From first to second and back to first intermediate host: the unusual transmission route of Curtuteria australis (Digenea: Echinostomatidae)

The trematode Curtuteria australis uses the whelk Cominella glandiformis as first intermediate host and the cockle Austrovenus stutchburyi as second intermediate host before maturing in shorebirds. The whelk also happen to be an important predator of cockles on intertidal mudflats. In this study we show that whelks can act as temporary paratenic hosts for the trematode. A single whelk feeding on 1 cockle can ingest large numbers of metacercariae, which remain within the whelk for 1-3 days before passing out in feces. The viability of these metacercariae assessed as the percentage capable of successfully excysting under conditions simulating those inside a bird's digestive tract, is lower after passage through a whelk (48%) than before (59%). Still, given that shorebird definitive hosts prey on whelks as well as cockles, survival inside the whelk allows C. australis to complete its life cycle: overall, though, whelk predation is likely to be an important sink for the trematode population. To our knowledge, this is the first report of a trematode using a snail as both first intermediate host and paratenic host, offering an alternative transmission route for the parasite as a result of the unusual trophic relationships of its hosts.     

Parasite‐induced surfacing in the cockle Austrovenus stuchburyi: adaptation or not?

Parasite manipulation of host behaviour is a compelling example of the extended phenotype. However, in many cases, such manipulation may be incorrectly assumed. Previous work has demonstrated that Austrovenus stuchburyi cockles stranded on mud-flat surfaces due to an inability to re-burrow both contain significantly more metacercariae of the trematode Curtuteria australis and are predated by the definitive host of this parasite at a faster rate than burrowed cockles. These results have been interpreted as strong evidence for a manipulation of cockle behaviour by the trematode to facilitate transmission to the definitive host. The model presented here, however, indicates that the selective advantage to the parasite of the altered host behaviour is currently of a negligible level at our study site that is highly unlikely to have been realized as an adaptation over evolutionary time. Hence, there are no grounds on which the more parsimonious explanation, that the altered host behaviour observed is simply an incidental side-effect of infection, can be rejected. We thus maintain that for any change in the behaviour of infected hosts to be confirmed as potentially a parasite trait that has evolved in response to selection, the adaptive benefit taking into account the entire parasite life cycle may need to be considered.     

Non-Specific Manipulation of Gammarid Behaviour by P. minutus Parasite Enhances Their Predation by Definitive Bird Hosts

Trophically-transmitted parasites often change the phenotype of their intermediate hosts in ways that increase their vulnerability to definitive hosts, hence favouring transmission. As a “collateral damage”, manipulated hosts can also become easy prey for non-host predators that are dead ends for the parasite, and which are supposed to play no role in transmission strategies. Interestingly, infection with the acanthocephalan parasite Polymorphus minutus has been shown to reduce the vulnerability of its gammarid intermediate hosts to non-host predators, whose presence triggered the behavioural alterations expected to favour trophic transmission to bird definitive hosts. Whilst the behavioural response of infected gammarids to the presence of definitive hosts remains to be investigated, this suggests that trophic transmission might be promoted by non-host predation risk. We conducted microcosm experiments to test whether the behaviour of P. minutus-infected gammarids was specific to the type of predator (i.e. mallard as definitive host and fish as non-host), and mesocosm experiments to test whether trophic transmission to bird hosts was influenced by non-host predation risk. Based on the behaviours we investigated (predator avoidance, activity, geotaxis, conspecific attraction), we found no evidence for a specific fine-tuned response in infected gammarids, which behaved similarly whatever the type of predator (mallard or fish). During predation tests, fish predation risk did not influence the differential predation of mallards that over-consumed infected gammarids compared to uninfected individuals. Overall, our results bring support for a less sophisticated scenario of manipulation than previously expected, combining chronic behavioural alterations with phasic behavioural alterations triggered by the chemical and physical cues coming from any type of predator. Given the wide dispersal range of waterbirds (the definitive hosts of P. minutus), such a manipulation whose efficiency does not depend on the biotic context is likely to facilitate its trophic transmission in a wide range of aquatic environments.     

