Phylogenetic position of the enigmatic genus Atherospio and description of Atherospio aestuarii sp. nov. (Annelida: Spionidae) from Japan

Introduction

Atherospio Mackie & Duff, 1986 is a small genus in the family Spionidae Grube, 1850 that currently consists of two species: A. disticha Mackie & Duff, 1986 and A. guillei (Laubier & Ramos, 1974). The genus is closely related to Pygospiopsis Blake, 1983 (including the recently synonymized genus Pseudatherospio Lovell, 1994: Blake & Maciolek, 2018) as it has similar prostomial shapes, an occipital antenna, modified anterior neurochaetae, branchiae that are either basally or entirely fused to the notopodial lamellae, and unusual bidentate neuropodial hooks. In other spionids, the small tooth (teeth) of the neuropodial hooded hooks is (are) superior to the main fang on the convex side, while for Atherospio and Pygospiopsis the neuropodial hooded or unhooded hooks have a small tooth or knob on the concave side, which is subapical to the terminal shaft or main fang (Blake & Maciolek, 2018). Atherospio, Pygospiopsis, and a recently established genus, Aciculaspio Blake & Ramey-Balci, 2020, are collectively called the Pygospiopsis-Atherospio group (Blake & Ramey-Balci, 2020) and currently consist of nine species. Atherospio and Pygospiopsis are distinguishable as the former have their first branchiae on chaetiger 7, while the latter simple or partially fused branchiae anterior to chaetiger 7 in a variety of patterns (Blake & Maciolek, 2018). Aciculaspio differs from both Atherospio and Pygospiopsis as it has enlarged flattened branchiae fused to the dorsal lamellae from chaetiger 10 instead of 7 and simple, unidentate-hooded hooks with curved and pointed fangs (Blake & Ramey-Balci, 2020).

Blake, Maciolek & Meißner (2020) divided the spionid genera into four clades following Blake & Arnofsky (1999) and Blake (2006): (1) Subfamily Nerininae Söderström, 1920; (2) Subfamily Spioninae Söderström, 1920; (3) Clade consisting of Pygospiopsis, Atherospio, and Pseudatherospio (= Pygospiopsis-Atherospio group); and (4) five monotypic genera with no strong affinity for other spionids (Glandulospio Meißner et al., 2014; Glyphochaeta Bick, 2005; Spiogalea Aguirrezabalaga & Ceberio, 2005; Spiophanella Fauchald & Hancock, 1981; and Xandaros Maciolek, 1981). Species belonging to the Pygospiopsis-Atherospio group are superficially similar to species in subfamily Spioninae (including the Polydora complex and the genera Pygospio Claparède, 1863, Microspio Mesnil, 1896, and Spio Fabricius, 1785), some of which were originally classified as separate genera within Spioninae. Pygospiopsis dubia (Monro, 1930) was originally described as Pygospio, and Blake (1983) later established the genus Pygospiopsis for this species. Atherospio guillei was originally described as Polydora Bosc, 1802 in the Polydora complex, and later, Meißner & Bick (2005) transferred this species to Atherospio. Atherospio guillei and the species belonging to the Polydora complex both have heavy spines in the fifth segment. However, this is not considered to be evidence of a close relationship between the two taxa as these heavy spines are not homologous sensu stricto, as in A. guillei they are neuropodial, while in Polydora complex they are notopodial (Mackie & Duff, 1986; Radashevsky & Fauchald, 2000; Radashevsky, 2012).

