dizij and Kurtenbach, 1995): the vole Clethrionomys (now Myodes) glareolus is better able than Apodemus spp. to protect itself against tick feeding. Many of the salivary proteins in an ever-growing list catalogued by cDNA library screening, PCR subtraction and transcriptome analysis are now recognized as the key to the tick’s ability to complete its blood meal. Putative assigned functions include anticoagulant, platelet aggregation inhibitor, antimicrobial defence, anti-haemostatic, anti-platelet, anti-inflammatory, vasodilatory, tick mouthpart maintenance and unspecified housekeeping, with still a great many of unknown function (Ribeiro et al., 2006; Anderson and Valenzuela, 2008). Functional genomic tools are urgently needed to disentangle the complexity, redundancy and variability of thousands of tick salivary gland protein fragment sequences now deposited in GenBank (Anderson and Valenzuela, 2008). Given the complexity of vertebrate immune responses thrown at ticks (Wikel, 1996), it is little wonder that ticks have responded counter-punch for punch (Table 9.1 in Brossard and Wikel, 2008) as they run the evolutionary arms race in pursuit of both their dinners and their lives (Dawkins and Krebs, 1979). Salivary gland extract (SGE) has now been shown to produce effective counter-measures to 11 classes of host defensive cells or molecules, most of which appear to show a degree of species-specificity, as would be expected from such intense dynamic duelling, but which may complicate the design of new, generic vaccines or therapeutic products.


307

The ability of a tick to complete a blood meal on a host is clearly the sine qua non of that host’s complete role in tick-borne pathogen (TBP) transmission. It is possible to dream up scenarios whereby a tick takes a partial meal, sufficient to acquire infection, before being rejected and then going on to feed on another host, but there is no evidence yet that this is anything other than a rare, minority event under natural field conditions. Host-feeding specificity by ticks is therefore a crucial factor in the patterns of pathogen transmission amongst free-living vertebrates. Its variability and causes are hotly debated; extensive survey data indicate opportunistic generalism (Cumming, 1998), while intensive observational studies provide evidence of specificity (Horak et al., 1991). The immunobattle may play some part. Tick defence against the host alternative complement cascade, that is activated very early against feeding ticks (Wikel and Allen, 1977), has been shown to be specific to tick–host relationships. Anti-complement activity in SGE taken from I. ricinus, I. hexagonus and I. uriae, for example, differed between host species and was correlated with the reported host range of each tick species (Lawrie et al., 1999). This suggests that different elements in the alternative complement cascade are targeted by the anti-complement activity of each Ixodes species; saliva from both unfed and feeding I. ricinus, the species with one of the most catholic host ranges known, specifically inhibits C3a generation and factor B cleavage (Lawrie et al., 2005). Is it this early inhibition near to the start of the alternative complement pathway that confers general host permissiveness to I. ricinus, with such significant vector consequences? Other recent studies support the idea that different ticks display different host complement sensitivities because they interfere with different steps in the pathway. For example, anticomplement proteins of both I. ricinus and I. scapularis inhibit the host alternative complement pathway by interacting with properdin, thereby preventing activation of the pathway and cleavage of C3 and fB (Couvreur et al., 2008; Tyson et al., 2008). Furthermore, amongst the soft ticks, Ornithodoros moubata secretes a complement inhibitor that binds C5, preventing its cleavage and the generation of C5a (Nunn et al., 2005). 3.2.2

Orchestrating pathogen transmission

Clearly, the long evolutionary battle between ticks, pathogens and hosts may result in a species-specific balance (Humair et al., 1999). In the example above, while voles are better able than mice to protect themselves immunologically against ticks, they mount a less effective immune response against tick-transmitted Lyme disease spirochaetes Borrelia burgdorferi s. l. (Kurtenbach et al., 1994) and piroplasms Babesia microti (Randolph, 1995). This differential ability of a pathogen to use particular hosts for transmission is also apparently mediated by saliva as pathogens enter the vertebrate host at a site that is highly modified by the presence of a biting tick. The presence of whole SGE or recombinant saliva protein Salp15 (Ramamoorthi et al., 2005) inoculated alongside the pathogens significantly increases the transmission of many

