• No results found

Ecology and evolution in a host-parasitoid system: Host search, immune responses and parasitoid virulence

N/A
N/A
Protected

Academic year: 2022

Share "Ecology and evolution in a host-parasitoid system: Host search, immune responses and parasitoid virulence"

Copied!
47
0
0

Loading.... (view fulltext now)

Full text

(1)

Ecology and evolution in a host- parasitoid system

Host search, immune responses and parasitoid virulence

Lisa Fors

(2)

©Lisa Fors, Stockholm University 2015 Cover photos: Robert Markus and Lisa Fors Back cover photo: Anna Erlandsson ISBN 978-91-7649-103-4

Printed in Sweden by Publit, Stockholm 2015

Distributor: Department of Ecology, Environment and Plant Sciences, Stockholm University

(3)

Sometimes you’re the windshield, sometimes you’re the bug

Mark Knopfler

(4)
(5)

List of Papers

This thesis is based on the following papers, which are referred to by their roman numerals:

I. Fors L, Liblikas I, Andersson P, Borg-Karlson A-K, Cabezas N, Mozuraitis R, Hambäck PA (2014) Chemical communication and host search in Galerucella leaf beetles. Chemoecology doi:

10.1007/s00049-014-0174-1

II. Fors L, Verschut T, Hambäck PA. Host search and host preference in Asecodes parviclava. Manuscript

III. Fors L, Markus R, Theopold U, Hambäck PA (2014) Differences in cellular immune competence explain parasitoid resistance for two coleopteran species. Plos One

doi: 10.1371/journal.pone.0108795

IV. Fors L, Markus R, Theopold U, Ericson L, Hambäck PA.

Geographic variation in parasitoid virulence and parasitoid host race formation. Submitted manuscript

(6)
(7)

Contents

Introduction ... 9

Search behaviour and host preference... 10

Host search in herbivores ... 10

Host search in parasitoids ... 11

Insect Immunity ... 12

Host defence strategies ... 12

Parasitoid counter-defence strategies ... 15

Aim of the thesis ... 16

Methods ... 17

Study system and study area ... 17

Chemical communication and host search (Paper I and Paper II) ... 19

Paper I ... 20

Paper II ... 21

Host immune response and parasitoid virulence (Paper III and Paper IV) .. 22

Paper III ... 22

Paper IV ... 23

Results and discussion ... 25

Concluding remarks ... 32

Acknowledgements ... 32

References ... 33

Svensk sammanfattning ... 40

Tack/Acknowledgements ... 45

(8)
(9)

Introduction

In nature, much of evolution is coevolution between interacting species, driven by natural selection [1]. Species can interact either through mutualistic processes where all species involved benefit, as for example in specialized plant-pollinator interactions [2, 3], or through antagonistic processes which include either competitive or trophic interactions [4, 5]. A prerequisite for coevolution is local adaptation, which can be defined as genetic change in a population, due to a variation in selective pressures across the landscape.

Local adaptation can result in a higher fitness in local individuals at their home site, compared to the fitness of nonlocal individuals, depending on their ability to cope with local biotic and abiotic conditions [6]. As gene flow may prevent populations of the same species to evolve independently, it can have great influence on the process of local adaptation [7, 8].

In systems with antagonistic interactions, for example host-parasite systems, there is a continuous coevolutionary arms race between the species, with each species imposing strong selection pressure on the other. Parasites are conventionally considered to be ahead of the hosts in the coevolutionary arms race, due to short generation times and large population sizes [9].

Consequently, if the parasites have the ability to rapidly adapt to new defence strategies in the host, it could result in local adaptation in the parasites, with a certain parasite population showing higher virulence on local compared to non-local host populations [10, 11]. However, there is a great variability in the outcome of empirical studies concerning local adaptation in host-parasite interactions [12]. In many studies, the parasites show local adaptation, but in some cases there seems to be no local adaptation or even maladaptation of the parasites [10].

A specific form of parasites are constituted by parasitoids; free-living adult insects whose progeny feed on the body of another arthropod, which unconditionally leads to the death of the host [13-15]. Parasitoids constitute a diverse group of insects that can cause high mortality in many host populations. All parasitoids are holometabolous with a four-stage life cycle of egg, larva, pupa and adult. Most parasitoids attack a particular life stage of the host insect. The juvenile stages (i.e. eggs, larvae, pupae) are most common to attack, but a few species attack only the adult insects, for example conopid wasps attacking adult bees of several genera [16]. Parasitoids are classified as

(10)

either idiobionts or koinobionts. Idiobiont parasitoids prevent further development of the host once it has been infected, whereas hosts of koinobionts continue their development after infection [17]. The parasitoids can either develop inside the host (endoparasitoids) or on the exterior of the host (ectoparasitoids). Most parasitoids fall into one or the other of these categories, but there are examples of species showing a transition between the strategies, most often from ectoparasitism to endoparasitism [18].

Endoparasitoids are usually considered to have a more complex relation to their hosts, as they must overcome the immune defence of the host but at the same time keep the host alive long enough for the parasitoid larvae to develop.

Parasitoids usually have a quite specific host range and the intimate interactions between parasitoids and their hosts make host-parasitoid systems particularly well suited for studying local adaptation and coevolution [19].

In order to better understand the evolution in host-parasitoid systems, different aspects of the interactions must be taken into account, including search behaviour and host selection in the parasitoid, development of defence strategies in the host and counter-defence mechanisms in the parasitoid.

Recently, there has been a growing interest to gain a better understanding of the complexity of trophic interactions by examining immunity-related traits in relation to evolution and ecology, a field known as ecological immunology [20]. However, there is still limited knowledge on the connections between host immunity, parasitoid virulence, host race formation and speciation in natural host-parasitoid systems. In this thesis, based on four papers, I have investigated interactions and possible coevolution in a host-parasitoid system, focusing on host search, parasitism success and host immune responses. The first part of the thesis is a general background regarding search behaviour and insect immunity, followed by a description of the four studies conducted.

Search behaviour and host preference

Host search in herbivores

Herbivore insects can use different methods to locate a host plant, including visual, olfactory and gustatory search cues [21]. The most important stimuli for many insects at a distance from the resource are olfactory and visual cues, often used in combination. Shape, size and spectral quality are examples of visual plant characteristics that may influence host selection [22]. Visual cues, unlike odour cues, are not likely to be affected by abiotic factors such as wind and temperature, and should thereby be quite stable [23]. However, in a dense and complex vegetation, olfactory signals may be more reliable than visual

(11)

characteristics. Odour cues are airborne chemical compounds of different origin, such as green leaf volatiles emitted by plants or pheromones emitted by insects. Insect pheromones are primarily used as intraspecific signals for social and sexual behaviour, but indirectly they are also used in the search for host plants [24]. Both plant odours and insect pheromones are usually complex blends of chemical components. Pheromones are often species specific, whereas many green leaf volatiles can be produced by a variety of plant species. However, many herbivorous insects have a highly evolved system for olfactory reception, enabling them to distinguish specific plant volatiles in a blend and translate them into a chemical message [25]. When herbivorous insects feed on their host plant, the green leaf volatiles released differ qualitatively and/or quantitatively from undamaged or artificially damaged plants [26]. Consequently, insects feeding on a host plant may attract more conspecifics and possibly also other herbivore species to the same plant, leading to herbivore aggregation.

