The genetic population structure of Thunnus thynnus (Linnaeus, 1758) in the Mediterranean Sea,a controversial issue

The Atlantic bluefin tuna (ABFT), Thunnus thynnus (Linnaeus, 1758), is an important commercial species managed as two different stocks, western and eastern Atlantic, with their spawning grounds in the Gulf of Mexico and in the Mediterranean Sea, respectively. The eastern Atlantic stock has been overexploited in the last decades, leading to the application of specific management measures introduced by the International Commission for the Atlantic Tuna (ICCAT). A clear understanding of the genetic structure of ABFT Mediterranean population should be pursued in order to support management decisions. To date the genetic studies on the Mediterranean ABFT, carried out with different molecular markers and sampling procedures, have produced unclear results. Here, we analysed ABFT samples from central and western Mediterranean Sea with mitochondrial sequences and 11 microsatellite loci to investigate, among the others, the area of the Strait of Messina, where environmental conditions seem to support a resident population of ABFT. Furthermore, genetic analyses of mitochondrial sequences were carried out including nucleotide sequences of Adriatic ABFT wild larvae retrieved from GenBank. Among the investigated areas a genetic differentiation was detected between the Strait of Messina and the Tyrrhenian Sea with microsatellite loci according to the exact G test, but not to the Bayesian analyses carried out with STRUCTURE. The analyses with mitochondrial sequences do not reveal any differentiation among sampled areas, however, a highly significant genetic divergence was observed between the Adriatic mitochondrial sequences retrieved from GenBank and the central‐western Mediterranean sequences obtained in the present work. Our results provide some evidence of population structure of Mediterranean ABFT adding pieces to a still unclear picture.     

Genetic identity of YOY bluefin tuna from the eastern and western Atlantic spawning areas

We used 320 young-of-the-year (YOY) specimens of the highly migratory and overfished Atlantic bluefin tuna, Thunnus thynnus, Linnaeus 1758, to evaluate the hypothesis that Atlantic bluefin tuna comprises 2 stocks with spawning grounds in the Gulf of Mexico and in the Mediterranean Sea. Significant genetic differentiation at 8 nuclear microsatellite loci (F(ST) = 0.0059, P = 0.0005) and at the mitochondrial control region (Phi(ST) = 0.0129, P = 0.0139) was detected among YOY Atlantic bluefin tuna captured on spawning grounds in the Gulf of Mexico (n = 40) versus the western (n = 255) and eastern (n = 25) basins of the Mediterranean Sea. The genetic divergence among spawning populations, combined with the extensive trans-Atlantic movements reported for juvenile and adult Atlantic bluefin tuna, indicates a high degree of spawning site fidelity. Recognition of genetically distinct populations necessitates independent management of Atlantic bluefin tuna on spawning grounds and warrants evaluation of the level of mixing of populations on feeding grounds. The genetic pattern might not have been detected unless juvenile specimens or actively spawning adults had been sampled.     

Microsatellite and mitochondrial DNA analyses of the genetic structure of blacktip shark (Carcharhinus limbatus) nurseries in the northwestern Atlantic, Gulf of Mexico, and Caribbean Sea

Abstract We investigated the genetic structure of blacktip shark (Carcharhinus limbatus) continental nurseries in the northwestern Atlantic Ocean, Gulf of Mexico, and Caribbean Sea using mitochondrial DNA control region sequences and eight nuclear microsatellite loci scored in neonate and young-of-the-year sharks. Significant structure was detected with both markers among nine nurseries (mitochondrial PhiST = 0.350, P < 0.001; nuclear PhiST = 0.007, P < 0.001) and sharks from the northwestern Atlantic, eastern Gulf of Mexico, western Gulf of Mexico, northern Yucatan, and Belize possessed significantly different mitochondrial DNA haplotype frequencies. Microsatellite differentiation was limited to comparisons involving northern Yucatan and Belize sharks with nuclear genetic homogeneity throughout the eastern Gulf of Mexico, western Gulf of Mexico, and northwestern Atlantic. Differences in the magnitude of maternal vs. biparental genetic differentiation support female philopatry to northwestern Atlantic, Gulf of Mexico, and Caribbean Sea natal nursery regions with higher levels of male-mediated gene flow. Philopatry has produced multiple reproductive stocks of this commercially important shark species throughout the range of this study.     