Two different strategies of host manipulation allow parasites to persist in intermediate–definitive host systems

Trophically transmitted parasites start their development in an intermediate host, before they finish the development in their definitive host when the definitive host preys on the intermediate host. In intermediate–definitive host systems, two strategies of host manipulation have been evolved: increasing the rate of transmission to the definitive host by increasing the chance that the definitive host will prey on the intermediate host, or increasing the lifespan of the parasite in the intermediate host by decreasing the predation chance when the intermediate host is not yet infectious. As the second strategy is less well studied than the first, it is unknown under what conditions each of these strategies is prevailed and evolved. We analysed the effect of both strategies on the presence of parasites in intermediate–definitive host systems with a structured population model. We show that the parasite can increase the parameter space where it can persist in the intermediate–definitive host system using one of these two strategies of host manipulation. We found that when the intermediate host or the definitive host has life‐history traits that allow the definitive host to reach large population densities, that is high reproduction rate of the intermediate host or high conversion efficiency of the definitive host (efficiency at which the uninfected definitive host converts caught intermediate hosts into offspring), respectively, evolving manipulation to decrease the predation chance of the intermediate host will be more beneficial than manipulation to increase the predation chance to enhance transmission. Furthermore, manipulation to decrease the predation chance of the intermediate host results in higher population densities of infected intermediate hosts than manipulation that increases the predation chance to enhance transmission. Our study shows that host manipulation in early stages of the parasite development to decrease predation might be a more frequently evolved way of host manipulation than is currently assumed.     

The effects of parasite age and intensity on variability in acanthocephalan-induced behavioural manipulation

Numerous parasites with complex life cycles are able to manipulate the behaviour of their intermediate host in a way that increases their trophic transmission to the definitive host. Pomphorhynchus laevis, an acanthocephalan parasite, is known to reverse the phototactic behaviour of its amphipod intermediate host, Gammarus pulex, leading to an increased predation by fish hosts. However, levels of behavioural manipulation exhibited by naturally-infected gammarids are extremely variable, with some individuals being strongly manipulated whilst others are almost not affected by infection. To investigate parasite age and parasite intensity as potential sources of this variation, we carried out controlled experimental infections on gammarids using parasites from two different populations. We first determined that parasite intensity increased with exposure dose, but found no relationship between infection and host mortality. Repeated measures confirmed that the parasite alters host behaviour only when it reaches the cystacanth stage which is infective for the definitive host. They also revealed, we believe for the first time, that the older the cystacanth, the more it manipulates its host. The age of the parasite is therefore a major source of variation in parasite manipulation. The number of parasites within a host was also a source of variation. Manipulation was higher in hosts infected by two parasites than in singly infected ones, but above this intensity, manipulation did not increase. Since the development time of the parasite was also different according to parasite intensity (it was longer in doubly infected hosts than in singly infected ones, but did not increase more in multi-infected hosts), individual parasite fitness could depend on the compromise between development time and manipulation efficiency. Finally, the two parasite populations tested induced slightly different degrees of behavioural manipulation.     

Equal partnership: two trematode species, not one, manipulate the burrowing behaviour of the New Zealand cockle, Austrovenus stutchburyi

Metacercariae of the trematode Curtuteria australis (Echinostomatidae) accumulate in the foot of the New Zealand cockle Austrovenus stutchburyi, severely impairing the cockle's ability to burrow under the sediments. This results in increased predation by birds on cockles, and thus enhanced transmission rates of the parasite to its bird definitive hosts. This host manipulation by the trematode is costly: fish regularly crop the tip of the foot of cockles stranded on the sediment surface, killing any metacercariae they ingest. A second, previously undetected trematode species (characterized by 23 collar spines) co-existing with C. australis, has been found in the foot of cockles in the Otago Harbour, South Island, New Zealand. The relative abundance of the two species varies among localities, with the identity of the numerically dominant species also changing from one locality to the next. Both C. australis and the new species have a strong preference for encysting in the tip of the cockle's foot, where their impact on the burrowing ability of the host is greatest, and where they both face the risk of cropping by fish. Results indicate that these two species are ecological equivalents, and their combined numbers determine how the cockle population is affected.     

Host manipulation by Ligula intestinalis: a cause or consequence of parasite aggregation?

