Host repertoire: Host evolution shapes community structure
We inferred two main predictors of cichlid-Cichlidogyrusinteractions: host phylogenetic relationships and host environment. The hosts’ evolutionary history can have a significant impact on the structure of host-parasite networks (e.g. Mouillot et al. 2008a; Braga et al. 2014, 2020). Here, we show that the realised host repertoire of most species of Cichlidogyrus is determined more by the hosts’ evolutionary history than their environment. First, host repertoires differed more regarding the ecological niche than the phylogenetic relationships as evidenced by a reduction of the mean functional-phylogenetic distances (FPDist) with increasing phylogenetic weight (Fig. 4). Second, FPDist estimates rarely differed from the null distribution for increasing functional weight (but see C. sp. ‘nyanza ; Fig. 4). Third, estimates outside the null distribution were underdispersed (clustered), i.e. lower than expected at random at both the ancient and recent evolutionary scale (measured as MPD and MNTD respectively). The strong phylogenetic influence and underdispersion are associated with co-divergent evolution (Clark & Clegg 2017). Co-divergence assumes that host and parasite phylogenies are phylogenetically congruent (Page 2003; Hoyal Cuthill & Charleston 2012), a pattern that has already been observed for a number of species of Cichlidogyrus (Vanhove et al. 2015). Congruence might arise especially in younger parasite lineages that have experienced a phase of isolation resulting in co-divergence (Agosta & Brooks 2020), e.g. species of Cichlidogyrus infecting tropheine cichlids (Vanhove et al. 2015), which arose 4.5–6.5 MYA (Schedel et al. 2019).
Additionally, phylogenetic tracking might explain the role of the host phylogeny on host repertoires. Even within the same ecosystem, parasites are constrained to compatible ecological niches created by host speciation (Russo et al. 2018). Consequently, phylogenetic relationships are determinants of neotropical (Braga et al. 2014) and Mediterranean (Desdevises et al. 2002) monogenean-fish communities but also plant-pollinator, plant-frugivore (Rezende et al. 2007), plant-mycorrhiza (Jacquemyn et al. 2011), and other host-parasite networks (Mouillot et al. 2008). Yet natural (Birgi & Euzet 1983; Birgi & Lambert 1986) and invasion-induced (Jiménez-García et al. 2001; Šimková et al. 2019; Jorissenet al. 2020) host switches of species of Cichlidogyrushighlight that fundamental host repertoires might differ considerably from the realised host repertoires estimated here. Monogeneans have the capacity to infect new hosts and adapt to new environments (Bragaet al. 2014; Messu Mandeng et al. 2015; Nack et al.2016). Thus, conclusions drawn based solely on analyses of host repertoires should be treated with caution.