Final ReportSummary - SELFDOTT (From capture based to SELF-sustained aquaculture and Domestication Of bluefin tuna, Thunnus thynnus)
Executive summary:
Continuing on the success of the Fifth Framework Programme (FP5) research programme REPRODOTT, which was the first ever study of the reproductive biology of the Atlantic bluefin tuna in captivity, and which developed a hormonal method for the induction of spawning, the programme SELFDOTT (From capture based to self sustained aquaculture and domestication of Bluefin tuna, Thunnus thynnus) implemented the knowledge on the reproduction of bluefin tuna in captivity. It established the knowledge-base required for controlled development of eggs, larvae and suitable and environmentally performing feeds.
Project Context and Objectives:
SELFDOTT proposed to implement knowledge already obtained on the artificial control of reproduction of the Atlantic bluefin tuna (BFT), Thunnus thynnus, to obtain viable eggs, and study embryonic and larval development for the production of fry (juveniles). At the same time, suitable and environmentally performing feeds for the grow out of BFT have been developed. It will contribute to reduce/eliminate in the future the practice of raw fish importation and feeding by the fattening industry. Wild juvenile and mature BFT were reared in captivity at two sites in the Mediterranean, and they have used to study puberty, gametogenesis, and the influence of diet on reproductive maturation and gamete quality. Mature fish were induced to spawn using hormone implants and the eggs were collected using devices designed specifically for cages. To establish the knowledge-base for controlled development of BFT larvae, the mesocosm and artificial larval rearing methods have been employed. The ontogenesis of essential biological functions was studied, including environmental perception, digestion, immunity and behaviour.
The main objectives of the SELFDOTT project have been:
1.- Substantiate the current knowledge on the reproduction of capture based aquaculture bluefin tuna
2.- Establish the knowledge-base required for controlled development of eggs and larvae
3.- Establish the knowledge-base for the development of suitable and environmentally performing feeds
Project Results:
One way to alleviate the pressure on the wild fishery of the Atlantic bluefin tuna (Thunnus thynnus, BFT) and aid in its conservation would be the domestication of the BFT and the development of a self-sustained industry. This aquaculture industry will propagate the BFT in captive conditions, rear the larvae and produce fingerlings for further grow-out on suitable, scientifically formulated and environmentally performing feed, as it is done successfully in the EU for species such as the Atlantic salmon (Salmo salar), European seabass (Dicentrarchus labrax) and gilthead seabream (Sparus aurata). Therefore, the overall objectives of SELFDOTT were to further study the reproductive biology and larval rearing of this species in captivity, and develop formulated feeds for the ongrowing of the produced juveniles to market size. The above studies were expected to also result in a better understanding of the life history and physiology of the BFT, contributing to the successful management of the wild stocks.
WP2 Reproduction
In the area of reproduction, the objectives of SELFDOTT were to:
(a) enhance the existing knowledge on the reproductive biology (onset of puberty, gametogenesis, and gamete quality) and artificial spawning-induction of BFT;
(b) optimize microsatellite markers to genotype BFT and their progeny, with applications to broodstock management and fisheries populations;
(c) obtain fertilized and viable eggs, and ship them to larval rearing facilities for the production of juveniles;
(d) develop appropriate broodstock diets for the support of reproductive maturation and gamete quality; and
(e) study the behaviour of long-term reared wild broodstock and establish monitoring procedures; in order to develop the knowledge for a self-sustained aquaculture of the BFT.
WP3 Larval rearing
In WP3, devoted to larval rearing, the first year of the project, 2008, was dedicated to the first husbandry attempts for defining a preliminary rearing protocol on a comparative basis between mesocosm (HCMR, Greece) and artificial rearing systems such as pseudo green (IEO, Spain, HCMR, Greece and MRRA-Aquaculture, Malta) and clear water rearing methods (IFREMER, France).
The Mesocosm, implemented by HCMR. This technique is described as an intermediate between the intensive and extensive methods of rearing that can be considered as a semi-intensive technique of mass production. Phytoplankton and live prey item concentrations and quality is partly related to the "maturation" of the middle previous the beginning of the rearing
The Pseudo-Green Water, implemented by HCMR, IEO and MRRA-Aquaculture. This technique is characterized by controlled conditions of light intensity, photophase and feeding. The phytoplankton is not reproduced neither bloom in the rearing tank (constant concentration by daily addition). The live prey quality is partly influenced by phytoplankton quality and the water quality control is partly dependant of the phytoplankton management.
The Clear Water, implemented by Ifremer and punctually by IEO. This technique is characterized by controlled conditions of water quality, light intensity, photophase and feeding. The live prey quality is fully dependant of the strain quality and their enrichment products and methods. The water quality control partly depends on the recycling system definition and efficiency.
In 2008, a sampling strategy were defined to deliver BFT larva batches from the husbandry sites to each WP3 partners in charge of describing the ontogeny of the main targeted functions.
Larvae of Bonito (Sarda sarda) considered as a model species were expected to train each husbandry sites and supply each scientific topics issues before the BFT season. The bonito fishing season (June 2008) was exceptionally poor and no Bonito egg batch was collected for in vitro fertilization. One Bonito natural spawning occurred from one of the Bonito captive broodstock in IEO facilities in Puerto de Mazarron (Spain). This unique egg batch (16,000 units) was collected and stocked in IEO 5,000 l rearing tanks. The feeding sequence, swimbladder inflation, behaviour, growth and survival patterns were described using a pseudo green rearing method until 40-45 DPH and several 2-4 g Bonito fingerlings were produced.
BFT tentative egg production from SELFDOTT cages located in Malta and Spain failed. Four successive BFT spawning events (mean: 3 millions eggs per day) occurred in ALLOTUNA (collaborative project) cages in Vibo (Italy) end June but only a few larva batches can have been delivered by the Panittica Pugliese Spa hatchery to IEO, MRRA-Aquaculture, IFREMER and HCMR. Rearing attempts failed in MRRA-Aquaculture and HCMR due to a dramatic early larva mortality during and just after transport. Only IEO and IFREMER succeeded in maintaining a few individuals alive until 14 DPH (pseudo green water method) and 9 DPH (clear water method) respectively. These first tentative trials implemented with a few hundreds (IEO) or a few tens (IFREMER) of larvae having suffered detrimental transport conditions allowed to fine tune each rearing system and define the environmental conditions leading the animals and living preys to be acceptably homogenisation. Despite a poor early growth, first feeding figures were observed during the early stages.
The different dummy transport attempts and the actual transport operations conducted with BFT larvae and eggs from the Panittica Pugliese Spa hatchery highlighted the importance of a strict management of egg and larva batches from the spawning events to the final destination. An optimal control of the temperature and other water quality parameters (pH, ammonia, bacterial concentration) during the transport time is crucial for meeting the best compromise between egg and larva survival potential and transport duration.
Small BFT larva samples were delivered for studying the lateral line development, the onset of digestive enzymes and the microbiotal diversity respectively from IFREMER (clear water) and Panittica Pugliese Spa hatchery (pseudo green water) husbandry tanks. BFT tissues were also sampled from adults during the broodstock sampling operations and delivered for studying the ontogeny of visual and neuroendocrine systems.
Regarding the ontogeny of visual and neuroendocrine systems issues, 3 genes have been cloned to date from BFT tissue extracts. The first, the arylalkylamine N-acetyltransferase-2 (Aanat2), was cloned after messenger ARN were extracted from the photosensitive pineal organ using a classical PCR approach and degenerated primers. The second and the third were cloned from mRNA extracted from the pituitary gland. The proopiomelanocortin (Pomc) was cloned using degenerated primers, whereas the growth hormone (Gh), had been cloned previously in BFT, and specific primers were used to obtained the full length sequence. Probes were made for the in situ hybridization (ISH).
Concerning the studies on lateral line development, preliminary results suggested a high level of conservation between zebrafish and BFT, greatly facilitating the analysis of BFT patterns. Labelling methods that were developed in zebrafish to reveal the position of lateral line neuromasts can be used in BFT for following the development of the lateral line system.
The behavioural pattern description for describing the BFT feeding and swimming behavior from hatching to juvenile stage with submersible cameras in mesocosm were initiated using sea bass larvae. The recorded images have been analyzed using special software that allows the estimation of swimming velocity and distribution in the water column.
Total mRNA was extracted from samples in BFT adults. We made purified by RT-PCR cDNA fragments of gene coding for elongation factor 1-alpha (EF1 alpha gene). EF1 is a housekeeping gene, necessary for quantifying the expression of genes of interest.
From BFT larvae, mRNA was extracted. cDNA fragments of genes coding for digestive enzymes were purified by RT-PCR, then cloned and sequenced: amylase and trypsine, which are some of the main digestive pancreatic enzymes involved in glucide and protein digestion, respectively, aminopeptidase and alkaline phosphatase, which are two of the main enzymes located on the brush border membrane of enterocytes, and a non specific dipeptidase, which is located in cytosol of enterocytes.
Relatively to microbiotal diversity, 19 and 21 DPH larvae were stored in a freezer with RNA later, and they became available for this purpose, only in 2009. They were treated for microbial analysis. The analytical methods were refined with intestinal samples from three ongrowing tuna euthanatized in the IEO facilities at Puerto de Mazarrón (Spain).
In 2009, larval rearing husbandry tasks were satisfactorily supplied in BFT viable eggs (IEO- 346,000, MRRA-Aquaculture - 450,000, IFREMER- 540,000 and HCMR- 386,000) from Tuna Graso (SELFDOTT-Spain: 1 486,000) and MareNostro (ALLOTUNA-Italy: 836,000). Previously bonito eggs spawned by a captive broodstock in IEO land-based facilities (Puerto de Mazarron) were delivered to IEO-Spain (198,000), IFREMER-France (48,000) and HCMR-Crete (52,000). The Bonito and BFT egg shipment results were notability improved and successful but still critical points have to be solved in terms of egg temperature pre-packing conditioning, transport boxes insulation related to transport ambient temperature and duration.
Several experiments were implemented in 2009 using the 3 targeted larval rearing methods: Mesocosm (HCMR, IEO), pseudo green water (IEO, HCMR and MRRA-Aquaculture) and clear water (IFREMER, IEO). In bonito, the rearing attempts were conducted until 23 (IEO) and 80 DPH (HCMR) in mesocosm, 80 DPH (HCMR) and 73 DPH (IEO) in pseudo green method, and 14 (IFREMER) and 20 DPH (IEO) in clear water method. In mesocosm swimbladder inflation rates were 60 (HCMR) and 78 % (IEO) while they ranged between 37 (IEO) and 60 % (HCMR) in pseudo green method and between 33 (IEO) and 85 % (IFREMER) in clear water method. In BFT, the rearing attempts have been conducted until 17 (IEO) and 60 DPH (HCMR) in mesocosm, 24 (MRRA-Aquaculture), 60 (HCMR) and 73 DPH (IEO) in pseudo green method, and 10 (IEO) and 20 DPH (IFREMER) in clear water method. In mesocosm swimbladder inflation rates was 50 % (HCMR), while ranging between 50 (HCMR) and 58 % (IEO) in pseudo green method, and between 1.2 (IFREMER) and 5 % (IEO) in clear water method. The presence or the addition of microalgae seemed to give a significant advantage to mesocosm and pseudo green rearing methods during the early development stages in terms growth and survival rates compared to the clear water rearing method. The feeding sequence including yolk sac larvae post Artemia strongly influenced both survival and growth highlighting the critical importance to satisfy the energetic and nutritional needs at that stage. A few BFT individuals ranging between 8 and 14 cm total length were produced and a small batch of 300 DPH old Bonito. Light intensity, lighting regime, hydrodynamic, surface cleaning, feed sequence and nutritional quality have been identified as the main factors to be fine tuned mainly during the larval rearing stage. As far as possible Bonito and BFT larvae were sampled, pre-processed and delivered to scientific topics according specifications related to each scientific issue allowing the first academic data monitoring in early stage development.
The study the ontogeny of the visual and neuroendocrine system partly launched in 2008 in BFT was complemented. In order to obtain molecular tools for the spatial and temporal localization of the corresponding transcripts, genes of interest (melatonin receptors, AANAT2) were partially cloned. Partial cloning of POMC and GH genes, as pituitary factors, was also obtained based on sequence fragments from NCBI data basis. For further qPCR experiments, primers were designed according to these partial sequences, and successfully tested. Conventional PCR revealed that MT1 and MT2 receptors both started being expressed in the larvae at 3 DPH, GH might be present in low quantity at 1 and 2 DPH and became more expressed at 3 DPH, POMC gene started being expressed at 3 DPH. Dig-labeled probes were used for in situ hybridization on BFT larvae collection starting from stage 0 DPH. Larvae submitted to POMC and GH probes exhibited the same pattern: the pituitary was clearly labeled at 3 DPH which suggests both the differentiation of this organ and its ability to express hormonal genes at this stage of development. Melatonin receptors hybridizations are currently being analyzed.
The early development of BFT posterior lateral line was examined. This sensory system analyzes the movements of water along the body surface, and is used in very many aspects of fish behaviour, notably prey catching or predator avoidance, and social behaviour (e.g. swimming in schools). We have described the embryonic and early larval development of this system. Many aspects of this development are highly conserved with other fish species examined so far, including the more basal model species Danio rerio (zebrafish), including the long-distance migration of a sensory primordium and deposition of five regularly spaced sense organs along the horizontal myoseptum, and of one terminal neuromast at the tip of the tail. Tuna is rather unique, however, in the size of its sensory organs at birth, suggestive of a remarkable sensitivity in prey detection. This may be related to the very fast assimilation of the embryo's vitellus, making early prey catching a vital necessity.
