key: cord-0034966-x5n4dyxp
authors: Hatcher, Melanie J.; Dick, Jaimie T. A.; Paterson, Rachel A.; Alexander, Mhairi E.; Bunke, Mandy; Dunn, Alison M.
title: Trait-Mediated Effects of Parasites on Invader-Native Interactions
date: 2015-07-20
journal: Host Manipulations by Parasites and Viruses
DOI: 10.1007/978-3-319-22936-2_3
sha: aa4a8bad5bd6d82204fb44944ca896cd7e77a27c
doc_id: 34966
cord_uid: x5n4dyxp

Parasites have a variety of behavioural effects on their hosts, which can in turn affect species with which the host interacts. Here we review how these trait-mediated indirect effects of parasites can alter the outcomes of invader-native interactions, illustrating with examples from the literature and with particular regard to the invader-native crustacean systems studied in our laboratories. Parasites may potentially inhibit or exacerbate invasions via their effects on host behaviour, in addition to their direct virulence effects on hosts. In several crustacean systems, we have found that parasites influence both host predation rates on intra- and inter-guild prey and host vulnerability to being preyed upon. These trait effects can theoretically alter invasion impact and patterns of coexistence, as they indirectly affect interactions between predators and prey with the potential for further ramifications to other species in the food web. The fitness consequences of parasite-induced trait-mediated effects are rarely considered in traditional parasitological contexts, but demand attention in the context of ecological communities. We can regard these trait effects as a form of cryptic virulence that only becomes apparent when hosts are examined in the context of the other species with which they interact.

Most work on parasite-induced behavioural change in hosts focuses on the mechanisms underlying these changes and their direct consequences for parasite and host fi tness (see other chapters in this volume). However, infected hosts do not exist in isolation; they form part of a community of organisms that interact through an array of trophic links. Here, we broaden the focus of parasite-modifi ed behaviour to examine the indirect consequences of such behavioural changes for species that interact with the host, and explore how this may impact community structure. In particular, we review our work on crustacean host-parasite systems where we fi nd that parasites affect a range of trophic interactions between invasive and native host species. We examine the potential signifi cance of these parasite-induced behavioural effects in the biological invasion process.

From a population perspective, we usually think of the direct effects of parasitism on the host -the morbidity and mortality induced by the parasite in hosts. However, recent theoretical and empirical research suggests that the indirect effects of parasitism may be equally important in structuring communities (Hudson et al. 2006 ; Hatcher and Dunn 2011 ) . Indirect interactions occur when the impact of one species on another affects populations of a third species, with these interactions being density-or trait-mediated. Change in population density has been regarded as the main mechanism driving such indirect interactions. For example, the decline in the Iberian Lynx was an indirect result of parasite (myxamatosis & rabbit haemorrhagic disease) induced decline in its rabbit prey (Ferrer and Negro 2004 ) . However, indirect interactions can also be caused by changes in behaviour, development or physiology; for example parasite (nematode) induced vulnerability to predation mediates population cycles in snowshoe hares (Murray et al. 1997 ) . Such effects, known as trait-mediated indirect effects (Werner and Peacor 2003 ) can theoretically be as important as density-mediated effects for community structure and function (Abrams and Matsuda 1996 ) and are now considered an important driving factor in community ecology (Ohgushi et al. 2012 ) .

Trait-mediated interactions may be especially relevant in parasite-host systems because parasites do not necessarily kill their hosts, at least not immediately, but frequently modify host behaviour or development (reviewed in Lefevre et al. 2009 ; Thomas et al. 2010 and chapters in this issue). Such trait changes may have no apparent deleterious effect on hosts measured in isolation. However, if these trait modifi cations reduce host fi tness when interacting with other species, we can nevertheless regard them as aspects of virulence; we have referred to such characteristics as "cryptic virulence" (MacNiel et al. 2003c ; Hatcher and Dunn 2011 ) .

In our laboratories at The University of Leeds and Queen's University Belfast, we are particularly interested in the roles played by parasites in biological invasions (Prenter et al. 2004 ; Hatcher and Dunn 2011 ; Dunn et al. 2012 ) . Parasites (here we use the term to include macro-and microparasites and parasitoids) are an important consideration in biological invasions for a number of reasons ( Fig. 3.1 ) . Firstly, some invasions may succeed partly because the introduced species leaves behind its parasites (and their deleterious effects) in the native range (a mechanism known as enemy release; Mitchell and Power 2003 ; Torchin et al. 2003 ) . In other cases, introduced species arrive with their parasites, but the parasites go on to infect populations of related host species in the recipient range (the phenomenon of spillover; Daszak et al. 2000 ) , often with highly deleterious effects on the new host. Parasite spillback or dilution may also occur (from an invasions perspective, where native parasites in the invaders new invaded range cross into introduced hosts; Kelly et al. 2009 ). These can be considered as direct effects in that they infl uence host (novel or original) survival directly via presence (or absence) of the parasite. In addition, many parasites are themselves invasive, and the process of disease emergence shares many similarities with that of biological invasion (Hatcher et al. 2012a , b ; Dunn and Hatcher 2015 ) .

