Our knowledge of the diet of wild Octopus vulgaris is restricted to the first two weeks of its planktonic phase when they are selective hunters found in the coastal area. These small paralarvae, bearing only three suckers per arm, are transported by oceanic currents from the coast towards offshore waters where they complete the planktonic phase in two months. Herein, we have investigated the trophic ecology of O. vulgaris paralarvae as they drift from the coast into the ocean of the Iberian Canary Current (ICC) eastern boundary upwelling system, to evaluate if their specialist trophic behaviour is held throughout the planktonic phase. Paralarvae (n = 100) were collected in Northwest Spain (n = 5, three suckers per arm), across the Northwest Iberian Peninsula (n = 60, three to five suckers per arm) and off the west coast of Morocco (n = 35, three to 15 suckers per arm). Using high-throughput amplicon sequencing up to 90 different prey was identified in 95 paralarvae. Diet varied along the ICC, with the most discriminant groups being crab larvae and siphonophores in the northern part of the ICC and crab larvae and cnidarians in the south. Ontogenetic changes in the diet were detected between coastal and oceanic communities, evidenced by the decreasing contribution of coastal meroplankton and an increase in holoplankton, including siphonophores, pteropods and krill. Averaged trophic niche breadth values ranged from 0.16 to 0.31, thus suggesting that O. vulgaris paralarvae are specialist predators throughout their planktonic phase, a finding that has important implications for the aquaculture of this species.
Article
Trophic ecology of Octopus vulgaris paralarvae along the Iberian Canary current eastern boundary upwelling system
https://doi.org/10.21203/rs.3.rs-2187875/v1
This work is licensed under a CC BY 4.0 License
Journal Publication
published 30 May, 2023
You are reading this latest preprint version
Octopus vulgaris
paralarvae
diet
COI
next generation sequencing
eastern boundary upwelling system
zooplankton
Understanding trophic links in the early life stages of cephalopods (e.g. octopus, squid cuttlefishes) is challenging for two primary reasons. Firstly, early life stage cephalopods are not abundant in the zooplankton community and secondly, because they quickly break down their prey. There are two modes of ingestion depending on the main prey: fish and crustaceans. Fish larvae are mechanically fragmented using the beak and radula and ingested in small pieces 1,2, whereas crustacean prey is digested using a complex array of enzymes 3, specifically evolved to remove the flesh from their exo- or endoskeletons. The beak and radula are then used to suck up the predigested prey leaving empty exoskeletons 4, and small pieces of prey are occasionally ingested 5. These strategies of ingestion make studying stomach contents difficult. Thus, there are only a few reports on the diet of wild octopod paralarvae. Prey fragments (n = 44) were found in 32 out of 98 octopod paralarvae collected from the eastern Gulf of Mexico1 of the families Amphitretidae (Japetella diaphana), Argonautidae (Argonauta argo) and Octopodidae (Octopus sp. 2, Macrotritopus defilippi and Scaeurgus unicirrhus). The major prey types were euphausiids (49%), fishes (14%), non-cephalopod molluscs (11%), chaetognaths (9%), copepods (7%) and other crustaceans including ostracods, decapods and hyperiid amphipods (11%) 1.
The common octopus or Octopus vulgaris have been observed in captivity to be active hunters immediately after hatching, armed with only three suckers on each arm 6. Prey has been detected via analysis of DNA from partially digested and liquefied contents of captive O. vulgaris hatchings 7. Since the visual analysis of the diet was precluded by the presence of amorphous material inside of the crop and stomach, the first molecular attempt to unveil the diet of wild O. vulgaris paralarvae revealed up to 20 different prey species, showing that early hatchlings preyed upon 12 families of decapods, two fish families and one family of euphausiid in the coastal area of Ría de Vigo, Northwest Spain 8. The prevalence of decapod families in the diet compared with the low abundance of these families in the zooplankton coastal communities9 suggested that O. vulgaris hatchlings were specialist predators, as shown by the positive values of the linear indices of food selection and the low values of trophic niche breadth (0.04–0.37, being 1 a generalist and 0 a specialist predator)10. Further molecular studies also carried out on early hatchings with three arm suckers over the continental shelf of northwest Spain, revealed up to 46 different prey species: mostly decapod larvae (16 families, 24 species), four families of copepods and ophiuroids, three families of bivalves, two families of cladocerans and cnidarians as well as an amphipod and a chaetognath 11.
Our current understanding of the trophic ecology of Octopus paralarvae is restricted to early stages with only three suckers on each arm 8,11. During the planktonic phase additional suckers are added until there is a total of 23–25 suckers on each arm, after which they begin the settlement phase to become benthic juveniles and adults 12. Largely all paralarvae collected to date over the continental shelves of the Northeastern Atlantic (n = 3254) had three suckers 10, 13–17. The only exceptions correspond to some O. vulgaris paralarvae collected in zooplankton tows in the English Channel 18,19, which were later used for histological studies20. Furthermore, four out of the 254 O. vulgaris paralarvae collected at stations beyond the continental shelf of Portugal16 with bottom depths ranging from 287 to 2281 m had more than three suckers (A. Moreno, pers. comm.). More recently, 49 out of 134 O. vulgaris paralarvae with arm suckers ranging from four to 15 were also collected beyond the continental shelf (> 200 m depth) during the oceanographic surveys CAIBEX-I and III, off Northwest Iberian Peninsula and West Morocco 21.
Cephalopod paralarvae drift with zooplankton communities during the first months of their life cycle and adjust their vertical behaviour to maximise survival and recruitment 22–25. Three different life strategies have been suggested for the planktonic phase of the different cephalopod paralarvae present along the Iberian Canary current eastern boundary upwelling system (ICC): coastal, oceanic and coastal-oceanic strategies 21. The coastal strategy is observed in the cephalopod families Loliginidae, Sepiolidae and some species of Ommastrephidae (including Todaropsis eblanae and Illex coindetii), where the paralarvae hatch and are retained over the continental shelf (< 200 m depth) being absent from oceanic waters. The oceanic strategy is found in paralarvae that hatch and develop in the open ocean, followed by oceanic squids, octopods and sepiolids. The coastal-oceanic strategy, which has only been observed in O. vulgaris paralarvae, requires hatching over the continental shelf and posterior drift into the open ocean (> 200 m depth), where they undergo their planktonic phase.
Such unique lifestyle, where O. vulgaris paralarvae are transported far from the coast by oceanic currents, together with the paucity of different developmental stages collected in the plankton, restricts the current knowledge of their planktonic phase beyond three suckers to the specimens collected in the English Channel 18, and off Northwest Iberian Peninsula and West Morocco, in CAIBEX-I and III respectively 21. Age estimation based on beak growth increments of paralarvae collected in these surveys revealed that the O. vulgaris paralarvae captured in the oceanic realm with more than three suckers were older (seven to 28 growth rings; ~ seven to 40 days old) than those collected in the coastal areas with only three suckers (0 to eight growth rings; ~ 0 to 10 days old) 26, thus supporting that O. vulgaris paralarvae develop their planktonic phase far from the coastal area, actively hunting in the oceanic zooplankton communities. Microbiome analysis of those paralarvae showed a significant increase in bacterial diversity correlated with an increase in sucker number, which was hypothesized to be partly because of the change in the prokaryotic assemblage along the coast-ocean gradient 27, and partly related with the ingestion of diverse prey in the oceanic areas 28.
However, the diet of the paralarvae collected in the oceanic realm have not yet been characterised and filling that gap will help to understand their trophic behaviour throughout their development in the plankton. Such knowledge would be of paramount importance for the aquaculture of this species, since the main bottleneck in aquaculture of O. vulgaris is the survival of the paralarvae owed to nutritional deficiencies 29,30. Accordingly, in this work we examined the diet of different developmental stages of O. vulgaris paralarvae found in zooplankton samples along the ICC to (i) evaluate the differences in the diet in three contrasting upwelling regions, and (ii) to evaluate their trophic behaviour as they travel along the coastal-oceanic gradient during the planktonic phase.