Ontogeny of digestive tract in Bonito and BFT was rather extensively described in a comparative way between the 2 species, bonito and BFT, and the 3 investigated rearing methods focusing on the digestive system description with emphasis on the appearance of the different organs, their physiological role and function and relations on larval feeding preferences. In BFT no difference existed in the time of organ appearance between the rearing methods but the rate of development seems to be significantly affected by the temperature level which differed between sites and rearing methods. In mesocosm, larvae preferred unidentified preys grown spontaneously until 7 DPH when rotifers were detected in the stomach while they were delivered from DPH 2. In pseudo green method the first rotifer presence in the stomach was not noticed before 4 DPH even thought they were delivered two days before. On the same way, Artemia and yolk sac larvae were detected between 2 and 3 days after their first delivery to rearing tank. Even though Bonito were reared in lower temperature their digestive tract development was notably faster than BFT as illustrated by the gastric glands which appeared in 3 DPH in bonito vs 11 DPH in BFT.
The activity pattern (gene expression assessed by q PCR) of digestive enzymes as indicators of physiological status of larvae and antioxidant enzymes as indicators of stress during development (3 to 20 DPH) was investigated and compared between the 3 targeted rearing methods. The different digestive evolution pattern between rearing methods suggested a poor physiological status in larvae reared in clear water associated with a poor growth compared with pseudo green ad mesocosm rearing method. Antioxidant enzyme profiles suggested a higher stress level in clear water reared larvae compared with the 2 other rearing methods.
Regarding the immune system ontogeny, a preliminary approach has been conducted with the bonito and BFT samples collected during the early stages in mesocosm (HCMR). Molecular probes are in preparation and the application of RT PCR related to immune system expression is being evaluated for further analysis in summer 2010.
The evaluation of the microbiotal diversity of the digestive tract was limited by the availability and size of the bonito and the BFT samples. Consequently, it was decided to upgrade the discrimination level of the study by working at the individual level. Individual genomic DNA was extracted from larvae (3 to 12 DPH) sampled in clear water rearing (IFREMER) and pseudo green rearing (IEO) This preliminary approach revealed a microbism in the 12 DPH larvae reared in the clear water method.
A bacterial characterization has been performed in Bonito larvae (1 to 20 DPH) reared in mesocosm (HCMR) and 28 bacterial strains were identified and phenotypically characterized before being identified later on. In addition, the total bacterial content in terms of total counts of the aerobic heterotrophic gut microbiota was determined. An exponential increase in numbers of bacteria per larva was found in bonito larvae as shown in studies with other species. The numbers of Vibrio were detected at notably very low levels in the mesocosm rearing units. Samples were taken as well for analysis for non-culturable bacteria. A bacterial strain belonging to the Phaeobacter genus was isolated from a mesocosm with bonito larvae 1 DPH. This strain exhibited in vitro inhibition against Vibrio bacteria, as shown with other representatives of its genus.
In 2010 and 2011, encouraging survival and growth results allow us to postulate that BFT domestication should be feasible in a next future. But, larval rearing is still at the research stage and has not reached the status of a commercial technology.
Mesocosm and pseudo-green methodologies implemented in large tanks have been substantially improved as it is evident from the survival of the best rearing trials (2.5 -3 % at 25 DPH). The knowledge and knowhow bases in larval and fingerlings rearing have been substantially documented both at the academic and at the rearing levels and gave us clear indications for improving rearing techniques and biological performances. But key issues still need further investigations.
a) Currents, luminosity parameters and feed quality which have a significant effect on vision, feed intake, growth and survival as soon as the early stages have to be fine tuned.
b) It is still impossible to sustain an acceptable growth and survival without YSL (prelarvae) post 17 DPH whatever the rearing methods. So, even if weaning of juveniles (post 25 DPH) on artificial diet was achieved, the later have to be improved and lead to better results.
c) Considering these results and comments we can propose a first set of recommendations for a BFT larval rearing protocol for implementing mesocosm and pseudo green water techniques in large tanks.
Several batches of fertilized BFT eggs were delivered at IFREMER (France), HCMR (Greece) and IOLR-NCM (Israel) facilities during 2010 and 2011 spawning seasons originated from the broodstock kept at Caladeros del Mediterráneo facilities at Mazarrón, Spain. Egg shipments were performed using the standard Cargo services of airlines facilitated by a specific company in Spain in the Greece and Israel cases and transported by car and bus, in the IFREMER case. Egg quality was significantly affected by the transportation conditions and in particular temperature. When variations overdraw 2 degrees, abnormal embryonic development and therefore abnormal individuals were observed.
Regarding Atlantic Bonito (Sarda sarda), a total of 232,900 larvae were cultured in IEO facilities during 2010 season. The swimbladder inflation started between 3 and 4 DPH. The swimbladder inflation rate ranged between 67 and 80%. Between 8 and 12 DPH, an increase in the mortality was noted in some of the batches. Bonito larvae lied in the surface of the tank with a reduce motility and some of them with swimbladder hyperinflation. An analyses looking for virus was done with negative results. After 3-4 days and a couple of prophylactic baths of formaline (40 ppm during 1 hour), the problem started to disappear, seemed that the problem was related with stress caused by high density and maybe an unbalanced feeding. A trial to test the effect on bonito larvae of two different enrichment methods: Origreen (Skretting) and DHA Protein Selco (Inve) for rotifer was carried out and results in survival rate and growth did not offer significant differences (p greater than 0.05). Bonitos were successfully weaned at 22-27 DPH with a paste made in the IEO and composed by minced bonito meat, krill meal, hydrolyzed fish meal, squid meal, fish oil and vitamin premix. From 40-45 DPH onwards the paste were replaced by 2 mm wet pellets manually distributed. In spite bonitos accepted the paste and had an acceptable survival (28.7%) and growth (if we compare with results obtained the previous years when weaning was done with raw fish), mortality continued too high, more than 70%). Besides we must to consider these trials began about 19 DPH and they finished about 31 DPH (average terms). Weaning onset was between 22 and 27 DPH (average 25 DPH) and as YSL were overlapped with paste for 4-5 days, we cut off YSL only a couple of days before being extracted and counted from the tanks. So, some mortality due to weaning was still coming in the following days.
Bonito eggs, obtained in 2010 from a natural spawning of captive broodstock from IEO Mazarrón, Spain, were delivered to the HCMR facilities (Crete-Greece) by air cargo. 60.000 eggs were received (T=21, pH=8.3 O2 =22 mg/l) and were incubated equally in three tanks 500 l tanks. During embryogenesis, egg hatching and the autotrophic larval stage, aeration was provided in the tanks by means of a wooden diffuser (150-200 ml min-1). Aeration was stopped one day after the first day of feeding. From the day of first feeding until 19 DPH, water circulation was autonomous for each tank. During this period the only source of water renewal of the tanks was the constant flow over the feeding tanks. From day 20 onwards, the tanks were connected to a biological filter and the inflow to the tanks was progressively increased. A surface skimmer was installed from 5 to 9 DPH to remove the oily layer from the water surface in order to allow proper swim bladder inflation. The photoperiod was 24L:0D from 0 to 40 DPH, when it was changed to 16h light: 8h dark.
Feeding was executed, using an automated feeding system. In this way feeding could be executed frequently, labor efficient and 24 hours a day. Feeding was based on enriched rotifers and Artemia, Instar II nauplii. Prior to distribution to the larvae, rotifers and Artemia were enriched with commercial products (Selco, INVE) and were maintained in a solution with phytoplankton. From 23 DPH onwards eggs of various species were incubated in a separate tank. The eggs and larvae were fed manually to the bonito's. On 10 DPH hand feeding of dry feed was started. The dry feed was hydrated in the feeding tanks before feeding. On 14 DPH an automatic dry feeder was installed, adding small amounts of dry feed (Proton 2/4, INVE) every 15 minutes. From 10 to 67 DPH defrosted seabream eggs were fed manually several times per day. From 27 to 67 DPH pieces of defrosted mackerel were fed manually several times per day. The pieces of mackerel were covered with dry feed in order to enrich the diet and to adapt the juveniles to dry feed. Water quality was similar in all tanks. The temperature in A2 was higher during the later stages of rearing when no fish were present in the other tanks. Mouth opening took place on 4 DPH. The swim bladder started to develop from 5 DPH and on 16 DPH the swim bladder was inflated in 100% of the individuals. The beginning of flexion started on 7 DPH. No deformities were observed. Growth was similar in all tanks. The exponential growth rate of TL was 5.6% day-1 and the one for BW for 30-58 DPH was 11% day-1. The initial mortality was high and it decreased with population size. Observed causes of mortality were cannibalism, the presence of histophages cilliet and the jumping out of the tanks. Cannibalism was observed between 8 and 40 DPH. Parasitic histophage ciliates were present in A1 between 14 and 19 DPH. They were first ineffectively treated by adding 25ppm formalin to the system. An increased water refreshment of the tank on 19 DPH successfully removed them. At later stages from 43 DPH onwards, fish jumping out of the tank became an important cause of death. This was the cause of death of the last surviving fish. During the rearing for a period of 25 days tanks A1 and A3 were also used for sample collection, causing an additional artificial mortality. During the rearing no developmental problems of the larvae were observed. Therefore the main focus of successive rearing trials should lie on reducing the mortality. A major problem during the rearing was the unsuccessful feeding resulting in inadequate cover of the larvae nutritional requirements. Cannibalism can be counteracted upon by providing good alternative food sources. The provision of de-frozen eggs and yolk sac larvae as alternative food sources was not effective in completely eliminating cannibalism, although it probably led to a significant reduction. A remarkable observation was that the juveniles hunted equally sized conspecifics, which they were not able to swallow and ultimately spit them out. This observation indicates that the juvenile bonito prefer a relatively big prey size, as smaller preys were plentiful at all times. Although dry feed was provided during most of the period, the actual feeding on dry feed was not observed. The juveniles did ingest small amounts of dry feed when feeding on mackerel covered with dry feed. The jumping out of fish can be relatively easily counteracted upon by heightening the walls of the tank. Although just a few fish jumped out, it was a relative big part of the total population at the times it occurred. Although the goal of rearing bonito to weaning was not completely reached, it was a rather successful attempt. The development and growth of the fish during this rearing trial were good, while an ongoing mortality caused a premature termination of the trial.
A total of 1 926,000 Bonito eggs were collected during 2011, all of them coming from Z3 tank. This broodstock tank was composed by about 50 fish born in May 2009 and May 2010. Bonitos spawned during a period of 35 days, and fertilization rate was 81.7%, ranging between 65 and 95%. Spawning occurred at nightfall and eggs were collected early morning every day. The Bonito larvae were fed with Rotifer, Brachionus plicatilis, which was completed up to 5 rot/ml. a couple of times every day. Rotifers were fed on Chlorella and three hours before being offered to the bonito larvae, they were enriched with Origreen +, a special formulation from Skretting based on Origreen but with an extra supplements of micro-nutrients. Artemia was added to satiety four times per day. Instar II of Artemia were enriched over a minimum of a 14 hours period with Origreen + and then offered to bonito larvae. Yolk sac larvae (YSL): Increasing amounts of sea bream YSL were added every day to the bonito larvae from 10-15 DPH onwards. Sometimes, they were fed on bonito YSL instead of sea bream YSL. The weaning were carried out at 22-26DPH with the special dry food developed for BFT. From these production batches, a total of 4,727 23 DPH bonito larvae were produced, which means a 2.7% of survival. In parallel, several trials to study the effect of light intensity, the different enriched diets for rotifers and the density of bonito larvae on growth and survival have been carried out.
A total of more than 4 million BFT eggs were incubated in the IEO facilities, in 2010. Incubators were 400 and 1,000 l tanks, were BFT eggs were placed to hatch and kept one or two days after hatching. Then they were counted and after separating dead eggs and larvae a part of them were moved to larval rearing tanks. A trial was done to assess the effect of iodine on embryonic development of eggs. Iodine treatment is widely use in aquaculture, and it is showed very efficient in the control of some diseases. Short period treatments of iodine are applied to eggs to disinfect and even are duty to transport eggs to some places (some countries like Israel demands it before introducing any kind of fish eggs). Four different dose of iodine were tested during 10 minutes: 0 ppm, 50 ppm, 100 ppm and 200 ppm. The results showed that egg survival rate decreases with the increase of iodine concentration, but the only significant decrease (p less than 0.05) is between 50 and 200 ppm. 50 ppm should be a recommendable dose and even when differences are not significant, 100 ppm it seems also quite high. A total of 1,204,000 newly hatched larvae were cultured in IEO facilities in 2010. From 1,204,000 eggs/larvae we obtained 2,776 larvae (25 to 32 DPH) ready to be weaned. In Mesocosm Tank of 45,000 l was inoculated with 18,000, 2 DPH bluefin tuna larvae. This low stocking density allows us to refer to this culture method as mesocosm. A total of 310 larvae, which means 1.7% survival, were produced. Using pseudogreen water method, the rest of larvae were cultured at greater density (6-12 larvae/l), which should be referred as pseudogreen techniques whichever would be the volume of the tanks. A high density trial (40 eggs/l) was also done in a couple of H tanks. Finally, a total of 2,466 larvae of about 26 DPH, with an average of 0.21% survival rate, were obtained.
