In present study, a previously unidentified but frequently encountered species of deep-sea protobranch, Yoldiella haimaensis sp. nov., is described new to science from the Haima Cold Seep on the northwestern slope of the South China Sea. A morphological analysis confirmed that this species belongs to a previously undescribed species of the genus Yoldiella A.E. Verrill & K.J. Bush, 1897. It differs morphologically from other known species within the genus in its shell shape, degree of inflation, beaks, and number of hinge teeth. Furthermore, we sequenced three gene segments of Y. haimaensis sp. nov., comprising a nuclear ribosomal gene (18S rRNA), a nuclear protein-coding gene (histone H3), and a mitochondrial gene (cytochrome c oxidase subunit I, COI). Our phylogenetic analysis performed on the superfamily Nuculanoidea and family Yoldiidae indicates that the genus Yoldiella is non-monophyletic, and the widely recognized families within the superfamily Nuculanoidea are also not monophyletic. Our results provide molecular insights into the Protobranchia and highlight the necessity for further samples and data to revise the classification of families and genera within the superfamily using an integrative approach that combines morphological analysis and molecular data.

Anatomy, chemosynthetic ecosystems, deep-sea, molecular analysis, morphology

Protobranchia, with a significant evolutionary history dating back to the Cambrian, represents an ancestral and basal group of Bivalvia. The protobranchs are primarily found in the subtidal zone, especially in the deep sea, and are generally deposit feeders that bury themselves in the soft sediment (Allen 1978). They have limited presence in the intertidal zone. So, it is difficult to collect specimens of this group (Xu 1999). The highly conserved and distinctive morphology and anatomy, including gill structure, hinge conformation, shell microstructure, as well as the pericalymma larval development, small in size and lifestyle of the group make it a difficult but fascinating taxon of Bivalvia (Zardus 2002). The simplicity in the form of protobranch bivalves veils the complexity of their phylogeny. The monophyly of Protobranchia has been discussed intensively and become a subject of controversy due to the extensive use of molecular methods (Smith et al. 2011; Sharma et al. 2012; Bieler et al. 2014; González et al. 2015; Combosch et al. 2017; Lemer et al. 2019). Phylogenetic analyses using four nuclear genes (Sharma et al. 2012) and an exemplar-based approach combining Sanger-based sequences and an extensive morphological data matrix (Bieler et al. 2014), as well as the phylogenomic analysis using genomes and transcriptomes (González et al. 2015), recovered the monophyly of Protobranchia. However, the subsequent analysis by Combosch et al. (2017), which utilized five genes and included more taxa, supported the polyphyly of Protobranchia. The latest research has divided Protobranchia into three orders and five superfamilies: Nuculanoidea and Sareptoidea in Nuculanida, Solemyoidea and Manzanelloidea in Solemyida, and Nuculoidea in the order Nuculida (Sharma et al. 2013; Sato et al. 2020). The reconstruction of the phylogeny has indicated that eight families (i.e. Nuculanidae, Bathyspinulidae, Malletiidae, Neilonellidae, Phaseolidae, Siliculidae, Tindariidae, and Yoldiidae) within the superfamily Nuculanoidea are all non-monophyletic (Sato et al. 2020). Shell microstructure of protobranchs play a crucial role in their classification at the superfamily level. Moreover, subtle differences in shell microstructure can aid in distinguishing similar species (Sato et al. 2020). Extensive research has revealed a multitude of unknown or cryptic protobranch species awaiting description and classification under the integrative taxonomy framework (e.g. Neulinger et al. 2006; Zardus et al. 2006). Additional samples and data are required to revise the families and genera within the superfamily Nuculanoidea through a combination of morphological diagnosis and molecular analysis.

Cold seeps are natural phenomena widely distributed across the globe. On the seafloor of these areas, the hydrocarbon-rich fluids and gases leak from cracks and enter the water column through sediment, forming a distinctive habitat (Dong et al. 2021). In March 2015, a newly active cold seep was discovered using the Haima remotely operated vehicle (ROV) on the northwestern slope of the South China Sea (SCS) (Zhao et al. 2020). Dong et al. (2021) documented 34 epibenthic macrofauna species collected from Haima Cold Seep with 24 species being identified. Yao et al. (2022) identified 12 macrobenthic species from five phyla and 12 families in the Haima Cold Seep, including two species first found in this location. Seven phyla, 14 classes, and 65 species were identified by He et al. (2023) in this cold seep. To date, more than 80 species of macrobenthic organisms have been collected from the Haima cold seeps (Wang et al. 2022).

