New insights into phylogenetic relationships of Rhabdocoela (Platyhelminthes) including members of Mariplanellida

Abstract

Background

Previous flatworm phylogenetic research has been carried out analysing 18S and 28S DNA markers. Through this methodology, Mariplanellinae subfamily has been recently re-classified as Mariplanellida status novus. This new classification implied that 3 genera belonged to Mariplanellida: Mariplanella, Lonchoplanella and Poseidoplanella. In this study, we aim to clarify some of the relationships within Rhabdocoela analysing 18S and 28S DNA markers of a total of 91 species through Maximum Likelihood and Bayesian Inference methodologies. A total of 11 species and genera, including Lonchoplanella, from the island of Sylt are included and had not previously been involved in any molecular phylogenetic analyses.

Results

Our phylogenetic results support Mariplanellida as an independent group within Rhabdocoela and its status as an infraorder. Our study suggests that Lonchoplanella axi belongs to Mariplanellida. Within Rhabdocoela, Haloplanella longatuba is nested within Thalassotyphloplanida, instead of Limnotyphloplanida. Within Kalyptorhynchia, the taxon Eukalyptorhynchia turned out to be paraphyletic including members of Schizorhynchia. These results also support the position of the genus Toia separate from Cicerinidae.

Conclusions

Lonchoplanella axi belongs to Mariplanellida, whose status as infraorder is herein confirmed. The genus Toia belongs separate from Cicerinidae. Further research is needed to clarify the phylogenetic relationships of Hoploplanella. Most of the species, genera and families included in this study with more than one terminal are monophyletic and well supported. Adding gene markers and complementary morphological studies will help to clarify those relationships that remain uncertain.

Background

Flatworms are a large group in terms of diversity, with more than 26,500 described species [1, 2]. Most of them are parasitic, while around 6500 species of them are “free-living Platyhelminthes”. The parasitic flatworms, Neodermata (comprising Trematoda, Monogenea, and Cestoda), are a well-defined and supported clade characterized by a syncytial, nonciliated epidermis whose nuclei-bearing parts lie sunken below the musculature. The rest of platyhelminthes are mostly free-living but also symbiotic flatworms, also known as turbellarians (non-cladistic group); most of them with a ciliated, cellular epidermis [3].

The free living Rhabdocoela and Proseriata are the two most diverse microturbellarian orders, which, in recent phylogenetic hypotheses, are basally branching within Euneoophora [4] (Sup. Figure 1). Proseriata has above 400 described species [5], and Rhabdocoela, has about 1530 described species [6]. Nevertheless, these numbers represent a scarce amount of the estimated total number of microturbellarian species present on Earth, which was estimated to be around 44.000 by Armonies [7]. Further research is needed in order to record their diversity more accurately [8].

Fig. 1

ML phylogenetic tree inferred from the concatenated data set (18S+28S). Species provided by this study in red. Bootstrap support values under/beside nodes. Values below 70% not represented

Proseriata, is a monophyletic order [9]; based on morphological traits Sopott-Ehlers [10] divided it into Unguiphora (taxa without cirrus and a statocyst but with pigment in the mantle cells of rhabdomeric receptors) and Lithophora (statocyst present and no pigment in the mantle cells). Later the monophyly of both clades was confirmed [9, 11]), though with some dispute in the Unguiphora as it was underrepresented in the studies supporting its monophyly and appears to be paraphyletic in others [11, 12]). Within Proseriata, the position of the genus Ciliopharyngiella Ax, 1952 has been recently debated. Curini-Galletti et al. [11] suggested that it belongs inside Proseriata clustered with Unguiphora, while Van Steenkiste & Leander [13] suggested Ciliopharyngiella being the sister lineage of Proseriata.