Also the effect of several parameters on larval growth an survival have been tested. A trial was developed to test two enrichment methods (Origreen, Skretting, and Spresso, Inve) and a cofeeding diet (Gemma, Skretting).The first approach indicates that it seems that Origreen offer better results than Spresso and that tuna larvae did not accepted dry food during the first 20 days.
Also the different conditions also upwelling (a recirculation system) versus aeration have been tested. The results seem to indicate: a) Increased aeration helps larvae to distribute homogeneously. This is particularly important from 17-18 DPH onwards, with the onset of cannibalism. b) It is possible to eliminate Artemia period but there are a reduced growth (1/2 or 1/3) which leads to delay the beginning of YSL period. c) While tuna has a relatively good performance with minced fish (bonito) as a weaning diet, wet paste used with bonito (see above in bonito trials) it seems offer a worse results and it is not a good diet for weaning.
A high density trial was done in two 1,500 l tanks. 60,000 just fertilized BFT eggs were inoculated in each one of both tanks, which means a stocking density of 40 eggs/l During the first week, larvae went ahead with an acceptable survival rate, but from 10 DPH onwards, a high mortality. This led to a very low survival rate, and finally we decided to finish the trial at 23 DPH, when only 10 larvae were alive in both tanks. In conclusion we can say that with the actual level of knowledge, it is very difficult to be able to culture BFT tuna larva at high density.
A trial to compare the two methods employed to avoid larvae sinking during the early stage of development in bluefin tuna larval rearing was conducted. Two methods deal with kept larvae in suspension using air current or water current. In order to watch larvae behaviour, two 500 l transparent tanks were used in spite of this prevent us from carrying out replicates of the trial. One had surface water inlet and stronger aeration, while the other had bottom water inlet and mild aeration. Both of them had water outlet in surface, so idea was to create an upwelling current inside the tank to prevent BFT larvae sinking to the bottom of the tank. Larvae tend to sink mainly in night time, when light is off and they decrease their swimming movements. In this moment it seems very
important to help the larvae to keep themselves in suspension by upwelling current. If current is too strong, larvae can injure by hitting against the walls of the tanks, but if it is too weak, larvae fall to the bottom of the tank and they die. At 25 DPH survival rate was 2.4% in bottom water inlet tank and 1.2% in surface water inlet tank. From this results it seems that is better to kept larvae in suspension with water instead with aeration, but it should be necessary to test every single tank, in order to avoid an excessive water flow.
The weaning of BFT larvae was carried out in 2010 in the IEO facilities on a minced fish basis. We used several species of fish (bonito, sand eel, mackerel, anchovy, sea bream, sardine, etc.), but mainly bonito. The conclusion is that it is possible to wean BFT larvae on a minced fish basis, and they prefer fish instead squid (Sepia sp.) as weaning diet. However mortality went on too is high and growth also should be improved. Another trial was conducted to test a dry food, Gema (Skretting). A total of 235 BFT around 34 DPH were placed in four 1,500 l tanks. Two tanks were weaned onto bonito meat and the others two were weaned onto Gema diet. Results obtained with dry food were quite worse, and improved diet should be used in order to wean suitably BFT larvae.
More than that, a couple of weaning attempts were done in square 12 m3 tanks. Weaning diet was bonito mix with another fish, mainly Scomber or Phycis. First attempt ended with bad results because a disease which caused great mortality (more than 90%), and even the tuna surviving had a increased mortality during the first ongrowing days Second attempt was made and it obtained better results, about 25% survival rates. BFT fingerlings had a weight of 3-4 g at 38 DPH. A careful cleaning of the tank, and the use of a greater rate of sand eel, could explain the best results obtained.
Several preongrowing attempts were carried out in IEO during 2010. Tuna were fed mainly on bonito and sand eel. In spite they accepted well the minced fish, mainly sand eel. Mortality went ahead due to several reasons, but mainly collision and maybe an unbalanced diet. In order to prevent collision, tank walls were painted on a bright green colour, and a dim light during night to keep continuous photoperiod was used. But collision went on and from 45 DPH was the responsible of the most of death. At the end of this trial, bluefin tuna had an average weight of 18 g, but the greater fish weighted more than 50 g.
Two transports to a cage of 25 m, diameter and 10 m depth were carried out on August 20th and 21st 2010. 7 died during transport (11% mortality), and a total of 56 fingerlings were introduced to the cage. During the first hours, a great mortality occurred in the cages: 24 tunas died immediately after arriving, in the first 2h (43%). 48 h after the transportation, accumulated mortality was 75%. 96 h after the transportation, mortality was 89%, and only there were 6 tunas in the cage. A week later, mortality was 93%. There were 4 surviving bluefin tuna. This tuna fed on minced fish (sand eel at the beginning and mainly sardine from September onwards). They started to accepted food from the day after transportation, but ingestion rates were lower than in the tank. Last tuna died after 45 days in the cage (110 DPH) with about 100 g.
To test the effect of transportation three dummy trials were conduced. A couple of fingerlings were placed in a plastic bag like those used for transport purpose during 24 h. After this, 1 tuna was died and the other was still alive. 2. Five BFT were subjected to handling to extract from the tanks and then they underwent a transport simulation (4 h). After this they were placed in a 5 m3 tank. 2 tuna died during transport and 3 during the next 48 h. Five BFT were subjected to handling to extract from a tank and then they were place in another 5 m3 tank. 3 tuna died during the next 72 h, and a couple of fingerlings survived.
Also another trial moving 15 fingerlings to a greater tank (a 100 m3 tank) early September 2010 was carried out. These were smaller tuna (average weight of 5 g), and mortality after 72 hours was 7%. Tunas were not too enthusiastic with minced fish and in spite of eating, hey were reducing their ingestion rate through following weeks. Their growth was not good and most of the fish died during September. At the beginning of October there were only two tuna, and the last one died on October 20th 2010, when it was 110 days old and it weighted 52 g.
These results seem to point out that handling is mainly traumatic for tuna greater than 5-10 gr. tunas. The only action of collecting from the tanks led to a great mortality that is increased if after this they undergo new stressing situation like a long transport. More than that, it is necessary to acquire knowledge about nutritional requirements of fingerlings and to improve nutritional quality of supplied feeding.
In 2011, from total of 162 millions eggs were collected during this year in both cages in Cartagena (Spain), and transported to the IEO facilities in Mazarrón. A sample of 72 eggs coming from every spawning event was incubated in three different plates. Each plate contained 24 wells of 2 ml capacity, and one egg was placed in each well. The hatching rate results were very similar in both cages (Mean ± SD): Cage R1 81.7±12.1% and 87.5±10.2% for Cage R2.
A total of 1.7 million eggs were cultured at the IEO facilities in 2011.All the BFT larvae were reared with similar rearing techniques and feeding schedule: a) Rotifer: From 2 to 16 DPH. Rotifer, Brachionus plicatilis, was completed up to 10 rot/ml. a couple of times every day. Rotifers were fed on Fresh Chlorella and three hours before being offered to the tuna larvae, they were enriched with Origreen+ made by Skretting and also tested in bonito. b) Artemia: From 13 to 19 DPH. Artemia was added to satiety four times per day. Instar II of Artemia were enriched over a minimum of a 14 hours period with Origreen + and then offered to tuna larvae. c) Yolk Sac Larvae (YSL): Increasing amounts of sea bream YSL were added every day to the tuna larvae between 17 and 30 DPH. Sometimes, they were fed on tuna YSL instead of sea bream YSL. d) Weaning: It started when larvae were between 25 and 30 DPH. Nevertheless there were some important differences according to the size of the tanks: Production tanks (40 m3), Intermediate tanks (5 m3) and research tanks (1.5 m3, and transparent tanks, 0.5 m3).
In a 40 m3 tanks, three production batches were cultured, with a total initial number of 1.25 million eggs. When larval rearing finished, tuna larvae were taken out, counted and placed in some new tanks to wean them. A total of 17,300 23 DPH larvae were produced. This means a 1.38% survival rate, but if we consider only the two first batches, the survival rate was 2.09% The main difference was the amount of yolk sac larvae, which was extremely low in the last batch.
Larvae coming from the first batch, weighted 107±48.5 mg (ranged between 48 and 107 mg) and they have a total length of 22.3±0.40 mm (Mean±SD). Larvae coming from the second batch, weighted 158±58.5 mg (ranged between 73 and 281 mg) and they have a total length of 24.9±0.3.41 mm.
The effect of four light intensity (0, 500, 1000 and 2000 luxes) on growth and survival of BFT larvae has been tested in twelve 1,500 l tanks, in triplicate. The initial density was 7.2 eggs/l. (10,800 eggs/tank). The complete darkness (0 lux) led to a fully mortality in the first week, and even when there were not significant differences in survival and growth rate at other intensities, it seems that the greater intensity led to slightly worst results.
A second experiment to study effects of two different enrichment diets for rotifers on growth and survival of tuna larvae has been carried out testing two diets: Origreen and Origreen+ (with extra supplement of micro-nutrients). The trial was carried out in 1,500 l cylindrical tanks in triplicate. The initial density was 10 larvae 0DPH/l. (15,000 just hatched larvae/tank) The feeding schedule was: 10 rot/ml adjusted twice per day. 20 ml of Fresh Chlorella V12, added also twice per day. The experiment when larvae were 15DPH. The results showed that using Origreen+ the survival rate was quite higher (3.01±1.64% versus 1.03±0.10%). In growth, differences were not significant (p greater than 0.05).
A third experiment to study the effects of three different enrichment diets for artemia on growth and survival of BFT larvae was carried out. Three different diets to enrich Artemia: DHA Selco (Inve), Origreen and Origreen+ (Skretting) were tested. The trial was carried out in 150 l cylindrical tanks in triplicate. Differences among three treatments were significant (p less than 0.05); the best results were achieved in BFT larvae fed on DHA Selco enriched Artemia, and the worst were those in larvae fed on Origreen enriched Artemia. The wet weight ranged between 7 and 11 mg, and it was significant lower (p less than 0.05) in tuna larvae fed on Origreen enriched Artemia than DHA enrichment, while there were not significant difference among three treatments with regards to total length (all the larvae reached about 10 mm).
A four experiment to test the effect of BFT larvae density was carried out. Two different stocking densities (10 and 30 eggs/l) in 500 l transparent tanks without replicates were tested. Even when we cannot obtain conclusive results, it seems that is possible to increase density achieving acceptable survival rates (1.4%), during the first stage of development (before start piscivorous feeding), so that obtained results in growth and survival rate were very similar. Nevertheless, further research must be implemented.
Several weaning attempts were implemented in IEO during 2011 season. In the same way that in larval rearing, two different techniques have been used: a production technique with 5, 15 and 20 m3 tanks in which a mixture of weaning diets were employed, and experimentation technique with 1.5 and 0.2 m3 tanks in with different weaning diets were tested. In production tanks, dry food consisted in a mixture of two different diets from Skretting (TD and LM). The LM diet was an experimental dry diet designed to support nutritional requirements in tuna larvae. In order to try to improve the results, the dry food was distributed to satiety using automatic belt feeders. Tuna lasted some days in being fully adapted to new food, but after 5-7 days they showed a good acceptance of the food. Survival at 39 DPH was 28.5%, ranging between 19.3 and 33%. At the end a total number of 1900 tunas (20.8% survival) were produced and sent to the cages when they were 39-44 DPH. Until 35DPH, most of mortality was due to weak fish unsuccessfully weaned, but from this moment onwards, main mortality was caused by collision.
Two of experiments were made to compare the different weaning diets, using tuna larvae 28 DPH. These experiments took ten/twelve days and the results were compared in survival and growth terms. First experiment was made in 150 l tanks to compare 3 Skretting diets: GM (Gemma Micro, a commercial diet), LM and TD (two experimental diets) and raw fish (minced sand eel previously frozen). A total of 320 larvae were used. They were 29DPH, weighted 153±23 mg and their total length was 22.3±1.6 mm. During the first five days, sea bream YSL and weaning diet was offered, and from 34 DPH onwards, weaning diet was the only source of feeding. Experiment was finished when larvae were 38 DPH. Best results were achieved with LM diet, which produced higher survival than TD diet and Fish diet (p less than 0.05). No significant differences (p greater than 0.05) were founded between LM and GM, and between TD and Fish diet. With regards to growth, the gain in total weight was significantly lower in TD (0.49±0.2 g) than in LM diets (0.18±0.1 g), meanwhile no differences were observed in the other diets. Second experiment was made in 1500 l tanks to compare the three different Skretting diets: GM (Gemma Micro, a commercial diet), TD and LM (two experimental diets). A total of 495 larvae were used. They were 29DPH, weighted 250±108 mg and their total length was 29.65±4.45 mm. During the first five days, sea bream YSL and weaning diet was offered, and from 34 DPH onwards, weaning diet was the only source of feeding. Experiment finished when larvae were 40 DPH. In spite no significant differences (p greater than 0.05%) have been observed among the different diets in weight gain and in survival, LM and GM diet produced the best results.
