Fish & Shellfish Immunology 36 (2014) 130e140 Contents lists available at ScienceDirect Fish & Shellfish Immunology journal homepage: www.elsevier.com/locate/fsi Full length article In vitro leukocyte response of three-spined sticklebacks (Gasterosteus aculeatus) to helminth parasite antigens Frederik Franke a, Anna K. Rahn b, Janine Dittmar a, Noémie Erin c, Jennifer K. Rieger d, David Haase d, Irene E. Samonte-Padilla c, Joseph Lange a, Per J. Jakobsen e, Miguel Hermida f, Carlos Fernández f, Joachim Kurtz a, Theo C.M. Bakker b, Thorsten B.H. Reusch d, Martin Kalbe c, Jörn P. Scharsack a, * a Department of Animal Evolutionary Ecology, Institute for Evolution and Biodiversity, University of Münster, Hüfferstrasse 1, 48149 Münster, Germany Institute for Evolutionary Biology and Ecology, University of Bonn, An der Immenburg 1, 53121 Bonn, Germany c Department of Evolutionary Ecology, Max-Planck Institute of Evolutionary Biology, August-Thienemann Str 2, 24306 Plön, Germany d Department of Evolutionary Ecology of Marine Fishes, GEOMAR Helmholtz Centre for Ocean Research, Düsternbrooker Weg 20, 24105 Kiel, Germany e Institute for Biology, University of Bergen, Thor Møhlensgate 55, 5020 Bergen, Norway f Departamento de Xenetica, Facultade de Veterinaria, Universidade de Santiago de Compostela, Campus de Lugo, 27002 Lugo, Spain b a r t i c l e i n f o a b s t r a c t Article history: Received 22 July 2013 Received in revised form 18 October 2013 Accepted 21 October 2013 Available online 29 October 2013 Helminth parasites of teleost fish have evolved strategies to evade and manipulate the immune responses of their hosts. Responsiveness of fish host immunity to helminth antigens may therefore vary depending on the degree of host-parasite counter-adaptation. Generalist parasites, infective for a number of host species, might be unable to adapt optimally to the immune system of a certain host species, while specialist parasites might display high levels of adaptation to a particular host species. The degree of adaptations may further differ between sympatric and allopatric host-parasite combinations. Here, we test these hypotheses by in vitro exposure of head kidney leukocytes from three-spined sticklebacks (Gasterosteus aculeatus) to antigens from parasites with a broad fish host range (Diplostomum pseudospathaceum, Triaenophorus nodulosus), a specific fish parasite of cyprinids (Ligula intestinalis) and parasites highly specific only to a single fish species as second intermediate host (Schistocephalus pungitii, which does not infect G. aculeatus, and Schistocephalus solidus, infecting G. aculeatus). In vitro responses of stickleback leukocytes to S. solidus antigens from six European populations, with S. solidus prevalence from <1% to 66% were tested in a fully crossed experimental design. Leukocyte cultures were analysed by means of flow cytometry and a chemiluminescence assay to quantify respiratory burst activity. We detected decreasing magnitudes of in vitro responses to antigens from generalist to specialist parasites and among specialists, from parasites that do not infect G. aculeatus to a G. aculeatus-infecting species. Generalist parasites seem to maintain their ability to infect different host species at the costs of relatively higher immunogenicity compared to specialist parasites. In a comparison of sympatric and allopatric combinations of stickleback leukocytes and antigens from S. solidus, magnitudes of in vitro responses were dependent on the prevalence of the parasite in the population of origin, rather than on sympatry. Antigens from Norwegian (prevalence 30e50%) and Spanish (40e66%) S. solidus induced generally higher in vitro responses compared to S. solidus from two German (<1%) populations. Likewise, leukocytes from stickleback populations with a high S. solidus prevalence showed higher in vitro responses to S. solidus antigens compared to populations with low S. solidus prevalence. This suggests a rather low degree of local adaptation in S. solidus populations, which might be due to high gene flow among populations because of their extremely mobile final hosts, fish-eating birds. Ó 2013 Elsevier Ltd. All rights reserved. Keywords: Parasite antigens In vitro leukocyte response Local adaptation Gasterosteus aculeatus Schistocephalus solidus * Corresponding author. Tel.: þ49 251 83 32550. E-mail address: joern.scharsack@uni-muenster.de (J.P. Scharsack). 1050-4648/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.fsi.2013.10.019 F. Franke et al. / Fish & Shellfish Immunology 36 (2014) 130e140 131 1. Introduction 1.2. Host-parasite local mutual adaption Helminths are frequent parasites of natural fish populations, but interactions of helminths with the piscine immune system are under-investigated. This might be attributed to the fact that helminth infections in aquaculture are often relatively easy to control (e.g. by control of invertebrate intermediate hosts) and rarely have prominent commercial impact [1]. However, in their natural habitat, parasites often drastically reduce host fitness and thus pose strong selection pressures on their hosts, which therefore have evolved powerful counter-measures to control infection [2]. The success of helminth parasites largely depends on their ability to evade and/or manipulate the generally efficient immune system of their fish hosts [3,4]. The evolutionary arms race of host-parasite counter adaptations (often described as Red Queen Dynamics) promote parasite virulence and infection success on the one hand [5e7], but host immunocompetence and prevention of infection on the other [8e11]. In cyprinids and salmonids, activation of granulocytes is considered to be an important part of the immune defence against parasitic helminths [12e16]. In sticklebacks, traits of cellular innate immunity, such as respiratory burst activity, were elevated in a population that was adapted to higher parasite infection pressure [9]. Adaptation of stickleback immunocompetence to local parasites is presumably supported by selection for certain MHC class II genotypes [11]. The basis of functional cellular immunity in such differential adaptive situations is not well investigated to date in teleost fish. A specialist like S. solidus might even have optimised its surface antigens (immune evasion) for a frequently infected local host population. If this is the case, immunity of sympatric hosts might have a weaker response to the parasites’ antigens compared to immunity of allopatric hosts. We thus hypothesize that antigenicity (strength of in vitro leukocyte response) decreases from generalist to specialist parasites and among specialists from parasites that do not infect G. aculeatus to G. aculeatus-infecting parasites, and among G. aculeatus-infecting parasites from allopatric to sympatric host-parasite combinations. Previous studies of local adaptation of teleost fish hosts and their parasites have mainly focussed at infectivity and host mortality, but have rarely included immunological patterns of adaptation [5e7,27e30]. Some of these studies observed local advantages of the (co-evolved) host population. In those studies, hosts were genetically best adapted to the local parasite population and showed inferior performance in preventing infections with nonlocal parasites of the same species [27,28]. Such situations would disadvantage immigrant hosts, but favour migrating parasites, thus promoting gene flow in the parasites. Other studies failed to detect local adaption in host-parasite systems [29] but a larger third group of studies describes a co-evolutionary local advantage of the parasite population, which became more infective for local compared to non-local hosts of the same species [5e7]. Such a constellation would promote immigrant hosts and disadvantage foreign parasites, which may promote gene flow among hosts. Therefore, parasites may play an important role in the dynamic process of diversification and speciation of their teleost fish hosts and vice versa [9]. 