University of Groningen
Coevolution in host–parasite systems Ashghali Farahani, Sajad
DOI:
10.33612/diss.128654818
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Ashghali Farahani, S. (2020). Coevolution in host–parasite systems: Behavioral strategies of native and invasive hosts. University of Groningen. https://doi.org/10.33612/diss.128654818
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
61
Chapter 5
62
Introduction
Parasites consist of approximately 50% of the total species in ecosystems (Hudson et al., 2006). In aquatic ecosystems their biomass, is greater than the biomass of their predators (Kuris et al., 2008; Lafferty et al., 2008; Preston et al., 2013). Parasites dominate the links in food webs. This means that food webs, which show the flow of energy throughout an ecosystem, include more parasite–host links, e.g. trophic transmission of the parasite from one intermediate host to the definitive host, than predator–prey links (Lafferty et al., 2006). This is true, even though the probability of parasites finishing their life cycle is controlled by several factors, which include intermediate and definitive hosts, as well as non-host predators. Parasites start their development in the intermediate host and can manipulate the physiology and behaviour of their intermediate hosts (Tain et al., 2006b). Parasites complete their life cycle in the definitive host to which the parasites are transmitted by predation of the intermediate host (Tain et al., 2006a). At the end of the life cycle, adult parasites typically mate and reproduce inside the definitive host. The resulting eggs or larvae are transmitted via defecation of the definitive host and subsequent infection of the intermediate host (de Vries and van Langevelde, 2018). The transmission of the parasites to their definitive host is facilitated by non-host predator avoidance by the intermediate host, which would otherwise prevent completion of the parasite’s life cycle (Vielma et al., 2019). Parasites could act as a digestive food source for non-host predator (Johnson et al., 2010; Thieltges et al., 2013) or as a non-digestive food source for a paratenic host predator – a vertebrate host that comes before the definitive host and does not require a certain developmental stage of the parasite (Médoc et al., 2011). Digestion by non-host predators means the end of the parasite’s life cycle and leads to no fitness for the parasite. The selection pressure on parasites to reach their definitive host predator is higher than reaching a non-host predator, because transmission to suitable definitive hosts is mandatory for the survival of the parasite (life-dinner principle, Dawkins and Krebs, 1979). The extinction or low fitness of intermediate or definitive hosts or both could lead to alternative mechanisms exhibited by parasites to increase their survival and the completion of their life cycle. For example, parasites have changed their intermediate hosts within the same family over evolutionary history (Lyndon, 2001), expanded the diversity of definitive host predators (Vanacker et al., 2012), or increased intermediate hosts abilities upon non-host predators avoidance behaviour (Médoc and Beisel, 2008) resulting in higher survival of parasites, completion of their life cycle and success in parasite colonization. Parasites have the ability to acquire both novel native and invasive intermediate hosts (Bauer et al., 2005). I discussed in first chapter about response traits of native species to invasive host-species and effect of parasites on native and invasive host-fitness. The arrival of invasive host species in a native host population may promote local parasite maladaptation (Moret et al., 2007). The main aim of my thesis was to study the ability of parasites to alter the native and invasive intermediate host’s behavior. I tested whether the “increased host ability” hypothesis in
non-host predators, was more pronounced in native, sympatric co-evolved gammarids in comparison to sympatric invasive gammarids. In this dissertation, I performed an experimental study to investigate the effect of infection by parasites on changes in behaviour of invasive and native intermediate hosts. I studied the thorny-headed worm parasite Polymorphus minutus and its native intermediate hosts Gammarus pulex, G. fossarum and invasive Echinogammarus berilloni on the Paderborn Plateau (Westphalia, Germany). P. minutus, comprises morphologically similar, but genetically divergent subspecies, in Westphalia (Grabner et al., 2020). In Westphalia, the P. minutus co-invaded the region with invasive gammarids, E. berilloni and G. roeseli from the Mediterranean area and Southeast Europe, respectively (Zittel et al., 2018). G. roeseli transmitted this invasive parasite to G. pulex. However, the native cryptic parasite P. minutus uses native G. fossarum as an
63 intermediate host (Zittel et al., 2018). Perhaps native P. minutus can potentially have different traits than invasive P. minutus in manipulating the intermediate host behaviour in such a way that it is increasing the chances of being transmitted to the definitive host. In chapter two, we investigated the effect of parasite infection on native and invasive intermediate hosts in non-host predator avoidance behaviour. Our findings revealed that native P. minutus were more successful in manipulating native G. fossarum behaviour to avoid the non-host predator, three-spined sticklebacks Gasterosteus aculeatus, than invasive P. minutus were in manipulating G. pulex and E. berilloni In chapter three, my coworkers and I investigated the role of the native and invasive parasite on rheotaxis – to swim against or with the current - in the intermediate hosts. The lack of differences in rheotaxis between infected native and invasive hosts may indicate that P. minutus enhances transmission to its definitive host, by making the intermediate host more stationary and hence more easily preyed upon by the definitive host. In the previous chapters, I described the importance of studies on invasive intermediate host species and its’ impact on parasite prevalence in food webs. I focused on the ecological effects in invasive gammarids in Europe. In chapter four, I assessed the effect of parasites on the survival of native intermediate hosts in water with different salinities. I showed that survival rate in infected, native gammarids was higher than that of uninfected native gammarids at higher salinities. This means that infected native gammarids who have drifted to downstream brackish waters, where the habitat of water fowl mostly is, have a chance to survive, and transmit the parasite successfully to definitive hosts. The key question in all chapters was if the ability of P. minutus to alter its intermediate host’s behaviour has evolved specifically to target sympatric gammarids, or gammarids in general.
