No abstract

Four hundred Gyrodactylus species have been formally described, but the estimated number of species in this fish ectoparasite genus of Monogenean Platyhelminthes is more than 20,000. The unusually high species richness has lead to the hypotheses of speciation and adaptive radiation via host switching. These hypotheses were tested by reconstructing a molecular phylogeny for the subgenus G. (Limnonephrotus) which is a group of freshwater parasites, including five species infecting wild and farmed salmonids. The highly variable ITS1 and ITS2 segments and the conservative 5.8S ribosomal gene were sequenced in 22 species plus two species representing the subgenus G. (Paranephrotus) as an outgroup. The phylogeny was compared with host systematics: the species were collected from six fish families (Cyprinidae, Salmonidae, Percidae, Esocidae, Gasterosteidae, and Gobitidae). The phylogenetic analysis demonstrated that G. (Limnonephrotus) is a monophyletic group that was originally hosted by cyprinids. The speciation has occurred in two episodes, the older one manifested in genetic distances 25-33% (4-6 Myr BP). The latter speciation burst occurred in one clade only, perhaps one million years ago. This clade has been morphologically identified as a wageneri species group. It is a monophyletic group of 18 species [studied here] and contains all five salmonid parasites, but also parasites, on cyprinids, percids, esocids, and gasterosteids. In G. (Limnonephrotus), eight host switches crossing the host family barrier were observed, and at least three of them were followed by repetitive speciation. Seven host-switch events were statistically confirmed by bootstrapping. The suggested model of speciation by host switch was accepted, and interestingly the adaptive radiation seems to be a consequence of host switch to a new family (key innovation model). The molecular and ecological evolution rate of Gyrodactylus parasites is manyfold in comparison to host species, and the phylogenies are largely independent and disconnected.

Abstract. Four hundred Gyrodactylus species have been formally described, but the estimated number of species in this fish ectoparasite genus of Monogenean Platyhelminthes is more than 20,000. The unusually high species richness has lead to the hypotheses of speciation and adaptive radiation via host switching. These hypotheses were tested by reconstructing a molecular phylogeny for the subgenus G. (Limnonephrotus) which is a group of freshwater parasites, including five species infecting wild and farmed salmonids. The highly variable ITS1 and ITS2 segments and the conservative 5.8S ribosomal gene were sequenced in 22 species plus two species representing the subgenus G. (Paranephrotus) as an outgroup. The phylogeny was compared with host systematics: the species were collected from six fish families (Cyprinidae, Salmonidae, Percidae, Esocidae, Gasterosteidae, and Gobitidae). The phylogenetic analysis demonstrated that G. (Limnonephrotus) is a monophyletic group that was originally hosted by cyprinids. The speciation has occurred in two episodes, the older one manifested in genetic distances 25-33% (4-6 Myr BP). The latter speciation burst occurred in one clade only, perhaps one million years ago. This clade has been morphologically identified as a wageneri species group. It is a monophyletic group of 18 species [studied here] and contains all five salmonid parasites, but also parasites, on cyprinids, percids, esocids, and gasterosteids. In G. (Limnonephrotus), eight host switches crossing the host family barrier were observed, and at least three of them were followed by repetitive speciation. Seven host-switch events were statistically confirmed by bootstrapping. The suggested model of speciation by host switch was accepted, and interestingly the adaptive radiation seems to be a consequence of host switch to a new family (key innovation model). The molecular and ecological evolution rate of Gyrodactylus parasites is manyfold in comparison to host species, and the phylogenies are largely independent and disconnected.

Host switching explains the high species number of ectoparasitic, viviparous, mainly parthenogenetic but potentially hermaphroditic flatworms of the genus Gyrodactylus. The starlike mitochondrial phylogeny of Gyrodactylus salaris suggested parallel divergence of several clades on grayling (also named as Gyrodactylus thymalli) and an embedded sister clade on Baltic salmon. The hypothesis that the parasite switched from grayling to salmon during the glacial diaspora was tested using a 493-bp nuclear DNA marker ADNAM1. The parasites on salmon in lakes Onega and Ladoga were heterozygous for divergent ADNAM1 alleles WS1 and BS1, found as nearly fixed in grayling parasites in the White Sea and Baltic Sea basins, respectively. In the Baltic salmon-specific mtDNA clade, the WS/BS heterozygosity was maintained in 23 out of the 24 local clones. The permanently heterozygous clade was endemic in the Baltic Sea basin, and it had accumulated variation in mtDNA (31 variable sites on 1600 bp) and in the alleles of the nuclear locus (two point mutations and three nucleotide conversions along 493 bp). Mendelian shuffling of the nuclear alleles between the local clones indicated rare sex within the clade, but the WS/BS heterozygosity was lost in only one salmon hatchery clone, which was heterozygous WS1/WS3. The Baltic salmon-specific G. salaris lineage was monophyletic, descending from a single historical hybridization and consequential host switch, frozen by permanent heterozygosity. A possible time for the hybridization of grayling parasite strains from the White Sea and Baltic Sea basins was during the Eemian interglacial 132 000 years bp. Strains having a separate divergent mtDNA observed on farmed rainbow trout, and on salmon in Russian lake Kuito were suggested to be clones derived from secondary and tertiary recombination events.