1.1. Parasites with a broad and a narrow host range e Diplostomum and Schistocephalus Helminths are experts in evasion and manipulation of their hosts’ immune functions and the respective strategies may depend on the host range. Generalist parasites might not be able to adapt their antigenicity (antigenic surface) for the immune system of a certain host species and might instead use other immune evasion strategies. An example is the trematode Diplostomum pseudospathaceum, which infects the immunological inert eye lens of various freshwater fish species. After penetrating the skin (or gills) of their fish host, the tissue migrating larval stage (diplostomulum) of D. pseudospathaceum finds its way along the blood vessels to the eye lenses. Antigens from such a generalist parasite might trigger stronger immune responses compared to antigens from specialist parasites, such as the cestode Schistocephalus solidus. The adult stage of the tapeworm S. solidus reproduces in the gut of warmblooded vertebrates, most often fish-eating birds. Eggs are released in the faeces of the final host. A first, free-swimming larval stage (coracidium) hatches in water, and develops to the second larval stage (procercoid) after ingestion by a cyclopoid copepod. The third larval stage (plerocercoid) develops in the body cavity of the obligatory and specific second intermediate host: the threespined stickleback (Gasterosteus aculeatus). Besides the immune system, the aggressive environment of the stickleback’s stomach may prevent infection [17,18], but once in the body cavity of the three-spined stickleback host, clearance of S. solidus plerocercoids is rare [19]. Experimental transfer of S. solidus plerocercoids to fish species other than three-spined sticklebacks lead to rapid death of the larvae [20,21], underlining the specific adaptation of S. solidus to three-spined sticklebacks, but suggesting that the immune system of fish is in principle able to clear S. solidus infections. In threespined sticklebacks, established plerocercoids of S. solidus take all the resources that the parasite needs from the host and grow to up to 20e30% (w/w) of their host’s body weight [19], thereby reducing the fitness of the hosts and resulting in decreased or even absent reproduction [22e26]. 1.3. The present study In this study, responses of three-spined stickleback head kidney leukocytes (HKL) to antigens of helminth fish parasites were investigated with an in vitro system, enabling large-scale comparisons between parasite species, as well as comparisons across different host populations. Since activation of granulocytes is important in the immune defence of fish against parasitic helminths [12e16] and in sticklebacks cellular innate immunity was elevated in a population with higher parasite infection pressure [9], we quantified the respiratory burst (RB) activity of HKL. We hypothesized that in vitro exposure of HKL to parasite antigens might influence leukocyte viability and the frequencies of cellular subsets and therefore analysed numbers of viable HKL and the granulocytes to lymphocytes ratio (G/L ratio) after in vitro stimulation. We investigated HKL in vitro responses to antigens from generalist parasites, such as the eye fluke D. pseudospathaceum that infects, among other fish species, also three-spined sticklebacks (G. aculeatus) [31], and the cestode Triaenophorus nodulosus, with several fish species including G. aculeatus as second intermediate hosts [32]. Furthermore, antigens from Ligula intestinalis were used, a parasite specific to cyprinids as second intermediate hosts, which does not infect G. aculeatus. Finally, tapeworm antigens from two highly specialised Schistocephalus species, Schistocephalus pungitii, specific for nine-spined sticklebacks (Pungitius pungitius) and S. solidus, specific for G. aculeatus, were used (Table 1). From the latter, seven hosts and corresponding parasite populations from across Europe were tested to investigate potential local adaptation in the G. aculeatuseS. solidus system. After in vitro stimulation, stickleback HKL were analysed by means of flow cytometry to determine the cell viability and the granulocyte to lymphocyte ratio. In addition, the respiratory burst 132 F. Franke et al. / Fish & Shellfish Immunology 36 (2014) 130e140 Table 1 Parasite antigen sources. Naturally- or laboratory-infected hosts originated from different populations: NO (lake ‘Skogseidvatnet’, Norway), SC3 (‘Loch Olabhat’, Scotland), GPS (lake ‘Grosser Plöner See’, Germany), NST (lagoon ‘Neustädter Binnenwasser’, Germany), IBB (brook ‘Ibbenbürener Aa’, Germany), SP (channel near ‘Xinzo de Limia’, Spain), LBT (lake ‘Lebrader Teich’, Germany), MGS (lake ‘Müggelsee’, Germany).* Total pool weight. Parasite species Host species Population Natural-/laboratory infection Pooled individuals Weight (mg) Used for experiment Parasitizes G. aculeatus S. solidus S. solidus S. solidus S. solidus S. solidus S. solidus S. solidus S. pungitii L. intestinalis T. nodulosus D. pseudospathaceum G. aculeatus G. aculeatus G. aculeatus G. aculeatus G. aculeatus G. aculeatus G. aculeatus P. pungitius R. rutilus P. fluviatilis L. stagnalis NO SC3 GPS NST IBB IBB SP LBT MGS MGS GPS Lab. inf. Lab. inf. Lab. inf. Lab. inf. Lab. inf. Nat. inf. Lab. inf. Nat. inf. Nat. inf. Nat. inf. Nat. inf. 6 10 7 10 19 11 12 7 6 41 e 85  30 168  32 987* 2267* 116  37 94  32 178  48 1333* 4200  1140 8.5  4.6 e 3 3 3 1, 3 3 2 3 1 1 1 1, 2 þ þ þ þ þ þ þ activity of in vitro stimulated HKL was analysed in a zymosaninduced chemiluminescence assay. In a first experimental set, in vitro responses of HKL from a single stickleback population to antigens of the five parasite species were tested (experiment 1: helminth species). In a second set of experiments, we compared in vitro responses of HKL derived from sticklebacks of seven different populations to S. solidus and D. pseudospathaceum antigens from a single origin each (experiment 2: host origins). With a third experimental set up, HKL from six stickleback populations were exposed in a fully crossed design to S. solidus antigens derived from the same habitats, to test if stickleback HKL responses differ between sympatric and allopatric host-parasite combinations (experiment 3: sympatric/allopatric combinations). 2. Materials and methods 2.1. Experimental sticklebacks We took advantage of the availability of seven three-spined stickleback (G. aculeatus) populations from across Europe, within the ‘stickleback cluster’ of the DFG priority programme 1399 ‘HostParasite Coevolution’. The majority of sticklebacks used for head kidney leukocyte (HKL) isolation were laboratory-raised, first generation offspring of wild caught individuals. Parental sticklebacks originated from seven European populations, a lake in the West of Norway (‘Skogseidvatnet’, NO), lakes on the Scottish island North Uist (‘Loch Sandary’, SC1 and ‘Loch Scadavay’, SC2), an inland lake and a brackish lagoon of the Baltic Sea in Northern Germany (‘Grosser Plöner See’, GPS and ‘Neustädter Binnenwasser’, NST), a brook in Western Germany (‘Ibbenbürener Aa’, IBB) and a drainage channel system in the Northwest of Spain near ‘Xinzo de Limia’ (SP). Investigated populations were sampled with support of local cooperators, which were holding necessary licences and helped to obtain sampling permits from the local authorities. Live sticklebacks were transported according to EU legislation for noncommercial and solely scientifically used material. Stickleback offspring were bred at the IEB Münster, Germany, except for the Scottish (SC1, SC2) populations (bread at the IEZ Bonn, Germany) and a German (GPS) population (bread at the MPI Plön, Germany). Wild caught sticklebacks of the Scottish (SC1, SC2), a German (IBB) and the Spanish (SP) population were used in experiment 2 (host origins) and second generation offspring of a German (NST) population in experiment 1 (parasite species) and 3 (sympatric/allopatric combinations). Upon arrival (SC1, SC2, GPS individuals), and at least two weeks before experimentation (NO, NST, IBB, SP individuals), sticklebacks were kept in 125 L glass tanks þ þ separated by populations (origins) at the IEB Münster. All tanks were connected to a water recirculation system which provided aerated and filtered water at 18e20  C. Sticklebacks were kept at 15/9 h light/dark cycles and fed daily ad libitum with frozen red mosquito larvae and dry food (TetraMin, Tetra). Sticklebacks were maintained and treated in accordance with the local animal welfare authorities and the EU Directive 2010/63/EU for animal experiments. S. solidus prevalence in the stickleback populations used in the present study were under constant surveillance for several years (at least 3 years), except for the Scottish populations (1 year). Prevalence ranged from <1% (NST, GPS; Kalbe, pers. comm.) over 3e5% (IBB; personal observation), 0e10% (SC1; Rahn, pers. comm.) 0e11% (SC3; Rahn, pers. comm.), 10% (SC2 [33]), 30e50% (NO; Kalbe, pers. comm.) to 40e66% (SP; personal observation [34],). 