Introduction

Corals are key components of sublittoral ecosystems not only in tropical (Spalding et al. 2001), but also subtropical (Czechowska et al. 2020), temperate, and cold ecosystems (Orejas et al. 2009; Bo et al. 2014; Buhl-Mortensen et al. 2018). While there are an increasing number in the studies on the classes of Scleractinia and Octocorallia, inhabiting various depth zones, Antipatharia (Cnidaria: Anthozoa: Hexacorallia) has received relatively less attention (Bo et al. 2012). Antipatharians, commonly known as black corals, encompasses over 300 accepted species (Molodtsova and Opresko 2023) and, under favorable environmental conditions, can form dense aggregations, creating “coral forests” (a type of animal forests, sensu Rossi et al. 2017) or “coral gardens” (sensu Hall-Spencer and Tasker 2006). These communities have some structural and functional similarities with terrestrial forests, with the main difference that they are dominated by animals instead of plants (Rossi et al. 2017). Their colony morphology varies from branched, bush-like, and feather-like to whip types (Wagner et al. 2012). The dense canopies they form can change local physical conditions and generate a three-dimensional habitat, which provide shelter to associated species and, ultimately, increase biodiversity (Freiwald et al. 2004; Buhl-Mortensen et al. 2010; De Clippele et al. 2019). One of the main components of the visible black coral forests-associated biodiversity is accounted by epifauna (i.e., crustaceans, polychaetes and molluscs, Lavelle 2012; Wagner et al. 2012), which can find habitat (Herler 2007), food (Angel 1990; Bo et al. 2012), and protection against predators (Lavelle 2012).

Amphipods are one of the most abundant and diverse taxa of marine macro-invertebrates (Arfianti et al. 2018), as well as a diverse and important component of deep-sea habitats worldwide (Arfianti and Costello 2020a, b). Amphipods include species with different trophic strategies (e.g., detritivores, omnivores, carnivores, and herbivores), and they are predated by other crustaceans, polychaetes, and fishes (Guerra-García et al. 2014; Jiménez Prada et al. 2015). Moreover, they play an important role in marine food webs, by directly or indirectly recycling nutrients and linking different trophic levels (Karlson et al. 2007; Havermans and Smetacek 2018).

The amphipod family Stenothoidae includes marine benthic species ranging from the shallow subtidal to depths over 3,000 m (Krapp-Schickel 2015; Krapp-Schickel and Vader 2015). Several species of Stenothoidae have received attention from the scientific community due to their biological association (i.e., commensalism) with marine sessile cnidaria, such as anemones (Vader and Krapp-Schickel 1996; Auster et al. 2011) and hydrozoans (Lewis 1992), on which they find food, nesting grounds, and shelter (Marin and Sinelnikov 2017). However, to date, there are no studies on amphipods and their ecological interactions with black corals.

In the present study, we investigated the epifaunal composition and abundance in mesophotic forests created by the black coral Antipathella wollastoni (Gray 1857) from Lanzarote Island over time (Canary Islands, eastern Atlantic Ocean). Identification of sampled specimens has been performed in a robust taxonomic framework and resulted in the description of a new amphipod genus and species within the Stenothoidae family, for which a complete key to genera is provided. Then, we described the first biological association between an amphipod species (i.e., Wollastenothoe minuta gen. nov., sp. nov.) and a black coral species. Specifically, the relationship between coral colony size and amphipod abundances was tested for both the entire amphipod assemblage and the new described species, suggesting an exclusive association.

Material and methods

Study region

The study was conducted in the Southeastern coast of Lanzarote Island (Canary Islands, eastern Atlantic Ocean) at Puerto del Carmen (28º55′26.81″ N, 13º 39′ 12.61″ W) (Fig. 1a, b). The site was selected based on previous records of A. wollastoni in the shallower limits of its depth distribution range (ca. 60 m depth, Fig. 1c, d) (Bianchi et al. 2000; Czechowska et al. 2020). In this area, local topography is characterized by narrow rocky shelves and steep slopes, typical of oceanic volcanic islands (Acosta et al. 2005; Tuya et al. 2021). Local hydrography is complex, with NE trade winds affecting shallow subtidal habitats by generating wind waves and near-bottom turbulence (Mann and Lazier 2013). The high nutrient load transported by currents may affect the upper limit of black coral distribution, by either providing food or physically smothering the corals (Wagner et al. 2012; Czechowska et al. 2020).