In this chapter, I will assess the research findings presented in this thesis in an evolutionary framework, and discuss the effect of the parasites on their sympatric native and invasive intermediate host survival and behaviour. Indeed, invasive species are interesting as “real-time” systems to study the succession of “new” species’ establishment in invaded ecosystems. A fundamental question is whether parasites may play a role in the successful establishment of invasive gammarids? For example, are invasive gammarids more successful because they are able to avoid infection by native cryptic P. minutus than native gammarids (Zittel et al., 2018)? Below, I will discuss the success of infection and invasion in aquatic ecosystem.
1. Success of infection
The success of infection by parasites during the host-parasite coevolution is related to increasing a parasite’s ability to alter intermediate host behaviour, resulting in higher fitness of parasites. In the first part of this chapter, I will discuss the parasite’s ability to alter the behaviour of the intermediate host in sympatric host species. Second, I will discuss why the infected gammarids actively swim against the water current (positive rheotaxis). Positive rheotaxis will also decrease downstream drifting rate of the intermediate host, where most of the non-host drifting predators are present. Karst streams (intermittent streams with temporal and spatially varying discharge regime) on Paderborn Plateau, i.e. the upper catchment of the Lippe, are susceptible to summer droughts and consequently have elevated salinity during this period. Positive rheotaxis is one mechanism to keep permanent up streams as reservoirs of parasite during the summer time. It is vital for parasite in infected gammarids to cope with higher salinity levels, both in drought area and downstream, compared to uninfected gammarids. Third, I will discuss the differences in survival rates observed in infected gammarids in different degrees of salinity. Finally, I will discuss these observations in terms of ‘adaptation’ and ‘exaptation’ in host-parasite co-evolution.
64
1-1. Increased sympatric intermediate hosts’ abilities hypothesis
Previous studies have demonstrated that parasites decrease non-host predator exposure by manipulating the behaviour of their intermediate host (Holomuzki et al., 1988; Rigby and Jokela, 2000; Matz and Kjelleberg, 2005; Friman et al., 2009). To reach optimal fitness a parasite must keep their infected intermediate hosts in an optimal balance between survival - i.e. protection from non-host predator - and foraging - i.e. seeking food or mate (Dianne et al., 2011, 2014). The “increased host abilities” hypothesis posits that parasites have evolved in response to selection pressures on transmission to the next host, i.e. the ability to manipulate the behaviour of the intermediate hosts in such a way that it increases the chance of transmission (Medoc and Beisel, 2008; Beisel and Médoc, 2010).
I introduced two aspects of predation in chapter two: (i) efficiency of predation on native versus invasive intermediate host species, and (ii) parasite-induced changes in non-host predator avoidance behaviour. Native gammarids have been living for a substantially longer time in sympatry with native parasites than invasive gammarids. I tested whether the
“increased host abilities” hypothesis, was higher in native, co-evolved gammarids in comparison to invasive gammarids. To test this hypothesis, I used P. minutus, an acanthocephalan parasite, two native intermediate hosts, G. pulex and G. fossarum, and one invasive host, E. berilloni.
The changes in behaviour observed in infected G. fossarum were consistent with the “increased host abilities” hypothesis. Our results indicated that infected native G. pulex and invasive E. berilloni avoided cues of three-spined stickle backs (i.e. chemical cues from mucous of fish skin or pheromone) significantly less compared to infected native G. fossarum. The results suggested that native cryptic P. minutus and native G. fossarum both are able to increase their transmission by avoiding predation by a non- definitive host species, thereby indirectly promoting transmission success to the definitive hosts.