Rhabdocoela is a monophyletic order [9] of free-living Platyhelminthes, traditionally subdivided into two major groups: Kalyptorhynchia and Dalytyphloplanida, with and without an anterior proboscis, respectively. Within Rhabdocoela, the family Mariplanellidae was initially included in Dalytyphloplanida though not fully fitting with the diagnostic characters. Recently, Steenkiste & Leander [13] re-classified the subfamily Mariplanellinae to Mariplanellida status novus, representing the monophyletic group sister to a large clade comprising Kalyptorhynchia and Dalytyphloplanida. According to this classification, Mariplanellidae (the only family of Mariplanellida) currently comprises three genera: Mariplanella Ax & Heller, 1970, Lonchoplanella Ehlers, 1974, and Poseidoplanella Willems et al., 2005. Morphologically, these genera share the characters single ovary and a double connection in the female reproductive system. However, the phylogenetic analyses by Van Steenkiste and Leander [13] only included two species of Mariplanella, M. piscadera Van Steenkiste & Leander, 2022 and M. frisia Ax & Heller, 1970, while members of Lonchoplanella and Poseidoplanella have not been included in any molecular phylogenetic analysis up to date.

Results

Trees obtained from independent Maximum Likelihood (ML) analyses of single markers (18S, 28S), as well as the one obtained from the analyses of the concatenated data matrix (18S + 28S) showed congruent results (SFig. 2, SFig. 3 and Fig. 1, respectively). The concatenated data matrix (18S + 28S) run through Bayesian Inference (BI) yielded highly congruent results (SFig. 4) with practically the same tree topology and high support values. The trees were rooted in Ciliopharyngiella constricta Martens & Schockaert, 1981, excepting the tree of 28S which was rooted in Proseriata.

In the ML concatenated analysis (Fig. 1), the taxon Proseriata is not well supported (68 B, where B represents bootstrap support values). Within Proseriata, Unguiphora (excepting Nematoplana sp.) and Lithophora are monophyletic and well supported (99B and 80B, respectively). Within Unguiphora, both Nematoplana coelogynoporoides Meixner, 1938 are in a monophyletic group; however, Nematoplana sp. is sister to Lithophora. The genus Polystyphora Ax, 1958, represented by three species, is monophyletic. Lithophora is subdivided into Otoplanidae and Monocelididae (100 B), and Coelogynoporidae and Calviriidae (100 B), respectively, as sister groups.

Rhabdocoela is well supported (100 B). Within Rhabdocoela, the three clades, Mariplanellida, Dalytyphloplanida and Kalyptorhynchia, are strongly supported (100 B). Mariplanellida is sister to Dalytyphloplanida and Kalyptorhynchia, which are joined in a not highly supported clade (69 B). In the concatenated analysis (Fig. 1), the species Lonchoplanella axi is located within Mariplanellida, sister to M. axi (86 B), showing the genus Mariplanella, as currently delineated, as paraphyletic. The 18S tree (SFig. 2) shows Mariplanella as monophyletic but is not supported (61 B) with Lonchoplanella axi as its sister group. Within Dalytyphloplanida, Neodalyellida, Limnotyphloplanida (excepting Haloplanella longatuba Ax & Heller, 1970) and Thalassotyphloplanida are monophyletic and, excepting the latter, well supported (100 B, 100 B, 96 B, and 84 B, respectively). Limnotyphloplanida and Thalassotyphloplanida are sister to each other (100B). The species Haloplanella longatuba is located within Thalassotyphloplanida, out of the Limnotyphloplanida.

Within Kalyptorhynchia, Eukalyptorhynchia is paraphyletic, including Schizorhynchia. The genus Toia Markus, 1952 is located as the sister group of a large clade (100 B) containing the rest of Eukalyptorhynchia (83 B) and Schizorhynchia, which is not supported (48B). These two latter groups are better supported in the BI analisis (SFig. 4) with a posterior probability value of 1. The family Cheliplanidae appears to be paraphyletic (100 B) with Cheliplanilla caudata Meixner 1938 being the sister taxon of a group compromised by Carcharodorhynchus Meixner, 1938 and Schizorhynchidae, represented by more than one species, and also well supported (100 B). The genera Carchadorhynchus, Schizochilus Boaden, 1963 and Proschizorhynchus Meixner, 1928, and Psammorhynchus tubulipenis Meixner, 1938, including more than one terminal, are respectively monophyletic, though herein represented by few species.

