Influence of broodstock nutrition on egg quality and fatty acid composition in California Yellowtail (2025)

Search for other works by this author on:

Article history

Abstract

INTRODUCTION

The California Yellowtail (CYT) Seriola dorsalis, also formerly known as the Yellowtail Kingfish Seriola lalandi, is a carangid fish found in the temperate waters of southern California, with recreational, commercial, and potential global culture value (Martinez‐Takeshita et al. 2015; Ben‐Aderet et al. 2020; Rotman et al. 2021). In 2021, the commercial landings of CYT in California amounted to 17.6 metric tons and had an exvessel value of US$111,690 (California Department of Fish and Wildlife 2022). Commercial landings in the state are limited by gear restrictions and have experienced a decrease overall since the 1980s, whereas the commercial value of CYT has steadily increased (California Department of Fish and Wildlife 2022). This increase in commercial value can be largely attributed to the popularity of a similar species, the Japanese Amberjack Seriola quinqueradiata, which is highly valued as sashimi. In 2021, Japanese export of frozen yellowtail Seriola sp. alone was valued at $76 million (Loew 2021). There are currently no existing commercial CYT farms in the United States and global commercial production of the species is limited to a few farms in Baja California Sur, Mexico. These farms largely export to the United States, with the farm Omega Azul reporting an export of about 150 metric tons in 2017 (Seafood Watch 2020).

Since 2003, research into the development of culture methods for CYT in the United States has been largely accomplished by Hubbs–SeaWorld Research Institute (HSWRI; Rotman et al. 2021). For every cultured species, the reliable production of juveniles is dependent on high‐quality eggs, which are especially influenced by nutrients obtained from the broodstock diet (Bromage and Roberts 1995; Brown et al. 2006; Migaud et al. 2013; Biswajit et al. 2020; El‐Gamal et al. 2020). Egg quality can be defined as the ability of the egg to be fertilized and subsequently survive and develop into larvae (Brooks et al. 1997; Bobe and Labbé 2010; Bobe 2015). For marine fish like the CYT, floating eggs are presumed to be more viable, as sinking eggs are more likely to be poorly developed and may be linked to lower amounts of the enzymes involved in yolk formation (Carnevali et al. 2001). In CYT culture research at HSWRI, standard egg quality metrics, such as egg and oil diameters, viability, egg hatching rate, survival to first feeding, and days posthatch (dph), are usually recorded for every spawn to help determine egg quality for spawning events (Stuart et al. 2020). In addition to egg quality metrics, the biochemical composition of eggs is important in assessing egg quality. This is because the nutrients in the oil droplet of the egg and yolk sac of the larvae must be able to support larval development, especially during the endogenous feeding period (Wilson 2009; Fuiman and Faulk 2013; Rodríguez‐Barreto et al. 2014). Among all of the nutrients, fatty acids (FAs) in particular have been used to evaluate egg quality because FA composition can greatly influence spawn characteristics, including fertilization rate, cell symmetry, hatching, larval growth, and survival (Salze 2005; Tocher 2010; Parma et al. 2015). This is especially evident for essential FAs (EFAs) like arachidonic acid (ARA), which has been shown to significantly improve egg hatch rates and viability in CYT (Stuart et al. 2018). Studies like these, along with other broodstock diet studies, have shown that the lipid composition of eggs is directly correlated and influenced by the lipid composition of the broodstock diet (Watanabe et al. 1984a, 1984b; Mourente and Odriozola 1990; Harel et al. 1994; Almansa et al. 1999). This influence of dietary lipids is especially evident in CYT because they are continuous synchronous spawners, exhibiting a prolonged spawning season from March to September (Schmidt et al. 2021). Unlike asynchronous spawners, CYT spawn multiple times per season and immediately transfer nutrients from their diet into their eggs through a short vitellogenic period (Izquierdo et al. 2001).

