Mitogenomics reveals phylogenetic relationships of caudofoveate aplacophoran molluscs

Highlights

• Six new mitochondrial genomes of aplacophoran molluscs are presented.

• Order of protein coding genes is conserved but tRNA order is unique among lineages.

• Gene order in Aculifera corresponds to the ancestral gene order in Mollusca.

• Mitochondrial genomes are useful molecular markers for aculiferan phylogenetics.

Abstract

The worm-shaped, shell-less aplacophoran molluscs Caudofoveata and Solenogastres have recently received attention as part of the clade Aculifera, but relationships within these two lineages are still largely unknown. Here, we use complete mitochondrial genomes to shed light on higher-level relationships within Caudofoveata. Mitochondrial genomes have been sequenced for many diverse molluscs, but only two mitochondrial genomes from aplacophoran molluscs (the caudofoveates Scutopus ventrolineatus and Chaetoderma nitidulum) are available to date. We sequenced and assembled complete or near complete mitochondrial genomes of five additional species of Caudofoveata (Falcidens acutargatus, Falcidens halanychi, Scutopus robustus, Psilodens balduri and Spathoderma clenchi) and one species of Solenogastres (Neomenia carinata) for comparison to available mitochondrial genomes of aculiferans. Comparison of mitochondrial gene order among different lineages revealed a highly conserved order of protein coding genes corresponding to the hypothesized ancestral gene order for Mollusca. Unique arrangements of tRNAs were found among lineages of aculiferan molluscs as well as among caudofoveate taxa. Phylogenetic analyses of amino acid sequences for the 13 protein-coding genes recovered a monophyletic Aplacophora. Within Caudofoveata, Chaetodermatidae, but not Limifossoridae, was recovered monophyletic. Taken together, our results suggest that mitochondrial genomes can serve as useful molecular markers for aculiferan phylogenetics, especially for more recent phylogenetic events.

Introduction

The shell-less, worm-shaped Solenogastres (=Neomeniomorpha) and Caudofoveata (=Chaetodermomorpha) have recently received attention as part of Aculifera, a clade grouping aplacophoran molluscs with chitons (Polyplacophora) as sister to the remaining Mollusca (Conchifera). Aculifera was first proposed based on morphology (Scheltema, 1993). Subsequently, several recent molecular phylogenetic studies (Kocot et al., 2011, Smith et al., 2011, Vinther et al., 2012, Osca et al., 2014), fossil evidence (Vinther et al., 2012, Vinther et al., 2017, Sutton and Sigwart, 2012, Sutton et al., 2012) and larval development (Scherholz et al., 2013, Scherholz et al., 2015, Redl et al., 2016) have provided strong support for the Aculifera hypothesis (reviewed by Kocot, 2013, Schrödl and Stöger, 2014).

Despite an increasing number of studies on the phylogenetic position of aplacophorans within Mollusca, many questions about the internal relationships of both Caudofoveata and Solenogastres remain (Todt et al., 2008, Todt, 2013). Only one analysis has been performed to investigate relationships within Aplacophora to date (Salvini-Plawen, 2003), but results of this cladistic morphological analysis focused only on Solenogastres and produced trees inconsistent with current taxonomy. Evolutionary relationships within Caudofoveata have never been tested with phylogenetic analysis. Except for sequences from aplacophorans that have been used in studies of deep molluscan phylogeny, little molecular data have been available for aplacophoran molluscs. Resolving relationships is necessary to infer plesiomorphic character states for aplacophoran taxa, which will aid aplacophoran taxonomy and the understanding mollusc evolution as a whole.