The close relationship between the Pygospiopsis-Atherospio group and the subfamilies Spioninae and Nerininae has not been consistently supported in previous studies. The first phylogenetic analysis of the Spionidae genera using morphology by Sigvaldadóttir, Mackie & Pleijel (1997) indicated that there were four clades in the family: (1) Aonidella López-Jamar, 1989 and Xandaros; (2) Prionospio complex, Laonice Malmgren, 1867, Spiophanes Grube, 1860, and Aonides Claparède, 1864; (3) a large unresolved assemblage of genera including the Polydora complex, Scolelepis Blainville, 1828, Malacoceros Quatrefages, 1843, and Spio; and (4) Atherospio, Pseudatherospio, and Pygospiopsis, but the support for these clades was weak and the selection of outgroups was subsequently deemed unfortunate (Blake, Maciolek & Meißner, 2020). Mackie (1996) re-examined the intergeneric relationships within Spionidae examined by Sigvaldadóttir, Mackie & Pleijel (1997) by adding several new taxa and eight taxa with questionable generic attribution using the same outgroups and indicated generally consistency with the previous results, but separated the third large unresolved assemblage by Sigvaldadóttir, Mackie & Pleijel (1997) into a ‘polydorid’ group (including Pygospio muscularis Ward, 1981 and excluding Tripolydora Woodwick, 1964 and Pseudopolydora primigenia Blake, 1983) and a large group including 11 genera. He also showed that the Pygospiopsis-Atherospio group, which includes two unnamed groups provisionally assigned as ‘Genus A’ and ‘Genus B’, is sister to a clade consisting of Pseudopolydora primigenia and Pygospio elegans Claparède, 1863. The third phylogenetic analysis of the spionid genera using morphological, reproductive, and developmental characteristics from Blake & Arnofsky (1999) indicated that there were three clades: two major clades consisting of the subfamily Spioninae and a larger clade consisting of all remaining spionid genera and the genera Heterospio Ehlers, 1874 (now considered to be a taxon closely related to cirratuliform polychaetes rather than spioniforms: Blake & Maciolek, 2019), Poecilochaetus Claparède in Ehlers, 1875, Trochochaeta Levinsen, 1884, and Uncispio Green, 1982, and a third, minor clade consisting of the enigmatic genus Pygospiopsis (including Atherospio). At present, because of several unusual morphological characteristics of the Pygospiopsis-Atherospio group, its taxonomic relationship with other spionids or spioniforms is not well understood. However, Blake & Maciolek (2018) noted that the large recurved hooded hooks of P. profunda Blake & Maciolek, 2018 have similarities with the giant modified neuropodial hooks or spines of some Uncispio species.

To date, there are no available molecular data on the Pygospiopsis-Atherospio group or Uncispio deposited in the DNA Data Bank of Japan (DDBJ), the European Nucleotide Archive (ENA), or GenBank databases. Therefore, these taxa were not included in the first and recent comprehensive molecular phylogenetic analyses of the spionid genera, which was conducted by Abe & Sato-Okoshi (2021) and Wang et al. (2022), respectively. Our field surveys have identified several specimens of the genus Atherospio, which have never been recorded from Japan before, from several study sites. In this study, we report the morphology of the specimens and compare it with that of other species of the genus, and describe a new species, Atherospio aestuarii sp. nov. We also evaluate the phylogenetic position of Atherospio by conducting the first molecular phylogenetic analysis including the genus, whose phylogenetic position has remained a question until now.

Materials and Methods

Specimen collection

Specimens of the Atherospio species were collected from bottom sediments in the intertidal zone of Otomo-ura (38.9958 N, 141.6817 E), Hirota Bay, Iwate Prefecture on August 6, 2017, August 18, 2018, and August 4, 2020; subtidal zones <1 m in depth in a nameless small inlet of Ago Bay (34.2985 N, 136.8311 E), Mie Prefecture on October 8, 2021; and a small fishing port at the mouth of the Kurio River (30.2741 N, 130.4214 E) on Yakushima Island, Kagoshima Prefecture, Japan on November 6, 2021 (Figs. 1 and 2). The water areas where the specimens were collected in this study are not protected, and no permission of any kind is required to collect the organisms. In the field survey of this study, we did not collect any commercially marine species and did not use any collection method that violated the prefectural fishery regulation, so we did not need any permission for the survey.

Results

Systematics

Family Spionidae Grube, 1850

Genus Atherospio Mackie & Duff, 1986

Type-species: Atherospio disticha Mackie & Duff, 1986

Diagnosis (Emended from Meißner & Bick, 2005). Prostomium deeply incised, longer than wide, posteriorly tapered and extended into a short caruncle; occipital antenna present or absent or minute process at the position of this antenna present. Nuchal organs small or indistinct. Dorsal branchiae from chaetiger 7; branchiae with distal digitiform process, outer branchial margin completely fused with notopodial postchaetal lamella. Parapodia biramous with well developed postchaetal lamellae and alimbate mostly hirsute capillaries in noto- and neuropodia. Chaetigers 4 and 5 or only chaetiger 5 with modified chaetae in the neuropodium being falcate and pointed or aristate spines, modified chaetae in regular or irregular short vertical rows superior to several capillary chaetae. Neuropodial hooks alongside capillaries; hooks uni-or bidentate, secondary tooth below main fang; hook distally with closely applied sheath. Notopodial hooks absent. Posterior spine-like notochaetae present or absent. Sabre chaetae absent but several capillaries in inferiormost position throughout the body. Genital pouches absent. Pygidium surrounded by several pairs of lateral cirri.

