Dry matter, protein, and energy digestibility of diets for juvenile Pacific white leg shrimps ( Litopenaeus vannamei ) reared at different... Ciência Rural, Dry matter, protein, and energy digestibility of diets for juvenile Pacific white leg shrimps ( Litopenaeus vannamei) reared at different salinity levels

: This study evaluated the effect of low, medium, and high-water salinity (5, 35, and 50 ppt) on the apparent dry matter, protein, and energy digestibility of two formulated and six commercial diets for juvenile whiteleg shrimp, Litopenaeus vannamei, in a 120-day trial. Digestibility was determined in vivo using chromic oxide as an inert diet marker. Hydrostability in pellets varied from 86.8% to 99.9%; dry matter digestibility varied from 49.1% to 64.1%; protein digestibility showed greater variations at all salinities (56.9%–85.8%); and energy digestibility ranged from 70.1 to 86.4%. Salinity had a significant effect on dry matter, protein, and energy digestibility. Using a principal component analysis (PCA) with a covariance matrix, our findings suggested that the E2 (fishmeal-based formulation) diet and 35 ppt salinity provided optimum hydrostability and digestibility to Pacific white leg shrimp juveniles.


INTRODUCTION
The Pacific white leg shrimp Litopenaeus vannamei is one of the most commonly farmed species at commercial scale because of its fast growth, high survival at high density, and resistance to disease (MARTÍNEZ-CÓRDOVA et al., 2009;ANAYA-ROSAS et al., 2017;ABDELRAHMAN et al., 2019). Their osmoregulatory capacity allows them to inhabit waters with salinity ranging from 0.5 to 60 ppt (ROY et al., 2007;JAIME-CEBALLOS et al., 2008;CHEN et al., 2019).
Given the recent increase in market demand, shrimp farming production has shifted from traditional extensive systems toward modern intensive and super-intensive systems. In the latter, environmental regulations require producers to use environmentally-friendly feed sources with low nitrogen and phosphate output (OLAUSSEN, 2018), and reducing waste load to environment (CHATVIJITKUL et al., 2017).
Studies of feedstuff digestibility in commercially-farmed aquatic organisms have gained increasing interest in recent years (CAMPAÑA-TORRES et al., 2006;HOSSEIN et al., 2017). Data on nutrient digestibility is particularly important to determine suitable and accurate diet formulations for white leg shrimp farming (YANG et al., 2009;HOSSEIN et al., 2017).
Salinity and temperature are two of the most important environmental factors controlling shrimp growth and survival, as they directly impact their physiological response (BÜCKLE, 2006;KIR et al., 2008;ABDELRAHMAN et al., 2019). If these parameters are controlled within their optimal range, it results in higher farm productivity (PÉREZ-VELÁZQUEZ et al., 2007;CHEN et al., 2019). The worldwide expansion of low salinity shrimp rearing, driven by increasing demand for this product, requires specific feed formulation for this environment. When shrimp are exposed to low salinity, they need to counteract the passive loss of Na + and Clthrough the active uptake of Na + from water in exchange for H + , which occurs in the apical membrane of osmoregulatory cells (PALACIOS et al., 2004;BÜCKLE, 2006;HURTADO et al. 2006). In arid and semi-arid zones, where high pond water evaporation is common, salinity may increase to 50 ppt or higher, especially by the end of the growing season (PÉREZ-VELÁZQUEZ et al., 2007). Thus environmental and management conditions interfere with the shrimp cultivation system, growth and production performance.
Therefore, this study determined the effect of low, medium, and high salinity on the apparent digestibility of dry matter, protein, and energy of two formulated and six commercial diets for juvenile Pacific white leg shrimp (L. vannamei).

Diet formulation and preparation
Experimental diets (E1 and E2) were formulated (Table 1) with different levels of ingredient inclusion using the MIXIT-WIN software (Agricultural Software Consultants, San Diego, CA, USA). All ingredients were pulverized and sieved through a 250-μm mesh. Both experimental diets were prepared by mixing macro-ingredients in a blender until a uniform mixture was obtained. The micro-ingredients (vitamin and mineral premixes, sodium alginate, chromic acid, and antioxidant BHT) were mixed in a plastic container, and then added to the macro-ingredients. Fish oil and soy lecithin were homogenized into an emulsion and added to the mixture. Distilled hot water was added to this mixture (~30% of dry weight of the ingredients), forming a dough, which was then passed through a meat grinder (Torrey MR , Monterrey, Nuevo Leon, MX) to form 2-mm diameter pellets. Then, the pellets were dried in an air flux oven at 45 °C for 8 h. Finally, the pellets were packed in plastic bags and stored at -2 °C until further use (MÉNDEZ-MARTÍNEZ et al., 2017;. Diet E2 was similar to the commercial fishmeal-based diet formulated according to AKIYAMA et al. (1991) (Table 1), which was first pelletized at an industrial shrimp feed factory (CRUZ-SUÁREZ et al., 2009); then, it was ground to obtain a maximum particle size of 500 µm. Subsequently, it was mixed with 1% chromic oxide as an inert marker and 1% sodium alginate (A-7128, Sigma, St. Louis, MO, USA) as a high viscosity binder. After that, distilled hot water (~30% of dry weight of the diets) was added to the mixture. The resulting dough was passed through a meat mill (Torrey MR , Monterrey, Nuevo Leon, MX) to form 2-mm diameter holes.