Caudofoveata, with about 130 species in three recognized families, represents the less diverse aplacophoran group, whereas Solenogastres includes about 280 species in 23 families and four orders (García-Álvarez and Salvini-Plawen, 2007, Todt, 2013; Fig. 1). Traditionally, Caudofoveata has been separated into three families, Prochaetodermatidae (Salvini-Plawen, 1975), Chaetodermatidae (Ihering, 1876) and Limifossoridae (Salvini-Plawen, 1970). A fourth family, Scutopidae, was suggested by Ivanov (1981), but was later rejected (Salvini-Plawen, 1992, Saito and Salvini-Plawen, 2014). Caudofoveata are classified based on characters of the oral shied surrounding the mouth opening, body shape, and most importantly, radula morphology. Prochaetodermatidae and Limifossoridae have a serial, distichous radula; a radula bearing transverse rows of two mirror image teeth. Prochaetodermatidae is additionally characterized by the presence of jaws and a middle row of central plates between the teeth, both unique among Caudofoveata. In Chaetodermatidae the radula has been reduced to a single pair of teeth supported by an unpaired cone (Salvini-Plawen, 1975). Limifossoridae has been regarded as sister to all other Caudofoveata, based on characters presumed to be plesiomorphic: the serial distichous radula and a simple, cylindrical body shape with externally scarcely pronounced body regions (Salvini-Plawen, 1977). Additionally, the ventral line found in several species of Limifossoridae, has been interpreted as a vestige of a ventral furrow, homologous to the ventral foot groove in Solenogastres (Ivanov, 1986, Salvini-Plawen, 2003). A ventral line is however also found anteriorly in the chaetodermatid Falcidens sagittiferus (Ivanov et al., 2009) and in certain Prochaetodermatidae (Scheltema, 1985). Chaetodermatidae has been inferred to be the most derived taxon because of the reduced radula and complex midgut morphology (Salvini-Plawen L.v., , 1975, Scheltema, 1981). Relationships within Caudofoveata, both among recognized genera and families, have been debated (e.g. Salvini-Plawen L.v., , 1975, Ivanov, 1981, Ivanov, 1986, Scheltema, 1981) and are still unclear (Todt et al., 2008).

Mitochondrial genomes have been widely used for phylogenetics in metazoans, including other groups of molluscs (e.g., Bernt et al. 2013). Phylogenetic reconstruction of deep molluscan relationships based on mitochondrial genomes has been hampered by problems with heterogenous evolutionary rates leading to long-branch attraction artifacts (e.g., Stöger and Schrödl, 2013, Osca et al., 2014), even when these issues are accounted for (Osca et al., 2014, Stöger et al., 2016). However, analyses of mitochondrial genomes have proven useful for inferring more recent evolutionary relationships in molluscs, including gastropods (e.g., Grande et al., 2008, Osca et al., 2015, Uribe et al., 2016, Williams et al., 2014), bivalves (e.g., Doucet-Beaupré et al., 2010, Yuan et al., 2012), and cephalopods (e.g., Akasaki et al., 2006, Allcock et al., 2011, Yokobori et al., 2004, Yokobori et al., 2007). Within Aculifera, sequences of full mitochondrial genomes have been scarce. Prior to this study, only two mitochondrial genomes of aplacophoran molluscs were available in GenBank, Scutopus ventrolineatus (Osca et al., 2014) and an unpublished mitochondrial genome of Chaetoderma nitidulum (EF211990, Dreyer and Steiner, unpublished). After recent additions, mitochondrial genomes are now available for seven species of Polyplacophora (Boore and Brown, 1994, Irisarri et al., 2014, Veale et al., 2014, Guerra et al., 2018). An analysis of mitochondrial genes and evidence from mitochondrial genome arrangement provided support for the monophyly of Aculifera (Osca et al., 2014) and, within Polyplacophora, Irisarri et al. (2014) used mitochondrial genome data for phylogeny reconstruction, showing that gene order changes are potentially useful phylogenetic markers in Polyplacophora.

In order to investigate the evolution of aculiferan mitochondrial genome organization and relationships within Caudofoveata, we sequenced the mitochondrial genomes of five additional species representing all three recognized families of Caudofoveata as well as one species of Solenogastres. These data were combined with data available from GenBank to infer relationships among caudofoveate lineages, allowing assessment of the current taxonomy and understanding of morphology.