Discussion

Our results show Mariplanellida as an independent group within Rhabdocoela as found by Van Steenkiste & Leander [13], and previously suggested by other molecular analyses [9]. Its higher-level status as an additional infraorder proposed by Van Steenkiste & Leander [13] within Rhabdocoela is supported herein as well. Our study included Lonchoplanella axi, a genus and species not incorporated before in a molecular phylogeny. Lonchoplanella axi shares a number of conspicuous characters with Mariplanella frisia, including two types of adenal rhabdites (needle-shaped and elongate viscous) and a muscular copulatory bursa with a sclerotized basal membrane [13] and had been described as a member of Mariplanellidae [14]. The topology of the concatenated tree confirms Lonchoplanella axi as belonging to the Mariplanellidae family and Mariplanellida infraorder. More information (terminals and markers) is needed to clarify the monophyly of both genera, and further analyses including Poseidoplanella halleti Willems et al., 2005 (presumably the only missing taxon of Mariplanellidae) will clarify the relationships between the three genera.

Our study included six new terminals and five species of Kalyptorhynchia in addition to those included by Van Steenkiste & Leander [13]. Results corroborate Eukalyptorhynchia as non-monophyletic supporting Willems et al. [9] and Tessens et al. [15] results. Our results also support the position of the genus Toia separate from Cicerinidae, as suggested by Tessens, et al. [15]. The rest of Cicerinidae are monophyletic and the sister group of Schizorhynchia. Cicerinidae (except Toia) were located in a polytomy with Schizorhynchia and most members of Eukalyptorhynchia in the analysis of Van Steenkiste & Leander [13].

Most of the species, genera and families included in this study with more than one terminal are monophyletic and well supported. However, the terminals Nematoplana sp. and Haloplanella longatuba need further consideration. Within Proseriata, Nematoplana sp. (downloaded from GenBank) is not closely related to N. coelogynoporoides; its sister relationship to Lithophora might indicate that this terminal could have been misidentified or the result of contamination in the 18S sequence. Additionally, the length of branches of Nematoplana coelogynoporoides from Sylt (included herein) and N. coelogynoporoides from Roscoff (downloaded from GenBank) might indicate they are not be the same species.

In Rhabdocoela, Haloplanella longatuba is one of the species we incorporate in a phylogenetic analysis for the first time with molecular information. The relationships of brackish and marine water Typhloplanidae species, such as Haloplanella longatuba have been previously discussed [6, 16, 17]. Rieger [16] describes the resemblance between certain genera in the family Typhloplanidae (Limnotyphloplanida parvorder) and various genera in the Thalassotyphloplanida parvorder, thus encouraging future reorganizations within these taxonomic groups. Hochberg & Cannon [17] remarks on the presence of an unusual character in some genera of the Typhloplanidae family, such as Haloplanella Luther, 1946 and Pratoplana Ax 1960, where a stylet is present in the copulatory apparatus instead of a cirrus, which is one of the family's ground pattern traits. Moreover, several similarities between this species and members of Thalassotyphloplanida have been found, such as the female genital canal and the proboscis structure. Van Steenkiste et al. [6] already suggested that several brackish water and marine Typhloplanidae taxa might be closely related to Byrsophlebidae (Thalassotyphloplanida). Our results show Haloplanella longatuba nested within Thalassotyphloplanida, which supports that the position of this taxon has to be taken into further consideration, and revised in future studies.

In this study, the selected gene markers, 18S and 28S were used because they are already available for a large number of species and have been previously found useful to discern phylogenetic relationships within Proseriata and Rhabdocoela [13]. Nevertheless, Next Generation Sequencing (NGS) techniques will be undoubtedly useful to discern and clarify the evolution of these groups.

Conclusions

Our results support the infraorder Mariplanellida as an independent group within Rhabdocoela and confirms, by the first time with molecular data, that Lonchoplanella axi belongs to Mariplanellida. Eukalyptorhynchia is paraphyletic including members of Schizorhynchia, and the genus Toia separate from Cicerinidae. Haloplanella longatuba is herein nested within Thalassotyphloplanida, and not in Typhloplanidae (Limnotyphloplanida), which suggests that further studies are needed to clarify its phylogenetic relationships.

More terminals and information from morphological studies, as well as new markers of NGS techniques may clarify the gaps and the still doubtful relationships.

