Sharks are among the most vulnerable species inhabiting the marine realm, and although they are characterised by a long lifespan, they exhibit slow growth, late maturity and low fecundity, and other biological traits that may pose a significant threat to their populations (Stevens et al., 2000; Kousteni and Megalofonou, 2019). Habitat selection (DiGiacomo et al., 2020) and complex behaviours such as philopatry or site fidelity (Pillans et al., 2021) can additionally limit their capacity to recover from population declines related to overexploitation (Dell’Apa et al., 2012) and habitat fragmentation and depletion (Barausse et al., 2014). In deep waters, these intrinsic features may be exacerbated, increasing their vulnerability to human impacts and particularly exploitation (Rigby and Simpfendorfer, 2015). Concerns about the overexploitation of the deep-sea waters, lead to the identification, assessment and management of stock units of several shark species in the Atlantic Ocean (Veríssimo et al., 2011a; Daley et al., 2019; ICES, 2021), where bottom trawling has been banned below 800 metres (European Parliament, Council of the European Union, 2016). Within the Mediterranean Sea, the General Fisheries Commission for the Mediterranean (GFCM) banned bottom trawling below 1,000 m depth, allowing a possible refuge zone for mature individuals of those species showing depth-dependent size segregation (Gouraguine et al., 2011; Ramírez-Amaro et al., 2020). However, such measures may have likely shifted the trawling effort towards deeper waters (Tserpes et al., 2011; Farriols et al., 2019). On top of that, the effects of bycatch on already poorly investigated deep-water species may become even more dangerous, hindering the quantification of indirect impacts such as cascade effects on trophic interactions and loss of biomass (Myers et al., 2007; Cashion et al., 2019; Serena, 2021).

The blackmouth catshark Galeus melastomus Rafinesque, 1810 is a common benthic shark whose nominal family is still debated (Iglésias et al., 2005; Musick and Ellis, 2005; Dymek et al., 2021). Currently the species belongs to the family Pentachidae (Naylor et al., 2012; Kousteni et al., 2021) and it is widely distributed on the outer continental shelves and upper slopes in the North-eastern Atlantic Ocean and the Mediterranean Sea (Kallianiotis et al., 2000; Follesa et al., 2011; Abella et al., 2017; Froese and Pauly, 2019). The species shows a wide bathymetric range (from 55 to 2000 m depth), dwelling between 200 and 500 m depth (Ebert et al., 2021). The distribution, feeding habits and biological features of G. melastomus are well-studied and many authors have been focussing on the size at first maturity, sex ratio, sexual dimorphism, and egg deposition (Costa et al., 2005; Ragonese et al., 2009; Anastasopoulou et al., 2013; Metochis et al., 2018; D’Iglio et al., 2021a; Scacco et al., 2022). The species has shown signs of population differentiation across its distribution range regarding the maximum body-size, the length at maturity (Massutí and Moranta, 2003; Barría et al., 2015) and sex-related body weight (D’Iglio et al., 2021b). Data reported in literature showed differences in the maximum total length (TL) in individuals caught in the Atlantic and in the Mediterranean areas (TL 90 versus 70 cm, respectively; Massutí and Moranta, 2003; Barría et al., 2015), with the Mediterranean Sea individuals reaching the first maturity more rapidly (Massutí and Moranta, 2003; Moore et al., 2013; Metochis et al., 2018). In general, both sexes increase in size and abundance with depth (Rey et al., 2004), with females growing larger than males and reaching sexual maturity at a larger size (Moore et al., 2013; Metochis et al., 2018). Reproduction through spawning was observed all over the year (Capapé et al., 2008), and peaks were recorded in late spring-summer (Ragonese et al., 2009).

