Resumen de Desarrollo de dietas semihúmedas formuladas para cefalópodos (Octopus vulgaris y Enteroctopus megalocyathus): alteraciones lipídicas en materias primas deshidratadas y suplementación con lípidos polares
- español
El interés del pulpo de roca como especie para producción acuícola creció durante las últimas décadas, pero su desarrollo fue ralentizado por la falta de dietas formuladas adecuadas que garantizaran el crecimiento de Octopus vulgaris. La primera sección de esta tesis está enfocada en la deshidratación que se lleva a cabo para la obtención de ingredientes. El Capítulo I trata de la selección de un método de deshidratación para materias primas incluidas en dietas artificiales para el engorde. Los resultados reflejaron que la inclusión de boga deshidratada a baja temperatura (≈ 60°C) como ingrediente tuvo un efecto similar a la inclusión de boga liofilizada en relación al crecimiento, la ingesta, la digestibilidad de nutrientes y la supervivencia sugiriendo este tipo de deshidratación como apropiado para ser aplicado en materias primas destinadas al engorde de pulpo. En el Capítulo II se analizó el perfil de clases lipídicas en materias primas de origen marino (moluscos, crustáceos y peces) y estándares de lípidos puros congelados y deshidratados por liofilización o exposición a aire caliente, a 60 °C y 100 °C. El calor aplicado causó una reducción del total de lípidos polares (TPL) y un incremento del total de lípidos neutros (TNL), de forma más marcada cuanto mayor fue la temperatura aplicada. La segunda sección aborda la selección de un protocolo de alimentación. Para ello, se evaluó el rendimiento de la misma dieta artificial (ingesta, crecimiento, eficacia alimentaria, digestibilidad y condición de los animales) aplicando dos protocolos de alimentación en los que se incluyeron dos o tres días de ayuno no consecutivos por semana. Ambos tratamientos fueron similares en relación al crecimiento y la digestibilidad de nutrientes (P>0,05), no registraron mortalidad y mostraron tasas de ingesta superiores el día posterior al ayuno. Sin embargo, con tres días de ayuno se observó una mejora de la eficacia alimentaria. Los resultados obtenidos sugieren que el pulpo podría tener la capacidad de compensar los días de ayuno aumentando la ingesta, en el día posterior al ayuno, u optimizando el uso de sus reservas. Además, en términos comerciales, la reducción de los días de alimentación supondría una reducción de los costes de producción. La tercera sección de este documento se centra en la suplementación de dietas con lípidos polares de origen vegetal y animal. La adición de fosfolípidos fue promotora de mejoras en el crecimiento, la supervivencia y funciones digestivas en otros organismos marinos. Sin embargo, la inclusión de lecitina, ya fuera de soja o marina, como fuente de lípidos polares no generó mejoras en el rendimiento. En la cuarta y última sección de esta tesis, se comprobó que los avances generados en dietas para O. vulgaris podían ser transferidos a otra especie. El suministro de piensos a juveniles de pulpo rojo patagónico (Enteroctopus megalocyathus) generó rendimientos satisfactorios indicando que el formato y la formulación de dietas para O.vulgaris pueden ser transferidos y considerados para el desarrollo futuro de dietas específicas para el engorde de esta especie.
- English
In the last decade, the common octopus (Octopus vulgaris) has aroused great interest in aquaculture as a candidate to be produced at commercial scale (García García and García García, 2011; Vaz-Pires et al., 2004). Despite the reduction of global fish landings, fisheries still supply high amounts of cephalopods for human consumption. However, both socio-economic and biological traits highlight the high potential of this species to be produced in captivity, especially in countries long-term related with common octopus (Octopus vulgaris). The species is highly appreciated and consumed, it has a broad demand from national and international markets, an attractive selling price. Moreover, it represents a chance to diversify the species which are industrially produced in aquaculture. On the other hand, biological aspects such as a short life cycle (1-2 years), rapid growth - above 5 % of body weight/day (Mangold, 1983), high feed conversion - 30-60 % of ingested food is incorporated into body weight (Iglesias Estévez et al., 1997) and fecundity rates (100,000-500,000 eggs by female), elevated protein content in muscle -70-90 % in dry weight (Lee, 1995), fast adaptation to captivity, good acceptance of inert food, low amount of pathologies and high resistance to handling and transport procedures also reinforce the potential of the species to the aquaculture industry (Rey-Méndez, 2015). However, the inexistence of feeds with a balanced nutritional profile for all life stages has hampered O.
vulgaris aquaculture development. Focused research on feeds development for this species started at 80s. Since then various research groups performed ongrowing trials feeding octopuses with natural (Fig. 1.3), moist (Fig. 1.4), extruded (Fig. 1.5) or semi-moist (Fig. 1.6) feeds, maintaining individuals in either isolated or grouped conditions. Traditionally, the best performance has been obtained through the supply of natural diets, with some proportion of crustaceans. However, in previous studies, semi-moist feeds based on freeze-dried raw materials and pure substances agglutinated with gelatin provided acceptable performance.