Host search in parasitoids

In host-parasitoid interactions, it is crucial for the parasitoid females to locate suitable hosts and to select between host individuals of different qualities. The process of host selection is conventionally divided into three steps: host habitat location, host location and host acceptance [27-29]. Parasitoids most frequently use odour cues when locating hosts [29], although some parasitoid species use other types of signals for detection, such as sound [30, 31], visual cues [32] and electromagnetic radiation [33]. The search strategies and the cues used for host location may differ depending on whether the parasitoid is a specialist or a generalist. The usability of a certain cue depends both on how reliable the information is and how easily it can be detected by the parasitoid.

Odour cues can either be released directly from the host itself or derive from the herbivore host plant, host products or the microhabitat [34]. Parasitoids in tritrophic systems often use a combination of odour cues from both lower trophic levels for host location. Many species are attracted to green leaf volatiles released due to feeding activity of their host herbivore, but do not show the same response to volatiles of mechanically damaged plants [35-40].

Some parasitoids respond to a mixture of odours from the host and the host plant, even if the host is not actively damaging the plant. This is for example seen in some species of egg parasitoids that respond to plant volatiles induced by host egg deposition on the plant, whereas no response is observed when volatiles are released by the same plant due to larval feeding [41, 42]. When parasitoids exploit volatile chemicals directly emitted from the host as a search cue, it is likely to be a selection pressure on the host species to avoid detection by reducing the emission of the specific compound [36]. The parasitoids in their turn are under selection to evolve more efficient ways to detect the

(12)

signals of their hosts, resulting in a coevolutionary arms race between host and parasitoid [43]. However, if the host-emitted compound is also used by the host to communicate with conspecifics (i.e. a pheromone) it is likely to be a strong selection pressure for keeping the compound. Sex pheromones in particular are expected to be under strong selection for stability, as even a slightly modified version of the compound is less likely to attract potential mates [44]. Usually, pheromones are produced in very small quantities, which could be a mechanism to avoid eavesdropping by natural enemies [45, 46].

But even in small quantities pheromones are often detectable at quite long distances and they are also often host specific, which would make them reliable cues for parasitoid host location. However, the presence of the preferred host stage is not necessarily indicated by adult pheromones [34].

Insect Immunity

Host defence strategies

In host insects there are a number of strategies developed in order to escape parasitism, such as using enemy-free space, concealment or physical counter- attack. In communities of social insects there are examples of external strategies to avoid spreading of pathogens, something often referred to as

“social immunity”. One example of external defence is the uptake of antifungal or antibacterial substances, such as the collection and use of plant resins by honey bees [47] or conifer saw flies [48]. Other strategies can be grooming to remove parasites from group members [49, 50], socially generated fevers to limit the proliferation of natural pathogens [51], detection and removal of infested and deceased individuals [52, 53] or relocation to abandon infested areas [54].

Even when successfully parasitized, the host insect can still defend itself through a potent immune defence. Insects only possess one level of immunity:

innate (or natural) immunity, which is present in both invertebrates and vertebrates, but lack the adaptive (or acquired) immunity, which is present only in vertebrates. Innate immunity refers to nonspecific defence mechanisms that start immediately or within hours after an antigen has appeared in the body. The innate immune system in insects consists of several different defence mechanisms, with both cellular and humoral contributions [55, 56].

(13)

Any organism trying to infest an insect first has to overcome the physical barrier of the insect cuticle. The cuticle can be divided into two layers, the outer epicuticle, which is a thin chitin-lacking layer, highly resistant to water and other solvents, and the inner thicker procuticle. The procuticle can again be divided into two layers, the outer exocuticle and the inner endocuticle, consisting of a large number of protein and chitin fibre layers, which creates a very tough and flexible substance [57]. Even if parasites or microorganisms enter the insect body cavity through the mouth, they still have to overcome physical barriers. The gut of most insects is lined by the peritrophic membrane, which is a grid-like structure composed of chitin and proteins [58].

Furthermore, ingested microorganisms activate the production of reactive oxygen species (ROS) in the insect midgut. ROS are multifunctional molecules involved in host defence, hormone biosynthesis, apoptosis, necrosis, and gene expression. The ROS response is strongly induced by microbial infections [59, 60].

The two major immune organs in insects are the fat body and the hemocytes.

The fat body is the largest organ of the insect body cavity, playing an essential role in nutrient storage and metabolism [61]. It is an endocrine organ unique to insects that has been studied extensively in Drosophila [62, 63], whereas there is still less knowledge of its specific functions in many other insects [64].

From the fat body, soluble effector molecules toxic to intruding parasites and pathogens are secreted into the open circulatory system [57, 65]. The soluble molecules recognize microbes and provide an early defence against pathogens present outside host cells. They can act either directly on the invader or by altering the insect’s immune response [66]. One type of soluble effectors are antimicrobial peptides (AMPs), small molecules (12-50 amino acids) with different target organisms (either bacteria or fungi). AMPs act by binding to bacterial or fungal membranes, which leads to disruption of the membrane and death of the cell [67]. The first induced AMP isolated from an insect was Cecropin, fully characterized by Boman and co-workers, in bacteria- challenged diapausing pupae of the moth Hyalophora cecropia [68]. Since then, over 150 AMPs have been characterized in insects, but the number and types vary between species [69]. Although most AMPs are secreted by the fat body, some are also produced by hemocytes [65, 70]. Other examples of processes mediated by soluble effectors are clotting and melanisation.

Clotting is the coagulation of hemolymph, leading to the rapid formation of a plug which seals wounds and keep bacteria from entering the body cavity, something that is especially important in the open circulatory system of insects [71]. Melanisation plays a central role in insect defence against a wide range of pathogens, participating in wound healing as well as in nodule and capsule formation. The melanisation pathway leads to the formation of melanin, which results in an immediate localized blackening of the tissue at the wound site or around an encapsulated object [70] (Fig. 1).

(14)

The cellular immune defence mechanisms in insects are directly mediated by the hemocytes, circulating freely in the insect hemolymph [72] or localized in specific hematopoietic organs. Until recently, hemocytes have been studied in most detail in Dipteran [73] and Lepidopteran [74] species, where several hemocyte classes have been characterised. The cellular defences include several mechanisms such as phagocytosis, encapsulation, clotting and nodulation. Phagocytosis is the process where intruding objects are engulfed and destroyed by individual, specialised hemocytes [75, 76]. Hemocytes can phagocytise a variety of invaders, such as fungi, yeast, bacteria and apoptotic bodies [72, 77]. Larger targets that cannot be engulfed by single cells, such as parasitoids or nematodes, are instead encapsulated by aggregating hemocytes (Fig. 1b-d). Following parasitoid attack, the capsule formation usually begins within 4-6 hours and is completed after approximately 48 hours [78]. During this process the capsule is often melanised [70] and due to crosslinking of its protein compounds it also hardens, which leads to the death of the encapsulated intruder [79, 80]. Another example of cellular defences, similar to the encapsulation process, is nodulation. Insect nodules are multicellular hemocytic aggregates, which are formed as a reaction against large numbers of bacteria or fungi [81, 82].