308

SARAH E. RANDOLPH

TBPs, including viruses (Jones et al., 1989; Alekseev et al., 1991; Labuda et al., 1993a) and bacteria (Pechova´ et al., 2002; Zeider et al., 2002; Krocˇova´ et al., 2003; Machaćkova´ et al., 2006). This so-called ‘‘saliva-activated (now assisted) transmission’’ (SAT) (reviewed in Nuttall and Labuda, 2008), that varies with the competence of the vector species for the pathogen in question and even the particular pathogen, helps to explain the mechanism behind pathogen transmission between co-feeding ticks in the absence of a systemic infection (Jones et al., 1987; Alekseev and Chunikhin, 1990; Labuda et al., 1993a; Gern and Rais, 1996). No longer are systemic infections above a certain threshold level seen as a necessary condition for transmission of certain pathogens, but rather viraemia and even bacteraemia may occur as an inconsistent, species-specific consequence of infection. Instead, the presence of uninfected ticks co-feeding with an infected tick is sufficient for transmission to take place. TBEV and B. burgdorferi s.s. appear to be recruited preferentially to the skin site of tick feeding, rather than reaching co-feeding ticks by generalized diffusion, because skin biopsies taken from hosts where ticks are not feeding remain negative while feeding ticks at other sites become infected (Labuda et al., 1996; Gern and Rais, 1996). When tested empirically on natural tick–hosts, different species varied inversely in the development of TBEV viraemia and their ability to support co-feeding transmission: amongst rodents, despite developing the lowest viraemic titres, Apodemus spp. mice showed the highest transmission probability (Labuda et al., 1993b). The epidemiological significance for tick-borne encephalitis virus (TBEV) lies in the fact that, by avoiding the higher virulence of systemic viraemia, more hosts survive long enough to allow ticks to complete their blood meals. The consequent ca. 50% increase in transmission potential from tick to tick via vertebrates is highly significant in a system whose conventionally perceived biology up to that point appeared too fragile to permit persistence (Randolph et al., 1996). Even this, however, is not sufficient alone (see Section 5.1). As a result of this dependence on the general but specific immunomodulatory property of tick saliva to permit tick feeding on the one hand and pathogen transmission on the other, several different types of hosts are commonly needed to support the persistence of any one pathogen. Natural cycles of B. microti, for example, require both mice and voles to perform complementary roles, one to sustain the tick population and the other to sustain the microbes. This situation is especially well recognized in systems where the adult reproductive tick stage feeds on large hosts, commonly deer or other ungulates, which are typically not competent to transmit the pathogens between ticks; it applies to most I. ricinusborne pathogens such as B. burgdorferi s. l., TBEV and Louping ill virus (Gilbert et al., 2001). The non-competence of deer in pathogen transmission has given rise to the idea that they introduce some degree of zoo-prophylaxis (i.e. reduction in the abundance of infected ticks) by wasting infected tick bites (LoGiudice et al., 2003). This needs very careful quantitative analysis of field data, not just models, to be upheld, because these same hosts also contribute to maintaining and enhancing tick populations. Hitherto, models suggest that


309

zoo-prophylaxis only occurs at unrealistically high densities of deer (Rosa et al., 2003). Indeed, contrary to causing zoo-prophylaxis, an increase in roe deer abundance due to changes in land and wildlife management practices appears to be one of the most crucial factors enhancing the circulation of TBEV and, consequently the risk of TBE emergence in humans in northern Italy, and possibly in many parts of Europe (Carpi et al., 2008; Rizzoli et al., 2009). The sudden increase in deer abundance in southern Sweden in the early 1980s as a result of a crash in fox populations due to the southward spread of sarcoptic mange (Lindstro¨m et al., 1994; Kjellander and Nordstro¨m, 2003) may have contributed to the equally sudden increase in TBE incidence in 1984 (Randolph, 2001). 3.3

PATHOGEN TRAFFIC

Given the passage of such large volumes of saliva from tick to host, it is not surprising that salivary glands are the site of development and replication for many TBPs. Nevertheless, TBPs vary in their degree of cell specificity within ticks. While viruses apparently show little specificity, either within the tick body in general or the salivary glands in particular, and may be transmitted very soon after tick attachment, others infect specific cells within the glands. Theileria annulata, for example, a protozoan of considerable veterinary significance, only infects cell type ‘‘e’’ of type III acini (Bowman et al., 2008). The detailed function of each cell type (a–f) within each acinus type (I, II and III) is beyond the scope of this review, but it is relevant that both types II and III increase greatly in size during feeding. Furthermore, Bowman et al. (2008) describe how ‘‘during blood meal concentration, fluid accumulates in an expanding acinar lumen and is expelled by contraction of the adluminal cell winding in web-like fashion around the apical side of cells in acini II and III.’’ It is thus to be expected that saliva, rather than any of the other potential routes of fluid transfer (leakage of fluid from coxal glands, regurgitation, or faecal excretion) should be the vehicle of TBP transmission, but this was not conclusively shown to be the case until 1996 (Kaufman and Nuttall, 1996). Furthermore, in North America, infective B. burgdorferi s.s. spirochaetes do not appear in salivary glands until >48 h after the attachment of infected nymphal I. scapularis, as shown by inoculating susceptible mice with glands that had been dissected from infected nymphs that had fed for increasing periods (Piesman, 1995) (Fig. 3). This was corroborated by results of feeding infected ticks on mice; the probability of transmission increased rapidly from 0.70 during the third day of tick feeding (des Vignes et al., 2001; Hojgaard et al., 2008). There appears to be some species-specific variation in the timing of transmissibility: while I. ricinus did not transmit European strains of B. burgdorferi s.s. by natural tick bite if attached for only 48 h, B. afzelii was transmitted with increasing probability after 24 h of attachment (0.14 at 24 h and 0.5 at 48 h) (Crippa et al., 2002). Nevertheless, inoculation of homogenates of

310

SARAH E. RANDOLPH Tick had detached

% mice infected by inoculation of salivary glands

100 Salivary glands degenerate

80

60

40

20

0 0

24 48 60 72 96 120 Timing of source of inoculated salivary glands: no. hours after tick had attached to host

240

FIG. 3 The timing of the dispersal of B. burgdorferi s.s. to the salivary glands of nymphal I. scapularis as revealed by mouse inoculation bioassay. Drawn from data given in Piesman (1995).