Furthermore, the importance of parasite-induced indirect effects in the process of biological invasion has now been recognized for a range of terrestrial and aquatic, plant and animal systems (White et al. 2006 ; Dunn et al. 2012 ). Parasite-induced trait effects have been reported as important in ongoing biological invasions in diverse taxa and ecosystems, including wild oat ( Avena fatua ) in Californian grasslands, fi re ants ( Solenopsis invicta ) in North America and amphipods in UK freshwaters Dunn and Hatcher 2015 ) . Here, we review the importance of trait-mediated effects, with particular focus on the crustacean invasions studied in our laboratories in the UK. We use a community module-based approach (Holt 1997 ; Hatcher et al. 2006 ) to structure this review, allowing us to examine systematically the potential "targets" of indirect effects.

Amphipod crustaceans are often keystone species in freshwater ecosystems. Through processing nutrients and providing prey for larger invertebrates and vertebrates, they provide important ecosystem services (MacNeil et al. 1997 ) . They process the primary basal energy resource (leafy detritus) through shredding, with strong impacts on community structure. They also consume smaller invertebrate species in the food web, infl uencing macroinvertebrate species richness and diversity, and are key prey for commercial and recreational fi sh stocks, and for wildfowl. Furthermore, amphipod species differ in their propensity towards predation and Alternative ways in which invaders cause ecological novelty through their interactions with parasites. Green circle : native community. Red arrow : invasive parasites can drive changes in host abundance and extinction and can mediate native/invader interactions leading to changes in biodiversity and community structure. Blue arrow : invasive hosts can co-introduce parasites which may act as novel weapons; invasive species may also acquire endemic parasites resulting in parasite amplifi cation or dilution. Purple arrow : extinctions and changes in community structure as a result of biological invasions can affect native parasite-host interactions. Green spiral : feedback between parasite prevalence and biodiversity (From Dunn and Hatcher 2015 ) detritus processing, and thus from an ecosystem perspective, there is unlikely to be equality of functionality between invader and native. Therefore, the impact of parasitism on amphipod population dynamics or invasion outcomes could have profound ramifi cations for the diversity and structure of aquatic communities as well as having eco-economic costs.

In rivers and streams in Ireland, the native amphipod Gammarus duebeni celticus is being replaced by the European Gammarus pulex. In addition to infl uencing native amphipod populations, G. pulex invaded rivers also have lower macroinvertebrate abundance and diversity (Kelly et al. 2006 ) . A further two species of non-native amphipods also co-occur; Gammarus tigrinus and Crangonyx pseudeogracilis , despite being weaker competitors and being predated by the other invasive and native amphipods. In mainland Europe, the invasive Dikerogammarus villosus is a voracious predator of macroinvertebrates including native and other invasive amphipods (Dodd et al. 2014 ) , with invaded regions showing a decrease in macroinvertebrate abundance and diversity (Rewicz et al. 2014 ) .

Comparison of invasive and native amphipods reveal differences in detritivory, in predation on native invertebrates, as well as a hierarchy of competition and intraguild predation (i.e., predation of ecological guild members, that is, potential competitors; Polis et al. 1989 ) between these amphipods ( Fig. 3 .2 ).

Amphipods are host to a suite of parasites including viruses (Bojko et al. 2014 ; Arundell et al. 2015 ) , microsporidia (Terry et al. 2003 ) , trematodes (Arundell et al. 2015 ) and acanthocephala (MacNeil et al. 2003a ) , which modify competitive and trophic interactions of their hosts. There is correlative evidence that enemy release may be a factor in both G. pulex and D. villosus invasions. In Ireland, Dunn and Dick ( 1998 ) have shown that parasite diversity is higher in the native G. duebeni celticus than in the invading G. pulex. Five species of parasite were detected, of which three were shared but two, the microsporidian Pleistophora mulleri and the acanthocephalan Polymorphus minutus , were restricted to the native host. For two species ( Embata parasitica and Epistylus sp.) that infected both native and invading species, parasite prevalence and burden was higher in native hosts. However, this pattern was not universal; prevalence of the acanthocephalan Echinorhynchus truttae was found to be higher in the invasive G. pulex (Dunn and Dick 1998 ; Hatcher and Dunn 2011 ) . A recent study of newly invasive D. villosus populations in the UK provides evidence of enemy release from viral, metazoan and microsporidian parasites that are highly prevalent in the (invasive) continental range (Bojko et al. 2014 ; Arundell et al. 2015 ) , although there is also evidence that parasite diversity has increased in the 5 years since this species has been reported in the UK (Bovey et al. unpublished).