Little information is available about blackmouth catshark genetic diversity and population structure (Ferrari et al., 2018; Ramírez-Amaro et al., 2018), being the molecular method previously applied for integrated taxonomy (Castilho et al., 2007) and food traceability (Clarke et al., 2006). With the refinement of genetic and genomic techniques, deepening the knowledge about species genetic diversity and structuring is becoming one of the pillars of elasmobranchs’ conservation (Ovenden et al., 2019; Domingues et al., 2021; Delaval et al., 2022). Finding population structure across wide geographical or global scales failed in pelagic sharks (Hoelzel et al., 2006; Veríssimo et al., 2017) although some exceptions are reported (Kraft et al., 2020), while the identification of isolated populations on smaller regional scales has been documented in multiple coastal species (Portnoy et al., 2015; Catalano et al., 2022).

The above-mentioned population dynamics and habitat use of the blackmouth catshark may influence gene flow, being the latter limited by hydrographic barriers (e.g., currents and frontal systems and water masses), and progressively leading to genetic differentiation (Shaw et al., 2004). Currently, estimating population fragmentation is of primary interest for the conservation of cartilaginous fish species, whose population structure must be carefully assessed. Despite a high discard rate of the species (Ferretti et al., 2005), the International Union for the Conservation of Nature (IUCN) suggested that populations of blackmouth catshark in the Alboran, Balearic, Ligurian and Tyrrhenian Seas can be considered as stable (Ferretti et al., 2005; Ramírez-Amaro et al., 2020), even if the current fishing mortality does not ensure stock sustainability (Abella et al., 2017). Among the main threats, bycatch has been demonstrated to be the heaviest, leading to a steady decrease in mature individuals. According to the IUCN Red List Status, the species is characterised as Least Concern globally (Finucci et al., 2021) and in the Mediterranean Sea (Abella et al., 2016). Nevertheless, the North-eastern Atlantic Ocean and the Mediterranean Sea have been subjected to extensive and intensive deep-water fishing across much of the species’ known depth and spatial distribution (Finucci et al., 2021). Despite the management measures adopted to reduce deep-water shark fishing mortality, the species is still captured in demersal fisheries (e.g., in deep water shrimp trawl fishery), with a bycatch rate steadily increasing in some locations (Pauly and Zeller, 2015; Moura et al., 2018). Discard mortality of G. melastomus is unknown, but is likely to be high based on severe injuries obtained when individuals are captured (Finucci et al., 2021).

The present study aims at assessing the population genetic structure of G. melastomus within a large part of its range, from the North-eastern Atlantic Ocean and from the Western to the Eastern Mediterranean Sea, surveying the genetic diversity of 194 specimens collected from eight geographically distant areas at 13 cross-amplified microsatellite loci. Our findings, in conjunction with the results from previous studies that analysed the species’ genetic diversity by using mitochondrial DNA markers, will advance current knowledge about the species’ population units and contribute to future species assessment in the region.

A total of 13 out of the 129 tested SSRs (10.1%) provided amplicons within the reference size and in both Atlantic and Mediterranean individuals ( SM2 ). Considering a call rate threshold of 90%, a total of 180 individuals constituted the final dataset. Loci Gg11 and McaB28 were discarded since their amplification patterns suggested the presence of multiple genic regions. The most polymorphic locus was MaD2X (Na = 9), while the less polymorphic ones were Cc21 and MaND5 (both with Na = 2). The presence of null alleles ( SM4 ) contributed to a moderate heterozygosity that was observed at the population level ( Table 1 ). The population sample from the Ligurian Sea showed the highest number of alleles (Na = 36) and the highest allelic richness (A_r = 3.219), while the sample from Scottish waters showed the lowest values (Na = 23, A_r = 2.077). In general, genetic diversity estimators showed samples from the Ligurian Sea and Scottish waters as the putative populations with the highest and lowest genetic variability, respectively ( SM4 ).

The pattern and the success of SSRs cross-amplification followed the levels of taxonomic relatedness between species since most of the loci effectively amplified in G. melastomus were those previously developed for species of the closest family Triakidae (Galeorhinus galeus and Mustelus spp.; extensive details and analyses are reported in SM5 ).