Hence, this type of diets was the basis for the development of feeds in this thesis Experiments were carried with wild captured octopuses at Marine Aquaculture Station of Murcian Institute of Agri-Food Research and Development (IMIDA). After an acclimation period of two weeks fed daily with crab (Carcinus mediterraneus) to satiety, octopuses were reared isolated (216 L tank; 83 cm diameter, 50 cm height and 40 cm column water) or grouped (1970L tank) in circular tanks with plain bottom and PVC pipes (110-120 mm diameter) as refugees. To assess diets performance octopuses were weighed at days 0, 28 and 56. Food remains were collected to quantify food intake. Fresh diet samples and faeces were also collected to determine the macronutrient composition and estimate digestibility coefficients (ADC).
Euthanized octopuses were dissected to obtain tissues samples (digestive gland, carcass).
Macronutrient analyses were performed according to the Association of Official Analytical Chemists International methodologies (methods n°: 930.15 for moisture, 942.05 for ashes, 955.01 for crude protein, 920.39 for crude lipid and carbohydrates by difference). Acid insoluble ashes were also determined in diets and faeces samples, to obtain the needed markers for digestibility estimations, according to Atkinson et al. (1984). Lipid classes were determined by High Performance Thin-layer Chromatography (HPTLC), following Olsen and Henderson (1989) methodology, after extracting lipids according to Folch et al. (1957).
Diets performance was assessed regarding growth, food intake, feed efficiency and nutrients Apparent Digestibility Coefficients according to expressions exposed in points 4.4.1, 4.4.2, 4.4.3, and 4.4.4, respectively. Variations in body composition, regarding macronutrients and lipid classes (only in some chapters), was also assessed in tissues samples.
The first section of this thesis was focused on raw materials dehydration. Chapter I focused on the selection of a dehydration method for raw materials to be included in formulated feeds for O. vulgaris. The study was performed with isolated juveniles that were fed along 56 days with two semi-moist diets based on freeze-dried ingredients following a feeding protocol with two non-consecutive days of starvation per week (Wednesdays and Saturdays). Both diets had similar basal composition varying only the thermal treatment applied to one ingredient: freeze-dried bogue at IMIDA facilities (FDb diet) and dehydrated bogue by exposition to convective air at 60 °C at a commercial enterprise (Mb diet).
Both feeds were accepted and promoted 100 % survival. Protein and mineral contents of Mb diet was slightly higher than FDb diet (P<0.05; 69.98 vs. 66.14 % and 6.20 vs. 5.77 %, respectively). Lipid and carbohydrates contents were similar (P>0.05). Diets disaggregation in seawater (WSI) presented high and similar values in both feeds (42.69-44.71 %).
The Mb diet had higher intake and digestive gland index than FDb (P<0.05; SFR de 2.14 and 1.59 %BW/day and DGI of 6.75 and 4.34 % in Mb and FDb, respectively). However, both diets promoted similar growth (SGR ≈ 0.78-0.85 % BW/day; P>0.05), feed efficiency (FE ≈ 39.22-48.31 %; P>0.05) and conversion (FCR ≈ 2.15-2.56; P>0.05).
Faeces proximate composition varied among treatments (P<0.01). Higher protein and mineral contents were verified in Mb faeces, while FDb presented higher lipid content. The fat content was also verified in-situ as floating faeces. The Mb experimental group presented higher Apparent Digestibility Coefficients (ADC) of both diet (dry matter) and nutrients (protein, lipid and carbohydrates; P<0.01).
Excluding the slightly higher ash content verified in FDb muscle (P<0.05), the proximate composition of tissues (muscle, digestive gland and whole animals) was similar among experimental treatments (P>0.05).
Results suggest that raw materials use in the manufacturing of octopus feeds can be dehydrated applying convective air at 60 °C, without diminishing the satisfactory performance obtained with freeze-dried ingredients. However, previous studies reported negative results with diets based on raw materials that were dehydrated by applying high temperature. Lehninger (1997) stated that the 3D-proteins structure collapse at temperatures above 57 °C. Hence, temperature applied in both dehydration methods tested in this study did not affect proteins native conformation. Considering these observations and the known relevance of lipids in cephalopods composition and nutrition (Koueta et al., 2002; Navarro and Villanueva, 2003, 2000; Sinanoglou and Miniadis-Meimaroglou, 1998) it was decided to evaluate changes in the lipid classes’ profile of seafood samples dehydrated with conventional methods: freeze-drying or convective air at low or high temperature.
In Chapter II, the effect of dehydration on lipids was assessed in pure standards (polar o neutral) and in ten seafood samples (cephalopod, crustacean and fish species) by comparing the lipid classes profile of frozen samples with those obtained after dehydration (freeze-dried or exposed to convective hot air at 60°C and 100°C).
The applied dehydration methods were effective in reducing moisture below 8 % (2.3-7.9 % in freeze-dried samples, 1.9-7.5 % at 60 °C and 0.1-4.6 % at 100 °C) without affecting total lipid content (1.5-15.9 g/ 100 g).