My study revealed that invasive E. berilloni do not seem to detect or avoid fish cues, irrespective of the level of infection with the P. minutus. My findings indicated that there was a difference in avoidance behaviour of infected E. berilloni on the Paderborn Plateau in comparison to their native origin area in France. For example, native E. berilloni infected by native P. minutus from the Lunain river (France) reduced their activity in the presence of fish cues (Jacquin et al., 2014). However, native G. pulex infected by invasive, cryptic P. minutus , transmitted from G. roeseli to them (Zittel et al., 2018), had a tendency of swimming towards fish cues.
1-2. Avoidance of infected intermediate host by drifting
Focusing on the comparison of rheotaxis behaviour of infected versus uninfected gammarids of different species, my results showed a high degree of positive rheotaxis in infected species. This positive rheotaxis means that infected intermediate gammarid hosts in all species avoid drifting. My findings were consistent with those reported by Macneil et al (2003) who found a higher prevalence of parasitism in the faster and shallower water bodies (upstream) compared to slower and deeper areas (downstream). Furthermore, I found that the effect of infection on rheotaxis behaviour in all my experimental gammarids species were similar (Chapter 3). Rheotaxis and geotaxis (Cezilly et al., 2000) are both mechanisms potentially favouring the trophic transition of the parasite to its definitive host (Lafferty, 1999). Overall, infection with P. minutus appeared to increase rheotaxis.
65
1-3. Survival rates of infected and uninfected gammarids in relation to salinity
It is essential for parasites to develop an effective system to cope with extreme variations in osmolarity. Osmolarity means the total number of solute particles per liter. Parasites move from an aquatic environment into the body of the intermediate host and afterwards to a terrestrial definitive host (Huang et al., 2016b). Parasites increase their own fitness by avoiding environments with high variations in salinity levels (Piscart et al., 2007). I showed that gammarids infected with P. minutus survived at higher rates compared to uninfected gammarids at higher salinity (Chapter 4). P. minutus thus increase the survival rate of its intermediate host under different salinities and hence the survival of itself. Parasites and their intermediate hosts are at lower trophic levels than the definitive hosts. Species at higher trophic levels typically have lower environmental tolerances compared to species at lower trophic levels (Montserrat et al., 2013). Moreover, infected aquatic organisms not only tend to be more tolerant to elevated salinity but also to global climate changes (e.g. acidification) than uninfected organisms (MacLeod and Poulin, 2016).
Parasites have been shown to alter the activity level of neuroendocrine factors (i.e. serotonin) in the brain of their intermediate hosts during developmental stage of parasites (Tain et al., 2006, Cézilly and Perrot-Minnot, 2010, Cézilly et al., 2013). Serotonin is a neurotransmitter
that affects water and salt movements in crustacean gills through osmoregulatory mechanisms, which in turn may make intermediate host more tolerant to higher salinity levels (Péqueux, 1995; Liu et al., 2008). Serotonin plays a key role in several behavioral traits, including thermotaxic behavior -movement of an organism in response to temperature- (Li et al., 2013; Wong and Rankin, 2019), phototaxis behavior -movement of an organism in response to light- (McPhee and Wilkens, 1989; Tain et al., 2006a; Rodriguez Moncalvo and Campos, 2009; Thamm et al., 2010), the geotaxis -swimming of an organism to the top or bottom of the water column- (Maximino et al., 2013), feeding behavior (Alanärä et al., 1998; Ortega et al., 2013), and oxygen consumption (Srinivasan et al., 2008; Pérez-Campos et al., 2012).
I interpreted my findings in the light of an adaptive scenario. It is necessary to further investigate the effects of salinity in sympatric or allopatric, infected species in terms of physiological plasticity, e.g. parasites might alter the trade-off between survival and reproduction in gammarids (Dunn et al., 2012).
1-4. Plasticity and consistency of behaviour differs between infected and uninfected gammarids
Recent studies showed that behavioural plasticity (individual differences in behaviour) and consistency (consistent among-individual variation in behaviour) of intermediate hosts differed among both host sex and infection status. For example, the male aquatic isopod Caecidotea intermedius showed higher plasticity for refuge use in both uninfected and infected individuals that is infected by the parasite Acanthocephalus dirus, and higher consistency for refuge use only in infected individuals. In contrast, female aquatic isopod exhibited plasticity in uninfected and consistency in infected individuals, in activity (Park and Sparkes, 2017). I showed that avoidance from non-host predator’s cues was exhibited consistency in infected individuals of G. fossarum (Chapter 2). Consistency of rheotaxis and salinity tolerance was present in infected individuals, but not in uninfected individuals of gammarids (Chapter 3 and Chapter 4). My results indicate that P. minutus infection may shape personalities in sympatric gammarids by increasing consistency of mentioned traits. P.
66 minutus appear to be able to exploit plasticity in sympatric intermediate host antipredator responses toward its own advantage.
