Evolutionary histories of symbioses between microsporidia and their amphipod hosts: contribution of studying two hosts over their geographic ranges

Abstract

Microsporidia are obligate endoparasites, exploiting their hosts with either vertical or horizontal transmission. While the former may promote co-speciation and host-specificity, the latter may promote shifts between host species. Freshwater amphipods are hosts for many microsporidian species, but no general pattern of host specificity and co-diversification is known. In my PhD work microsporidian infections, identified with SSU rDNA, were assessed in two Gammarus species complexes, G. roeselii and G. balcanicus , over their full geographic ranges (each c. 100 sites and 2000 individuals) in aim of (i) exploring the microsporidian diversity present in both hosts and their phylogenetic relationships; (ii) testing if the host phylogeographic history might have impacted host-parasite association (co-diversifications or recent host-shifts from local fauna); (iii) proposing the host-parasite evolutionary history scenarios to explain the diversity and co-bio-geographical pattern observed in the two host species between using N. granulosis as a model. The SSU rDNA marker revealed a high number of microsporidian variants (i.e. haplogroups, 24 and 54, respectively), clustered into 18 species-level taxa, almost all being shared between the two host species. However, many microsporidian haplogroups within a given parasite species are host-specific, suggesting host-parasite co-variation. Within each host species-complex, while the confrontation between hosts and parasites phylogeography suggested some degrees of co-diversification, these patterns remain to be confirmed, mainly as SSU rDNA reached its limits in phylogenetic information content in that matter. Strikingly, almost all of these microsporidia taxa were previously detected in other gammarids, mainly within the genus Gammarus, but also in other genera of Gammaridae. Some were already clearly recognised parasite taxa associated with amphipods: Nosema granulosis, Dictyocoela roeselum, D. muelleri, D. roeselum, D. duebenum, D. berillonum, Cucumispora roeselum, C. ornata, C. dikerogammari, Microsporidium sp 515 and Microsporidium sp 505). Many times, my results increased host taxonomic spectrums and extended geographic ranges (often widely). Some other taxa were known to be extremely rare, having scarce literature records often with few or even very few geographic records and being not fully described. My PhD work either extend host taxonomic spectrum and/or deeply extend geographic ranges for these taxa. It allowed a reappraisal for such taxa, changing their status from puzzling anecdotic association to potentially overlooked established associations for amphipods. Among the most common parasite species detected in G. roeselii and G. balcanicus, a few were clearly established in the literature as vertically transmitted and having a feminizing effect on their hosts. In G. balcanicus, the only species for which it could be logistically tested, we were unable to confirm this feature, showing that a single microsporidia species cannot be considered as totally vertically-transmistted through its entire host spectrum. Furthermore, focusing on N. granulosis, the use of an additional marker, i.e. RPB1, allowed: i) to clearly identify divergent clades (while SSU rDNA was unconclusive) and ii) suggested that some peculiar strains of these supposedly feminizing parasites are not transmitted vertically and not feminizing at all. In addition, for G.roeseli, a within-parasite-species polymorphism could exist relative the transmission routes, one lineage being associated with vertical transmission, the other with horizontal transmission. These results are still preliminary, mainly associated with incomplete gammarid host and parasite taxa sampling, and need running further experiments to confirm our hypotheses.