It is also important to examine the FA composition of eggs in terms of polar and neutral lipid fractions because they utilize and reflect dietary lipids differently. Neutral lipid classes are hydrophobic and include sterols, wax esters, and triacylglycerols, which provide important energy reserves that may be stored in the oil droplet (Wiegand 1996; Couturier et al. 2020). On the other hand, polar lipid classes have a hydrophilic head with a hydrophobic tail; they include phospholipids and phosphoglycerides, and they are responsible for membrane structure and fluidity (Higgs and Dong 2000; Couturier et al. 2020).

Fresh fishery products (FFPs; i.e., fish and squid) that are frozen and then thawed are typically the standard diet used for feeding many marine broodfish, including CYT. Absent formulated broodstock feeds, FFPs have proven to be the most effective way of meeting the nutritional requirements of the broodfish and ensuring good‐quality eggs. However, the use of FFPs requires careful management of the quality of the product, which is often outside the control of the producer since the nutritional composition can vary seasonally (Bruce et al. 1999). Additionally, from a biosecurity standpoint, FFPs that are thawed may be a source of harmful bacteria if rigorous sanitary practices are not followed, or they may harbor infectious diseases that can be transferred to the broodstock. Considering these risks, the use of a formulated diet instead of FFPs could give producers an advantage and would allow for species‐specific nutrients to be added when necessary. This study compares a control FFP diet (FFP‐D) to commercial diet alternatives (a reference diet [REF‐D] and a commercial diet [COM‐D]) in CYT broodstock to determine differences in egg production, quality, and biochemical composition.

METHODS

Broodstock

The CYT broodstock used for this study were part of an F₁ generation broodstock population produced at the HSWRI laboratory (San Diego, California) under protocols similar to those described by Stuart and Drawbridge (2013). The F₁ broodfish were produced in 2014 and therefore were 3 years old at the start of this study. In 2017, we conducted a pilot trial consisting of one female (6.14 kg) and two males (mean ± SD = 6.43 ± 0.49 kg) in a single spawning tank. The following year, six females and 12 males with initial weights of 8.40 ± 1.50 kg and 7.90 ± 1.46 kg, respectively, were placed into six replicate spawning tanks (one female and two males per tank). The spawning tanks were 10 m³ (3.7 m diameter × 1.0 m depth). The fish were exposed to shaded natural light and ambient seawater temperatures ranging from 13°C to 25°C. Flow‐through seawater (34–35%) was supplied to all tanks at a flow rate of 30 L/min, and 500‐μm egg collector nets were placed on the outside of each spawning tank to collect all eggs from the effluent stream.

Experimental diet

Three different experimental diets were used in this study. The one tank of CYT in the 2017 pilot trial was fed the REF‐D, which was a commercial larval starter feed (BioVita Starter; Bio‐Oregon); it is referred to here as a “reference diet” because it was not specifically formulated for broodfish. Three tanks of CYT in the 2018 trial were fed the COM‐D, which was a powdered commercial broodstock diet (Breed‐M; INVE), and the other three tanks were fed FFPs composed of 70% sardine and 30% squid supplemented with vitamins as described by Stuart and Drawbridge (2013). Both the REF‐D and the COM‐D were formed into a moist paste by adding 23% of the diet's weight in water using a Hobart mixer. This paste was shaped into a cylindrical form of 3.0 cm diameter using a sausage tube attachment and was cut to 8.0 cm in length. Fish were fed experimental diets three times per week during January–March and then five times per week during the spawning season from April through October. Broodstock were fed to satiation with the FFP‐D at 3–5% of body weight per day or with the REF‐D or COM‐D at 1–3% of body weight per day. The differences in feed rations were calculated to account for the difference in moisture content of the experimental diets. All feeds were sampled five times during the spawning season and stored at −80°C for biochemical analysis.