Experimental design
Juvenile white leg shrimp (3.6 ± 0.3 g) were obtained from a local hatchery and transported in 100-L containers to the laboratory of Centro de Investigaciones Biológicas del Noroeste (La Paz, B.C.S., MX). They were acclimated to a salinity of 35 ppt in three 1500-L fiberglass tanks for seven days and fed with commercial feed with 35% crude protein (PIASA TM , La Paz, BCS, MX). Molts, dead shrimp, and unconsumed feed were removed daily.
A completely randomized 8 × 3 (diet × salinity) factorial experimental design was used, with four replicates per treatment. Treatments consisted of six commercial diets for shrimp (D1, D2, D3, D4, D5, and D6), and the two formulated diets described above (E1 and E2) and tested at each salinity level (5, 35, and 50 ppt). The feeding rate during the culture period was set at 5% total shrimp biomass in each experimental tank. Shrimp were fed twice daily (09:00 and 17:00 h). Unconsumed feed, exuviates, overnight feces, and dead shrimp were removed daily (08:00 h) by siphoning with a plastic hose (0.5 cm in diameter). Feces were collected twice daily (10:30 and 18:30 h), by gently siphoning fecal strands with a Pasteur pipette. Feces were gently rinsed with distilled water, transferred to 30-mL conic tubes, and frozen at −20 °C. Pooled samples of frozen fecal matter from each day and tank were freeze-dried, ground, thoroughly mixed, and kept frozen at −80 °C until further analysis.
The trial was conducted in an experimental unit consisting of 60-L rectangular indoor plastic tanks (58 × 48 × 25 cm), filled with filtered (5-μm), and UV sterilized marine water, under constant aeration. Each tank was stocked with 10 juvenile shrimp (mean: 4.0 ± 0.5 g). To reach a salinity of 5 ppt, fresh water was gradually added to seawater, reducing salinity at a rate of ~5 ppt per day, as recommended by JAIME-CEBALLOS et al. (2008). Seawater was used for the second set of trials, adjusting salinity as needed. For high salinity trials (50 ppt), a 1100-L tank was filled with seawater and iodide-free salt was added until the desired salinity was obtained. Salinity was recorded using a refractometer (Model RF20, Extech Instruments, Waltham, MA, USA) in 15-20 min intervals until 50 ppt was reached.
Water temperature and O 2 concentration were measured daily using a multi-parameter probe (Model 55, YSI, Yellow Springs, OH, USA), while nitrites and nitrates were recorded once a week using a portable spectrophotometer (Hach DREL 2800, Loveland, CO, USA). Alkalinity was determined volumetrically with phenolphthalein and bromocresol, using sulfuric acid in a 25-mL digital burette with precision of ±30 μL. Total phosphorus, orthophosphates and ammonium ion content were measured according to MURPHY & RILEY (1962) & SOLÓRZANO (1969, with a spectrophotometer (Hach DR/2000, Loveland, CO, USA). The photoperiod was kept to a 12:12 h light:dark cycle.

Diet hydrostability analysis
Diet hydrostability was measured by determining the amount of retained dry matter after water immersion according to OBALDO et al. (2002) and GUCIC et al. (2013). Two grams of feed was placed in 250-mL Erlenmeyer flasks with 200 mL water at 5, 35, and 50 ppt salinity for one hour. During this time, the flasks were gently swirled on a Table 1 -Formulation (g/100 g of dry matter) of the experimental diets (E1 and E2) used to measure in vivo digestibility of dry matter, protein, and energy in juvenile Litopenaeus vannamei shrimp.
Ingredients platform shaker at 100 rpm at room temperature (~27 °C) to disperse and submerge feed pellets.