Prey composition and abundance
Digestive tract dissections showed that 91 of 100 O. vulgaris paralarvae had empty crops and stomachs with partially digested prey, none of which could be identified. Despite this, only five paralarvae had no prey DNA in their digestive tract, four collected in Morocco (with three, four (n = 2) and six suckers per arm), and one collected off the coast of Portugal with five suckers per arm. High-throughput amplicon sequencing generated 2,923,539 high-quality reads, with an average of 30,057 reads per paralarvae (ranging from 22 to 339,756). Classification of the reads showed that 2,745,495 (93.9%) corresponded to Octopus vulgaris, 56,180 (1.92%) to contamination, and 121,835 (4.17%) were identified as prey/ASVs. Prey was detected in 95 out of 100 O. vulgaris paralarvae, with 1,219 prey reads per paralarva on average (ranging from 1 to 29,677). The negative sample revealed 29 reads that corresponded to Octopus vulgaris.
Analysis of prey reads/ASVs revealed 90 different taxa, 32 of which were exclusively detected in the coastal area, 39 in the ocean, and 19 in both environments (Fig. 1). An average of 3.6 prey were detected in each paralarva (ranging from one to 13 prey). The most abundant prey groups based on read frequency were crabs (64%, 77,945 reads), pteropods (15.1%, 18,424 reads), polychaetes (5.9%, 7,190 reads), ostracods (5.6%, 6,869 reads), euphausiids (3.8%, 4,594 reads) and siphonophores (2.2%, 2,700 reads) (Fig. 2a). The most frequently observed prey based on prevalence in O. vulgaris paralarvae were crabs (71.6%, 68 paralarvae), siphonophores (35.8%, 34 paralarvae), cnidarians (30.5%, 29 paralarvae), copepods (29.5%, 28 paralarvae), euphausiids and pteropods (15.8%, 15 paralarvae), fish (10.5%, ten paralarvae) and polychaetes and cladocerans (7.4%, seven paralarvae) (Fig. 2b).
The most abundant prey reads were the crabs Liocarcinus marmoreus (21.2% of prey reads), L. navigator (11%), Lophozozymus incisus (9.3%) and Goneplax rhomboides (7.7%), and the pteropod Cavolinia inflexa (15.1%), which together accounted for 64.4% of total prey reds (Suppl. Table S1). The most common prey was the crab Goneplax rhomboides (detected in 26.3% of paralarvae), the siphonophore Muggiaea atlantica (24.2% paralarvae), the crabs Liocarcinus navigator and Lophozozymus incisus (22.1% paralarvae) and the cnidarian Clytia sp. (17.9% paralarvae). Contamination reads included fungus, humans, vertebrates, insects, and some marine species that were analysed in the laboratory like lobsters (see supplementary files for details). These reads were removed from downstream analyses to avoid confusion.
Diet along the Iberian Canary current upwelling system (ICC)
PERMANOVA test showed significant differences in the diet in the three areas sampled in the ICC (p < 0.001). Figure 3 shows data from 95 O. vulgaris paralarvae arranged in multidimensional space according to sampling areas, which are correlated with the PCO 2 axis (Fig. 3a). PCO 1 axis is strongly related to the locations where the paralarvae were collected, with positive values strongly correlated with paralarvae collected in the coastal area (marked as the coast, Fig. 3b) or in oceanic areas affected by the advection of upwelled coastal waters and the upwelling filament (indicated as coast-ocean); negative values are represented by samples collected in the ocean (Fig. 3b). Given that the diets are markedly different in the different areas of the ICC, a detailed description is provided per region.
Northwest Iberian Peninsula
In this section, the five paralarvae collected in the Ría de Vigo were analysed with the 60 samples from the survey CAIBEX-I. A total of 1,850,867 merged sequencing reads were retained for analysis, which were classified 348.065 / 1,417,606 as Octopus, 8,244 / 70,585 prey, and 28 / 6,339 contamination reads, from the Ría de Vigo / CAIBEX-I respectively. An average of 3.8 (from two to seven, Ría de Vigo) and four (from one to 13, CAIBEX-I) prey were detected in all octopus paralarvae except one sample with no prey DNA. Overall, 59 prey were identified: 35 crustaceans (15 crabs, eight copepods, three cladocerans, two hermit crabs, two galatheids, one porcellanid, two euphausiids, one hyperiid and one ostracod), seven fishes, five siphonophores, four cnidarians, four molluscs (one pteropod, two cephalopods and one bivalve), two polychaetes, one ophiuroid and a chaetognath (Fig. 1).
In terms of frequency of the different prey, the most common prey detected in the paralarvae collected from the Ría de Vigo were polychaetes (87.1% of prey reads, detected in all the five paralarvae), followed by siphonophores and chaetognaths (0.3 and 0.1% of prey reads, from three of five paralarvae), copepods and cnidarians (0.5 and 0.2% of prey reads, from two of five paralarvae) and crabs and echinoderms (9.7 and 2.1% of prey reads, from one of five paralarvae) (Fig. 2b). Similarly, the most common prey detected in the paralarvae collected in CAIBEX-I were crabs (69.3% of prey reads, 49/59 paralarvae), siphonophores (0.8% of prey reads, 27/59 paralarvae), cnidarians (0.5% of prey reads, 18/59 paralarvae), copepods (0.3% of prey reads, 17/59 paralarvae), pteropods (26.1% of prey reads, 13/59 paralarvae), fish (0.6% of prey reads, 7/59 paralarvae), cladocerans and euphausiids (1.3 and 0.5%, 6/59 paralarvae) and hyperiid amphipods (0.02% of prey reads, 4/59 paralarvae) (Suppl. Table S1; Fig. 2a).
The diet of the paralarvae was significantly different according to the location collected (coast vs ocean, p < 0.001; strongly correlated with PCO axis 1 in Fig. 3b), the strata (p < 0.001) and marginally significant for the factor day-night (p = 0.054). Pairwise analysis of the strata showed significant differences for the pairs S-M (p = 0.01) and S-DSL (p = 0.004) and no difference between the deeper strata M-DSL (p = 0.714). PERMANOVA results for the nested analysis of the location and strata showed marginally significant results for the interaction (p = 0.059). Pairwise analysis within the levels of the factor location x strata levels revealed significant diet differences between the two strata sampled in the coastal area (p = 0.044) and no significant differences between the three strata in the ocean.
The most discriminant prey that defined the two locations sampled off Northwest Iberian Peninsula (coast and ocean) are represented as pie charts in Fig. 4. The crabs Liocarcinus navigator and Goneplax rhomboides were the prey contributing the most to the diets of coastal paralarvae, whereas Lophozozymus incisus and Liocarcinus corrugatus were the most discriminant contributors in the ocean (Fig. 4). Seven species are listed in the coastal area accounting for 91% of the variability observed and six species in the ocean account for 91.6%. The marked differences between the two locations sampled is evidenced by the average dissimilarity of the diets on the coast (85.4%) and ocean (88.4%, Fig. 4). The ten most discriminant organisms account for 59.1% of the differences observed between the coastal and oceanic regions, which are markedly different as indicated by the 95.6% of dissimilarity observed between them. DistLM results showed that 84.5% of the diet of the paralarvae collected along the Northwest Iberian Peninsula is represented by crabs 56.4% > siphonophores 11.9% > copepods 9.5% > cnidarians 4% and pteropods 2.8% (Table 2).
Cape Ghir, W Morocco
A total of 1,072,643 reads were obtained from the 35 paralarvae analysed from CAIBEX-III, of which 979,824 corresponded to Octopus, 43,006 to prey, and 49,813 contamination. An average of 2.9 prey were detected on each paralarva (from one to 13 different prey), except for four that had no prey reads. Overall, 45 prey were identified: 29 crustaceans (12 crabs, eight euphausiids, seven copepods, one cladoceran and one ostracod), three fishes, five siphonophores, four cnidarians, three molluscs (one pteropod and two cephalopods) and one chaetognatha (Fig. 1). The most common prey detected in the paralarvae were crabs (65.6% of prey reads, 18/31 paralarvae), euphausiids, cnidarians and copepods (9.9, 2.4, and 0.4% of prey reads, 9/31 paralarvae), siphonophores (4.8% of prey reads, 4/31 paralarvae), and fish and cephalopods (0.2% of prey reads, 3/31 paralarvae) (Fig. 2b, Suppl. Table S1).
