Experimental animals and bioassay conditions
Juvenile Nile tilapia (1.3 ± 0.7 g) were acquired from a commercial hatchery located in Soto la Marina (Tamaulipas, Mexico). Fish were placed in plastic bags containing oxygen saturated water and immediately transported to the Aquatic Production Laboratory of the Veterinary and Zootechnics School (Autonomous University of Nuevo Leon). Fish were allowed to adapt to local conditions for 15 days in glass aquaria kept under a natural photoperiod (12:12 h dark:light). Over the conditioning period, mean water parameters remained as follows: temperature 29.2 ± 0.6°C, pH 8.4 ± 0.2, total ammonia nitrogen 0.09 ± 0.04 mg L− 1, nitrite was maintained below detection limits, and mean values for nitrate were 10.7 ± 2.9 mg L− 1. To establish known, basal nitrogen isotopic values in fish muscle, during the acclimation period animals were supplied with a previously analyzed, commercial tilapia feed (44% crude protein and 15% crude lipid, NutriPec, Purina®).
Experimental diets
The main ingredients employed to formulate the experimental diets were fish meal (Monterey sardine) and Madagascar cockroach meal, which constituted the only sources of dietary protein in the mixed diets. Both main ingredients were analyzed before the experiment to verify their proximal, elemental, and isotopic profiles. Care was taken to ensure that the conditioning diet and the experimental diets had contrasting isotopic values to promote isotopic changes in muscle tissue. The insect meal was obtained from Madagascar cockroaches grown in plastic boxes fitted with coconut fiber and peat moss substrates (Universidad Autónoma de Aguascalientes, Mexico). Insects were maintained at 26°C and fed a mixture of discarded vegetables and water containing 10% cane sugar. Once grown, insects were sacrificed by freezing, dehydrated (55°C / 96 h) and finally ground to obtain a fine powder.
Five experimental diets were formulated to gradually replace the fish meal-derived protein with insect meal. Diets were isonitrogenous (40% crude protein) and isoenergetic (3.7 kcal/g) and were manufactured as follows: diet 1 contained 100% fish meal as protein source (diet 100F), diets 2 to 4 included fish meal substitutions of 25, 50 and 75% cockroach meal (diets 75F/25C, 50F/50C, 25F/75C). Diet 5 was formulated with 100% insect meal as protein source (diet 100C). Diets 100F and 100C were used as nutritional and isotopic controls to eventually estimate the isotopic differences between diets and fish (isotopic discrimination factors, Δ15N) (Table 1).
Table 1
Nutrient formulation (gr ingredient 1000 gr diet− 1) of five experimental diets for juvenile Nile tilapia, which were used to compare the contribution of nutrients supplied by fish meal (F) and Madagascar cockroach meal (C) to fish growth. The proximal and isotopic composition of each diet is presented.
Ingredient | 100F | 75F/25C | 50F/50C | 25F/75C | 100C |
Fish meala | 597.9 | 453.3 | 305.3 | 151.7 | 0.0 |
G. portentosa meal | 0.0 | 177.3 | 354.0 | 520.6 | 696.9 |
Wheat starchb | 286.0 | 259.3 | 220.3 | 180.3 | 158.7 |
Fish oil | 63.1 | 55.2 | 45.0 | 43.8 | 32.1 |
Binder (CMC) c | 20 | 20 | 20 | 20 | 20 |
Cellulosec | 3.2 | 5.1 | 25.3 | 53.9 | 62.3 |
Constant ingredientsd | 30 | 30 | 30 | 30 | 30 |
Total | 1000 | 1000 | 1000 | 1000 | 1000 |
Proximal and isotopic analyses | | | | | |
Crude protein (g kg− 1) | 397 | 406 | 403 | 398 | 401 |
Lipids (g kg− 1) | 123 | 119 | 122 | 124 | 118 |
Gross energy (Kcal g− 1) | 3.7 | 3.6 | 3.7 | 3.6 | 3.7 |
Moisture (%) | 6.2 | 5.6 | 5.9 | 6.4 | 5.8 |
Ash (g kg− 1) | 92 | 85 | 78 | 74 | 65 |
δ15N (‰) | 16.52 | 15.48 | 13.07 | 12.15 | 8.54 |
aAlimentos Costamar (Sonora, Mexico). |
bAlmidones y gluten S.A. (Monterrey, Mexico). |
cSigma-Aldrich (St. Louis, MO, USA). |
dMineral and vitamin mix, 20 g; calcium phosphate, 10 g. |
[Table 1.]