Social kin associations and genetic structuring of striped dolphin populations (Stenella coeruleoalba) in the Mediterranean Sea

We investigated hierarchical patterns of genetic subdivision, and assessed kinship within and between social groups of striped dolphins (Stenella coeruleoalba) in the Tyrrhenian Sea. A total of 165 samples were analysed at eight microsatellite DNA loci, including outgroup samples from the Adriatic, Scotland and Spain for population-level comparisons. We found population genetic structure within the Mediterranean basin, including small but significant differentiation between the Adriatic and Tyrrhenian Seas (FST=0.0047, P=0.008), and between putative 'inshore' and 'offshore' (FST=0.0217, P=0.005) populations in the Tyrrhenian Sea. Assessment of kinship within and among 12 association groups showed higher average kinship for females within than between groups, and smaller groups showed higher average kinship. Comparisons of relatedness for both sexes showed a significant difference between males and females, with females more likely to associate with adult kin. Together these data emphasize the importance of the social cohesion of kin in small groups to the structuring of striped dolphin populations in this environment.     

Phylogeography of the common goby, Pomatoschistus microps, with particular emphasis on the colonization of the Mediterranean and the North Sea

The phylogeographical patterns of a small marine fish, the common goby, Pomatoschistus microps, were assessed at 12 sites along the northeastern Atlantic coasts and the western Mediterranean Sea. A combination of two genetic markers was employed: cellulose acetate allozyme electrophoresis (CAGE) and sequence analysis of a 289 bp fragment of the mitochondrial locus cytochrome b. Both markers were congruent in revealing significant differences between samples (global FST = 0.247 for the allozymes and PhiST = 0.437 for the mitochondrial DNA data) and a pattern of isolation-by-distance. Phylogeographical analyses yielded a shallow branching structure with four groups. Three of those were confined to the Atlantic basin and showed a star-like pattern. The fourth group contained a central haplotype occurring at the edges of the species' distribution, accompanied by a few more rare variants, which were restricted to the Mediterranean Sea. A genetic break was observed around the British Isles, with distinct haplotypes dominating at either side of the English Channel. A significantly negative correlation between the degree of genetic diversity and latitude was recorded both for mitochondrial DNA (mtDNA) and allozymes in the Atlantic basin. Gene flow analysis suggested that recolonization of the North Sea and the coasts of western Scotland and Ireland may have taken place from a glacial refugium in the Southern Bight of the North Sea. These results are discussed in the perspective of possible postglacial migration routes of marine fish along the northeastern Atlantic coasts.     

Into the depth of population genetics: pattern of structuring in mesophotic red coral populations

Deep-sea reef-building corals are among the most conspicuous invertebrates inhabiting the hard-bottom habitats worldwide and are particularly susceptible to human threats. The precious red coral (Corallium rubrum, L. 1758) has a wide bathymetric distribution, from shallow up to 800 m depth, and represents a key species in the Mediterranean mesophotic reefs. Several studies have investigated genetic variability in shallow-water red coral populations, while geographic patterns in mesophotic habitats are largely unknown. This study investigated genetic variability of C. rubrum populations dwelling between 55 and 120 m depth, from the Ligurian to the Ionian Sea along about 1500 km of coastline. A total of 18 deep rocky banks were sampled. Colonies were analyzed by means of a set of microsatellite loci and the putative control region of the mitochondrial DNA. Collected data were compared with previous studies. Both types of molecular markers showed high genetic similarity between populations within the northern (Ligurian Sea and Tuscan Archipelago) and the southern (Tyrrhenian and Ionian seas) study areas. Variability in habitat features between the sampling sites did not affect the genetic variability of the populations. Conversely, the patchy distribution of suitable habitats affected populations’ connectivity within and among deep coral banks. Based on these results and due to the emphasis on red coral protection in the Mediterranean Sea by international institutions, red coral could be promoted as a ‘focal species’ to develop management plans for the conservation of deep coralligenous reefs, a reservoir of marine biodiversity.     