Spawning

Eggs were collected from each broodstock tank in the morning after spawning events, which usually occurred in the late evening (Stuart et al. 2020). Viability was calculated for each spawn as the ratio of the floating portion of the spawning event to the total eggs produced, multiplied by 100. For each spawning event, the following egg and larval measures were recorded: total egg production, egg diameter (±0.01 mm), oil diameter (±0.01 mm), percent oil volume, hatch rate (%), survival to first feeding (SFF; %), yolk sac volume (±0.01 mm³), and notochord length at hatch (±0.01 mm). The total egg production was enumerated volumetrically in 1‐L graduated cylinders by totaling up the milliliters of floating, sinking, and neutral fractions in the water column. Finally, the number of eggs per milliliter was multiplied by each volumetric fraction to obtain the numbers of eggs per spawn. A digital photomicrograph was taken of subsamples of eggs from the floating, neutral, and sinking portions of each spawning event at 16× magnification (Leica MZ16 Macro). Once the digital image was taken, the egg and oil diameters were measured for 20 eggs by using calibrated software (Image‐Pro Plus). Eggs and oil globules (each CYT egg has a single oil globule) were assumed to be spherical, and volumes were calculated from single measurements of diameter according to the formula for the volume of a sphere (i.e., volume = 4/3 × π × radius³). The egg hatching rate was determined by filling five 1.0‐L beakers with 800 mL of filtered seawater and placing 20 floating eggs into each beaker. On the day of hatch (0 dph), the number of hatched larvae was counted in each beaker. Hatch success was calculated as the average for all five replicate beakers (totaling 100 eggs) for each spawning event. Similar to the hatch rate, SFF was measured at 2–3 dph (depending on spawning temperature) by counting the number of surviving larvae in five replicate beakers (totaling 100 larvae). Broodstock were weighed before and after the experiment, and the average of these two weights was taken to calculate fecundity. Fecundity (eggs/kg of females) was calculated by dividing the total egg production by the total weight of females that contributed to the spawns in a respective diet treatment. Average batch fecundity (eggs·kg⁻¹·spawn⁻¹) was calculated by dividing the egg production from each spawn by the weight of the corresponding female broodstock for that spawn.

Fatty acid analysis

Egg samples used for FA analysis were from alternating spawns of each diet to represent potential seasonal spawn variation. Eight of 15 spawns were analyzed from the REF‐D broodstock, 7 of 11 spawns were analyzed from the COM‐D group, and 12 of 22 spawns were analyzed from the FFP‐D group. A total of 27 spawns from the three different diet treatments were weighed and freeze‐dried in preparation for FA extraction. Frozen eggs were thawed to room temperature, and total lipids were extracted and suspended in 4 mL of dichloromethane using the methods described by Folch et al. (1957) and modified by Parrish (1999). From the total lipid extract, lipids were separated into neutral and polar FA fractions by using silica solid phase extraction columns (Thermo Fisher Scientific Inc). Total lipid extract from each sample was loaded into an individual column and eluted first with 10 mL of 98:2 dichloromethane : methanol solution to extract neutral FAs and then 10 mL of 100% methanol to extract polar FAs. After separation, all solvents were evaporated, and the remaining FAs were methylated using the methods from Lepage and Roy (1984) to produce FA methyl esters (FAMEs) for analysis. Polar and neutral FA samples were treated identically in the methylation process.

Samples of the REF‐D and COM‐D, as well as samples of the FFP‐D with and without vitamins added, were freeze‐dried and homogenized for analysis. The FAs were extracted from diets and methylated using a direct esterification process. Samples were sonicated in dichloromethane to extract FAs and methylated to produce FAMEs for analysis. The FFP‐D proportions were calculated accordingly to account for discrepancies between feeding schedules during different spawn years. Corresponding FA concentrations for each diet treatment were also calculated based on these proportions and the FA concentrations detected in each type of whole sardine or squid.