In IFREMER facilities, the common live prey and larval rearing conditions in bonito and BFT have been studied. To evaluate the effect of probiotoic addition via the enrichment media of live preys we added Bactocell®, Pediococcus acidilactici MA 18/5M (1011 CFU g-1) at the concentration and time exposition of 1g.m-3 2 hours and 5g.m-3 12 hours respectively in rotifers and Artemia. The total length of bonito larvae was significantly higher as soon as 4 DPH when Bactocell enrichment was used under 24L:0D lighting regime (LL) but due to lack of repetition in treatements it was not possible to conclude. In BFT no significant effect of probiotic enriched preys on larval growth (total length) was observed. In BFT, it was demonstrated that between 2 lighting regimes 18L:6D (LD) and 24L:0D (LL) the first one led to significantly higher growth (total length).
During this experiment we globally observed a significant improvement in term of early behaviour and in some extent reduction of early mortality. We hypothesize that is due to a better hydrodynamic management related to water and air injection from the bottom of the tank at an acceptable flow rate. The new surface auto cleaning device we used improved the surface cleanliness but the low swimbladder inflation rate we got could be explained by a relatively limitation of its efficiency during 2 and 3 DPH when we tried to improve the setting of this new device. Mortality picks remained unexplained. Chronic mortality could be related to inadequate prey global quality taking into account that bonito seems to significantly take advantage in terms of growth and survival in feeding YSL as soon as 10-12 DPH.
Clear water rearing experiments was carried out in bonito at Palavas facilities, in 2011.Probiotic enriched living prey effect was tested under LL lighting regime in one experiment and LL lighting regime was compared against LD in a second one without any probiotic enrichment. From these two trials, we concluded that 1) probiotic enrichment didn't promote a higher growth and survival by comparison with the control and 2) LL lighting condition seems to promote a better early survival and delayed the hyperinflation crisis which was observed in the two treatments, 3) in each experiment we confirmed that is was not possible to sustain a correct growth and survival more than 15 days without using any yolk sac larvae to replace enriched artemia. In a third experiment we demonstrated that an early weaning using Gema Micro (Skretting) allowed a successful transition between rotifer and micropellets within two days from dph 5 to 7 but this micropellets as single feed source in our feeding conditions was inappropriate to sustain a correct growth and survival
A second study was realized for the evaluation of the effect of volume on rearing performance: a comparison between 500 l and 2000 l tanks. The duration of the trial was 20 days. There was no significant effect of tank volume within treatments on TL evolution and in the case of 500 l the individuals grew exponentially with a rate of 0.253/d while in the 2000 l the rate was 0.175. There was a significant difference between the two conditions. In terms of survival a high variation between treatments was observed. Mean survival was higher in the 2000 l tanks although not significantly different.
In the knowledge on the immune system of the BFT larvae and for the purposes of gene expression studies using the methodology of real time PCR/quantitative PCR (qPCR), adequate reference genes have to be identified and primers for qPCR have to be developed. A first set of 6 candidate genes known to be involved in the immune response were selected for gene expression analysis. Some of the genes are not yet described for tuna, therefore homologues sequences for several perciform teleost species were retrieved from the NCBI nr and EST databases and primer were designed at conserved sites. Primers for the reference genes and genes related to the immune system have been constructed.
Biodiversity in microbiota associated with larval tuna and bonito have been assessed by the end of rearing experiments. The variability of microbial profiles was observed even at 2-3 DPH, possibly due to individual variations in the timing of mouth opening among the larvae. It seemed that this initial microbiota originated in the rearing water, with likely poor influence of the environment before hatching. Indeed, there was a clear difference at mouth opening between the samples from Mazarrón (Spain), and those from Palavas (France), though the eggs were incubated in Mazarrón in both cases. There was also a clear difference between the samples of 2009 and 2010 in these two hatcheries, confirming the unspecific nature of the initial microbial community in larval tuna. A general feature observed in all cases was the increase in individual dissimilarity between mouth opening and 10-12 DPH. This increase seemed to occur relatively early, as it was already observed at 7 DPH in Palavas in 2009. The highest difference in the microbial profiles between groups was observed between 7 DPH and the two other dates of sampling. In this particular case the individuals surviving at 12 DPH did not seem healthy, and these results should be interpreted cautiously. The general trend seemed that the microbiota settling in the gut of the larvae were different from those observed initially at mouth opening, when the microbes associated with external surfaces were likely still dominant. These settled communities appeared yet unspecific, are they remained different between years and hatcheries. There was no general trend of microbial settlement in terms of diversity, except a higher dominance, and lower evenness, equitability and richness in 2009, compared with 2010 in Mazarrón and Palavas, while the indices from Heraklion in 2010 resembled those from the other hatcheries in 2009. Microbiota associated with tuna larvae appeared therefore unspecific, and likely selected among the environmental microbes present in the hatchery during the rearing period. It seems possible to exert some influence on the selection process by modifying the rearing conditions. This may be illustrated in Heraklion by the higher intra-group similarity among the individuals reared in mesocosm, compared with those in intensive conditions. Future replications are necessary to confirm this finding. The triplicates available in Palavas gave more weight to the observation that the probiotic treatment seemed to regulate the microbial diversity, and the intra-group similarity of microbiota associated with tuna larvae, but that need also to be confirmed in further replications and sites of experiment.
Sampling for analysis of bacterial colonization has been done in Mesocosm systems in the case of bonito and in both Mesocosm and artificial rearing system in the case of BFT larvae. In most marine fish larvae there is an exponential increase of bacteria in the gut during the first weeks of rearing when the larvae consume rotifers and Artemia. In the case of BFT larvae there is an unusual pattern compared with colonization of bacteria in other fish larvae, as the larvae reached a peak 5 DPH and thereafter lower values were observed. This could be explained by the fact that rotifers are a large source of bacteria for fish larvae. In the case of tuna, the larvae may switch over faster to consumption of different prey items such as yolk-sac larvae instead of rotifers. Yolk-sac larvae probably have a much lower bacterial load than rotifers. In the case of bonito larvae reared in mesocosm in HCMR (Greece), the pattern was closer to what is normally observed in rearing of marine fish larvae. The culturable microbiota of tuna larvae was investigated during the season of 2011. A quantification of microbiota in tuna larvae sampled from mesocosm in terms of counts in marine agar and thiosulfate-citrate-bile salts-sucrose (TCBS) agar. In the case of yolk-sac larvae the bacteria present in the larvae were showed a variability between different groups, so TCBS counts were relatively low. The numbers of total counts increased exponentially in larger larvae, and members of the Vibrio group became dominant.
WP4 Nutrition and feeding
In the part devoted to Nutrition of juveniles, 25 Young BFT 0+ (YBFT) were captured alive on the shore of Murcia (SE Spain) in Autumn 2008 with the method so called in Spanish "curricán" (a term meaning to fish by trailing a baited line along behind a boat). They were placed on board in cylindrical plastic tanks (1.45 -1.65 m Ø, 1.20 m height, containing 1 200 litres of sea water), transported to the project facilities in El Gorguel (Cartagena, SE Spain) and then released in a circular 25 m-diameter floating cage which was 20 m deep. Whilst close to 50% of the captured fish arrived alive to the cage, the survival rate after one month in the cage was high (75%). The main mortality in this process was due to difficulties in freeing fish from the hooks, although the hooks were barbless. No YBFT grouping was observed in any case, so it was impossible to use the purse seine method for massive capture as planned. Another 29 YBFT were captured dead. The stomach contents from these 29 dead YBFT have been analyzed and the whole fish has been minced and lyophilized for later body composition analysis. The first stomach analysis show that cephalopods are the main prey (53% in weight), followed by fishes (28%) and crustaceans (2%). The main prey species is the cuttlefish Illex coindetti.
One cage of 25m of diameter x 20m deep which contained 15 YBFT captured in Autumn 2008, was used to start the artificial feeding trials in January 2009. In February 2009 we started the weaning trials using different kinds of artificial food. Weaning was achieved. For weaning purposes and may be for general feeding we have a feed that the tuna will eat. Due to the small number of fish in the cage, we were not able to draw conclusions on growth in this year.
From the body composition analyses it seems that YBFTs have nutritional requirement that are different from most of other Mediterranean aquaculture species. YBFT have high protein and low fat content. Comparing the total n-3 and EPA+DHA concentrations among various marine species we determined significantly higher values for YBFTs. Comparing the amino acid profiles of YBFT and other teleosts we observe that in most cases YBFT have higher amino acid concentrations. The most significant differences appear for histidine, methionine, lysine, leucine and aspartic acid. The tuna feed formulation must consider these differentiations and include appropriate ingredients in order to achieve higher dietary amino acid and protein concentrations, lower total lipid content, higher dietary n-3 concentrations and higher vitamin C and E concentrations. These results will form the guidelines for the formulation of the next nutritional trial of YBFTs.
The stomach analysis carried out in 2009, showed that most of the organic content in weight consisted of fish, but cephalopods and small crustaceans were also present in the sampled stomachs. Fish are the main prey (37% in weight), followed by cephalopods (11%) and crustaceans (5%). The main prey species is the fish round sardinella (Sardinella sp.). These results contrast with the ones obtained in 2008 where cephalopods were the main prey (53% in weight), followed by fish (28%) and crustaceans (2%). The main prey species was the shortfin squid Illex coindetti. In conclusion, juvenile BFT predate on a huge variety of preys. The diet composition may vary greatly from one year to another depending on the availability or abundance of preys. Therefore, this age class (0+) can be considered a generalist feeder.
WP5 Dissemination
In WP5 devoted to Dissemination, a project website has recently been commissioned (see http://www.selfdott.org online) and the first newsletter containing details of the first annual meeting together with a guest workshop are in the process of being completed. In 2010 and 2011, dissemination activities were carried out including specialized sessions at the EAS meeting in Rhodes and a stakeholder workshop in Madrid.
Potential Impact:
RESEARCH AND INNOVATION
Type of Exploitable Foreground: General advancement of knowledge
Exploitable foreground (description): The project SELFDOTT has maintained broodstocks of bluefin tuna over a period of three years in captivity and has successfully obtained fertilised eggs in large numbers. These eggs have been hatched and larvae successfully reared and fingerlings which had successfully been transferred to open sea cages for on-growing. Young tuna larger than 2kg have survived for a period of greater than 9months DPH.
Confidential: no
Exploitable products or measures: Monitoring of broodstock populations. Induction techniques for providing "eggs on demand", transfer protocols for eggs. Larval rearing of tuna, fingerling production. Grow-out of tuna.New feeds for tuna larval stages, weaning diets.
Sector of Application: Sustainable Aquaculture
Timetable for commercial use: The results will be used by individual participants within the framework of both national and EU coordinated programmes. This foresees the use of these technologies within 2012 and also continuing fall at least the next two years as part of a follow-on project TRANSDOTT
Patents or other IPR exploitation: At the present time there are no patent applications under consideration since much of the work has been or is the subject of scientific publications or has been presented at international conferences. All partners of the SELFDOTT have access to the results and may continue to use them. There are no licences granted at the present time. A own and other.
Owner and other Beneficiary: As laid down in the consort to agreement foreground shall be the property of the beneficiary carrying out the work generating that foreground. The owners and beneficiaries are therefore the full partners of the SELFDOTT consortium.
CONCERNING OWNERSHIP AND ACCESS FOR SELFDOTT PARTNERSHIP IN GENERAL
Regarding the feedback from the evaluators, here is what the colleagues of the IPR service want to stress 'what matters is not to be the OWNER of the results but to ENJOY the RIGHT to USE them (i.e. have access to them). '
1) SELFDOTT - CA
The CA is based on the DESCA model, it has rather standard clauses. Those relevant for our purposes are from art.8 to art.10 (plus attachment 1).
The basic principles applies as regards the ownership of the foreground (art.II.26 of the Annex II to the Grant Agreement: ftp://ftp.cordis.europa.eu/pub/fp7/docs/fp7-ga-annex2-v6_en.pdf): Foreground shall be the property of the beneficiary carrying out the work generating that foreground
As regards results developed jointly by two or more partners, art.9 allows each joint owner to use the jointly owned foreground and to non-exclusively licence it to third parties (provided that notification and compensation conditions are fulfilled).
As regards access rights, our interest is only focused on access rights for Use: each participant enjoy access rights on the "available" background and on the whole foreground owned by other parties under fair and reasonable conditions (provided that it submit a written request to the foreground/background owner and explains it needs the concerned piece of foreground/background for exploiting his own foreground). "Fair and reasonable conditions" may include financial and non-financial compensation. In case of request of access rights for internal research, access is granted royalty-free (non financial burden on the requesting party). If not agreed in writing, access rights do not allow to grant sublicences to third parties and are normally granted on a non-exclusive basis.
To sum it up: As regards the SELFDOTT project, the members of the TRANSDOTT consortium who participated to the SELFDOTT project enjoy access rights on the "available" background and on the whole foreground owned by the whole SELFDOTT consortium. Such access can take place under fair and reasonable conditions (royalty free for internal reaserch purposes - confidentiality issues to be considered!) to be negotiated with the relevant owner and it is a right to be exercised by ONE YEAR from the termination of the project. If this time expires, then the relevant owner of the results of the SELFDOTT project is not obliged to grant access to the other consortium members. [see below for a more detailed explanation of the clauses contained in the SELFDOTT consortium agreement].