The simulation obtained with POWSIM highlighted the high-resolution power of the dataset. Both Chi-square and Fisher’s tests revealed genetic differentiation in most runs and were able to detect F _st values of as low as 0.025 in 100% of the runs (t₁₀₀: χ² = 1.000, F = 1.000, P_value < 0.05).

The mean pairwise F _st value among samples was 0.159 (P_value < 0.001) while the pairwise F _st values ranged from 0.037 (POR-ADR) to 0.348 (TYR-AEG) ( SM6 ). All the pairwise F _st values were significant (P_value < 0.001) except for the comparison SPA-ADR. The AMOVA ( Table 2 and Table 3 ) with the highest percentage of molecular variation among groups (13.62%) and the lowest value of molecular variation among populations within group (3.99%) was the one performed with the subdivision of geographical samples into five groups (SCO; POR; SPA-LIG-ADR; SIC-TYR; AEG).

The DAPC scatter plot showed a separation into three main groups ( Figure 2 ). The SCO group was clearly separated from the rest of the samples along axis 2, the TYR-SIC group separated along axis 1, while the remaining samples only partially separated (e.g., POR, SPA, LIG, ADR and AEG).

The STRUCTURE analyses inferred K = 2 and K = 3 as the optimal number of genetic clusters that best fitted the data with STRUCTURE HARVESTER (ΔK) and STRUCTURESELECTOR (K), respectively ( Figure 3 , Figure 4 , SM7 and SM8 ). In Figure 3 , the barplot of K = 2 showed all individuals except one from the Mediterranean TYR and SIC assigned to a genetic cluster different with respect to that of SCO and AEG individuals. The rest of the samples (POR, SPA, LIG and ADR) were genetically admixed with different proportions of individuals assigned to the two genetic clusters. The barplot of K = 3 improved the separation of SCO from AEG, as well as the admixed composition of POR, SPA, LIG and ADR with individuals associated to the three identified clusters in different proportions.

This research work revealed the first evidence of a detectable genetic structure of the necto-benthic, deep-water G. melastomus populations in the Mediterranean Sea and neighbouring North-East Atlantic areas, whose stocks are potentially affected and impacted by the demersal fishery (Tiralongo et al., 2021). Here, we described the first evidence of a weak albeit detectable genetic structure and the development of a technological tool represented by a panel of cross-specific microsatellite loci which were successful in detecting and measuring species genetic diversity and differentiation of blackmouth catshark over a large, regional geographic scale and a significant number of samples and individuals.

Our results are partially in contrast with previous findings obtained using multiple mitochondrial DNA markers and assessing genetic differentiation based on phylogeographic signals at a similar geographical scale, that suggested high connectivity with past population expansion of the species (Ferrari et al., 2018; Ramírez-Amaro et al., 2018). Indeed, our study newly unravels the genetic complexity of G. melastomus populations on a large geographical scale by using nuclear, and polymorphic DNA markers. Such recorded genetic complexity can be attributable to i) a detectable but weak level of genetic differentiation and population structure that ii) speaks in favour of the occurrence of groups of subpopulations and environmental and/or ecological barriers to gene flow and iii) a great amount of genetic diversity represented by a high number of alleles per locus, high allelic richness and moderately high proportion of heterozygous multilocus genotypes. Given the clearcut evidence of genetic differentiation of the blackmouth sharks from Scottish waters emerged from all the tests and analyses performed, the rest of the population samples clustered depending mostly by the test and the analysis performed but, all in all, indicating that the species is genetically structured in the Mediterranean Sea and in the adjacent North-eastern Atlantic. Both the DAPC and STRUCTURE analyses identified a genetic unit formed by the individuals collected in the Tyrrhenian Sea and the Strait of Sicily, which resulted quite differentiated from the rest of the Mediterranean and Portuguese samples. In addition, the STRUCTURE analyses resolved a certain degree of separation of the easternmost Aegean sample and the admixed nature of the other Mediterranean and the Portuguese samples. The separation of the Central Southern Mediterranean samples could be a combination of different factors, related on one hand to specific environmental conditions in this area as hydrodynamic regimes and front systems limiting the species migration or, on the other hand, related to the modifications induced by the fishing pressure. Although intensive fishing currently involves the whole Mediterranean basin, the Strait of Sicily has been suffering from intense overfishing for decades (Fiorentino and Vitale, 2021). Such impact has been found to contribute to the modification of relevant biological traits of G. melastomus as the decrease of size at first maturity (D’Iglio et al., 2021b). Although the species seems to be resilient and adaptable to human exploitation, mitigation measures based on population dynamics or mating success should not be excluded.