Figure 1. Melanisation and encapsulation in Galerucella. A) Melanised wound site in the cuticle of G. calmariensis. B) G. pusilla larva with encapsulated parasitoid eggs visible through the cuticle (arrow). C) Encapsulated and melanised parasitoid eggs inside G. pusilla larva. D) Encapsulated parasitoid egg dissected from G. pusilla, with layers of hemocytes attached to the surface. Photos: Robert Markus and Lisa Fors.

(15)

Recognition of intruding pathogens is a crucial aspect of immunity. This task is even more challenging when the pathogen is intracellular, as in viral infections of a host [83]. One effective immune response against intruding viruses is the destruction of infected cells by the tightly regulated process called apoptosis or programmed cell death. Apoptosis is mediated by caspases; proteolytic enzymes that are present in all cells as inactive precursors. Another important immune response towards viruses and other foreign genetic material is RNAi interference, first described in insects in the model nematode Caenorhabditis elegans [84, 85]. RNAi interference is a specific gene silencing process in which double-stranded RNA is used to inhibit gene expression, mainly through the degradation of specific mRNA molecules.

Parasitoid counter-defence strategies

Just like host insects have to develop defence mechanisms in order to avoid parasitism, parasitoids are under strong selection to evolve counter-defence strategies to protect their progeny from being rejected [86]. There is a variety of mechanisms used by parasitoids in order to manipulate their host and create a favourable environment for the developing parasitoid offspring [15].

Many parasitoid species have developed different methods to avoid the encapsulation of eggs or larvae, which is otherwise the most common and often successful defence against parasitoids. To avoid or reduce encapsulation, the parasitoid female can hide her eggs in an organ that is inaccessible to circulating hemocytes, such as the host brain, or produce eggs that can adhere to the host fat body, which protects them from being completely surrounded by hemocytes [86, 87]. Parasitoid eggs or larvae can also be “camouflaged” by a protective layer in order to mimic the host tissue, thereby preventing recognition by the host [36, 88]. Some parasitoid species can even survive being encapsulated, by modifying capsule formation. Many parasitic tachinid flies develop inside the host, but have spiracles armed with hooks that penetrate the cuticle or trachea of the host. If encapsulated, the parasitoid larva is not killed as long as it can use this respiratory funnel to avoid asphyxiation [89].

The encapsulation process can also be disrupted by active destruction of host immune cells that are required for capsule formation. Many parasitoids manipulate the host and promote parasitism by injecting venoms or symbiotic viruses into the host during oviposition [86, 90]. Injected substances can suppress the immune response, cause tissue necrosis, or paralyze the host.

(16)

Venoms used by parasitoids typically consist of one or more proteins that function by acting on nerve synapses or disrupting cell membranes. A common strategy in many parasitoids is to quickly sting the host to inject venom and then withdraw to let paralysis set in. Once the host is paralyzed the parasitoid returns to oviposit undisturbed [91]. Some parasitoid species inject venoms that keep the host alive but leave it immobile or incapable of moulting for weeks after the attack [36]. Parasitoids belonging to the Braconidae and Ichneumonidae families are known to form associations with viruses, which have received the name polydnaviruses (PDVs). The name refers to the arrangement of the viral genome, typically composed of 15-35 double- stranded DNA circles. PDVs act as delivery vectors of genes that disrupt humoral signalling pathways or hemocyte function, thereby altering the host’s immune defence, growth or development to promote the development of parasitoid offspring [15, 92]. Parasitoids from several families (in particular Braconidae, but also Scelionidae and Trichogrammatidae) produce special cells called teratocytes that are released into the host’s hemolymph by the serosa membrane surrounding the parasitoid embryo [15]. Teratocytes do not multiply once in the hemolymph, but they are often released in very large numbers and they continue to grow, sometimes dramatically, within the host body [93]. The main function of teratocytes is trophic; they absorb nutrients from the host hemolymph and are eventually consumed by the developing parasitoid larvae. However, the teratocytes can also secrete compounds that may manipulate the host or interfere with the immune response in order to favour parasitoid development [94].

Aim of the thesis

The overall aim of this thesis was to investigate interactions and possible coevolution in host-parasitoid systems, by combining ecological studies with chemical and cellular investigations. The studies have been carried out in a system consisting of Galerucella leaf beetles (Coleoptera: Chrysomelidae) and Asecodes parasitoids (Hymenoptera: Eulophidae). Specifically, I wanted to investigate chemical communication in the host species (Paper I), parasitoid search behaviour and host preference (Paper II) and parasitoid virulence in relation to the immune response in the attacked host (Paper III and IV). In the following part of the thesis, I will describe the study system and go through the methods and results of each study, followed by a general discussion of the combined results.

(17)

Methods

Study system and study area

There are five species of Galerucella included in the studies of this thesis: G.

calmariensis (L.), G. pusilla (Duftschmid), G. tenella (L.), G. sagittariae (Gyllenhal) and G. lineola (Fabricius) (Fig. 2). The differentiation of the species is fairly recent: G. pusilla and G. calmariensis are most closely related with an estimated divergence date 77 000 years ago [95]. G. tenella is also quite closely related to G. pusilla and G. calmariensis, whereas G. sagittariae and G. lineola are further apart in the phylogeny. While G. pusilla and G.

calmariensis are monophagous and share L. salicaria as exclusive host plant, the other species are polyphagous. G. tenella uses F. ulmaria as primary host, but can also be found on other Rosaceae species. G. sagittariae shares some host plant species with G. tenella, but also uses additional Rosaceae and Primulaceae species, whereas G. lineola uses different species of Salicaceae as host plants [95]. In many localities, several host plant species can be present within close distance, which means that several Galerucella species can sometimes co-occur in the same area. However, the geographic distribution of the beetles in Sweden differs a bit between the species. For example, G. pusilla is not present in the northern localities where G. calmariensis is found, even though G. pusilla and G. calmariensis often co-occur in the same localities in the south (often even on the same plant). For the studies of this thesis, beetles and larvae from the five Galerucella species were collected from various localities in Sweden each season of the experiments (see Fig. 3 for details).

Figure 2. Study system showing adults of the five Galerucella species and names of the attacking parasitoid species. A) G. calmariensis B) G. pusilla C) G. tenella D) G.

lineola and E) G. sagittariae. Photos: Robert Markus and Lisa Fors.

(18)

The beetle species all have similar life cycles, over-wintering as adults and emerging during spring. Mating takes place on the host plants and the eggs are deposited directly on the leaves or stem in early summer, hatching after a few weeks. Both larvae and adults feed on the plant, which can often lead to quite severe damage. After 3-4 weeks the larvae pupate in the ground and the new adults emerge from the pupae 2-3 weeks later [96].