The diet of the paralarvae collected off Cape Ghir, Morocco, during CAIBEX-III showed significant differences according to the location sampled (p = 0.018, Fig. 3b) and no differences according to strata (p = 0.089) and day-night (p = 0.148) factors. Pairwise analysis of the factor location revealed significant differences in diets for the pairs C-O (p = 0.01) and C-C ~ O (p = 0.034). SIMPER analysis revealed that only three to five prey species account for more than 90% (between 92.6–95.6%) of the variability detected within the different locations sampled in Cape Ghir, Morocco (Fig. 4). Contrary to the Northwest Iberian Peninsula, the most discriminant prey was not the most abundant in each location (pie charts, Fig. 4). The diets detected at the three locations are more dissimilar as the paralarvae drift offshore from the coastal area (79.7%) within the upwelling filaments and its periphery (C-O samples, 93.5%), into the open ocean (95.6%).
The ten most discriminant prey species listed for comparing coastal (C) and oceanic (O) locations accounts for 65.7% of the total variation observed. In contrast, those listed for the comparison between coastal-oceanic (C-O) and oceanic (O) locations accounted for 52.9% (Fig. 4). The diets of the paralarvae at the three locations had very little in common because the average dissimilarity between locations ranged from 96.3 to 96.7%, except for the cnidarian Liriope tetraphylla. DistLM results showed that 90% of the diet of the paralarvae collected off Cape Ghir, Morocco is represented by crabs 36.6% > cnidarians 23% > copepods 16.7% > euphausiids 9.8% and ostracods 3.9% (Table 2).
Ontogenetic changes in diet: trophic behaviour
The number of suckers per arm was used to investigate ontogenetic changes in the diet (represented in Fig. 5 as N below the column charts). PERMANOVA results showed significant differences for the NW Iberian Peninsula (Ría de Vigo and CAIBEX-I grouped) (p < 0.001). Pairwise analysis showed significant differences in the diet of paralarvae between three and four suckers (p = 0.001), three and five (p = 0.004) and between four and five suckers (p = 0.001). SIMPER analysis within the different groups of paralarvae revealed a decrease in prey diversity as the paralarvae grow, clearly seen in the pie charts and the number of different species detected (S) in Fig. 5. Nine species are responsible for 90.9% of the diet of paralarvae with three suckers collected in the coast and the ocean, whereas the diet of the paralarvae with four suckers collected in the ocean is represented by six preys (91.8% of the variability), and only two preys represent up to 91% of the diet of paralarvae with five suckers. The ten most discriminant prey responsible for the differences between the diets is shown in Table 3, where the high dissimilarities obtained (from 88.1 to 94.6%) underline the large differences in the prey selection detected.
Off Cape Ghir, Morocco, PERMANOVA results also showed statistical differences in the diets according to the factor suckers (p = 0.041). Since the representation of paralarvae with different suckers was not even (see N under the column chart of Fig. 5), the following three groups were created for PERMANOVA tests and identification of discriminant species: three suckers (n = 19), four to five suckers (n = 7) and more than six suckers (n = 9). Pairwise analysis showed that the only significant comparison was between three and more than six suckers (p = 0.008). SIMPER analysis within the different groups of paralarvae showed that three species were responsible for 92.8% of the diet of paralarvae with three suckers (collected both on the coast and in the ocean). The diet of the paralarvae with four to five suckers found in the ocean is represented by five prey, accounting for 91.8% of the variability. Lastly, 96.2% of the diet of paralarvae with more than six suckers only collected in the ocean is explained by only two prey species (Fig. 5, pie charts). SIMPER analysis revealed even higher dissimilarities (from 91.9 to 97.1%, Table 3) manifesting the differences in prey selection detected. Overall, certain groups like porcellanid crabs, hermit crabs, hyperiids, ophiuroids and polychaetes were exclusively consumed in the coastal area by recently hatched paralarvae. Contrarily, cladocerans, ostracods and cephalopods were exclusively detected in paralarvae collected in the ocean with more than four suckers.
Natural mesozooplankton abundances of the prey groups ingested can be found in the supplementary table S2. The linear index of food selection (L) was calculated for every prey group (Suppl. table S3), thus highlighting a preference for brachyuran larvae, cnidarians, siphonophores and pteropods. Negative values were obtained for euphausiids, ophiuroids and the copepod families Centropagidae and Temoridae, thus indicating ingestion of prey groups that were abundant in the samples of Ría de Vigo (ophiuroids), NW Iberian peninsula (euphausiids and Centropagidae) and W Morocco (Temoridae and Centropagidae). Trophic niche breadth calculated with the Czekanowski´s Index (CI) ranged from 0.67 to 0.06 (Fig. 6). No statistical differences were detected when paralarvae were grouped by sucker number (with averaged values of 0.21 ± 0.14 for 3 suckers, 0.29 ± 0.15 for 4–5 suckers and 0.22 ± 0.07 for > 6 suckers). This evidence shows that most O. vulgaris paralarvae are very selective hunters, feeding on prey groups that are not abundant in the plankton, although some of them won't lose an opportunity to catch prey that might be abundant (CI > 0.5, like euphausiids or copepods).
The first insights into the diet of O. vulgaris paralarvae during their planktonic phase beyond the continental shelf in three different sub-regions of the ICC are presented in this study. Our data reveal marked differences in their diet within and between different sub-regions and ontogenic changes throughout development. Such changes in the diet are related to changes in zooplankton communities along the coastal-ocean gradient, from a coastal embayment like the Ría de Vigo 9, the continental shelf of the Northwest Iberian Peninsula 31–34, to the subtropical waters off Morocco 35–38. These environments are markedly different in their physico-chemical (temperature, salinity, transparency) and biological properties (primary production, zooplankton abundance and diversity), creating strong biological and physical clines that paralarvae of O. vulgaris are subject to.
Octopus vulgaris paralarvae hatch in coastal areas and are transported by upwelling filaments hundreds of kilometres offshore to complete their planktonic phase 21. As the paralarvae drift with the currents from the coast into the ocean, they are surrounded by zooplankton communities that gradually become more diverse but less abundant. In the oceanic environment beyond the continental shelf, 58 prey species were detected (39 being exclusive of this environment, Suppl. Table S1), including pteropods, ostracods, and oceanic cephalopods that were not detected before in gut contents. The prevalent faunal composition in the ocean and upwelled waters of the ICC, shows a clear dominance of small and medium-size copepods, accounting for 82.5–87% during the annual cycle 39. Other groups, such as ostracods and appendiculariaceans, can reach 8% and 12% in the zooplankton communities, respectively, while cladocerans, pteropods and euphausiid larvae did not represent more than 1%. Despite being the dominant taxa in the zooplankton, copepods were not the most important group in the diet of O. vulgaris paralarvae, altogether representing between 9.5% and 16.7% of the diet in the Northwest Iberian Peninsula and Morocco, respectively. These data, together with the trophic niche breadth values shown in Fig. 6, suggest that the paralarvae do not capture prey opportunistically based on their abundance (as expected for a generalist predator) but on other prey characteristics like movement or nutritional aspects. Copepods are extremely fast swimmers with erratic movements that are difficult to predict and might represent a difficult prey for O. vulgaris. Copepod capture is a skill acquired in an experience-dependent manner throughout the planktonic stage, as observed with Loligo opalescens in captivity 40. Cephalopod paralarvae that have a pair of feeding tentacles, such as loliginids or ommastrephids 1,2,11, 41–43, are better copepod hunters than O. vulgaris paralarvae which do not possess feeding tentacles. Herein, we observed that the paralarvae collected in the ocean consumed significantly more copepods (specifically, Centropages typicus and Paracalanus gracilis) than those in the coastal area (Table 3). Since the paralarvae found in the open ocean are older than those found near the coast 26, this increase in copepod predation suggests that the paralarvae may learn to hunt on these elusive prey as they develop.