P. minutus adopts different strategies to exploit native and invasive intermediate gammarids hosts. One outcome of this adaptation could be that the parasites modify non-host predator avoidance behaviour in native G. fossarum but not in invasive E. berilloni. Consistent with “increased intermediate host abilities” hypothesis, infected invasive gammarids showed less drift compared to uninfected invasive species. Differences in ‘personality’ and behavioural plasticity between native and invasive species contribute to invasion success (Bierbach et al., 2016). Personality traits affect dispersal tendencies of invasive species (Chapple et al., 2012). A highly adaptable, invasive species may survive and reproduce thereby facilitating colonization and establishment (Plantamp et al., 2019). Contrary to my expectation, the invasive uninfected E. berillonoi drifted more than uninfected native G. pulex and G. fossarum.
1-5. Modulation of host behaviour evolved through both ‘adaptation’ and ‘exaptation’
The ultimate cause of “increased parasite fitness through transmission to a definitive host-predator” occurs in combination with proximate causes of the success or failure of infected intermediate hosts, such as avoidance of non-host predators (Chapter 2), positive rheotaxis (Chapter 3) and increased salinity tolerance of the intermediate hosts (Chapter 4). Evidence in support of my results suggested that modulation of intermediate host behaviour has evolved through ‘adaptation’ (Baldauf et al., 2007) and/or ‘exaptation’ by parasites (Cézilly and Perrot-Minnot, 2005; Combes, 2005; Beisel and Médoc, 2010). This trait can potentially be explained as an exaptation of a parasite manipulating it’s intermadite host behaviour in a manner that places the intermediate host in the vicinity of the definitive host. Hence, parasite evolution may be explained by a shift in function, from regulation of survival in the intermediate host to reproduction in the defenitive host. Predation of infected intermediate hosts by the definitive hosts, is crucial for the parasite to complete its life cycle (Harvey and Pagel, 1991). However, modulating the intermediate host’s behaviour is not always adaptive for the parasite (Bakker et al., 2017). For example, acanthocephalan parasite Pomphorhynchus laevis manipulates the behaviour of its native intermediate host, G. pulex, but is unable to manipulate the behaviour of the invasive G. roeseli (Moret et al., 2007). On the other hand, exaptation may also occur, when behavioural traits of infected intermediate host by parasite ultimately facilitates trophic transmission to the definitive host by natural selection (Cezilly et al., 2012).
As emphasized at several occassions, natural selection in a host-parasite system may not necessarily target host traits directly, but instead on the ability of parasites to alter hosts traits in a manner enhancing the trophic transmission of the parasite (Seppälä and Jokela, 2008; Seppälä et al., 2008; Cézilly and Perrot-Minnot, 2010).This means that the host –parasite coevolution is directly related to the concept of the “extended phenotype” introduced by Richard Dawkins in 1982 (Seppälä and Jokela, 2008; Seppälä et al., 2008; Cézilly and Perrot-Minnot, 2010).Examples of exaptation are: avoidance of non-definitive host predators, positive rheotaxis and increased salinity tolerance that evolved both in adaptation and exaptation by natural selection in response to predation, flooding and salinization.
67
2. Factors association with invasion success to dominate invaded
communities by invasive species compared to native species
There are behavioural differences between native and invasive gammarids, e.g. less activity in invasive species (Piscart et al., 2011a). Factors known to affect the invasion progress of invasive species compared to native species, resulting in a wider distribution in aquatic ecosystems compared to native species, are: parasitic infection (Comont et al., 2014; Haelewaters et al., 2017), presence of predators (Jermacz and Kobak, 2017), river type (Boets et al., 2015), physical-chemical and hydro-morphological variables such as conductivity, orthophosphate and dissolved oxygen concentration (Boets et al., 2013), artificial hard substrates (Van Riel et al., 2006), local adaptation (Dlugosch and Parker, 2008; Colautti and Lau, 2015), rapid adaptation of invasive species to new environments (Whitney and Gabler, 2008; Sotka et al., 2018; Stutz et al., 2018) and increased competitive ability compared to native species (Blossey and Nötzold, 1995). However, the behavioural traits that enhance invasion success, such as fast growth and high reproductive rates in invasive species (‘invasion syndrome’, Chapple et al., 2012), have received less attention compared to research on native species. Behavioural differences between native and invasive species may contribute to enhances in invasion success of invasive species. Successful invasion means that invasive species were able to take advantage of available resources and outcompete local populations, e.g. through decreased competition from native species or increased resources through eutrophication from nutrient discharge (Davis et al., 2000; Wikström and Hillebrand, 2012).