Figure 3. Map of Sweden showing field localities used for collection of Galerucella adults and larvae. G. pusilla is not present north of the dashed line. The species were collected from the following localities, indicated by numbers: G. calmariensis (1, 3- 11, 23-26), G. pusilla (2, 4-11, 21-31), G. tenella (1, 8-12, 14-22, 24, 26, 27, 29), G.

lineola (8-12, 14, 15, 23, 24, 27, 29) and G. sagittariae (8-11, 13, 23, 24, 26-28).

(19)

The Galerucella spp. are attacked by three species of Asecodes (Hymenoptera:

Eulophidae): Asecodes parviclava (Thomson), A. lucens (Nees) and A.

lineophagum Hansson & Hambäck. Genetic studies of Asecodes suggest that speciation of the parasitoid has followed the host speciation [95]. The Asecodes parasitoids are small (<1mm) and morphologically similar but each species has a unique host range. Asecodes parviclava parasitizes G.

calmariensis, G. pusilla and G. tenella, whereas A. lucens parasitizes only G.

sagittariae and A. lineophagum only G. lineola [97] (Fig. 2). Asecodes are koinobiont endoparasitoids that can cause a high level of mortality in respective host species. The parasitoids attack the beetles in the larval stage, laying one or more eggs inside the larva. When the eggs hatch, the parasitoid larvae start consuming the interior of the host. Parasitized larvae develop normally until pupation, when they are unable to form pupae. Instead the larvae turn into black mummies from which the adult parasitoids subsequently hatch (Fig. 4), usually during the next summer [96, 98].

Figure 4. Development stages in Asecodes parasitoids. A) Parasitoid larvae dissected from an infected G. calmariensis larva. B) Live parasitoid larva. C) Parasitized, mummified G. calmariensis larva showing pupating parasitoids inside. D) Parasitoid pupae. E) Adult parasitoid (A. parviclava). Photos: Robert Markus and Lisa Fors.

Chemical communication and host search (Paper I and Paper II)

Earlier studies have shown that males of G. pusilla and G. calmariensis both emit the same aggregation pheromone (dimethylfuran-lactone) when feeding on their host plant [99, 100]. In other Galerucella species no pheromones have previously been identified. Based on these previous observations, I wanted to further investigate the pheromone production in Galerucella, and also study the behavioural responses in the beetles to different search cues (Paper I).

(20)

Additionally, I wanted to find out more about the search behaviour in Asecodes, and whether the parasitoids can exploit the pheromone of the beetles as a search cue when locating a host (Paper II).

Paper I

The first part of this study was to investigate the production of and the response to pheromones in Galerucella. In this part all five beetle species were included: G. calmariensis, G. pusilla, G. tenella, G. sagittariae and G. lineola (Fig. 2). As it had previously been reported that G. pusilla and G. calmariensis produce the pheromone dimethylfuran-lactone only when feeding on their host plant (Bartelt et al. 2006), the first step was to collect volatile compounds from beetles of all five species feeding on their respective host plant. Thus, volatiles were collected of G. calmariensis and G. pusilla feeding on L.

salicaria, G. tenella and G. sagittariae feeding on Fragaria x ananassa and G. lineola feeding on Salix viminalis. To distinguish the compounds produced by the beetles from those produced by the plants, half of the plants for each species pair (beetle-host plant) were mechanically damaged and the other half were damaged by feeding beetles (10 beetles/plant). There was also an additional set-up with larvae instead of beetles. The plants (with or without beetles or larvae) were then enclosed in polyester cooking bags and the volatiles were collected during 24 hours using solid phase micro extraction (SPME), which is a sampling technique where a polymer-coated fibre is used to absorb analytes. After collection the chemical compounds were separated and analysed by using a gas chromatograph-mass spectrometer (GCMS).

The next step was to test whether the beetles showed any attraction to the pheromone. For this experiment, the pheromone was produced synthetically, using a newly developed and improved method. The behavioural responses were studied in two-armed olfactometers (see [101] for description), with a cut out arena in the middle where the beetle was introduced. A small rubber dispenser was placed in each arm of the olfactometer, one loaded with 100 µg of synthetic dimethylfuran-lactone diluted in 50 µl hexane, and the other loaded with 50 µl hexane (serving as control). The beetle’s position in the arena was recorded every minute for a period of 30 min.

In the second part of the study the behavioural responses towards blends of pheromone and respective host plant were investigated. This part included two of the beetle species: G. pusilla (host plant: L. salicaria) and G. tenella (host plant: F. ulmaria). The experimental procedure was the same as in the first trial, but now the responses towards different odour combinations were tested.

In the first treatment there was a choice between a blend of pheromone and host plant odour vs the pheromone alone and in the second treatment there

(21)

was a choice between the same blend vs host plant odour alone. The amount of dimethylfuran-lactone was 100 µg diluted in 50 µl hexane for all combinations. For treatments with plants, 10-15 cm branches of L. salicaria and F. ulmaria were used. A few leaves on the branches were mechanically damaged prior to the test, as there is usually a stronger response to damaged compared to non-damaged plants.

Paper II

The aim of this study was to investigate the behavioural responses to different host cues in A. parviclava, the parasitoid species attacking G. calmariensis, G. pusilla and G. tenella (Figs. 4 and 5). First, the response to the pheromone dimethylfuran-lactone, produced by males of G. calmariensis, G. pusilla and G. tenella (Paper I) was investigated, in order to find out if A. parviclava can exploit the pheromone as a host cue kairomone. In this part of the study parasitoid females hatching from all three host species were included. The behavioural responses of A. parviclava were studied in two-arm Y-tube olfactometers, with an airflow of approximately 30 ml/s. Each arm of the Y- tube was connected to a gas bottle, where a small rubber dispenser was placed.

In the first arm the dispenser was loaded with 100 g synthetic pheromone (dimethylfuran-lactone) diluted in 50 l hexane and in the other arm the dispenser was loaded with 50 l hexane (as a control). The parasitoids were introduced to the Y-tube and given a couple of minutes to acclimatize without airflow. Each parasitoid was observed for 5 min or until making a decisive choice, which was recorded if the parasitoid passed two-thirds of an arm of the Y-tube and stayed there for at least 5 seconds.

The second part of the study was to investigate the ability of the parasitoids to distinguish between larvae of the different host species, based only on odour cues. Due to insufficient numbers of parasitoids from G. pusilla, only parasitoids from G. calmariensis and G. tenella were included in this part. The general set-up was the same, using two-armed Y-tube olfactometers, but each gas bottles was now loaded with a 10-15 cm host plant branch with 4 feeding larvae (G. calmariensis and G. pusilla with L. salicaria and G. tenella with F.

ulmaria). Three different combinations were used: i) G. calmariensis vs G.

tenella larvae, ii) G. calmariensis vs G. pusilla larvae and iii) G. tenella vs G.

pusilla larvae.

