Echiniscus

Guil, Noemí, Jørgensen, Aslak, Giribet, Gonzalo & Kristensen, Reinhardt Møbjerg, 2013, Congruence between molecular phylogeny and cuticular design in Echiniscoidea (Tardigrada, Heterotardigrada), Zoological Journal of the Linnean Society 169 (4), pp. 713-736 : 732-733

DESIGN IN ECHINISCUS

Contrary to the conflict found among the dorsal plate configuration in echiniscid genera, and molecular phylogenetic data (using 18S rRNA and 28S rRNA), cuticular design seems to contain evolutionary signal within Echiniscus . However, the traditional cuticular design groups (see Ramazzotti & Maucci, 1983; Peluffo et al., 2002; Pilato et al., 2007, 2008), i. e. bigranulatus , blumi-canadensis , merokensis , arctomys, and viridis ( Fig. 1 and Figs S1 –7), did not coincide with the groups found in our parsimony network ( Fig. 5 ). The three groups of Echiniscus , supported by the AMOVA (named I, II, and III; Fig. 5 ), show different types of cuticular design. One group (group I, Fig. 5 ) is for the E. blumi-canadensis group, with the typical polygonal sculpture of blumicanadensis . Another group (group II) includes species with granulation in their cuticles, from the large, roundish, densely distributed granulation of E. viridissimus to E. spiniger and E. testudo with small roundish pores of different sizes, regularly but not densely distributed, E. jenningsi and E. wendti with similar cuticular designs of very minute granulation regularly and densely distributed ( Dastych, 1984), and finally E. bigranulatus , with mixed large and fine granulation evenly distributed. In the third group (group III), we include two E. blumicanadensis , which could be a case of misidentification ( Guil & Giribet, 2009), E. granulatus (pores structured in polygonal areas; Ramazzotti & Maucci, 1983) and E. merokensis , which shows a cuticle with pores of various sizes and shapes and smooth cuticle between pores ( Fig. 5 ). In contrast to the evolutionary signal in cuticular design, we found little biogeographical signal in our data even at broad geographical scale: for example, E. blumi from Chile and Greenland share 18S rRNA and/or 28S rRNA haplotypes; this is also the case of E. merokensis from Spain and Greenland, and of E. testudo from Greece, France, and Denmark. This may support the idea that microscopic animals can achieve broad distributions mediated by long-distance passive dispersal ( Fenchel & Finlay, 2004).

Monophyly of the different Echiniscus species analysed is phylogenetically supported ( Fig. 4 ), but with different data sets, depending on the species: 28S rRNA supports monophyly of four Echiniscus species ( E. merokensis , E. wendti , E. spiniger , and E. testudo ) while 18S rRNA confirms monophyly of three species ( E. granulatus , E. spiniger , and E. testudo ), and one complex of species ( Echiniscus blumi-canadensis ). Two species or complexes of species remain problematic. Echiniscus bigranulatus is monophyletic except for specimen Tar756 (GenBank accession number: JX114856 ; Fig. 3 and Fig. S1 ). The Echiniscus blumi-canadensis complex (comprising E. blumi , E. canadensis , E. mediantus Marcus, 1930 , E. trisetosus , E. dearmatus Bartoš, 1935 , and probably E. marleyi Li, 2007 ) has not been supported by 28S rRNA data, as opposed to the rest of the Echiniscus species complexes studied ( Fig. 4B ). In contrast, the blumi-canadensis complex finds 18S rRNA support ( Fig. 4A ). However, this complex of species has been problematic for a long time, due to high morphological variability ( Guil, 2008) not reflected in the molecular information (at least for COI data; Guil & Giribet, 2009), and apparently supported by the present study (morphospecies of the complex, E. blumi , E. canadensis , and E. trisetosus are not phylogenetically differentiated either by 18S rRNA data or 28S rRNA). Two specimens of the blumi-canadensis complex had different sequences (coded as Tar612 and Tar635; GenBank accession numbers: FJ435717 , FJ435782 , FJ435718 , and FJ435783 ; Table 1) when compared with the rest of the blumi-canadensis individuals. These two specimens (Tar612 and Tar635) were closely related to E. granulatus , indicating a possible misidentification (for these GenBank sequences there is no morphological voucher).

With the current sampling two clear trends are noted: (1) the distribution of plates within the family Echiniscidae is in conflict with the phylogenetic information derived from 18S and 28S rRNA sequence data; and (2) the cuticular design contains evolutionary signal congruent with the 18S rRNA information within Echiniscus . Together with morphological and any other source of information, this would contribute towards a more integrative taxonomic approach within this group of minute animals. We also emphasize the importance of generating and making available morphological information for the study of these tiny animals, as argued previously by Pleijel et al. (2008).

Figures

Figure 1. The five types of cuticular designs traditionally used to group Echiniscus species and the recently described genus Diploechiniscus (Ramazzotti & Maucci, 1983; Peluffo et al., 2002; Pilato et al., 2007, 2008): A, bigranulatus; B, blumi-canadensis; C, merokensis; D, arctomys; E, viridis; F, D. oihonnae. A to D and F were made using phase contrast. E was made using differential interference contrast. Scale bars: 10 μm.

Figure 3. Bayesian phylogram obtained with combined 18S and 28S rRNA, aligned with MUSCLE and trimmed with GBlocks, using all taxa considered in the present study (i.e. outgroups and eutardigrades from Table 3, heterotardigrades from Table 2, and Echiniscus species from Table 1). Above branches are posterior probabilities obtained in the Bayesian analysis are provided. Below branches are bootstrap support values from the ML analysis.

Figure 4. Bayesian phylogram obtained from the analysis of 18S rRNA (A) and 28S rRNA (B) information for Echiniscus species and Diploechiniscus oihonnae, using Testechiniscus spitzbergensis as outgroup. Values above branches indicate posterior probabilities obtained with the Bayesian analysis. Bootstrap support values from ML analysis are provided below

Figure 5. Parsimony network obtained with TCS for the 18S rRNA information from Echiniscus species. A photo with the cuticular design for each species is provided. E. merokensis SP, Spanish Echiniscus merokensis merokensis. E. merokensis Tar759 SUE, subspecies Echiniscus merokensis suecicus. The three supported groups found among the Echiniscus species, based on cuticle design, are identified with dotted squares, and named as I, II, and III.