Previous investigations suggest that the infection of the cyprinid roach, Rutilus rutilus, with the larval plerocercoid forms of the cestode, Ligula intestinalis, creates behavioural and morphological changes in the fish host, potentially of adaptive significance to the parasite in promoting transmission to definitive avian hosts. Here we consider whether these behavioural changes are important in shaping the distribution of parasite individuals across the fish population. An examination of field data illustrates that fish infected with a single parasite were more scarce than expected under the negative binomial distribution, and in many months were more scarce than burdens of two, three or more, leading to a bimodal distribution of worm counts (peaks at 0 and >1). This scarcity of single-larval worm infections could be accounted for a priori by a predominance of multiple infection. However, experimental infections of roach gave no evidence for the establishment of multiple worms, even when the host was challenged with multiple intermediate crustacean hosts, each multiply infected. A second hypothesis assumes that host manipulation following an initial single infection leads to an increased probability of subsequent infection (thus creating a contagious distribution). If manipulated fish are more likely to encounter infected first-intermediate hosts (through microhabitat change, increased ingestion, or both), then host manipulation could act as a powerful cause of aggregation. A number of scenarios based on contagious distribution models of aggregation are explored, contrasted with alternative compound Poisson models, and compared with the empirical data on L. intestinalis aggregation in their roach intermediate hosts. Our results indicate that parasite-induced host manipulation in this system can function simultaneously as both a consequence and a cause of parasite aggregation. This mutual interaction between host manipulation and parasite aggregation points to a set of ecological interactions that are easily missed in most experimental studies of either phenomenon.     

Host manipulation by parasites in the world of dead-end predators: adaptation to enhance transmission?

Trophically transmitted parasites often alter their intermediate host's phenotype, thereby predisposing the hosts to increased predation. This is generally considered a parasite strategy evolved to enhance transmission to the next hosts. However, the adaptive value of host manipulation is not clear as it may be associated with costs, such as increased susceptibility to predators that are unsuitable next hosts for the parasites. We examined the ratio between the benefits and costs of host manipulation for transmission success of Acanthocephalus lucii (Acanthocephala), a parasite that alters the hiding behaviour and pigmentation of its isopod hosts. We experimentally compared the susceptibility of infected and uninfected isopods to predation by perch (Perca fluvialis; definitive host of the parasite) and dragonfly larvae (dead end). We found that the parasite predisposed the isopods to predation by both predators. However, the increased predation vulnerability of the infected isopods was higher towards perch. This suggests that, despite the costs due to non-host predation, host manipulation may still be advantageous for the parasite.     

An acanthocephalan parasite boosts the escape performance of its intermediate host facing non-host predators

Among the potential effects of parasitism on host condition, the 'increased host abilities' hypothesis is a counterintuitive pattern which might be predicted in complex-life-cycle parasites. In the case of trophic transmission, a parasite increasing its intermediate host's performance facing non-host predators improves its probability of transmission to an adequate, definitive host. In the present study, we investigated the cost of infection with the acanthocephalan Polymorphus minutus on the locomotor/escape performance of its intermediate host, the crustacean Gammarus roeseli. This parasite alters the behaviour of its intermediate host making it more vulnerable to predation by avian definitive hosts. We assessed the swimming speeds of gammarids using a stressful treatment and their escape abilities under predation pressure. Despite the encystment of P. minutus in the abdomen of its intermediate host, infected amphipods had significantly higher swimming speeds than uninfected ones (increases of up to 35%). Furthermore, when interacting with the non-host crustacean predator Dikerogammarus villosus, the highest escape speeds and greatest distances covered by invertebrates were observed for parasitized animals. The altered behaviour observed among the manipulated invertebrates supported the 'increased host abilities' hypothesis, which has until now remained untested experimentally. The tactic of increasing the ability of infected intermediate hosts to evade potential predation attempts by non-host species is discussed.     

Parasite-induced suppression of aggregation under predation risk in a freshwater amphipod: Sociality of infected amphipods

Recent findings suggest that grouping with conspecifics is part of the behavioural defences developed by amphipod crustaceans to face predation risk by fish. Amphipods commonly serve as intermediate hosts for trophically transmitted parasites. These parasites are known for their ability to alter intermediate host phenotype in a way that promotes predation by definitive hosts, where they reproduce. If aggregation in amphipods dilutes the risk to be preyed on by fish, then it may dilute the probability of transmission for the parasite using fish as definitive hosts. Using experimental infections, we tested whether infection with the fish acanthocephalan Pomphorhynchus laevis alters attraction to conspecifics in the amphipod intermediate host Gammarus pulex. We also measured G. pulex's activity and reaction to light to detect potential links between changes in aggregation and changes in other behaviours. The attraction to conspecifics in the presence of predator cue, a behaviour found in uninfected gammarids, was cancelled by the infection, while phototaxis was reversed and activity unchanged. We found no correlation between the three behaviours in infected amphipods, while activity and aggregation were negatively correlated in uninfected individuals after the detection of predation cue. The physiological causes and the adaptive value of aggregation suppression are discussed in the context of a multidimensional manipulation.     