In the second part of current chapter, I will discuss the role of infection of intermediate host on avoidance of non-definitive host predators. Second, I will discuss the mechanism of higher negative rheotaxis in invasive species compared to native species. Third, I will discuss the negative effect of increasing salinity on survival rates in freshwater gammarids. Finally, I will discuss the competitive exclusion mechanism of related invasive and native species.
2-1. Gammarids do not appear to avoid non-efficient predators
Invasion can be facilitated if the invasive species are less susceptible to non-definitive host predators than native species (Schmidt-Drewello et al., 2016). I have tested non-definitive host predator avoidance behaviour of uninfected native and invasive gammarids from three-spined stickleback cues in an experimental setting. I showed that uninfected individuals of both species preferentially swim towards fish cues. This behaviour was especially noticeable in the native gammarids compared to the invasive gammarids. In contrast, invasive gammarids reduced their locomotion activity in the presence of fish cues (Chapter 2). The main question is why would the absence of avoidance from fish (predator) cues be higher in uninfected native gammarids compared to uninfected invasive gammarids?
Uninfected gammarids appear able to discriminate between the chemical cues emitted by scavengers or inefficient predatory species (e.g. three-spined sticklebacks (Jacquin et al., 2014) and those emitted by the efficient predator, the trout, (Abjornsson et al., 2000)). Moreover, fish cues could act as indications of food resources to gammarids. The gammarids may behave as omnivore scavengers in aquatic ecosystems (Wilhelm and Schindler, 2011; Jermacz et al., 2017), whereas the three-spined sticklebacks used in my experiment, were not fed with gammarids or any commercial food in order to remove cues stemming from conspecific injuries, residual food and predator faces (Wisenden et al., 1999; Ward, 2012; Smith and Webster, 2015). It is likely that the signal produced by food is only one of many cues that predator relies upon.
68 Invasive species often show greater behavioural plasticity compared with native species in order to increase their ability to evade predation (Guo et al., 2019). For example, invasive species may be better at predator recognition. The ability of invasive species to avoid predators has been demonstrated in several previous studies, for example invasive gammarids reduce their time spent in open water in the presence of fish cues (Jermacz and Kobak, 2017). My results (Chapter 3) revealed that uninfected invasive gammarids compared to native gammarids on the Paderborn plateau limit their movements to the upstream area where the efficient predators prevail (e.g. sculpin, Cottus gobio and trout, Salmo trutta (Abjornsson et al., 2000)). The native G. pulex is an active and explorative amphipod (Truhlar and Aldridge, 2015). G. pulex shows local adaptation in the presence of the non-efficient predator. G. pulex lives in shallow muddy streams, where low water depth and dense benthic detritus may reduce the risk of detection by fish (Haddaway et al., 2014).
2-2. High drifting behaviour in invasive species.
My study detected higher levels of negative rheotaxis (drifting) in invasive gammarids compared to native gammarids (chapter 3). Active drifting is the primary anti-predation mechanism (Naman et al., 2016). Individuals that passively drift constitute the primary prey for drift-foraging predators, such as salmonids (Nielsen, 1992). G. pulex, is an abundant amphipod in the lower (Piscart et al., 2010) and middle parts (Karaman and Pinkster, 1977; Siegismund and Müller, 1991) of small streams with slow flow or standing waters (Verberk et al., 2018). G. pulex has evolved living in shallow muddy streams (Haddaway et al., 2014) and drought area’s (Meyer et al., 2004) where there is low water depth. It isexpected that G. pulex use both positive and negative rheotaxis for leaving the areas with high degree of drought. G. pulex use mechanisms such as recruitment with flood (negative rheotaxis) in rainy seasons to compensate their lost population in dry area. Positive rheotaxis is necessary for G. fossarum to compensate their drift ed population from the upper parts of streams or in the mountainous areas (Scheepmaker and van Dalfsen, 1989; Siegismund and Müller, 1991; Müller 2000; Pöckl et al., 2003). Rheotaxis may enable invasive gammarids to avoid severe temporary changes in environmental conditions thereby facilitate the response to a wide range of environmental stressors, such as predation, pollution and erosional habitats (Meijering, 1991; Marvier et al., 2004; Hänfling et al., 2011; Macneil and Dick, 2014). The invasive species may leave the areas with high environmental stressors and search for a new foraging area with fewer environmental stressors.