The diets analysed showed that the diversity of prey is slightly higher in the oceanic area (57 different prey species detected in 66 paralarvae) than in the coastal area (51 different prey species detected in 34 paralarvae). However, this pattern is not consistent along the ICC: in the coastal and oceanic areas of the NW Iberian Peninsula, 44 and 25 prey species were detected in 25 and 40 paralarvae, respectively. Contrarily, 17 and 41 prey species were detected in nine and 26 paralarvae of the coastal and oceanic areas in Western Morocco. The overall prey diversity detected in the coastal area is higher than in previous studies, where 46 different prey taxa were detected in 56 out of 64 O. vulgaris with three suckers per arm with the COI gene 11. The groups were very similar, except for polychaetes and cephalopods, which were detected for the first time in this work. It is important to mention the absence of decapod families like Processidae, Alpheidae, Crangonidae and Thalassinidae, formerly detected in other dietary studies in O. vulgaris 8,11 but not present in this work. One possibility is that the modified primers, which included inosines to bind to any of the four nucleotides, rather than degenerated bases 44,45, may have prevented the amplification of those families.
The high diversity of euphausiids and siphonophores detected as prey in the ocean was remarkable, with siphonophores and cnidarians being the second and third more frequent prey ingested by O. vulgaris paralarvae (Fig. 2b). The contribution of gelatinous zooplankton like cnidarians and siphonophores is significant inside the upwelling filaments, and their abundance gradually decreases as these mesoscale structures venture into the open ocean 46. In coastal areas, gelatinous zooplankton have also been detected in O. vulgaris with three suckers 11 and loliginid paralarvae, like Alloteuthis media and Loligo vulgaris 11,41. Different facts might help to explain the predation of these organisms. These organisms are protein-rich (18.3 ± 7.8% of dry mass 47), thus providing an easily digestible source of soluble nutrients. Furthermore, cnidarians and siphonophores are predators of copepods 48, and the paralarvae may benefit from their foraging because when they ingest gelatinous plankton, they would also ingest their copepod prey. Finally, compared with highly motile prey like copepods, gelatinous plankton are relatively large slow drifters with predictable movements that likely make an easy capture by the paralarvae, like that of fish larvae 49 or crustaceans 50,51. Once captured, a large gelatinous prey would be easier to subdue than a more vigorous prey, saving energy. The energy expenditure may be further reduced by relying on the buoyancy of the prey rather than actively swimming. Another important aspect of ingesting gelatinous prey is that transparency plays a key role in the open ocean for epipelagic organisms. This strategy is the most common solution to the dilemma of having nowhere to hide, a trait present in almost all marine phyla (e.g. salpids, crustaceans, leptocephalus, heteropods, polychaetes, chaetognaths or ctenophores, among others) inhabiting the epipelagic realm. Octopus paralarvae are completely transparent except for membranes enclosing the eyes and digestive gland that are covered by reflective cells called iridophores. This reflective surface acts as an ambient light reflector, concealing the opaque body organs 12. The ingestion of transparent prey will make the paralarvae less conspicuous and more difficult to target by predators. This strategy of transparent predators ingesting transparent prey has also been suggested for the European eel larvae 52 and lobster larvae 47,50.
Crabs were the most important prey in all octopus analysed, even though their abundances in the ocean were an order of magnitude less than in shelf communities 9,10,37. This supports the hypothesis that planktonic O. vulgaris are specialist predators, specifically targeting prey that is rare in the zooplankton communities. Consistent with this hypothesis, another interesting group detected is cephalopods, whose abundance is lower than decapod larvae. To our knowledge, it is the first time that the squid, Alloteuthis media has been detected as prey in O. vulgaris paralarvae; only one paralarva in the coastal region of the Iberian Peninsula was positive for A. media, where it is the main loliginid paralarvae present in the zooplankton 53,54. Similarly, Alloteuthis subulata, O. vulgaris and the octopod Eledone cirrhosa have been detected as prey of Loligo vulgaris paralarvae collected in the Northwest Iberian Peninsula 41. The oceanic squid Brachioteuthis was also detected in three O. vulgaris paralarvae collected in the oceanic area off Morocco, where Brachioteuthis was one of the most abundant cephalopod paralarvae sampled 21. The methodology applied to characterise the diet does not allow the detection of predation in conspecifics, but it may also constitute a potential prey since cannibalism has been detected in captivity in Octopus species 55,56
Other groups detected in the digestive tracts of the paralarvae, like rotifers, sponges or phytoplankton, might result from secondary predation 57. Octopus paralarvae are visual predators that attack large prey compared with their size 12, and these groups are too small to be captured by the paralarvae. It is more likely that these were ingested by other prey, like copepods or decapod larvae, shortly before these were captured by the paralarvae. The longer a prey is inside a predator's gut, the harder it is to detect it because of DNA degradation. If the prey ingested by the paralarvae also ingested a prey shortly before being captured, it is possible to detect the DNA of both organisms inside the predator 58. Molecular techniques are powerful tools to unravel trophic links even in small organisms like O. vulgaris paralarvae. When dissections were carried out, only 9% of the paralarvae had amorphous contents inside the crop and stomach. However, prey DNA was successfully detected in 95%, highlighting the importance of molecular studies even when no prey contents are present.
Current aquaculture practices of octopus suffer from high levels of mortality, which may in part be due to sub-optimal diets in captivity. These diets are based primarily on Artemia spp. (commonly used in aquaculture) that might be enriched with commercial products or supplemented with other organisms like crustacean zoeae, copepods or amphipods (reviewed in Vidal et al., 2014). The increasing diversity of bacterial families detected in O. vulgaris has been related to increased prey diversity as they develop in the open ocean (Roura et al., 2017). This work shows that O. vulgaris paralarvae ingest numerous prey during their development, each of which has its microbiome mixed with that of the paralarvae; therefore, it is expected that the microbiome of the paralarvae will increase in complexity during the planktonic phase. This natural prey diversity heavily contrasts with the mono-diets commonly used in aquaculture, which offer a less enriched nutrition and a limited microbial community. It is known that recently hatched octopus paralarvae possess a diverse microbial community. However, diets based on Artemia have a profound impact on the microbiota after a few days, altering the nutritional landscape of the gut and leading to the expansion of pathogenic populations 28. This has also been observed in other organisms reared in captivity, such as olive flounder 59 or white shrimp 60, and suggests that resilient paralarvae would be those that harbour a diverse microbial community because of a diverse diet. The O. vulgaris paralarvae studied in this work provide invaluable data about their biology during early development amongst diverse prey and will be of great interest in addressing the high levels of mortality in captivity that are currently constraining octopus aquaculture.
Oceanographic context
The Iberian–Canary current eastern boundary upwelling system (ICC) constitutes one of the four main eastern boundary upwelling systems of the world’s oceans. The ICC covers the latitudinal range between 12–43º N and can be broadly divided into five sub-regions according to their biogeographical characteristics 61. This work is centred in three out of these five sub-regions: the Galician (SR1), Portuguese (SR2) and the Moroccan sub-regions (SR4). The western Iberian Peninsula sub-region is the northernmost part of the ICC and includes the Galician and Portuguese sub-regions that are characterized by seasonal upwelling. During spring and summer (from March–April to September–October), north-easterly winds predominate in the Iberian basin, and mesoscale upwelling filaments develop intermittently in association with irregularities in the coastline like capes 62–64. On the other hand, the Moroccan sub-region experiences year-round upwelling that vary seasonally, with extended upwelling filaments, absence of freshwater inputs and massive dust inputs from the adjacent Sahara Desert 61,65.
The seasonal upwelling filament of Cape Silleiro (Northwest Iberian Peninsula, 42–43º N) and the permanent one located off Cape Ghir, Morocco (30–31º N) export 4x108 and 31x108 kg/year of organic carbon to the adjacent shelf representing 20 and 60% of the total phytoplankton primary production in these regions, respectively 66. These filaments break the strongly stratified oceanic waters creating clear community gradients that contrast with the open ocean communities 39. Zooplankton organisms have developed multiple strategies to thrive in such dynamic ecosystems, mostly related to changes in their vertical position. Most larvae cannot swim against horizontal currents but can swim faster than vertical currents to maintain a preferred depth, thus limiting their cross-shore dispersal 67. Studies of larvae dispersal in the ICC suggest that vertical behaviour is one of the main biological mechanisms for crustacean larval retention over the shelf 32,33, fish larvae retention 68 or fish larvae dispersal 69,70. In this work, we sampled three different areas of the ICC during the upwelling season to understand the transference of zooplankton between the continental shelf and the adjacent ocean and, specifically, evaluate the dispersal of paralarvae by these upwelling filaments.