The resulting FAMEs from both extraction methods were suspended in 1.5 mL of hexane in amber headspace vials that were compatible with the autosampler carousel. Samples were then analyzed for FAME content on a Perkin‐Elmer Clarus 680/600 gas chromatograph–mass spectrometer with a 30‐m Thermo Fisher TR‐5 general purpose column. Each sample in the autosampler was injected into the column at a volume of 1.0 μL, was slowly heated to 250°C, and was held at this temperature for 10 min as it moved through the column. Resulting chromatograms were interpreted using TurboMass quantification software. The quantification method was calibrated using a 37‐component FAME standard mix (Supelco 37 FAME Mix; Millipore Sigma), which contained saturated FAs (SFAs), monounsaturated FAs (MUFAs), and polyunsaturated FAs (PUFAs) of interest. The final software output provides FA totals for each of the 37 target compounds in micrograms per milliliter. For each sample, these totals were converted into micrograms per milligram, and relative percentages were calculated for each of the FAs.

Fatty acid ratios

One method used to determine the broodstock nutritional quality in terms of EFAs is the ratio of specific FAs in eggs from the broodstock to the total FAs present in the diet fed to the broodstock (the egg : diet [ED] ratio; Mejri et al. 2021). The ED ratio indicates whether a specific FA in the diet is selectively incorporated by the eggs. If the relative proportion of a specific FA in eggs compared to the broodstock diet is less than or equal to 1.0, then the specific requirement for this FA is considered to have been satisfied. In contrast, if the ED ratio is greater than 1.0, then we consider this FA to have been selectively incorporated into the egg by the female broodfish, which may indicate a potential dietary deficiency.

Proximate analysis

Freeze‐dried and homogenized subsamples of CYT eggs and experimental diets were sent to a partnering laboratory for proximate analysis (New Jersey Feed Laboratory). All samples were analyzed for total protein (mg/g), total fat (mg/g), total fiber (mg/g), total ash (mg/g), total carbohydrates (mg/g), and total calories (kcal/kg).

Statistical analysis

Every spawn from each diet treatment was treated as a replicate. If tank replicates were available, a permutational analysis of variance (PERMANOVA; with 9999 permutations) was performed on tank replicates to identify potential interfemale variation in the FA profiles of eggs, with tank as a factor. Permutational ANOVA, including a posteriori pairwise comparisons, was performed on the FA profiles of the three diets and separately on the polar and neutral FA profiles of CYT eggs. Each PERMANOVA was tested with one factor: diet treatment (REF‐D, COM‐D, and FFP‐D).

Assumptions of multivariate homoscedasticity were verified with a PERMDISP test, and data were transformed (arcsine–square root) when necessary. Canonical analysis of principal coordinates (CAP) of the Bray–Curtis similarity matrix was run to visualize FA profiles in a multidimensional scale. A follow‐up one‐way ANOVA was performed on total SFAs, MUFAs, PUFAs, and each EFA, including the precursors linoleic acid (LA; 18:2[n‐6]) and α‐linolenic acid (LNA; 18:3[n‐3], where the number to the left of the colon is the number of carbon atoms, the number to the right of the colon is the number of double bonds, and the number after the hyphen indicates the position of the first double bond from the methyl end).

If tank replicates were available, a one‐way ANOVA was performed on the tank replicates to identify potential interfemale variation in egg quality metrics (egg diameter, oil diameter, viability, average batch fecundity, hatch rate, and SFF), with tank as a factor. Analysis of variance was performed on each egg quality metric and on proximate components (total protein, total fat, total fiber, total ash, carbohydrates, and calories), with diet treatment as a fixed factor. Assumptions of homoscedasticity and normality were tested with Levene's test and the Shapiro–Wilk test, respectively. If variances were heteroscedastic, a Kruskal–Wallis test was applied. A follow‐up Tukey test was run for pairwise analysis of all ANOVA results. For all statistical analyses, significance was assessed at p‐values less than 0.01. Permutational ANOVAs and visualization were done with PRIMER 7 (version 7.1.12) and PERMANOVA+ (version 1.0.2), and ANOVAs were performed with RStudio.