List of Websites:
http://www.selfdott.org
Continuing on the success of the Fifth Framework Programme (FP5) research programme REPRODOTT, which was the first ever study of the reproductive biology of the Atlantic bluefin tuna in captivity, and which developed a hormonal method for the induction of spawning, the programme SELFDOTT (From capture based to self sustained aquaculture and domestication of Bluefin tuna, Thunnus thynnus) implemented the knowledge on the reproduction of bluefin tuna in captivity. It established the knowledge-base required for controlled development of eggs, larvae and suitable and environmentally performing feeds.
Project Context and Objectives:
SELFDOTT proposed to implement knowledge already obtained on the artificial control of reproduction of the Atlantic bluefin tuna (BFT), Thunnus thynnus, to obtain viable eggs, and study embryonic and larval development for the production of fry (juveniles). At the same time, suitable and environmentally performing feeds for the grow out of BFT have been developed. It will contribute to reduce/eliminate in the future the practice of raw fish importation and feeding by the fattening industry. Wild juvenile and mature BFT were reared in captivity at two sites in the Mediterranean, and they have used to study puberty, gametogenesis, and the influence of diet on reproductive maturation and gamete quality. Mature fish were induced to spawn using hormone implants and the eggs were collected using devices designed specifically for cages. To establish the knowledge-base for controlled development of BFT larvae, the mesocosm and artificial larval rearing methods have been employed. The ontogenesis of essential biological functions was studied, including environmental perception, digestion, immunity and behaviour.
The main objectives of the SELFDOTT project have been:
1.- Substantiate the current knowledge on the reproduction of capture based aquaculture bluefin tuna
2.- Establish the knowledge-base required for controlled development of eggs and larvae
3.- Establish the knowledge-base for the development of suitable and environmentally performing feeds
Project Results:
One way to alleviate the pressure on the wild fishery of the Atlantic bluefin tuna (Thunnus thynnus, BFT) and aid in its conservation would be the domestication of the BFT and the development of a self-sustained industry. This aquaculture industry will propagate the BFT in captive conditions, rear the larvae and produce fingerlings for further grow-out on suitable, scientifically formulated and environmentally performing feed, as it is done successfully in the EU for species such as the Atlantic salmon (Salmo salar), European seabass (Dicentrarchus labrax) and gilthead seabream (Sparus aurata). Therefore, the overall objectives of SELFDOTT were to further study the reproductive biology and larval rearing of this species in captivity, and develop formulated feeds for the ongrowing of the produced juveniles to market size. The above studies were expected to also result in a better understanding of the life history and physiology of the BFT, contributing to the successful management of the wild stocks.
WP2 Reproduction
In the area of reproduction, the objectives of SELFDOTT were to:
(a) enhance the existing knowledge on the reproductive biology (onset of puberty, gametogenesis, and gamete quality) and artificial spawning-induction of BFT;
(b) optimize microsatellite markers to genotype BFT and their progeny, with applications to broodstock management and fisheries populations;
(c) obtain fertilized and viable eggs, and ship them to larval rearing facilities for the production of juveniles;
(d) develop appropriate broodstock diets for the support of reproductive maturation and gamete quality; and
(e) study the behaviour of long-term reared wild broodstock and establish monitoring procedures; in order to develop the knowledge for a self-sustained aquaculture of the BFT.
WP3 Larval rearing
In WP3, devoted to larval rearing, the first year of the project, 2008, was dedicated to the first husbandry attempts for defining a preliminary rearing protocol on a comparative basis between mesocosm (HCMR, Greece) and artificial rearing systems such as pseudo green (IEO, Spain, HCMR, Greece and MRRA-Aquaculture, Malta) and clear water rearing methods (IFREMER, France).
The Mesocosm, implemented by HCMR. This technique is described as an intermediate between the intensive and extensive methods of rearing that can be considered as a semi-intensive technique of mass production. Phytoplankton and live prey item concentrations and quality is partly related to the "maturation" of the middle previous the beginning of the rearing
The Pseudo-Green Water, implemented by HCMR, IEO and MRRA-Aquaculture. This technique is characterized by controlled conditions of light intensity, photophase and feeding. The phytoplankton is not reproduced neither bloom in the rearing tank (constant concentration by daily addition). The live prey quality is partly influenced by phytoplankton quality and the water quality control is partly dependant of the phytoplankton management.
The Clear Water, implemented by Ifremer and punctually by IEO. This technique is characterized by controlled conditions of water quality, light intensity, photophase and feeding. The live prey quality is fully dependant of the strain quality and their enrichment products and methods. The water quality control partly depends on the recycling system definition and efficiency.
In 2008, a sampling strategy were defined to deliver BFT larva batches from the husbandry sites to each WP3 partners in charge of describing the ontogeny of the main targeted functions.
Larvae of Bonito (Sarda sarda) considered as a model species were expected to train each husbandry sites and supply each scientific topics issues before the BFT season. The bonito fishing season (June 2008) was exceptionally poor and no Bonito egg batch was collected for in vitro fertilization. One Bonito natural spawning occurred from one of the Bonito captive broodstock in IEO facilities in Puerto de Mazarron (Spain). This unique egg batch (16,000 units) was collected and stocked in IEO 5,000 l rearing tanks. The feeding sequence, swimbladder inflation, behaviour, growth and survival patterns were described using a pseudo green rearing method until 40-45 DPH and several 2-4 g Bonito fingerlings were produced.
BFT tentative egg production from SELFDOTT cages located in Malta and Spain failed. Four successive BFT spawning events (mean: 3 millions eggs per day) occurred in ALLOTUNA (collaborative project) cages in Vibo (Italy) end June but only a few larva batches can have been delivered by the Panittica Pugliese Spa hatchery to IEO, MRRA-Aquaculture, IFREMER and HCMR. Rearing attempts failed in MRRA-Aquaculture and HCMR due to a dramatic early larva mortality during and just after transport. Only IEO and IFREMER succeeded in maintaining a few individuals alive until 14 DPH (pseudo green water method) and 9 DPH (clear water method) respectively. These first tentative trials implemented with a few hundreds (IEO) or a few tens (IFREMER) of larvae having suffered detrimental transport conditions allowed to fine tune each rearing system and define the environmental conditions leading the animals and living preys to be acceptably homogenisation. Despite a poor early growth, first feeding figures were observed during the early stages.
The different dummy transport attempts and the actual transport operations conducted with BFT larvae and eggs from the Panittica Pugliese Spa hatchery highlighted the importance of a strict management of egg and larva batches from the spawning events to the final destination. An optimal control of the temperature and other water quality parameters (pH, ammonia, bacterial concentration) during the transport time is crucial for meeting the best compromise between egg and larva survival potential and transport duration.
Small BFT larva samples were delivered for studying the lateral line development, the onset of digestive enzymes and the microbiotal diversity respectively from IFREMER (clear water) and Panittica Pugliese Spa hatchery (pseudo green water) husbandry tanks. BFT tissues were also sampled from adults during the broodstock sampling operations and delivered for studying the ontogeny of visual and neuroendocrine systems.
Regarding the ontogeny of visual and neuroendocrine systems issues, 3 genes have been cloned to date from BFT tissue extracts. The first, the arylalkylamine N-acetyltransferase-2 (Aanat2), was cloned after messenger ARN were extracted from the photosensitive pineal organ using a classical PCR approach and degenerated primers. The second and the third were cloned from mRNA extracted from the pituitary gland. The proopiomelanocortin (Pomc) was cloned using degenerated primers, whereas the growth hormone (Gh), had been cloned previously in BFT, and specific primers were used to obtained the full length sequence. Probes were made for the in situ hybridization (ISH).
Concerning the studies on lateral line development, preliminary results suggested a high level of conservation between zebrafish and BFT, greatly facilitating the analysis of BFT patterns. Labelling methods that were developed in zebrafish to reveal the position of lateral line neuromasts can be used in BFT for following the development of the lateral line system.
The behavioural pattern description for describing the BFT feeding and swimming behavior from hatching to juvenile stage with submersible cameras in mesocosm were initiated using sea bass larvae. The recorded images have been analyzed using special software that allows the estimation of swimming velocity and distribution in the water column.
Total mRNA was extracted from samples in BFT adults. We made purified by RT-PCR cDNA fragments of gene coding for elongation factor 1-alpha (EF1 alpha gene). EF1 is a housekeeping gene, necessary for quantifying the expression of genes of interest.
From BFT larvae, mRNA was extracted. cDNA fragments of genes coding for digestive enzymes were purified by RT-PCR, then cloned and sequenced: amylase and trypsine, which are some of the main digestive pancreatic enzymes involved in glucide and protein digestion, respectively, aminopeptidase and alkaline phosphatase, which are two of the main enzymes located on the brush border membrane of enterocytes, and a non specific dipeptidase, which is located in cytosol of enterocytes.
Relatively to microbiotal diversity, 19 and 21 DPH larvae were stored in a freezer with RNA later, and they became available for this purpose, only in 2009. They were treated for microbial analysis. The analytical methods were refined with intestinal samples from three ongrowing tuna euthanatized in the IEO facilities at Puerto de Mazarrón (Spain).
In 2009, larval rearing husbandry tasks were satisfactorily supplied in BFT viable eggs (IEO- 346,000, MRRA-Aquaculture - 450,000, IFREMER- 540,000 and HCMR- 386,000) from Tuna Graso (SELFDOTT-Spain: 1 486,000) and MareNostro (ALLOTUNA-Italy: 836,000). Previously bonito eggs spawned by a captive broodstock in IEO land-based facilities (Puerto de Mazarron) were delivered to IEO-Spain (198,000), IFREMER-France (48,000) and HCMR-Crete (52,000). The Bonito and BFT egg shipment results were notability improved and successful but still critical points have to be solved in terms of egg temperature pre-packing conditioning, transport boxes insulation related to transport ambient temperature and duration.
Several experiments were implemented in 2009 using the 3 targeted larval rearing methods: Mesocosm (HCMR, IEO), pseudo green water (IEO, HCMR and MRRA-Aquaculture) and clear water (IFREMER, IEO). In bonito, the rearing attempts were conducted until 23 (IEO) and 80 DPH (HCMR) in mesocosm, 80 DPH (HCMR) and 73 DPH (IEO) in pseudo green method, and 14 (IFREMER) and 20 DPH (IEO) in clear water method. In mesocosm swimbladder inflation rates were 60 (HCMR) and 78 % (IEO) while they ranged between 37 (IEO) and 60 % (HCMR) in pseudo green method and between 33 (IEO) and 85 % (IFREMER) in clear water method. In BFT, the rearing attempts have been conducted until 17 (IEO) and 60 DPH (HCMR) in mesocosm, 24 (MRRA-Aquaculture), 60 (HCMR) and 73 DPH (IEO) in pseudo green method, and 10 (IEO) and 20 DPH (IFREMER) in clear water method. In mesocosm swimbladder inflation rates was 50 % (HCMR), while ranging between 50 (HCMR) and 58 % (IEO) in pseudo green method, and between 1.2 (IFREMER) and 5 % (IEO) in clear water method. The presence or the addition of microalgae seemed to give a significant advantage to mesocosm and pseudo green rearing methods during the early development stages in terms growth and survival rates compared to the clear water rearing method. The feeding sequence including yolk sac larvae post Artemia strongly influenced both survival and growth highlighting the critical importance to satisfy the energetic and nutritional needs at that stage. A few BFT individuals ranging between 8 and 14 cm total length were produced and a small batch of 300 DPH old Bonito. Light intensity, lighting regime, hydrodynamic, surface cleaning, feed sequence and nutritional quality have been identified as the main factors to be fine tuned mainly during the larval rearing stage. As far as possible Bonito and BFT larvae were sampled, pre-processed and delivered to scientific topics according specifications related to each scientific issue allowing the first academic data monitoring in early stage development.
The study the ontogeny of the visual and neuroendocrine system partly launched in 2008 in BFT was complemented. In order to obtain molecular tools for the spatial and temporal localization of the corresponding transcripts, genes of interest (melatonin receptors, AANAT2) were partially cloned. Partial cloning of POMC and GH genes, as pituitary factors, was also obtained based on sequence fragments from NCBI data basis. For further qPCR experiments, primers were designed according to these partial sequences, and successfully tested. Conventional PCR revealed that MT1 and MT2 receptors both started being expressed in the larvae at 3 DPH, GH might be present in low quantity at 1 and 2 DPH and became more expressed at 3 DPH, POMC gene started being expressed at 3 DPH. Dig-labeled probes were used for in situ hybridization on BFT larvae collection starting from stage 0 DPH. Larvae submitted to POMC and GH probes exhibited the same pattern: the pituitary was clearly labeled at 3 DPH which suggests both the differentiation of this organ and its ability to express hormonal genes at this stage of development. Melatonin receptors hybridizations are currently being analyzed.
The early development of BFT posterior lateral line was examined. This sensory system analyzes the movements of water along the body surface, and is used in very many aspects of fish behaviour, notably prey catching or predator avoidance, and social behaviour (e.g. swimming in schools). We have described the embryonic and early larval development of this system. Many aspects of this development are highly conserved with other fish species examined so far, including the more basal model species Danio rerio (zebrafish), including the long-distance migration of a sensory primordium and deposition of five regularly spaced sense organs along the horizontal myoseptum, and of one terminal neuromast at the tip of the tail. Tuna is rather unique, however, in the size of its sensory organs at birth, suggestive of a remarkable sensitivity in prey detection. This may be related to the very fast assimilation of the embryo's vitellus, making early prey catching a vital necessity.