Spatial patterns of genetic diversity in Posidonia oceanica, an endemic Mediterranean seagrass

Posidonia oceanica is an endemic seagrass species in the Mediterranean Sea. In order to assess levels of genetic structure in this species, the microsatellite polymorphism was analysed from meadows collected in several localities, along the coasts of the Tyrrhenian Sea (Mediterranean Sea). The existence of single population units and the recruitment of seedlings collected in some localities were investigated. Moreover, genetic structure at different spatial scales and biogeographic relationships among populations were also assessed. Our analysis showed the existence of clear patterns of genetic structure in P. oceanica in the area considered in the analysis. P. oceanica, in fact, is present in separate meadows that represent discrete populations, characterized by low genetic diversity. Comparable levels of genetic variability between mature meadows and seedlings were found. Patterns of genetic relatedness among populations seem to be in accord with direction of dominant current flux in the whole area, separating South Tyrrhenian from North Tyrrhenian populations. Moderate levels of gene flow between populations and genetic substructure within populations, together with the finding of the limited role of sexual reproduction in increasing genetic variability, should be a cause for concern for the persistence of this essential resource in the Mediterranean basin.     

Comparative phylogeography of Atlantic bluefin tuna and swordfish: the combined effects of vicariance, secondary contact, introgression, and population expansion on the regional phylogenies of two highly migratory pelagic fishes

Comparative phylogeography has revealed remarkable patterns of concordance in the maternal phylogenies of many species. The phylogeography and historical demography of the mitochondrial control region I for 607 Atlantic bluefin tuna (Thunnus thynnus) and 275 swordfish (Xiphias gladius) were analyzed to clarify the complex phylogenetic signals in the North Atlantic-Mediterranean region where they are sympatric. Atlantic bluefin tuna mtDNA is polyphyletic, and includes rare sequences sister to Pacific bluefin tuna (Thunnus orientalis) and introgressed albacore (Thunnus alalunga) sequences. There is no geographic partitioning between Atlantic and Mediterranean samples of Atlantic bluefin tuna (Phi(ST)=0.002). In contrast, Atlantic and Mediterranean swordfish are differentiated (Phi(ST)=0.091) due to the combined effects of vicariance, secondary contact, and dissimilar regional demographic histories. Mediterranean swordfish has substantially less variation, and a more recent history (tau=2.42) than that of Atlantic swordfish (tau=7.02). In spite of the discordant phylogenetic and phylogeographic signals, the demographic history of Atlantic swordfish and Atlantic bluefin tuna (tau=7.51) suggests concordance in the timeline of population expansion. Possible scenarios of cladogenesis, expansion, and contraction, influenced by glacial cycles during the Pleistocene, are formulated.     

Genetic consequences of the ice ages on nurseries of the bat Myotis myotis: a mitochondrial and nuclear survey