In Europe, the decline of the native white-clawed crayfi sh ( Austropotamobius pallipes ) and its replacement by the North American signal crayfi sh (Pacifastacus leniusculus) is at least partly mediated by parasitism ). The fungal parasite Aphanomyces astaci (which causes crayfi sh plague) was co-introduced with the signal crayfi sh (Holdich and Poeckl 2007 ) , whilst an endemic microsporidian parasite Thelohania contejeani affects the endangered native species (Oidtmann et al. 1997 ).

Simplifi ed food web depicting amphipod invader-native interactions in Ireland. The native Gammarus duebeni celticus ( N ) sits mid-way in an intraguild predation hierarchy between the invasives G. pulex ( I1 ) and G. tigrinus ( I2 ). Amphipods prey on smaller invertebrates (e.g. Asellus aquaticus , shown) and process leafy detritus. They are prey to salmoniid fi sh (e.g. Salmo trutta , shown). Two parasites modify gammarid interactions: Pleistophora mulleri infects G. d. celticus , and Echinorhynchus truttae infects G. d. celticus and G. pulex as intermediate hosts, utilizing salmoniid fi sh as the defi nitive host. Arrows depict direct trophic interactions; direction of arrow indicating direction of energy fl ow

To examine how parasites infl uence invasion dynamics and impact, we need an understanding of the potential for indirect effects of parasitism on other species in the invaded community. Analysis of community modules helps to categorize all the potential indirect effects that may ramify from a focal host-parasite pairing ( Fig. 3 .3 ). Community modules (Holt 1997 ) interacting species, providing a bridge between simple single population or twospecies dynamics and the inherent complexity of complete communities. Modules provide a basis for analysing indirect interactions mediated by a third species, and how interactions such as predation and parasitism themselves interact. All the major trophic interactions in an ecosystem may be affected by parasitism (Hatcher and Dunn 2011 ) . From the perspective of understanding invasions, we can distinguish between two relevant facets of module structure where parasites may have impact ):

Horizontal interactions between species on the same trophic level. These may include competitive interactions (which may be infl uenced by parasites in one or more host species at that level) or apparent competition and related interactions (for instance, where two species do not compete directly for resources, but are host to the same parasite species). Via their effects on competitive strength or apparent competition, parasites may therefore infl uence invasion rate/success, and coexistence/species replacement dynamics. Vertical interactions between species at different trophic levels. These include consumptive predator-prey, herbivory and detritivory interactions. Such vertical interactions are relevant to understanding ecological impact in invasion biology, as some of the most devastating invasive species have been demonstrated as more voracious consumers than native counterparts ). As we will review, parasites infl uence predation rates on or by their hosts, and therefore may strongly infl uence the strength of these interactions.

Here, we review evidence for each of these indirect interactions in the crustaceanparasite systems we study, along with a brief examination of evidence from other systems.

In many aquatic ecosystems, leafy detritus provides the primary basal energy resource. Amphipods and isopods play an important role in processing this detritus, converting it from coarse to fi ne particulate organic matter that can be consumed by fi lter feeders and collector-gatherers (MacNeil et al. 1997 ) . Parasitism has been shown to modify detritivore interactions in some of these systems. Infection by the acanthocephalan parasite Acanthocephalus tahlequahensis was found to cause a reduction in detritus processing by its isopod host, Ceacidotea communis in streams in North America (Hernandez and Sukhdeo 2008 ) . This trait effect has communitywide ramifi cations, with parasites predicted to reduce detritus processing by up to 50 %, affecting energy transfer throughout the ecosystem. Biological invaders may have different rates of detritus processing than the native species that they replace. For example, leaf shredding function of the invasive Dikerogamarus villosus was lower than that of 3 native species MacNeil et al. ( 2011 ) , and current work is exploring whether parasites can also infl uence detrital processing.

Trait-mediated effects can be both subtle and context-dependent. For example, larval Drosophila melanogaster compete with the fungus Aspergillus niger for dead organic matter; D. melanogaster larvae suppress fungal growth, whilst the fungus secretes toxins that inhibit the fl y larvae. Infection with the parasitoid Asobara tabida reduces feeding by D. melanogaster larvae, and this infl uences its interaction with the fungus; this form of cryptic virulence leads to complex interactions between the competitors (Rohlfs 2008 ).