Ontogeny of digestive tract in Bonito and BFT was rather extensively described in a comparative way between the 2 species, bonito and BFT, and the 3 investigated rearing methods focusing on the digestive system description with emphasis on the appearance of the different organs, their physiological role and function and relations on larval feeding preferences. In BFT no difference existed in the time of organ appearance between the rearing methods but the rate of development seems to be significantly affected by the temperature level which differed between sites and rearing methods. In mesocosm, larvae preferred unidentified preys grown spontaneously until 7 DPH when rotifers were detected in the stomach while they were delivered from DPH 2. In pseudo green method the first rotifer presence in the stomach was not noticed before 4 DPH even thought they were delivered two days before. On the same way, Artemia and yolk sac larvae were detected between 2 and 3 days after their first delivery to rearing tank. Even though Bonito were reared in lower temperature their digestive tract development was notably faster than BFT as illustrated by the gastric glands which appeared in 3 DPH in bonito vs 11 DPH in BFT.
The activity pattern (gene expression assessed by q PCR) of digestive enzymes as indicators of physiological status of larvae and antioxidant enzymes as indicators of stress during development (3 to 20 DPH) was investigated and compared between the 3 targeted rearing methods. The different digestive evolution pattern between rearing methods suggested a poor physiological status in larvae reared in clear water associated with a poor growth compared with pseudo green ad mesocosm rearing method. Antioxidant enzyme profiles suggested a higher stress level in clear water reared larvae compared with the 2 other rearing methods.
Regarding the immune system ontogeny, a preliminary approach has been conducted with the bonito and BFT samples collected during the early stages in mesocosm (HCMR). Molecular probes are in preparation and the application of RT PCR related to immune system expression is being evaluated for further analysis in summer 2010.
The evaluation of the microbiotal diversity of the digestive tract was limited by the availability and size of the bonito and the BFT samples. Consequently, it was decided to upgrade the discrimination level of the study by working at the individual level. Individual genomic DNA was extracted from larvae (3 to 12 DPH) sampled in clear water rearing (IFREMER) and pseudo green rearing (IEO) This preliminary approach revealed a microbism in the 12 DPH larvae reared in the clear water method.
A bacterial characterization has been performed in Bonito larvae (1 to 20 DPH) reared in mesocosm (HCMR) and 28 bacterial strains were identified and phenotypically characterized before being identified later on. In addition, the total bacterial content in terms of total counts of the aerobic heterotrophic gut microbiota was determined. An exponential increase in numbers of bacteria per larva was found in bonito larvae as shown in studies with other species. The numbers of Vibrio were detected at notably very low levels in the mesocosm rearing units. Samples were taken as well for analysis for non-culturable bacteria. A bacterial strain belonging to the Phaeobacter genus was isolated from a mesocosm with bonito larvae 1 DPH. This strain exhibited in vitro inhibition against Vibrio bacteria, as shown with other representatives of its genus.
In 2010 and 2011, encouraging survival and growth results allow us to postulate that BFT domestication should be feasible in a next future. But, larval rearing is still at the research stage and has not reached the status of a commercial technology.
Mesocosm and pseudo-green methodologies implemented in large tanks have been substantially improved as it is evident from the survival of the best rearing trials (2.5 -3 % at 25 DPH). The knowledge and knowhow bases in larval and fingerlings rearing have been substantially documented both at the academic and at the rearing levels and gave us clear indications for improving rearing techniques and biological performances. But key issues still need further investigations.
a) Currents, luminosity parameters and feed quality which have a significant effect on vision, feed intake, growth and survival as soon as the early stages have to be fine tuned.
b) It is still impossible to sustain an acceptable growth and survival without YSL (prelarvae) post 17 DPH whatever the rearing methods. So, even if weaning of juveniles (post 25 DPH) on artificial diet was achieved, the later have to be improved and lead to better results.
c) Considering these results and comments we can propose a first set of recommendations for a BFT larval rearing protocol for implementing mesocosm and pseudo green water techniques in large tanks.
Several batches of fertilized BFT eggs were delivered at IFREMER (France), HCMR (Greece) and IOLR-NCM (Israel) facilities during 2010 and 2011 spawning seasons originated from the broodstock kept at Caladeros del Mediterráneo facilities at Mazarrón, Spain. Egg shipments were performed using the standard Cargo services of airlines facilitated by a specific company in Spain in the Greece and Israel cases and transported by car and bus, in the IFREMER case. Egg quality was significantly affected by the transportation conditions and in particular temperature. When variations overdraw 2 degrees, abnormal embryonic development and therefore abnormal individuals were observed.
Regarding Atlantic Bonito (Sarda sarda), a total of 232,900 larvae were cultured in IEO facilities during 2010 season. The swimbladder inflation started between 3 and 4 DPH. The swimbladder inflation rate ranged between 67 and 80%. Between 8 and 12 DPH, an increase in the mortality was noted in some of the batches. Bonito larvae lied in the surface of the tank with a reduce motility and some of them with swimbladder hyperinflation. An analyses looking for virus was done with negative results. After 3-4 days and a couple of prophylactic baths of formaline (40 ppm during 1 hour), the problem started to disappear, seemed that the problem was related with stress caused by high density and maybe an unbalanced feeding. A trial to test the effect on bonito larvae of two different enrichment methods: Origreen (Skretting) and DHA Protein Selco (Inve) for rotifer was carried out and results in survival rate and growth did not offer significant differences (p greater than 0.05). Bonitos were successfully weaned at 22-27 DPH with a paste made in the IEO and composed by minced bonito meat, krill meal, hydrolyzed fish meal, squid meal, fish oil and vitamin premix. From 40-45 DPH onwards the paste were replaced by 2 mm wet pellets manually distributed. In spite bonitos accepted the paste and had an acceptable survival (28.7%) and growth (if we compare with results obtained the previous years when weaning was done with raw fish), mortality continued too high, more than 70%). Besides we must to consider these trials began about 19 DPH and they finished about 31 DPH (average terms). Weaning onset was between 22 and 27 DPH (average 25 DPH) and as YSL were overlapped with paste for 4-5 days, we cut off YSL only a couple of days before being extracted and counted from the tanks. So, some mortality due to weaning was still coming in the following days.
Bonito eggs, obtained in 2010 from a natural spawning of captive broodstock from IEO Mazarrón, Spain, were delivered to the HCMR facilities (Crete-Greece) by air cargo. 60.000 eggs were received (T=21, pH=8.3 O2 =22 mg/l) and were incubated equally in three tanks 500 l tanks. During embryogenesis, egg hatching and the autotrophic larval stage, aeration was provided in the tanks by means of a wooden diffuser (150-200 ml min-1). Aeration was stopped one day after the first day of feeding. From the day of first feeding until 19 DPH, water circulation was autonomous for each tank. During this period the only source of water renewal of the tanks was the constant flow over the feeding tanks. From day 20 onwards, the tanks were connected to a biological filter and the inflow to the tanks was progressively increased. A surface skimmer was installed from 5 to 9 DPH to remove the oily layer from the water surface in order to allow proper swim bladder inflation. The photoperiod was 24L:0D from 0 to 40 DPH, when it was changed to 16h light: 8h dark.
Feeding was executed, using an automated feeding system. In this way feeding could be executed frequently, labor efficient and 24 hours a day. Feeding was based on enriched rotifers and Artemia, Instar II nauplii. Prior to distribution to the larvae, rotifers and Artemia were enriched with commercial products (Selco, INVE) and were maintained in a solution with phytoplankton. From 23 DPH onwards eggs of various species were incubated in a separate tank. The eggs and larvae were fed manually to the bonito's. On 10 DPH hand feeding of dry feed was started. The dry feed was hydrated in the feeding tanks before feeding. On 14 DPH an automatic dry feeder was installed, adding small amounts of dry feed (Proton 2/4, INVE) every 15 minutes. From 10 to 67 DPH defrosted seabream eggs were fed manually several times per day. From 27 to 67 DPH pieces of defrosted mackerel were fed manually several times per day. The pieces of mackerel were covered with dry feed in order to enrich the diet and to adapt the juveniles to dry feed. Water quality was similar in all tanks. The temperature in A2 was higher during the later stages of rearing when no fish were present in the other tanks. Mouth opening took place on 4 DPH. The swim bladder started to develop from 5 DPH and on 16 DPH the swim bladder was inflated in 100% of the individuals. The beginning of flexion started on 7 DPH. No deformities were observed. Growth was similar in all tanks. The exponential growth rate of TL was 5.6% day-1 and the one for BW for 30-58 DPH was 11% day-1. The initial mortality was high and it decreased with population size. Observed causes of mortality were cannibalism, the presence of histophages cilliet and the jumping out of the tanks. Cannibalism was observed between 8 and 40 DPH. Parasitic histophage ciliates were present in A1 between 14 and 19 DPH. They were first ineffectively treated by adding 25ppm formalin to the system. An increased water refreshment of the tank on 19 DPH successfully removed them. At later stages from 43 DPH onwards, fish jumping out of the tank became an important cause of death. This was the cause of death of the last surviving fish. During the rearing for a period of 25 days tanks A1 and A3 were also used for sample collection, causing an additional artificial mortality. During the rearing no developmental problems of the larvae were observed. Therefore the main focus of successive rearing trials should lie on reducing the mortality. A major problem during the rearing was the unsuccessful feeding resulting in inadequate cover of the larvae nutritional requirements. Cannibalism can be counteracted upon by providing good alternative food sources. The provision of de-frozen eggs and yolk sac larvae as alternative food sources was not effective in completely eliminating cannibalism, although it probably led to a significant reduction. A remarkable observation was that the juveniles hunted equally sized conspecifics, which they were not able to swallow and ultimately spit them out. This observation indicates that the juvenile bonito prefer a relatively big prey size, as smaller preys were plentiful at all times. Although dry feed was provided during most of the period, the actual feeding on dry feed was not observed. The juveniles did ingest small amounts of dry feed when feeding on mackerel covered with dry feed. The jumping out of fish can be relatively easily counteracted upon by heightening the walls of the tank. Although just a few fish jumped out, it was a relative big part of the total population at the times it occurred. Although the goal of rearing bonito to weaning was not completely reached, it was a rather successful attempt. The development and growth of the fish during this rearing trial were good, while an ongoing mortality caused a premature termination of the trial.
A total of 1 926,000 Bonito eggs were collected during 2011, all of them coming from Z3 tank. This broodstock tank was composed by about 50 fish born in May 2009 and May 2010. Bonitos spawned during a period of 35 days, and fertilization rate was 81.7%, ranging between 65 and 95%. Spawning occurred at nightfall and eggs were collected early morning every day. The Bonito larvae were fed with Rotifer, Brachionus plicatilis, which was completed up to 5 rot/ml. a couple of times every day. Rotifers were fed on Chlorella and three hours before being offered to the bonito larvae, they were enriched with Origreen +, a special formulation from Skretting based on Origreen but with an extra supplements of micro-nutrients. Artemia was added to satiety four times per day. Instar II of Artemia were enriched over a minimum of a 14 hours period with Origreen + and then offered to bonito larvae. Yolk sac larvae (YSL): Increasing amounts of sea bream YSL were added every day to the bonito larvae from 10-15 DPH onwards. Sometimes, they were fed on bonito YSL instead of sea bream YSL. The weaning were carried out at 22-26DPH with the special dry food developed for BFT. From these production batches, a total of 4,727 23 DPH bonito larvae were produced, which means a 2.7% of survival. In parallel, several trials to study the effect of light intensity, the different enriched diets for rotifers and the density of bonito larvae on growth and survival have been carried out.
A total of more than 4 million BFT eggs were incubated in the IEO facilities, in 2010. Incubators were 400 and 1,000 l tanks, were BFT eggs were placed to hatch and kept one or two days after hatching. Then they were counted and after separating dead eggs and larvae a part of them were moved to larval rearing tanks. A trial was done to assess the effect of iodine on embryonic development of eggs. Iodine treatment is widely use in aquaculture, and it is showed very efficient in the control of some diseases. Short period treatments of iodine are applied to eggs to disinfect and even are duty to transport eggs to some places (some countries like Israel demands it before introducing any kind of fish eggs). Four different dose of iodine were tested during 10 minutes: 0 ppm, 50 ppm, 100 ppm and 200 ppm. The results showed that egg survival rate decreases with the increase of iodine concentration, but the only significant decrease (p less than 0.05) is between 50 and 200 ppm. 50 ppm should be a recommendable dose and even when differences are not significant, 100 ppm it seems also quite high. A total of 1,204,000 newly hatched larvae were cultured in IEO facilities in 2010. From 1,204,000 eggs/larvae we obtained 2,776 larvae (25 to 32 DPH) ready to be weaned. In Mesocosm Tank of 45,000 l was inoculated with 18,000, 2 DPH bluefin tuna larvae. This low stocking density allows us to refer to this culture method as mesocosm. A total of 310 larvae, which means 1.7% survival, were produced. Using pseudogreen water method, the rest of larvae were cultured at greater density (6-12 larvae/l), which should be referred as pseudogreen techniques whichever would be the volume of the tanks. A high density trial (40 eggs/l) was also done in a couple of H tanks. Finally, a total of 2,466 larvae of about 26 DPH, with an average of 0.21% survival rate, were obtained.