(22)

Host immune response and parasitoid virulence (Paper III and Paper IV)

Previous observations in the Galerucella-Asecodes system have shown differences in parasitism rates between localities, with larvae in northern localities generally showing a much higher parasitism rate (>70%) than larvae in southern localities (<10%) [102], Hambäck, unpublished data. There have also been observations of differences in parasitism rates between the species [102, 103], indicating that G. pusilla (that does not occur in the north) experiences a lower parasitism rate than the other two species. Based on these observations, I wanted to investigate the structure of the immune system in Galerucella (Paper III) and to detect whether the differences in the level of parasitism were due to differences in the immune response in the three species (Paper III and IV) or were caused by geographic variation in the ability of A.

parviclava to infect the hosts (Paper IV). I also wanted to investigate the possibility that parasitoids hatching from one host species would have a higher success when attacking larvae of the same species (Paper IV).

Paper III

In the first study (Paper III), only G. calmariensis and G. pusilla were included. The A. parviclava used in the experiments all derived from northern populations of G. calmariensis, as parasitoid abundance was much higher in this area. The study was started by performing controlled parasitism experiments, where laboratory-reared larvae of each species were put together with a fixed number of A. parviclava females. After 24 h the parasitoids were removed and the larvae were examined in a stereo microscope to detect melanisation of wound sites (black dots) in the cuticle that would indicate parasitoid attack. 96 h later the larvae were dissected in order to find out whether they were successfully parasitized (containing live parasitoid larvae) or showing a successful immune response (containing exclusively melanised eggs).

The next step was to investigate the cell composition of the larvae, to find out if there were any detectable differences between the species at the cellular level. In connection to the dissections, hemolymph samples were prepared from all larvae used in the parasitism experiments, as well as from non- infested larvae reared in the laboratory and from larvae collected in the field.

To begin with, we had some troubles to establish a well-functioning method for hemocyte preparation, as we found that the hemocytes of Galerucella larvae rupture when in contact with air and the hemolymph coagulates very quickly. To avoid hemolymph clotting the larvae were dissected submerged in a well containing PBS mixed with a small amount of phenylthiourea (PTU).

(23)

All hemolymph samples were stained with blue-fluorescent nucleic acid stain (DAPI) to reveal the nuclei in the cells. The samples were then studied in a phase contrast, epifluorescent microscope connected to a Hamamatsu camera with Axio Vision 4.6. Nine random images were taken from each individual.

As there were no previous cytological studies in these species, the first step was to classify the different cell types in the hemolymph. Based on this classification, differential hemocyte counts were performed for naïve and parasitized larvae of both species. To identify phagocytic cells, fluorescently labelled bacteria (E. coli) were injected into live larvae of both beetle species 30 min prior to dissection. To investigate which cell types were involved in the encapsulation process, live and encapsulated eggs from infested Galerucella were permeabilised with Triton-X and incubated with Phalloidin and DAPI diluted in PBS.

Figure 5. Study species. A) The parasitoid A. parviclava attacking larvae of B) G.

calmariensis C) G. pusilla and D) G. tenella. Photos: Robert Markus and Lisa Fors.

Paper IV

In the second immunological study (Paper IV), G. calmariensis, G. pusilla and G. tenella were included, as well as A. parviclava originating from all three host species (Fig. 5). This study had two aims: to investigate possible geographic variation in parasitoid virulence and host immune response in the Galerucella-Asecodes system, and to find out whether the former host species of the parasitoid might have an effect on future parasitism success. As in Paper III, the study was carried out by combining controlled parasitism experiments with an investigation of the cell composition in the larval hemolymph. The general set-up for the parasitism experiments was the same as in Paper III, but in this study there was a distinction between parasitoids hatching from the

(24)

three host species, as well as between northern and southern populations of both parasitoids and larvae. As far as possible, Galerucella larvae from all three species and from both geographic areas (with an exception for G. pusilla that is only present in the south) were parasitized with A. parviclava hatching from all host species deriving from both geographic areas. However, there were some gaps not possible to fill in the scheme of combinations, due to insufficient numbers of parasitoids from southern populations of G.

calmariensis and G. tenella, as well as low numbers of G. tenella larvae. The dissections and preparations of hemocyte samples were performed in the same way as in Paper III, but only individuals that proved to be infected at dissection were included in the cellular study.

(25)

Results and discussion

The first study, concerning chemical communication in the beetle host, resulted in the finding that G. tenella both produces and responds to dimethylfuran-lactone, the male aggregation pheromone previously observed in G. pusilla and G. calmariensis [99], whereas G. lineola and G. sagittariae were not found to produce or respond to the same pheromone (Paper I). Due to these results, G. lineola and G. sagittariae were not included in the behavioural study with pheromone and host plant odours. Unfortunately, G.

calmariensis also had to be excluded from this experiment, since the number of beetles of this species available at the time of the experiment was too low for meaningful analyses. The study showed that in the choice between the blend and only the pheromone, males of G. pusilla preferred the blend. In G.

tenella, it was instead the females that preferred the blend over the pheromone alone. In the choice between the blend and only the host plant, both sexes of both species were significantly attracted to the blend.

It is likely that males of G. lineola and G. sagittariae produce other pheromones even though they were not detected in this study. That the pheromone was produced by G. tenella and not by G. lineola and G.

sagittariae seems fairly logical, since G. tenella is the species most closely related to G. pusilla and G. calmariensis [95]. This suggests that pheromone production and response may be connected to the phylogenetic relatedness between the Galerucella species. The result was particularly interesting as the three species producing the pheromone are all attacked by the same parasitoid (A. parviclava), whereas G. lineola and G. sagittariae are attacked by two separate Asecodes species. This led to the idea that A. parviclava might exploit the adult host pheromone as a host cue kairomone. However, the results of the following study suggested that the parasitoid females are unable to use the pheromone to locate host larvae (Paper II). The lack of response to the pheromone in A. parviclava could possibly be due to the fact that the parasitoids attack the larval stage, which makes the male pheromone a less reliable cue for locating a suitable host. Kairomones emitted by host adults are commonly used by egg parasitoids, but more rarely by larval or pupal parasitoids [34, 104, 105]. However, there are some studies showing that both egg and larval parasitoids use pheromones for host habitat location and then reside in the neighbourhood until the preferred host stage is available for egg- laying [34, 106-108]. In Galerucella there is usually a period of 2-3 weeks

(26)

from pheromone emittance until the beetle larvae emerge, but there is occasionally an overlap where adult beetles and larvae can be found concurrently on the same plant, which could promote the idea of the adult pheromone as a search cue.

The next part in Paper II suggested that A. parviclava hatching from both G.

calmariensis and G. tenella can detect hosts based on odour cues from the host larvae and that they also have the ability to distinguish between host species from a distance. These results were even more interesting in the light of the results from the immunological studies, where larvae of G. pusilla were found to have a much more potent immune defence towards A. parviclava than larvae of G. tenella and G. calmariensis (Paper III and Paper IV).

Furthermore, parasitoids from both G. calmariensis and G. tenella were found to have much higher success rates when attacking larvae of their former respective host than larvae of G. pusilla (Paper IV). The results from Paper II revealed that parasitoids from both G. calmariensis and G. tenella have a preference for their former respective host species over the well defended host G. pusilla. Parasitoids from G. calmariensis also had an ability to distinguish larvae of G. tenella from larvae of G. pusilla, whereas parasitoids from G.

tenella did not distinguish between larvae of G. calmariensis and G. pusilla.

