The abyssal plains of the Clarion-Clipperton Zone (CCZ) are famous for their fields of polymetallic nodules, which are inhabited by benthic invertebrates. Ten specimens from the Interoceanmetal Joint Organisation (IOM) licence area in the CCZ were collected in 2014 and submitted to a short-read genome skimming sequencing. In total, mitochondrial genomes and nuclear ribosomal genes were retrieved for nine different organisms belonging to Ophiuroidea, Holothuroidea, Polychaeta, Bryozoa, Porifera, and Brachiopoda (assigned to these phyla immediately upon retrieval from the seafloor). As many of these samples were partial and physically deteriorated following their seven-year storage in IOM’s collections, their morphology-based taxonomic identification could rarely be performed at the lowest possible level (species or genus) prior to preparing the samples for molecular or genomic investigations. Therefore, it was not possible to apply the reverse identification scheme recommended for such investigations. However, several of these specimens represent poorly studied groups for which few molecular references are available as of now. In two cases, the presence of introns in the mitochondrial genome questions the practicability of using the cox1 gene for further routine molecular barcoding of these organisms. These results might be useful in future DNA primers design, molecular barcoding, and phylogeny or population genetic studies when more samples are obtained.

Brachiopoda, Bryozoa, genome-skimming, Holothuroidea, introns, Mitochondrial genomes, Ophiuroidea, Polychaeta, Porifera, ribosomal RNA genes

Located in the Pacific Ocean, the Clarion-Clipperton Zone (CCZ) spans 4.5 million km² between Hawaii and Mexico. The abyssal plain of this area has recently become a focus of attention due to the massive presence of polymetallic nodule deposits on its floor, which hold potential for exploitation. Far from being a lifeless environment, the floor of the CCZ is inhabited by benthic fauna (Rabone et al. 2023), mostly composed of invertebrates (e.g., Amon et al. 2016; Christodoulou et al. 2019) and large benthic foraminifera (e.g., Stachowska-Kamiǹska et al. 2022; Gooday and Wawrzyniak-Wydrowska 2023). Although the scale of environmental impacts of nodule exploitation activity (deep-sea mining) is yet to be fully understood, the retrieval of the resource from the seafloor is likely to affect benthic fauna, especially the sessile species which live attached to the nodules.

Environmental considerations have led the International Seabed Authority (ISA) to issue a certain number of recommendations in order to assess benthic biodiversity in the licence areas as part of the baseline studies. Environmental baseline studies are conducted by ISA contractors holding exploration licences with the aim to describe this benthic environment. The findings of the studies will constitute a basis (the baseline) against which exploitation impacts will be assessed, and are to be incorporated into an informed decision-making process by both ISA (e.g., while preparing standards and guidelines and defining environmental thresholds) and contractors (e.g., while preparing environmental management and monitoring plans).

Barcoding and more generally genetic studies are some of the tools used to identify and/or support taxonomic identification of fauna – in particular, key and representative species that could be used as indicator for assessing impacts – collected during exploration activities (ISBA/25/LTC/6/Rev.3 2023). This taxonomic work is of primary importance in species cataloguing and biodiversity assessment. In addition, molecular studies help to unravel the ecological functions and connectivity of species or assemblages (Danovaro et al. 2017). Although reverse taxonomy is advocated (ISBA/25/LTC/6/Rev.3 2023), this approach is not always possible when revisiting legacy data and specimen collections, as in the case in this article. Fully aware of the limitations resulting from the lack of proper morphology-based taxonomic identification of specimens retrieved from the IOM licence area in 2014, we nevertheless decided to try a genome-skimming approach on these samples.

IOM conducts exploration activities in the area located in the eastern part of the CCZ under the contract signed with ISA in 2001. Recently, IOM has developed its own protocol for the molecular study of the CCZ benthic fauna (Gastineau et al. 2023). This protocol emphasizes the use of the genome-skimming approach based on short-read sequencing whenever possible, with the aim of obtaining the largest amount of data possible within a single sequencing. In the best-case scenario, the outcome could be a complete cluster of nuclear rRNA genes and/or a complete mitochondrial genome. This is exemplified by the aforementioned article on Abyssoprimnoa gemina Cairns, 2015 (Gastineau et al. 2023). This deep-sea coral was collected during the IOM cruise to the CCZ in 2014, together with the specimens described in the present article. All the specimens were documented by macrophotography immediately upon retrieval but were not taxonomically identified to the species level at that time, except in a few cases, including A. gemina. The samples had been stored in ethanol 96% for seven years before the molecular and genomic protocol was implemented. This, in addition to the fact that several of these samples were partial, resulted in many cases in their poor physical conservation, which made it impossible to perform proper morphology-based taxonomic identification. Moreover, it has to be stressed that in the case of the CCZ fauna, there are still many species that remain undescribed (Rabone et al. 2023). Even if references exist for some of these taxa, it might still be difficult to identify them at the species level, considering the scarcity of taxonomic expertise.