When looking at the interaction between some biological traits of G. melastomus and physical barriers of the marine realm, knowledge about the use of nursery areas (Capapé et al., 2008; Metochis et al., 2018) and sex-based migratory behaviour should be encouraged, since this species might be an important key-taxon for understanding deep-sea ecosystems (Rey et al., 2004). Moreover, if the large geographic distance between sampling locations is not sufficient to explain the lack of connectivity in deep-sea species (Veríssimo et al., 2011b; Catarino et al., 2015; Kousteni et al., 2015a; Kousteni et al., 2015b; Gubili et al., 2016), factors as bathymetry or geomorphology (e.g., the presence of canyons; Fernandez-Arcaya et al., 2017) and temperature may be influencing the genetic structuring of the species. As a matter of fact, large adults of G. melastomus appear to undertake periodical migration from deeper waters toward shallow environments for mating (D’Iglio et al., 2021b). It is reasonable to suppose that abiotic factors such as water circulations and bathymetry might be decreasing the contact potential between populations, contributing to the differentiation of samples from the Tyrrhenian Sea and the adjacent Strait of Sicily and to the homogenization of their genetic variability against other Mediterranean areas (Spanish, Ligurian, Adriatic and Aegean waters). As a matter of fact, within the Mediterranean Sea, the Levantine Intermediate Water and the Deep-water circulations (El-Geziry and Bryden, 2010) might be contributing to the differentiation of samples from the Tyrrhenian Sea and the adjacent Strait of Sicily and the homogenization of genetic variability between other sectors of the Basin (Spanish, Ligurian, Adriatic and Aegean waters). In addition, the important depth of the Strait of Sicily (over 1300 m deep) and the Ionian Sea (average depth near to 4000 m) can explain the very low similarity between samples from Tyrrhenian Sea and Strait of Sicily and the other groups. Specific dispersal potential related to species’ habits and life-history traits, such as the reproductive strategy, the use of mating and nursery areas and sex-based migratory behaviour, may contribute to population structuring (Hirschfeld et al., 2021; Catalano et al., 2022). Migration events or site fidelity may be strongly related to the optimal temperature at which the species is able to maximise growth rate and the bioenergetic efficiency, as already described in other sharks like S. canicula (Sims et al., 2006). The search for optimal temperature by G. melastomus is consistent with its reproduction hot-spot discovered in the deep waters 200 miles off the western Irish coast (O’Sullivan et al., 2018) and agrees with similar observations for other sharks, like Negaprion brevirostris Poey, 1868 (Kessel et al., 2014). High dispersal potential and panmixia have been inferred for some migratory shark species, including Squalus acanthias Linnaeus, 1758 (Veríssimo et al., 2010), Squalus blainville Risso, 1827 (Kousteni et al., 2015a), Centroscymnus coelolepis Barbosa du Bocage & de Brito Capello, 1864 and Centrophorus squamosus Bonnaterre, 1788 (Veríssimo et al., 2011a; Veríssimo et al., 2012). On the contrary, strong genetic differentiation has been described for the less vagile S. canicula when comparing samples from the Ionian and Aegean Seas with Atlantic, Spanish, and Algerian ones (Kousteni et al., 2015b) and within the Alboran Sea and Western Mediterranean (Ramírez-Amaro et al., 2018). According to Ramírez-Amaro et al. (2018), results for G. melastomus collected in the same areas did show genetic connectivity, being the latter well supported by a high number of migrants per generation crossing the Alboran and Balearic Seas.