Host manipulation as a parasite transmission strategy when manipulation is exploited by non-host predators

Trophically transmitted parasites often alter their intermediate host's phenotype, thereby predisposing hosts to increased predation. This is generally considered to be a parasite strategy evolved to enhance transmission to the next host. However, the adaptive value of host manipulation is not clear, as it may be associated with costs, such as increased susceptibility to predator species that are unsuitable next hosts for the parasites. Thus, it has been proposed that, to be adaptive, manipulation should be specific by predisposing hosts more strongly to predation by target hosts (next host in the life cycle) than to non-hosts. Here we formally evaluate this prediction, and show that manipulation does not have to be specific to be adaptive. However, when manipulation is nonspecific, it needs to effectively increase the overall predation risk of infected hosts if it is to increase the parasite transmission probability. Thus, when initial predation risk is low, even highly nonspecific manipulation strategies can be adaptive. However, when initial predation risk is high, manipulation needs to be more specific to increase parasite transmission success. Therefore, nonspecific host manipulation may evolve in nature, but the adaptive value of a certain manipulation strategy can vary among different parasite populations depending on the variation in initial predation risk.     

Intra-host competition between co-infecting digeneans within a bivalve second intermediate host: Dominance by priority-effect or taking advantage of others?

We experimentally investigated the interactions between two parasites known to manipulate their host's phenotype, the trematodes Acanthoparyphium sp. and Curtuteria australis, which infect the cockle Austrovenus stutchburyi. The larval stages of both species encyst within the tissue of the bivalve's muscular foot, with a preference for the tip of the foot. As more individuals accumulate at that site, they impair the burrowing behaviour of cockles and increase the probability of the parasites' transmission to a bird definitive host. However, individuals at the foot tip are also vulnerable to non-host predators in the form of foot-cropping fish which selectively bite off the foot tip of exposed cockles. Parasites encysted at the foot base are safe from such predators although they do not contribute to altering host behaviour, but nevertheless benefit from host manipulation as all parasites within the cockle are transmitted if it is ingested by a bird. Experimental infection revealed that Acanthoparyphium sp. and C. australis have different encystment patterns within the host, with proportionally fewer Acanthoparyphium metacercariae encysting at the foot tip than C. australis. This indicates that Acanthoparyphium may benefit indirectly from C. australis and incur a lower risk of non-host predation. However, in co-infections, not only did C. australis have higher infectivity than Acanthoparyphium, it also severely affected the latter's infection success. The asymmetrical strategies and interactions between the two species suggest that the advantages obtained from exploiting the host manipulation efforts of another parasite might be offset by traits such as reduced competitiveness in co-infections.     

Body size,trophic level,and the use of fish as transmission routes by parasites

Within food webs, trophically transmitted helminth parasites use predator–prey links for their own transfer from intermediate prey hosts, in which they occur as larval or juvenile stages, to predatory definitive hosts, in which they reach maturity. In large taxa that can be used as intermediate and/or definitive hosts, such as fish, a host species’ position within a trophic network should determine whether its parasite fauna consists mostly of adult or larval helminths, since vulnerability to predation determines an animal’s role in predator–prey links. Using a large database on the helminth parasites of 303 fish species, we tested whether the proportion of parasite species in a host that occur as larval or juvenile stages is best explained by their trophic level or by their body size. Independent of fish phylogeny or habitat, only fish body length emerged as a significant predictor of the proportion of parasites in a host that occur as larval stages from our multivariate analyses. On average, the proportion of larval helminth taxa in fish shorter than 20 cm was twice as high as that for fish over 100 cm in length. This is consistent with the prediction that small fishes, being more vulnerable to predation, make better hosts for larval parasites. However, trophic level and body length are strongly correlated among fish species, and they may have separate though confounded effects on the parasite fauna exploiting a given species. Helminths show varying levels of host specificity toward their intermediate host when the latter is the downstream host involved in trophic transmission toward an upstream definitive host. Given this broad physiological compatibility of many helminths with fish hosts, our results indicate that fish body length, as a proxy for vulnerability to predators, is a better predictor of their use by helminth larvae than their trophic level based on diet content.     