whole ticks that had fed for only 24 h was infective (probability 0.6–0.7 for each strain), showing that spirochaetes are infectious in the tick before they reach the salivary glands. As Hojgaard et al. (2008) explain, this delay in infectivity may be at least partially due to the decreased production of outer surface protein A (OspA) by spirochaetes, and increased production of OspC, that both start when ticks begin to feed (Schwan and Piesman, 2000). This in turn allows spirochaetes to be released from the tick midgut protein TROSPA, migrate to the salivary glands, bind to a tick salivary gland protein Salp15, and achieve transfer to the vertebrate host (Pal et al., 2004; Ramamoorthi et al., 2005; Rosa, 2005; Hovius et al., 2007). Whatever the precise mechanism, the epidemiological consequences are clear. Human patients are at lower risk of infection if a tick is removed soon after attachment, and this may limit the need for prophylactic antibiotic treatment (Sood et al., 1997; Nadelman et al., 2001; Wormser et al., 2006).

4

Seeking a host – Where and when?

Even though ticks feed very rarely, their search for food (hosts) occupies a great deal of their time. Constraints on host-questing behaviour, driven principally by moisture and temperature conditions, are a major determinant of the variable threat they pose to humans, livestock and wildlife. The most common (but see


311

Section 2) strategy is to sit on vegetation and await passing hosts, but limited tolerance to water stress and the need to maintain their water balance forces them to interrupt their questing activity to a greater or lesser extent according to environmental conditions. In addition to short-term constraints of water balance, there are longer term constraints imposed by temperature-dependent rates of development from the previous stage, that is rates of recruitment, and also by energy reserves. Questing behaviour therefore varies diurnally and seasonally with climate (Lees and Milne, 1951; Belozerov, 1982). 4.1 4.1.1

WATER BALANCE CONSTRAINTS

Water vapour uptake by unfed ticks

The dehydration that ticks suffer while questing for hosts is reversible, but at a cost. First there is the metabolic cost of active water absorption (Fielden and Lighton, 1996a) as ticks deploy several mechanisms for extracting water vapour from unsaturated atmospheres. Central to this process is secretion of hygroscopic material by the salivary glands, specifically the type I acini, that performs two functions. First, in dehydrated unfed ticks, surplus ions and other substances are temporarily removed from the haemolymph by the production of hypertonic saliva and its deposition on the gnathosoma, where it dries and remains as a crystalline substance while RH is low (Rudolph and Knulle, 1974, 1979). The lowest RH at which water vapour uptake is possible, the so-called critical equilibrium humidity (CEH), is typically 85–90% for ticks. Once RH rises and exceeds the CEH, the crystalline substance dissolves and is swallowed (Needham and Teel, 1991). Second, it appears that water vapour absorbed at sub-saturated humidities by the hydrophilic cuticle in the hypostome can be released as salivary secretions that temporarily reduce the water affinity of this cuticle, allowing the condensed water to be sucked into the gut by a powerful sucking action of the pharynx (Gaede and Knu¨lle, 1997). Ticks do not, however, drink from free water (Kahl and Alidousti, 1997), and indeed have been shown to avoid contact with water and even walking on wet surfaces (Kro¨ber and Guerin, 1999; Guerin et al., 2000). 4.1.2

Questing behaviour and consequent tick–host relationships

Another cost associated with water balance in unfed ticks is the need to interrupt questing and return to the moist parts of their habitat, typically at the base of the vegetation where RH commonly exceeds CEH. Under favourable conditions, ticks may remain in questing positions on the vegetation for periods of several days (Lees and Milne, 1951; Loye and Lane, 1988), but they descend much more frequently in response to increased saturation deficit (SD – a measure of the drying power of the atmosphere that depends on both temperature and RH) so that, commonly, fewer ticks quest during the middle of the day (Fig. 4B).

312

SARAH E. RANDOLPH

A

Dry arena

Max saturation deficit

40 30

30

20

20

10

10

0

0 0

No. nymphs counted per hour

B

4

8

12 16 20 24 28 32

30

30

20

20

10

10 0

4

8

12 16 20 24 28 32

No. larvae counted per hour

15

10

5

5

8

12 16 20 24 28 32

0

4

8

12 16 20 24 28 32

0

4

8

12 16 20 24 28 32

0

4

8

12 16 20 24 28 32

0

0 0

No. nymphs and larvae per rodent

4

15

10

D

0

0

0 C

Wet arena

40

4

8

12 16 20 24 28 32

4

4

3

3

2

2

1

1

0

0 0

4

8

12 16 20 24 28 32

Fresh ticks added: LL + NN

NN

Day no.

LL + NN

NN

Arena watered

FIG. 4 Questing activity and attachment to rodents of Ixodes ricinus nymphs and larvae in relation to degrees of moisture stress in dry (left) and wet (right) experimental arenas. (A) maximum saturation deficit; (B) numbers of nymphs and (C) larvae counted per hour at 09:00 and 21:00 h (filled symbols) and 12:00, 15:00 and 18:00 h (open symbols); (d) numbers of nymphs (closed squares) and larvae (open squares) attached per rodent. Fresh ticks were added to the arenas on days 17 and 24 as indicated, and the dry arena was watered on day 24, both of which showed that quiescence rather than death was the cause of low questing activity. With permission from Randolph (2004).