This study presents the description of a new Yoldiella species, Y. haimaensis sp. nov., from the Haima Cold Seep in the SCS. This species had been identified as Malletia sp. or Yoldiella sp. (Dong et al. 2021; Ke et al. 2022; He et al. 2023). Additionally, we provide the sequences of three gene segments of the new species, including a nuclear ribosomal gene (18S rRNA), a nuclear protein-coding gene (histone H3), and a mitochondrial gene (cytochrome c oxidase subunit 1, COI). A phylogenetic analysis was conducted on the superfamily Nuculanoidea and family Yoldiidae, providing molecular data for the study of Protobranchia and enhancing understanding of macrobenthos at the Haima Cold Seep.

The genus Yoldiella is believed to be ubiquitous in all the world’s oceans, with a particularly high abundance in deep waters (Benaim and Absalão 2011a). Environmental differences are likely to have a significant impact on the weight of their shell and the rate of their growth (Reed et al. 2013). However, their small size and offshore habitat contribute to their rarity in collections. Prior to this study, no species of this genus had been identified in the South China Sea. The discovery of Y. haimaensis sp. nov. bridges this geographical gap. In fact, this new species has been encountered more than once during the investigation and research on cold-seep fauna in the South China Sea, but in these studies, it has been misidentified as Malletia sp. or as Yoldiella not to species (Dong et al. 2021; Ke et al. 2022; He et al. 2023). The detailed description of Y. haimaensis sp. nov. contributes to the understanding of macrobenthos in the Haima Cold Seep.

Yoldiella is a difficult taxon to define because the morphological differences within this genus are mostly subtle and there are many closely related species. This genus was established by Verrill and Bush (1897), and it is usually characterized by ovate or wedge-shaped shells, which always have a slight postero-ventral sinuosity. The internal cartilage is often relatively large and occupies a simple notch. The external ligament is weak, and the pallial sinus is usually indistinct. Since then, its classification and composition have been controversial. Allen and Hannah (1986) and Allen et al. (1995) redefined the genus Yoldiella considering shell shape, hinge morphology, musculature, and the extent and course taken by the hindgut, which have been widely accepted by researchers. The redescription of Yoldiella by Coan and Valentich Scott (2012) limited it to forms having an elongate, amphidetic ligament with an internal section, an obscure or absent escutcheon and lunule, and a small pallial sinus.

It is widely acknowledged that soft-tissue analysis plays a crucial role in contemporary malacology. However, there is a significant lack of detailed morphological descriptions for this taxon. Yonge and Calman (1939) presented a comprehensive summary of the anatomical characteristics of Protobranchia. Purchon (1956) surveyed the structure and function of the stomach throughout the Protobranchia to establish evidence for phylogenetic relationships between the families and orders within the subclass. Xu (1999) highlighted a significant distinction in the anatomical features of Protobranchia compared to other bivalves, emphasizing the muscular nature of their feet for crawling. The ctenidium assumes a lamellar shape primarily for respiratory purposes, while the well-developed labial palp functions as a distinct feeding organ. Okutani and Fujiwara (2005) illustrated the soft part of Yoldiella kaikonis Okutani & Fujiwara, 2005. Reed et al. (2014) described and deliberated on the gonad morphology and oocyte size of four species of Yoldiella. Yonge and Calman (1939) and Allen et al. (1995) provided detailed accounts of the anatomical characteristics of Y. lucida, the type species of Yoldiella, noting that the organs in this species are more compactly arranged due to its relatively shorter shell, which exhibits a greatly abbreviated posterior rostrum. The labial palps are moderately large and extended, with well-developed, long, and muscular palp proboscides. The foot is relatively larger, as are the ctenidia, the filaments of which are broader and deeper. The shell and anatomical features of the new species align with the aforementioned characteristics of Y. lucida.