Herbivory per se has not been studied in our crustacean systems. However, as herbivory is key to processing the basal energy resource in many ecosystems, it would be remiss not to discuss some key fi ndings from other systems, especially as parasitism of herbivores is well documented as having profound effects on community structure. Disease outbreaks in vertebrate populations have led to some dramatic examples of cascading effects. For instance, myxomatosis and rinderpest have led to population declines of dominant grazers (rabbits Oryctolagus cuniculus and wildebeest, Connochaetes taurinus, respectively), with consequent effects on both plant and predator communities (reviewed in Hatcher and Dunn 2011 ) . In these examples, parasitism has its impact via classical density-mediated effects (reduced host survival); however, recent work suggests that less obvious trait-mediated effects on host fi tness may also have cascading effects in communities. For example, endophytes of grasses which reduce host palatability to herbivores can have knock-on effects for plant community composition and succession, and also appear to infl uence herbivore, detritivore and predator communities. Because these endophytes protect hosts from herbivory, infection with endophytes may underlie the invasive success of some grass species (Rudgers et al. 2007 ) . The trematode C ryptocotyle lingua does not kill its intermediate host, the algal grazing periwinkle Littorina littorea , but damages the host digestive system. As a result, per capita algal grazing was reduced by 40 % in infected periwinkles, leading to a 65 % increase in algal cover, which is likely to infl uence the recruitment of other species that require a rocky surface for colonization (Wood et al. 2007 ).

Parasite-mediated effects on competition provide some of the most-cited cases of parasite impact on community structure. Parasite-induced density effects have long been recognized as having potential to drive community structure via their effects on competitors (Elton 1958 ) . Indeed, seedling mortality induced by specialist pathogens could be one of the chief drivers of plant biodiversity in tropical rainforests (Freckleton and Lewis 2006 ) and may explain aspects of global diversity patterns today (Kulmatiski et al. 2008 ) . In contrast, multi-host parasites may reduce biodiversity, preventing coexistence or speeding the replacement of one host species by another. One well-documented case is that of squirrel poxvirus (SQPV) in the red squirrel Sciurus vulgaris , which is under threat of replacement by the grey squirrel Sciurus carolinensis in the UK (Tompkins et al. 2003 ) . SQPV has a direct, strongly pathogenic effect on red squirrels (Tompkins et al. 2003 ) and can potentially be spread through multiple routes (Collins et al. 2014 ) . In England, where the virus is present, decline in red squirrels is up to 25 times more rapid than in Scotland and Italy, where the virus is absent (Rushton et al. 2006 ) .

However, trait-mediated competitive interactions are also known to affect both animal and plant invasions. For instance, plant pathogens can potentially infl uence plant community composition and grassland diversity via trait-mediated indirect effects on competition; Barley Yellow Dwarf Virus (BYDV) modifi es competition between native perennial bunchgrasses and invasive annual grass species in the Californian plains. In fi eld plot experiments, BYDV infection did not affect bunchgrass survivorship when grown alone, but halved survivorship of bunchgrass grown with invasive species (Malmstrom et al. 2006 ). This effect may facilitate the ecological replacement of bunchgrasses by invasive competitors (Borer et al. 2007 ) . A further example of the power of parasite-induced trait effects concerns parasitic phorid fl ies. Phorids have surprising effects on their ant hosts, causing direct virulence effects in infected hosts, but indirectly exerting trait effects in uninfected hosts that exhibit parasite avoidance behaviour. These parasitoids develop in the head capsule of worker ants -leading to decapitation before the fl y pupates and emerges as an adult. The presence of the phorid Apocephalus sp. induces refuge seeking behaviour in potential hosts, preventing defensive behaviour towards the competing ant species (Lebrun and Feener 2007 ) . In the presence of phorid fl ies, an otherwise dominant species may be outcompeted by an inferior competitor. Interestingly, ant species that are dominant competitors experience higher levels of infection as host alarm and foraging trail pheromones are used by the parasitoid to detect hosts in which they will lay their eggs. Fear of parasitism may be more widespread than we appreciate: in the 2003 SARS pandemic, in addition to state-imposed contact control measures, there were notable decreases in social behaviour (such as restaurant usage and attendance of social events) as people attempted to reduce their risk of infection (Blendon et al. 2004 ).

In the Ireland Gammarus system, we suspect that trait effects of parasites may act via competition on the invasion process. For instance, parasite-induced changes in predation on shared invertebrate prey could release or exacerbate invader-native competition. However, intraguild predation is also key to invader-native interactions, thus it is diffi cult to tease apart the importance of competition versus predation (including intraguild predation between competing gammarids) in these community modules; these processes are discussed below.

Parasites can infl uence predation via trait effects in two ways; parasites of predators may alter the predatory behaviour of such hosts, whereas parasites of prey species can alter the likelihood of the host being consumed. Both scenarios are found in crustacean systems we study.