An early observation that I. ricinus shows positive geo-tropism at saturation deficits above 4.4 mmHg (equal to 80% RH at 24 C, or 71% RH at 18 C) (Macleod, 1935) was corroborated by counts of reduced numbers of questing I. ricinus in Swiss woodlands when maximum SD exceeded 4.4 mmHg


313

(Perret et al., 2000), with adults less affected than nymphs. Likewise, experiments in quasi-natural arenas (Randolph and Storey, 1999) revealed that under such dry conditions questing activity was diminished more amongst larvae than nymphs, but for both stages it was reversible once moist conditions were restored, that is high SD, even up to 15–20 mmHg, induced quiescence rather than direct mortality (Fig. 4A–C). Nevertheless, high SD may shorten a tick’s lifespan indirectly. In the dry arena, nymphs used up their fat twice as fast as those in the wet arena, presumably largely related to the increased metabolic costs of walking, which increases in dry conditions (see below), and the active water absorption described above, thereby reducing the estimated maximum questing period from 4 to 2 months (Randolph and Steele, 1985; Randolph and Storey, 1999). If ticks run out of fat before finding a host they will die. The reduced longevity recorded under increasing SD conditions for six species of African ticks housed in glass tubes in the laboratory was partly due to direct desiccation, with ticks from largely drier habitats better able to withstand low humidities. At RH above the CEH, however, death was more likely to be due to energy depletion (Fielden and Lighton, 1996b). Accordingly, longevity at 85% RH increased with tick size, presumably reflecting the energy reserve and metabolic advantages of larger organisms as well as the relatively smaller surface area of larger bodies. The absence of spiracles in the smallest stages, larvae, presumably helps in this respect. Furthermore, aggregation amongst Haemaphysalis longicornis larvae, but not nymphs or adults, increases longevity when exposed to low humidity (Tsunoda, 2008). This, of course, arises naturally as larvae hatch from egg batches, without any additional effort required by the ticks. Presumably as a result of these physical and physiological constraints, a general feature of many tick species is vertical separation in questing positions between life stages, with sub-adults sitting nearest to the base of the vegetation, commonly with larvae lower than nymphs, and adults questing very much higher (Gigon, 1985). Inter-stadially, it seems as if ticks ascend as high as possible within the limitations of their size-related tolerance to moisture stress (Rechav, 1979), locomotory powers and costs, and energy reserves, but the benefits of height are not entirely obvious. Making contact with the larger circumference of the hosts’ upper body parts will certainly facilitate attachment there (Andra´s Lakos, personal communication), but large numbers of nymphal and larval I. ricinus, for example, attach and engorge successfully on the lower body and lower legs, respectively, of large ungulate hosts (Gilot et al., 1994; Talleklint and Jaenson, 1994; Ogden et al., 1998). Any increased ability of later life stages to exploit higher, unoccupied feeding niches is off set by the lost opportunity to exploit smaller hosts, particularly rodents that are abundant and ubiquitous. Such apparently inherent stage-specific host relationships are affected by the impact of climate on questing heights. In the same arena experiments referred to above (Randolph and Storey, 1999), dry conditions forced nymphs of I. ricinus

314

SARAH E. RANDOLPH

to quest lower down the vegetation from where they contacted and attached to small rodents in greater numbers than in wet conditions (Fig. 4D). Very few larvae in the same dry conditions fed on the rodents. Once the dry arena was watered, the same host relationships were established as in the wet arena, with the same low nymph:larva ratio on rodents as is typically seen in the wild. These climatic effects will clearly have an impact on the potential for pathogen transmission between nymphs and larvae feeding on rodents, and perhaps also on other trans-stadial routes via other host species, and involving other tick species. The risk to humans of infection with TBPs thus depends on the outcome of a balance between several extrinsic abiotic and biotic factors acting on individual ticks as they quest for hosts. The population of questing ticks will be adversely affected by hot dry conditions, and also desiccating winds, that impose mortality directly and indirectly on unfed ticks. Conversely, however, under favourable warm, moist conditions, prolonged questing will increase the probability of finding a host, thereby exhausting the questing tick population more rapidly. High host availability will have a similar effect (Randolph and Steele, 1985). Standard field sampling data alone can rarely distinguish between tick quiescence/mortality and tick feeding as a cause for the end of the active questing season (Eisen et al., 2002). 4.2

SENSORY SYSTEMS

As implied above (Section 2), haematophagous arthropods live on an energetic knife-edge; they necessarily minimize the high costs of locomotion by having evolved systems enabling them to find hosts quickly and efficiently at optimal intervals. In the case of tsetse, their sophisticated host-detection system, using a combination of olfactory and visual stimuli (Vale, 1974, 1977; Torr, 1989), gives them a 0.8 per day probability of finding a host under natural conditions of dense bush once they start their search (Randolph et al., 1992). At the same time, their strong flight apparatus permits them to redistribute themselves across large areas in response to environmental conditions; on a short timescale (days) they move between different parts of the habitat that are differentially suitable for feeding and breeding (Randolph and Rogers, 1984), and on longer timescales they track seasonal habitat suitability (Davies, 1967). Ticks are well equipped with sensory apparatus, comprising setiform sensilla, nonsetal sensilla and photoreceptors sensitive to CO2, ammonia, hydrogen sulphide, heat, sound, gravity, humidity, pheromones, host odours and light (described in detail in Coons and Alberti, 1999), but do not have the luxury of long-distance travel independent of their hosts. Recent work, however, suggests that even sit-andwait strategists are not as limited in their directed horizontal (as opposed to vertical, see above) locomotion as is commonly imagined. Ticks in water deficit will orientate towards zones of high humidity (Lees, 1948; Hair et al., 1975) and approach water drops (but avoid contact) (Kahl and