Analyses of mitochondrial DNA (mtDNA) control region polymorphism and of variation at 10 nuclear microsatellite loci were used to investigate the mechanisms and genetic consequences of postglacial expansion of Myotis myotis in Europe. Initial sampling consisted of 480 bats genotyped in 24 nursery colonies arranged along a transect of approximately 3000 km. The phylogeographical survey based on mtDNA sequences revealed the existence of major genetic subdivisions across this area, with several suture zones between haplogroups. Such zones of secondary contact were found in the Alps and Rhodopes, whereas other potential barriers to gene flow, like the Pyrenees, did not coincide with genetic discontinuities. Areas of population admixture increased locally the genetic diversity of colonies, which confounded the northward decrease in nucleotide diversity predicted using classical models of postglacial range expansion. However, when analyses were restricted to a subset of 15 nurseries originating from a single presumed glacial refugium, mtDNA polymorphism did indeed support a northwards decrease in diversity. Populations were also highly structured (PhiST = 0.384). Conversely, the same subset of colonies showed no significant latitudinal decrease in microsatellite diversity and much less population structure (FST = 0.010), but pairwise genetic differentiation at these nuclear markers was strongly correlated with increasing geographical distance. Together, this evidence suggests that alleles carried via male bats have maintained enough nuclear gene flow to counteract the effects of recurrent bottlenecks generally associated with recolonization processes. As females are highly philopatric, we argue that the maternally transmitted mtDNA marker better reflects the situation of past, historical gene flow, whereas current levels of gene flow are better reflected by microsatellite markers.     

Trophic ecology of Atlantic bluefin tuna Thunnus thynnus larvae

Feeding intensity, diet composition, selectivity, energy ingestion and dietary niche breadth of larval Atlantic bluefin tuna Thunnus thynnus were studied on the eastern (Mediterranean) spawning grounds of the species. Larval T. thynnus were collected in the Balearic Archipelago (north-west Mediterranean Sea) during 2004 and 2005 using surveys specific for larval scombrids. Larvae between 2·6 and 8·7 mm standard length (L(S) ) are diurnal feeders, and 94% of the guts collected during daylight hours were full. The mean ±s.d. number of prey per gut was 7·1 ± 5·7, with mean ±s.d. ranging from 3·0 ± 1·6 in the smallest T. thynnus larvae to 11·1 ± 5·8 in 5·0-6·0 mm L(S) larvae. Up to 21 prey were found in a single larval gut (5·0-6·0 mm L(S) ) at the end of the day. Larvae progressively selected larger prey and exhibited increased carbon content concurrent with preflexion development of feeding and locomotory structures. Larvae of 5·0-6·0 mm L(S) exhibited positive selection of cladocerans over other prey (Chesson's index), whereas copepod nauplii dominated the diets of earlier stages. The dietary niche breadth measured increased initially but decreased at c. 5·5 mm L(S) . Appendicularians were found in the diet of larger larval sizes, but no piscivory was observed. Results are discussed in light of the sparse existing data for larval T. thynnus and other larval tuna species.     

Population genetics of Atlantic bluefin tuna,Thunnus thynnus (Linnaeus, 1758), in the Mediterranean: implications for its conservation management

The movement of Atlantic bluefin tuna (ABFT) across international boundaries necessitates traceability strategies that would provide more accurate information needed for stock assessment. The Mediterranean Sea is one of the main contributors to ABFT reproduction and global population genetic diversity. In the present study this genetic variability was investigated using 193 samples of adult bluefin tuna from Spain, Turkey and Malta – a longitudinal distance of 3400 km. Analysed were 13 microsatellite loci (eight of which were newly‐tested) as genetic markers for the population study. Allele richness measured per locus and sampling location varied from 1.89 to 8.88, taking into account rarefaction. ABFT private alleles were detected in each of the three sampling sites. No significant spatial genetic divergence was found between pairs at the studied locations (F_ST values <0.0001; P‐values >0.05). Bayesian clustering analysis corroborated a single and panmictic ABFT population in the Mediterranean Sea. Statistical power analyses indicated a high probability of detecting genetic differentiation and population structure with the sample size and microsatellites used, even at an F_ST value of 0.005. From the results it may be postulated that migrating ABFT during the spawning season are allowing gene flow within the Mediterranean Sea. The complex interplay of movements, including plasticity in the selection of spawning sites with increasing age and environmental conditions, require multiple and new fisheries monitoring and management techniques in order to target the ABFT long‐term conservation effectively.     