Remarks. The morphology of the new species described below is generally consistent with the diagnosis for the genus Atherospio by Meißner & Bick (2005). Since the description of “Dorsal branchiae on chaetiger 7 and following 4–6 chaetigers” and “Postbranchial neuropodial hooks” in the diagnosis provided by Meißner & Bick (2005) does not apply to the new species, we emended these parts of the diagnosis as “Dorsal branchiae from chaetiger 7” and “Neuropodial hooks”, respectively, to include the new species. Following the definition of caruncle by Wong et al. (2014), “a dorsal extension of the prostomium, taking the form of an elevation or a distinct crest separating the nuchal organs one from another”, the posterior tapered and elongated part of the prostomium present in the known Atherospio species is regarded as a short caruncle. Therefore, we modified the posterior part of prostomium, which is considered “not extended into a distinct caruncle” in the diagnosis by Meißner & Bick (2005), to be a short caruncle. We also emended the arrangement of the modified spines from “a irregular short row” to “regular or irregular short vertical rows” since the modified spines are in very regular vertical rows in A. disticha and A. aestuarii sp. nov. and two forms of spines gathered in irregular, short rows in A. guillei. We added to the diagnosis that posterior needle-like notochaetae are found in A. guillei and the new species.

Atherospio aestuarii sp. nov.

Japanese name: Irie-nogi-supio

LSID. urn:lsid:zoobank.org:act:287692C4-C105-41BC-8718-37C6BBE10B7C

(Figs. 3 and 4)

Type material. Holotype: NSMT-Pol H-858, small fishing port at the mouth of the Kurio River, 30.2741 N, 130.4214 E, Yakushima Island, Kagoshima Prefecture, subtidal, <1 m depth, muddy sand, November 6, 2021 (complete specimen). Paratypes: NSMT-Pol P-859, Otomo-ura, 38.9958 N, 141.6817 E, Hirota Bay, Iwate Prefecture, intertidal, gravelly muddy sand, August 6, 2017 (incomplete 1 specimen); NSMT-Pol P-860, Otomo-ura, 38.9958 N, 141.6817 E, Hirota Bay, Iwate Prefecture, intertidal, gravelly muddy sand, August 18, 2019 (incomplete 1 specimen); NSMT-Pol P-861, Otomo-ura, Hirota Bay, 38.9958 N, 141.6817 E, Iwate Prefecture, intertidal, gravelly muddy sand, August 4, 2020 (incomplete 1 specimen); NSMT-Pol P-862 (incomplete 1 specimen), NSMT-Pol P-863 (incomplete 7 specimens), NSMT-Pol P-864 (incomplete 1 specimen), nameless small inlet in Ago Bay, 34.2985 N, 136.8311 E, Mie Prefecture, subtidal, <1 m depth, gravelly muddy sand, October 8, 2021; NSMT-Pol P-865 (incomplete 1 specimen), NSMT-Pol P-866 (incomplete 2 specimens), small fishing port at the mouth of the Kurio River, 30.2741 N, 130.4214 E, Yakushima Island, Kagoshima Prefecture, subtidal, <1 m depth, muddy sand, November 6, 2021.

Description. Holotype complete (pygidium damaged) with 64 chaetigers, measuring 9.5 mm long and 1.2 mm wide at chaetiger 5 (Fig. 3); paratypes incomplete up to 14.4 mm long, 1.5 mm wide for 40 chaetigers. Body wide, dorsoventrally flattened for first six chaetigers (Figs. 3B and 4A), then gradually narrower and becoming cylindrical in cross-section. Body white to light tan in preserved specimen (Fig. 3A), translucent white to light tan when alive with red blood vessels and pale orange to brown digestive tract internally (Figs. 3C, 3E and 4A); body and palp pigmentation absent.

Prostomium longer than wide, anteriorly incised; extends posteriorly as caruncle to middle of chaetiger 1 (Figs. 3B and 4A). Eyes dark red, two pairs arranged in trapezoidal shape, lateral pair situated anteriorly, kidney-shaped, larger than medial ones (Figs. 4A and 4B). Occipital antenna absent. Nuchal organs U-shaped with outward curving posterior part, located just behind prostomium and between notopodial lamellae of chaetiger 1 (Figs. 3B and 4B). Palps arising lateral to prostomium (Fig. 4A). Peristomium extending lateral to prostomium, forming upper lip of mouth and extending ventrally forming ventral lip of mouth; thick everted proboscis or pharynx present; oral lips relatively smooth; peristomial papillae (see Blake & Maciolek, 2018) absent.