315

Alidousti, 1997). Triggered by the onset of darkness, ticks appear to make use of moist night-time conditions to walk, thought to be related to selecting suitable questing locations (Perret et al., 2003). Continuous automated observations of I. ricinus walking up and down within vertical channels were interpreted as the equivalent of horizontal displacement. In these experimental conditions, nymphs walked further after periods of quiescence (median 43 cm, max 9.7 m) than after questing (median 17 cm, max 2.9 m), while those denied atmospheric moisture walked between 5 and 31 m until they died (Perret et al., 2003). Nevertheless, walking occupied only 6.6% of the 10 days of continuous observations. In horizontal arenas, more limited displacement was observed, with 27% of nymphs not walking at all within 24 h and 30–50% of those that did walk not moving beyond 14 cm (Crooks and Randolph, 2006). In field experiments, I. scapularis Say nymphs dispersed average distances of 2–3 m, and adults >5 m, over a period of several weeks, apparently by their own locomotion (Carroll et al., 1996). If questing ticks persistently fail to contact hosts, conserving any remaining energy is clearly an essential part of their sit-and-wait strategy, but ticks that do not move may be caught in a downward spiral towards starvation. Host finding would be more efficient if any locomotory activity were directed towards hosts. When adult I. scapularis were released centrally within a 2 m diameter circle of upright wooden skewers, within 2 weeks they accumulated on skewers anointed with substances rubbed from the external glands on the legs of white-tailed deer to a significantly greater extent than on control skewers (Carroll et al., 1996). Observations on I. ricinus in more controlled laboratory conditions suggested that walking is not entirely random followed by responses to stimuli encountered by chance, but rather the onset and direction of walking is stimulated by both intrinsic and extrinsic factors (Crooks and Randolph, 2006). Comparing ticks with different levels of fat content, those with lower energy reserves were indeed less likely to walk, but overall 66% of even the low-fat nymphs (typical of field ticks in June) did move horizontally over short distances. Intrinsic moisture conditions of ticks also exerted an influence. Whereas a mild degree of dehydration did not stimulate greater walking activity, it did increase the probability of ticks moving up a humidity gradient. Biologically, this is as would be expected. Walking was also directed towards secretions from dog skin, but only when humidity was high (Crooks and Randolph, 2006). Why this should be is still a mystery. High atmospheric vapour content may possibly have impeded the dissemination of the odour molecules and so created a stronger gradient, or alternatively aided the detection of the molecules by the tick’s olfactory sensilla within the Haller’s organ (Guerin et al., 2000; Leonovich, 2004). The volatile rumen metabolites to which ticks are attracted (Donze´ et al., 2004) would reach ticks within a naturally moist air stream. Alternatively, I. ricinus might be more responsive to host odour when the moisture stress of walking is less, consistent with Perret et al. (2003) observations of night-time walking.

316

SARAH E. RANDOLPH

Thus it seems that under certain circumstances host odour can act as a kairomone (a host-produced substance that stimulates tick appetence behaviour), but does so by attracting ticks to move towards the source, rather than merely acting as an arrestant on contact as shown for other tick species (Carroll, 2002). This behaviour may allow adult ticks to adopt species-specific preferred questing heights, loosely correlated with the size of their principal host animal (Lees and Milne, 1951; Gigon, 1985; Loye and Lane, 1988). Laboratory experiments with artificial ‘‘vegetation’’, such as glass rods or wooden dowels, suggest an inherent preference (ibid ), which may have been established under natural conditions through a response to scent-markings from glands on various parts of the host’s body. Alternatively, the ‘‘choice’’ of hosts may be determined purely mechanistically by the tick’s questing height. This does not resolve the debate about whether the variable degree of host-specificity observed amongst tick species is adaptive or merely the result of opportunism operating within differential constraints. Either way, adaptations to maximize tick feeding success on the particular hosts encountered may reinforce stageand species-specific host detection and/or location mechanisms. 4.3 4.3.1

RECRUITMENT OF UNFED TICKS TO THE QUESTING POPULATION

Development rates

The backdrop to the variable host-seeking behaviour in relation to environmental conditions is the seasonal timing of questing determined principally by rates of development from one tick stage to the next. This dictates the timing of recruitment of unfed ticks into the questing population. Development rate is more useful than the birth rate as a demographic input parameter for ticks because a female ixodid tick lays a single large egg mass (typically several thousand eggs) over a short time period, but the annual recruitment of unfed larvae, nymphs and adults must be treated separately. In common with most physiological processes in poikilothermic animals, inter-stadial development rates of ticks increase non-linearly with increasing ambient temperature, with quantitative relationships that vary between stages and species (see figure 2 in Randolph 2004). It might be tempting to interpret the rather slow development rates, and consequent long inter-stadial development periods (typically months), as a trade-off against the advantages of low metabolic rate (see Section 2.3), but some tick species show exceptions indicating that slow development is not a universal constraint. In Slovakia, for example, two sympatric species within the same habitat show very different patterns: I. ricinus develops slowly so that each stage appears in successive years and the whole life cycle takes 3 years, whereas Dermacentor reticulatus develops much more rapidly, passing from engorged larva to moulted nymphs within one month during the summer so that each tick stage follows sequentially and the life cycle is completed within a year. This has significant consequences for the ticks’ role as vector of TBEV