Isolation and characterization of seven tetranucleotide microsatellite loci from Atlantic northern bluefin tuna Thunnus thynnus thynnus

Microsatellite‐enriched genomic libraries were obtained from the Atlantic northern bluefin tuna, Thunnus thynnus thynnus and seven tetranucleotide markers were successfully isolated and characterized from this library. These markers were found to have between 1 and 17 alleles in Atlantic northern bluefin tuna and heterozygosity ranged from 0 to 0.85. No deviations from the expectation of Hardy‐Weinburg equilibrium were found for any marker. Several of these markers amplify reliably in other tuna species.     

Distinct Yellowfin Tuna (Thunnus albacares) Stocks Detected in Western and Central Pacific Ocean (WCPO) Using DNA Microsatellites

The yellowfin tuna, Thunnus albacares (Bonnaterre, 1788), covers majority of the Philippines’ tuna catch, one of the major fisheries commodities in the country. Due to its high economic importance sustainable management of these tunas has become an imperative measure to prevent stock depletion. Currently, the Philippine yellowfin tuna is believed to be part of a single stock of the greater WCPO though some reports suggest otherwise. This study therefore aims to establish the genetic stock structure of the said species in the Philippines as compared to Bismarck Sea, Papua New Guinea using nine (9) DNA microsatellite markers.DNA microsatellite data revealed significant genetic differentiation between the Philippine and Bismarck Sea, Papua New Guinea yellowfin tuna samples. (FST = 0.034, P = 0.016), which is further supported by multilocus distance matrix testing (PCoA) and model-based clustering (STRUCTURE 2.2).With these findings, this study posits that the yellowfin tuna population in the Philippines is a separate stock from the Bismarck Sea population. These findings add evidence to the alternative hypothesis of having at least 2 subpopulations of yellowfin tuna in the WCPO and calls for additional scientific studies using other parameters to investigate this. Accurate population information is necessary in formulating a more appropriate management strategy for the sustainability of the yellowfin tuna not only in the Philippines but also in the WCPO.     

Expression of vitellogenin receptor gene in the ovary of wild and captive Atlantic bluefin tuna (Thunnus thynnus)

The cDNA sequences of vitellogenin receptor proteins (VgR(+) and VgR(-)), containing or lacking the O-linked sugar domain, were determined in Atlantic bluefin tuna (Thunnus thynnus L.). VgR(-) gene expression in the ovary was compared in captive-reared and wild Atlantic bluefin tuna during the reproductive cycle. Gonad samples from adult fish were sampled from 2008 to 2010 from stocks reared in captivity at different commercial fattening operations in the Mediterranean Sea and from wild individuals caught either by traditional tuna traps during their migration towards the spawning grounds in the Mediterranean Sea or by the long-line artisanal fishery. In addition, juvenile male and female Atlantic bluefin tuna were sampled from a farming facility, to obtain baseline information and pre-adulthood amounts of VgR(-). The total length of VgR(+) cDNA was 4006 nucleotides (nt) and that of VgR(-) was 3946 nt. Relative amounts of VgR(-) were greater in juvenile females and in those adults having only previtellogenic oocytes (119 ± 55 and 146 ± 26 folds more than juvenile males, respectively). Amounts of VgR(-) were less in individuals with yolked oocytes (ripening stage, May-June) and increased after spawning in July (92 ± 20 and 113 ± 13 folds more than juvenile males in ripening and post-spawning fish, respectively). These data suggest that regulation of VgR(-) is not under oestrogen control. During the ripening period, greater VgR(-) gene expression was observed in wild fish than in fish reared in captivity, possibly because of (a) differences in water temperature exposure and/or energy storage, and/or (b) an inadequate diet in reared Atlantic bluefin tuna.     