Chemical analysis of ingredients and diets
Diet samples (Table 2) were finely ground and sieved, then analyzed in triplicates. The following variables were analyzed according to the corresponding method (AOAC, 2005): dry matter (weight difference, dried in an oven at 105 ºC for 24 h); crude protein (%N × 6.25) using the Kjeldahl nitrogen method (Foss, Hillerød, DK); crude lipid using the ether-extraction method (Soxtec Avanti, Höganäs, SWE); crude fiber using the method of Weende and van Soest (Fibertec, Foss, Hillerød, DK); ash and nitrogen-free extract with standard methods (AOAC, 2005); and gross energy using an adiabatic calorimeter (Parr Instrument, Moline, IL, USA).
Protein was determined using the micro Kjeldahl method (Tecator, Höganäs, SWE). The concentration of chromic oxide (Cr 2 O 3 ) was determined by digesting organic matter with nitric acid and oxidizing Cr 2 O 3 to Cr 2 O 7 with perchloric acid, followed by a colorimetric analysis of the dichromate ion with diphenylcarbazide (FURUKAWA & TSUKAHARA, 1966).

Statistical analysis
Kolmogorov-Smirnov and Bartlett tests were applied to the data to determine normality and homogeneity of variance, respectively. The statistical significance level was set at P < 0.05. A two-way analysis of variance (ANOVA) was used to determine significant difference between groups (3 salinities; 8 diets). In the presence of significant interactions, one-way analysis of variance (ANOVA) was used. Tukey's multiple-range test was used to identify significant differences among digestibility coefficients. A principal component analysis (PCA) using a covariance matrix was performed to explore differences between treatments as a function of diet and salinity interaction (ZAR, 1999). Data were analyzed using the software Statistica 11.0 (StatSoft, Tulsa, OK, USA). Data expressed in percentages were transformed using the square root of the arcsine before statistical analyses.
The diets E2 and D4 used in shrimp feed were significantly better (P <0.05) in coefficient of digestibility of dry matter. With respect to protein digestibility the highest values (P <0.05) were observed for the diets E2 and D4. Energy digestibility was significantly (P <0.05) affected by the different diets. They were significantly higher in dietary treatments D2, D3, D4, D5, and E2.
The ANOVA analysis demonstrated that the salinity factor had an important effect on stability in water and apparent digestibility of diets, which were significantly better (P <0.05) at a salinity of 35 ppt for energy digestibility, 35 and 50 ppt for hydrostability, 35 and 5 ppt for digestibility of dry matter and protein.
In this investigation, the interaction of diet and salinity factors showed a total of 24 interactions (independent variables). ANOVA was applied, and significant differences (P <0.05) were found in response variables of hydrostability in pellets, digestibility of dry matter, proteins and energy on diet. Hydrostability in pellets varied from 86.8 to 99.9%; dry matter digestibility varied from 49.1% to 64.1%; protein digestibility showed greater variations at all salinities (56.9%-85.8%); and energy digestibility ranged from 70.1 to 86.4%. However, the Principal Component Analysis (PCA) was applied and significant differences (P <0.05) were reported between the 24 interactions.
The PCA using a covariance matrix to explore differences in the response of hydrostability in pellets, digestibility of dry matter, proteins and energy on diet from juvenile shrimps to diet-salinity interaction is shown in figure 1. The plot PC1 vs. PC2 (using their multipliers as a variable) showed a clear pattern. PC1 (48.3%) and PC2 (24.9%) jointly accounted for 73.2% of the differences in hydrostability, digestibility of dry matter, protein, and energy responses. For PC1, the main differences were observed in dry matter, protein, and energy. For PC2, the main difference was the hydrostability.
The diet -salinity ratio plot suggested that the formulated diet E2 at 35 ppt (Interaction -J) was optimal as shrimp feed, given its position at the top of the plot to the right (highly positive for PC1 and PC2). Conversely, the ratio for E1 and salinity at 5 ppt was collected (Interaction -A) around the zero point at PC1 but still positive in the upper part of the plot to the right (highly positive for both PC1 and PC2). This result was not the case for the other diets; for instance, the diet -salinity ratio of D6 at 50 ppt (Interaction -X) was more negative in PC1 than PC2.