In addition, ecological traits, such as feeding habits, may also contribute to population structuring. The relationship between predator abundance and prey density has been discussed in Barría et al. (2018); Ramírez-Amaro et al. (2020) and more recently in Sbrana et al. (2022) when describing marine predators’ foraging maximisation. In their works, these authors indicated the possibility that population structure and dynamics may depend on the adoption of opportunistic habits and a highly diverse diet, a trait that may enhance their susceptibility to spatial-temporal changes in prey abundance (Campana and Joyce, 2004; Anastasopoulou et al., 2013).

For some deep-sea elasmobranch species such as C. coelolepis (Catarino et al., 2015), Etmopterus spinax Linnaeus, 1758 (Gubili et al., 2016) and hypothetically also for G. atlanticus (Rey et al., 2010), the Strait of Gibraltar represents a breakpoint in the connectivity of the species. Differently, our results did not highlight such a strong barrier to the gene flow between Mediterranean and adjacent Atlantic, since samples collected along the Portuguese coasts were strongly different from the northernmost Scottish sample but were similar to those collected in the western Mediterranean (Spanish and Ligurian) waters. These results appear discordant with the identification of water circulation assumption as a physical barrier (see Roque et al., 2019 for Atlantic currents), but not with the bathymetry since these areas are divided by the Bay of Biscay characterised by an average depth of 1744 m, much deeper than the common depths inhabited by G. melastomus and reached only occasionally (Ebert et al., 2021). In addition, in the Atlantic Ocean temperature decreases with latitude, likely limiting the migration of individuals along the Atlantic coast and influencing their spatial dispersion (Hemmer-Hansen et al., 2007). Strong genetic differences were also detected by outlier SNPs between Northern and Southern North-eastern Atlantic samples of the European hake Merluccius merluccius Linnaeus, 1758 (Milano et al., 2014), a species with several ecological traits similar to the blackmouth shark.

The relationship existing between sensitive ecosystems (e.g., canyons and cold-water coral habitats, nurseries and refugia) and inhabiting species, should be preserved with effective conservation strategies (Carluccio et al., 2021). In this perspective, further in-depth research focusing on the monitoring of natural populations of G. melastomus and on the comparison of the genetic variability and differentiation of populations over multiple generations are required. Over temporal replicates, any difference in allelic richness and allele frequencies could reveal the existence of a founder effect or past bottleneck events. Additionally, including individuals of G. melastomus from other areas in the Mediterranean Sea as the North African coasts, Western Mediterranean waters and the Levantine Sea could be useful to describe any further differentiation in the Basin, especially in those areas with canyon systems and other essential fish habitats. Similarly, targeting the Atlantic Iberian coasts, the Bay of Biscay, the West of Ireland and the United Kingdom and Norway waters would pinpoint the species’ connectivity in the North-eastern Atlantic Ocean.

Maximising the monitoring of Mediterranean and Atlantic populations of blackmouth catshark both in its temporal and spatial coverage is advisable and it could be achieved by capitalising established scientific surveys where G. melastomus represents one of the target species (i.e., MEDITS and PNAB; Spedicato et al., 2020; Vieira et al., 2020). In addition, the employment of non-invasive sampling protocols (i.e., environmental DNA; Aglieri et al., 2021) represent a valuable approach to obtain data without sacrificing individuals for research purposes, even though such techniques have not been extensively applied to threatened and deep water species (Dugal et al., 2022; Merten et al., 2021). The improvement of the genotyping approach (i.e., the application of an increased number of polymorphic microsatellite loci or the development of panels of Single Nucleotide Polymorphisms, SNPs) will provide more statistical power for the robust delineation of the genetic structure of the species. The application of high throughput sequencing techniques would likely enhance the detection of randomly distributed genomic markers. In species where large numbers of SNPs have been screened, even a small fraction of these markers has been demonstrated to be very informative for population structure analysis, sometimes outperforming highly polymorphic microsatellites (Liu et al., 2005; Lao et al., 2006).

