Introduction
A growing demand for seafood and the ensuing environmental effects are pushing the need for innovation in aquaculture and a critical examination of technological advancement in several key areas, such as novel feed composition and feed technologies. The increasing need for aquafeeds and limited fish meal supply has led to alternative ingredients in feed formulations. This may result in higher carbohydrate content in diets, affecting fish digestion and environmental waste production. As known, soy protein is the major source of protein as fishmeal replacement in fish feed because of its worldwide availability and low price. However, to our knowledge there is a lack of information regarding the use of X-ray microtomography to evaluate change in microstructure of feed blends added different concentration of soybean protein concentrate (SPC). Nonetheless, significant knowledge gaps persist regarding the long-term physiological and nutritional effects of these protein sources in carnivorous species maintained in RAS. The aims of the study, therefore, were to investigate the pellet's characteristics (bulk density, expansion ratio, diameter and pellet's porosity) and their usage properties (oil leakage rate, durability, hardness and floatability) and to assess growth performance, digestibility, gut microbiome profile of European sea bass fed with experimental diets containing increasing levels of SPC over 16 weeks long-term feeding.
Material and Methods
Juvenile European seabass were obtained from a local hatchery and transferred to a marine RAS until reaching an initial mean weight of 32 g. Fish were stocked in a fully controlled RAS with 15 tanks (250 L capacity; 30 fish per tank). Five iso-nitrogenous (48% CP) and iso-lipidic (18% CL) extruded diets were formulated, including a control diet with low fishmeal content. Soybean Protein Concentrate replaced fishmeal at 25% (SPC25), 50% (SPC50), 75% (SPC75), and 100% (SPC100) levels. Fish were hand-fed twice daily to apparent satiety during acclimation and throughout the 16-week experimental period. Micro-CT analysis was performed using a Skyscan 1275 X-ray microtomograph. Faeces were collected via passive settlement systems attached to each tank outlet during the final 30 days of the trial. Total RNA was extracted from intestinal samples using TRIzol reagent according to the manufacturer's protocol. RNA quality and quantity were assessed using spectrophotometric and fluorometric methods. Raw paired-end FASTQ files were quality-checked using FastQC (v0.12.1) and MultiQC (v1.21). Taxonomic profiling was conducted using a Na��ve Bayes classifier in QIIME2 trained on the SILVA 138 SSU rRNA database (99% identity, 338F/515R region). Diversity analyses were performed using QIIME2 core-metrics and the phyloseq package (v1.44) in R.
Results
Overall, increasing SPC inclusion leads to a progressive transition from a compact and uniform structure to a more porous and heterogeneous matrix, particularly for SPC75, and SPC100. This trend suggests that higher SPC levels negatively affect pellet densification and cohesion, likely influencing key functional properties such as mechanical strength, water stability, and hydration behavior. Percent porosity (total, open, closed), mean cell size (MCS) and cell size and cell wall thickness will be presented in detail. No significant differences in growth, feed intake, and feed conversion ratio were observed among the experimental diets, further indicating that high levels of SPC inclusion had a negligible effect on overall growth performance. Collectively, while dry matter digestibility remained unaffected, increasing dietary inclusion of SPC significantly reduced both protein and lipid digestibility. The reduction was more pronounced at higher inclusion levels, particularly in the SPC75 and SPC100 groups, suggesting that elevated SPC incorporation may negatively impact nutrient availability and digestive efficiency in fish. With respect to microbiome analysis, mean Shannon diversity ranged from 1.85 (SPC0) to 2.26 (SPC75), with no significant differences among groups. Samples generally clustered without clear dietary separation; however, outliers were observed, including samples 6323 and 6324 (SPC50) along the negative PC1 axis and sample 6329 (SPC100) along the positive PC1 axis. The gut microbiota was dominated across all dietary groups by Firmicutes, Proteobacteria, Actinobacteriota, and Bacteroidota. At the genus level, the most abundant taxa were Lactobacillus, Ruminococcaceae (unclassified), Pseudomonas, and Photobacterium. Genus-level abundances showed high inter-individual variability, consistent with elevated intra-group beta diversity dispersion.
Discussion
The present study showed that, although no significant differences in growth performance were observed, pellets with high Soybean Protein Concentrate inclusion exhibited a clear transition from a dense, well-integrated matrix to a highly porous and structurally heterogeneous structure, indicating reduced pellet integrity at higher inclusion levels. The host genotype appears to exert a stronger influence on gut microbiota composition than diet, which may partly explain the observed outliers, potentially reflecting genetic variability or differential sensitivity to SPC anti-nutritional factors. Individual differences in intestinal transit time, mucus composition, and innate immune status may also contribute to these microbiome deviations under high SPC dietary stress. The numerical increase in ASV richness in high-SPC groups (SPC75, SPC100) may be associated with greater availability of plant-derived substrates, such as complex polysaccharides and secondary compounds, potentially supporting a wider range of microbial taxa capable of metabolizing these compounds in the intestine of European seabass. Further investigation is required to clarify the functional implications of these observations.
Acknowledgement
The authors would like to thank U.S. Soybean Export Council for their support to this study.