2.2. Experimental parasites For in vitro stimulation of stickleback head kidney leukocytes (HKL), antigen preparations of four cestode species and a trematode species were used (Table 1). S. solidus plerocercoids originated from the stickleback populations used for HKL isolation and were partly provided by the members of the ‘stickleback cluster’ (see 2.1 Experimental sticklebacks). Scottish S. solidus originated from sticklebacks of a third Scottish lake (‘Loch Olabhat’, SC3), about 7 km away from ‘Loch Sandary’ (SC1) and 10 km from ‘Loch Scadavay’ (SC2). For antigen preparation, S. solidus plerocercoids were grown in sympatric host-parasite combinations in laboratoryraised and -infected three-spined sticklebacks. Only in experiment 2 (host origins) S. solidus antigens were prepared from naturally infected sticklebacks from a German (IBB) population. Plerocercoids of a second Schistocephalus species (S. pungitii), were collected from wild caught nine-spined stickleback (P. pungitius), from lake ‘Lebrader Teich’ (LBT) about 6 km away from lake ‘Grosser Plöner See’ (GPS). Plerocercoids of L. intestinalis were collected from roach (Rutilus rutilus) caught in the lake ‘Müggelsee’ (MGS) close to Berlin, Germany. Plerocercoids of the pike tapeworm, T. nodulosus were collected from twenty European perches (Perca fluviatilis) that originated also from the lake ‘Müggelsee’ (MGS). Cercariae of D. pseudospathaceum were isolated from ten infected snails (Lymnaea stagnalis) from the lake ‘Grosser Plöner See’ (GPS) as described in Hibbeler et al. [35]. 2.3. Antigen preparations Parasite antigens were prepared from pools of at least six individuals (Table 1). Antigen preparations were kept on ice during preparation and phosphate-buffered saline (PBS, pH 7.4, F. Franke et al. / Fish & Shellfish Immunology 36 (2014) 130e140 Calbiochem 524650) was used at 4  C. Plerocercoids of S. solidus, S. pungitii, L. intestinalis and T. nodulosus were collected from fish body cavities under sterile conditions, and weighed. Plerocercoids were washed and frozen with PBS at 20  C (T. nodulosus: 0.25 g wet weight mL 1, all other species: 0.5 g wet weight mL 1). After thawing, teguments of the cestodes S. solidus, S. pungitii and L. intestinalis were detached from the worm body by vortexing. Tegument antigen preparations were decanted and worm bodies were washed with PBS to remove remaining tegument fragments. The remaining worm bodies (in case of L. intestinalis sections of worm bodies) were homogenised manually in a 1.5 mL tube with a pestle and adjusted with PBS to a concentration of 0.25 g original wet weight mL 1. Tegument and body antigen preparations were sonicated (Sonoplus 2070, Bandelin, Germany) for 120 s (duty cycle 10%, power 60%) on ice. Solid material was removed by centrifugation (600  g, 4  C, 10 s). Instead of centrifugation, antigen preparation from the tegument of S. solidus from naturally infected sticklebacks of a German (IBB) population used in experiment 2 (host origins) was 0.45 mm filtered. The protein content of each preparation was determined with a Bradford assay and adjusted to 400 mg L 1 (protein fraction) with Leibovitz 15 medium (PAA E15020) with 10% (v/v) distilled water and 10 mmol L 1 HEPES buffer (Lonza 17e737) and stored at 80  C. Antigens of T. nodulosus were prepared as whole body preparation as described above, without detaching the tegument beforehand. The antigen preparation from D. pseudospathaceum was prepared as described by Hibbeler et al. [35], diluted and stored as described above. Fish protein preparations were produced from muscle tissue and pooled from six individual laboratory-raised sticklebacks (G. aculeatus), originating from a German (NST) population and three individual roaches (R. rutilus) originating from a pond at the IEB Münster, Germany, following the protocol for parasite body antigens as described above. Each antigen/protein preparation was controlled microscopically for sterility after incubation of 25 mL subsamples in wells of a 96-well half-area flat bottom microtitre plate (Greiner Bio-One) at 20  C in a water vapour saturated atmosphere with 3% CO2 for 4 days. 2.4. Leukocyte isolation and in vitro stimulation Cells and media were kept refrigerated during the preparations. Basic medium was Leibovitz 15 (PAA E15-020) supplemented with 10 mmol L 1 HEPES buffer (Lonza 17e737) and 10% (v/v) distilled water to adjust osmotic pressure according to stickleback serum osmolarity (subsequently named L-90). Fish were anesthetized by a blow on the head, decapitated and the body cavities were opened. Head kidneys were removed under sterile conditions and transferred to 40 mm cell strainers (BD Falcon) in petri dishes (3.5 cm Ø) with 1 mL L-90 with heparin (2  104 IU L 1, Applichem A3004). Single cell suspensions of head kidney leukocytes (HKL) were prepared by forcing the tissues through the strainers with a plunger of a syringe. HKL were washed once with heparinised L-90 (600  g, 4  C, 5 min), once with L-90 without heparin, and resuspended in culture medium (L-90 with 1  105 IU L 1 penicillin, 100 mg L 1 streptomycin (PAA P11-010), 4 mmol L 1 L-Glutamin (PAA M11006), 5% (v/v) foetal bovine serum (PAA A11-103) and 1% (v/v) heat inactivated, pooled carp serum). Total cell numbers in head kidney isolates were determined by means of flow cytometry (see 2.5 Flow cytometric analysis) and cell suspensions were adjusted to 4  106 cells mL 1 with culture medium. For in vitro stimulation, HKL were seeded out in 96-well half-area flat bottom microtitre plates (Greiner Bio-One) at a density of 1  105 cells well 1 in a final volume of 100 mL culture medium well 1. From each individual fish, HKL were cultured with 133 medium alone as a negative control and cultured with lipopolysaccharides (LPS, 20 mg L 1, Sigma L7895) and pokeweed mitogen (PWM, 2 mg L 1, Sigma L8777) as positive controls. Protein preparations from fish muscle tissues and bovine serum albumin (BSA, Carl Roth CP77) were used to test potential effects of allogeneic and xenogeneic proteins (protein control, 10 mg L 1). Parasite antigens were added to final concentrations of 10 mg L 1 (protein fraction) each. All cultures were incubated for 4 days at 20  C in a water vapour saturated atmosphere with 3% CO2 and were controlled microscopically for sterility afterwards. 2.5. Flow cytometric analysis Subsamples of freshly isolated and cultured head kidney leukocytes (HKL) were analysed by means of flow cytometry (FACSCanto II, BD, USA). Total cell numbers (per sample/culture) were determined with the standard cell dilution assay (SCDA [36],) modified by Scharsack et al. [37]: after in vitro culture, HKL culture plates were placed on ice (30 min) to detach adherent cells. Resuspended HKL (5 mL cell suspension per sample of freshly isolated HKL, 25 mL from each culture well) were transferred to individual wells of a 96 well round bottom microtiter plate (BD Falcon). Samples were supplemented with propidium iodide (2 mg L 1, Sigma 81845) and green fluorescent reference particles (4.5 mm, Polysciences 16592-5, 1.5  104 particles well 1 for freshly isolated and 3  104 particles well 1 for cultured HKL) and measured with the automated sampling unit of the flow cytometer. Forward- and side scatter (FSC/SSC) characteristics of up to 1  104 events in the single cell gate were acquired in linear mode. Fluorescence intensities at 530 nm and 585 nm were measured using log-scale. Flow-cytometric data were analysed with the FacsDiva software (version 6.1.2, BD, USA). Dead cells (propidium iodide positive) and cellular debris (low FSC/SSC characteristics) were excluded from further evaluation. Lymphocyte- and granulocyte populations were identified according to their characteristic FSC/SSC profiles [37]. Cell viability (absolute numbers of viable cells) before and after in vitro culturing were calculated according to N (viable cells) ¼ events (viable cells)  number (standard beads)/events (standard beads). Granulocyte to lymphocyte ratio (G/L ratio) was calculated according to G/L ratio ¼ number (viable granulocytes in culture)/number (viable lymphocytes in culture). 