Fig. 1
figure 1

Map of the Canary Islands in the eastern Atlantic Ocean (a), including the study site located in the Southeastern coast of Lanzarote Island (b). Mesophotic black coral forest of Antipathella wollastoni (c and d)

Full size image

Sampling design and collection of samples

Epifauna was collected in mesophotic monospecific forests of A. wollastoni at three times during 2021: T1 (February), T2 (April), and T3 (October). At each time, 10 colonies spanning between 40 and 180 cm (total height from the seafloor to the upper tip of the colony) were selected randomly, for a total of 30 colonies. All sampling was carried out at the same depth (ca. 60 m), taking care to not sample twice the same area. Sampling was carried out by two divers using mixed-gas rebreather diving. On each colony, samples of associated epifauna were collected by sucking up on their top branches through a vacuum cleaner for 20 s (Fig. 2). The temperature was monitored between February and October 2021 by a temperature data logger (Hobo data-logger Pedant Temp-Light, Onset Computer Corporation, USA; sampling frequency 15 min) positioned next to the black corals.

Fig. 2
figure 2

Epifauna was collected in mesophotic forests of A. wollastoni by sucking up coral branches through a small vacuum cleaner for 20 s

Full size image

Specimen processing

Collected epifauna was directly brought to the laboratory after surfacing in sealed containers filled with seawater. Then, each sample was carefully inspected, rinsed, and sorted to separate epifauna from marine debris and coral mucus. Each organism was then stored in 96% ethanol and identified to the lowest taxonomical level. Identifications were implemented by using a stereoscopic microscope (Leica, EZ4W, Wetzlar, Germany). The taxonomic guide provided by Barnard and Karaman (1991), Ruffo (1998), and specific references from the Stenothoidae family (i.e., Krapp-Schickel 2000, 2011, 2015) was used for amphipod identification. Specifically, 68 specimens of the new amphipod genus/species were dissected in alcohol and mounted on microscope slides using dimethylhydantoin–formaldehyde resin for morphological description. Appendages were observed under a Nikon SMZ800N stereomicroscope and a Nikon Eclipse Ci microscope and photographed with a Nikon DS-Fi2 camera. Body length (BL) was measured with NIS-Elements Analysis software from the anterior margin of the head to the posterior end of the telson. Drawings were carried out from pictures using Inkscape software (v.0.48). For scanning electron microscope (SEM) observations, 12 specimens were dehydrated in a graded ethanol series, critical point dried, using carbon dioxide as a medium, mounted on stubs, coated with gold for 60 s sputter coated with gold, and photographed with a Hitachi tabletop microscope TM3030Plus.

DNA sequencing

DNA extraction was performed on a pool of 23 ethanol-preserved specimens using the MagAttract DNA Extraction Kit (Qiagen) following the manufacturer's protocol. Folmer’s 658-bp fragment of the first subunit of cytochrome c oxidase mitochondrial gene (COI) was amplified using primers LCO1490 and HCO2198 (Folmer et al. 1994). In addition, a fragment of the 28S rRNA gene (28S) was amplified using primers from Verovnik et al. (2005). All amplifications were performed using DreamTaq Green PCR Master Mix (ThermoScientific). For COI, the PCR protocol involved an initial denaturation period at 94 °C for 3 min, followed by 50 cycles of denaturation at 94 °C for 30 s, annealing at 45 °C for 1 min, and elongation at 72 °C for 1 min. For 28S, the PCR protocol comprised an initial denaturation period at 94 °C for 3 min, then 40 cycles of denaturation at 94 °C for 20 s, annealing at 50 °C for 50 s, and elongation at 72 °C for 1 min. PCR products were checked on 1% agarose gel, and subsequently, 1–2 µL of each PCR product (depending on band intensity) was pooled (together with other PCR products from unrelated research projects). The pool was then cleaned up using the InnuPREP PCRpure Kit (Innuscreen) and used to prepare a SKQ-LSK114 library that was sequenced on a FLO-FLG114 Nanopore Flongle flow cell featuring R10.4.1 pore proteins. The resulting reads (1264 for COI and 1932 for 28S) were assembled using amplicon_sorter (Vierstraete and Braeckman 2022).

Phylogenetic analyses

For the COI marker, all Stenothoidae sequences available in GenBank on 19 November 2023 were collected and complemented with sequences of Iphimedia obesa and Gitana sarsi as outgroups. Alignment was performed by hand in MEGA11 (Tamura et al. 2021) taking into account the amino acid translation of the sequences, followed by maximum likelihood phylogenetic analysis using IQtree2 (Minh et al. 2020) with automatic model selection (Kalyaanamoorthy et al. 2017) and 100,000 ultrafast boostraps (Hoang et al. 2018). The resulting Newick tree was displayed, re-rooted, and turned into PDF in MEGA11.

Statistical analyses

All modeling and testing were conducted using R (Rstudio Team 2022). Mixed effects generalized linear models (GLMs) were fitted to univariate responses, including the total abundance of amphipods and the abundance of the new genus and species, Wollastenothoe minuta gen. nov., sp. nov. GLMs were fitted by means of the ‘lme4’ (Bates et al. 2014) and ‘lmerTest’ packages (Kuznetsova et al. 2017) considering time (three levels) as a random factor and the size of colonies of A. wollastoni as a covariate. Models were fitted using a ‘negative binomial’, as the family distribution of residuals, with a ‘log’ link function. For all fitted GLMs, diagnosis plots of residuals and Q-Q plots were visually inspected to check the appropriateness of the fitted models (Harrison et al. 2018). We realized simple linear regressions tested whether the total abundance of amphipod and the abundance of W. minuta gen. nov., sp. nov., were predicted by the colony size.

