Population genetic structure of the salmon louse, Lepeophtheirus salmonis (Kroyer) on wild and farmed salmonids around the Pacific coast of Canada.

摘要

In this study, the hypothesis of whether population genetic structure existed among salmon lice (Lepeophtheirus salmonis) populations collected from different geographical regions of British Columbia, Canada, existed because of restricted gene flow was tested. Samples of the salmon louse were obtained from three different wild Pacific salmon species (Oncorhynchus tshawytscha, O. gorbuscha and O. kisutch) by sports fishing and during creel (sports fishery catch) surveys and from farmed Atlantic salmon (Salmo salar) by netting. To test for population genetic structure with the largest sample size possible, we decided to use a fragment of the mitochondrial cytochrome c oxidase subunit 1 (COI). Template DNA was extracted from individual salmon lice using modified proteinase K methods. A 710-bp fragment of the COI gene was polymerase chain reaction (PCR) amplified using the primer pairs LCO1490 and HCO2198 or LepF1and LepR1, and visualized in a 96-well E-Gel. A total of 194 COI sequences in the combined dataset between our 175 new samples and 19 previously published from Vancouver Island were included in our population structure analysis. The dataset contained 29 variable sites that were present in at least two salmon lice and comprised 45 haplotypes. We observed considerable population structure not only between lice from wild salmon hosts caught on the east and west coasts of Vancouver Island (Barkley Sound Wild vs. Broughton Wild) but also between lice from wild and farmed salmon hosts within a region (Broughton Wild vs. Broughton Farmed). This later result suggests that migration between the Broughton farmed and Broughton wild populations was low enough to allow the two populations to maintain different haplotype frequencies. Analysis of 185 of our new COI sequences from British Columbia, combined with all previously published COI sequences for L. salmonis, using the Kimura two-parameter distance method and the neighbour-joining tree-building algorithm implemented on BOLD shows two different clades with considerable sequence divergence (4.8-7.7%) between samples collected from the Pacific and samples collected from the Atlantic. The presence of population structure among salmon louse from different regions of British Columbia supports our hypothesis that gene flow is restricted during the salmon louse larva's 4-14-day planktonic period. We recommend that several salmon lice from each fish be preserved and genotyped for the 'BIO barcoding' COI fragment. This will enable the patterns of genetic variation within a host fish to be separated from those among host fish.