In laboratory experiments, the microsporidian P. mulleri causes little direct mortality or reduction in fecundity to its host, the native G. d. celticus . However, this parasite leads to reduced predatory ability on smaller sympatric amphipods and isopods (MacNeil et al. 2003b ; Fielding et al. 2005 ) . The effects of a second parasite, the acanthocephalan E. truttae, on its host the invasive G. pulex, are more complex and depend on prey size and species (Dick et al. 2010 ; Paterson et al. 2015 ) . Analyses of predator functional responses (the per capita prey consumption as a function of prey density ; Holling 1966 ; Dick et al. 2014 ) reveal that, when provided with juvenile A aquaticus , E. truttae -infected G. pulex exhibited a signifi cantly higher functional response asymptote (i.e., exhibit higher predation rates) than uninfected individuals (Dick et al. 2010 ) .

For both the invasive G. pulex and the native G. d. celticus , parasitism (with E. truttae and P. mulleri respectively) infl uenced foraging mostly through effects on prey handling time . Whilst functional responses to some more mobile species (e.g. mayfl y nymphs, Baetis rhodani ) were lower for infected amphipods, predation was increased on the more sessile Simulium species. Furthermore, additional trait-mediated effects (the presence of higher-order fi sh predators) further modifi ed foraging behaviour and the impact of parasites . These fi ndings, together with those for detritivory and cannibalism (see below), are strong indicators that parasites not only infl uence appetite, they infl uence prey choice and may induce trophic shifts in the host. This may be particularly relevant for our understanding of parasite impact in communities in general, as well as their role in invasion success and impact Dunn and Hatcher 2015 ) .

Parasitism also mediates interactions between invasive and native crayfi sh and their native prey ). The invasive signal crayfi sh, P. leniusculus, which was introduced to Europe for aquaculture, is driving extirpation of native European crayfi sh including the endangered white clawed crayfi sh A. pallipes . Crayfi sh plague (caused by the fungus Aphanomyces astaci ) is asympotamatic in the signal crayfi sh. However, the invader acts a reservoir for the parasite which causes mass mortality in the native species (Holdich and Poeckl 2007 ) . In contrast, the microsporidian parasite Thelohania contejeani (Imhoff et al. 2011 ) mediates predation by the host through trait effects. T contejeani causes porcelain disease, a chronic infection in the native crayfi sh, which suffer muscle damage and a reduction in their predatory abilities (Haddaway et al. 2012 ) . Comparison of predatory functional responses revealed that native crayfi sh had a lower prey intake rate than did the invasive species, refl ecting a longer prey handling time. This difference is exacerbated by the parasite, which induced muscle damage leading to a reduced attack rate and increased prey handling time, contributing to a 30 % reduction in prey intake by parasitised native crayfi sh (Haddaway et al. 2012 ) .

The effects of parasites on predator survival and/or behavior can have knock-on effects for prey species (Hatcher et al. 2006 ) . In some cases, the effects can long outlive the duration of active infection in a population, and would be missed in contemporary experiments. For instance, an outbreak of canine parvovirus in grey wolves ( Canis lupus ) in the US National Park of Isle Royale 20 years ago precipitated a population crash from which the wolf population has never recovered (probably because the crash reduced genetic diversity in the wolf population). This density-mediated effect has shifted the balance of population regulation of the chief prey species (moose, Alces alces ) away from predation and towards correlation with climatic factors (Wilmers et al. 2006 ) . Whether trait-mediated effects can have such lasting impact has yet to be examined.

In some cases, predators attack infected prey more frequently than healthy prey because parasitism weakens hosts making them easier to catch. This parasiteinduced vulnerability to predation may determine population dynamics and community structure in some natural systems (Packer et al. 2003 ; Holt and Roy 2007 ) . If infected prey provide an "easy" resource for predators to exploit, parasitism can enhance persistence of a predatory species (Hethcote et al. 2004 ). Alternatively, predators and parasites can interact to determine prey population persistence and dynamics. For instance, parasite-predator interactions can drive cyclic population dynamics in the snowshoe hare ( Lepus americanus ). Murray et al. ( 1997 ) found that hare populations prone to strong predation more than doubled their survival rates when treated with anthelminthics. Infection with nematode parasites had no other measurable effects on host fecundity or direct mortality, thus increased predation on infected hosts appears to be the sole cause of virulence in this system. Similarly, red grouse Lagopus lagopus killed by predators had signifi cantly higher burdens of the caecal nematode Trichostrongylus tenuis and worm burden refl ected the level of gamekeeper control of predators (Hudson et al. 1992 ) .