Chaetigers 1–6 abranchiate (Figs. 3B, 3C and 4A). Notopodial postchaetal lamellae long, digitiform or lanceolate on chaetiger 1 (Figs. 3B and 4C), broader on chaetiger 2, and becoming broad triangular or oval on chaetigers 3–6 (Fig. 4E). Neuropodial lamellae digitiform or lanceolate on chaetiger 1, broad triangular on chaetiger 2, and oval to triangular on chaetigers 3–6. Chaetiger 5 of same size as neighboring chaetigers. Midventral series of white rectangular pads in anterior chaetigers, indistinct in fixed specimens.

Branchiae from chaetiger 7 to 18–23 (File S2), long and cirriform, with digitiform process at distal end (Fig. 4F); overlapping mid-dorsal or not, full-sized from chaetigers 10–12; fully fused with notopodial postchaetal lamellae in outer margin (Fig. 4F); ciliation along inner margin, extending to a transverse ciliated band across the whole width of the chaetiger. In branchial chaetigers, notopodial postchaetal lamellae foliated and often wavy, especially when alive (Figs. 3C and 3E); neuropodial postchaetal lamellae rounded, larger dorsoventrally than that of chaetigers 1–6 (Fig. 4F). In postbranchial chaetigers, both postchaetal lamellae smaller, rather more subtriangular.

Notochaetae in most chaetigers long slender capillaries without limbations; some posterior notopodia with bundles of needle-like capillaries raised dorsally (Fig. 3F); notopodial hooks absent. Neurochaetal capillaries without limbations in anterior chaetigers. Neuropodia of chaetiger 5 double vertical rows of aristate spines dorsal to small bundle of capillaries; spines in posterior row slightly thicker than those of closely applied anterior row, taper steeply towards tip with short aristae part; spines in the anterior row taper gradually towards tip with long aristae part (Figs. 4D and 4E). Hooded hooks in neuropodia from chaetigers 16–19 (File S2) to the posterior-most chaetiger, accompanied by capillaries in all chaetigers; numbering up to 6 in a series, reduced in posterior chaetigers, shafts S-curved, hooks bidentate with secondary tooth on concave side at right angle to and below main fang (Fig. 4G). Neuropodial sabre chaetae absent.

Pygidium without anal cirri probably due to damage.

Methyl green staining. Anterior half of the prostomium deeply stained (Fig. 3B). Peristomium stained with vertical stripes (Figs. 3D and 4C). Tips of some post-chaetal lamellae deeply stained. Chaetigers 1–6 diffusely stained with scattered deeply stained cells on both dorsal and ventral sides; chaetiger 7 onward more strongly stained than chaetigers 1–6 on dorsal ventral, and lateral sides (Figs. 3B, 3D and 4C). Unstained ventral large white spots, one pair per chaetiger, present from chaetiger 2 to posterior middle-body chaetigers (Fig. 4C).

Remarks. Atherospio aestuarii sp. nov. closely resembles A. disticha and A. guillei and is intermediate in morphology between these species. Atherospio aestuarii sp. nov. is similar to A. disticha and differs from A. guillei in having branchiae fused to the notopodial lamellae on a restricted number of segments from chaetiger 7, modified neurochaetae on chaetiger 5, and at least some bidentate neuropodial hooks with the secondary tooth below the main fang (Table 2). The form and arrangement of the modified aristate neurochaetae in double vertical rows closely resemble those found on chaetigers 4 and 5 of A. disticha. The new species lacks the occipital antenna present in A. disticha. In this respect it resembles A. guillei, however, that species differs in having robust neuropodial spines on chaetiger 5 and peristomial papillae (see Meißner & Bick, 2005: fig. 2C), and a preponderance of unidentate neurochaetae. Both A. guillei and the new species have slender needle-like notochaetae in their posteriormost chaetigers. Atherospio aestuarii sp. nov. is distinguished from both congeneric species by its branchial and neuropodial hook distributions; as the last branchial chaetiger and the first chaetiger with neuropodial hook are more posterior in the former species. The other two nominal Atherospio species were collected from ≥27 m depths in the subtidal zone (Table 2), whereas the new species was unique in that it was recorded at relatively shallow depths, which included intertidal zones.