Also the effect of several parameters on larval growth an survival have been tested. A trial was developed to test two enrichment methods (Origreen, Skretting, and Spresso, Inve) and a cofeeding diet (Gemma, Skretting).The first approach indicates that it seems that Origreen offer better results than Spresso and that tuna larvae did not accepted dry food during the first 20 days.
Also the different conditions also upwelling (a recirculation system) versus aeration have been tested. The results seem to indicate: a) Increased aeration helps larvae to distribute homogeneously. This is particularly important from 17-18 DPH onwards, with the onset of cannibalism. b) It is possible to eliminate Artemia period but there are a reduced growth (1/2 or 1/3) which leads to delay the beginning of YSL period. c) While tuna has a relatively good performance with minced fish (bonito) as a weaning diet, wet paste used with bonito (see above in bonito trials) it seems offer a worse results and it is not a good diet for weaning.
A high density trial was done in two 1,500 l tanks. 60,000 just fertilized BFT eggs were inoculated in each one of both tanks, which means a stocking density of 40 eggs/l During the first week, larvae went ahead with an acceptable survival rate, but from 10 DPH onwards, a high mortality. This led to a very low survival rate, and finally we decided to finish the trial at 23 DPH, when only 10 larvae were alive in both tanks. In conclusion we can say that with the actual level of knowledge, it is very difficult to be able to culture BFT tuna larva at high density.
A trial to compare the two methods employed to avoid larvae sinking during the early stage of development in bluefin tuna larval rearing was conducted. Two methods deal with kept larvae in suspension using air current or water current. In order to watch larvae behaviour, two 500 l transparent tanks were used in spite of this prevent us from carrying out replicates of the trial. One had surface water inlet and stronger aeration, while the other had bottom water inlet and mild aeration. Both of them had water outlet in surface, so idea was to create an upwelling current inside the tank to prevent BFT larvae sinking to the bottom of the tank. Larvae tend to sink mainly in night time, when light is off and they decrease their swimming movements. In this moment it seems very
important to help the larvae to keep themselves in suspension by upwelling current. If current is too strong, larvae can injure by hitting against the walls of the tanks, but if it is too weak, larvae fall to the bottom of the tank and they die. At 25 DPH survival rate was 2.4% in bottom water inlet tank and 1.2% in surface water inlet tank. From this results it seems that is better to kept larvae in suspension with water instead with aeration, but it should be necessary to test every single tank, in order to avoid an excessive water flow.
The weaning of BFT larvae was carried out in 2010 in the IEO facilities on a minced fish basis. We used several species of fish (bonito, sand eel, mackerel, anchovy, sea bream, sardine, etc.), but mainly bonito. The conclusion is that it is possible to wean BFT larvae on a minced fish basis, and they prefer fish instead squid (Sepia sp.) as weaning diet. However mortality went on too is high and growth also should be improved. Another trial was conducted to test a dry food, Gema (Skretting). A total of 235 BFT around 34 DPH were placed in four 1,500 l tanks. Two tanks were weaned onto bonito meat and the others two were weaned onto Gema diet. Results obtained with dry food were quite worse, and improved diet should be used in order to wean suitably BFT larvae.
More than that, a couple of weaning attempts were done in square 12 m3 tanks. Weaning diet was bonito mix with another fish, mainly Scomber or Phycis. First attempt ended with bad results because a disease which caused great mortality (more than 90%), and even the tuna surviving had a increased mortality during the first ongrowing days Second attempt was made and it obtained better results, about 25% survival rates. BFT fingerlings had a weight of 3-4 g at 38 DPH. A careful cleaning of the tank, and the use of a greater rate of sand eel, could explain the best results obtained.
Several preongrowing attempts were carried out in IEO during 2010. Tuna were fed mainly on bonito and sand eel. In spite they accepted well the minced fish, mainly sand eel. Mortality went ahead due to several reasons, but mainly collision and maybe an unbalanced diet. In order to prevent collision, tank walls were painted on a bright green colour, and a dim light during night to keep continuous photoperiod was used. But collision went on and from 45 DPH was the responsible of the most of death. At the end of this trial, bluefin tuna had an average weight of 18 g, but the greater fish weighted more than 50 g.
Two transports to a cage of 25 m, diameter and 10 m depth were carried out on August 20th and 21st 2010. 7 died during transport (11% mortality), and a total of 56 fingerlings were introduced to the cage. During the first hours, a great mortality occurred in the cages: 24 tunas died immediately after arriving, in the first 2h (43%). 48 h after the transportation, accumulated mortality was 75%. 96 h after the transportation, mortality was 89%, and only there were 6 tunas in the cage. A week later, mortality was 93%. There were 4 surviving bluefin tuna. This tuna fed on minced fish (sand eel at the beginning and mainly sardine from September onwards). They started to accepted food from the day after transportation, but ingestion rates were lower than in the tank. Last tuna died after 45 days in the cage (110 DPH) with about 100 g.
To test the effect of transportation three dummy trials were conduced. A couple of fingerlings were placed in a plastic bag like those used for transport purpose during 24 h. After this, 1 tuna was died and the other was still alive. 2. Five BFT were subjected to handling to extract from the tanks and then they underwent a transport simulation (4 h). After this they were placed in a 5 m3 tank. 2 tuna died during transport and 3 during the next 48 h. Five BFT were subjected to handling to extract from a tank and then they were place in another 5 m3 tank. 3 tuna died during the next 72 h, and a couple of fingerlings survived.
Also another trial moving 15 fingerlings to a greater tank (a 100 m3 tank) early September 2010 was carried out. These were smaller tuna (average weight of 5 g), and mortality after 72 hours was 7%. Tunas were not too enthusiastic with minced fish and in spite of eating, hey were reducing their ingestion rate through following weeks. Their growth was not good and most of the fish died during September. At the beginning of October there were only two tuna, and the last one died on October 20th 2010, when it was 110 days old and it weighted 52 g.
These results seem to point out that handling is mainly traumatic for tuna greater than 5-10 gr. tunas. The only action of collecting from the tanks led to a great mortality that is increased if after this they undergo new stressing situation like a long transport. More than that, it is necessary to acquire knowledge about nutritional requirements of fingerlings and to improve nutritional quality of supplied feeding.
In 2011, from total of 162 millions eggs were collected during this year in both cages in Cartagena (Spain), and transported to the IEO facilities in Mazarrón. A sample of 72 eggs coming from every spawning event was incubated in three different plates. Each plate contained 24 wells of 2 ml capacity, and one egg was placed in each well. The hatching rate results were very similar in both cages (Mean ± SD): Cage R1 81.7±12.1% and 87.5±10.2% for Cage R2.
A total of 1.7 million eggs were cultured at the IEO facilities in 2011.All the BFT larvae were reared with similar rearing techniques and feeding schedule: a) Rotifer: From 2 to 16 DPH. Rotifer, Brachionus plicatilis, was completed up to 10 rot/ml. a couple of times every day. Rotifers were fed on Fresh Chlorella and three hours before being offered to the tuna larvae, they were enriched with Origreen+ made by Skretting and also tested in bonito. b) Artemia: From 13 to 19 DPH. Artemia was added to satiety four times per day. Instar II of Artemia were enriched over a minimum of a 14 hours period with Origreen + and then offered to tuna larvae. c) Yolk Sac Larvae (YSL): Increasing amounts of sea bream YSL were added every day to the tuna larvae between 17 and 30 DPH. Sometimes, they were fed on tuna YSL instead of sea bream YSL. d) Weaning: It started when larvae were between 25 and 30 DPH. Nevertheless there were some important differences according to the size of the tanks: Production tanks (40 m3), Intermediate tanks (5 m3) and research tanks (1.5 m3, and transparent tanks, 0.5 m3).
In a 40 m3 tanks, three production batches were cultured, with a total initial number of 1.25 million eggs. When larval rearing finished, tuna larvae were taken out, counted and placed in some new tanks to wean them. A total of 17,300 23 DPH larvae were produced. This means a 1.38% survival rate, but if we consider only the two first batches, the survival rate was 2.09% The main difference was the amount of yolk sac larvae, which was extremely low in the last batch.
Larvae coming from the first batch, weighted 107±48.5 mg (ranged between 48 and 107 mg) and they have a total length of 22.3±0.40 mm (Mean±SD). Larvae coming from the second batch, weighted 158±58.5 mg (ranged between 73 and 281 mg) and they have a total length of 24.9±0.3.41 mm.
The effect of four light intensity (0, 500, 1000 and 2000 luxes) on growth and survival of BFT larvae has been tested in twelve 1,500 l tanks, in triplicate. The initial density was 7.2 eggs/l. (10,800 eggs/tank). The complete darkness (0 lux) led to a fully mortality in the first week, and even when there were not significant differences in survival and growth rate at other intensities, it seems that the greater intensity led to slightly worst results.
A second experiment to study effects of two different enrichment diets for rotifers on growth and survival of tuna larvae has been carried out testing two diets: Origreen and Origreen+ (with extra supplement of micro-nutrients). The trial was carried out in 1,500 l cylindrical tanks in triplicate. The initial density was 10 larvae 0DPH/l. (15,000 just hatched larvae/tank) The feeding schedule was: 10 rot/ml adjusted twice per day. 20 ml of Fresh Chlorella V12, added also twice per day. The experiment when larvae were 15DPH. The results showed that using Origreen+ the survival rate was quite higher (3.01±1.64% versus 1.03±0.10%). In growth, differences were not significant (p greater than 0.05).
A third experiment to study the effects of three different enrichment diets for artemia on growth and survival of BFT larvae was carried out. Three different diets to enrich Artemia: DHA Selco (Inve), Origreen and Origreen+ (Skretting) were tested. The trial was carried out in 150 l cylindrical tanks in triplicate. Differences among three treatments were significant (p less than 0.05); the best results were achieved in BFT larvae fed on DHA Selco enriched Artemia, and the worst were those in larvae fed on Origreen enriched Artemia. The wet weight ranged between 7 and 11 mg, and it was significant lower (p less than 0.05) in tuna larvae fed on Origreen enriched Artemia than DHA enrichment, while there were not significant difference among three treatments with regards to total length (all the larvae reached about 10 mm).
A four experiment to test the effect of BFT larvae density was carried out. Two different stocking densities (10 and 30 eggs/l) in 500 l transparent tanks without replicates were tested. Even when we cannot obtain conclusive results, it seems that is possible to increase density achieving acceptable survival rates (1.4%), during the first stage of development (before start piscivorous feeding), so that obtained results in growth and survival rate were very similar. Nevertheless, further research must be implemented.
Several weaning attempts were implemented in IEO during 2011 season. In the same way that in larval rearing, two different techniques have been used: a production technique with 5, 15 and 20 m3 tanks in which a mixture of weaning diets were employed, and experimentation technique with 1.5 and 0.2 m3 tanks in with different weaning diets were tested. In production tanks, dry food consisted in a mixture of two different diets from Skretting (TD and LM). The LM diet was an experimental dry diet designed to support nutritional requirements in tuna larvae. In order to try to improve the results, the dry food was distributed to satiety using automatic belt feeders. Tuna lasted some days in being fully adapted to new food, but after 5-7 days they showed a good acceptance of the food. Survival at 39 DPH was 28.5%, ranging between 19.3 and 33%. At the end a total number of 1900 tunas (20.8% survival) were produced and sent to the cages when they were 39-44 DPH. Until 35DPH, most of mortality was due to weak fish unsuccessfully weaned, but from this moment onwards, main mortality was caused by collision.
Two of experiments were made to compare the different weaning diets, using tuna larvae 28 DPH. These experiments took ten/twelve days and the results were compared in survival and growth terms. First experiment was made in 150 l tanks to compare 3 Skretting diets: GM (Gemma Micro, a commercial diet), LM and TD (two experimental diets) and raw fish (minced sand eel previously frozen). A total of 320 larvae were used. They were 29DPH, weighted 153±23 mg and their total length was 22.3±1.6 mm. During the first five days, sea bream YSL and weaning diet was offered, and from 34 DPH onwards, weaning diet was the only source of feeding. Experiment was finished when larvae were 38 DPH. Best results were achieved with LM diet, which produced higher survival than TD diet and Fish diet (p less than 0.05). No significant differences (p greater than 0.05) were founded between LM and GM, and between TD and Fish diet. With regards to growth, the gain in total weight was significantly lower in TD (0.49±0.2 g) than in LM diets (0.18±0.1 g), meanwhile no differences were observed in the other diets. Second experiment was made in 1500 l tanks to compare the three different Skretting diets: GM (Gemma Micro, a commercial diet), TD and LM (two experimental diets). A total of 495 larvae were used. They were 29DPH, weighted 250±108 mg and their total length was 29.65±4.45 mm. During the first five days, sea bream YSL and weaning diet was offered, and from 34 DPH onwards, weaning diet was the only source of feeding. Experiment finished when larvae were 40 DPH. In spite no significant differences (p greater than 0.05%) have been observed among the different diets in weight gain and in survival, LM and GM diet produced the best results.