Many of the behavioural changes described above represent "by products" of infection (Poulin 1995 ) . Nonetheless, in the community context, their outcomes may be far from "boring" (Poulin 1995 ) . Other changes in prey/host behavior may be adaptive, resulting from selection on the host to avoid infection, or selection on the parasite to enhance transmission. For example, parasites can infl uence predation via effects on the avoidance behavior of potential hosts. Tadpoles of the toad Bufo americanus exhibit an adaptive "fear" response of hyperactivity to chemical cues from cercariae of the trematode Echinostoma trivolvis , which reduce their chances of becoming infected (Rohr et al. 2009 ). This increased activity could potentially offset adaptive predator avoidance strategies (predator presence elicits a decrease in activity) and lead to suboptimal foraging and developmental rates (Raffel et al. 2010 ) .

Trophically transmitted parasites have indirect life cycles and are transmitted to the defi nitive host when it preys upon an infected intermediate host. Many species of acanthocephalan and trematode parasites have been found to manipulate the anti-predator behaviour of the intermediate host, thereby increasing its vulnerability to predation (Moore 2002 ; Lefevre et al. 2009 ). For example, the acanthocephalan parasites E. truttae and P. minutus both induced changes in activity and photophilic behaviour of their amphipod hosts: infected hosts show higher motility and are more photophilic than uninfected hosts (MacNeil et al. 2003a ) , enhancing transmission to the defi nitive fi sh and duck hosts. Although both the native and invasive species are used as intermediate hosts, E. truttae prevalence is higher in the invasive G. pulex, which could have ramifi cations for invasion dynamics as well as having the potential to increase parasite frequency in commercially important fi sh (Hatcher and Dunn 2011 ) .

Intraguild predation, that is, predation between potential competitor species, is widespread in ecological communities (Polis et al. 1989 ) . For example, wolves compete with hyenas for prey, but also prey upon them, whilst invasive and native amphipods show mutual intraguild predation whilst perhaps also competing for other invertebrate prey. Intraguild predation is a special case wherein predation by and of focal host species need to be considered together. In the N. Ireland amphipod system, intraguild predation is key to understanding invasion and coexistence (Dick et al. 1993 ) . The two species of parasite (the microsporidium P. mulleri and the acanthocephalan E. truttae. ) have been shown to infl uence both the prey and predatory components of intraguild predation between these amphipods, with P. mulleri infecting the abdominal musculature in the native G. d. celticus, thus reducing host motility (MacNeil et al. 2003b ) . In fi eld enclosure experiments, P. mulleri had no impact on G. d. celticus survival whether in single populations or in mixed species populations with the smaller G. tigrinus . However, infection of G. d. celticus resulted in increased survival of the smaller intraguild prey species ; whilst G. tigrinus kept in mixed populations with G. d. celticus showed 90 % mortality over 2 weeks, this species was able to coexist with infected G. d. celticus . Laboratory experiments confi rmed that coexistence was facilitated by a reduction in intraguild predation by infected G. d. celticus on G. tigrinus (Fig. 3.4 ) . Similarly, infected G. d. celticus showed less intraguild predation of another smaller amphipod, C. pseudogracilis. In addition to modifying the intraguild predatory abilities of its host, P. mulleri infection also increased host vulnerability to becoming the intraguild prey of the larger invasive species G. pulex .

For G. pulex , infection by the acanthocephalan E. truttae not only modifi es host predatory abilities on smaller invertebrate species, it also reduces predation on its intraguild prey, including the native G. d. celticus (MacNeil et al. 2003a ) . Field enclosure experiments were used to assess the impact of parasitism on the native and invading hosts in single and mixed species populations. In single species populations, E. truttae did not directly affect the survival of the native or the invader. In mixed species populations, the survival of G. pulex was again unaffected. However, parasitized G. pulex were less likely to attack the native species leading to higher survival of the native species (MacNeil et al. 2003a ). This reduction in intraguild predation can theoretically facilitate species coexistence, limiting spread of the invader (Hatcher et al. 2008 . Mathematical models of this process demonstrate that the trait-mediated effects of parasites on predation can act as a strong force on co-existence dynamics, equal to that generated by density effects (Hatcher et al. 2008 . Intraguild predation is a curious population module because it is theoretically rather unstable, predicted in many cases to disintegrate to extinction of either the intraguild predator or prey. Whilst parasites with density (mortality) effects can promote coexistence of predators and prey (Hatcher et al. 2008 ) , we have also found that parasites with trait effects can have an equally strong impact on coexistence . Indeed, parasites with zero classical virulence, but cryptic virulence in the form of altered vulnerability to or propensity for predation, can have as pronounced an impact as density in promoting or inhibiting coexistence (Fig. 3.5 ) . Parasites can thus have strong ecological impacts, even if they have negligible classical virulence, underscoring the need to consider trait-mediated effects when predicting effects of parasites on community structure in general and biological invasions in particular.