Material and methods

In this study we aim to provide a more robust phylogenetic hypothesis of Rhabdocoela relationships. For this purpose, we introduced the species Lonchoplanella axi to phylogenetic analyses to test whether or not it belongs to Mariplanellida. With respect to Kalyptorhynchia and Dalytyphloplanida we added species of Cystiplana Karling, 1964, Haloplanella, Marirhynchus Schilke, 1970, Lonchoplanella, and Schizorhynchoides Meixner, 1928 (Table 1).

Table 1 Terminals included, genes and GenBank Accession numbers. Species provided by this study are represented in bold

Sampling and species identification

A total of 37 samples were taken during two days sampling from intertidal sand in the island of Sylt, Germany. Samples were obtained by digging on the substrate with a 10 cm long shovel. The substrate was kept in zip bags and stored in fridges at 4ºC.

Sampling sites were the beach besides List Harbour (55.015337N, 8.435999E) and the beach in front of Alfred-Wegener-Institute building (55.023745, 8.439049), always during low tide. The collected sediment samples were all coarse sand enriched with variable amounts of organic material. Meiofauna was separated from the sediment using the MgCl₂ decantation method [18].

Flatworms were morphologically identified under Leica S APO stereomicroscope and Leica DM 2500 microscope and photographed (stylets) under a portable Leica MC 190 HD attached camera (Fig. 2).

Fig. 2

Close−up picture of the copulatory organs, stylets belonging to the identified species from Sylt. Pharynx rosulatus (p), stylet (st), testis (t), vesicula seminaris (sv), prostate vesicle (pv), genital atrium (ga), glandular organ (gl), copulatory organ (co). A. Marirhynchus longasaeta; B. Mariplanella frisia; C. Carcharodorhynchus listensis; D. Proschizorhynchus gullmarensis; E. Psammorhynchus tubulipenis; F. Cystiplana paradoxa; G. Schizorhynchoides aculeatus; H. Schizochilus caecus; I. Haloplanella longatuba. J. Lonchoplanella axi

DNA extraction, amplification and sequencing

For the DNA extraction the DNeasy® Blood & Tissue Kit (QIAGEN) was used. Manufacturer’s instructions were followed, with the exception that DNA was eluted in 60µL of preheated AE elution buffer (60 °C). For samples with low concentration, this protocol was followed by the Amplification of purified genomic DNA protocol from QIAGEN REPLI-g® kit. Thermocycling conditions and primers from Van Steenkiste & Leander [13] were used to sequence markers 18S and 28S (see supplementary material, S1). Sequencing was carried out by Eurofins Genomics (Konstanz, Germany). All new sequences were deposited in GenBank, and sequence accession numbers are provided in Table 1.

Phylogenetic analyses

Once the sequences were obtained, they were blasted using blastn through ncbi-blast + v.2.12.0 to confirm that platyhelminthes’ DNA was amplified during the PCRs. The rest of the sequences were obtained from GenBank attempting to gather a broad representation of the different families and subfamilies. Those terminals for which both markers (18S and 28S) were available were selected for this study. Several terminals of Proseriata, including Ciliopharyngiella constricta were also included to root the tree.

Sequences were visually checked in Geneious v10.2.3 and aligned using MAFFT v.7.305b [19] using the iterative refinement method E-INSI. Single genes (18S, 28S) were concatenated using FASconCAT-G [20, 21]. The maximum likelihood (ML) analysis of the single markers, as well as the concatenated matrix was performed through IQtree v.1.3.11.1 [22, 23], with best fitting models selected by Modelfinder [24] (18S: GTR + F + I + G4; 28S: GTR + F + I + G4). In all analyses, each partition was allowed to have its own set of branch lengths. (-spp option). Support values were estimated based on 1000 bootstrap pseudo replicates (B). iTol v.6. and Adobe Illustrator (2020) were used to edit the phylogenetic trees. The concatenated matrix (18S + 28S) was also analysed through Bayesian inference (BI). For BI analyses, two independent runs of 1,342,000 generations and four chains, each (one cold, three heated) were run in MrBayes 3.2.7 [25]. The most similar models available in MrBayes (-mset option) to those selected by Modelfinder for each partition were applied. All parameters were unlinked, rates were allowed to vary freely over partitions andtrees were sampled every 1000 generations. The runs were stopped when the standard deviation reached the value of 0,007. After discarding 25% first trees as burn-in, trees from the stationary phase were combined to obtain a majority rule consensus and posterior node probabilities [26].