Dehydration in pure standards: Freeze-drying conserved the initial lipid profile of pure standards while heat dehydration provoked changes. At low temperature (60 °C) 88-98 % and 95-100 % of polar and neutral lipids were conserved, respectively. In contrast, high temperature (100 °C) generated more pronounced variations with conservation ranges of 27-98 % and 22-52 % at the polar and neutral fractions, respectively. In general terms, given polar lipid class standards (PC, PS and PE) were broken giving place to other polar and neutral lipid classes while neutral standards (CHO, FFA and TG) were decomposed in other neutral lipid classes.
PC experienced minimal variations. PC dehydration at 60 °C generated negligible LPC (≈ 0.1 %).
In contrast, although 98.5 % of PC was conserved after 100 °C drying, LPC proportion was ten times higher and FFA was also generated. About 90 % of PS was conserved after 60 °C drying with the appearance of PA (≈10 %) while elevated temperature (100 °C) generated a high proportion of neutral lipid classes (about 50 % of MG, DG and mainly FFA). Low temperature drying caused a reduction of 12 % of PE and the appearance of other polar lipid classes (mainly PA, ≈4.4 %) and low proportion of other neutral lipids (MG, DG y FFA, ≈4.5 %). However, at 100 °C, the neutral lipids proportion was markedly increased (41.7 %). CHO was not affected by any of the applied drying thermal treatments. Heat application in FFA standard provoked mono- and di- glycerides apparition (73 % MG and 5 % DG, at 100 °C).
Dehydration in seafood samples: The chromatograms of the analyzed mollusk, crustacean and fish samples revealed mainly fourteen different bands of which eight corresponded to polar lipids (LPC, SM, PC, LPE, PS, PI, PA and PE) and six to neutral lipids (MG, DG, CHO, FFA, TG and SE). Another band was detected in the C. mediterraneus samples between TG and SE, named as “Other neutral lipids, ONL”. Other diffuse bands observed in all samples were associated with pigments according to its reference position between PE and MG and included into an “Other polar lipids, OPL” group.
Total polar and neutral lipids: All the samples suffered a loss of polar lipids when dehydrated applying heat (60⁰C or 100⁰C). That loss was reflected in a reduction of the detected bands, but also in a density smoothing. In this sense, polar and neutral lipids displayed an inverse relation in all analyzed samples but also high variability on appearance, depending on the species and dehydration treatment.
There were no detected significant differences in total PL between frozen (68.6 %), freeze-dried (68.5-73.7 %) or air-dried at 60 °C (61.4 %) in T. sagittatus, with a sharp decrease at 100 °C treatment (P<0.05). Contrarily, T. sagittatus freeze-dried by Vegenat S.A. company (Id. 2) had a markedly low PL but high NL proportions, with a similar pattern to dehydrated T. sagittatus at 100 ⁰C.
C. mediterraneus samples displayed between 3.92-16.20 % of total PL and 80.33-94.94 % of NL.
There were no detected significant differences in PL between frozen and freeze-dried (P>0.05), while both air-drying-treatments (60 °C and 100 °C) reduced PL below 10 %. Frozen fish samples exhibited a common trait: the higher proportion of PL (27.4-50.9 %) and the lower of NL (49.1-71.8 %) compared with dehydrated samples. Some freeze-dried samples showed a slight decrease of PL (P<0.05) while NL increased (P<0.05). Air-drying at 60 ⁰C decreased PL proportions (16.9-40.4 %) compared with frozen samples (P<0.05) but conserved freeze-dried samples proportions. However, the most relevant changes were produced by air- drying at 100 ⁰C, which caused an abrupt descent on the PL proportion of all fish samples (5.1- 23.6 %) compared to the other dehydration treatments.
Principal Component Analysis (PCA): Regarding cephalopod and crustacean samples, the three selected principal components explained 78.44 % of the total variance. Frozen and freeze-dried cephalopod samples were characterized by high LPT, PC, PS+PI and CHO but low LPC percentages. Low temperature (60 °C) dried samples suffered an increase on LPC but conserved the lipid profile of frozen and freeze-dried samples while high temperature (100 °C) caused a reduction on polar lipids (LPT, PC, PS+PI and PE) and an increase on neutral lipids (LNT, MG, DG, FFA and SE). Cephalopod samples freeze-dried by Vegenat S.A. were distinguished between them by the percentage of polar lipids and the appearance of LPC in both samples, compared to IMIDA freeze-dried sample. Frozen, freeze-dried and 60 °C dehydrated crab samples were related with high LNT, MG, FFA, TG and SE percentages, and the presence of LPC. In contrast, the 100°C dehydrated sample suffered a marked loss of LPC and other polar lipids.