Strong population genetic structure and contrasting demographic histories for the small-spotted catshark (Scyliorhinus canicula) in the Mediterranean Sea

Coastal and demersal chondrichthyans, such as the small-spotted catshark, are expected to exhibit genetic differentiation in areas of complex geomorphology like the Mediterranean Basin because of their limited dispersal ability. To test this hypothesis, we used a fragment of the mitochondrial cytochrome c oxidase subunit I gene and 12 nuclear microsatellite loci in order to investigate the genetic structure and historical demography of this species, and to identify potential barriers to gene flow. Samples were collected from the Balearic Islands, the Algerian Basin, the Ionian Sea, the Corinthian Gulf and various locations across the Aegean Sea. Additional sequences from the Atlantic and the Levantine Basin retrieved from GenBank were included in the mitochondrial DNA analysis. Both mitochondrial and nuclear microsatellite DNA data revealed a strong genetic subdivision, mainly between the western and eastern Mediterranean, whereas the Levantine Basin shared haplotypes with both areas. The geographic isolation of the Mediterranean basins seems to enforce the population genetic differentiation of the species, with the deep sea acting as a strong barrier to its dispersal. Contrasting historical demographic patterns were also observed in different parts of the species'' distribution, most notably a population growth trend in the western Mediterranean/Atlantic area and a slight decreasing one in the Aegean Sea. The different effects of the Pleistocene glacial periods on the habitat availability may explain the contrasting demographic patterns observed. The current findings suggest that the small-spotted catshark exhibits several genetic stocks in the Mediterranean, although further study is needed.     

Genetic homogeneity among colonies of the white-nest swiftlet (Aerodramus fuciphagus) in Thailand

The white-nest swiftlet, Aerodramus fuciphagus, originally lived in large colonies in natural caves, but now it also occurs in man-made buildings. We investigated the patterns of genetic differentiation in two mitochondrial DNA genes (cyt-b and ND2) and eight microsatellite loci among and within colonies of A. fuciphagus from across recently established man-made colonies in Thailand. Ten white-nest swiftlet colonies were sampled along the coast of the Gulf of Thailand and the Andaman Sea in Thailand during 2003-2006. The genetic diversity of mtDNA was very low, and few significant PhiST values were found between pairs of colonies. Analyses of haplotype relationships did not show genetic structure across the sampled distribution. The level of genetic diversity for microsatellite loci was high, but FST values were not significant. However, due to small sample sizes for some colonies that could limit conclusions on genetic differentiation from PhiST and FST, we also analyzed the microsatellite data using STRUCTURE and found that number of subpopulations of white-nest swiftlets in sampled colonies was one. The lack of genetic differentiation among swiftlet house colonies could be a result of high gene flow between colonies and large population sizes. Our results suggest that A. fuciphagus living in recently established man-made colonies in Thailand should be considered members of a single panmictic population. Future work will be necessary to determine whether this panmixia is stable or a temporary result of the recent explosive expansion of the number of colonies, and comparisons to natural colonies may provide an understanding of mechanisms producing the lack of genetic structure in swiftlet house colonies.     

Population structure and colonization history of the olive fly, Bactrocera oleae (Diptera, Tephritidae)

The olive fly, Bactrocera oleae, is the major pest of olives in most commercial olive-growing regions worldwide. The species is abundant in the Mediterranean basin and has been introduced recently into California and Mexico, creating problems for quarantine protection and international trade. Here, we use nuclear microsatellite markers and mitochondrial sequences to examine the history of olive fly range expansion and colonization. Sampled populations span the current distribution of the olive fly worldwide, including South and Central Africa, Pakistan, Mediterranean Europe and Middle East, California, and Mexico. The Pakistani populations appear to be genetically well differentiated from the remaining populations, though rooting the origins of the species is problematic. Genetic similarity and assignment tests cluster the remaining populations into two genetic groups--Africa and a group including the Mediterranean basin and the American region. That Africa, and not the Mediterranean, is the origin of flies infesting cultivated olive is supported by the significantly greater genetic diversity at microsatellite loci in Africa relative to the Mediterranean area. The results also indicate that the recent invasion of olive flies in the American region most likely originated from the Mediterranean area.     