Results

Taxonomy

Class Malacostraca Latreille 1806

Order Amphipoda Latreille 1816

Suborder Senticaudata Lowry & Myers 2013

Family Stenothoidae Boeck 1871

Genus Wollastenothoe Gouillieux & Navarro-Mayoral gen. nov.

urn:lsid:zoobank.org:act:5FFE2706-D2F1-4EB9-B6A8-B341FF6D10D5.

Type species. Wollastenothoe minuta Gouillieux & Navarro-Mayoral gen. nov., sp. nov., here designated.

Diagnosis of the new genus

Body dorsally smooth. Head without rostrum. Antenna 1 article 1 not nasiform; accessory flagellum with 1 article. Mandible palp with 1 article, molar process conical. Maxilla 1 palp with 2 articles. Gnathopod 1 and 2 subchelate, subequal. P5 basis rectolinear without posterodistal lobe. P6-7 basis widened.

Etymology

The genus name, Wollastenothoe, refers the combination of host name corresponding to the species of black coral (i.e., Antipathella wollastoni) with the genus name Stenothoe belonging to the Stenothoidae family.

Remarks on genus assignation and Stenothoidae diagnosis

Characters of the present new genus agree with the diagnosis of the Stenothoidae family except for the molar process of the mandible. In the present new genus, molar process is slightly developed, conical, whereas the diagnosis for the family Stenothoidae includes a molar process evanescent. This conical shape of the molar process is already present in other genera of Stenothoidae, such as Antatelson, Pseudothaumatelson, Ptychotelson, Raukumara or Thaumatelsonella. According to Horton et al. (2022), 46 genera belong to the family Stenothoidae, and only 12 present a mandibular palp uniarticulate: Ausatelson J.L. Barnard 1972; Metopelloides Gurjanova 1938; Paraprobolisca Ren in Ren & Huang 1991; Prostenothoe Gurjanova 1938; Prothaumatelson Schellenberg 1931; Pseudothaumatelson Schellenberg 1931; Ptychotelson Krapp-Schickel 2000; Stenothoides Chevreux 1900; Stenula J.L. Barnard 1962; Victometopa Krapp-Schickel 2011; Vonimetopa Barnard & Karaman 1987 and Zaikometopa Barnard & Karaman 1987. Wollastenothoe gen. nov. can be distinguished from Ausatelson, Metopelloides, Prostenothoe, Prothaumatelson, Ptychotelson, Stenula, Vonimetopa and Zaikometopa by the presence of an uniarticulate accessory flagellum (vs absence), with Paraprobolisca and Pseudothaumatelson by pereopod 7 basis expanded (vs rectolinear), with Victometopa by pereopod 5 without posterodistal lobe (vs with) and with Stenothoides by 2-articulate maxilla 1 palp (vs 1-articulate). In the case of Paraprobolisca, the only species of this genus, Paraprobolisca leptopoda Ren in Ren & Huang 1991 was synonymized with Probolisca ovata (Stebbing 1888) by Krapp-Schickel and Koenemann (2006). They based their diagnosis on the morphological similarities of the gnathopods and the mouthparts and considered that the uropod 3 described as being one-articulated in Paraprobolisca leptopoda, a character of the genus, has simply been overlooked and should be two-articulated as for Probolisca ovata. Even if most of species have uropod 3 ramus two-articulated, some have an one-articulated uropod 3 ramus as some Raumahara J.L. Barnard 1972 species. Furthermore, Probolisca ovata presents uropods 1 and 2 peduncles marginally bares, whereas Paraprobolisca leptopoda has uropods 1 and 2 peduncles with many robust setae along outer margin. Thus, based on the original descriptions, we consider Paraprobolisca leptopoda as a valid species that should not be synonymized with Probolisca ovata.

Wollastenothoe minuta Gouillieux & Navarro-Mayoral sp. nov.

urn:lsid:zoobank.org:act:7A70A805-5499-4FC6-93AD-1A8CE17C53DE Description of Wollastenothoe minuta.

Type material.

Holotype: brooding female, BL = 1.38 mm, 1 egg (MNHN-IU-2016–3389). Paratypes: brooding female, BL = 1.21 mm, 1 egg, dissected specimen, 11 slides (MNHN-IU-2016–3388); brooding female, BL = 1.45 mm, 2 eggs, dissected specimen, 11 slides (MNHN-IU-2016–3387); female with oostegites, BL = 1.37 mm, dissected specimen, 12 slides (MNHN-IU-2021–8807). Atlantic Ocean, European waters, Canary Islands, Puerto del Carmen in Lanzarote Island. 28′55′26″81″ N, 13º’39′ 12″61″ W, October 2021, 60 m depth, collected by technical diving and rebreathers with the air-vacuum method on black coral branches. Collectors: Francisco Otero-Ferrer, Lorenzo Bramanti and Lucas Terrana.