Zooplankton sampling
Plankton samples were collected during the multidisciplinary project “Canaries-Iberian Marine Ecosystem Exchanges (CAIBEX)” off the coast of Northwest Iberian Peninsula (CAIBEX-I, June 7th − 24th ) and Cape Ghir, Morocco (CAIBEX-III, August 16th – September 5th ) in 2009 onboard the research vessel (RV) "Sarmiento de Gamboa” (Fig. 7a). During CAIBEX-I, two lagrangian experiments that followed two different water masses identified with drifting buoys were carried out: (i) an upwelling-relaxation event in the open ocean in front of Portugal (L1: July 10–13 inclusive Fig. 7b, samples S1-S7 in blue Fig. 7c); and (ii) an upwelling event with alongshore transport along the continental shelf of Spain and Portugal, that can be identified by the sea surface temperature (SST) obtained at the end of the experiment with the violet colours representing the cold upwelled waters (L2: July 17–20 inclusive, Fig. 7b, samples S13-S20 in green, Fig. 7c). Between L1 and L2, zooplankton samples were collected following a coastal-ocean gradient off the Portuguese coast (S8-S10) to observe the changes between these two environments and samples from the continental shelf of Galicia (S11 and S12). In contrast, during CAIBEX-III, a single lagrangian experiment was carried out off the coast of Cape Ghir, Morocco (31ºN) following a strong upwelling filament 71,72 forced by the prevailing trade winds from the coast into open ocean (L3: August 23–31 inclusive Fig. 7d, samples in violet, Fig. 7e). The magnitude of the cold upwelling filament was visible from the SST obtained from satellite data, as shown in Fig. 7d. Samples were also collected over the continental shelf (in green, Fig. 7e), in an area affected by the upwelled water over the continental slope (in orange, Fig. 7e), and in the open ocean (in blue, Fig. 7e).
Mesozooplankton samples were collected day and night with two 750 mm diameter bongo nets equipped with 375 µm mesh and a mechanical flow meter. Three double-oblique towings were carried out at a ship speed of 2.5 knots over the continental slope (> 200 m depth): (i) at the deep scattering layer (DSL: 500 m), (ii) at 100 m, and (iii) at the surface (0–5 m). Over the continental shelf (< 200 m), only two double oblique towings were collected at 100 m (when sea-bottom was < 100 m, otherwise 10 m above it) and at the surface (0–5 m). The bongo net was first lowered to the desired depth, towed for 30 min and subsequently hauled at 0.5 m s–1. The net was recovered, cleaned on board, and placed back into the sea for the next towing. Plankton samples were fixed with 96% ethanol and stored at − 20◦C to facilitate DNA preservation. Mesozooplankton abundance was estimated for each sample, by subsampling the original sample into an amount suitable for examination using a Folsom splitter. The subsample was made up to 300 ml, several aliquots of 3 ml were obtained with a Stempel pipette, then identified and counted until at least 500 individuals were enumerated. Organisms were identified under a binocular (Nikon SMZ800) or inverted microscope (Nikon Eclipse TS100) to the lower taxonomic level possible.
All cephalopod paralarvae were separated from the zooplankton samples in the laboratory and stored individually in 70% ethanol at − 20◦C. In total, 134 O. vulgaris paralarvae were collected during CAIBEX-I (n = 99 specimens) and CAIBEX-III (n = 35 specimens), and the number of suckers on the arms was counted. Of these, 60 paralarvae were chosen from CAIBEX-I (ranging from three to five suckers per arm) and 35 from CAIBEX-III (ranging from three to 15 suckers per arm) to study their diet (Table 1). Separately, five recently hatched O. vulgaris paralarvae were collected at night at the surface of the Ría de Vigo (September 22nd 2010) onboard the RV “Mytilus” and were included in the diet analysis. More detail of the sampling carried out in the Ría de Vigo in 2010 can be found in 10.
Library preparation and sequencing
The digestive tract of O. vulgaris paralarvae, which included the oesophagus, crop, stomach, caecum, digestive gland, and intestine was dissected using entomological needles that were rinsed in ethanol and burned between every dissection. DNA was extracted using the QIAGEN DNeasy Blood and Tissue Kit, according to the manufacturer’s instructions. A slight modification was made at the final elution stage of the extraction; the elution was repeated twice using two 20 µL aliquots of 45ºC ultrapure water and stored as a combined 40 µL eluate prior to use. A 300 bp region of the mitochondrial COI gene was amplified with NICO-F and NICO-R primers. The designed primers NICO-F and NICO-R (Suppl. Table S4) are a modification of the degenerated primer mICOIintF 73 and HCO 74, respectively. These primers were modified to add an inosine (I) that complements all four nucleotides 45 to capture a greater fraction of the prey. The primers contained a 5’ tail containing a universal Illumina adapter sequence (italics) to enable multiplex indexing via primer extension PCR (Suppl. Table S4).
PCR reactions (12.5 µL) contained 10–30 ng of template, 6.25 µL of RedTAQ ReadyMix (Sigma-Aldrich), 0.35 µL NICO-F (10 µM), 0.2 µL NICO-R (10 µM) and 0.1 µL MgCl2 (25 nM). Touchdown-PCR cycling conditions consisted of an initial denaturation at 95°C for 3 min followed by ten cycles of denaturation at 95°C for 30 s, annealing at 57°C for 30 s - decreasing 1°C per cycle - and an extension at 72°C for 40 s, then 25 cycles of 95°C for 30 s, annealing at 47°C for 30 s and an extension at 72°C for 40 s, with a final step at 72°C 4 min. PCR products were cleaned with AMPure beads and quantified with a Qubit™ fluorometer. Illumina multiplex indexes were added to the sequences in a second-round PCR before pooling 0.5 ng of each PCR product in a metasample at a final concentration of 2.5 nM. After adding 10% PhiX and diluting to 12.5 pM, the pooled library was sequenced using a v3 2✕300 bp sequencing kit on an Illumina MiSeq. DNA extractions and PCR reactions were carried out in different rooms to avoid cross contamination. Negatives were included for each round of PCR (with miliQ water instead of DNA) and were included in the metasample for sequencing.
Quality filtering and bioinformatic analyses
Quality filtering was carried out following recommendations for Illumina platforms 75. Reads that did not meet the following standards were removed: (i) Phred score below 30 (i.e., one error in 1,000 bases), (ii) less than 75% of target length, (iii) less than three consecutive low-quality calls, (iv) reads with ambiguous calls and (v) reads with less than 5 identical copies. The remaining paired-end reads were merged with PEAR v0.9.4 76 using a 99% sequence similarity threshold (minimum of five sequences per cluster). Merged reads were demultiplexed into individual sample read sets based on their corresponding adapter combination. Reads for which the indexes/primers did not match the expected sequences were discarded. The remaining reads were then filtered against a Kraken (v0.10.4) database to exclude archaeal and viral contamination 77, and the UCHIME algorithm of USEARCH (v 6.0.307) 78 was used to check for chimeric sequences. Consensus sequences were BLASTed against sequences on GenBank, and those with similarities above 80% were recovered. Amplicon sequencing variants (ASVs) corresponding to potential prey were assigned using the following criteria to taxonomical categories: ASVs with an identity higher than 97% were determined at the species level, ASVs between 93 and 97% were assigned to the genus, ASVs between 93 − 90% were assigned to subfamily, and those below 90% were assigned to family 11.
Multivariate analysis of the diet
The software PRIMER6 and PERMANOVA + 79 were used to analyse the diet of O. vulgaris paralarvae. Prey reads/ASVs were log-transformed (x + 1), and these were used to generate a Bray-Curtis resemblance matrix. An unbiased representation of the prey detected in the multidimensional space was visualized using principal coordinate analysis (PCO) plots. Permutational multivariate analysis of variance (PERMANOVA) tests using 9999 permutations were run to test the independent factors described in Table 1: ICC (three levels: Ría de Vigo, Northwest Iberian Peninsula: CAIBEX-I,and Cape Ghir, Morocco: CAIBEX-III), location (three levels: C, coast for all the samples collected over the continental shelf; C ~ O, for the samples collected over the continental slope but under the influence of the upwelling filament or coastal upwelled water; and O, ocean for all the samples collected in the open ocean), strata (three levels; S, surface; M, samples collected < 100 m; and DSL, deep scattering layer for those samples collected < 500 m), day/night (two levels) and suckers (six levels, according to the number of suckers present on each arm). PERMANOVA analysis was performed using the ‘unrestricted permutation of raw data’ option for these independent factors. Furthermore, a two-factor PERMANOVA test was also performed for the factors location and strata, with the factor strata nested within the different locations, using the “permutation of residuals under the full model”.