The three selected principal components obtained for fish samples explained 64.10 % of the total variance. Frozen and 60°C dried fish samples (Id. 5-10) presented a similar distribution in PCA analysis. These samples were related with polar lipids presence (LPT, PC, PE, SM, PS+PI) but lower CHO percentages. Freeze-dried samples suffered a loss of polar lipids and increased CHO percentages. Dried samples at 100 °C had lower presence of polar lipids and higher of neutral lipids (LNT, TG, MG and CHO) compared with frozen, freeze-dried and 60 °C dehydrated samples. The results obtained in Chapter II showed a general change of the lipid profile, regarding polar and neutral lipid classes, caused by dehydration treatments, especially when high temperature was applied. In general terms, dehydration caused a decrease on the polar lipid fraction and, consequently, a proportional increase on the neutral lipids fraction of raw materials commonly used as ingredients for cephalopods diets.
The second section of this thesis focused at food management. The frequency of feeding was evaluated as strategy to enhance culture profitability as benefits were checked previously in salmonids and other aquatic vertebrates’ cultures. Controlled fasting could be advantageous for the common octopus in biological (to avoid excessive lipid storage in the digestive gland;
(Eroldoǧan et al., 2006; Lovell, 1998) and economic (reduction of production costs related with feeding expenses) terms.
With this aim, in Chapter III we assessed performance applying two feeding schemes. Isolated juveniles of common octopus were fed along 56 days with the semi-moist diet FDb (supplied previously in Chapter I) including two (Wednesdays and Saturdays, 2FDb group) or three (Mondays, Wednesdays and Saturdays; 3FDb group) non-consecutive starvation days per week.
The diet was accepted and promoted 100 % survival. Both treatments had similar growth and ingestion rates between biological samplings (days 0-28 or 29-56) and during the whole experiment (days 0-56; SFR ≈ 1.47-1.59 % BW/day; SGR ≈ 0.78-0.88 % BW/day). As previously observed in fish (Ali et al., 2003), the daily intake raised the day after starvation. However, three days of fasting enhanced protein productive value and feed efficiency, compared with only two days of fasting (PPV of 23.90 vs. 16.80 % and FE of 58.65 vs. 48.31 %, respectively).
The applied feeding schemes did not affect tissues proximate composition (muscle, digestive gland and whole animals) nor nutrients digestibility (dry matter, protein, lipids or carbohydrates). Hence, results suggest that the inclusion of non-consecutive fasting days in feeding protocols should be a viable methodology for the development of common octopus’ culture as it might reduce operational production costs (quantity of food and staff for feeding). Consequently, fasting in feeding schemes were incorporated in subsequent ongrowing experiences.
The third section of this thesis focused on trying to correct possible lipid deficiencies of raw materials included as ingredients, after being dehydrated, in formulated diets. As referred in the first section of this document, dehydration treatments partially destroyed polar lipids. The use of polar lipids as dietary supplement resulted in satisfactory results regarding growth, survival and digestion in other marine species. In addition, previous studies pointed the relevance of phospholipids for cephalopods, in composition and nutrition terms. Following this reasoning, 2% of soybean or marine lecithins were included in diets to test the effect of polar lipids supplementation in Octopus vulgaris. These lecithins vary on their origin and the richness of long chain polyunsaturated fatty acids (PUFA) omega-3, i.e. eicosapentaenoic (EPA) and docosahexaenoic (DHA) acids, or omega-6, i.e. arachidonic acid (AA).
In Chapter IV, soybean lecithin dietary supplementation was tested on O. vulgaris performance in individual (Experiment 1) or grouped (Experiment 2) kept animals. Applying previous knowledge about feeding, food was supplied on alternate days during 56 days.
Individually reared animals were fed two semi-moist diets (N= 8) without (VEGENAT-LS0) or with supplementation (VEGENAT-LS2, substituting 2 % of starch by lecithin). Formulated feeds presented low disintegration rates in water (12-25 % dry matter loss in 24 h). However, VEGENAT-LS2 showed higher disaggregation rates (P<0.05) for a combined effect between lecithin and a reduction on starch. These differences also translated into lower lipid content, and consequently lower energy content, while higher CHOT and P/E ratio was verified in VEGENAT- LS0 compared with VEGENAT-LS2 (P<0.05). Soybean lecithin supplement increased PA, PE, PIG and TPL in VEGENAT-LS2 compared to VEGENAT-LS0 but also caused a reduction on TG and TNL (P<0.05).
Formulated feeds were ingested providing animal growth and faeces production. Mortality was detected in both treatments along the first month of rearing reaching a final survival of 87.5 % at the end of the experiment.
Soy lecithin supplementation did not generate statistical differences on growth (9.0-10.5 g/day), intake (AFR of 18.2-19.1 g/day), feed efficiency (FE of 48.32-60.15 %) or digestive gland condition (DGI of 4.6 % for both treatments) at the end of the experiment (P>0.05). Furthermore, there were no differences between those indices considering the experimental period separately (Days 1-28 or 29-56; P>0.05). SFR and SGR were reduced in both treatments in the second period (days 29-56) compared to the first (days 1-28).