Control region haplotype variation in the central Mediterranean common sole indicates geographical isolation and population structuring in Italian stocks

Nine common sole Solea vulgaris samples from five fishery Management Units (MUs) of the central Mediterranean exhibited differences in a control region sequence marker. Parsimony network showed two low-divergent haplotype phylogroups. The former predominated in samples from the Ligurian, Tyrrhenian, Adriatic and north-western Ionian Seas (MUs 9, 10, 17, 18 and northern part of 19, respectively), whereas the latter was abundant in the south-western Ionian (southern part of MU 19). The geographical clustering of maternal lineages accounts for population structuring and indicates geographical isolation of common sole stocks in the Mediterranean. Several life-history traits of common sole and hydro-geographical features of the basin might support this pattern of differentiation. Haplotype frequency differences were detected among samples either from different MUs or within the same MU. This situation indicates the presence of partially subdivided or nearly panmictic population units, whose exploitation should be based on appropriate bio-ecological features. The usefulness of control region sequence marker enhances the routinely use of the genetic stock structure analysis in low-dispersal demersal marine resources.     

RAPD profiles distinguish <Emphasis Type="Italic">Paracentrotus lividus</Emphasis> populations living in a stressing environment (Amvrakikos Gulf,Greece)

Random amplified polymorphic DNA (RAPD) analysis was performed to assess genetic markers of Paracentrotus lividus populations living in stressing environment in the Amvrakikos Gulf (Western Greece, Ionian Sea) where two populations distinguishable in body size, smaller than the open sea ones, were detected. The UPGMA dendrogram, constructed from pairwise Φ_st values among population nuclear DNA markers, revealed that the small and medium-sized populations living inside the Amvrakikos presented a lower polymorphism, and form a cluster that shows the genetic distance with normal-sized populations (Ionian and Tyrrhenian Seas) living in open sea. AMOVA analysis indicated a genetic distance among the sea urchin populations from the Tyrrhenian sea and Ionian sea. The article is published in the original.     

Comparative study of liver vitellogenin gene expression and oocyte yolk accumulation in wild and captive Atlantic bluefin tuna (Thunnus thynnus L.)

The sequence of vitellogenin A (VgA) and vitellogenin B (VgB) cDNAs in Atlantic bluefin tuna (Thunnus thynnus L.) were determined, and vitellogenin expression levels in the liver and oocyte yolk accumulation were compared in wild and captive-reared individuals. Liver and ovary samples were taken from 31 individuals reared experimentally in three commercial Atlantic bluefin tuna fattening sites in the Mediterranean Sea and from 33 wild individuals caught by commercial traps during the fish's migration towards their Mediterranean spawning grounds. The total length of VgA cDNA was 5585 nucleotides and that of VgB was 5267 nucleotides. The identity and similarity between deduced amino acid sequences of VgA and VgB were 60% and 78%, respectively. The Atlantic bluefin tuna VgA and VgB amino acid sequences have high similarities with those of other teleost fishes. Relative levels of VgA and VgB mRNAs were low in April, increased significantly during the reproductive period in May and June, and declined in July. There was a trend towards higher relative levels of VgA and VgB mRNAs in captive fish compared to wild individuals during the reproductive period. The surface occupied by eosinophilic yolk granules in fully vitellogenic oocytes, as well as the frequency of oocytes in late vitellogenesis, was significantly higher in captive compared to wild individuals. The study suggests that the experimental conditions under which Atlantic bluefin tuna individuals were reared allowed the occurrence of normal vitellogenesis, based on gene expression of VgA and VgB in the liver and yolk accumulation in the oocytes. The higher yolk accumulation and frequency of vitellogenic oocytes observed in the ovaries of captive fish suggest that improvements in feeding practices may result in an improved vitellogenic process.