Additional material: 5 juveniles, 53 females and 10 specimens sex not determined (MNHN-IU-2016–3359), 12 specimens used for SEM pictures (MNHN-IU-2021–8808). Same data as holotype and paratypes.

Diagnosis

Body length less than 1.5 mm. Antenna subequal, shorter than half length of body. Antenna 1 accessory flagellum with 1 small article. Gnathopod 1 and 2 subchelate, subequal. Pereonite 4 slightly longer than pereonite 3. Coxa 4 ventral margin concave. Coxae 5–7 posterior margin with a notch. P5 basis rectolinear without posterodistal lobe. P6-7 basis widened with posterodistal lobe reaching along half of ischium, merus posterodistal lobe reaching more than half length of carpus. Telson with dorsal spines.

Description based on holotype and paratypes (Figs. 3, 4, 5, 6, 7).

Body (Fig. 6a) dorsally smooth, very compressed laterally. Pereonite 4 slightly longer than pereonite 3. Urosomites free.

Head. Antennae subequal in length, setose. Antenna 1 (Fig. 3aA) peduncle article 1 slightly tapering distally, about 2 times longer than wide; article 2 cylindrical, about half length of article 1; article 3 cylindrical, subequal to article 2; flagellum consists of 4 articles increasing in length, accessory flagellum 1 articulate, very small, with 1 to 3 distal setae. Antenna 2 (Fig. 3bB) peduncular article 3 cylindrical, as long as wide; article 4 and 5 subequal, about 2.8 times longer than article 3; flagellum consists of 5 articles of decreasing length.

Fig. 3
figure 3

Wollastenothoe minuta gen. nov., sp. nov. a, b, e, f, h: Paratype MNHN-IU-2016–3387, BL: 1.37 mm. c, d: Paratype MNHN-IU-2016–3388, BL: 1.21 mm. g: Based on SEM picture, MNHN-IU-2021–8808. a Right antenna 1 with a focus on accessory flagellum; b Right antenna 2; c Left maxilla 1; d Left maxilla 2, e Mandible left; f Mandible right; g Upper lip; h Maxilliped. Scale bars: a, b: 0.1 mm; c–g: 0.01 mm; h: 0.025 mm

Full size image

Mouthparts. Upper lip (Fig. 3g) cleft at the apex, asymmetric lobes. Mandible (Fig. 3e–f) Palp uni-articulate with 1 or 2 distal setae, about 1.4 to 2.6 longer than wide, partial suture line sometimes visible at the base of the palp; incisor and lacinia mobilis multi-dentate; accessory setarow consists of 3 blades with comb-shaped distal part; molar process slightly developed, conical, finely setose. Lower Lip (Fig. 7c) Lower lip with many setae, inner lobes coalesced, mandibular lobes well developed. Maxilla 1 (Fig. 3c) inner plate inner margin with a single simple setae, outer plate distal part with 6 large stiff robust setae; palp two-articulate, article 1 smooth, about 2 times shorter than article 2; article 2, with three distolateral robust setae, a subdistal simple seta and some little simple setae on distal part and inner plate. Maxilla 2 (Fig. 3d, 7b) inner plate with 4 long robust setae; apical margin rounded and armed with 3 setae; outer plate inner margin with row of 6 simple setae, distal part with 4 long and 1 short robust setae. Maxilliped (Fig. 3h) with reduced outer plate, inner plate with 2 distal simple setae; palp 4-articulate: article 3 the longest, with many very small setae distally; article 4 with lateral row of setae.

Gnathopods. Gnathopod I (Figs. 4a, b, 7e) subchelate, slightly smaller to gnathopod 2; coxa tapering distally, about 1.5 times longer than wide; basis and ischium with few setae; merus posterior and distal margins with dense small simple setae, posterodistal margin with long simple and plumose setae; carpus widening distally, posterodistal corner with long simple and plumose setae; propodus about twice as long as wide, dorsal margin bare except 2 dorsodistal setae, inner face with row of fine and small plumose setae and 2 or 3 long subdorsal setae; propodus palmar edge serrate, 3 stout robust setae at the base of propodus; dactylus with serrate cutting margin and 2 or 3 simple setae, subdistal notch with 4 setae. Gnathopod 2 (Figs. 4c–e, 7d) subchelate; coxa anteroventral margin regularly rounded, 1.8 times longer than wide, posteroventral corner notched; basis to carpus similar to gnathopod 1; propodus about twice as long as wide, dorsal margin bare except 2 dorsodistal setae; propodus palmar edge not serrated except a small part between 4 stout robust setae at the base of propodus; dactylus with 1 or 2 simple setae along cutting margin, not serrate, subdistal setae present without subdistal notch.