DISCUSSION
All tanks were maintained according to recommended aquaculture practices (BUCKLE et al., 2006;CRUZ-SUÁREZ et al., 2009;ABDELRAHMAN et al., 2019;ALVES et al., 2019;CHEN et al., 2019), ensuring that environmental conditions did not influence our results. Good acceptance for all commercial and formulated diets was observed throughout the trial.
The findings in this study indicated that salinity level has a significant effect on shrimp's digestibility of dry matter, protein, and energy, concurring with previous studies (ROSAS et al., 2001;OBALDO et al. 2002, CRUZ-SUÁREZ et al., 2009. Previous research has also shown that salinity significantly influenced the metabolism of penaeid shrimps (ROSAS et al., 2001, ZHU et al., 2006. As salinity deviates from the iso-osmotic point, the osmotic balance in shrimp is modified and the organism expends more energy on osmoregulation Table 4 -Hydrostability of dry matter, apparent digestibility coefficients of dry matter, protein, and energy for two experimental (E1 and E2) and six commercial (D1, D2, D3, D4, D5 and D6) diet formulations for juvenile Litopenaeus vannamei shrimp cultivated at three different salinity levels (5, 35, and 50 ppt).  98.   and less on growth (HURTADO et al., 2006). Shrimp are excellent osmo-regulators if ionic ratios in water are adequate (GONG et al., 2004, ROY et al., 2007. Osmoregulation is determined by the interaction between physiological requirements for water nutrients and mineral profile (DAVIS et al., 2002).
Our results also showed that salinity had a significant effect on DMR: leaching was significantly higher at low salinity levels. At the highest salinity level (50 and 35 ppt), ionic concentrations (Cl -, HCO 3 -, and SO 4 2-) were higher than in lower salinity treatments, which delayed the rapid leaching of feed, thereby influencing the digestibility coefficient of dry matter. The observed values in this study were similar to findings by CRUZ-SUAREZ et al. (2009) at 35 ppt salinity, but lower than those reported by XIA et al. (2010) at equal salinity. The high DMR in all diets indicated that the ingredients used for both formulation and elaboration methods were adequate.
Protein is the most critical ingredient in shrimp feed, both in terms of cost and organism's growth response. Proteins are crucial from an ecological point of view, as they are a major source of nitrogen in aquaculture ponds, which may cause hyper-nutrification. The protein APD coefficients showed differences between commercial and experimental feed: significantly higher APDs were observed at 35 ppt than those described by CRUZ-SUÁREZ et al. (2009). Apparently, the optimal salinity for growth is associated with protein metabolism, as free amino acids are involved in the regulation and maintenance of cell volume.
Energy AED coefficients showed significant differences between experimental and commercial diets within each salinity level. The highest obtained value was for diet D4 (86.4%) at 35 ppt interaction; although, it did not differ significantly (P > 0.05) from diet -salinity ratio E2: 35 ppt, D5: 35 ppt, E1: 50 ppt and D2: 50 ppt, which was related to the shrimp's energy expenditure at higher salinity levels.
For all diets, the best hydrostability values were observed at higher salinities; however, the E1 diet was the most hydrostable out of all the tested diets. That being said, high hydrostability does not necessarily indicate high diet performance -on the   (2017) reported that the ADC of dry matter and protein ranged from 58.74 to 74.31% and 62.43 to 80.25%, respectively, in soybean diets for L. vannamei. GUCIC et al. (2013) evaluated the apparent digestibility of carbohydrates and lipids in white shrimp feeds at different salinities (5, 35, and 50 ppt) and reported better stability (>86%) for diets at 5 ppt, with the lowest lipid digestibility at 50 ppt (74%) while carbohydrate digestibility was greater at 5 ppt (90%). They suggested that the high hydrostability observed at low salinity levels did not necessarily imply a better production response but could induce a lower digestibility.
In this study, diets with greater hydrostability showed a lower protein and dry material digestibility; although, this result could also be explained by the chemical composition of the diet, as well as the digestive characteristics of the species and environmental conditions in which the study was conducted.
In this study, the interaction between salinity and diet was observed to have a significant effect on digestibility of dry matter, protein, and energy in shrimp, suggesting the possibility of manipulating dry matter, protein, and energy in shrimp feeds at different culture salinities to achieve a desired response of farmed shrimp. However, further research is needed to analyze the digestive enzyme activity and hepatopancreas histology of reared shrimp, enabling a better understanding of the processes and mechanisms involved. Furthermore, researchers should also investigate the influence of different management practices on shrimp aquaculture production (both in terms of output quantity and quality).

CONCLUSION
The results of this study allowed us to conclude that diet formulation and water salinity has a significant influence on hydrostability and digestibility of shrimp feed. The PCA suggested that a diet -salinity ratio between formulated diet E2 and salinity at 35 ppm promotes hydrostability, digestibility and proper nutrient use in white shrimp juveniles, which could be set as recommendable for further studies on feed formulation for the species.