Faeces from octopus fed the VEGENAT-LS0 feed had higher CHOT content while that of those fed VEGENAT-LS2 presented higher protein and ash contents (P<0.05). In contrast, similar lipid content was quantified between faeces from both experimental groups. ADC dry matter was higher in VEGENAT-LS0 (76 %) than VEGENAT-LS2 (63%; P<0.05). VEGENAT-LS0 also had higher ADC protein and ADC lipid (Table 4; P<0.05). The polar fraction presented similar ADC between experimental treatments (95.4 %; P>0.05) but the neutral fraction was better digested in VEGENAT-LS0 (88.9 %) compared with VEGENAT-LS2 (83.2%; P<0.05). In general terms, polar lipid classes were better digested (ADCs above 80 %, excluding PS+PI; Fig. 1a) compared with neutral lipid classes (ADCs above 60 %). LPC, SM, PS+PI ADCs were higher in VEGENAT-LS2 while LPE, PE and PIG ADCs were higher in VEGENAT-LS0 (P<0.05). There were no differences between DG, CHO, FFA and TG ADCs while MG ADC was higher in VEGENAT-LS0 group. The low SE concentration detected either in diets and faeces did not permit an accurate calculation of SE ADC. No differences were verified in the macronutrient composition of the digestive gland between treatments (P>0.05). However, the animals fed VEGENAT-LS2 displayed higher lipid content in the carcass and whole composition compared to VEGENAT-LS0 (P<0.05). Both experimental groups presented higher lipid and lower ash contents compared with the initial macronutrient composition of octopuses prior to the essay (P<0.05).
The analysed tissues did not present statistical differences on the composition of polar and neutral lipids regardless of the diet, with small exceptions. In this sense, higher PS+PI and lower DG contents were detected in the digestive gland of VEGENAT-LS2 fed animals (P<0.05). Higher LPC was observed in the carcass of VEGENAT-LS0 animals (P<0.05). The digestive gland was a rich neutral lipid organ (84.2-89.8 % lipids content). In this sense, TG and FFA represented together around 60 % lipids while other neutral lipids individually represented less than 12 % lipids. The polar fraction (10-15 % total lipids) was mainly formed by PC (4.15-6.52 % lipids), PS+PI (1.98-3.09 % lipids) and PE (2.60-4.62 % lipids). In contrast, the carcass was rich in polar lipids (68.2-69.7 % lipids) and was mainly composed by PC (27.56-29.79 % lipids), PE (18.38- 20.02 % lipids) and PS+PI (14.30-15.47 % lipids) while the neutral fraction was almost entirely CHO (18.0-20.3 % lipids) followed by FFA (6.01-6.21 % lipids).
In experiment 2, 40 juveniles of common octopus were maintained in group (10 octopuses per tank). They were fed along 56 days (alternate days; 3-4 days of starvation per week) with a mixed natural diet composed by bogue and crab (2BOGUE:1CRAB; cycles of two-days bogue and one- day crab) or the supplemented diet with soybean lecithin VEGENAT-LS2. The natural diet had higher moisture, protein and mineral content while very low CHOT compared to VEGENAT-LS2 (P<0.05). Both diets were similar in lipid content (10.5-10.8 %; P>0.05) and energetics (2227- 2284 KJ/100 g), but the formulated feed P/E was 5 g protein/MJ lower than those calculated for the natural diet. The soy lecithin supplement promoted a higher polar lipid content. In this sense, VEGENAT-LS2 presented higher percentages of TPL, PC, PS+PI, PA, PIG, MG, CHO and lower PE, TNL, DG, FFA, TG and SE compared to 2BOGUE:1CRAB (P<0.05). Mortality was registered only in VEGENAT-LS2 fed group from the beginning of the experience with a final survival of 50 %. In contrast the 2BOGUE:1CRAB diet promoted 100% survival.
Regarding the whole experience (days 1-56), 2BOGUE:1CRAB presented higher feeding rates and promoted higher growth than VEGENAT-LS2. Again, AFRdw was similar (10.5-10.9 g /day). The same values were observed for FE (53.6-59.9 %) and FCR (1.8-1.9) but 2BOGUE:1CRAB presented better PPV and LPV. Both diets provided similar digestive gland index (DGI) values.
Faeces from animals fed natural diet had higher protein, lipid and ash content than VEGENAT- LS2 (P<0.05). In contrast, the inclusion of starch was reflected in a higher amount of CHOT in VEGENAT-LS2. Similar ADC dry matter (49.0-57.0 %) and ADC protein (75.9-77.9 %) were verified between treatments. In contrast, a markedly lower ADC lipid was detected in 2BOGUE:1CRAB (31.8 %) compared to the tested supplemented formulated diet (65.4 %). Polar lipids were more efficiently digested than neutral lipids in both treatments: ADC TPL values were 84.3±4.3 % and 73.2±3.3 % (P<0.05) and ADC TNL were 55.1±10.1 % and 15.5±10.5 % (P<0.05) for VEGENAT-LS2 and 2BOGUE:1CRAB, respectively. There were no significant differences in the ADC of LPC, LPE, PA and FFA (P>0.05). However, 2BOGUE:1CRAB presented better PC, PS+PI, PE and SE ADCs while VEGENAT-LS2 showed better ADCs for SM, PIG, MG, DG, CHO and TG (P<0.05).