Fig. 4
figure 4

Wollastenothoe minuta gen. nov., sp. nov. a, d: Paratype MNHN-IU-2016–3387, BL: 1.37. b, c, e: Based on SEM picture. a Gnathopod 1 right, outer view; b Gnathopod 1 left, propodus and dactylus, inner view; c Gnathopod 1 distal part of dactylus, outer view; d Gnathopod 2 left outer view; e Gnathopod 2 right, propodus and dactylus, outer view. Scale bars: a, b, e: 0.05 mm; c: 0.01 mm; d: 0.1 mm

Full size image

Pereopods. Pereopods 3–7 subequal in length, with few setae; ischium posterodistal seta absent for pereopods 3–4, present for pereopods 5–7; merus with distal lobe increasing in length, dactylus with dorsal seta. Pereopod 3 (Fig. 5a) coxa elongate, somewhat rectangular, posteroventral corner notched; length ratio of articles from basis to dactylus about 3.3:1:1.9:1.8:2.7:1.5; basis rectolinear, about 3.8 times longer than wide; merus about twice as long as wide, with small distodorsal lobe reaching 0.2 length of carpus; carpus 3.5 as long as wide; propodus elongated, about 5.5 times longer than wide. Pereopod 4 (Fig. 5b) similar in shape to pereopod 3 except for coxa very large, subtriangular, about the same length as the first four segments of the thorax, reaching half-length of ischium, smooth, unarmed, ventral margin slightly concave; length ratio of articles from basis to dactylus about 4.5:1:2.7:1.8:3.3:1.7. Pereopods 5–7 ischium to dactylus similar in shape, merus increasingly expanded. Pereopod 5 (Fig. 5c) coxa mostly developed posteriorly, posterior margin notched with a simple seta inside; length ratio of articles from basis to dactylus about 4.3:1:2.1:1.4:2.8:1.8; basis rectolinear, without posterodistal lobe. Pereopod 6 (Fig. 5d) coxa posterior margin notched with a simple seta inside; length ratio of articles from basis to dactylus about 5:1:2.6:1.6:4:2; basis slightly expanded, about 2 times longer than wide, posterior margin notched with a simple seta inside, posterodistal lobe reaching half-length of ischium. Pereopod 7 (Fig. 5e) coxa rounded, posterior margin notched with a simple seta inside; length ratio of articles from basis to dactylus about 6.3:1:2:1.6:3.8:2.1; basis expanded, slightly longer than wide, posterior margin notched with a simple seta inside, posterodistal lobe reaching two-thirds of ischium length; merus about 1.4 longer than wide, posterodistal lobe reaching 0.7 length of carpus.

Fig. 5
figure 5

Wollastenothoe minuta gen. nov., sp. nov. a, b: Paratype MNHN-IU-2016–3387, BL: 1.37 mm. c, e: Paratype MNHN-IU-2016–3388, BL = 1.21 mm. a–e: Left pereopods 3–7. Scale bars: a–e: 0.1 mm

Full size image

Epimeral plates (Fig. 6a) 1–3 smooth, posteroventral corner weakly produced in a blunt lobe, more or less pronounced; posterior margin weakly convex.

Fig. 6
figure 6

Wollastenothoe minuta gen. nov., sp. nov. a: Paratype MNHN-IU-2016–3388. b: Based on SEM picture, MNHN-IU-2021–8808. c, d, e: Paratype MNHN-IU-2016–3388, BL: 1.21 mm. a Posterior part, lateral view; b Telson, dorsal view; c Right uropod 1; d Left uropod 2, distal part of outer ramus broken; e right uropod 3. Scale bars: a: 0.1 mm; b–e: 0.05 mm

Full size image

Uropods (Figs. 6a, 7f). Uropod 1 (Fig. 6c) biramous; peduncle about 1.2 times longer than rami, with 2 robust setae on outer margin and a distal one, and 1 distal robust seta on inner margin; rami subequal, outer ramus with 1 dorsal robust seta, inner ramus with or without 1 dorsal robust seta. Uropod 2 (Fig. 6d) biramous; peduncle with 2 robust setae on outer margin and 1 distal robust seta on inner margin; rami unequal, inner ramus slightly longer than peduncle, 1.3 times longer than outer ramus, with 1 dorsal robust seta, inner ramus slightly smaller than peduncle, with or without 1 dorsal robust seta. Uropod 3 (Fig. 6e) uniramous, ramus 2-articulate, length ratio of peduncle and articles variable between 1:1:1.1 and 1.4:1:1.2; peduncle with 1 dorsodistal robust seta, ventrodistal corner produced into a small tooth; article 1 with 1 dorsodistal robust seta, ventrodistal corner produced into a small tooth with 1 simple seta, article 2 unarmed, smooth.