When applying a basic molecular barcoding protocol to the 2014 samples to amplify the 18S and cox1 genes, we faced several challenges. As much as we could generally obtain positive results for the 18S gene, we mostly failed to amplify the cox1 gene, regardless of the phyla. This was likely a consequence of the DNA primers not being sufficiently specific rather than insufficient quality or quantity of the DNA extracted. Indeed, in several cases, the amount of DNA retrieved qualified the samples for a next-generation type of sequencing as previously performed on A. gemina (Gastineau et al. 2023).

In the current article, we present the results of a genome-skimming strategy applied to ten samples from the CCZ that represent two species of Ophiuroidea, one Holothuroidea, one Polychaeta, two Bryozoa, two Porifera, and one Brachiopoda. Of these samples, only the Ophiuroidea specimens could be identified at the species level (morphological identification confirmed by molecular results). It has not been possible to identify or describe the other samples at the species level so far, but our findings still hold some potential for the scientific community involved in the exploration of the CCZ. Some of these samples may represent poorly studied phyla for which few molecular references are available. Some others have mitogenomes with complex features that could not be resolved by the usual PCR and Sanger sequencing protocol, which in some cases render the amplification of their cox1 gene impossible.

It is assessed that there is a substantial gap between the CCZ biodiversity and the described metazoan species. Many species remain not only to be described, but also to be discovered (Amon et al. 2016; Christodoulou et al. 2019; Rabone et. al 2023). The gap is being slowly filled in, thanks to the increasing number of sampling efforts undertaken by scientist over the past years (including scientists working with ISA contractors), but at the same time new areas of knowledge gaps are being identified. To a certain extent, this paradox is reflected in the increasing amount of environmental data that ISA contractors are required to collect. Comparison of the LTC recommendations from previous years with the most recent ones clearly underlines the gaps in our knowledge (ISBA/25/LTC/6/Rev.3). This includes but is not limited to the application of genetic studies in assessing benthic biodiversity and population connectivity of organisms.

Although molecular studies have been making rapid progress over past years, advancing our knowledge on benthic metazoans, there are still phyla that have received limited attention from a genomic point of view. This is the case for Brachiopoda for example, which has been scarcely documented so far. Among the more than 400 known non-fossil species of Brachiopoda, fewer than ten have had their mitogenomes sequenced, with the majority of the sequences belonging to inarticulate brachiopods (Karagozlu et al. 2021; Niaison et al. 2021; Breton 2024). Only four species (and three genera) of articulate taxa, to which specimen IOM_2014_62 is likely to belong, have had their mitogenomes sequenced and published (Stechmann and Schlegel 1999; Helfenbein et al. 2001; Karagozlu et al. 2017; Noguchi et al. 2000). The percentage of identity of the cox1 gene between specimen IOM_2014_62 and the four other species ranges between 63.27% and 72.31%. There is a practical implication of such differences: DNA primers designed using previously published reference mitogenomes may not anneal correctly on the DNA of a specimen such as IOM_2014_62, and possibly other Brachiopoda from the CCZ. Documenting these taxa with more mitogenome data could help solve this problem, with the subsequent possibility to design more efficient primers.

Among our specimens, half of the Porifera and Bryozoa have introns in their mitogenomes, in each case in the cox1 gene. Although rare and nearly absent in other taxa groups, this is not the first time that introns have been found within the mitogenomes of these two phyla (Rot et al. 2006; Jenkins et al. 2022). In the case of Porifera, intron content within a single species has been proven to vary across populations (Cranston et al. 2021). Not only are these findings interesting from the evolutionary genomics point of view but, due to the unpredictable presence of introns, they also challenge the use of the cox1 gene for routine molecular barcoding of the CCZ Bryozoa and Porifera (Neal et al. 2022, 2023). Owing to the presence of introns, amplification of this gene by PCR might fail, or at least will require adoption of a protocol for longer elongation time and possibly the use of the Taq polymerase suitable for a long PCR.