Notable is however that all A. parviclava individuals used in this study derived from northern localities, where G. pusilla is not present. Thus, a positive attraction for G. calmariensis and G. tenella larvae must have evolved in A.

parviclava in the north, since there cannot be any selection pressure on these parasitoids to avoid larvae of G. pusilla.

At this point we have no information on which specific odour cues Asecodes uses for host search and host selection, although our study indicates that the most important cue is likely to be produced by the larvae. Some chemical analysis have been performed on the odour from feeding larvae, but so far no key differences have been found. Furthermore, the study does not show whether A. parviclava can detect larval odours from a further distance. One possibility is that the parasitoids use different cues to first locate the host habitat and larval odours only when in closer range of the host. Even though no response to the adult pheromone was observed in this study, it is still possible that A. parviclava could respond to the pheromone in combination with odours from the correct host plant. No such tests were performed (due to low numbers of parasitoids) but it is something that could be worth investigated further. A. parviclava has previously been shown to respond to green leaf volatiles released from damaged L. salicaria and F. ulmaria, with a strong preference for F. ulmaria [103]. However, in the study by Stenberg et al., there was no distinction between parasitoids from the two species, which makes it unclear whether the preference for F. ulmaria is true for A. parviclava from both G. calmariensis and G. tenella. Further studies on search behaviour

(27)

in A. parviclava should preferably include also parasitoids from G. pusilla and parasitoids from southern populations of G. calmariensis and G. tenella.

The immunological studies (Paper III and Paper IV) revealed large differences in the level of successful immune response in the three host species G. pusilla, G. calmariensis and G. tenella. Larvae of G. pusilla showed a much more potent immune defence towards A. parviclava than the other two species, and G. tenella showed a stronger defence than G. calmariensis (although still much weaker than G. pusilla). This was in accordance with previous field observations, suggesting that G. pusilla in general experiences a lower parasitism rate than G. calmariensis and G. tenella. In the first study (Paper III) where only G. pusilla and G. calmariensis were included, the difference between the species was striking; no infected larvae of G. calmariensis managed to suppress the parasitoid attack whereas all infected G. pusilla showed a successful immune response. One thing that could potentially influence the results of this study, was the fact that only parasitoids deriving from G. calmariensis were used for the parasitism experiments. This suggested an adaptation in the parasitoids to the former host species (G.

calmariensis), that would potentially explain the low parasitism success in G.

pusilla.

To a large extent, this idea proved to be true in the following study (Paper IV), where G. tenella was included, as well as parasitoids deriving from all three host species. In Paper IV it was clear that former host species of the parasitoid had an effect on parasitism success. Accordingly, parasitoids with G. pusilla as former host had much higher success rates when attacking G. pusilla larvae than parasitoids from the other two species. Parasitoids from G. tenella also had much higher success rates when attacking G. tenella larvae than parasitoids from G. calmariensis. However, parasitoids from G. pusilla was almost equally successful as parasitoids from G. tenella when attacking G.

tenella larvae. When combining data from all parasitism experiments (regardless of parasitoid origin), the results showed an overall strong immune defence in larvae of G. pusilla, a somewhat intermediate defence in G. tenella and a very poor defence in G. calmariensis (Paper III and Paper IV). It is likely that parasitoids from G. pusilla have developed an ability to overcome the strong defence of the host and thereby become more effective also when infecting larvae of the other two species. Accordingly, parasitoids from G.

calmariensis are generally less successful when attacking larvae of both G.

pusilla and G. tenella, as they developed in the host with the overall poorest immune defence.

Although there was strong evidence that G. pusilla had a more potent immune response than G. calmariensis and G. tenella, we did not know the underlying cause for this difference, which led to further investigations on the cellular

(28)

level. Based on morphological characteristics, six types of hemocytes were distinguished in non-infested and infested individuals: granulocytes, phagocytes, prohemocytes, oenocytoids, lamellocytes and lamellocyte precursors (Fig. 6).When studying the parasitoid eggs in the microscope, no cell activity was found on the live eggs in G. calmariensis, whereas several layers of cells were attached to the surface of the encapsulated eggs in G.

pusilla and G. tenella. The cytological studies revealed that the successful encapsulation of parasitoid eggs in Galerucella involves at least three different cell types: lamellocytes, phagocytes and granulocytes (Paper III).

Granulocytes could be detected by their autofluorescence in the green channel, lamellocytes by phalloidin staining and phagocytes by engulfed fluorescent bacteria. Large clusters of granulocytes were observed surrounding the melanised capsules, as well as granulocyte content on the surface of the eggs, suggesting cell rupture. When dissecting larvae in paraformaldehyde (PFA), we found that the rupture of the granulocytes takes approximately 50 ms, indicating that it is a rapid way to deliver the content of the cell to the wound.

Thus, we suggest that granulocytes in Galerucella, similar to crystal cells in Drosophila [109], ruptures and delivers its cargo locally at the wound or infection site.The results in Paper III indicated that lamellocytes are crucial for the capsule formation to be completed, something that is supported by previous findings in Drosophila melanogaster, where lamellocytes are essential for encapsulation of parasitoid eggs [110]. Phagocytes also contribute to the encapsulation in D. melanogaster, which further supports our observations in Galerucella. We believe that also the oenocytoids may have an active role in the immune defence, even if we have not yet been able to reveal their specific function in Galerucella (Paper III and Paper IV).

Oenocytoids have been shown to participate in the melanisation process in other species [111, 112].

Figure 6. Morphology of hemocytes in Galerucella. A) Granulocytes B) Phagocytes C) Prohemocyte D) Oenocytoid E) Lamellocyte F) Lamellocyte precursors. Cell nuclei are stained with DAPI (blue). Photos: Robert Markus and Lisa Fors.

The differential hemocyte counts showed that the hemocyte ratios differ between the species and that there is strong connection between hemocyte composition and the ability of the larva to mount an effective immune response against A. parviclava (Paper III and Paper IV). The results in Paper III revealed significant differences in hemocyte ratios between naïve and parasitized individuals, indicating a cellular response. In particular, the levels

(29)

of the cell types shown to be active in the encapsulation process, phagocytes, lamellocytes and granulocytes, changed upon infection. Lamellocytes and phagocytes were found to increase in numbers following parasitoid infection, whereas granulocytes decreased. There were also significant differences in hemocyte composition between the two species (G. pusilla and G.

calmariensis), both in naïve and parasitized individuals, reflecting their ability to defend against the parasitoid. The strong defence in G. pusilla was connected to high levels of lamellocytes and phagocytes. In general, the same pattern was found in the following study (Paper IV), but here only infested individuals from G. pusilla and G. tenella were included, as the immune response in G. calmariensis did not differ depending on former parasitoid host species or geographic origin. Larvae of G. pusilla showed higher levels of both lamellocytes and phagocytes when infected by parasitoids from G.

calmariensis or G. tenella, towards which the strongest immune response was observed, than when infected by parasitoids from G. pusilla. Accordingly, larvae of G. tenella showed a much higher level of lamellocytes when infected by parasitoids from G. calmariensis, towards which the strongest immune response was observed, whereas the level of phagocytes was equally high in larvae infected by parasitoids from G. calmariensis or G. tenella.