RESULTS

Egg production and egg quality

Only fish that were fed the FFP‐D produced spawns in all three replicate tanks. The FFP‐D replicate groups produced spawns with similar egg quality metrics except for viability (Table 1; F = 7.7424, p = 4.07 × 10⁻³).

The REF‐D, COM‐D, and FFP‐D treatments produced spawns with varying egg production and egg quality results (Table 2). In 2018, all three replicate tanks that were fed the FFP‐D spawned, but only one out of three tanks that were fed the COM‐D spawned. The brood female that spawned in the COM‐D treatment was the same female that received the REF‐D treatment in our 2017 pilot experiment. When comparing between the spawns produced by the same female when fed the COM‐D versus the REF‐D, the COM‐D treatment spawns had significantly larger egg diameters (F = 14.102, p = 2.97 × 10⁻⁴) and oil diameters (F = 31.857, p = 3.66 × 10⁻⁷). However, the COM‐D spawns had a significantly poorer hatch rate (F = 14.26, p = 5.32 × 10⁻⁶) and a significantly lower SFF (F = 12.905, p = 3.89 × 10⁻⁵), with only 2 of 11 spawns successfully hatching. These two spawns had hatch rates of 45% and 30% and SFF values of 7% and 0%. When comparing the COM‐D and FFP‐D treatments from 2018, both treatments produced similar egg quality except that the FFP‐D achieved a significantly better hatch rate and SFF than the COM‐D.

Proximate analysis of diets and eggs

Proximate analysis results for the diets suggested that the REF‐D had a higher carbohydrate content than the COM‐D and FFP‐D (Table 3). Egg proximate composition was statistically similar overall among the three diet treatments (Table 4).

Fatty acid profiles of diets and eggs

Multivariate analysis showed that the FA composition was significantly different among the three diets (pseudo‐F_{2, 6} = 42,800, p = 0.001; Table 3). The CAP showed that the data points of the relative percentages of FA profiles from the REF‐D, COM‐D, and FFP‐D treatments were well separated in the multidimensional space. Overall, the plot showed that the three treatments had significantly different FA composition (Figure 1).

Permutational ANOVA showed that overall neutral FA composition (pseudo‐F = 3.1493, p = 0.0138) and polar FA composition (pseudo‐F = 1.5006, p = 0.2269) for the FFP‐D treatment replicates were similar. However, neutral FA composition of the eggs from each diet treatment was significantly different overall (pseudo‐F_{2, 6} = 46.39, p = 0.001), with pairwise comparisons showing significant differences in relative percentages of SFAs (F = 10.58, p = 5.07 × 10⁻⁴), MUFAs (χ² = 16.29, p = 2.9 × 10⁻⁴), PUFAs (F = 38.643, p = 3.13 × 10⁻⁸), LNA (χ² = 22.619, p = 1.23 × 10⁻⁵), and LA (χ² = 19.90, p = 4.78 × 10⁻⁵; Table 5). The CAP plot showed three distinct clusters based on diet treatment (Figure 2). Eggs from the REF‐D treatment were significantly correlated with high total SFA and MUFA percentages. The REF‐D eggs had significantly higher LA, LNA, total SFA, and total MUFA percentages but lower ARA, eicosapentaenoic acid (EPA), and total PUFA percentages when compared to spawns from fish that were fed FFP‐D. Spawns from the COM‐D treatment were significantly correlated with high docosahexaenoic acid (DHA) percentages than the eggs spawned from broodstock that were fed the two other diets. The COM‐D eggs had similar total SFA, MUFA, and PUFA percentages but lower ARA and EPA percentages when compared to spawns from fish that were fed the FFP‐D. Eggs from the FFP‐D treatment were significantly correlated with high ARA and EPA percentages.