A solution to such issues might be our genome-skimming approach. However, this approach has its limitations, one of them being that obtaining the required amount of DNA could result in the destruction of the smallest samples. There could thus be a risk of not leaving a correct specimen voucher behind, which is not in line with the ISA recommendations that advocate for reverse taxonomy followed by curation of voucher specimens and molecular samples in order to maintain the link between morphology-based and molecular-based identifications (ISBA/25/LTC/6/Rev.3 2023). Otherwise, such approach might require a preliminary treatment such as whole genome amplification. Regardless of the above limitation, when specimens qualify in terms of biomass, or are expendable because of their limited further use (which was the case with some of our own material), our approach could still be applied.

Among the ten specimens in this study, five of the sequences obtained matched the sequences stored in GenBank. Sequencing confirmed that specimen IOM_2014_38 is S. daleus and that specimens IOM_2014_54 and IOM_2014_58 were O. glabrum. We regard as especially promising the results obtained on O. glabrum, for further studies in the emerging field of population genetics and connectivity in the CCZ (Taboada et al. 2018; Riehl and De Smet 2020). With 12 SNPs out of 1602 bp, the cox1 gene would be a useful population marker for this species, as already suggested by the works of Christodoulou et al. (2020). It could be noted that none of the sequences obtained by Christodoulou et al. (2020) were identical to the cox1 gene of IOM_2014_54 and IOM_2014_58, which suggests a large polymorphism of this gene among this species.

The two other specimens which matched to some degree with GenBank references were identified as Demospongiae (IOM_2014_13) and Polychaeta (IOM_2014_17). Both were far more degraded, especially IOM_2014_17, which was torn into two pieces of ~ 1.5 cm each. Neither of the specimens was suitable for taxonomy, neither preliminary nor reverse. Megablast queries returned a 99.28% identity of IOM_2014_17 with Nicomache cf. benthaliana NHM_058, an organism that has been previously found in the licence areas UK-1A and UK-1B (UK Seabed Resources Ltd.), BGR (Federal Institute for Geosciences and Natural Resources of the Federal republic of Germany) and OMS (Ocean Mineral Singapore PTE Ltd.), all of which are located to the East of IOM claim area (Stewart et al. 2023). It is also worth reminding that queries of partial 18S, 28S, and cox1 returned a complete identity between IOM_2014_13 and Spinularia sp. voucher RC1570 which was also sampled in the Eastern part of the CCZ, as for the aforementioned Polychaeta.

The results of genome comparison obtained for specimen of Demospongiae IOM_2014_13 are more intriguing. The 100% identity with the partial cox1 gene of S. sarsii poses some problems. As far as we know, this species has never been reported in the CCZ. It is mostly found in the Atlantic Ocean, and a few locations have also been reported for the South-West Pacific (de Voogd et al. 2024). If the presence of S. sarsii is confirmed in the CCZ, it will raise questions about its global distribution. However, no further assessment should be done for this species based on our sequencing results, as we are faced with two taxonomic issues. First, it was impossible to perform a correct morphology-based identification of a partial and degraded specimen. Second, it should be noted that the 100% identity of the cox1 gene was returned for a 658 bp fragment deposited in GenBank, which is less than half the length of the complete cox1 gene of Demospongiae IOM_2014_13, leaving room for informative polymorphisms outside this 658 bp fragment. Moreover, the differences in length between queries and the sequences registered in GenBank may lead to further difficulties. When results are sorted based on their ‘Max Score’ or ‘Total Score’, different lengths and the impact they have on the ‘Query Cover’ parameter affect the result returned by such query. If the reference sequence deposited in GenBank is considerably shorter than the query, it may lead to the exclusion of the reference from the sequences producing significant alignment, whose number is limited to a maximum of one hundred. This in fact could serve as an argument in favour of our genome-skimming approach: submitting complete mitogenomes to GenBank means that querying a partial gene belonging to an identical species from the CCZ will return complete sequences as top results. With this in mind, we hope that the results presented here could be used as references in future studies on phylogeny, distribution of species and possibly population genetics of benthic organisms inhabiting the CCZ. We also hope that further investigations by other teams would lead to a more formal identification or description of the unidentified taxa here studied, and that such studies will benefit from the genomic results here presented.