Aphragmophora, Tokioka, 1965

B, Sam Peter, B, Manoj Kumar Bhaskaran Nair & Pillai, Devika, 2020, Evolutionary analyses of phylum Chaetognatha based on mitochondrial cytochrome oxidase I gene, Turkish Journal of Zoology 44 (6), pp. 508-518 : 514-516

4.1. Division I- Aphragmophora and Ctenodontina/ Flabellodontina hypothesis

Studiesontheinternalsystematicsinchaetognaths ( Nielsen, 2001; Papillon et al., 2006; Perez et al., 2014) revealed two major groups, Phragmophora and Aphragmophora , on the basis of the occurrence of the phragms. Throughout the debate on chaetognath evolutionary trends, authors like Tokioka (1965a) and Casanova (1985) agreed to consider the presence of phragms as a plesiomorphic state but with slightly different hypotheses. Salvini-Plawen (1986) suggested a radically different concept which contradicted the primitiveness of phragms and identified Pterosagittidae as the sister group to all remaining families.

Later, Bieri (1991a) pointed out a possible relationship between P. draco and species belonging to the family Sagittidae . The inclusion of P. draco within Sagittidae has been corroborated by many reports ( Harzsch et al., 2009; Gasmi et al., 2014). In agreement with these reports, our study also showed an assemblage of P. draco

PETER et al. / Turk J Zool

( Pterosagittidae ) to that of Sagittidae species. Although there is only one species that was taken into account from Krohnittidae , the K. subtilis ascended as sister-species to S. enflata by both analyses and showed a close assemblage to that of Sagittidae species. As stated by Gasmi et al. (2014) using both morphological and molecular data, monophyly of Sagittidae were not retrieved in our analyses and revealed that Sagittidae is strictly paraphyletic. Hence, we propose that the Aphragmophora division encompassed Sagittidae comprising Pterosagittidae and Krohnittidae families and our analyses revives the concept of Aphragmophora , a clade invalidated by Papillon et al. (2006). In parallel to our findings, the first molecular study conducted by Telford and Holland (1997) using LSU rRNA gene upheld the concept Aphragmophora by including Sagittidae , Ptreosagittidae, and Krohnittidae under a unique clade. Again, a recent phylogenetic study conducted by Gasmi et al. (2014) using both SSU and LSU rRNA genes were also supported the monophyly of Aphragmophora with the Pterosagittidae included in the Sagittidae . However, our findings undermined an earlier hypothesis proposed by Papillon et al. (2006) using 26 sequences of the SSU rRNA isolated from members of six extant families. According to them, the order Aphragmophora is monophyletic without Pterosagitta draco , the only living representative of pterosgittidae family.

Finally, moving on to Tokioka’s biclassification concept of Aphragmophora into two sub-orders ( Flabellodontina containing the family Krohnittidae and Ctenodontina containing families Sagittidae and Pterosagittidae ), our study established that Sagittidae sensu stricto is a paraphyletic assemblage from which P. draco and K. subtilis derives. Morphological studies conducted by many scientists were already disproved this concept and added that further division of Aphragmophora into Ctenodontina/ Flabellodontina is not relevant (Salvini-Plawen, 1986; Casanova, 1996 and Gasmi et al., 2014). Later, Papillon et al. (2006) and Gasmi et al. (2014) using the molecular phylogeny of a portion of ribosomal (rRNA) genes also disproved this biclassification concept. Hence, the Ctenodontina and Flabellodontina concept and the hypothesis based on the structure of the cephalic armature were not supported.

4.2. Division II- Phragmophora and validity of Biphragmophora/ Monophragmophora and Syngonata / Chorismogonata hypotheses

According to our results, earlier classification which included Eukrohnia , Heterokrohnia, and Spadella in a single family viz., Eukrohniidae as proposed by Von Ritter-Zahony (1911) and Hyman (1959) is invalid. In parallel to the statement proposed by Gasmi et al. (2014) who used SSU and LSU rRNA genes, both the Bayesian and ML trees formed by COI gene were able to separate the species of Eukrohniidae , Spadellidae , and Heterokrohniidae in three separate clades. As stated by Telford and Holland (1997) who used the LSU rRNA gene, the grouping of Eukrohniidae , Spadellidae , and Heterokrohniidae under the monophyletic division of Phragmophora is found well supported for the available molecular datasets studied and thereby invalidated Gasmi’s concept of paraphyly of Phragmophora ( Gasmi et al., 2014) . Again, our results underscored an earlier morphological hypothesis proposed by Tokioka (1965a, 1965b) and Salvini-Plawen (1986) regarding the monophyly of Phragmophora and undermined their concept of inclusion of Heterokrohniidae under Eukrohniidae .

Our study unambiguously confirmed the monophyly of Eukrohniidae , since Eukrohnia bathyantarctica , E. fowleri , E. hamata , and E. macroneura produced a unique assemblage with support values 1/81.9. This result was in accordance with recent phylogenetic analyses where a close relationship was observed in species under the family Eukrohniidae ( Jennings et al., 2010, Gasmi et al., 2014). The molecular analyses supported the division of Phragmophora into two monophyletic groups, the Monophragmophora and Biphragmophora. Phylogenetic trees showed Casanova’s concept of Monophragmophora ( Eukrohniidae and Spadellidae ) as a natural group, yet with low robust values (0.64/54). In agreement with the Casanova’s hypothesis, when placed Heterokrohniidae under the sub-division Biphragmophora, the available set of sequences of Heterokrohnia species produced a distinctive clade. Hence, the subdivisional concept of Biphragmophora was found true and rejected the statement proposed by Papillon et al. (2006). However, to definitely conclude such a sister-group relationship between these three families ( Eukrohniidae , Spadellidae and Heterokrohniidae ), broader COI gene sequences from various species of Heterokrohniidae , meso-bathyplanktonic Eukrohniidae , and representative of Hemispadella genus, a link between the families Heterokrohniidae and Spadellidae, ( Casanova, 1996) need to be studied. Moving on to the biclassification concept of Casanova in to Syngonata and Chorismogonata, a clear separation was detected between the species under Phragmophora and Aphragmophora , and thereby the Syngonata and Chorismogonata hypothesis found undermined. Earlier studies conducted by Papillon et al. (2006) and Gasmi et al. (2014) already rejected the Syngonata and Chorismogonata hypothesis.

Although this study provides some coverage of species of phylum Chaetognatha, it is not a complete analysis of ca. 130 chaetognath species from the global oceans ( Miyamoto et al., 2014). Taxonomic coverage was uneven for Heterokrohniidae , Krohnittidae , and Spadellidae families. Hence, an expanded database of chaetognaths COI barcodes is needed to improve the accuracy of species identification and phylogeny of this complex group of organisms. Further, it is well known that an evolutionary tree (gene tree) constructed from DNA sequences for a genetic locus does not necessarily approve with the tree that represents the real evolutionary pathway of the species involved (species tree). Therefore, one has to use DNA sequences from various loci that have evolved independently of each other to predict the actual evolutionary relationship of organisms ( Pamilo et al., 1988). Although we used only a single set of gene locus (COI) in our analyses, we were able to compare our results with previously proposed major hypotheses using various molecular loci and thereby provided new insights into the evolutionary relationships of chaetognaths.