Magnitude and direction of parasite‐induced phenotypic alterations: a meta‐analysis in acanthocephalans

Several parasite species have the ability to modify their host's phenotype to their own advantage thereby increasing the probability of transmission from one host to another. This phenomenon of host manipulation is interpreted as the expression of a parasite extended phenotype. Manipulative parasites generally affect multiple phenotypic traits in their hosts, although both the extent and adaptive significance of such multidimensionality in host manipulation is still poorly documented. To review the multidimensionality and magnitude of host manipulation, and to understand the causes of variation in trait value alteration, we performed a phylogenetically corrected meta‐analysis, focusing on a model taxon: acanthocephalan parasites. Acanthocephala is a phylum of helminth parasites that use vertebrates as final hosts and invertebrates as intermediate hosts, and is one of the few parasite groups for which manipulation is predicted to be ancestral. We compiled 279 estimates of parasite‐induced alterations in phenotypic trait value, from 81 studies and 13 acanthocephalan species, allocating a sign to effect size estimates according to the direction of alteration favouring parasite transmission, and grouped traits by category. Phylogenetic inertia accounted for a low proportion of variation in effect sizes. The overall average alteration of trait value was moderate and positive when considering the expected effect of alterations on trophic transmission success (signed effect sizes, after the onset of parasite infectivity to the final host). Variation in the alteration of trait value was affected by the category of phenotypic trait, with the largest alterations being reversed taxis/phobia and responses to stimuli, and increased vulnerability to predation, changes to reproductive traits (behavioural or physiological castration) and immunosuppression. Parasite transmission would thereby be facilitated mainly by changing mainly the choice of micro‐habitat and the anti‐predation behaviour of infected hosts, and by promoting energy‐saving strategies in the host. In addition, infection with larval stages not yet infective to definitive hosts (acanthella) tends to induce opposite effects of comparable magnitude to infection with the infective stage (cystacanth), although this result should be considered with caution due to the low number of estimates with acanthella. This analysis raises important issues that should be considered in future studies investigating the adaptive significance of host manipulation, not only in acanthocephalans but also in other taxa. Specifically, the contribution of phenotypic traits to parasite transmission and the range of taxonomic diversity covered deserve thorough attention. In addition, the relationship between behaviour and immunity across parasite developmental stages and host–parasite systems (the neuropsychoimmune hypothesis of host manipulation), still awaits experimental evidence. Most of these issues apply more broadly to reported cases of host manipulation by other groups of parasites.     

Can helminth community patterns be amplified when transferred by predation from intermediate to definitive hosts?

Helminth communities in definitive hosts are formed by the acquisition of packets of larvae arriving each time an intermediate host is consumed. It is thus possible that associations between parasite species or other aspects of community structure get transferred from intermediate to definitive hosts. Earlier computer simulations showed that associations between 2 parasite species, in particular positive associations, could be transferred up the food chain. Here, we alter some of the assumptions of previous models and generate new simulations of several ways in which source infracommunities in intermediate hosts can be transferred to target infracommunities in definitive hosts. In particular, we introduced nonrandom selection of intermediate hosts by predatory definitive hosts, to mimic the phenomenon of host manipulation by parasites; this consisted in biasing predation toward intermediate hosts harboring a certain parasite species. Overall, our results show that positive covariances between 2 parasite species can not only be transferred but can also be amplified during transmission to definitive hosts; significant covariance between parasite species can even appear in the definitive hosts when none existed in the intermediate hosts. Negative covariance was not as readily transferred to definitive hosts and amplified, in part because of properties of the presence-absence covariance index. Amplification of covariance results from intermediate host manipulation as well as from other processes taking place during transmission. These results suggest that the patterns of association between helminth species in definitive hosts cannot be taken to reflect the processes acting inside those hosts: they may simply be inherited, with amplification, from intermediate hosts.     

When should a trophically and vertically transmitted parasite manipulate its intermediate host? The case of Toxoplasma gondii

Parasites with complex life cycles are expected to manipulate the behaviour of their intermediate hosts (IHs), which increase their predation rate and facilitate the transmission to definitive hosts (DHs). This ability, however, is a double-edged sword when the parasite can also be transmitted vertically in the IH. In this situation, as the manipulation of the IH behaviour increases the IH death rate, it conflicts with vertical transmission, which requires healthy and reproducing IHs. The protozoan Toxoplasma gondii, a widespread pathogen, combines both trophic and vertical transmission strategies. Is parasite manipulation of host behaviour still adaptive in this situation? We model the evolution of the IH manipulation by T. gondii to study the conflict between these two routes of transmission under different epidemiological situations. Model outputs show that manipulation is particularly advantageous for virulent strains and in epidemic situations, and that different levels of manipulation may evolve depending on the sex of the IH and the transmission routes considered. These results may help to understand the variability of strain characteristics encountered for T. gondii and may extend to other trophically transmitted parasites.     