One further trophic interaction could be important in understanding the dynamics of these amphipod systems. Cannibalism is a frequent occurrence in many systems, being both stage-and sex-specifi c; adults frequently consume immature conspecifics, and adults at moult -in particular females that may be held in precopula -are also vulnerable to cannibalism (MacNeil et al. 2003c ; Bunke et al. 2015 ) . In G. d. celticus , infection with P. mulleri can cause increased vulnerability to cannibalism, which also leads to enhanced transmission of the parasite as it is transmitted directly per os (MacNeil et al. 2003b ) . Although more vulnerable to being cannibalized themselves, recent work has shown that infected individuals become more cannibalistic, eating twice as many conspecifi cs in laboratory experiments. This may refl ect the metabolic demand of the parasite; infected individuals are less able to consume prey of different species, and it may be that smaller conspecifi cs are an easier food item to capture and handle (Bunke et al. 2015 ) . Interestingly, choice experiments indicate that uninfected G. d. celticus preferentially cannibalise uninfected conspecifi cs (Bunke et al. 2015 ) . The parasite is transmitted via cannibalism, thus it seems that this behavior is an adaptive mechanisms to avoid the risk of parasitic infection.

Our crustacean study system is one of few systems where the impact of parasites on a range of different community modules has been studied, and we have seen that trait-mediated indirect effects have the potential to ramify both horizontally and vertically in these modules. It is likely, therefore, that such effects underpin other systems. In particular, since parasites are known to have a wealth of effects on feeding behaviour and vulnerability to predation, it seems likely that they will strongly infl uence vertical trophic interactions. We have good evidence that, for a broad range of systems, invasion impact is related to the predatory propensity or resource consumption by the introduced species . Hence, parasites could potentially be important in modifying predation/consumption in a number of these systems, contributing to invasion success and impact. It is therefore important to study trait-mediated effects from the practical perspective of understanding and predicting invasion outcomes and also to understand the impact of parasites on biological invasion in particular and community ecology in general. Trait-mediated effects of parasitism on IGP systems. Equilibrium population densities ( solid lines ) and parasite prevalence ( dashed lines ) in the more predatory species ( intraguild predator ; red ) and the weaker predator ( intraguild prey , blue ) plotted against the relative competitive advantage of the intraguild prey. ( a ) in the absence of parasitism, a strong intraguild predator excludes competitively weak intraguild prey; ( b ) virulent parasites (inducing 10 and 30 % mortality in the prey and predator respectively) enable coexistence of predator and prey over a broader parameter range; ( c ) parasites with no mortality effect but which reduce predatory ability strongly enhance coexistence whereas ( d ) parasites that increase predation strongly reduce coexistence (From Hatcher et al. 2014 ) With the increasing realization that parasites are important components of ecosystems (Hudson et al. 2006 ; Lafferty et al. 2008 ) , the potential importance of behavioural effects of parasites in a community context now deserves recognition (Lefevre et al. 2009 ). We have shown these community effects to be important both empirically and theoretically at the community module level (MacNeil et al. 2003c ; Dick et al. 2010 ; Hatcher et al. 2014 ) . Parasitologists and ecologists need to examine further the community and ecosystem-level consequences of trait-mediated effects and the cryptic virulence to which they lead. There is increasing recognition that parasites play important roles in ecosystem functioning, by infl uencing species coexistence patterns, biological invasions, energy fl ow and community stability (Hudson et al. 2006 ; Hatcher et al. 2012a , b ) . Underpinning all these ecosystemlevel effects are the interactions between species in community modules. However, whilst we can demonstrate parasite-induced trait-mediated interactions at the level of modules, scaling up to the ecosystem level is a challenging proposition, both theoretically and empirically.

Positive indirect effects between prey species that share predators

Enemy release and genetic founder effects in invasive killer shrimp populations of Great Britain

The public's response to severe acute respiratory syndrome in Toronto and the United States

Baseline histopathalogical survey of a recently invading island population of 'killer shrimp' Dikerogammarus villosus

Pathogen-induced reversal of native dominance in a grassland community

Eaten alive: cannibalism is enhanced by parasites

Squirrelpox virus: assessing prevalence, transmission and environmental degradation

Wildlife ecology -emerging infectious diseases of wildlife -threats to biodiversity and human health

Replacement of the indigenous amphipod Gammarus duebeni celticus by the introduced Gammarus pulex -differential cannibalism and mutual predation

Parasitism may enhance rather than reduce the predatory impact of an invader

Advancing impact prediction and hypothesis testing in invasion ecology using a comparative functional response approach