The prey contributing most to similarities and dissimilarities between the paralarvae collected at different locations of the ICC and the paralarvae grouped by sucker count was determined using the program SIMPER 80. This analysis uses the Bray-Curtis dissimilarity matrix of log-transformed reads (reads + 1) to recognize the main prey consumed in the different groups, their contribution to the total variability observed, and the discriminative power of the main prey driving the differences observed. Within each survey, the contribution of the different prey groups was analysed with distance linear models (DistLM), using an indicator that grouped the prey species (grouping was done at different taxonomic levels). A stepwise selection procedure was chosen using the adjusted R2 as the selection criterion for retaining variables that summarised the diets of the different paralarvae at the different places of the ICC sampled.
Two trophic indices were analysed to evaluate the behavioural changes in trophic selection: the linear index of food selection (L) for the different prey ingested and the Czekanowski´s index (CI) for each paralarva. L ranges from − 1 to + 1, with positive values indicating a preference, negative values indicating avoidance or inaccessibility, and zero values showing random feeding 81. This index is obtained for every prey detected with the formula (L = ri - pi), where “ri” is the relative proportion of prey item i in the gut and “pi” the relative proportion of prey item i in the zooplankton. Furthermore, we examined the trophic niche breadth for every octopus paralarvae using the CI 82. Trophic niche breadth was calculated with the formula (CI = 1–0.5 ∑i | rii - pi |), where ri and pi are the relative abundances of resource item i eaten by the paralarvae (ri) and in the zooplankton (pi). Values of CI range from 1 for the broadest possible niche (a population uses resources in proportion to their availability) to [min pi] for the narrowest possible niche (a population is specialized exclusively on the rarest resource). Individual values of CI for each paralarvae were compared in the different locations sampled, to test whether the foraging tactics of O. vulgaris changed with development.
ACKNOWLEDGEMENTS
We thank the captain and crew of the R/V “Sarmiento de Gamboa” (IIM, CSIC). This study was supported by the project CAIBEX (CTM2007-66408-C02). AR was supported by a postdoctoral fellowship “Fundación Barrié” (3003197/2013) and “Securing Food, Water and the Environment” Research Focus Area funds (La Trobe University, Melbourne). SRD is supported by a UKRI Future Leaders Fellowship [MR/T020733/1] and the Wellcome Trust [108413/A/15/D]. For the purpose of Open Access, the authors have applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission.
DATA AVAILABILITY
All data generated or analysed during this study are included in this published article (and its Supplementary Information files).Author contributions
Conceived and designed the experiment: AR, AFG; Sample collection: AR, AFG; processed the samples: AR, ACB; Genomic design and analysis: AR, SD; Contributed reagents/materials/analysis tools: SD, JMS; zooplankton analysis ACB, AR; AR wrote the first draft of the manuscript. All authors contributed substantially to revisions and have approved the final version of the manuscript.
Competing Interests
All authors declare no conflict of interest.- Passarella, K. C. & Hopkins, T. L. H. Species composition and food habits of the micronektonic cephalopod assemblage in the Eastern Gulf of Mexico. Bull Mar Sci 49, 638–659 (1991).
- Vecchione, M. A method for examining the structure and contents of the digestive tract in paralarval squids. Bull Mar Sci 49, 300–308 (1991).
- Fingerhut, L. C. H. W. et al. Shotgun proteomics analysis of saliva and salivary gland tissue from the common octopus Octopus vulgaris. J Proteome Res 17, 3866–3876 (2018).
- Hernández-García, V., Martín, A. Y. & Castro, J. J. Evidence of external digestion of crustaceans in Octopus vulgaris paralarvae. Journal of the Marine Biological Association of the UK 80, 559–560 (2000).
- Iglesias, J., Otero, J. J., Moxica, C., Fuentes, L. & Sánchez, F. J. The completed life cycle of the octopus (Octopus vulgaris, Cuvier) under culture conditions: paralarval rearing using Artemia and zoeae, and first data on juvenile growth up to 8 months of age. Aquaculture International 12, 481–487 (2004).
- Villanueva, R. Decapod crab zoeae as food for rearing cephalopod paralarvae. Aquaculture 128, 143–152 (1994).
- Roura, Á., González, Á. F., Pascual, S. & Guerra, Á. A molecular approach to identifying the prey of cephalopod paralarvae. ICES Journal of Marine Science 67, 1408–1412 (2010).
- Roura, Á., González, Á. F., Redd, K. & Guerra, Á. Molecular prey identification in wild Octopus vulgaris paralarvae. Mar Biol 159, 1335–1345 (2012).
- Roura, Á. et al. Short-term meso-scale variability of mesozooplankton communities in a coastal upwelling system (NW Spain). Prog Oceanogr 109, 18–32 (2013).
- Roura, Á. et al. Life strategies of cephalopod paralarvae in a coastal upwelling system (NW Iberian Peninsula): Insights from zooplankton community and spatio-temporal analyses. Fish Oceanogr 25, 241–258 (2016).
- Olmos-Pérez, L., Roura, Á., Pierce, G. J., Boyer, S. & González, Á. F. Diet composition and variability of wild Octopus vulgaris and Alloteuthis media (Cephalopoda) paralarvae through a metagenomic lens. Front Physiol 8, (2017).
- Villanueva, R. & Norman, M. D. Biology of the planktonic stages of benthic octopuses. Oceanography and Marine Biology: An Annual Review 46, 105–202 (2008).
- Rocha, F., Guerra, Á., Prego, R. & Piatkowski, U. Cephalopod paralarvae and upwelling conditions off Galician waters (NW Spain). J Plankton Res 21, 21–33 (1999).
- González, A. F. et al. Distribution of common octopus and common squid paralarvae in a wind-driven upwelling area (Ria of Vigo, northwestern Spain). J Plankton Res 27, 271–277 (2005).
- Otero, J., Álvarez-Salgado, X. A., González, Á. F., Gilcoto, M. & Guerra, Á. High-frequency coastal upwelling events influence Octopus vulgaris larval dynamics on the NW Iberian shelf. Mar Ecol Prog Ser 386, 123–132 (2009).
- Moreno, A., dos Santos, A., Piatkowski, U., Santos, A. M. P. & Cabral, H. Distribution of cephalopod paralarvae in relation to the regional oceanography of the western Iberia. J Plankton Res 31, 73–91 (2009).
- Olmos-Pérez, L. Ecology of cephalopod paralarvae in a seasonal upwelling system. (Universidad de Vigo, 2018).
- Rees, W. J. The distribution of Octopus vulgaris Lamarck in British waters. Journal of the Marine Biological Association of the United Kingdom 29, 361–378 (1950).
- Rees, W. J. & Lumby, J. R. The abundance of octopus in the English Channel. Journal of the Marine Biological Association of the UK 33, 515–536 (1954).
- Nixon, M. & Mangold, K. The early life of Octopus vulgaris (Cephalopoda: Octopodidae) in the plankton and at settlement: A change in lifestyle. J Zool 239, 301–327 (1996).
- Roura, Á. et al. Oceanographic processes shape genetic signatures of planktonic cephalopod paralarvae in two upwelling regions. Prog Oceanogr 170, 11–27 (2019).
- Martins, R. S. et al. Modelling transport of chokka squid (Loligo reynaudii) paralarvae off South Africa: reviewing, testing and extending the ‘Westward Transport Hypothesis’. Fish Oceanogr 23, 116–131 (2014).
- Roberts, M. J. & van den Berg, M. Recruitment variability of chokka squid (Loligo vulgaris reynaudii)—role of currents on the Agulhas Bank (South Africa) in paralarvae distribution and food abundance. Bull Mar Sci 71, 691–710 (2002).
- Robin, J. P. et al. Transitions during cephalopod life history: The role of habitat, environment, functional morphology and behaviour. Adv Mar Biol 67, 361–437 (2014).