Ecologists have divided rheotaxis into two categories: (1) active rheotaxis; behavioural and continuous as a form of active patch selection when gammarids’ predation risk is high and food availability is low (Naman et al., 2016), (2) passive rheotaxis; constant and catastrophic as a means of passive response to high (hydropeaking) flow conditions, when macro invertebrates, i.e. gammarids, cannot hold on to the bottom substrate as a result of floods (Bruno et al., 2016). I conducted my experiment without presence of predators or food in order to assess the effects of water velocity on rheotaxis, only. The water velocity in a circulatory canal was kept at a rate of ca. 25 cm per second, which is similar to the water velocity at the sampling site. I detected higher levels of negative rheotaxis in invasive E. berilloni compared to native gammarids. However, my experimental setup was unsuitable to discern between active or passive rheotaxis. I recommend another experiment with different water velocities to further gain insights into active or passive rheotaxis. Rheotaxis is necessary for gammarids reproduction and community structure during mating seasons during the late spring and summer. The middle part of the Paderborn Plateau was almost completely dry during the summer. Such a severe drought season could lead to high drift of E. berilloni.
69 E. berilloni could increase the connectivity between downstream to upstream con-specific populations via rheotaxis.
2-3. Lower survival of freshwater gammarids in saline waters
Survival rates are decreased in freshwater amphipods when osmoregulation is disturbed and the hemolymph osmolality increases above a critical point (Hart et al., 1991). Survival, locomotion and ventilation are all impacted by salinity stress in gammarids (Sronom et al., 2010). As expected, my study showed that increasing salinity had an adverse effect on the survival rates in G. pulex and G. fossarum (Chapter 4). Most gammarids are hyper-regulators of body ion concentrations at low salinity levels (Piscart et al., 2007). The lowered ability of freshwater gammarids to regulate chloride, and the decreasing intra- and extracellular free amino acids, resulted in a lower survival rate in gammarids at elevated salinity (Sutcliffe, 1971; Jordan et al., 1999; Patrick and Bradley, 2000; Pickens et al., 2019). It has also been suggested that salinity tolerance may differ with sex and age due to difference in metabolic rates (Schmidt-Nielsen, 1984).
Haemolymph volume regulation in invertebrate animals depends on the balance between water uptake via osmotic water influx, drinking, and loss via urination. For example, in the crab Cancer borealis, if the haemolymph volume during moulting rises, the drinking rate decreases and urination rate increases to compensate ion exchange (Greco et al., 1986). Although invasive gammarids have a wider salinity tolerance compared to native gammarids (Holopainen et al., 2016), it is currently unknown whether the tolerance of the invasive E. berilloni is any different from native G. pulex and G. fossarum. However, it is possible that invasive species could regulate their haemolymph at a higher salinity compared to native species.
2-4. Realized niches and survival of native gammarid species
My findings revealed no significant differences in salinity tolerance between the native gammarids sampled on the Paderborn Plateau (chapter 4). The ecological similar salinity tolerance in the two native species G. pulex and G. fossarum, which overlap in their realized niches resulted in an increased likelihood of competitive exclusion (Cardoso et al., 2018). It is clear that there are fitness differences among the two native species during drought seasons and invasion by E. berilloni (Meyer et al., 2004). The drought seasons is observed during summer and late spring. The salinity levels during a drought are far below the salinity during our experiments.There is overlap in the realized niches of G. pulex and E. berilloni ( see below; Meyer et al., 2004). Both native gammarids have equal salinity tolerance, but G. pulex occurs in high salty and dried out areas and G. fossarum not. The competitive exclusion by G. fossarum appeared to increase their fitness and abundance. G. fossarum seems to be better able to cope with changing environments: (i) my findings showed increasing salinity tolerance in G. fossarum and not in G. pulex; (ii) parental exposure induce transgenerational plasticity in G. fossarum and better acclimation to contaminants (Vigneron et al., 2019).
Several previous studies have shown an effect of temperature and salinity on the distribution of organisms in aquatic ecosystems (Einarson, 1993; Neuparth et al., 2002; Delgado et al., 2011). The drought season may change the number of available niches to both cold-adapted, native co-occurring species, and may limit the protrusion of warm-adapted invasive species. More research about genome wide data is needed, in order to detect possible signs of selective sweeps between two populations, both in native and invasive, in temporary and permanent
70 streams of the Paderborn plateau and could provide precise clarification about salinity tolerance of gammarids with different temperature preferences.
2- 5. Invasive species win anti-predator behavior: Competitive, but not comprehensive, exclusion
Some invasive species completely replace native species through competitive exclusion (Mooney and Cleland, 2001). My findings showed that invasive species compared to native species reduce their locomotion activity in the presence of non-efficient predators (i.e. three spined sticklebacks), resulting in a lowered predation risk (Chapter 2). Higher predation rates of native gammarids by three spined sticklebacks compared to invasive species on the Paderborn Plateau provides advantages for invasive E. berilloni from sympatric occurrence with native gammarid species, especially when refuge options, such as leaf disks, are available (Schmidt-Drewello et al., 2016). These two findings show competitive exclusion between invasive E. berilloni and native G. pulex and G. fossarum as a consequence of invasive species’ success. It is difficult to draw exact conclusions about the impacts of invasive species on native species either through direct (i.e. resource) or indirect (i.e. apparent) competition. Indirect competition is, for example, when an invasive species, as novel prey species, enhances growth rates of the predator population. Native gammarids are more heavily predated upon large predators relative to invasive species. This indirect competition could be more harmful to native species than direct competition (Noonburg and Byers, 2005).