Within the technological framework, this research work advanced by providing a new panel of 13 heterospecific microsatellite markers that were deeply verified to cross-amplify genetic loci in G. melastomus. Very positive results of cross-amplification tests confirmed that the microsatellite loci investigated in G. melastomus are particularly well-conserved over time in related families (i.e., Triakidae with G. galeus and Mustelus spp.), but also in phylogenetically more distant families and species as Hexanchus griseus, within the family Hexanchidae, and C. crepidater, within the family Somniosidae, which are the most distant taxa from Pentachidae and G. melastomus (Naylor et al., 2012). Overall, similar rates of success in cross-amplification were reached here compared to the literature (Griffiths et al., 2011; Helyar et al., 2011; Veríssimo et al., 2011b; see SM9 for comparison of cross-amplification results on G. melastomus). Most importantly, since these loci resulted largely polymorphic at the pan-European scale, this novel panel of microsatellites could be considered an effective analytical tool for assessing and monitoring genetic diversity and differentiation among populations and stocks either over the whole species range and at the subregional level.

With this research work we investigated and assessed the first evidence of genetic structure over a large, regional geographic scale of G. melastomus, a potentially vulnerable species highly represented in the bycatch composition of demersal fisheries. Understanding the intrinsic characteristics of both Mediterranean basin (e.g., bathymetry, water currents and deep-water temperature) and the species life-history traits and ecology will likely disclose the drivers influencing species differentiation, as well as the potential decline in species abundance, meaning the loss of biodiversity at a local scale with important long-term effects leading to local depletion (Dell’Apa et al., 2012). For these reasons, the effective panel of cross-specific and polymorphic microsatellite loci developed here will enhance the monitoring of natural populations of G. melastomus. The results described enhance the refining and integration of the current knowledge on the diversity and connectivity of a key species for deep-sea ecosystems (D’Iglio et al., 2021b). As a mesopelagic predator with a diverse diet, the blackmouth catshark is becoming one of the most important sentinels of the health of deep habitats and an indicator of human impacts on the marine environment that should be effectively managed (Sbrana et al., 2022). The improvement and employment of genetic and genomic tools coupled with wide research actions (e.g., integrating data on biology, reproductive strategies, feeding and migratory behaviours) should be directed at gathering information on spawning and nursery areas, pursuing scientific surveys at sea to detect discrete groups of populations (Domingues et al., 2018). Robust data mining along with strengthened international networks of collaborations maximising the monitoring effort, will likely fill the gaps in our knowledge of this benthic shark, and encourage effective management and conservation plans for the key species and its habitat.

The data presented in this study are included in the supplementary material (SM1; SM10), further inquiries can be directed to the corresponding authors.

Ethical review and approval was not required for the animal study because the following information was supplied relating to ethical approvals (i.e., approving body and any reference numbers): samples of shark individuals analyzed in the present work were obtained from commercial and scientific fisheries. The activity was conducted with the observation of the Regulation of the European Parliament and the Council for fishing in the General Fisheries Commission for the Mediterranean (GFCM) Agreement area and amending Council Regulation (EC) No. 1967/2006. This Regulation is de facto the unique authorization needed to conduct this type of activity.

SC, AF, AC, and FT conceived and designed the study. SC performed the molecular and data analyses. RC and IM contributed to statistical revision. SC and AF wrote the first draft of the manuscript. CB, JD, LF-P, DG, MH, VK, DM, TM, JR, PS, US, FS, and MS provided the samples used in this work. All authors contributed to the article and approved the submitted version.

This research was funded by RFO and Canziani grants given to FT and AC, and by the RFO grant of the University of Bologna for the research fellowships of AF. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

We thank all contributors to this research. In particular, we are grateful to all the participants in the scientific surveys, as well as the crew of each research vessel for their sampling effort.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The reviewer FT is currently organizing a Research Topic with the author CB.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.