317

(see Section 5.1). Such inter-specific differences suggest an element of adaptation evolved by each species, which may extend to each geographical race of the same species. It is essential, therefore, to establish the correct rates for each species, preferably under controlled conditions in the laboratory for application to natural temperature conditions in the field, laborious though this is. Then daydegree summation methods may be used to estimate which ticks, and therefore how many, of one stage give rise to which and how many ticks of the next stage, as counted in the field. Not only does this permit further analysis of the seasonal population dynamics of the tick (Randolph, 1997; Randolph et al., 2002), it also allows the temporal course of pathogen transmission to be constructed. The variable prevalence of infection in ticks or vertebrates at certain points in time may then be attributed to factors operating at the correct interval before. The simple fact of geographical variation in absolute ambient temperatures and their seasonal variability results in contrasting patterns of recruitment to questing tick populations, potentially continuous in the tropics and even in the sub-tropical regions of South Africa (Randolph, 1997), but intermittent in temperate regions. Where development rates drop very low and even to zero for large parts of the year (see figure 3 in Randolph, 2004), development periods are telescoped and so emergence of unfed ticks becomes more synchronized. This immediately raises the question of the potential impact of increased temperatures as part of more general climate change on seasonal patterns of tick abundance, which could be non-linear and not necessarily proportional to any simple acceleration of development. Precise seasonal patterns, that are highly variable in space and time, can be critical to pathogen transmission potential and exposure of humans (see Section 5). 4.3.2

Diapause

Few animals show only simple graded responses to environmental conditions, and ticks are no different. Below certain temperature thresholds neither development of engorged stages nor questing activity of unfed stages occurs at all, and in addition there are day-length triggers of diapause that operate while temperatures would otherwise be permissive. Diapause is particularly important in temperate species as it resets the clock each year according to the highly regular and predictable pattern of day-length change. The physiology of light detection thus becomes a central factor, although it is poorly understood for ticks. Paired eyes of variable structure at the lateral margins of the scutum have been described for Amblyomma and Hyalomma species (Coons and Alberti, 1999), while all stages of I. ricinus have no eyes but up to 21 bilaterally arranged photoreceptors immediately below the cuticle along the dorsolateral margins behind the second coxa (Perret et al., 2003). Tick eyes represent ocelli, capable of responding to shadows and to variation in light intensity and they may vaguely discriminate shapes, but cannot discriminate colour (Kopp and Gothe, 1995; Coons and Alberti, 1999). The preference of I. ricinus for walking at night

318

SARAH E. RANDOLPH

(see above) was shown to be a response to darkness irrespective of climatic conditions (Perret et al., 2003), but the exact light sensitivity of the photoreceptor cells is unknown. In the induction of diapause, ticks have been shown to respond to simulated natural changes in day length (Beattie and Randolph, unpublished observations), which are much more gradual (ca. 20–30 min per week) than those used in most experiments investigating diapause (typically an 8-h difference within a few days, either side of feeding). Defining what a tick perceives as day length (the limiting position of the sun relative to the horizon) is a separate unresolved issue. The latitudinal variation in the date of diapause induction in I. ricinus (see below), however, is best correlated with day length defined by when the sun is 9 below the horizon. Two distinct sorts of diapause are recognized for ticks, so-called behavioural diapause that switches off questing activity in unfed stages, and morphogenetic diapause that delays development in fed ticks (Belozerov, 1982). Both are presumably adaptations to ensure that the exposure of vulnerable stages to unfavourable abiotic conditions is minimized so that overall mean daily survival more than compensates for the fitness costs of a prolonged generation time. Behavioural diapause in the African tick Rhipicephalus appendiculatus illustrates nicely the plastic, presumably adaptive nature of diapause. It is confined to unfed adult ticks, which do not always quest and feed as soon as they have hardened after emergence from the engorged nymphal stage. In South Africa, this diapause appears to be induced by short day lengths so that adults that emerge after July each year (winter solstice in June in the southern hemisphere) do not start questing until some time after November of the same year (Rechav, 1981; Short and Norval, 1981; Pegram and Banda, 1990; Randolph, 1997; Madder et al., 1999). As a result, there are periods of near absence of each tick stage sequentially during the year, and the majority of the vulnerable eggto-larval stage occurs during the warm wet season (December–May), which would favour rapid development and good survival. In contrast, in equatorial Africa, where long cold dry seasons do not occur, the same species does not show any diapause (Branagan, 1973). Here, all tick stages feed throughout the year, interrupted only by abrupt declines in abundance of all stages simultaneously during the dry season, allowing continuous overlapping generations each of ca. 8–9 months duration. The propensity to diapause shows a latitudinal gradient (Madder et al., 1999), and the mechanism of diapause termination is thought to change from a long day-length trigger in the south to gradual physiological ageing further north (Berkvens et al., 1995). This variable pattern of diapause is significant not only for the survival of tick populations, but also for the epidemiology of associated infections. Diapause in southern populations of R. appendiculatus may prevent the circulation of the virulent strain of Theileria parva that causes East Coast Fever in cattle in eastern and central Africa (Norval et al., 1991). Only where there is sufficient overlap between the more or less continuous activity periods of the different tick stages can the virulent strain of T. parva be acquired from infected cattle by larvae and


