Global invasion genetics of two parasitic copepods infecting marine bivalves

Abstract

Invasive species, and especially invasive parasites, represent excellent models to study ecological and evolutionary mechanisms in the wild. To understand these processes, it is crucial to obtain more knowledge on the native range, invasion routes and invasion history of invasive parasites. We investigated the consecutive invasions of two parasitic copepods (Mytilicola intestinalis and Mytilicola orientalis) by combining an extensive literature survey covering the reported putative native regions and the present-day invaded regions with a global phylogeography of both species. The population genetic analyses based on partial COI sequences revealed significant population differentiation for M. orientalis within the native region in Japan, while introduced populations in North America and Europe could not be distinguished from the native ones. Thus, M. orientalis' invasion history resembles the genetic structure and recent spread of its principal host, the Pacific oyster, Crassostrea gigas, while M. intestinalis lacks population genetic structure and has an overall low genetic diversity. Therefore, the native origin of M. intestinalis remains unclear. With this study, we demonstrate that even highly related and biologically similar invasive species can differ in their invasion genetics. From this, we conclude that extrapolating invasion genetics dynamics from related invasive taxa may not always be possible.

Figure 1

Temporal development of the known distributions and invasions of (A) Mytilicola intestinalis and (B) Mytilicola orientalis from time of first observation and species description (i.e., 1902 for M. intestinalis and 1935 for M. orientalis) until December 2017. Locations where Mytilicola spp. presence has been recorded are indicated with red dots, showing the species distribution, invasion route and spread, as well as research effort through time. For temporal development of Mytilicola spp. invasions decade-by-decade, including species absence records, see Supplementary Figs S1 and S2. For animated temporal development of Mytilicola spp. year-by-year, see the .gifs in Online Resources 1 and 2. This figure is based on data from this study and from data acquired through our literature surveys. See Supplementary Information for reference lists.

Figure 2

Minimum spanning networks for COI haplotypes (circles) in (A) Mytilicola intestinalis and (B) M. orientalis. Branch length is proportional to the number of nucleotide differences (the shortest branches represent one nucleotide difference). The black dots on the branches represent missing haplotypes. Frequency of observation is proportional to circle area. Colors represent large scale differentiation between clusters of haplotypes and correspond with the pie charts in Fig. 4. Figure 2 was drawn in Adobe Illustrator based on the Arlequin output list of OTU differences.

Figure 3

Mismatch distributions of Mytilicola intestinalis (top row) and Mytilicola orientalis (lower row). From left to right: mismatch distributions based on population differences (M. intestinalis: all locations pooled; M. orientalis: Iwaya separate from the rest of the locations), mismatch distributions in the putative native region and in the invaded region(s). The sharp decline in M. intestinalis is indicative of population expansion after a bottleneck or selective sweep; the bimodal distributions in M. orientalis are indicative of stable demographics.

Figure 4

Maps showing pie charts of cytochrome-c-oxidase 1 (COI) haplotype frequencies in the invasive parasite species Mytilicola intestinalis and Mytilicola orientalis. Shown plotted overlaying the maps are pie charts for the different sampling locations, while the large pie charts in the corner of the maps represent the grouped frequencies for all locations in that map. On the side, the large pie charts represent the grouped frequencies for all invaded European locations for M. intestinalis and M. orientalis (“Europe”), and for all invaded locations for M. orientalis that includes the North American samples (“Invaded”). Note that the pie charts from the Mediterranean and Japan represent the overall frequencies of all sampled native sampling locations for M. intestinalis and M. orientalis, respectively. Pie charts are smaller for sampling locations with fewer than 15 individuals, which were not included in our analyses. Colors correspond to large scale differentiation between clusters of haplotypes in the minimum spanning networks presented in Fig. 2. Pie charts were made in R statistical environment version 3.4.3, maps were plotted with R packages rworldmap and ggplot and Fig. 4 was assembled in Adobe Illustrator.

Figure 5

Haplotype diversity h and nucleotide diversity π of Mytilicola intestinalis (MI) and Mytilicola orientalis (MO). Putative native ranges are indicted in grey and invaded ranges in black. Bars represent standard deviations. Only samples with more than 15 individuals were included. See Supplementary Tables S1 and S2 for π and h values of individual locations underlying this figure.