The genus Yoldiella may encompass species from other genera due to its small size and the potential for confusion with immature specimens. La Perna (2004) observed that, based on the traditional description at the time, Yoldiella was regarded as a provisional “pigeon-hole”, where numerous species were temporarily allocated in a kind of waiting list, rather than a natural group. It was proposed that within this genus, some clusters of morphologically similar species could be recognized, with the morphological differences among these clusters suggesting distinct systematic ranks. Benaim and Absalão (2011a) followed this conjecture by analyzing the morphological characteristics of some Atlantic species using empty shell specimens and proposed three distinct clusters, grouping together Y. nana, Y. inconspicua, and Y. americana Allen, H.L. Sanders & F. Hannah, 1995. However, in our molecular phylogenetic analysis of these three species, they did not cluster in a single branch. The feasibility of this hypothesis requires validation through the combination of anatomy, morphometry, molecular data, and other methods. In addition, hinge plate features are considered a significant diagnostic feature in descriptions of Yoldiella species, especially the width of the posterior hinge plate (Benaim and Absalão 2011b; Benaim et al. 2011). Further meticulous examination of all taxa is essential before achieving confidence in the accuracy of classification.

Based on anatomical and morphological characteristics, Protobranchia has traditionally been regarded as a monophyletic group (Pelseneer 1889; Xu 1999), but the widespread application of molecular methods in phylogenetic analysis has not provided support for the monophyletism of the Protobranchia (Combosch et al. 2017). Recent studies have supported the monophyly of Protobranchia and five superfamilies (Nuculoidea, Manzanelloidea, Solemyoidea, Sareptoidea, and Nuculanoidea) (Smith et al. 2011; Sharma et al. 2012; Bieler et al. 2014; González et al. 2015; Lemer et al. 2019; Sato et al. 2020). Notably, the monophyly of lower taxa (family and below) is still uncertain, particularly within the superfamily Nuculanoidea. In this study, all currently recognized families were found to be non-monophyletic, which is consistent with previous research findings (Sharma et al. 2013; Sato et al. 2020). In addition, the limited support for the majority of nodes within the Nuculanoidea has resulted in largely ambiguous internal relationships. This phenomenon may be the result of a combination of factors, possibly the lack of genetic information, deficient taxon sampling, or both, or even the possibility that this taxon may have undergone rapid radiation (Lemer et al. 2019).

Phylogenetic analysis of the Yoldiidae revealed differences between the evolutionary relationships inferred from the COI and the 18S rRNA genes. Specifically, the phylogenetic tree based on COI gene shows that Yoldia is polyphyletic, in which species were interspersed among other genera rather than forming a distinct clade. In contrast, the phylogenetic tree derived from the 18S rRNA gene shows that Yoldia is a monophyletic group within a single clade. Such incongruities may be attributed to variations in the evolutionary rates of the two genes. The genetic markers employed in this study, such as COI and H3, exhibit rapid evolutionary rates, particularly at the third codon position, which are almost certainly saturated when applied across multiple taxonomic families. In addition, Mallatt et al. (2010) explored the metazoan tree using almost complete rRNA genes (18S and 28S), which suggested that non-monophyly of Mollusca in analyses based on 18S rRNA might be explained by these non-homologous forms, and indeed paraphyly/polyphyly of Mollusca has been found in many studies of metazoan phylogeny based on 18S rRNA. In addition, Mallatt et al. (2010) investigated the metazoan phylogenetic tree by analyzing nearly complete 18S rRNA genes. Their findings did not support the monophyly of Mollusca, indicating that the non-monophyly of Mollusca in 18S rRNA-based analyses could be attributed to non-homologous forms. Indeed, the paraphyly/polyphyly of Mollusca has been observed in numerous studies on metazoan phylogeny utilizing 18S rRNA (e.g. Winnepenninckx et al. 1996; Mallatt et al. 2010, 2012). Based on the findings derived from the analysis of 18S rRNA, the potential scenario cannot be ruled out. Consequently, great care needed to be taken in the application of 18S rRNA for phylogenetics to avoid the incorporation of non-homologous variants of 18S rRNA in the assessments.

There may be more problems with traditional identification that relies solely on anatomy and morphology. Challenges persist due to the difficulty in obtaining protobranch samples, resulting in relatively limited molecular data with insufficient resolution. As shown in our phylogenetic analysis, the bootstrap values are low which may be unreliable. Nonetheless, this work represented a minor step forward and demonstrated the complexity of the taxonomic classification system within this taxon and made it clear that a greater coverage of taxa and more informative genetic markers (not limited to a single gene, but based on high-throughput sequencing technologies for massive orthologous genes from transcriptomes or even genomes) will provide great potential for improving the resolution of phylogenetic classification in this group in the future.