Fig. 7
figure 7

Wollastenothoe minuta gen. nov., sp. nov. SEM pictures, MNHN-IU-2021–8808. a Lateral view; b Maxilla 2, left; c Lower lip; d Gnathopod 2, outer face, dactylus and propodus; e Gnathopod 1, inner face, dactylus and propodus; f Urosome, lateral view. Scale bars: a: 0.25 mm; b: 0.01 mm; c: 0.01 mm, d, e: 0.02; f: 0.1 mm

Full size image

Telson (Fig. 6b) entire, tapering distally, with 1 or 2 lateral robust setae and a small subdistal seta on each side.

Male: unknown.

Color in vivo: Whitish brown.

Type locality. Puerto del Carmen (28º55′26.81″ N, 13º 39′ 12.61″ W), Lanzarote, Canary Islands, Spain.

Etymology. The epithet specific of the species, minuta, refers to its small size.

Key to Genera of Stenothoidae (based on Barnard & Karaman 1991 key with inclusion of recent subsequent new genera descriptions)

figure a
figure b

Abundance patterns

In total, 1,736 epifaunal individuals were sampled, with amphipods representing a 98.3% of the total epifauna (Table 1); there were 1,721 amphipods belonging to 6 species (Table 1). Significant variation in the total abundance of amphipod was mainly explained by colony size (Pseudo-R2 = 0.85; Table S1), while the random factor time hardly explained variability (Pseudo-R2 = 0.13). The highest abundances of amphipods (mean ± SE) were recorded in T3 (163.3 ± 36.2 ind. colony−1), relative to T1 and T2 (27.6 ± 8.7 and 35.1 ± 9.4 ind. colony−1, respectively; Fig. 8a; Table S2). Wollastenothoe minuta gen. nov., sp. nov., was the most abundant species (Table 1) over time, accounting for 86.6% of amphipods in T1, 59.3% in T2 and 92% in T3 (Fig. 8b). These differences were mostly explained by colony size (Pseudo-R2 = 0.97; Table S1) and, to a lesser extent, by different sampling times (Pseudo-R2 = 0.10). A significant positive correlation between the total abundance of amphipods and the size of the coral colonies was found, i.e., the bigger the colonies, the larger the abundances of amphipods (Fig. 8c). A similar pattern was found for W. minuta gen. nov., sp. nov. (Fig. 8d). The highest abundances found in T3 (October) coincided with the higher average monthly temperature recorded in our study (Fig. S1).

Table 1 Epifauna living on colonies of the black coral Antipathella wollastoni. For each taxon, the total abundances recorded during the study, and their proportion with respect to the total epifaunal abundance, are shown
Full size table
Fig. 8
figure 8

Temporal variation in the total abundance of amphipods (a) and Wollastenothoe minuta gen. nov., sp. nov. (b), at T1 (February), T2 (April) and T3 (October). Relationship between colony size of Antipathella wollastoni and the total abundance of amphipods (c) and W. minuta gen. nov., sp. nov., at different times (d)

Full size image

The new genus and species were found with other amphipod species belonging to five different families: Caprellidae (i.e., Phtisica marina Slabber 1769), Pleustidae (Buchholz, 1874), Ischyroceridae Stebbing, 1899 (i.e., Jassa Leach, 1814), Lysianassidae (Dana, 1849) and Oedicerotidae (Lilljeborg, 1965).

Phylogeny

Molecular data were successfully obtained at the two independent loci investigated. The COI sequence (GenBank accession number PP595991) was 658 bp length, and molecular phylogenetic analyses were consistent with the position of the new genus and species within the Stenothoidae family (Fig. 9), confirming its distinctiveness from other genera for which COI sequence data are currently available. The closest sequence (p-distance = 0.25) was from a specimen identified as Metopa boeckii G.O. Sars, 1892, but sequences attributed to this species were found at three different places in our Stenothoidae COI tree (with p-distances among them of about 0.25), suggesting that some of these sequences come from misidentified specimens. The 28S sequence fragment (GenBank accession number PP594429) was 1025 bp long—due to the dearth of Stenothoidae 28S sequences currently available in GenBank; no phylogenetic analysis was conducted for this marker.

Fig. 9
figure 9

Maximum-likelihood tree obtained using IQtree2 with 100,000 ultrafast bootstraps showing the relationships between Wollastenothoe minuta gen. nov., sp. nov., and other Stenothoidae COI sequences available in GenBank. Iphimedia obesa (Iphimediidae) and Gitana sarsi (Amphilochidae) were used as outgroups. These sequences attributed to Stenothoidae grouped with Gitana sarsi suggesting that these sequences may come from misidentified specimens, but as the corresponding nodes have low ultrafast bootstrap support values (< 65%), this grouping is not strongly supported

Full size image

Discussion

This is the first study to date describing the temporal patterns of a new genus and species of amphipod associated with a black coral (i.e., Antipathella wollastoni). Wollastenothoe minuta gen. nov., sp. nov., was the most abundant and recurrent over time amphipod associated with this habitat, and we found that the abundances of this new genus and species were directly related to colony size.