2.6. Production of reactive oxygen species The respiratory burst (RB) activity of head kidney leukocytes (HKL) after in vitro cultivation was quantified in a lucigeninenhanced chemiluminescence (CL) assay, modified after Scott & Klesius [38] as described by Kurtz et al. [39]. After removing a subsample for the flow cytometric analysis (25 mL well 1, see 2.5 Flow cytometric analysis), 65 mL of cell suspension was transferred from each well of an HKL culture plate to a well of a white CL 96-well plate (Nunc), prefilled with 20 mL of 2.5 g L 1 lucigenin (N,N0 -dimethyl-9,90 -biacridiniumdinitrate, Sigma M8010) in PBS and 95 mL medium (RPMI-1640, PAA E15-039) supplemented with 10 mmol L 1 HEPES buffer (Lonza 17e737) and 10% (v/v) distilled water. CL plates were incubated for 30 min at 20  C in a water vapour saturated atmosphere with 3% CO2 to enable lucigenin uptake by the cells. The RB was induced by addition of 20 mL zymosan suspension from Saccharomyces cerevisiae (7.5 g L 1, Sigma Z4250) in PBS to every well. Relative luminescence units (RLU) of individual wells were measured for 3 s, in 5 min intervals during 3 h incubation at 20  C in an Infinite 200 multimode reader (Tecan, Switzerland). For data analyses the area under the kinetic curve (RLU area, integral from t0 to t3h of kinetic RLU curve) was determined using Magellan 6.5 software (Tecan, Switzerland). 134 F. Franke et al. / Fish & Shellfish Immunology 36 (2014) 130e140 2.7. Statistics SPSS Statistics software (version 20, IBM, USA) was used for statistical analyses. Normal distributions of data were verified by visual inspection of residual histograms. Since head kidney leukocytes (HKL) from a single stickleback population were used in experiment 1 (helminth species), those data were analysed using a Greenhouse-Geisser corrected repeated measurement analyses of 14 respiratory burst (RLU × 105) A n = 35 h 12 10 c c c 8 6 b a f d a e d e d e b g f g e d a 4 3. Results 2 3.1. Controls B 4 HKL (number of viable cells × 10 ) 0 6 n = 34 5 4 g b a 3 c a e d a e d c f e e e b a e b a e b c f 2 1 0 5 C n = 34 e 4 G/L ratio variance (RM ANOVAs). Greenhouse-Geisser corrected two-way RM ANOVAs with antigen treatment as the within-subject factor and the stickleback population as the between-subject factor were used to analyse experimental data with HKL from more than one stickleback population. If significant effects were detected with the ANOVAs, pairwise comparisons with sequential Bonferroni corrected paired-sample t-tests [40] were used as post-hoc tests. Data of sympatric G. aculeatus-S. solidus combinations were compared to the mean of the data from the allopatric combinations in a pairedsample t-test and compared separately for stimulation with tegument and body antigens. Correlations of in vitro leukocyte responses and geographical distances as well as different S. solidus prevalences across populations were analysed by sequential Bonferroni corrected Mantel-tests [41]. Mantel-tests were computed using the ade4 statistical package in R (version 3.0.2, R Core Team 2013, Austria) with means of allopatric stimulated host populations and allopatric combined parasite populations. Results of the statistical analysis are summarized in Table 2 of the Supplementary material. 3 b a b a b a d c b a a c a a b a b a d d 2 1 C t BS r A LP G .a P S cu W M S. R lea so . r tus l u S. idu tilu S. so s bo s pu lidu dy n s L. S. git te in pu ii b g te n od L. stin git y in al ii te te is g D .p s b se T tina od ud . n lis y os od te pa ulo g th su ac s eu m 0 Fig. 1. Experiment 1 (helminth species): In vitro responses of stickleback leukocytes to antigens from five helminth parasite species. Respiratory burst activity (A), cell viability (B), and granulocyte to lymphocyte ratio (C) was analysed after 4 days of incubation of head kidney leukocytes (HKL) from laboratory-raised second-generation sticklebacks from a German (NST) population. HKL from individual fish were cultured in medium alone (Ctr), with BSA (10 mg L 1), LPS (20 mg L 1), PWM (2 mg L 1), fish muscle proteins (10 mg L 1) or parasite antigens (10 mg L 1). Parasite antigens were prepared from the tegument (teg), the body without tegument (body), or the whole bodies (T. nodulosus, D. pseudospathaceum). Different letters above error bars (mean þ standard error) indicate significant differences (p < 0.05). Medium controls (leukocytes in medium alone) and positive controls (LPS, PWM) were included in each experiment. Head kidney leukocytes (HKL) from positive controls showed increased respiratory burst (RB) activity (Figs. 1A, 2A and 3A, Fig. 5A Supplementary material) compared to medium controls, whereas the cell viability was only higher with LPS (Figs. 1B, 2B and 3B, Fig. 5B Supplementary material). The granulocyte to lymphocyte ratio (G/L ratio) was reduced by the addition of LPS (Figs. 1C, 2C and 3C, Fig. 5C Supplementary material). In experiment 1 (helminth species) and experiment 2 (host origins) a protein control (BSA) was included. The RB activity, the cell viability, and the G/L ratio did not differ between HKL from protein and medium controls (Figs. 1 and 2). Therefore the protein control was not used in experiment 3 (sympatric/allopatric combinations). Proteins from G. aculeatus and R. rutilus served as additional protein controls in experiment 1 (helminth species). Xenogeneic proteins from R. rutilus merely decreased the cell viability (Fig. 1B), but did not change the respiratory burst activity (Fig. 1A) and G/L ratio (Fig. 1C) compared to medium controls. Allogeneic stickleback proteins increased the RB activity of HKL to the level of LPS stimulation (Fig. 1A) and the cell viability was elevated compared to medium controls (Fig. 1B). The G/L ratio was more prominently influenced by allogeneic proteins as by the positive controls or parasite antigen stimulations (Fig 1C). The strong in vitro responses to allogeneic proteins might be explained by a transplant rejection response mediated by major histocompatibility complex (MHC) cell surface antigens of the same species, whereas xenogeneic MHC antigens were not recognised. Similar observations were made with mixed cultures of rat lymphocytes, which were fully reactive to MHC alloantigens, but displayed no detectable primary reactivity to surface antigens from xenogeneic mammalian cells [42]. 3.2. Experiment 1 (helminth species): in vitro responses of stickleback head kidney leukocytes to antigens from five helminth parasite species To investigate reactions of three-spined stickleback (G. aculeatus) head kidney leukocytes (HKL) to different parasite species, HKL of sticklebacks from a German (NST) population were stimulated in vitro for 4 days with antigens from four cestode and one trematode species. 135 F. Franke et al. / Fish & Shellfish Immunology 36 (2014) 130e140 14 5 respiratory burst (RLU × 10 ) A 12 10 8 6 4 2 B 4 HKL (number of viable cells × 10 ) 0 6 5 4 3 2 1 0 3 C G/L ratio 2 1 0 NO SC2 wild SC1 wild GPS n = 22 n = 20 n = 30 n = 32 NST n = 28 IBB n = 27 IBB wild SP wild n = 18 n=8 fish population Ctr BSA LPS PWM SSb SSt Diplostomum Fig. 2. Experiment 2 (host origins): In vitro responses of head kidney leukocytes from seven stickleback populations to parasite antigens. Respiratory burst activity (A), cell viability (B), and granulocyte to lymphocyte ratio (C) was analysed after 4 days of incubation of head kidney leukocytes (HKL) from laboratory-raised and wild caught (wild) sticklebacks from Norway (NO), Scotland (SC1, SC2), Germany (GPS, NST, IBB) and Spain (SP). HKL were cultured in medium alone (Ctr), with BSA (10 mg L 1), LPS (20 mg L 1), PWM (2 mg L 1), or parasite antigens (10 mg L 1). S. solidus from a German (IBB) population were used to prepare antigens from the tegument (SSt, 0.45 mm filtered) and the body without tegument (SSb). D. pseudospathaceum antigens were prepared from full bodies (Diplostomum). Means þ standard errors are given and asterisks indicate significant differences (p < 0.05) to corresponding medium controls. Incubation of HKL with antigens from the specialist, G. aculeatus-infecting cestode S. solidus significantly elevated the respiratory burst (RB) activity of HKL. Comparable RB activities were observed with antigens of the close relative of S. solidus, S. pungitii, which specifically infects nine-spined sticklebacks (P. pungitius) (Fig. 1A). In contrast, the cell viabilities were decreased by S. solidus but increased by S. pungitii antigens compared to medium controls (Fig. 1B). The granulocyte to lymphocyte (G/L) ratios were not affected by S. solidus and S. pungitii antigens (Fig. 1C). Antigens from the cestode L. intestinalis, which does not infect G. aculeatus, stimulated the RB activity of HKL to a higher extent than antigens from the Schistocephalus species (Fig. 1A). HKL stimulated with L. intestinalis antigens did not differ significantly in 136 F. Franke et al. / Fish & Shellfish Immunology 36 (2014) 130e140 A Respiratory burst fish population SP IBB NST GPS SC1 NO B Total viable cells fish population SP IBB NST GPS SC1 NO C G/L ratio fish population SP IBB NST GPS SC1 body antigens > control = control SP IBB NST GPS SC3 NO SP IBB NST GPS NO mitogens SC3 LPS PWM NO tegument antigens < control Fig. 3. Experiment 3 (sympatric/allopatric combinations): In vitro responses of leukocytes from six stickleback populations to S. solidus antigens from identical populations in a fully crossed design. Respiratory burst activity (A), cell viability (B), and granulocyte to lymphocyte ratio (C) after 4 days of incubation of head kidney leukocytes (HKL) from laboratory-raised sticklebacks from Norway (NO, n ¼ 19), Scotland (SC1, n ¼ 24), Germany (GPS, n ¼ 31e32; NST, n ¼ 28; IBB, n ¼ 29) and Spain (SP, n ¼ 23). HKL were cultured in medium alone (Ctr), with LPS (20 mg L 1), PWM (2 mg L 1), or S. solidus tegument or body antigens (10 mg L 1) from the corresponding populations. Means of stimulated cultures were normalised to corresponding medium controls (see also Fig. 5, Supplementary material). their G/L ratios from medium controls, but G/L ratios were significantly higher with tegument antigens compared to body antigens (Fig. 1C). Stimulation with antigens of the pike-tapeworm T. nodulosus, which besides other fish species also infects the three-spined stickleback as a second intermediate host [43], led to elevated RB activity of HKL comparable to the Schistocephalus stimulations (Fig. 1A). Antigens of T. nodulosus strongly decreased the cell viability (Fig. 1B) and the G/L ratio (Fig. 1C). Antigens from the eye fluke D. pseudospathaceum, which also infects G. aculeatus, induced the strongest RB activity of HKL across all parasite species (Fig. 1A) and reduced the G/L ratio comparable to stimulation with LPS or T. nodulosus antigens (Fig. 1C). 3.3. Experiment 2 (host origins): in vitro responses of head kidney leukocytes from seven stickleback populations to parasite antigens To compare the in vitro responses of three-spined stickleback (G. aculeatus) head kidney leukocytes (HKL) from different populations, HKL were cultured with S. solidus and D. pseudospathaceum antigens. Parasite antigens from the body of S. solidus (specialist, infective for G. aculeatus) and those from D. pseudospathaceum (generalist, infective for G. aculeatus) significantly increased the respiratory burst (RB) activities of HKL from the investigated populations, except for HKL from the Spanish (SP) population and HKL from the Scottish (SC1, SC2) populations that were stimulated with S. solidus body antigens (Fig. 2A). In contrast to experiment 1 (helminth species, Fig. 1A), S. solidus antigens from the tegument did not increase the RB activity of HKL, which might be explained by removal of particles by the 0.45 mm filtration of the antigen preparation used in the present experiment. The cell viability varied between stickleback populations and was lowest in one of the Scottish (SC2) populations. Incubation with S. solidus body antigens and those of D. pseudospathaceum, usually resulted in higher cell viabilities than in medium controls, except for HKL from one Scottish (SC2) population, a German (NST) population and the Spanish (SP) population after stimulation with D. pseudospathaceum antigens. With S. solidus antigens from the tegument, higher cell viability was observed with HKL of sticklebacks from a German (GPS) population only compared to medium controls (Fig. 2B). The parasite antigens that stimulated the RB activity and the viability of HKL (S. solidus body antigens and those of D. pseudospathaceum) did not affect the granulocyte to lymphocyte (G/L) ratios. Instead, stimulation with antigens from the tegument of S. solidus resulted in a significant decrease of G/L ratios except for HKL of the Spanish (SP) population (Fig. 2C). The in vitro responsiveness of stickleback HKL was relatively similar across populations overall, with the exception of the Spanish (SP) population, where only the cell viability deviated significantly from the corresponding control after stimulation with S. solidus body antigens (Fig. 2B). The Scottish (SC2) population was notable because the viability of cells (Fig. 2B) and correspondingly the respiratory burst activity after the culture (Fig. 2A), was lower compared to the other populations. The G/L ratios varied across populations to a higher extent than across treatments within populations. A similar pattern of variation across populations was already present in the fresh HKL isolates before in vitro treatments (e.g. NO, NST e high; SC1, SC2, GPS e low; data not shown), thus was not an effect of the cell culture. 3.4. Experiment 3 (sympatric/allopatric combinations): in vitro responses of head kidney leukocytes from different stickleback populations to sympatric and allopatric Schistocephalus solidus antigens Three-spined stickleback (G. aculeatus) head kidney leukocytes (HKL) and S. solidus antigens from six populations (origins) across Europe with varying S. solidus prevalence (see 2.1 Experimental sticklebacks) were cultured in sympatric and allopatric in vitro combinations in a fully crossed experimental design. The stimulation with S. solidus antigens significantly increased respiratory burst (RB) activity in the majority of treatments. Among parasite populations, the highest RB activity was induced by antigens from the body of the Norwegian (NO), German (IBB) and Spanish (SP) S. solidus. This was consistent with all six stickleback origins tested here (Figs. 3A and Fig. 5A Supplementary material). Mantel-tests revealed that across allopatric combinations of both hosts and parasites, the RB activity was positively correlated with the parasite prevalence matrix, but not with the geographical distance matrix. Similarly to RB activity, HKL stimulation with S. solidus antigens from the body of the Norwegian (NO), a German (IBB) and the F. Franke et al. / Fish & Shellfish Immunology 36 (2014) 130e140 Spanish (SP) populations resulted in the highest viability of HKL. Highest HKL viability within the tegument antigen treatment was observed for the Norwegian (NO) and the Scottish (SC3) S. solidus populations (Figs. 3B and Fig. 5B Supplementary material). Granulocyte to lymphocyte (G/L) ratios were lowest in cultures with HKL from the Spanish (SP) population (Fig. 5C Supplementary material). Stimulation with S. solidus antigens altered the G/L ratio to a minor extent, but overall every stickleback populations’ stimulation with tegument antigens of Spanish (SP) parasites significantly reduced the G/L ratio (Figs. 3C and Fig. 5C Supplementary material). Significant differences between sympatric and allopatric hostparasite combinations were only detected in the cell viability of HKL cultures with S. solidus tegument antigens. In comparison to sympatric combinations, the cell viability was increased in HKL cultures that were stimulated with allopatric S. solidus antigens (Fig. 4). 