In this study, we investigated metazoan parasite communities in two intergeneric cyprinid species, native and endemic Parachondrostoma toxostoma and invasive Chondrostoma nasus, sampled in allopatric and sympatric populations. Using mtDNA and microsatellites, we showed that the frequencies of C. nasus and P. toxostoma in the localities situated in sympatric zones varied from the more abundant C. nasus or more abundant P. toxostoma to an approximately balanced representation of the two species. This pattern of distribution of these two fish species in the sympatric zone of the Durance River was previously shown by Costedoat et al.. We showed that the frequencies of C. nasus and P. toxostoma in the localities studied determine the composition of metazoan parasite communities, i.e. the localities with the highest frequencies of C. nasus were the localities with the highest proportion of Monogenea.
Specimens of P. toxostoma from the allopatric population were rarely infected by Monogenea. Instead, endoparasite species (i.e. mainly Trematoda) formed the dominant component of parasite communities. By contrast, C. nasus specimens from the allopatric population were not parasitized by endoparasite species and, in this case, Monogenea (especially Dactylogyrus species) formed the dominant component of parasite communities. The difference in endoparasite species richness between allopatric P. toxostoma and allopatric C. nasus could be explained by their different feeding preferences[26–29] (which are linked to different mouth morphology, following Corse et al.) or by different abiotic and biotic factors of their habitats (i.e. the presence of invertebrates like mollusca, isopoda, ostracoda and copepoda, which serve as intermediate hosts for endoparasites). The difference in ectoparasite species richness observed between the allopatric population of P. toxostoma and the allopatric population of C. nasus could also indicate that allopatric P. toxostoma in southern France is rarely infected by monogenean species or is free of several monogenean species widely infecting C. nasus. However, to verify such a hypothesis, the further sampling of P. toxostoma from allopatric populations is needed. Our sampling was limited only to one allopatric population, taking into consideration the threatened and protected status of P. toxostoma in southern France. Up to now, the investigation of parasite fauna in the allopatric population of P. toxostoma has only been performed in the Viaur River (southwest France) by Loot et al., who also found the low prevalence and abundance of Dactylogyrus and Gyrodactylus parasites. However, they found that P. toxostoma was infected only by two endoparasite species, which suggests that the endoparasite infection in our allopatric P. toxostoma is more likely the result of environmental conditions of the habitat.
In our study, we showed the high similarity between allopatric and sympatric populations of C. nasus, which may suggest that this species expands together with its original parasite fauna. The ectoparasites (especially Dactylogyrus species) maintain a similar intensity of infection in both allopatric and sympatric populations of C. nasus; thus, these parasites do not represent a factor limiting the survival and dispersal activity of their host species and their presence is probably the result of co-evolutionary host-parasite interactions, as shown for Dactylogyrus species and cyprinid fish by Šimková et al.. On the other hand, low similarity in parasite communities between the allopatric and sympatric populations of P. toxostoma may indicate that P. toxostoma secondarily acquired the parasites (especially Dactylogyrus) after coming into contact with C. nasus. From this point of view, the changes in parasite communities in P. toxostoma linked to the invasion of C. nasus into areas originally inhabited solely by P. toxostoma could represent a danger for native endemic species. However, following the general trend of low Dactylogyrus abundance in P. toxostoma in relation to C. nasus observed in sympatric zones, it seems that Dactylogyrus infection probably has a genetic basis limiting the intensity of infection in endemic P. toxostoma. Among Dactylogyrus species, Dactylogyrus chondrostomi was found to be a species infecting C. nasus living in allopatric and sympatric populations with a high prevalence. This parasite species was absent in four populations and found only rarely in one population of P. toxostoma in sympatric zones, which indicates that D. chondrostomi is specific to C. nasus. It seems that higher parasite fitness (measured by parasite abundance) in C. nasus is associated with a system of co-adaptation genes.