Interestingly, the constitutive level of oenocytoids was found to be much higher in G. tenella than in G. calmariensis and G. pusilla, indicating that the cellular immune defence might be somewhat different in this species.

Moreover, the levels of oenocytoids were particularly high in G. tenella individuals infected with parasitoids from G. calmariensis (towards which the strongest immune response was seen), indicating a cellular response (Paper IV). In some of the infected G. tenella larvae we also observed a type of large cell structures which had not been seen previously in Galerucella. The structures contained several nuclei and had the appearance of long ribbons of connected cells (Paper IV). These cells we believe to be similar to the multinucleated giant hemocytes that were recently described in Drosophila, where they have a function in parasite elimination [113]. However, as no functional tests were performed on the multinucleated cells observed in G.

tenella, we were unable to detect their potential role in the immune response.

Clearly, the very poor immune defence in G. calmariensis is reflected by the cell composition in both naïve and parasitized larvae. However, there are some things worth to keep in mind connected to the immune response. Even though the cell ratios differ between the species, all cell types are shown to be present in G. calmariensis (Paper III). This means that the poor immune defence cannot simply be due to the lack of a crucial cell type, i.e. the lamellocytes.

When going through hemocyte samples from numerous individuals, we have occasionally come across samples with fairly high ratios of lamellocytes also in G. calmariensis, suggesting some variation within the species. Further, both

(30)

melanisation and encapsulation are possible processes in G. calmariensis.

Melanisation at the wound site is very common in G. calmariensis, in fact it is often more easily detected than in G. tenella or G. pusilla. We have also observed proper capsule formation and melanisation of the parasitoid eggs in a few G. calmariensis individuals (both from the parasitism experiments and from the field), even though the encapsulation is very rare compared to what is observed in the other species.

There are also other aspects to take into account regarding the defence towards the parasitoid. A strong immune defence is costly on two levels; one is the cost of the actual defence after attack and one is the cost of having the ability to mount a successful immune response [86]. Many studies have shown life history trade-offs between growth, reproduction and immune competence in insects [20, 114]. We know from previous studies that larvae of G.

calmariensis has a higher growth rate (15%) than G. pusilla in the field and the adults are also larger in size [96]. The possible result of a high larval growth rate is a shorter exposure of G. calmariensis larvae to parasitoids, and consequently a lower risk for parasitism.

Even though the cellular immune response in G. calmariensis does not protect it against the parasitoid, the larva is not completely defenceless. When monitoring Galerucella larvae in parasitism experiments in the laboratory, we have often observed a type of behaviour that resembles the active defence seen in other parasitized species, for example the European pine sawfly, Neodiprion sertifer [115]. If attacked by a parasitoid, N. sertifer larvae can be seen vigorously flipping the upper part of the body in order to scare the parasitoid away. In connection to this, the larvae sometimes regurgitate part of the gut contents in sticky droplets. Galerucella larvae often show a similar behaviour when encountered by Asecodes parasitoids; they flip or twist both head and tail, sometimes in combination with blowing sticky bubbles from the mouth (Fig. 7). Whether this behaviour differs between the Galerucella species is not investigated, but so far we have mainly observed it in larvae of G. calmariensis.

Figure 7. Active defence mechanism in G. calmariensis when attacked by A.

parviclava. Photo: Robert Markus.

(31)

An additional aim with Paper IV was to investigate geographical variation in parasitism success. Due to missing combinations in the parasitism experiments the geographic effect on the immune response was a bit more difficult to interpret. Previous field observations have shown that parasitism rates on G. calmariensis are much higher in the north (>70%) than in the south (<10%). Due to the results of Paper III, we initially suspected that this difference was due to differences in the strength of the immune response.

However, as revealed in Paper IV, G. calmariensis has a very weak immune defence regardless of geographic origin. In G. tenella, there was an indication of a stronger defence in larvae from northern compared to southern populations towards parasitoids from G. tenella and G. calmariensis, but no significant differences could be detected. The immune defence in G. pusilla, on the other hand, was found to be affected by the geographic origin of the parasitoids, which suggests geographic differences in parasitoid virulence.

Larvae of G. pusilla (which only occurs in the south of Sweden) showed a weaker defence towards parasitoids deriving from southern localities of G.

calmariensis than towards parasitoids deriving from northern localities. In larvae of G. tenella, the effect of the parasitoids’ geographic origin on the immune defence was hard to interpret due to missing test combinations. Thus, it would be very interesting to further investigate the geographic variations of immune defence and parasitoid virulence in this system, especially in connection to further experiments on host preference in the parasitoid.

(32)

Concluding remarks

This thesis shows the importance of gathering information from different fields, in order to better understand the interactions and evolution in natural host-parasitoid systems. In particular, through combining studies of behavioural ecology with chemical ecology and molecular biology, a lot of knowledge can be gained, linking host selection to parasitoid virulence and host immune response. In the studies performed, I have found that there are large differences in immune competence between the Galerucella species, which can be linked to differences in parasitism rates observed in the field.

The differences in immune defence closely correspond to the hemocyte composition in the species. Further, the results suggest that parasitism success in A. parviclava is strongly affected by former host species of the parasitoid.

In connection to this I have also observed a potential ability in A. parviclava to select a host with weaker immune response from a distance. Taken together, the results of this thesis suggest that there is an on-going evolution in both parasitoid virulence and host immune responses in this system. Although many questions remain to be answered, the Asecodes-Galerucella system has proven to be a useful model system for investigating processes that may lead to host race formation and speciation in host-parasitoid systems.

Acknowledgements

I want to thank Peter Hambäck and Ulrich Theopold for constructive comments on earlier versions of this text, Robert Markus for help with editing the photos and Mathilda Arnell for help with the map of field localities.

(33)

References

1. Thompson, J.N. 1994 The coevolutionary process. Chicago, The University of Chicago Press.

2. Vazquez, D.P., Chacoff, N.P. & Cagnolo, L. 2009 Evaluating multiple determinants of the structure of plant-animal mutualistic networks. Ecology 90, 2039-2046. (doi:10.1890/08-1837.1).

3. Vazquez, D.P., Lomascolo, S.B., Belen Maldonado, M., Chacoff, N.P., Dorado, J., Stevani, E.L. & Vitale, N.L. 2012 The strength of plant-pollinator interactions.

Ecology 93, 719-725.

4. Thompson, J.N. 2009 The Coevolving Web of Life. The American Naturalist 173, 125-140. (doi:10.1086/595752).

5. Kniskern, J. & Rausher, M.D. 2001 Two modes of host-enemy coevolution.

Population Ecology 43, 3-14. (doi:10.1007/pl00012012).

6. Biere, A. & Verhoeven, K.J.F. 2008 Local adaptation and the consequences of being dislocated from coevolved enemies. New Phytologist 180, 265-268.

(doi:10.1111/j.1469-8137.2008.02632.x).

7. Kawecki, T.J. & Ebert, D. 2004 Conceptual issues in local adaptation. Ecology Letters 7, 1225-1241. (doi:10.1111/j.1461-0248.2004.00684.x).