4. Discussion 4.1. Parasite species induce different immune reactions in stickleback HKL cultures 6 12 n = 154 10 8 6 n.s. n.s. 4 2 0 5 3 B n = 155 4 2 n.s. p= 0.014 3 n.s. n.s. 1 2 1 0 body teg C n = 155 G/L ratio A 4 14 HKL (number of viable cells × 10 ) 5 respiratory burst (RLU × 10 ) In the present study, we investigated in vitro responses of head kidney leukocytes (HKL) from three-spined sticklebacks (G. aculeatus) to antigens from helminth parasite species, which partly infect G. aculeatus naturally. Among the parasite species tested here, antigens from D. pseudospathaceum, which has a relatively broad range of fish host species, induced the strongest in vitro responses of stickleback HKL, followed by antigens from L. intestinalis (specific for cyprinids) and the two Schistocephalus species; S. pungitii, specific for the nine-spined stickleback (P. pungitius) and S. solidus, specific for G. aculeatus. Cestodes often suppress (manipulate) the immune reactions of their hosts, which minimize harmful effects on themselves but may also reduce pathological effects on the hosts [44]. L. intestinalis and the Schistocephalus species gain most of their final weight in the body cavity of their second intermediate fish hosts, and are confronted with the host’s immunity for months, or even years. In contrast, larvae of digenean Diplostomum are confronted with the 0 body teg sympatric body teg allopatric Fig. 4. Experiment 3 (sympatric/allopatric combinations): Comparison of in vitro leukocyte responses to antigens from sympatric and allopatric S. solidus parasites. Respiratory burst activity (A), cell viability (B), and granulocyte to lymphocyte ratio (C) of the in vitro test with head kidney leukocytes (HKL) from six stickleback and S. solidus antigens from the identical populations (Fig. 3) were compared for differences between sympatric and allopatric host-parasite combinations (mean þ standard error). Antigen preparations from the tegument (teg) and the body without tegument (body) were treated separately (n. s. e not significant). 137 immune system of their fish host only for a short term (<24 h), during the migration from the skin to the immunological inert eye lenses [45,46]. In the present study, the in vitro respiratory burst (RB) activity of HKL was highest upon stimulation with D. pseudospathaceum antigens and consecutively lower with L. intestinalis, S. pungitii and S. solidus antigens. This suggests that maintenance of low immunogenicity requires specific adaptation to the host’s immune system, which is achievable for specialist parasites, such as S. solidus, but not for generalists, such as D. pseudospathaceum. However, antigens from digenean parasites, such as D. pseudospathaceum, might generally be less adapted to hide from a host’s immune response and therefore might induce higher in vitro responses. Although the tapeworm T. nodulosus is highly specific to pike as final hosts, it parasitizes approximately 72 fish species [43] and also the three-spined stickleback as a second intermediate host [32]. In the present study, stimulation of stickleback HKL with whole body antigens of T. nodulosus resulted in the lowest overall cell viability compared to other parasite antigens tested here. This was predominantly due to low granulocyte viability (data not shown). Nevertheless, T. nodulosus antigens induced an RB activity per culture, which was comparable to Schistocephalus antigens, presumably a consequence of higher activity of individual granulocytes. T. nodulosus plerocercoids are encapsulated in cysts in the liver tissue of their intermediate fish hosts [43,47,48]. In burbots (Lota lota) and perches (P. fluviatilis), encapsulation of T. nodulosus plerocercoids was followed by parasite degeneration, whereas Arctic charrs (Salvelinus alpinus) also encapsulated, but failed to degenerate T. nodulosus larvae [43]. Degeneration of encapsulated T. nodulosus is presumably facilitated by granulocytes. In the present in vitro study, granulocytes from sticklebacks were activated upon exposure to T. nodulosus antigens, but decreased prominently in viability. The strong activation of granulocytes by T. nodulosus antigens, might have exhausted granulocyte viability during the culture, but in vivo, T. nodulosus might reduce the viability of granulocytes immigrating the cysts to avoid damage by their RB activity. The Schistocephalus species S. solidus and S. pungitii, although closely related and able to hybridize in the lab [49] are specific for either the three-spined stickleback or the nine-spined stickleback (P. pungitius). Infection of nine-spined sticklebacks with S. solidus infected copepods failed [21], and transplantation of S. solidus plerocercoids to P. pungitius stopped parasite growth and ultimately leads to death while homo-transplants of both S. solidus and S. pungitii plerocercoids survived [20,21]. Despite the high in vivo specificity of Schistocephalus parasites for their stickleback hosts, in vitro responses of HKL from three-spined sticklebacks to antigens of the two species were relatively similar. Only cell viability was elevated in cultures with S. pungitii antigens, while it remained at the level of protein controls in cultures with S. solidus antigens. Lower response levels of HKL to S. solidus antigens might be indicative of the higher degree of adaption of S. solidus to the immune system of the three-spined stickleback. Absence of differential RB activities of HKL to antigens from the two Schistocephalus species might depend on their close relatedness and similarity in immune evasion strategies. Taken together, antigens from the two generalist parasites, D. pseudospathaceum and T. nodulosus, and the cestode L. intestinalis, which is specialised for cyprinids, induced relatively strong in vitro responses. This is presumably a cost of the ability to infect a high variety of hosts, respectively, the lack of adaptation to the stickleback’s immune system by L. intestinalis. Vice versa, antigens from stickleback specific plerocercoids (Schistocephalus) excited lower in vitro activation, which is likely a sign of specific adaptation to the host’s immune repertoire. 138 F. Franke et al. / Fish & Shellfish Immunology 36 (2014) 130e140 4.2. Comparison of parasite tegument and body antigens Generally lower in vitro HKL responses were observed with Schistocephalus tegument antigens when comparing antigen preparations from the tegument and the body. By contrast, antigens from the tegument of L. intestinalis induced higher in vitro responses of stickleback HKL, compared to body antigens, indicating that surface antigens in particular are decisive for the strength of the host’s immune response. These findings support the assumption that S. solidus tegument antigenicity is best adapted to its specific host, since it induced only low immune activity of G. aculeatus HKL. 4.3. Influence of host populations In experiment 2 (host origins), we investigated variations of the in vitro responses of HKL from different stickleback populations. It is known that assortative mating limits gene flow, even between closely neighbouring three-spined stickleback populations [50e55] resulting in a number of different stickleback ecotypes [10,50,56,57] with varying parasite susceptibility [9,10]. Consequently we expected to detect differences across populations in responses of HKL to in vitro stimulation with parasite antigens. In the present study, HKL from a Scottish (SC2) population responded with lower respiratory burst (RB) activity and cell viability compared to HKL from the other stickleback populations. The wild caught Scottish (SC2) sticklebacks used in experiment 2 (host origins), might have been close to their natural age limit (Rahn, pers. comm.) and low in vitro responsiveness might be attributed to immuno-senescence [58]. Generally RB activity and cell viability of HKL from the different stickleback populations responded in a similar pattern to the in vitro stimulation. This was also the case in a comparison of wild caught and laboratory-raised individuals of a German (IBB) population. The granulocyte to lymphocyte (G/L) ratio exhibited prominent cross population differences and was elevated in the Norwegian (NO) and a German (NST) population and lowest in another German (GPS) population. However, a similar pattern of differences in the G/L ratios across populations was observed in fresh HKL isolates before the in vitro culture (data not shown) and might be a sign of adaptations of immunocompetence to varying habitat conditions rather than a consequence of in vitro stimulation. 