Host manipulation of a freshwater crustacean (Gammarus roeseli) by an acanthocephalan parasite (Polymorphus minutus) in a biological invasion context

Several gammarid species serve as intermediate hosts for the acanthocephalan parasite Polymorphus minutus. This parasite influences gammarid behaviour in order to favour transmission to its ultimate host, generally a bird. We investigated this host manipulation in Gammarus roeseli, a gammarid species introduced in France 150 years ago which now coexists with several exotic species from different origins. In the field, vertical distribution of G. roeseli revealed a higher proportion of infected individuals close to the water's surface and the size distribution of infected gammarids revealed predation pressure on infected individuals. However, under laboratory conditions both infected and non-infected individuals remained benthic. The addition of a second gammarid, Dikerogammarus villosus, to the experimental device involved a vertical displacement of infected G. roeseli towards the water's surface. Dikerogammarus villosus, originating from the Ponto-Caspian basin, can be considered as an aggressive predator. The substitution of D. villosus with Atyaephyra desmarestii, a planktivore decapod, did not alter the gammarids' distribution, with both infected and uninfected G. roeseli staying benthic. Thus, biotic interactions between D. villosus and G. roeseli represent selective pressure encouraging the expression of manipulated behaviour in infected amphipods. Through manipulation, P. minutus was found to increase the survival of infected G. roeseli when faced with non-host predators and to make it more vulnerable to predation by the parasite's definitive host.     

Inertia of parasite infection versus host biomass fluctuation

Infection by parasites with complex life cycles such as trematodes depends on many environmental factors which may result in a time-lag between host biomass fluctuations and parasite density in hosts. A cockle (marine bivalve, second intermediate host) population and its associated parasite community were monitored over 15 years. A time-shift correlation analysis suggests that trematode abundance in cockles responds to cockle biomass after a long delay (8 year time-lag). Thus, these parasites can sustainably support a deficit of their intermediate host.     

Parasite-induced change in host behaviour and susceptibility to predation in an eye fluke-fish interaction

Trophically transmitted parasites may increase their transmission efficiency by altering the behaviour of infected hosts to increase their susceptibility to predation by target hosts (the next host in the life cycle). The parasite Diplostomum spathaceum (Trematoda) reduces the vision of its fish intermediate hosts: its metacercariae lodge themselves in the eyes of fish and induce cataract formation, which gives them the opportunity to affect fish behaviour. We examined whether D. spathaceum eye flukes change the preference of fish for the surface layers of the water column or their escape behaviour, which could make the fish more vulnerable to predation by bird hosts. We also studied the influence of parasites on the susceptibility of fish to artificial aerial predators that were able to catch fish from the water surface. Infected and control fish did not differ in their preference for the surface layers but infected fish showed less escape behaviour when a black plate was drawn over the water surface. They were also more easily caught by human ‘predators’ dipping a net into the tank. Thus, infected fish should be easier prey for gulls and terns, implying that the ability of D. spathaceum eye flukes to alter fish behaviour may be a parasite strategy evolved to enhance transmission.     

Manipulation of host behaviour by parasites: ecosystem engineering in the intertidal zone?

Understanding the influence of parasites on the community ecology of free-living organisms is an emerging theme in ecology. The cockle Austrovenus stutchburyi is an abundant mollusc inhabiting the sheltered shores of New Zealand. This species, which lives just few centimetres under the surface, plays a key role for many benthic invertebrate species, because in these habitats the cockle shell is the only available hard surface where invertebrates can establish. However, the behaviour of this cockle can be altered locally by a parasite, the trematode Curtuteria australis. Indeed, heavily infected cockles are unable to bury perfectly and typically lie entirely exposed at the surface of the mud. In this study, we investigated the ecological consequences of this behavioural alteration for two invertebrates species commonly associated with cockles, the anemone Anthopleura aureoradiata and the limpet Notoacmea helmsi. A field study first demonstrated that in both infected and non-infected populations of cockles, there was a negative relationship between the number of anemones and limpets found on cockles. In the laboratory, we showed that predation of limpets by anemones is possible when they share the same cockle shell. In a heavily infected population of cockles, limpets were significantly more frequent and more abundant on cockles manipulated by C. australis than on cockles with a normal behaviour. A colonization test conducted in natural conditions demonstrated that the predominance of limpets on manipulated cockles results from a direct habitat preference. Conversely, anemones were significantly less frequent and less abundant on manipulated cockles than on cockles manipulated by C. australis. A desiccation test revealed that, relative to limpets, they had a lower resistance to this physical stress. We discuss our results in relation to current ideas on ecosystem engineering by organisms.