8. Hereford, J. 2009 A Quantitative Survey of Local Adaptation and Fitness Trade- Offs. The American Naturalist 173, 579-588. (doi:10.1086/597611).

9. Gandon, S. & Michalakis, Y. 2002 Local adaptation, evolutionary potential and host-parasite coevolution: interactions between migration, mutation, population size and generation time. Journal of Evolutionary Biology 15, 451-462.

(doi:10.1046/j.1420-9101.2002.00402.x).

10. Greischar, M.A. & Koskella, B. 2007 A synthesis of experimental work on parasite local adaptation. Ecology Letters 10, 418-434. (doi:10.1111/j.1461- 0248.2007.01028.x).

11. Kaltz, O. & Shykoff, J.A. 1998 Local adaptation in host-parasite systems.

Heredity 81, 361-370.

12. Cogni, R. & Futuyma, D.J. 2009 Local adaptation in a plant herbivore interaction depends on the spatial scale. Biological Journal of the Linnean Society 97, 494- 502.

13. Godfray, H.C.J. & Shimada, M. 1999 Parasitoids as model organisms for ecologists. Researches on Population Ecology 41, 3-10.

(doi:10.1007/pl00011980).

14. Pennacchio, F. & Strand, M.R. 2006 Evolution of developmental strategies in parasitic hymenoptera. Annual Review of Entomology 51, 233-258.

(doi:10.1146/annurev.ento.51.110104.151029).

15. Beckage, N.E. & Gelman, D.B. 2004 Wasp parasitoid disruption of host development: Implications for new biologically based strategies for insect control. Annual Review of Entomology 49, 299-330.

(doi:10.1146/annurev.ento.49.061802.123324).

(34)

16. Santos, A.M., Serrano, J.C., Couto, R.M., Rocha, L.S.G., Mello-Patiu, C.A. &

Garofalo, C.A. 2008 Conopid Flies (Diptera: Conopidae) Parasitizing Centris (Heterocentris) analis (Fabricius) (Hymenoptera: Apidae, Centridini).

Neotropical Entomology 37, 606-608. (doi:10.1590/s1519- 566x2008000500018).

17. Askew, R.R., Shaw, M. R. 1986 Parasitoid communities: their size, structure and development. In Insect Parasitoids, 13th Symposium of Royal Entomological Society of London (eds. J. Waage & D. Greathead), pp. 225-264. London, Academic Press.

18. Quicke, D.L.J. 2015 The Braconid and Ichneumonid Parasitoid Wasps: Biology, Systematics, Evolution and Ecology. Chichester, John Wiley & Sons, Ltd.

19. Henter, H.J. & Via, S. 1995 The Potential for Coevolution in a Host-Parasitoid System I. Genetic Variation within an Aphid Population in Susceptibility to a Parasitic Wasp. Evolution 49, 427-438. (doi:10.2307/2410267).

20. Schulenburg, H., Kurtz, J., Moret, Y. & Siva-Jothy, M.T. 2009 Ecological immunology. Philosophical Transactions of the Royal Society B-Biological Sciences 364, 3-14. (doi:10.1098/rstb.2008.0249).

21. Miller, D.R., Strickler, K.L. 1984 Finding and accepting host plants. In Chemical Ecology of Insects (ed. W.J. Bell, Carde, R.T.), pp. 127-157. London, Chapman

& Hall.

22. Prokopy, R.J. 1986 Visual and olfactory stimulus interaction in resource finding by insects. In Mechanisms in insect olfaction (ed. T.L. Payne, Birch, M.C., Kennedy, C.E.J.), pp. 81-89. Oxford, Oxford University Press.

23. Schoonhoven, L.M., van Loon, J.J.A. & Dicke, M. 2005 Insect-Plant Biology. 2 ed. New York, Oxford University Press.

24. Morris, W., Grevstad, F. & Hertzig, A. 1996 Mechanisms and ecological functions of spatial aggregation in chrysomelid beetles. In Chrysomelidae Biology: Ecological Studies (eds. P.H.A. Jolivet & M.L. Cox), pp. 303-322.

Amsterdam, SPB Academic Publishing.

25. Visser, J.H. 1986 Host odour perception in phytophagous insects. Annual Review of Entomology 31, 121-144.

26. Dicke, M., Van Loon, J.J.A. & Soler, R. 2009 Chemical complexity of volatiles from plants induced by multiple attack. Nature Chemical Biology 5, 317-324.

(doi:10.1038/nchembio.169).

27. Salt, G. 1935 Experimental studies in insect parasitism III. Host selection.

Proceedings of the Royal Society B: Biological Sciences 117, 413-435.

28. Doutt, R.L. 1959 The biology of parasitic Hymenoptera. Annual Review of Entomology 4, 161-182. (doi:10.1146/annurev.en.04.010159.001113).

29. Vinson, S.B. 1976 Host selection by insect parasitoids. Annual Review of Entomology 21, 109-133. (doi:10.1146/annurev.en.21.010176.000545).

30. Cade, W.H. 1984 Effects of fly parasitoids on nightly calling duration in field crickets. Canadian Journal of Zoology-Revue Canadienne De Zoologie 62, 226- 228.

31. Cade, W.H. & Wyatt, D.R. 1984 Factors affecting calling behavior in field crickets, Teleogryllus and Gryllus (age, weight, density, and parasites).

Behaviour 88, 61-75. (doi:10.1163/156853984x00489).

32. Glas, P.C.G. & Vet, L.E.M. 1983 Host-habitat location and host location by Diachasma alloeum Muesebeck (Hym, Braconidae), a parasitoid of Rhagoletis pomonella Walsh (Dipt, Tephritidae). Netherlands Journal of Zoology 33, 41-54.

33. Vinson, S.B. 1981 Habitat location. In Semiochemicals: their role in pest control (eds. D.A. Nordlund, R.L. Jones & W.J. Lewis), pp. 51-77. New York, John Wiley & Sons.

References

Related documents

6 summarizes the effects of LL-37 on bacterial phagocytosis, intracellular bacteria killing, and eicosanoids production in human macrophages, as well as the

We observed that the pneumococcal pore forming toxin pneumolysin had profound effects on cytokine responses in human dendritic cells and macrophages.. We found a cell death

When combining data from all infected individuals independent of parasitoid origin, larvae of G. However, the immune responses were affected by parasitoid origin. Galerucella

5 All 17 type I IFNs signal through the IFN alpha receptor (IFNAR) and induce a type I IFN gene signature (IFNGS), which is present in a proportion of patients with SLE and

We added 25 different bacterial strains to pharyngeal, gastric, intestinal or cervical epithelial cells, depending on the site of colonization within the human body,

(2016) Lactobacilli reduce Helicobacter pylori attachment to host gastric epithelial cells by inhibiting adhesion gene expression.. (2019) Lactobacillus gasseri

Since pDCs play a crucial role in triggering humoral immunity (12,13) increased pDC responses in Fulani may support antibody production which could explain the higher Ig levels

• No correlation can be found between the microsclerotia density in soils and disease incidence, but it is possible to predict yield of winter oilseed rape based