4.4. Sympatric and allopatric host-parasite combinations In experiment 3 (sympatric/allopatric combinations), HKL from six stickleback populations across Europe were exposed to S. solidus antigens derived from the same six habitats in a fully crossed experimental design, to test if stickleback HKL respond differentially to antigens from sympatric and allopatric parasites. The sticklebacks used for this experiment were of similar ages as opposed to those used in experiment 2 (host origins). This might explain why differences in HKL responses between stickleback populations were less abundant than in the prior experiment. We hypothesized that in vitro responses of HKL would be higher to antigens from allopatric parasites compared to sympatric parasites, which was significant in the cell viability after stimulation with S. solidus tegument antigens. This might be attributed to the fact that parasites are under selection pressure to adapt their ‘antigenic’ surface for their respective host population, while body antigens are ‘hidden’ from host immunity anyways. Differential coevolution across host-parasite population pairs can only occur if gene flow between host and parasite populations is prevented. Genetically separated populations of G. aculeatus are well documented, and investigations on the relationships of S. solidus populations indicate that genetic divergence between populations from Alaska, Oregon and Wales is lower than that of their threespined stickleback hosts [59]. The lack of strong differences between HKL responses to sympatric and allopatric S. solidus antigens might be explained by low or absent divergent co-evolution of different host-parasite population pairs. This might be caused by higher dispersal rates of S. solidus genotypes with its mobile final hosts, fish eating birds, compared to relatively low dispersal rates of the stickleback intermediate hosts. 4.5. The effect of parasite prevalence S. solidus infects G. aculeatus populations with varying prevalence from <1% (Kalbe, pers. comm.) to up to 79% [60]. Parasites like S. solidus and L. intestinalis manipulate the behaviour of their intermediate fish hosts [61,62]. This results in higher host predation rates [63e65], which consequently diminishes host population sizes [60,66]. If the predator is the parasite’s definite host, it will thereby accelerate its life cycle completion rate, which may increase parasite population size and parasite-to-host biomass ratio [60]. Such dynamics may fluctuate, but often stabilise in a certain range, depending on the habitat specific host-parasite interactions and successful parasite transmissions. In vitro, HKL from stickleback populations with the highest S. solidus prevalence (Norway, NO; Spain, SP) increased their respiratory burst (RB) activity and cell viability more prominently after stimulation with S. solidus antigens compared to HKL from two German (GPS, NST) populations with the lowest prevalence. This suggests that HKL of sticklebacks are genetically predisposed to react more strongly to S. solidus in populations with high prevalence of the parasite. Conversely, S. solidus antigens from populations with high S. solidus prevalence (NO, SP) induced higher in vitro responses of HKL, compared to parasite antigens from populations with low S. solidus prevalence (GPS, NST). Thus, parasite virulence, or at least antigenicity, seems to increase in populations with high parasite prevalence. Taken together, both, responsiveness of stickleback leukocytes to S. solidus antigens, as well as immunogenicity of parasite antigens (virulence) increase with the parasite prevalence in populations. 5. Conclusions The present study demonstrates that immunogenicity of helminth antigens is highest in generalist parasite species that are infective for different fish species. Lowest immunogenicity was observed with antigens from the highly specialised parasite S. solidus with leukocytes from its specific second intermediate host. Investigations of six stickleback and corresponding S. solidus origins in sympatric and allopatric combinations revealed that high prevalence of the parasites rather than sympatric interaction primes leukocyte responsiveness and immunogenicity of the parasite antigens. Nevertheless, differences in the in vitro responses between populations with high and low S. solidus prevalence indicate that local adaptation does occur, but specific sympatric adaptations might be overruled by quantitative effects (high/low S. solidus prevalence). Selection pressure by high frequency (likelihood) of infections might be combined with the necessity to interact with a relatively high number of parasite genotypes, which counterselects specialisation for certain parasite genotypes, as would be expected in specific sympatric adaptations. Acknowledgements We are grateful to K. Knopf (IGB Berlin) for making parasites from the lake ‘Müggelsee’ available for the present study. We thank F. Franke et al. / Fish & Shellfish Immunology 36 (2014) 130e140 B. Hasert, W. Niermann and G. Plenge for technical assistance and support in the maintenance of the experimental sticklebacks. We also like to thank J. Howard (Institute for Genetics, University of Cologne) for fruitful discussion of the results on the response to allogeneic fish proteins. We are grateful to the anonymous reviewers for their relevant suggestions. The project was enabled by the stickleback cluster of the DFG priority programme 1399 “Host Parasite Coevolution” and funded by DFG grants to T. B. H Reusch (RE 1108/13-1), T. C. M. Bakker (BA 2885/5-1), M. Kalbe and J. P. Scharsack (SCHA 1257/2-1). [23] [24] [25] [26] [27] Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.fsi.2013.10.019. [28] [29] References [30] [1] Lafferty KD, Allesina S, Arim M, Briggs CJ, De Leo G, Dobson AP, et al. Parasites in food webs: the ultimate missing links. Ecol Lett 2008;11:533e46. [2] Summers K, McKeon S, Sellars J, Keusenkothen M, Morris J, Gloeckner D, et al. Parasitic exploitation as an engine of diversity. Biol Rev 2003;78:639e75. [3] Secombes CJ, Chappell LH. Fish immune responses to experimental and natural infection with helminth parasites. Annu Rev Fish Dis 1996;6:167e77. [4] Sitja-Bobadilla A. Living off a fish: a trade-off between parasites and the immune system. Fish Shellfish Immunol 2008;25:358e72. [5] Ballabeni P, Ward PI. Local adaptation of the tremadote Diplostomum phoxini to the European minnow Phoxinus phoxinus, its 2nd intermediate host. Funct Ecol 1993;7:84e90. [6] Voutilainen A, Valdez H, Karvonen A, Kortet R, Kuukka H, Peuhkuri N, et al. Infectivity of trematode eye flukes in farmed salmonid fish e effects of parasite and host origins. Aquaculture 2009;293:108e12. [7] Raeymaekers JAM, Wegner KM, Huyse T, Volckaert FAM. Infection dynamics of the monogenean parasite Gyrodactylus gasterostei on sympatric and allopatric populations of the three-spined stickleback Gasterosteus aculeatus. Folia Parasitol 2011;58:27e34. [8] Kalbe M, Wegner KM, Reusch TBH. Dispersion patterns of parasites in 0þ year three-spined sticklebacks: a cross population comparison. J Fish Biol 2002;60: 1529e42. [9] Kalbe M, Kurtz J. Local differences in immunocompetence reflect resistance of sticklebacks against the eye fluke Diplostomum pseudospathaceum. Parasitology 2006;132:105e16. [10] Scharsack JP, Kalbe M, Harrod C, Rauch G. Habitat-specific adaptation of immune responses of stickleback (Gasterosteus aculeatus) lake and river ecotypes. Proc R Soc B 2007;274:1523e32. [11] Eizaguirre C, Lenz TL, Kalbe M, Milinski M. Divergent selection on locally adapted major histocompatibility complex immune genes experimentally proven in the field. Ecol Lett 2012;15:723e31. [12] Hoole D, Arme C. Ultrastructural studies on the cellular response of fish hosts following experimental infection with the plerocercoid of Ligula intestinalis (cestoda, pseudophyllidea). Parasitology 1983;87:139e49. [13] Sharp GJE, Pike AW, Secombes CJ. Sequential development of the immuneresponse in rainbow trout [Oncorhynchus mykiss (Walbaum, 1792)] to experimental plerocercoid infections of Diphyllobothrium dendriticum (Nitzsch, 1824). Parasitology 1992;104:169e78. [14] Taylor MJ, Hoole D. Ligula intestinalis (L.) (Cestoda: pseudophyllidea): polarization of cyprinid leucocytes as an indicator of host- and parasite-derived chemoattractants. Parasitology 1993;107(Pt 4):433e40. [15] Taylor MJ, Hoole D. The chemiluminescence of cyprinid leucocytes in response to zymosan and extracts of Ligula intestinalis (Cestoda). Fish Shellfish Immunol 1995;5:191e8. [16] Nie P, Hoole D. Effects of Bothriocephalus acheilognathi on the polarization response of pronephric leucocytes of carp, Cyprinus carpio. J Helminthol 2000;74:253e7. [17] Hammerschmidt K, Kurtz J. Schistocephalus solidus: establishment of tapeworms in sticklebacks e fast food or fast lane? Exp Parasitol 2007;116:142e9. [18] Barber I, Scharsack JP. The three-spined stickleback-Schistocephalus solidus system: an experimental model for investigating host-parasite interactions in fish. Parasitology 2010;137:411e24. [19] Scharsack JP, Koch K, Hammerschmidt K. Who is in control of the stickleback immune system: interactions between Schistocephalus solidus and its specific vertebrate host. Proc R Soc B 2007;274:3151e8. [20] Bråten T. Host specificity in Schistocephalus solidus. Parasitology 1966;56: 657e64. [21] Orr TSC, Hopkins CA, Charles GH. Host specificity and rejection of Schistocephalus solidus. Parasitology 1969;59:683e90. [22] Arme C, Owen RW. Infections of the three-spined stickleback, Gasterosteus aculeatus L., with the plerocercoid larvae of Schistocephalus solidus (Müller, [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] 139 1776), with special reference to pathological effects. Parasitology 1967;57: 301e14. Pennycui L. Seasonal variations in parasite infections in a population of 3spined sticklebacks, Gasterosteus aculeatus L. Parasitology 1971;63:373e88. Tierney JF, Huntingford FA, Crompton DWT. Body condition and reproductive status in sticklebacks exposed to a single wave of Schistocephalus solidus infection. J Fish Biol 1996;49:483e93. Bagamian KH, Heins DC, Baker JA. Body condition and reproductive capacity of three-spined stickleback infected with the cestode Schistocephalus solidus. J Fish Biol 2004;64:1568e76. Heins DC. Fecundity compensation in the three-spined stickleback Gasterosteus aculeatus infected by the diphyllobothriidean cestode Schistocephalus solidus. Biol J Linnean Soc 2012;106:807e19. Ballabeni P. Experimental differences in mortality patterns between European minnows, Phoxinus phoxinus, infected with sympatric or allopatric trematodes, Diplostomum phoxini. J Fish Biol 1994;45:257e67. Bergeron M, Marcogliese DJ, Magnan P. The parasite fauna of brook trout, Salvelinus fontinalis (Mitchill), in relation to lake morphometrics and the introduction of creek chub, Semotilus atromaculatus (Mitchill). Ecoscience 1997;4:427e36. Sasal P, Durand P, Faliex E, Morand S. Experimental approach to the importance of parasitism in biological conservation. Mar Ecol-prog Ser 2000;198: 293e302. Francova K, Ondrackova M. Host-parasite interactions in sympatric and allopatric populations of European bitterling. Parasitol Res 2011;109:801e8. Reusch TBH, Rauch G, Kalbe M. Polymorphic microsatellite loci for the trematode Diplostomum pseudospathaceum. Mol Ecol Notes 2004;4:577e9. Poulin R, Valtonen ET. Interspecific associations among larval helminths in fish. Int J Parasit 2001;31:1589e96. De Roij J, MacColl ADC. Consistent differences in macroparasite community composition among populations of three-spined sticklebacks, Gasterosteus aculeatus L. Parasitology 2012;139:1478e91. Prieto MH, López JCF, González RA, Salán ES. Geographical and temporal FA variation in threespine stickleback populations from Galicia (NW Spain). Evol Ecol 2005;19:517e32. Hibbeler S, Scharsack JP, Becker S. Housekeeping genes for quantitative expression studies in the three-spined stickleback Gasterosteus aculeatus. BMC Mol Biol 2008;9. Pechhold K, Pohl T, Kabelitz D. Rapid quantification of lymphocyte subsets in heterogeneous cell populations by flow cytometry. Cytometry 1994;16: 152e9. Scharsack JP, Kalbe M, Derner R, Kurtz J, Milinski M. Modulation of granulocyte responses in three-spined sticklebacks Gasterosteus aculeatus infected with the tapeworm Schistocephalus solidus. Dis Aquat Org 2004;59:141e50. Scott AL, Klesius PH. Chemiluminescence: a novel analysis of phagocytosis in fish. Dev Biol Standard 1981;49:243e54. Kurtz J, Kalbe M, Aeschlimann PB, Haberli MA, Wegner KM, Reusch TBH, et al. Major histocompatibility complex diversity influences parasite resistance and innate immunity in sticklebacks. Proc R Soc B 2004;271:197e204. Holm S. A simple sequentially rejective multiple test procedure. Scand J Stat 1979;6:65e70. Mantel N. The detection of disease clustering and a generalized regression approach. Cancer Res 1967;27:209e20. Wilson DB, Fox DH. Quantitative studies on mixed lymphocyte interaction in rats: reactivity of lymphocytes from conventional and germfree rats to allogeneic and xenogeneic cell surface antigens. J Exp Med 1971;134:857e70. Hoffmann RW, Meder J, Klein M, Osterkorn K, Negele RD. Studies on lesions caused by plerocercoids of Triaenophorus nodulosus in some fish of an alpine lake, the Königssee. J Fish Biol 1986;28:701e12. Maizels RM, Balic A, Gomez-Escobar N, Nair M, Taylor MD, Allen JE. Helminth parasites e masters of regulation. Immunol Rev 2004;201:89e116. Erasmus DA. The migration of Cercaria X Baylis (Strigeida) within the fish intermediate host. Parasitology 1959;49:173e90. Whyte SK, Secombes CJ, Chappell LH. Studies on the infectivity of Diplostomum spathaceum in rainbow trout (Oncorhynchus mykiss). J Helminthol 1991;65:169e78. Kuperman BI. Tapeworms of the genus Triaenophorus, parasites of fishes. New Delhi, India: Amerind Publishing Company; 1981. Schöttler J. Untersuchungen zum Verhalten von Plerocercoiden des Fischbandwurms Triaenophorus nodulosus (Pallas, 1781) (Cestoda, Pseudophyllidea) in verschiedenen Fischarten des Königssees. Dissertation. University of Munich; 1984. Henrich T, Benesh DP, Kalbe M. Hybridization between two cestode species and its consequences for intermediate host range. Parasites Vectors 2013;6:33. Reusch TBH, Wegner KM, Kalbe M. Rapid genetic divergence in postglacial populations of threespine stickleback (Gasterosteus aculeatus): the role of habitat type, drainage and geographical proximity. Mol Ecol 2001;10: 2435e45. Hendry AP, Taylor EB. How much of the variation in adaptive divergence can be explained by gene flow? An evaluation using lake-stream stickleback pairs. Evolution 2004;58:2319e31. McKinnon JS, Mori S, Blackman BK, David L, Kingsley DM, Jamieson L, et al. Evidence for ecology’s role in speciation. Nature 2004;429:294e8. 140 F. Franke et al. / Fish & Shellfish Immunology 36 (2014) 130e140 [53] Olafsdottir GA, Ritchie MG, Snorrason SS. Positive assortative mating between recently described sympatric morphs of Icelandic sticklebacks. Biol Lett 2006;2:250e2. [54] Vines TH, Schluter D. Strong assortative mating between allopatric sticklebacks as a by-product of adaptation to different environments. Proc R Soc B 2006;273:911e6. [55] Jones FC, Grabherr MG, Chan YF, Russell P, Mauceli E, Johnson J, et al. The genomic basis of adaptive evolution in threespine sticklebacks. Nature 2012;484:55e61. [56] McPhail JD. Ecology and evolution of sympatric sticklebacks (Gasterosteus) e morphological and genetic evidence for a species pair in Enos Lake, British Columbia. Can J Zool 1984;62:1402e8. [57] Bell MA, Foster SA. The evolutionary biology of the threespine stickleback. Oxford, UK: University Press; 1994. [58] Aw D, Silva AB, Palmer DB. Immunosenescence: emerging challenges for an ageing population. Immunology 2007;120:435e46. [59] Nishimura N, Heins DC, Andersen RO, Barber I, Cresko WA. Distinct lineages of Schistocephalus parasites in threespine and ninespine stickleback hosts revealed by DNA sequence analysis. PLOS ONE 2011;6:e22505. [60] Heins DC, Birden EL, Baker JA. Host mortality and variability in epizootics of Schistocephalus solidus infecting the threespine stickleback, Gasterosteus aculeatus. Parasitology 2010;137:1681e6. [61] Giles N. Behavioral-effects of the parasite Schistocephalus solidus (cestoda) on an intermediate host, the 3-spined stickleback, Gasterosteus aculeatus. Anim Behav 1983;31:1192e4. [62] Barber I, Huntingford FA. The effect of Schistocephalus solidus (Cestoda: pseudophyllidea) on the foraging and shoaling behaviour of three-spined sticklebacks, Gasterosteus aculeatus. Behaviour 1995;132:1223e40. [63] Milinski M. Risk of predation of parasitized sticklebacks (Gasterosteus aculeatus L.) under competition for food. Behaviour 1985;93:203e15. [64] Giles N. Predation risk and reduced foraging activity in fish - experiments with parasitized and non-parasitized 3-spined sticklebacks, Gasterosteus aculeatus L. J Fish Biol 1987;31:37e44. [65] Godin JGJ, Sproul CD. Risk taking in parasitized sticklebacks under threat of predation effects of energetic need and food availability. Can J Zool 1988;66:2360e7. [66] Kennedy CR, Shears PC, Shears JA. Long-term dynamics of Ligula intestinalis and roach Rutilus rutilus: a study of three epizootic cycles over thirty-one years. Parasitology 2001;123:257e69.