The whole animal and carcass proximal composition were similar between treatments (P>0.05).
However, a higher ash and CHOT contents were verified in the digestive gland of animals fed VEGENAT-LS2 (P<0.05). A higher lipid content was verified in reared octopus compared to the initial composition of wild individuals (P<0.05).
In the digestive gland around 90.0 % lipids were neutral lipids. In this sense, FFA was higher in VEGENAT-LS2 while 2BOGUE:1CRAB presented higher TG content in the digestive gland (P<0.05). The polar fraction (9.08-13.22 % lipids) was similar between treatments (P>0.05) and mainly PC (3.84-5.90 % lipids), PS+PI (1.91-2.08 % lipids) and PE (1.46-3.63 % lipids). In contrast, the carcass was rich in polar lipids and was mainly composed by PC (22.73-23.92 % lipids), PS+PI (14.75-16.11 % lipids) and PE (18.73-20.65 % lipids). 2BOGUE:1CRAB presented a slightly higher PE content (P<0.05). No differences were verified in the neutral fraction and was remarkable the richness in the content of CHO in both treatments (21.13-21.2 % lipids). The other lipid classes were below 10.0 % and DG was non-detected.
In Chapter V, the effect of marine phospholipids as dietary supplement was evaluated. With this aim, marine lecithin was included in formulated diets to assess O. vulgaris performance in individual (Experiment 1) or grouped (Experiment 2) kept animals. Animals were fed three semi- moist diets (44-47 % moisture) without marine lecithin (CALPRO) or replacing 2 % of starch by lecithin (CALPRO-LM and VEGENAT-LM). Diets were made with dry ingredients exclusively freeze-dried at IMIDA (CALPRO and CALPRO-LM) or included some ingredient freeze-dried by a company (VEGENAT-LM). The applied feeding protocol included 3 non-consecutive days of fasting per week (Mondays, Wednesdays and Saturdays).
All tested diets were consistent prior to be placed in water. However, supplemented feeds presented higher disintegration in seawater (WSI = 20.3-20.8 % after 24 hours) compared to CALPRO feed (WSI = 42 %; P<0.05). Diets had 44.5-46.6 % moisture. Protein content was similar between CALPRO and CALPRO-LM feeds (67.3-67.4 %) while lower in VEGENAT-LM (64.0 %;
P<0.05). Marine lecithin supplement (2 %) increased crude lipid content in supplemented diets, i.e. CALPRO-LM and VEGENAT-LM (P<0.05). The ash content was significantly higher in VEGENAT-LM (P<0.05). CHOT was similar between supplemented diets while higher in CALPRO due to its high starch content in the basal composition (P<0.05). The inclusion of 2 % of marine lecithin provided higher energy, about 1MJ/kg, to CALPRO-LM and VEGENAT-LM diets compared with CALPRO feed. In contrast, the CALPRO P/E ratio was about 2g/MJ higher than lecithin supplemented diets. All feeds presented higher content of TNL (67.56-87.14 % of total lipid) than TPL (12.87-32.45 % of total lipid). The addition of marine lecithin (2%) did not generate significant differences in TPL or TNL between CALPRO and CALPRO-LM feeds (P>0.05). However, VEGENAT-LM presented significantly higher PC and TPL percentages (P<0.05). In addition, this diet also presented lower FFA, SE and TNL percentages compared with CALPRO and CALPRO-LM feeds.
In experiment 1, subadult octopus (684-769 g BW) were reared in isolation. Along the whole period (day 0-56) supplemented treatments (CALPRO-LM and VEGENAT-LM) presented similar survival (87.5 %), growth, feed efficiency and animal condition (AGR, SGR, FE and DGI; P=0.05) but VEGENAT-LM presented a higher intake (AFR and SFR; P<0.05). CALPRO provided higher growth (P<0.05) than both supplemented diets. The feed efficiency of this diet was similar to CALPRO-LM and higher than VEGENAT-LM (P=0.05). Except for the content of acid insoluble ashes (P<0.05), the proximate composition of faeces was similar in all treatments (P>0.05). Both supplemented diets presented similar digestibility to CALPRO (P>0.05). However, VEGENAT-LM was better digested than CALPRO-LM (P<0.05). No differences were verified between treatments in protein digestibility (≈94-95 %; P>0.05). CALPRO and CALPRO-LM diets presented similar and lower lipid digestibility (≈90-91 %; P>0.05) than VEGENAT-LM (94 %; P<0.05). Total polar lipids ADC was significantly higher in VEGENAT-LM (97 %; P<0.05) compared with both CALPRO (90 %) and CALPRO-LM (91 %). VEGENAT-LM presented the higher ADCs of almost all of the listed polar lipid classes (excluding LPC ADC, which was higher in CALPRO-LM; P<0.05).