The 46 valid genera of Stenothoidae include around 280 species (Horton et al. 2022), among which the genera Metopa Boeck 1871 and Stenothoe Dana 1852 are represented by more than 40% of the total species. Among these, 17 genera include only one species, and 10 genera are represented by only 2 species (Horton et al. 2022). W. minuta gen. nov., sp. nov., represents the eleventh monotypic genus and the tenth Stenothoid genus in European waters. Its small size (i.e., 1.21—1.45 mm), combined with the difficult accessibility of the mesophotic black coral habitat where the species lives, may explain why this genus had not been yet discovered. Our results suggest a high specificity of W. minuta gen. nov., sp. nov., for A. wollastoni as a host. To date, studies conducted in other habitats from the Canary Islands such as rhodoliths (Otero-Ferrer et al. 2019; Navarro-Mayoral et al. 2020), seagrasses (i.e., Cymodocea nodosa; Navarro-Mayoral et al. 2023), seaweeds (i.e., Caulerpa prolifera; Tuya et al. 2014), or sediments (Riera et al. 2012) have not revealed the presence of this genus and species.

Many species of amphipods live in close association with a wide variety of cnidarians (e.g., sea anemones, gorgonians and corals), through different types of specializations with their hosts (Vader 1983), which vary depending on the family and genus. Amphipods on black corals have been described as opportunistic (Wagner et al. 2012), based on the observations of caprellid amphipods feeding on living tissue of Antipathes sp., which led to the death of the entire colony (Tazioli et al. 2007). However, these observations are limited to caprellids and cannot be extended to all amphipods. Furthermore, it is not clear if the cause of death of Antipathes sp. is directly linked to the presence of amphipods, or to any previous disturbance. In our study with A. wollastoni, we reported a different pattern relative to Tazioli et al. (2007), with large abundances of amphipods that varied over time, while we did not observe any tissue necrosis of colonies throughout the study (almost 1 year). In fact, the most abundant species, W. minuta gen. nov., sp. nov., belongs to Stenothoidae, a family that include some species which tends to establish a biological association and display a strongly specificity with their hosts, as it has been observed with hydroids, for example (Vader and Krapp-Schickel 1996). This fact implies that amphipods spend their entire life cycle on their hosts, where they find shelter and food, such as Stenothoe brevicornis with the sea anemone Actinostola callosa (Vader 1983). Regarding the relationships between corals and amphipods, multiple ecological roles between them have been reported. Stenula nordmanni, for example, was observed in mutualistic relationship with some octocorals (e.g., Gersemia rubiformis, Caulier et al. 2021) and Stenothoe valida in commensalism with some hydrocorals (e.g., Millepora complanate, Lewis 1992). Understanding the interspecific relationships between black corals and amphipods is of great importance, especially in the case of W. minuta gen. nov., sp. nov., which was the dominant species over time in all A. wollastoni colonies. Regarding their population structure, we did not find any males among the dissected individuals. This result was expected, considering that we dissected 68 specimens out of 1496, and other studies have observed a skewed sex ratio toward females in species of the family Stenothoidae, such as Stenothoe valida that exhibited a sex ratio of up to 0.79 for females (Lee and Park 2021).

We found that amphipods were the dominant epifauna on A. wollastoni colonies over time, accounting for 98.3% of the total abundances, with W. minuta gen. nov., sp. nov., as the most abundant species by far (87.6% of the amphipods). Contrary to our results, the few previous studies on epifauna associated with black corals did not report a dominance of amphipods (Bo et al. 2012; Wagner et al. 2012; Deidun et al. 2015; Matamoros-Calderón et al. 2021), except Love et al. (2007), who reported a dominance of the amphipod Ericthonius rubricornis Stimpson 1853 on Antipathes dendrochristos Opresko 2005. However, this result was obtained from a single dead coral colony, and it is not easily comparable with our results, as in this case it is not sure if amphipods may be found in living colonies. Moreover, in our study, the abundances of amphipods were consistently large throughout the study, despite some variation among times. We observed the colonization of the near sea bottom by epibionts, such as sponges (e.g., Axinella spp.), and the presence of ascidians (e.g., Stolonica sp. and Pycnoclavella sp.) on the branches of A. wollastoni, but in the latter case the presence in the colonies was relatively infrequent or rare. Temporal variation in the abundances of amphipods can be attributed to changes in habitat structure throughout varying time scales, which alter habitat complexity via increased occurrence of epiphytes and associated algae (Jacobucci et al. 2009; Navarro-Mayoral et al. 2020). However, unlike other ecosystem engineers, such as seagrasses (e.g., Zostera noltii; Vermaat and Verhagen 1996), the primary habitat generated by A. wollastoni is stable over time and does not experience seasonal changes.