Polar FA composition of the eggs was also significantly different among dietary treatments (pseudo‐F_{2, 6} = 16.12, p = 0.001). In terms of polar FAs, or FAs in the cell membranes, the clustering was less obvious than that for the neutral FA fraction, but three distinct groups with a proximity between the clusters were observed, highlighting the more conservative nature of this portion of lipids. All groups were situated within the space, where relative percentages of the three EFAs did not vary much within their respective dietary treatment groups (Figure 2). Pairwise comparison showed that the three EFA relative percentages in all three diets were significantly different from each other (Table 6). Apart from this, the polar FA results were comparable to the results for neutral FAs.

Fatty acid ratios

The nutritional requirements for LA and LNA were satisfied in the polar lipid fraction (ratio < 1). The same FAs were highly retained in the neutral lipids (ratio > 1) for all diets, suggesting potential dietary deficiencies for these FAs (Figure 3). Requirements for two EFAs (EPA and ARA) were fulfilled for the REF‐D and COM‐D, with a lower percentage of incorporation found in the polar lipids, as indicated by both neutral and polar ED ratios (ED < 1). However, for the FFP‐D treatment, these EFAs were highly retained in both lipid fractions, which suggests a potential deficiency in the diet. We observed that DHA was highly retained in both neutral and polar lipid fractions for all diets except in the neutral lipids for the FFP‐D. There was a transfer of DHA from neutral to polar lipids, indicating a potential deficiency in this FA for all diets.

DISCUSSION

Each broodstock group was reared under similar conditions using standardized protocols to prevent variation in spawning as much as possible. However, replication in broodstock diet studies is difficult to achieve because the reproductive biology of each fish can vary depending on internal and external factors. Therefore, broodstock diet research for large marine fishes is scarce because it is expensive, time consuming, and challenging to conduct. We treated each spawning event as a replicate in our FA analyses, as CYT are multiple continuous spawners that can quickly incorporate what they consume into egg production. Previous studies with Gilthead Bream Sparus auratus and Almaco Jack Seriola rivoliana utilizing spawns as replicates have also shown a significant reflection of dietary lipids in their corresponding spawns (Almansa et al. 1999; Roo et al. 2015). When we compared our FFP‐D tank replicates, we found that neutral and polar egg FA compositions were statistically similar. Although the REF‐D and COM‐D treatments could not be tested for potential interfemale variability due to a lack of replication, the results from our FFP‐D replicates suggest that there may not be significant variation in terms of biochemical composition between broodfish.

Although it had no tank replication, we have included results from the REF‐D (BioVita Starter; Bio‐Oregon) pilot trial in 2017 because it represents an essential starting reference point for producers in exploring other commercially available broodstock diets. The REF‐D was not originally formulated as a broodstock diet, but it has previously been used as such in CYT studies due to its market accessibility and nutritional composition (Stuart et al. 2018; Salze et al. 2019). The COM‐D (Breed‐M; INVE) has been used both as a successful diet replacement and as a supplement in other broodstock marine finfish like the Atlantic Halibut Hippoglossus hippoglossus, Atlantic Cod Gadus morhua, and Yellowtail Kingfish (Salze 2005; Brown 2009; Fielder et al. 2020).

Values of egg quality metrics in previous CYT studies were comparable to what was achieved in this study apart from the poor hatch rate and SFF, especially in the COM‐D treatment. The average hatch rate and SFF in other CYT studies ranged from 60% to 75% and from 63% to 68%, respectively, whereas our study achieved ranges of 6–60% for hatch rate and 0–49% for SFF (Stuart and Drawbridge 2013; Stuart et al. 2018, 2020).