Higher PA and PE ADCs and lower PC and PS+PI ADCs were registered in CALPRO compared with CALPRO-LM feed (P<0.05). There were no differences on total neutral lipids digestibility (P>0.05). DG ADC was similar between diets. CHO ADC was significantly lower in CALPRO-LM feed (P<0.05). CALPRO and CALPRO-LM presented higher FFA and SE ADCs compared with VEGENAT-LM (P<0.05). TG ADC was high in all treatments (above 99 %) with significantly higher values in CALPRO-LM and VEGENAT-LM feeds (P<0.05). SE ADC was markedly low in VEGENAT- LM compared with CALPRO and CALPRO-LM treatments (P<0.05).
The proximate composition of the digestive gland was similar between all treatments at the end of the experiment. The 2 % marine lecithin supplementation was also not translated in crude lipid differences in the carcass (15.1-17.9 %; P>0.05). CALPRO and VEGENAT-LM fed animals presented similar moisture, crude protein, ash and CHOT (P>0.05) percentages in the carcass.
CALPRO-LM presented lower crude protein than the other treatments (P<0.05). Both supplemented treatments had similar CHOT percentages than CALPRO (P>0.05) but, among them, CALPRO-LM registered the highest (P<0.05).
Regarding the final macronutrient composition of whole animals, CALPRO-LM presented higher moisture than CALPRO (P<0.05) but no differences were observed when comparing both with VEGENAT-LM (P>0.05). CALPRO and VEGENAT-LM presented similar protein percentages (P>0.05) and both presented significantly higher content compared with CALPRO-LM (P<0.05).
There were similar lipid and ash contents between all treatments. All treatments displayed higher lipid content (3.56-3.96 %) compared with the initial composition (2.02 %; P<0.05). In addition, both CALPRO and CALPRO-LM displayed lower protein content than initial animals (77.86-81.11 vs. 83.58 %, respectively; P<0.05).
The lipid profile in tissues was practically not modified by dietary marine lecithin supplementation. Higher FFA were observed in the digestive gland of CALPRO fed animals compared with either CALPRO-LM or VEGENAT-LM (P<0.05). There were no significant differences in lipid classes from the carcass (P>0.05). The digestive gland and carcass presented similar percentages of total polar (TPL) and total neutral lipids (TNL) regardless the supplied diet (P>0.05).
In experiment 2, 18 juveniles of common octopus (≈ 700 g) were maintained in group (6 octopuses per tank). They were fed along 56 days applying a feeding protocol of three days of fasting per week (Mondays, Wednesdays and Saturdays). Along the whole period (days 0-56) mortality was only registered in CALPRO (survival of 83.33 %). Final weights and growth (AGR, SGR, Wg) were lower in CALPRO-LM, especially in VEGENAT- LM, compared with CALPRO (22.03 g/day, 1.71 % BW/day and 1234 g, respectively). Feeding rates (AFR, SFR, APFR and ALFR) were similar between treatments so the most effective feed, i.e. upper FE and lower FCR, was CALPRO (60 % and FCR of 1.65, respectively). The digestive gland index, commonly related with animal condition, was similar between CALPRO and VEGENAT-LM (P=0.704) and both lower than CALPRO-LM (P<0.05).
Both supplemented feeds (CALPRO-LM and VEGENAT-LM) generated higher crude protein, crude lipid and ash (P<0.05) faecal contents than CALPRO fed animals. In contrast, CHOT was higher in CALPRO (P<0.05). All diets were effectively digested (66-78 %) and presented high protein (89-93 %) and lipid ADCs (80-92 %). Among diets, the lower protein and lipid ADCs were registered at CALPRO-LM treatment. Accordingly, TPL and TNL digestibility were also lower in CALPRO-LM (75 and 81 %, respectively; P<0.05). VEGENAT-LM presented similar TPL and TNL (P>0.05) ADCs than CALPRO. All treatments presented similar LPC and PA digestibility. CALPRO and VEGENAT-LM presented similar ADCs for SM, PC, MG, CHO and TG (P>0.05) while ADCs of these lipid classes were always lower in CALPRO-LM (P<0.05). DG, FFA and SE were better digested in CALPRO diet (P<0.05). PS+PI and PE registered the higher ADC values in VEGENAT- LM (P<0.05). TG were effectively digested, above 97 % in all groups.
Moisture, crude protein, crude lipid and ash contents were similar in the digestive gland (P>0.05). Similar CHOT was verified between supplemented treatments (P>0.05), but lower content was detected in VEGENAT-LM than in CALPRO (P<0.05). There was lower moisture in the carcass and the whole composition of supplemented treatments (P<0.05). Furthermore, higher crude lipid content was found in CALPRO-LM, either in the carcass or in complete animals (P<0.05). The only treatment that preserved the composition obtained from wild animals (prior to the experiment) was VEGENAT-LM. CALPRO increased moisture, lipid and ash contents (P<0.05). A decrease of protein content and an increase of lipids was verified in CALPRO-LM.