319

nymphs and transmitted onwards by nymphs and adults. The less virulent strain of T. parva that causes January or Corridor disease survives for longer periods in carrier cattle, and so can be acquired by nymphs and transmitted by adults where there is less inter-stadial overlap; hence its occurrence in southern Africa. In the more markedly seasonal environments of temperate zones, ticks show both behavioural and morphogenetic diapause. The inter-stadial periods may be prolonged (i.e. adjusted) by a variable delay in the onset of development after engorgement (Campbell, 1948; Kemp, 1968). In Palaearctic (I. ricinus and I. persulcatus) and Nearctic (I. scapularis) members of this widespread species complex, morphogenetic diapause is evidently triggered by the change from long (pre-feed) to short (post-feed) day length (Belozerov, 1998; Belozerov and Naumov, 2001), either as experienced by the questing stages or as the difference between pre- and post-engorgement conditions. Thus I. ricinus, for example, that quest in the field after July enter diapause whether they feed and are held in quasi-natural field conditions (Chmela, 1969; Cerny´ et al., 1974; Daniel et al., 1976, 1977; Gray, 1982) or in constant laboratory conditions (Campbell, 1948; Kemp, 1968). There is a latitudinal gradient, with the date of diapause onset varying from the end of July in Scotland (ca. 56 N) to the end of August in the Czech Republic (ca. 49 N). This appears to be due simply to the latitudinal gradient in sun position relative to the horizon during late summer (see above), because in light conditions the simulated rapid or slow day length decrease equivalent to July–October at 60 and 49 N, respectively, and under 18 C constant temperature conditions, ticks entered diapause as photoperiods reached the same absolute length, ca. 13 h light, irrespective of the rate of decrease (Beattie and Randolph, unpublished observations). Temperature rather than day length appears to be instrumental in breaking morphogenetic diapause in I. ricinus. Even a brief period of exposure to 0 C induces rapid development without delay similar to that seen in spring-fed (i.e. pre-July) ticks (Campbell, 1948). This ensures that all ticks, whenever they feed after July/August, start development once temperatures increase after the winter (see figure 3 in Randolph, 2004). Under a wide range of conditions across the United Kingdom, the combination of diapause and seasonal temperaturedependent development rates acts to synchronize the emergence of virtually all unfed ticks of each stage to within a couple of months every autumn (Randolph et al., 2002) (which is likely to act as a brake on any impact of climate change). This conclusion was validated when emergence dates of unfed ticks predicted from a day-degree summation development model coincided precisely with the first appearance in the field of ticks with high fat content. Fat is a source of energy derived from each blood meal, which can be used as a marker of physiological ageing in the field (Uspensky, 1995; Walker, 2001). Its natural rate of usage varies with seasonal activity and climatic conditions according to the tick’s locomotory and physiological activities (Steele and Randolph, 1985; Randolph and Storey, 1999), allowing a distinction between the calendar age and physiological age of ticks. The frequency distribution of fat

320

SARAH E. RANDOLPH

contents in questing ticks collected regularly throughout several years at three sites in the United Kingdom indicated clearly that under a wide range of climatic conditions, a single cohort of each stage of I. ricinus is recruited each year in the autumn and then survives for 1 year until the ticks have either fed or died of fat exhaustion. Crucially, however, this timing of recruitment does not coincide with the timing of peak numbers of ticks seen questing in the field. Most newly emergent ticks delay their questing activity until the following spring in many parts of Europe (inter alia Lees and Milne, 1951; Daniel et al., 1976; Gray, 1985; Randolph et al., 2002). This may be true ‘‘behavioural diapause’’ (Belozerov, 1982), similar to that seen in adult R. appendiculatus and I. pacificus (Padgett and Lane, 2001), although the triggers will be species-specific. Again, short day length has been implicated (Belozerov, 1982), but in the field in mild climates increasing numbers of ticks may be active from January onwards (when day length is at its minimum, but increasing) (Fig. 5A), suggesting that decreasing day length beyond a certain level, rather than absolute day length, may be the cue for diapause. If it is indeed true diapause, it is nevertheless not absolute, as a small and variable percentage of I. ricinus do become active in the autumn. It is not yet known whether this represents a low probability of all ticks questing at this time, or the unequal division of the tick population into sub-sets with different physiological behaviour according to, perhaps, their developmental history or some genetic characteristics. 4.3.3