Predicting the ecological impacts of a new freshwater invader: functional responses and prey selectivity of the 'killer shrimp', Dikerogammarus villosus, compared to the native Gammarus pulex

Parasites and biological invasions

Parasitism and epibiosis in native and non-native gammarids in freshwater in Ireland

Parasites and biological invasions; parallels, interactions and control

Competition and parasitism in the native White Clawed Crayfi sh Austropotamobius pallipes and the invasive Signal Crayfi sh Pacifastacus leniusculus in the UK

Indirect effects of parasites in invasions

The ecology of invasions by animals and plants

The near extinction of two large European predators: super specialists pay a price

Ecological impacts of the microsporidian parasite Pleistophora mulleri on its freshwater amphipod host Gammarus duebeni celticus

Pathogens, density dependence and the coexistence of tropical trees

Predatory functional response and prey choice identify predation differences between native/invasive and parasitised/unparasitised crayfi sh

Parasites in ecological communities: from interactions to ecosystems

How parasites affect interactions between competitors and predators

A keystone effect for parasites in intraguild predation?

Disease emergence and invasions

Diverse effects of parasites in ecosystems: linking interdependent processes

Parasites that change predator or prey behaviour can have keystone effects on community composition

Parasite effects on isopod feeding rates can alter the host's functional role in a natural stream ecosystem

A predator-prey model with infected prey

Invasive crustaceans in European inland waters. In: Gherardi F (ed) Biological invaders in inland waters: profi les, distribution and threats

The functional response of invertebrate predators to prey density

Multitrophic interactions in terrestrial systems

Predation can increase the prevalence of infectious disease

Do parasites make prey vulnerable to predation -red grouse and parasites

Is a healthy ecosystem one that is rich in parasites?

Porecelain disease in white-clawed ( Austromotamobius pallipes (Lereboullet)) and signal ( Pacifastacus lenuisculus (Dana)) crayfi sh in the UK. Crayfi sh Conservation in the British Isles

Invasion by the amphipod Gammarus pulex alters community composition of native freshwater macroinvertebrates

Parasite spillback: a neglected concept in invasion ecology

Plant-soil feedbacks: a meta-analytical review

Parasites in food webs: the ultimate missing links

When trade-offs interact: balance of terror enforces dominance discovery trade-off in a local ant assemblage

Ecological signifi cance of manipulative parasites

The trophic ecology of freshwater Gammarus aspp. (Crustacea: Amphipoda): problems and perspectives concerning the functional feeding group concept

An acanthocephalan parasite mediates intraguild predation between invasive and native freshwater amphipods (Crustacea)

Parasite-mediated predation between native and invasive amphipods

Parasite transmission and cannibalism in an amphipod (Crustacea)

Direct and indirect effects of species displacements: an invading freshwater amphipod can disrupt leaf-litter processing and shredder efficiency

Virus infection and grazing exert counteracting infl uences on survivorship of native bunchgrass seedlings competing with invasive exotics

Release of invasive plants from fungal and viral pathogens

Parasites and the behavior of animals

Interactive effects of sublethal nematodes and nutritional status on snowshoe hare vulnerability to predation

Light microscopy of Astacus astacus L. under normal and selected pathological conditions, with special emphasis to porcelain disease and crayfi sh plague

Keeping the herds healthy and alert: implications of predator control for infectious disease

Predicting invasive species impacts: a community module functional response approach reveals context dependencies

The ecology and evolution of intraguild predation -potential competitors that eat each other

Adaptive" changes in the behaviour of parasitized animals: a critical review

Roles of parasites in animal invasions

Parasitism in a community context: traitmediated interactions with competition and predation

The profi le of a 'perfect' invader -the case of killer shrimp, Dikerogammarus villosus

Host-parasitoid interaction as affected by interkingdom competition

Parasites, info-disruption, and the ecology of fear

Forest succession suppressed by an introduced plantfungal symbiosis

Disease threats posed by alien species: the role of a poxvirus in the decline of the native red squirrel in Britain

Resolution of a taxonomic conundrum: an ultrastructural and molecular description of the life cycle of Pleistophora mulleri

Host manipulation by parasites: a multidimensional phenomenon

Ecological replacement of native red squirrels by invasive greys driven by disease

Introduced species and their missing parasites

A review of trait-mediated indirect interactions in ecological communities

Biotic indirect effects: a neglected concept in invasion biology

Predator disease out-break modulates topdown, bottom-up and climatic effects on herbivore population dynamics

Parasites alter community structure

Acknowledgements This work was supported by NERC (NE/G01/5201/1). We thank Prof. Dr. Chris Tofts for insightful discussions and Prof. Dr. Heinz Melhorn for his patience.