Distinct genetic differentiation and species diversification within two marine nematodes with different habitat preference in Antarctic sediments

Abstract

Background: Dispersal ability, population genetic structure and species divergence in marine nematodes are still poorly understood, especially in remote areas such as the Southern Ocean. We investigated genetic differentiation of species and populations of the free-living endobenthic nematode genera Sabatieria and Desmodora using nuclear 18S rDNA, internal transcribed spacer (ITS) rDNA, and mitochondrial cytochrome oxidase I (COI) gene sequences. Specimens were collected at continental shelf depths (200-500 m) near the Antarctic Peninsula, Scotia Arc and eastern side of the Weddell Sea. The two nematode genera co-occurred at all sampled locations, but with different vertical distribution in the sediment. A combination of phylogenetic (GMYC, Bayesian Inference, Maximum Likelihood) and population genetic (AMOVA) analyses were used for species delimitation and assessment of gene flow between sampling locations.

Results: Sequence analyses resulted in the delimitation of four divergent species lineages in Sabatieria, two of which could not be discriminated morphologically and most likely constitute cryptic species. Two species were recognised in Desmodora, one of which showed large intraspecific morphological variation. Both genera comprised species that were restricted to one side of the Weddell Sea and species that were widely spread across it. Population genetic structuring was highly significant and more pronounced in the deeper sediment-dwelling Sabatieria species, which are generally less prone to resuspension and passive dispersal in the water column than surface Desmodora species.

Conclusions: Our results indicate that gene flow is restricted at large geographic distance in the Southern Ocean, which casts doubt on the efficiency of the Weddell gyre and Antarctic Circumpolar Current in facilitating circum-Antarctic nematode species distributions. We also show that genetic structuring and cryptic speciation can be very different in nematode species isolated from the same geographic area, but with different habitat preferences (surface versus deeper sediment layers).

Keywords: Antarctica; Continental shelf; Cryptic species; Desmodora; Dispersal; Nematoda; Phylogeny; Population genetics; Sabatieria.

Fig. 1

Map of Antarctica highlighting the geographic location of the five sampling stations. Box 1: Scotia Sea: SG = South Georgia, SO = South Orkneys; Box 2: Antarctic Peninsula: KG = King George; Box 3: eastern Weddell Sea: AUS = off Auståsen, BX = Bendex. The same colour code is maintained in figures and graphs throughout the manuscript. Adapted from cruise plot ANT-XXVII/3 [41] © Alfred Wegener Institute

Fig. 2

Phylogeny and population genetic haplotype networks for ITS of Sabatieria. Upper left corner: Log-lineages through time plot (LTT) indicating position of threshold time T (red line). Middle: Bayesian tree of ITS haplotypes of Sabatieria; numbers above branches indicate posterior probabilities, numbers below (where indicated) are ML bootstrap percentages (only when values >50%). Number of populations (i.e. geographical locations) and haplotypes are indicated next to each clade. Right: corresponding TCS haplotype networks of all four ITS clades for Sabatieria. Values along branches indicate the number of base pair differences between the two connecting haplotypes. When this number is not indicated, there was only 1 mutation. Black dots represent missing haplotypes. Size of circles is proportional to the amount of individuals belonging to that specific haplotype. Colour code based on the different locations

Fig. 3

Bayesian trees for a 18S and b COI of Sabatieria. Numbers above branches indicate posterior probabilities as calculated by beast procedure, while numbers below branches depict ML bootstrap percentages from RAxML files. Only values above 50 are included in the graphs. Scale length represents number of substitutions per site. Colours represent location

Fig. 4

Bayesian trees of a ITS, and b COI haplotypes for Desmodora. Names of specimens are linked to morphological groups (here indicated as ‘morphospecies’). Numbers above branches are posterior probabilities; number below branches are maximum likelihood bootstrap values (only indicated when >50%). Colours represent locations. Scale length represents number of substitutions per site. The COI plot also includes the log-lineages through time plot with threshold time T indicated in red, and the haplotype network for Desmodora species I. Numbers along branches indicate the amount of mutations/base pair differences between the two connecting haplotypes. When this number is not indicated, there was only 1 mutation. Black dots represent missing haplotypes. Size of circles is proportional to the amount of individuals belonging to that specific haplotype. Colour code based on the different locations