Temperature thresholds

Ticks also vary their questing activity in response to their immediate climatic conditions, which should be interpreted as quiescence rather than true diapause. Interacting with the putative inhibitory effects of decreasing day length, and the supposed converse permissive effects of increasing day length, low winter temperatures in temperate regions inhibit tick activity. Yet even where temperatures are permissive throughout the year, fewer ticks quest before the winter solstice than after it (Fig. 5A). It appears that at the end of each year, decreasing day length reduces the probability of questing and low temperatures may inhibit activity altogether (Fig. 5B), while at the start of each year, increasing day length is permissive, but only if temperatures are high enough. Gradual recruitment to the questing population will occur as temperatures reach the threshold level (see below) in more and more parts of the ticks’ over-wintering micro-habitats (Eisen et al., 2002). Consistent with this, altitude influences the timing of the onset, more than the cessation, of tick activity (Jouda et al., 2004). Climate change may therefore affect tick activity in the spring more than in the autumn, because diapause induced by day length will continue unchanged. In the unusually warm conditions over the winter of 2006/2007, following unusually cold conditions the year before (i.e. a case of weather variability rather than climate change), ticks were seen questing within enclosed arenas in a Berlin forest in January and February, which is earlier than usual (Dautel et al., 2008).

EPIDEMIOLOGY FROM PHYSIOLOGY A

321

25

Max Ta ⬚C

20 15 10 5 0 Nymphs and larvae/3 per 100 m2

J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D 600 400 200 0 J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D

B

Max Ta ⬚C

20 15 10 5 0

Nymphs and larvae per 100 m2

J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D 60 40 20 0 J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D 1998

1999

2000

FIG. 5 Seasonal variation in maximum air temperature (Ta) and numbers of questing larval (○) and nymphal (●) Ixodes ricinus at two sites in the United Kingdom, (A) Dorset, a warm site and (B) Wales, a cooler site (note that larval density is presented at 33% of observed level at site A). Dotted and dashed lines indicate the temperature thresholds for the onset of seasonal activity of larvae and nymphs, respectively. With permission from Randolph (2004).

The threshold temperature for activity by questing nymphs and adults of I. ricinus has been estimated approximately as a weekly mean daily maximum temperature of ca. 7 C (Macleod, 1936; Perret et al., 2000; Randolph, 2004) (Fig. 5). This is a useful estimate for analysis, interpretation and modelling, and appears to apply widely across Europe even at coarse spatial and temporal resolutions (Randolph et al., 2008), but is only approximate for a number of reasons. First, temperature is usually recorded at standard meteorological stations some distance from the tick-monitoring site and virtually never

322

SARAH E. RANDOLPH

precisely as the tick experiences it. In any case, temperature varies considerably over tiny distances within natural microhabitats. Secondly, activity threshold temperatures vary inter-specifically, interstadially, and even intra-stadially, which appears to be size-dependent (Clark, 1995). In the United Kingdom, the onset of larval activity coincides with a threshold of 10 C, rather than 7 C, mean maximum temperature (Fig. 5). Furthermore, smaller nymphs and adults start questing later each year (Randolph et al., 2002). Each new cohort of ticks that appeared on the vegetation in the autumn in the United Kingdom showed a normal distribution of size (fat-free weight), virtually symmetrical about a median of 75–85 mg (range 40–120), which then changed seasonally (Fig. 6 for an example from one site over 1 year). A sub-set of larger nymphs (and adults, data not shown) started questing the following February, with only 15–30%

Physiology of Human and Animal Disease Vectors

Animal Disease and Human Trauma

Thrombin: Physiology and Disease

Human physiology

Human Physiology

Human Physiology

Introduction to Animal Physiology

Introduction to Animal Physiology

Animal and Human Calorimetry

The Biology of Disease Vectors html

Physiology and Behaviour of Animal Suffering (UFAW Animal Welfare)

Immunogenomics and Human Disease

Worms and Human Disease

Viruses and Human Disease

Human Demography and Disease

Viruses and human disease

Animals, Disease and Human Society: Human-animal Relations and the Rise of Veterinary Medicine

Animal Encounters (Human-Animal Studies)

Netter's Atlas of Human Physiology

Schaum's Outline of Human Anatomy and Physiology

Vectors

Vectors

Vectors

Schaum's Outline of Human Anatomy and Physiology

Epilepsy: Animal and Human Correlations

Animal Welfare and Human Values

Vectors

Vectors

Vectors

Epilepsy: Animal and Human Correlations

Efficiency and Economy in Animal Physiology

Physiology of Human and Animal Disease Vectors

Animal Disease and Human Trauma

Thrombin: Physiology and Disease

Human physiology

Human Physiology

Human Physiology

Introduction to Animal Physiology

Introduction to Animal Physiology

Animal and Human Calorimetry

The Biology of Disease Vectors html

Physiology and Behaviour of Animal Suffering (UFAW Animal Welfare)

Immunogenomics and Human Disease

Worms and Human Disease

Viruses and Human Disease

Human Demography and Disease

Viruses and human disease

Animals, Disease and Human Society: Human-animal Relations and the Rise of Veterinary Medicine

Animal Encounters (Human-Animal Studies)

Netter's Atlas of Human Physiology

Schaum's Outline of Human Anatomy and Physiology

Vectors

Vectors

Vectors

Schaum's Outline of Human Anatomy and Physiology

Epilepsy: Animal and Human Correlations

Animal Welfare and Human Values

Vectors

Vectors

Vectors

Epilepsy: Animal and Human Correlations

Efficiency and Economy in Animal Physiology

Recommend Documents