In my study system, home ranges differed among species: E. berilloni was more prevalent down streams than native species (Chapter 3). Nevertheless, Meyer et al. (2004) has
demonstrated that both native and invasive gammarids on Paderborn Plateau are well adapted to share the same environment in the middle stream (i.e. the drought regions) despite similar ecological traits. It seems that the coexistence between native and invasive gammarids could be achieved through a combination of competition and habitat specialization (Baumart et al., 2015). E. berilloni was reported on Paderborn Plateau for the first time in 1925 (Schellenberg, 1925; Boecker, 1926). It could be during the initial phase of an invasion by E. berilloni in this ecosystem. Native species may share habitat with invasive species during the initial phases of an invasions, and eventually be partly of fully replaced by the invasive species.
Conservation strategies and environmentally friendly cultured
fish-domestic water fowls farms in Paderborn plateau (Westphalia, Germany)
In heterogeneous and temporarily highly variable environments, such as the Paderborn Plateau waterways, invasive species have higher fitness (Luo et al., 2019b) and a wider range of phenotypic plasticity compared to native species (Agrawal, 2001; Davidson et al., 2011; Szabó et al., 2018). The potential role of invasive species in providing ecosystem services will increase in a scenario of climate change. Furthermore, increasing temperatures that are due to global warming could facilitate the invasion and establishment of invasive species originating from warmer areas, e.g. E. berilloni originate from Spain and France (Montserrat et al., 2013). The probability of successful invasion is positively related to the number of introductions, and high genetic diversity of invasive species. An invasive species with high levels of genetic diversity is presumed able to adapt rapidly to environmental change (Riis et al., 2010). Conservation strategies can focus on preventing the establishment of invasive species in new locations, or reduce the expansion of already present invasive species to new locations (Drury and Rothlisberger, 2008; Stewart-Koster et al., 2015). However, there are contrasting views in71 regards to the threats that invasive species have on biodiversity (Douglas et al., 1990; Wilson and Pärtel, 2003; Renne et al., 2006). For example, invasive perennial grass Agropyron cristatum (crested wheatgrass) would enhance biodiversity and increase belowground carbon storage in the northern Great Plains in USA and Canada (Wilson and Pärtel, 2003). One point of view is that biological invasions should be understood as a socio-ecological phenomenon (Carballo-Cárdenas, 2015). Negative consequences of invasive species for the future of native habitats are biodiversity loss (Perrings, 2002) or altering the structure and functioning of ecosystems (Hooper et al., 2005). In some instances, invasive species can also have positive effects on an ecosystem, e.g. invasive species can provide shelter and habitats for native species (Wonham et al., 2005). Invasive species can provide prey for endangered predators (Tablado et al., 2010). Invasive species act as catalysts for ecosystem restoration (Ewel and Putz, 2004) or as biofilters (Elliott et al., 2008).
Waterfowl migrate over large distances compared to other vertebrates (Clausen et al., 2002) acting as long-range vectors of parasite dispersal (Clausen et al., 2002). Water fowl act as a key dispersal vector for spreading two populations of native (upstream) and invasive (middle and downstream) P. minutus in Paderborn Plateau.
The karst streams, e.g. upper catchment of the Lippe in the Paderborn Plateau, include high relevant stress factors for aquatic organisms, such as inflow and seasonal droughts (Meyer et al., 2004). Drought will cause desiccation of aquatic organisms. The fitness consequences of stress factors for aquatic organisms in karst streams (e.g. reduced immune system, growth and mortality due to temperature, osmotic and hypoxic stress, losing the habitat due to hydraulic stress, disease due to bacteria, viruses, fungi and parasites as biotic stress) do not provide a suitable system for movement of gammarids, but mechanisms like rheotaxis and local and regional diversity contribute to the gammarids population connectivity, and this makes it a suitable habitat for them (Schofield et al., 2018).