In IFREMER facilities, the common live prey and larval rearing conditions in bonito and BFT have been studied. To evaluate the effect of probiotoic addition via the enrichment media of live preys we added Bactocell®, Pediococcus acidilactici MA 18/5M (1011 CFU g-1) at the concentration and time exposition of 1g.m-3 2 hours and 5g.m-3 12 hours respectively in rotifers and Artemia. The total length of bonito larvae was significantly higher as soon as 4 DPH when Bactocell enrichment was used under 24L:0D lighting regime (LL) but due to lack of repetition in treatements it was not possible to conclude. In BFT no significant effect of probiotic enriched preys on larval growth (total length) was observed. In BFT, it was demonstrated that between 2 lighting regimes 18L:6D (LD) and 24L:0D (LL) the first one led to significantly higher growth (total length).
During this experiment we globally observed a significant improvement in term of early behaviour and in some extent reduction of early mortality. We hypothesize that is due to a better hydrodynamic management related to water and air injection from the bottom of the tank at an acceptable flow rate. The new surface auto cleaning device we used improved the surface cleanliness but the low swimbladder inflation rate we got could be explained by a relatively limitation of its efficiency during 2 and 3 DPH when we tried to improve the setting of this new device. Mortality picks remained unexplained. Chronic mortality could be related to inadequate prey global quality taking into account that bonito seems to significantly take advantage in terms of growth and survival in feeding YSL as soon as 10-12 DPH.
Clear water rearing experiments was carried out in bonito at Palavas facilities, in 2011.Probiotic enriched living prey effect was tested under LL lighting regime in one experiment and LL lighting regime was compared against LD in a second one without any probiotic enrichment. From these two trials, we concluded that 1) probiotic enrichment didn't promote a higher growth and survival by comparison with the control and 2) LL lighting condition seems to promote a better early survival and delayed the hyperinflation crisis which was observed in the two treatments, 3) in each experiment we confirmed that is was not possible to sustain a correct growth and survival more than 15 days without using any yolk sac larvae to replace enriched artemia. In a third experiment we demonstrated that an early weaning using Gema Micro (Skretting) allowed a successful transition between rotifer and micropellets within two days from dph 5 to 7 but this micropellets as single feed source in our feeding conditions was inappropriate to sustain a correct growth and survival
A second study was realized for the evaluation of the effect of volume on rearing performance: a comparison between 500 l and 2000 l tanks. The duration of the trial was 20 days. There was no significant effect of tank volume within treatments on TL evolution and in the case of 500 l the individuals grew exponentially with a rate of 0.253/d while in the 2000 l the rate was 0.175. There was a significant difference between the two conditions. In terms of survival a high variation between treatments was observed. Mean survival was higher in the 2000 l tanks although not significantly different.
In the knowledge on the immune system of the BFT larvae and for the purposes of gene expression studies using the methodology of real time PCR/quantitative PCR (qPCR), adequate reference genes have to be identified and primers for qPCR have to be developed. A first set of 6 candidate genes known to be involved in the immune response were selected for gene expression analysis. Some of the genes are not yet described for tuna, therefore homologues sequences for several perciform teleost species were retrieved from the NCBI nr and EST databases and primer were designed at conserved sites. Primers for the reference genes and genes related to the immune system have been constructed.
Biodiversity in microbiota associated with larval tuna and bonito have been assessed by the end of rearing experiments. The variability of microbial profiles was observed even at 2-3 DPH, possibly due to individual variations in the timing of mouth opening among the larvae. It seemed that this initial microbiota originated in the rearing water, with likely poor influence of the environment before hatching. Indeed, there was a clear difference at mouth opening between the samples from Mazarrón (Spain), and those from Palavas (France), though the eggs were incubated in Mazarrón in both cases. There was also a clear difference between the samples of 2009 and 2010 in these two hatcheries, confirming the unspecific nature of the initial microbial community in larval tuna. A general feature observed in all cases was the increase in individual dissimilarity between mouth opening and 10-12 DPH. This increase seemed to occur relatively early, as it was already observed at 7 DPH in Palavas in 2009. The highest difference in the microbial profiles between groups was observed between 7 DPH and the two other dates of sampling. In this particular case the individuals surviving at 12 DPH did not seem healthy, and these results should be interpreted cautiously. The general trend seemed that the microbiota settling in the gut of the larvae were different from those observed initially at mouth opening, when the microbes associated with external surfaces were likely still dominant. These settled communities appeared yet unspecific, are they remained different between years and hatcheries. There was no general trend of microbial settlement in terms of diversity, except a higher dominance, and lower evenness, equitability and richness in 2009, compared with 2010 in Mazarrón and Palavas, while the indices from Heraklion in 2010 resembled those from the other hatcheries in 2009. Microbiota associated with tuna larvae appeared therefore unspecific, and likely selected among the environmental microbes present in the hatchery during the rearing period. It seems possible to exert some influence on the selection process by modifying the rearing conditions. This may be illustrated in Heraklion by the higher intra-group similarity among the individuals reared in mesocosm, compared with those in intensive conditions. Future replications are necessary to confirm this finding. The triplicates available in Palavas gave more weight to the observation that the probiotic treatment seemed to regulate the microbial diversity, and the intra-group similarity of microbiota associated with tuna larvae, but that need also to be confirmed in further replications and sites of experiment.
Sampling for analysis of bacterial colonization has been done in Mesocosm systems in the case of bonito and in both Mesocosm and artificial rearing system in the case of BFT larvae. In most marine fish larvae there is an exponential increase of bacteria in the gut during the first weeks of rearing when the larvae consume rotifers and Artemia. In the case of BFT larvae there is an unusual pattern compared with colonization of bacteria in other fish larvae, as the larvae reached a peak 5 DPH and thereafter lower values were observed. This could be explained by the fact that rotifers are a large source of bacteria for fish larvae. In the case of tuna, the larvae may switch over faster to consumption of different prey items such as yolk-sac larvae instead of rotifers. Yolk-sac larvae probably have a much lower bacterial load than rotifers. In the case of bonito larvae reared in mesocosm in HCMR (Greece), the pattern was closer to what is normally observed in rearing of marine fish larvae. The culturable microbiota of tuna larvae was investigated during the season of 2011. A quantification of microbiota in tuna larvae sampled from mesocosm in terms of counts in marine agar and thiosulfate-citrate-bile salts-sucrose (TCBS) agar. In the case of yolk-sac larvae the bacteria present in the larvae were showed a variability between different groups, so TCBS counts were relatively low. The numbers of total counts increased exponentially in larger larvae, and members of the Vibrio group became dominant.
WP4 Nutrition and feeding
In the part devoted to Nutrition of juveniles, 25 Young BFT 0+ (YBFT) were captured alive on the shore of Murcia (SE Spain) in Autumn 2008 with the method so called in Spanish "curricán" (a term meaning to fish by trailing a baited line along behind a boat). They were placed on board in cylindrical plastic tanks (1.45 -1.65 m Ø, 1.20 m height, containing 1 200 litres of sea water), transported to the project facilities in El Gorguel (Cartagena, SE Spain) and then released in a circular 25 m-diameter floating cage which was 20 m deep. Whilst close to 50% of the captured fish arrived alive to the cage, the survival rate after one month in the cage was high (75%). The main mortality in this process was due to difficulties in freeing fish from the hooks, although the hooks were barbless. No YBFT grouping was observed in any case, so it was impossible to use the purse seine method for massive capture as planned. Another 29 YBFT were captured dead. The stomach contents from these 29 dead YBFT have been analyzed and the whole fish has been minced and lyophilized for later body composition analysis. The first stomach analysis show that cephalopods are the main prey (53% in weight), followed by fishes (28%) and crustaceans (2%). The main prey species is the cuttlefish Illex coindetti.
One cage of 25m of diameter x 20m deep which contained 15 YBFT captured in Autumn 2008, was used to start the artificial feeding trials in January 2009. In February 2009 we started the weaning trials using different kinds of artificial food. Weaning was achieved. For weaning purposes and may be for general feeding we have a feed that the tuna will eat. Due to the small number of fish in the cage, we were not able to draw conclusions on growth in this year.
From the body composition analyses it seems that YBFTs have nutritional requirement that are different from most of other Mediterranean aquaculture species. YBFT have high protein and low fat content. Comparing the total n-3 and EPA+DHA concentrations among various marine species we determined significantly higher values for YBFTs. Comparing the amino acid profiles of YBFT and other teleosts we observe that in most cases YBFT have higher amino acid concentrations. The most significant differences appear for histidine, methionine, lysine, leucine and aspartic acid. The tuna feed formulation must consider these differentiations and include appropriate ingredients in order to achieve higher dietary amino acid and protein concentrations, lower total lipid content, higher dietary n-3 concentrations and higher vitamin C and E concentrations. These results will form the guidelines for the formulation of the next nutritional trial of YBFTs.
The stomach analysis carried out in 2009, showed that most of the organic content in weight consisted of fish, but cephalopods and small crustaceans were also present in the sampled stomachs. Fish are the main prey (37% in weight), followed by cephalopods (11%) and crustaceans (5%). The main prey species is the fish round sardinella (Sardinella sp.). These results contrast with the ones obtained in 2008 where cephalopods were the main prey (53% in weight), followed by fish (28%) and crustaceans (2%). The main prey species was the shortfin squid Illex coindetti. In conclusion, juvenile BFT predate on a huge variety of preys. The diet composition may vary greatly from one year to another depending on the availability or abundance of preys. Therefore, this age class (0+) can be considered a generalist feeder.
WP5 Dissemination
In WP5 devoted to Dissemination, a project website has recently been commissioned (see http://www.selfdott.org online) and the first newsletter containing details of the first annual meeting together with a guest workshop are in the process of being completed. In 2010 and 2011, dissemination activities were carried out including specialized sessions at the EAS meeting in Rhodes and a stakeholder workshop in Madrid.
Potential Impact:
RESEARCH AND INNOVATION
Type of Exploitable Foreground: General advancement of knowledge
Exploitable foreground (description): The project SELFDOTT has maintained broodstocks of bluefin tuna over a period of three years in captivity and has successfully obtained fertilised eggs in large numbers. These eggs have been hatched and larvae successfully reared and fingerlings which had successfully been transferred to open sea cages for on-growing. Young tuna larger than 2kg have survived for a period of greater than 9months DPH.
Confidential: no
Exploitable products or measures: Monitoring of broodstock populations. Induction techniques for providing "eggs on demand", transfer protocols for eggs. Larval rearing of tuna, fingerling production. Grow-out of tuna.New feeds for tuna larval stages, weaning diets.
Sector of Application: Sustainable Aquaculture
Timetable for commercial use: The results will be used by individual participants within the framework of both national and EU coordinated programmes. This foresees the use of these technologies within 2012 and also continuing fall at least the next two years as part of a follow-on project TRANSDOTT
Patents or other IPR exploitation: At the present time there are no patent applications under consideration since much of the work has been or is the subject of scientific publications or has been presented at international conferences. All partners of the SELFDOTT have access to the results and may continue to use them. There are no licences granted at the present time. A own and other.
Owner and other Beneficiary: As laid down in the consort to agreement foreground shall be the property of the beneficiary carrying out the work generating that foreground. The owners and beneficiaries are therefore the full partners of the SELFDOTT consortium.
CONCERNING OWNERSHIP AND ACCESS FOR SELFDOTT PARTNERSHIP IN GENERAL
Regarding the feedback from the evaluators, here is what the colleagues of the IPR service want to stress 'what matters is not to be the OWNER of the results but to ENJOY the RIGHT to USE them (i.e. have access to them). '
1) SELFDOTT - CA
The CA is based on the DESCA model, it has rather standard clauses. Those relevant for our purposes are from art.8 to art.10 (plus attachment 1).
The basic principles applies as regards the ownership of the foreground (art.II.26 of the Annex II to the Grant Agreement: ftp://ftp.cordis.europa.eu/pub/fp7/docs/fp7-ga-annex2-v6_en.pdf): Foreground shall be the property of the beneficiary carrying out the work generating that foreground
As regards results developed jointly by two or more partners, art.9 allows each joint owner to use the jointly owned foreground and to non-exclusively licence it to third parties (provided that notification and compensation conditions are fulfilled).
As regards access rights, our interest is only focused on access rights for Use: each participant enjoy access rights on the "available" background and on the whole foreground owned by other parties under fair and reasonable conditions (provided that it submit a written request to the foreground/background owner and explains it needs the concerned piece of foreground/background for exploiting his own foreground). "Fair and reasonable conditions" may include financial and non-financial compensation. In case of request of access rights for internal research, access is granted royalty-free (non financial burden on the requesting party). If not agreed in writing, access rights do not allow to grant sublicences to third parties and are normally granted on a non-exclusive basis.
To sum it up: As regards the SELFDOTT project, the members of the TRANSDOTT consortium who participated to the SELFDOTT project enjoy access rights on the "available" background and on the whole foreground owned by the whole SELFDOTT consortium. Such access can take place under fair and reasonable conditions (royalty free for internal reaserch purposes - confidentiality issues to be considered!) to be negotiated with the relevant owner and it is a right to be exercised by ONE YEAR from the termination of the project. If this time expires, then the relevant owner of the results of the SELFDOTT project is not obliged to grant access to the other consortium members. [see below for a more detailed explanation of the clauses contained in the SELFDOTT consortium agreement].
List of Websites:
http://www.selfdott.org