Mackie, Oliver & Rees (1995) and Mackie & Garwood (1995) reported two provisionally unnamed spionid taxa closely related to A. disticha from Cardigan Bay in the Irish Sea as ‘Spionidae gen. A’ and ‘Spionidae gen B’ and mentioned that ‘Spionidae gen. B’ is morphologically similar to A. guillei (as Polydora). Several Atherospio related taxa collected from Europe and Hong Kong including ‘Spionidae gen. A’ and ‘Spionidae gen B’ were referred as ‘Genus A’ and ‘Genus B’ (include A. guillei, but may also involve two separate taxa) in Mackie (1996). In his character matrices which provided the main characteristics of the morphology of these two groups (Mackie, 1996: Tables 2 and 3), ‘Genus A’ and ‘Genus B’ are distinguished by the former lacking and the latter having posterior modified notochaetae and the former having one type of anterior modified neurochaetae but the latter having two types. Atherospio aestuarii sp. nov. does not fall into either group because it has posterior needle-like notochaetae and one type of anterior modified neurochaetae.

Etymology. The specific name aestuarii is from the Latin word aestuarium, which means the estuary, inlet, and intertidal zone, thus referring to the habitat of this species.

Habitat. Muddy and gravelly muddy sand sediment in the intertidal to subtidal zone, <1 m in depth.

Distribution. Currently identified in Otomo-ura, Hirota Bay (Iwate Prefecture), Ago Bay (Mie Prefecture), and Yakushima Island (Kagoshima Prefecture), Japan.

Molecular phylogeny

The intraspecific p-distances in the 18S, 28S, and 16S rRNA gene sequences of the seven A. aestuarii sp. nov. specimens were 0%, 0–0.26%, and 0–2.20%, respectively. In the molecular phylogenetic analyses based on the concatenated sequences, the Polydora complex + Pygospio, subfamily Spioninae sensu Blake, Maciolek & Meißner (2020) + Glandulospio, and that plus Marenzelleria Mesnil, 1896, Rhynchospio Hartman, 1936, Atherospio, Dispio Hartman, 1951, Scolelepis, and Malacoceros were recovered as clades with robust statistical support (SH-aLRT ≥ 80%, aBayes ≥ 0.95, UFBoot ≥ 95%) (Fig. 5). Atherospio aestuarii sp. nov. formed a clade with Dispio and Scolelepis, however, the support value for the node was not robust (SH-aLRT = 76.3, aBayes = 0.99, UFBoot = 54).

Discussion

In contrast to the results of phylogenetic analyses by Sigvaldadóttir, Mackie & Pleijel (1997) and Blake & Arnofsky (1999), our molecular phylogenetic analysis indicated that Atherospio aestuarii sp. nov. did not form a clade distinct from the subfamilies Spioninae and Nerininae, but rather could be included within a clade that included the genera of the subfamily Spioninae sensu Blake, Maciolek & Meißner (2020) plus Glandulospio, Marenzelleria, Rhynchospio, Scolelepis, Dispio, and Malacoceros (Fig. 5). This clade corresponds to that referred to as the subfamily Spioninae in the alternative classification of the subfamily suggested by Wang et al. (2022) based on the results of molecular phylogenetic analysis. Monophyly of Spioninae sensu Wang et al. (2022) was supported also by Abe & Sato-Okoshi (2021) and the present study. However, the alternative subfamily classification suggested by Wang et al. (2022) has the following problems: (1) Nerininae sensu Wang et al. (2022) has been recovered as either monophyletic with low support (Wang et al., 2022) or as paraphyletic (Abe & Sato-Okoshi, 2021, this study) and (2) if Nerininae does not include Scolelepis, then this subfamily is not valid because the type-genus is Nerine which is a junior synonym of Scolelepis. The paraphyly of Nerininae sensu Blake, Maciolek & Meißner (2020) is also clearly indicated by the previous (Abe & Sato-Okoshi, 2021; Wang et al., 2022) and the present study. Our understanding of the phylogenetic relationships among the spionid genera is still in a precocious phase, and therefore the subfamily classification of the Spionidae should be revisited with more comprehensive and robust molecular phylogenetic tree as well as non-molecular data such as morphology, development, and reproduction about representatives of many genera. Nevertheless, our molecular phylogenetic analysis supports previous recognitions by Mackie & Duff (1986), Radashevsky & Fauchald (2000), and Radashevsky (2012) which indicate that the members belonging to the Pygospiopsis-Atherospio group are not closely related to the superficially similar taxa, that is, Polydora and Pygospio, and that the heavy spines in the fifth segments of Polydora and Atherospio are not homologous sensu stricto. The possibility of a close relationship between Atherospio and Dispio/Scolelepis is worth further investigation through molecular phylogenetic analysis with the addition of potential closely related taxa such as Australospio Blake & Kudenov, 1978 (Sigvaldadóttir, Mackie & Pleijel, 1997), Lindaspio Blake & Maciolek, 1992 (Mackie, 1996), and Pygospiopsis, as the statistical support for the clade was not robust in the present study.

Supplemental Information