P. minutus does not pose a threat to cultured fish, a non-host predator, but P. minutus parasite has been shown to result in septicemia in the intestines of domestic water fowl (Crompton and Harrison, 1965). Parasite ecology could help reduce the anthropogenic impact with a strategy similar to the conservation strategy on the Paderborn Plateau, e.g. connecting water resources through channelization where less waterfowl and parasites are present. Control of acanthocephaliasis in waterfowl farms, e.g. not having poultry activity near the P. minutus area, is necessary. P. minutus has been detected in poultry, which have been kept with ducks (Lingard and Crompton, 1972). Waste from poultry and domestic waterfowl into ecosystems housing wild population have shown to cause a high prevalence of acanthocephaliasis in wild waterfowl (Lingard and Crompton, 1972). One strategy could be to farm non-host fish species thereby decreasing the incidence of P. minutus in farmed water fowl. Considering the predation of farmed fish species and the gammarid species used for live feed will be beneficial to aquaculture. Use of behavioural traits of gammarids in aquaculture (e.g. rheotaxis and avoidance from a non-host predator), which could have economic and ecological benefits for aquaculture activities, could improve methods of feeding technology by live foods, e.g. gammarids. For example, high negative rheotaxis of invasive gammarids makes them a suitable live feed in non-intensive fish farming using -drift feeding fish, such as Atlantic salmon Salmo salar and rainbow trout Oncorhynchus mykiss, that feed facing the water flow. A reduced anti-predatory avoidance in native gammarids makes them more suitable as live feed in circular ponds used for intensive fish farming (e.g. salmon fish farms). Using live feed could reduce the negative environmental effects associated with feed pellets. In addition, increasing the temperature and salinity would decrease acanthocephaliasis in
72 farmed fish and hence reduce the need for chemical or antibiotic treatments. My findings should be considered in the development of environmentally friendly aquaculture and poultry systems in the future.
Constructing artificial, small salt water ponds near water fowl freshwater habitats could reduce ectoparasite load as well as the occurrence of intermediate hosts that attach to waterfowl feathers. Salt water ponds also negatively affect P. minutus access to intermediate hosts.
Conclusions and recommendations for future research
Assessing the effect of non-host predators on avoidance behaviour of intermediate hosts showed that the type of predator (i.e. efficient and efficient or definitive and non-definitive predator) may play an important role in the regulation of native and invasive amphipod populations via parasite populations (Abjornsson et al., 2000; Baldauf et al., 2007; Jacquin et al., 2014; Szokoli et al., 2015b). The parasite P. minutus studied here has co-evolved with sympatric native gammarids, and not with invasive gammarids, which has resulted to a system to promote avoidance behaviour in native gammarids towards non-host predators. Furthermore, we found that the effects of infection in rheotaxis in native and invasive species is not different and there was no differences in salinity tolerance between infected native gammarids (Chapters 3 and 4). It is important to note, however that, the interaction of the parasite with non-host predators, co-inhabiting the same ecosystems, will influence infection dynamics in final hosts (Roberts and Heesterbeek, 2013).
Parasite populations are regulated by predation. My results revealed that the proximate causes of the success or failure of infected intermediate hosts on anti-predatory behaviour (Chapter 2). These proximate causes seem to point towards effects on the regulation of host predator populations through ultimate cause: increased sympatric intermediate hosts’ ability by parasites. The invasive, intermediate host also affects the availability of native prey and predators. It has been estimated that 2.5% of the parasite population in intermediate hosts with manipulated behaviour are transmitted successfully, whereas 17.1% are lost to non-host or paratenic host predators (Mouritsen and Poulin, 2003). Stomach content analyses of waterfowl in future studies could improve our knowledge of digested or not digested parasites and the relative preferences for invasive and native gammarids. Such analyses could confirm whether parasite-induced trophic transmission is successful.
The lower susceptibility to non-final host predators and increased negative rheotaxis behaviour of invasive species could lead to occupation of the whole ecological niche by invasive parasite and intermediate host species due to the competitive exclusion. Predicting invasion success without any consideration of host-parasite co-evolution is impossible. I must emphasize that invasion biology complex and highly context-dependent making generalizations and hence predictability difficult (Elliott-Graves, 2016). According to Daehler (2003), invasive species exhibit a higher degree of phenotypic plasticity compared to native species occupying the same niche. The elevated phenotypical plasticity of invasive species must be incorporated into predictive models of invasive species distribution (Clements and Ditommaso, 2011). Also, the evolutionary potential of invasive species should be considered in future research. Future studies, therefore, appear to be essential for a better prediction of the long-term consequences of biological invasions for restoration and conservation of the waterways on worldwide Plateau. Further sampling in different seasons as well as laboratory experiments of predators’ stomach would be required to describe the dietary preferences of the final host predators and non-host predators in more detail. I recommend future studies to
73 focus on final host behaviour in relation to infected intermediate hosts and their migration patterns as a consequence of global warming. The increased water temperatures due to global warming may, in turn, affect infected intermediate hosts’ temperature tolerance and increase acanthocephaliasis in fish or domestic waterfowl farm. Consequently, the need for chemical or antibiotic treatments is likely to increase.