The REF‐D treatment produced spawn quality comparable to that achieved with our FFP‐D treatment and had significantly better viability (mean ± SD = 89.57 ± 6.18%) and batch fecundity (57,160 ± 16,840 eggs/kg of female). However, previous studies using the REF‐D showed the opposite and achieved poor viability (33.7 ± 39.6%; Stuart et al. 2018). This difference may be explained by potential interfemale variability, as viability was significantly different among our replicate FFP‐D tanks. The REF‐D spawns in 2017 were especially comparable to COM‐D spawns in 2018 because they were produced from the same brood female and thus were not subjected to interfemale variability. However, data from the REF‐D treatment may be affected by other external factors, like ambient light and temperature and broodstock age, because the pilot trial was conducted 1 year before the other diet trials. There were significant differences between the two commercial diets, with the COM‐D achieving significantly larger egg and oil diameters. Previous studies of Turbot Scophthalmus maximus found that a high content of highly unsaturated fatty acids (HUFAs) in broodstock diets was correlated with larger egg and oil diameters, as was evident in this study (Lavens et al. 1999). The COM‐D and FFP‐D diets and eggs had higher HUFA levels and produced significantly larger egg and oil diameters. Typically, a large egg diameter would be a reliable predictor and indication of a good‐quality egg, but the COM‐D spawns had extremely poor hatch rates (mean ± SD = 6.82 ± 10.6%) and SFF (0.64 ± 4.90%; Nazari et al. 2009; Stuart et al. 2020). In the previous year, the same brood female was able to produce spawns yielding a typical average egg hatching rate (60.10 ± 23.3%) and SFF (49.3 ± 28.3%) range while being fed the REF‐D treatment. Although the eggs were produced by the same female, they could have been fertilized by a different male, as there were two males in the tank. Stuart et al. (2020) recorded egg quality from a population of CYT broodstock over a period of 3 years and recorded spawns with much lower annual variation in egg metrics as seen in this study, which suggests that the large decrease in egg hatching rate and SFF during the COM‐D treatment was mostly likely caused by a nutritional deficiency.

However, all diets had distinct lipid profiles that were largely reflected in the lipid profiles of the eggs, especially in total PUFAs, which further emphasized the correlation between broodstock diet and egg FA composition (Figures 1 and 2). This correlation was more apparent in the neutral FA profiles than in the polar FA profiles, which was consistent with broodstock nutrition studies of the Gilthead Bream, Olive Flounder Paralichthys olivaceus, and White Bass Morone chrysops (Almansa et al. 1999; Furuita et al. 2002; Lane and Kohler 2006). This is because dietary lipids are reflected more in neutral lipids, which are an important energy reserve to fuel early development activities like hatching (Vetter et al. 1983; Wiegand 1996; Furuita et al. 2002). One of the most relevant dietary lipids in broodstock diets are the n‐3 HUFAs, which have been positively correlated with an improvement in spawn characteristics, including fecundity, hatch rate, and SFF, in Gilthead Bream and Cobia Rachycentron canadum (Nguyen et al. 2010; Fernández‐Palacios et al. 2015). The importance of n‐3 HUFAs was reflected partially in the FFP‐D, which had the most balanced ratio of EPA : DHA relative percentages in the neutral lipid and yielded desirable values for spawn variables. However, the ARA, EPA, and DHA percentages of the FFP‐D spawns were very different than those reported by Stuart et al. (2020), which suggests an inconsistent FA composition when using the FFP‐D (e.g., seasonal variations in FFP quality).

The FA ratios of the commercial diets (REF‐D and COM‐D) met qualitative requirements for ARA and EPA in both polar and neutral FA fractions but were insufficient in DHA for both. The FFP‐D only met the DHA requirement in the neutral FA fraction and was insufficient in ARA and EPA for both the neutral and polar lipid fractions. The FA ratio can be used to provide qualitative insight into the nutritional value of a diet by using an ED ratio, where a specific FA in the egg is compared to the same FA in the diet (Mejri et al. 2021). The ratios of LA and LNA in polar fractions of the eggs were less than 1.0, which resulted in a low retention in the cell membranes; however, both the LA and LNA ratios were higher than 1.0 in the neutral lipid fraction, suggesting a potential deficiency of those FAs in the diet. Linoleic acid and LNA may not be necessary for carnivorous marine fish like the CYT because they lack the Δ5 desaturase enzyme needed to convert these FAs into HUFAs. Other studies with juvenile Cobia have shown that dietary LNA and LA can promote the expression of this enzyme (Xu et al. 2018). Currently, there are no studies regarding LNA and LA supplements in broodstock diets for Seriola, but the potential benefits of EFA supplementation would reduce the need for dietary HUFAs. Arachidonic acid and EPA can affect reproductive performance of broodstock fish and have been reported to improve egg quality in cultured CYT (Rodríguez‐Barreto et al. 2012; Stuart et al. 2018).