- Zeidberg, L. D. & Hamner, W. M. Distribution of squid paralarvae, Loligo opalescens (Cephalopoda: Myopsida), in the Southern California Bight in the three years following the 1997–1998 El Niño. Mar Biol 141, 111–122 (2002).
- Perales-Raya, C. et al. Comparative study of age estimation in wild and cultured Octopus vulgaris paralarvae: Effect of temperature and diet. Mar Ecol Prog Ser (2017) doi:10.3354/meps12218.
- Baltar, F., Arístegui, J., Gasol, J. M., Hernández-León, S. & Herndl, G. J. Strong coast-ocean and surface-depth gradients in prokaryotic assemblage structure and activity in a coastal transition zone region. Aquatic Microbial Ecology 50, 63–74 (2007).
- Roura, Á., Doyle, S. R., Nande, M. & Strugnell, J. M. You are what you eat: a genomic analysis of the gut microbiome of captive and wild Octopus vulgaris paralarvae and their zooplankton prey. Front Physiol 8, (2017).
- Iglesias, J. et al. First feeding of Octopus vulgaris Cuvier, 1797 paralarvae using Artemia: Effect of prey size, prey density and feeding frequency. Aquaculture 261, 817–822 (2006).
- Vidal, E. A. G. et al. Cephalopod culture: Current status of main biological models and research priorities. Adv Mar Biol 67, 1–98 (2014).
- Blanco-Bercial, L., Álvarez-Marqués, F. & Cabal, J. A. Changes in the mesozooplankton community associated with the hydrography off the northwestern Iberian Peninsula. ICES Journal of Marine Science 63, 799–810 (2006).
- Marta-Almeida, M., Dubert, J., Peliz, Á. & Queiroga, H. Influence of vertical migration pattern on retention of crab larvae in a seasonal upwelling system. Mar Ecol Prog Ser 307, 1–19 (2006).
- Queiroga, H. et al. Oceanographic and behavioural processes affecting invertebrate larval dispersal and supply in the western Iberia upwelling ecosystem. Prog Oceanogr 74, 174–191 (2007).
- Santos, A. M. P. et al. Physical-biological interactions in the life history of small pelagic fish in the Western Iberia Upwelling Ecosystem. Prog Oceanogr 74, 192–209 (2007).
- Hernández-León, S. et al. Vertical distribution of zooplankton in Canary Island waters: Implications for export flux. Deep Sea Res 1 Oceanogr Res Pap 48, 1071–1092 (2001).
- Irigoien, X. et al. Large mesopelagic fishes biomass and trophic efficiency in the open ocean. Nat Commun 5, 3271 (2014).
- Landeira, J. M., Lozano-Soldevilla, F., Hernández-León, S. & Barton, E. D. Spatial variability of planktonic invertebrate larvae in the Canary Islands area. Journal of the Marine Biological Association of the United Kingdom 90, 1217–1225 (2010).
- Rodríguez, J. M., Barton, E. D., Eve, L. & Hernández-León, S. Mesozooplankton and ichthyoplankton distribution around Gran Canaria, an oceanic island in the NE Atlantic. Deep Sea Res 1 Oceanogr Res Pap 48, 2161–2183 (2001).
- Hernández-León, S., Gómez, M. & Arístegui, J. Mesozooplankton in the Canary Current System: The coastal-ocean transition zone. Prog Oceanogr 74, 397–421 (2007).
- Chen, D. S., van Dykhuizen, G., Hodge, J. & Gilly, W. F. Ontogeny of copepod predation in juvenile squid (Loligo opalescens). Biological Bulletin (1996) doi:10.2307/1542676.
- García-Mayoral, E., Roura, Á., Moreno, A. & González, Á. Diet composition of wild Loligo vulgaris paralarvae along the West Iberian Peninsula coast. Mar Ecol Prog Ser 681, 71–85 (2022).
- Uchikawa, K., Sakai, M., Wakabayashi, T. & Ichii, T. The relationship between paralarval feeding and morphological changes in the proboscis and beaks of the neon flying squid Ommastrephes bartramii. Fisheries Science 75, 317–323 (2009).
- Venter, J. D., van Wyngaardt, S., Verschoor, J. A., Lipinski, M. R. & Verheye, H. M. Detection of zooplankton prey in squid paralarvae with immunoassay. J Immunoassay Immunochem 20, 127–149 (1999).
- Corell, J. & Rodríguez-Ezpeleta, N. Tuning of protocols and marker selection to evaluate the diversity of zooplankton using metabarcoding. Revista de Investigación Marina 21, 19–39 (2014).
- Geller, J., Meyer, C., Parker, M. & Hawk, H. Redesign of PCR primers for mitochondrial cytochrome c oxidase subunit I for marine invertebrates and application in all-taxa biotic surveys. Mol Ecol Resour 13, 851–861 (2013).
- Hernández-León, S. et al. Zooplankton biomass and indices of feeding and metabolism in relation to an upwelling filament off northwest Africa. J Mar Res 60, 327–346 (2002).
- Wang, M. & Jeffs, A. G. Nutritional composition of potential zooplankton prey of spiny lobster larvae: A review. Rev Aquac 6, 270–299 (2014).
- Boero, F. et al. Gelatinous plankton: Irregularities rule the world (sometimes). Mar Ecol Prog Ser 356, 299–310 (2008).
- Diaz Briz, L., Sánchez, F., Marí, N., Mianzan, H. & Genzano, G. Gelatinous zooplankton (ctenophores, salps and medusae): an important food resource of fishes in the temperate SW Atlantic Ocean. Marine Biology Research 13, 630–644 (2017).
- O´Rorke, R. et al. Determining the diet of larvae of western Rock Lobster (Panulirus cygnus) using high-throughput DNA sequencing techniques. PLoS One 7, e42757 (2012).
- Suzuki, N., Hoshino, K., Murakami, K., Takeyama, H. & Chow, S. Molecular diet analysis of phyllosoma larvae of the Japanese spiny lobster Panulirus japonicus (Decapoda: Crustacea). Marine Biotechnology (2008) doi:10.1007/s10126-007-9038-9.
- Ayala, D. J. et al. Gelatinous plankton is important in the diet of European eel (Anguilla anguilla) larvae in the Sargasso Sea. Sci Rep 8, 1–10 (2018).
- Olmos-Pérez, L., Pierce, G. J., Roura, Á. & González, Á. F. Barcoding and morphometry to identify and assess genetic population differentiation and size variability in loliginid squid paralarvae from NE Atlantic (Spain). Mar Biol 165, 136 (2018).
- García-Mayoral, E., Roura, Á., Ramilo, A. & González, Á. F. Spatial distribution and genetic structure of loliginid paralarvae along the Galician coast (NW Spain). Fish Res 222, 105406 (2020).
- Espinoza, V., Farías, A. & Uriarte, I. Characterization of cannibalism in early paralarvae of Patagonian red octopus, Enteroctopus megalocyathus (Gould, 1852), grown in a controlled environment. J World Aquac Soc 1–7 (2020) doi:10.1111/jwas.12738.
- Miranda, R. M., Espinoza, V., Dörner, J., Farías, A. & Uriarte, I. Sibling cannibalism on the small octopus Robsonella fontaniana (d’Orbigny, 1834) paralarvae. Marine Biology Research 7, 746–756 (2011).
- Harwood, J. D., Phillips, S. W., Sunderland, K. D. & Symondson, W. O. C. Secondary predation: Quantification of food chain errors in an aphid-spider-carabid system using monoclonal antibodies. Mol Ecol (2001) doi:10.1046/j.0962-1083.2001.01349.x.
- Sheppard, S. K. et al. Detection of secondary predation by PCR analyses of the gut contents of invertebrate generalist predators. Mol Ecol (2005) doi:10.1111/j.1365-294X.2005.02742.x.
- Kim, D.-H. & Kim, D. Microbial diversity in the intestine of olive flounder (Paralichthys olivaceus). Aquaculture 414–415, 103–108 (2013).
- Huang, Z., Li, X., Wang, L. & Shao, Z. Changes in the intestinal bacterial community during the growth of white shrimp, Litopenaeus vannamei. Aquac Res 47, 1737–1746 (2016).
- Arístegui, J. et al. Sub-regional ecosystem variability in the Canary Current upwelling. Prog Oceanogr 83, 33–48 (2009).
