Prionospio, Malmgren, 1867

Hektoen, Atlantic Martin M., Radashevsky, Vasily I., Bakken, Torkild, Dunshea, Glenn & Ekrem, Torbjørn, 2024, Species delimitation and phylogenetic relationships of the Prionospio complex (Annelida, Spionidae) in the Northeast Atlantic, Zoologica Scripta 53 (3), pp. 358-375 : 369

4.3 | Phylogenetic relationships in Prionospio complex

Analysis of mitochondrial genomes and nuclear ribosomal DNA yielded information regarding the phylogenetic relationships of species in the Prionospio complex. Aurospio banyulensis was found nesting among Prionospio species. The monotypic genus Aurospio was erected for Aurospio dibranchiata Maciolek, 1981a , possessing three morphological characters unique for the Prionospio complex: two pairs of branchiae on chaetigers 3 and 4, branchiae being thin and flat, and branchiae basally fused to the notopodial postchaetal lamellae. Subsequent authors simplified the diagnosis of Aurospio to include all species with branchiae beginning from chaetiger 3. Blake et al. (2020, 58–59) re-established the original diagnosis of Aurospio and noted that “… subsequent researchers ( Mincks et al., 2009; Paterson et al., 2016; Sigvaldadóttir, 1998) have misconstrued the differences between Aurospio and Prionospio and have taken species that clearly belong to Prionospio and referred them to Aurospio .” Nevertheless, Blake et al. (2020, 59) included all of them in the list of Aurospio species (six in total) with a comment that the “issue will be addressed more fully in a subsequent study”. Guggolz et al. (2020) analysed 16S rDNA from 21 deep-sea species of the Prionospio complex, including A. cf. dibranchiata , A. foodbancsia , and four unidentified species with branchiae from chaetiger 3, which they referred to Aurospio . Although the analysis was not well supported, they found Aurospio species appearing in four different clades mixed with Prionospio species, indicating that branchiae from chaetiger 3 probably do not mark a monophyletic clade. Together with two other species referred to Aurospio , A. cf. dibranchiata formed a well supported clade, which, however, was deep inside among Prionospio species ( Guggolz et al., 2020, fig. 2). No unique morphological character shared by members of this clade was noted. We studied two species with branchiae from chaetiger 3 ( A. banyulensis and P. sp. 6) and found them nested among different Prionospio species, thus showing support to the idea that branchiae from chaetiger 3 evolved more than one time within Prionospio . We did not include A. dibranchiata in our analysis, and therefore we cannot comment any further on the status of the genus. Nevertheless, at this point, we suggest returning banyulensis to Prionospio as it was originally assigned by Laubier (1966).

Generic divisions within the Prionospio complex have historically been established primarily based on morphological characters of branchiae. Within Prionospio sensu lato, worms with pinnate and apinnate branchiae from chaetiger 2 were assigned to Prionospio , while those with only apinnate branchiae from chaetiger 2 were assigned to Minuspio . However, our analysis of molecular data proposed that this crucial character has evolved (or been lost) more than once in the evolution of Prionospio . Difficult-to-place species are often referred to Prionospio (e.g., Peixoto & Paiva, 2019) which could artificially inflate the number of species in the genus. Prionospio sp. 6 in our analyses represents such a species, with characters typical for Prionospio , Aurospio , and Laubieriellus . We tentatively consider this species as a member of Prionospio due to its sister relationship with P. cf. dubia . Still, it illustrates problems with the generic systematisation of the Prionospio complex.

Incongruence between mitochondrial and nuclear gene trees has commonly been reported in phylogenetic studies (e.g., Platt et al., 2018), and can usually be attributed to complex evolutionary processes such as incomplete lineage sorting ( Pamilo & Nei, 1988), lack of recombination ( Ballard & Whitlock, 2003), and introgression ( Toews & Brelsford, 2012). These processes can cause the evolutionary history of the mitochondrion to not accurately reflect the group's evolutionary history ( Edwards & Bensch, 2009). In the present study, the incongruence was minor, where only three species were placed differently between the mitochondrial and nuclear trees, and both analyses show traditional taxonomic groups to be polyphyletic. The incongruence could also be an artefact of the conserved nature of 18S and 28S rDNA, and indicates the need for expanded nuclear datasets in future research. All issues in molecular ( Guggolz et al., 2020; Abe & Sato-Okoshi, 2021; present study) and morphological ( Sigvaldadóttir, 1998; Yokoyama, 2007) analyses show that the phylogenetic relationships within the Prionospio complex are still poorly understood and the position of members of this complex requires further study.