In terms of ARA and EPA levels, the commercial diets tested here were potential broodstock diet candidates, but supplementation with DHA, LA, and LNA should be considered. Improvements to spawn quality in CYT can be achieved through the added supplementation of HUFAs like DHA to a pelleted diet, as reported in studies of the European Bass Dicentrarchus labrax, or through an occasional supplement of cut bait (Bruce et al. 1999).

CONCLUSION

Spawns from the three CYT broodstock diet treatments yielded varying physiological and biochemical results. Although the COM‐D treatment produced egg quality variables (egg and oil diameters, viability, and average batch fecundity) comparable to those generated by the FFP‐D, only one tank successfully spawned, and it yielded very poor egg hatching rates and SFF. These data suggest that the COM‐D is unsuitable as a complete replacement for the FFP‐D and should be studied further as a potential supplemental source of ARA and EPA in a broader formulation mixture.

Spawns from the CYT broodstock that received the REF‐D treatment had significantly lower egg and oil diameters but higher viability and average batch fecundity than spawns from broodstock that were fed the FFP‐D. The REF‐D was also able to satisfy ARA and EPA levels—but not DHA—in both neutral and polar fractions. Although the REF‐D has promising potential, further studies with tank replications and added EFA supplementation are needed to determine whether it can be a replacement for the FFP‐D. This pilot study was especially valuable because it showed that a single female broodfish can spawn volitionally with only two other males in a relatively small‐volume tank, which will allow for future broodstock diet studies to be feasibly conducted with more replicates.

Our results show EFA deficiencies in the REF‐D, COM‐D, and FFP‐D, and we recommend supplementation of specific EFAs to further improve egg quality and biochemical composition. This study offers the first look into potential commercial replacements for the FFP‐D as a diet for CYT broodstock, which is particularly relevant for commercial producers. It provides a foundation for future CYT broodstock studies to explore new or commercially available diets.

ACKNOWLEDGMENTS

The broodstock diet and spawning trials conducted at HSWRI were partially funded by the National Oceanic and Atmospheric Administration's Saltonstall–Kennedy Program (Award NA15NMFS4270304), with additional support from SeaWorld San Diego and the Chevron Corporation. Biochemical analyses of eggs and associated data analyses and reporting were supported by the U.S. Department of Agriculture (USDA) Agricultural Research Service via Cooperative Agreement 59‐6034‐9‐007 with Florida Atlantic University's Harbor Branch Oceanographic Institute (HBOI). We are grateful to Victoria Uribe for her contributions toward sample processing of eggs in the laboratory at HBOI, and we appreciate the husbandry staff at HSWRI for spawning, egg samples, and egg quality data. Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the USDA. The USDA is an equal opportunity provider and employer.

CONFLICT OF INTEREST STATEMENT

The authors declare that the study was conducted without any existing commercial or financial affiliations that could be interpreted as a potential conflict of interest.

ETHICS STATEMENT

This work complied with the Guide for the Care and Use of Laboratory Animals (National Research Council 2010), Policy on Humane Care and Use of Laboratory Animals (U.S. Public Health Service 1996), and Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources 1986). The Institutional Animal Care and Use Committee (IACUC) protocol used for this study was A18‐39.

DATA AVAILABILITY STATEMENT

The data supporting the findings of this study can be obtained from the primary author, Li Sun Chin (lchin2021@fau.edu), upon request.

REFERENCES

© 2024 American Fisheries Society (AFS)

Issue Section:

FEATURE ARTICLE

Email alerts

• Google Scholar

Citing articles via

More from Oxford Academic