- Barton, E. D., Inall, M. E., Sherwin, T. J. & Torres, R. Vertical structure, turbulent mixing and fluxes during Lagrangian observations of an upwelling filament system off Northwest Iberia. Prog Oceanogr 51, 249–267 (2001).
- Joint, I. Lagrangian studies of the Iberian upwelling system-an introduction. A study of the temporal evolution of surface production and fate of organic matter during upwelling on and off the NW Spanish continental margin. Prog Oceanogr 51, 217–220 (2001).
- Peliz, Á., Dubert, J., Santos, A. M. P., Oliveira, P. B. & le Cann, B. Winter upper ocean circulation in the Western Iberian Basin - Fronts, Eddies and Poleward Flows: An overview. Deep Sea Res 1 Oceanogr Res Pap 52, 621–646 (2005).
- Pelegrí, J. L. et al. Coupling between the open ocean and the coastal upwelling region off northwest Africa: water recirculation and offshore pumping of organic matter. Journal of Marine Systems 54, 3–37 (2005).
- Álvarez-Salgado, X. a., Arístegui, J., Barton, E. D. & Hansell, D. a. Contribution of upwelling filaments to offshore carbon export in the subtropical Northeast Atlantic Ocean. Limnol Oceanogr 52, 1287–1292 (2007).
- Sponaugle, S. et al. Predicting self-recruitment in marine populations: biophysical correlates and mechanisms-. Bull Mar Sci 70, 341–375 (2002).
- Santos, A. M. P. et al. Impact of a winter upwelling event on the distribution and transport of sardine (Sardina pilchardus) eggs and larvae off western Iberia: A retention mechanism. Cont Shelf Res 24, 149–165 (2004).
- Bécognée, P., Almeida, C., Barrera, A., Hernández-Guerra, A. & Hernández-León, S. Annual cycle of clupeiform larvae around Gran Canaria Island, Canary Islands. Fish Oceanogr 15, 293–300 (2006).
- Rodríguez, J. M., Hernández-León, S. & Barton, E. D. Mesoscale distribution of fish larvae in relation to an upwelling filament off Northwest Africa. Deep Sea Res 1 Oceanogr Res Pap 46, 1969–1984 (1999).
- Sangrà, P. et al. The Cape Ghir filament system in August 2009 (NW Africa). Journal of Geophysical Research C: Oceans 120, 4516–4533 (2015).
- Troupin, C., Mason, E., Beckers, J. M. & Sangrà, P. Generation of the Cape Ghir upwelling filament: A numerical study. Ocean Model (Oxf) 41, 1–15 (2012).
- Leray, M. et al. A new versatile primer set targeting a short fragment of the mitochondrial COI region for metabarcoding metazoan diversity: application for characterizing coral reef fish gut contents. Front Zool 10, 34 (2013).
- Folmer, O., Black, M., Hoeh, W., Lutz, R. & Vrijenhoek, R. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol Mar Biol Biotechnol 3, 294–299 (1994).
- Bokulich, N. A. et al. Quality-filtering vastly improves diversity estimates from Illumina amplicon sequencing. Nat Methods 10, 57–60 (2013).
- Zhang, J., Kobert, K., Flouri, T. & Stamatakis, A. PEAR: a fast and accurate Illumina Paired-End reAd mergeR. Bioinformatics 30, 614–620 (2014).
- Wood, D. E. & Salzberg, S. L. Kraken: ultrafast metagenomic sequence classification using exact alignments. Genome Biol 15, R46 (2014).
- Edgar, R. C., Haas, B. J., Clemente, J. C., Quince, C. & Knight, R. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27, 2194–2200 (2011).
- Anderson, M. J. PERMANOVA Permutational multivariate analysis of variance. Austral Ecol 1–24 (2005) doi:10.1139/cjfas-58-3-626.
- Anderson, M. J., Gorley, R. N. & Clarke, K. R. PERMANOVA + for PRIMER: Guide to Software and Statistical Methods. vol. 1 (PRIMER-E, 2008).
- Strauss, R. Reliability Estimates for Ivlev’s electivity Index, the forage ratio, and a proposed linear index of food selection. Trans Am Fish Soc 108, 344–352 (1979).
- Feinsinger, P., Spears, E. E. & Poole, R. W. A simple measure of niche breadth. Ecology 62, 27–32 (1981).
Table 1. Number of Octopus vulgaris paralarvae (n) collected during the different surveys and those analysed for diet (d) along the Iberian Canary Current (ICC) eastern boundary upwelling system. The distance shown in km refers to the horizontal length from the collection sites to the coastline.
|
Survey |
ICC |
Location |
n |
d |
Suckers |
Depth (m) |
Distance (km) |
|
Ría de Vigo 2010 |
Spain |
Coast |
209 |
5 |
3 |
30 - 115 |
5 - 10 |
|
CAIBEX-I |
Northwest Iberian Peninsula |
Coast |
51 |
20 |
3 |
62 - 147 |
10 - 31 |
|
Ocean |
48 |
40 |
3 - 5 |
1940 - 3105 |
62 - 75 |
||
|
CAIBEX-III |
Cape Ghir, Morocco |
Coast |
9 |
9 |
3 |
88 - 90 |
19 |
|
Upwelling |
4 |
4 |
3 - 4 |
787 - 2328 |
48 - 93 |
||
|
Filament |
10 |
10 |
3 - 6 |
1526 - 2720 |
50 - 162 |
||
|
Ocean |
12 |
12 |
4 - 15 |
2418 - 3110 |
140 - 171 |
Table 2. DistLM results for the analysis of the main prey groups contributing to the diet of O. vulgaris paralarvae in the ICC zones sampled. The first two columns correspond to the marginal tests evaluating the independent contribution (indep.) of the different prey groups. The following columns correspond to the sequential analysis of the groups retained by the model that maximized the % of explained variance (Adj. R2). Prop. Indicates the individual % of variance explained by the groups as they are added into the model. %. Indicates the cumulative % of variance explained as prey groups are added into the model.
|
|
Northwest Iberian Peninsula |
|||||||
|
Group |
Indep. |
p-value |
Adj R2 |
SS(trace) |
Pseudo-F |
P-value |
Prop. |
% |
|
Brachyura |
56.4% |
0.001 |
39.8% |
137810 |
3.3931 |
0.001 |
56.4% |
56.4% |
|
Siphonophores |
17.3% |
0.001 |
50.2% |
28984 |
2.7629 |
0.001 |
11.9% |
68.2% |
|
Copepods |
16.9% |
0.001 |
58.3% |
23104 |
2.1893 |
0.006 |
9.5% |
77.7% |
|
Pteropods |
5.8% |
0.001 |
62.3% |
6855.7 |
4.3146 |
0.028 |
2.8% |
80.5% |
|
Cnidarians |
11.4% |
0.001 |
66.7% |
9739.3 |
2.311 |
0.027 |
4.0% |
84.5% |
|
|
Cape Ghir, Morocco |
|||||||
|
Cnidarians |
23.0% |
0.001 |
11.1% |
30883 |
1.937 |
0.001 |
23.0% |
23.0% |
|
Brachyura |
43.1% |
0.001 |
24.2% |
49243 |
1.4486 |
0.013 |
36.6% |
59.6% |
|
Copepods |
28.0% |
0.003 |
35.3% |
22487 |
1.5507 |
0.107 |
16.7% |
76.3% |
|
Euphausiids |
20,7% |
0.311 |
48.0% |
13238 |
1.8914 |
0.173 |
9.8% |
86.1% |
|
Ostracods |
3.9% |
0.212 |
57.3% |
5264.7 |
2.7503 |
0.16 |
3.9% |
90.0% |
Table 3. SIMPER analysis showing the averaged log (abundance +1) of the ten most discriminant species that contributed to the ontogenic changes between paralarvae with different suckers (s) in two sub-regions of the ICC. The paralarvae collected off Cape Ghir, Morocco, were grouped in three groups (three suckers, 3s: n = 18; four to five suckers, 4-5s: n=5; more than suckers, >6s: n=8). Abbreviations: Diss/SD, shows the discriminant power of each species; Var%, shows the % of total variability that is explained by each species for the pair analysed; Av. Diss. is the averaged dissimilarity observed between the diets analysed (i.e., how different is the diet of the paralarvae with a different number of suckers).
No competing interests reported.