Our results showed that variation in amphipod abundances was correlated by the size of the colonies of A. wollastoni, with a significant (positive) relationship between coral colony size and amphipod abundances, both for the entire amphipod assemblage and for W. minuta gen. nov., sp. nov. Moreover, there is a colony size influence on the high variability of abundances within the times, especially in T3, where a greater difference in heights between the randomly selected colonies led to larger variations in abundances between replicates. Therefore, this suggests that the presence of amphipods is determined by habitat availability (i.e., colony size). Colonies of A. wollastoni dwell on rocky platforms, where they provide three-dimensional complexity to the substratum, offering a variety of microhabitats for several species (Czechowska et al. 2020). When A. wollastoni colonies are dense enough, they form a canopy (marine animal forest, sensu Rossi et al. 2017) that gives shelter and protection against strong currents and predators (Buhl-Mortensen et al. al. 2018). Thus, amphipods can benefit from different habitats provided by black corals, including (1) branches surface, (2) cavities within tissues or skeletons, and (3) free space between branches (Buhl-Mortensen and Mortensen 2004). This is consistent with the relatively large body of research that relates differences in abundances of marine invertebrates, in general, and amphipods in particular, with changes in the availability of habitat (Osman 1977; Aikins and Kikuchi 2001), so a higher abundance is observed in habitats with greater complexity and more available surface. Moreover, at these depths, there is a more limited availability of biogenic habitats compared to shallow waters, e.g., coral reefs, kelp forests, or seagrass meadows (Arfianti and Costello 2020a, b). Thus, the presence of A. wollastoni provides a 3-dimensional habitat for many species.

We found that the highest amphipod abundances were recorded in October (T3), which coincided with the highest temperature recorded. Some amphipods in shallow water environments show a turnover (i.e., growth and reproduction), which it is commonly adjusted to seasonal variation in terms of environmental conditions (e.g., temperature, photoperiod) and food resources (e.g., epiphytic biomass for herbivorous amphipods) (Neuparth et al. 2002; Martins et al. 2002; Navarro-Mayoral et al. 2020; Fernandez-Gonzalez et al. 2021). It is known that the temperature-induced changes in amphipod metabolic rates can lead to important physiological constraints to reproduction and determine of life history patterns of amphipods (Sainte-Marie 1991). For example, under laboratory conditions, Gammarus locusta showed faster individual growth, anticipated age at maturity, and higher population growth rate at a temperature of 20ºC compared to 15ºC (Neuparth et al. 2002). In our case, the difference between the minimum and maximum temperatures was 3.9ºC and controlled studies would be necessary to know exactly the direct relationship between temperature and the abundances of W. minuta gen. nov., sp. nov. Of course, temperature is typically correlated with a range of environmental and ecological processes. For example, an interesting observation was that the highest abundances of W. minuta gen. nov., sp. nov., coincided with a high amount of mucus generated by the colonies (personal observations). Mucus production is a strategy adopted by corals to obtain protection against sedimentation, which also plays a role in the food chain of habitats generated by corals (Galil 1987; Vytopil and Willis 2001; Wee et al. 2019; Fraser et al. 2020). Thus, mucus can catch detritus and phytoplankton that facilitate the feeding of amphipods, as observed in decapod taxa that feed on particles trapped by coral mucus (Galil 1987). Most of the species of the Stenothoid family are detritivorous and/or carnivorous species (Krapp-Schickel 1993; Moore et al. 1994; Sano et al. 2003; Tandberg et al. 2010; Auster et al. 2011; Vázquez-Luis et al. 2013; Guerra-García et al. 2014; Sedano et al. 2020), and W. minuta gen. nov., sp. nov., could take advantage of A. wollastoni secretions for feeding. Therefore, the amount of food available in the form of mucus could also be playing a role in the abundance pattern of amphipods, and specially of W. minuta gen. nov., sp. nov. However, more studies are needed to clarify the biological associations between deep-sea corals and epifauna, particularly focusing on the relationship between this black coral and this new genus and species of amphipod that dominated the associated epifauna community over time. Overall, this study supports that black coral forests in mesophotic depths have enormous potential for overlooked biodiversity and that much of the associated diversity remains undescribed.

Conclusions

We here reported patterns in composition and abundances of epifauna associated with Antipathella wollastoni over time, showing that amphipods were the dominant group. Our results confirmed that marine animal forests host an important, but somehow unknown, associated biodiversity. The new genus and species of the Stenothoidae family seem to be coral specific, displaying a direct relationship with the availability of habitat generated by the coral host. Our study reinforces the necessity to increase research effort in animal forest to improve management and conservation of these key habitats.