On the basis of mtDNA and microsatellite data, different genotypes of P. toxostoma x C. nasus hybrids were identified. However, our results indicate that the proportion of hybrids and their genotypes across the two sympatric zones (the Durance River and the Ardeche River) are different. These results confirm the findings of a previous study in the Durance hybrid zone and represent the first data obtained from the Ardeche River. The proportions of hybrids were unequal and sometimes low (ranging from 5 to 20%) depending on the considered localities. This fact together with the absence of P. toxostoma or C. nasus in some localities and the low frequencies of one of the Chondrostoma or Parachondrostoma species from other localities in the sympatric zones of the Durance and Ardeche Rivers did not allow us to test simultaneously the effect of locality and host on parasite species richness or abundance. Although parasite load was significantly influenced by the effect of locality, we showed the same pattern of parasite infection in hybrids at two different levels (1) using pooled data and (2) selecting one locality with the largest sample size within the Durance sympatric zone. Thus, P. toxostoma x C. nasus hybrids were less infected by ectoparasites (and especially by monogeneans of Dactylogyrus genus) than C. nasus. However, as no difference in ectoparasite abundance between hybrids and P. toxostoma was found, it seems that Dactylogyrus infection in C. nasus is more likely the result of co-evolutionary history between C. nasus and Dactylogyrus parasites, which limits the infection in both P. toxostoma and hybrids. Thus, both P. toxostoma and hybrids probably serve as additional hosts for Dactylogyrus. The susceptibility of Salmo salar x Salmo trutta hybrids to Gyrodactylus salaris (highly virulent) and G. derjavini (viviparous Monogenea) was experimentally tested and the intermediate pattern of hybrid susceptibility to that of the parents was shown. They suggested that resistance was transferred through interspecific crosses as a dominant trait. However, our study seems to indicate that the low susceptibility of hybrids to Dactylogyrus infection is linked to the presence of P. toxostoma genes in recombinant genotypes. Metazoan parasite communities in another cyprinid hybrid system (Alburnus alburnus x Rutilus rutilus hybrids from Lake Micri Prespa, Northern Greece), were investigated by Dupont and Crivelli. A higher susceptibility to metazoan parasite infection in hybrids compared to pure species was found for Dactylogyrus and Diplozoon species (Monogenea), Bolbophorus confusus (larval stages of Trematoda) and Pomphorhynchus bosniacus (Acanthocephala). This was explained by the spatial and trophic positions of the hybrids, which were intermediate between the two pure species. Concerning P. toxostoma x C. nasus, Corse et al. concluded that pure and hybrid specimens in the hybrid zone exhibit more diverse feeding behaviour than in the allopatric zone consistent with generalist behaviour. P. toxostoma x C. nasus hybrids are not intermediate between pure species and hybrids have “super P. toxostoma” feeding behaviour i.e. they feed on fewer diatomes and more invertebrates than both pure species. However, in our study, there was a trend towards lower endoparasite abundance in hybrids compared to both P. toxostoma and C. nasus, suggesting that there is no link between the frequencies of invertebrates (as the potential intermediate hosts) in the food and infection by endoparasite species in sympatric zones.
Concerning Monogenea, the strictly host-specific Dactylogyrus in the study by Dupont and Crivelli achieved lower prevalence in Alburnus alburnus x Rutilus rutilus hybrids compared to pure species. A similar result, i.e. the effect of introgression rate on Diplozoon infection (Monogenea) in the hybrids of Barbus barbus and B. meridionalis from a hybrid zone in southern France, was found, in which parasite prevalence correlated with the percentage of B. meridionalis genes. In our study, Dactylogyrus chondrostomi, a parasite specific to C. nasus, was absent in hybrids reported in five localities where both host fish species live in sympatry, and its presence was confirmed only in hybrids at Avignon, a single locality investigated in the sympatric zone on the Durance River, where 92% of collected specimens were determined as C. nasus. The high frequency of C. nasus seems to represent a more plausible explanation for the presence of D. chondrostomi in hybrids.