The lipid profile in tissues was practically not modified by dietary marine lecithin supplementation. Higher FFA were observed in the digestive gland of CALPRO fed animals compared with either CALPRO-LM or VEGENAT-LM (P<0.05). There were no significant differences in lipid classes from the carcass (P>0.05). The digestive gland and carcass presented similar percentages of total polar and total neutral lipids regardless the supplied diet (P>0.05).
In the digestive gland, the main lipid classes of the polar fraction were PC and PE while FFA, TG and SE were mainly found in the neutral fraction. PE content was lower in CALPRO-LM (P<0.05) than in CALPRO but both treatments presented similar percentages than VEGENAT-LM (P>0.05).
FFA was lower in CALPRO-LM than either CALPRO or VEGENAT-LM (P<0.05). TG were significantly higher in CALPRO-LM than CALPRO (P<0.05) but similar to VEGENAT-LM (P>0.05).
TPL percentages were lower in CALPRO-LM (P<0.05) than in CALPRO but both treatments presented similar percentages than VEGENAT-LM (P>0.05). CALPRO-LM individuals had higher TNL percentages compared with CALPRO fed animals (P<0.05), but no differences were detected in regard to VEGENAT-LM (P>0.05). The major polar lipid classes found in the carcass were PC (23 %), PS+PI (12-14 %) and PE (18-20 %) while in the neutral fraction were CHO (21-22 %) and FFA (10-13 %). The lipid profile was similar between treatments excepting sterols (SE). In this sense, the carcass of non-supplemented fed animals (CALPRO treatment) presented higher SE percentage than VEGENAT-LM (P<0.05). TPL and TNL percentages were similar between treatments (P>0.05).
Therefore, results do not suggest a beneficial effect on performance related with polar lipids dietary supplementation neither when including soybean (vegetal origin) nor marine (marine animal origin) lecithins.
Finally, the fourth section of this document consists of Chapter VI and embody the application of previous knowledge on artificial diets development acquired for O. vulgaris in other cephalopod species: the southern or Patagonian red octopus Enteroctopus megalocyathus.
After the satisfactory results obtained with CALPRO diet (Chapter V) in O. vulgaris, we decided to test if it was feasible to use the same feed/formulation in E. megalocyathus, another cephalopod species with interest for aquaculture.
The experiment was performed in 24 juveniles (≈320 g) during 58 days, applying a feeding protocol with two days of fasting per week (Wednesdays and Saturdays), of a given diet: a natural diet of Chilean crab (Cancer edwardsii; diet Jaiba N=6), one artificial diet developed for E. megalocyathus (diet HIM37; N=6) and two diets based on O. vulgaris previously tested feeds based on freeze-dried ingredients (diets IMIDA30 and IMIDA37; N=6 respectively). Jaiba had higher moisture content than formulated diets (P<0.05). The protein content was above 60 % DW in all supplied diets but IMIDA37 and HIM37 presented higher content of this macronutrient, above 75 % in both cases (P<0.05). The lipid content was similar between feeds (P>0.05).
IMIDA30 had more carbohydrates, while the other diets (Jaiba, IMIDA37 e HIM37), which had higher P/E presented a significantly lower content (P<0.05). The mineral content was below 14 %; the lowest values were verified in Jaiba and IMIDA30 (P<0.05), while IMIDA37 or HIM37 presente intermediate percentages (P<0.05). All artificial feeds were energetically similar (2024- 2111 KJ/100g) with the lowest value registered in Jaiba (1882,40 KJ/100g). Consequently, IMIDA37 and HIM37 presented closer P/E to Jaiba (≈37 g/MJ), while IMIDA30 presented a lower value (29,81 g/MJ). The different origin of raw materials included in IMIDA37 or HIM37 diets did not generate marked differences in terms of proximate composition or stability in water (P>0.05).
Mortality was recorded in all experimental groups. Survival was lower in IMIDA37 (66,67%) while it was similar in the remaining treatments (83,33%). Unless HIM37, which generated negative values, formulated diets promoted growht (Wg, AGR y SGR; P<0.05). Jaiba promoted higher growth and ingestion rates than the remaining feeds (P<0.05). In fact, intake was similar among artificial diets (AFR y SFR; P>0.05). IMIDA30 and IMIDA37 promoted similar growth rates (AGR and SGR; P>0.05), while HIM37 registered negative growth (P<0.05). Food conversion (FCR) and feed efficiency (FE) of Jaiba was lower (P<0.05), compared to IMIDA diets, whic were similar (P>0.05). Jaiba’s FCR was two or three times higher than IMIDA30 or IMIDA37 diets, respectively (P<0.05). The nutritive use of HIM37 by octopus was negative. Protein retention (PPV) was similar between Jaiba and IMIDA feeds (P>0.05). These artificial feeds presented PPV above 65 % while the value was negative in HIM37. IMIDA37 presented a LPV notably higher than the other diets (P<0.05). Animals’ condition was assessed looking at the digestive gland index (DGI), which was similar between formulated diets (P>0.05) but lower than the control Jaiba (P<0.05).
Results suggest that semi-moist diets developed for O. vulgaris (format and formulation) can be supplied to E. megalocyathus with acceptable performance.