Growth parameters and sampling procedures
A total of 160 juvenile Nile tilapia having a mean weight of 1.01 ± 0.2 g were randomly allocated into five duplicate treatments (16 fish per replicate) consisting in 20 L glass tanks. At the onset of the bioassay, the conditioning diet was replaced by the different experimental diets, which were ad libitum three times a day (9:00, 13:00 and 17:00 h). Uneaten feed and faces were siphoned out on a daily basis and before the first feeding ration. To determinate the effect of consuming diets on growth, all fish from every replicate tank were individually weighed on the initial and final experimental days. Fish were immobilized with an absorbent cotton cloth and transferred to an analytical balance. Production parameters were estimated as weight gain = [(average final weight – average initial weight)/average individual initial weight] X 100, specific growth rate = [loge average final weight – loge average initial weight] / (time) (100) and survival = (final number of fish / initial number of fish) × 100. A second sampling scheme was established to collect muscle tissue for stable isotope analysis. Sampling was fitted to the exponential isotopic change frequently observed in rapidly growing aquatic organisms receiving a diet shift (Winter et al. 2019). To this end, fish were sampled on experimental days 0, 3, 7, 14, 21, and 29. In every sampling point, one fish from every replicate was euthanized in ice/water slurry and immediately dissected to obtain muscle tissue. Muscle samples were dried in a convection oven (60°C/24h). Dried samples of muscle, individual ingredients and experimental diets were finely ground with pestle and mortar and stored in desiccators until pretreatment for stable isotope analyses (SIA).
Elemental and stable isotope analyses
Diet and fish muscle samples of 1 ± 0.2 mg were packed in tin microcapsules that were in turn organized in 96-well plastic microplates. Isotope analysis for nitrogen, at natural abundance levels, were conducted at the Stable Isotope Facility, University of California, (Davis, CA, USA) as described in García-Pérez et al. (2020). Reported results were expressed in delta (δ) notation as per mill (‰) deviations from the isotopic values of the international standard reference (atmospheric nitrogen, Eq. 1). Instrument precision (SD) was 0.06‰ for δ15N values, as indicated by the consistency of internal reference materials (e.g., glutamic acid and chitin).
δ15N = (Rsample/Rstandard – 1) · 1000 (1)
where R = 15N/14N
Nitrogen turnover and half times in muscle tissue
Diet-elicited isotopic changes were registered over time and the isotopic discrimination factors (Δ15N) were estimated at the end of the bioassay. It was considered that Δ15N values = ± 0.5‰, were indicators of an isotopic equilibrium being reached between fish and their respective diets. Isotopic values determined at different times were introduced into an exponential model (Eq. 2, Hesslein et al. 1993) that allows separating the isotopic change caused by growth (k) and metabolic turnover (m).
CSAMPLE= Cn + (Co – Cn)e −(k+m)t (2)
Where CSAMPLE is the isotopic value in fish tissue at time t, Co is the isotope value of fish tissue in equilibrium with the initial diet (conditioning diet in this study), Cn is the isotope value reached when fish are in equilibrium with a new diet (experimental diets in this study). The growth rate constant, k, was obtained by fitting an exponential growth model to observed weight data, k = log (final weight/initial weight)/time(d), while parameter m was estimated using iterative non-linear regression. By using the latter two coefficients, provides an indicator of the period necessary for half of the constituent nitrogen to be replaced in muscle tissue (t50, half times, Eq. 3) (MacAvoy et al. 2005).
t 50 = In2 / m + k (3)
Estimation of nutritional contributions to growth
From the δ15N values measured in fish meal (source 1), insect meal (source 2) and fish muscle tissue (isotopic mixture), a two-source, one-isotope mass-balance mixing model (Eq. 4; Phillips and Gregg 2001) was used to estimate the relative proportions of dietary nitrogen that were incorporated from both main ingredients into muscle tissue. Underlying assumptions required by isotopic mixing models (Gamboa-Delgado 2022) were met or considered for the calculations. Among these, verification of contrasting isotopic values in the sources, estimation of elemental content in ingredients (N) and verification of Δ15N values. The latter values were obtained from the isotopic differences observed between the two control diets and the respective control fish (muscle). Instead of using literature values, the use of observed Δ15N values as correction coefficients tends to increase the precision of results obtained from the isotopic mixing models (Phillips 2012). δ15N values, sample number and standard errors were introduced into the model to estimate the assimilation proportions and their 95% confidence intervals.
f1= (δ15Nfish muscle- δ15N source2)/( δ15N source1- δ15N source2) and f2 = 1-f1 (4)
Nitrogen contents in insect meal and fish meal were slightly, but statistically different; therefore, in order to calculate the total amount of dry matter contributed by each feeding source, the following equation (Fry 2006; Eq. 5) was used:
ftotal1 = f1·W2/(f1·W2 + f2·W1) and ftotal2 = 1-ftotal1 (5)
where ftotal1= is the total percent contribution of source 1 in a two-source mixing model, W1 and W2 represent the nitrogen content in each of the two sources (main experimental ingredients).
Statistical Analysis
After data homoscedasticity and normality were verified, student's t tests for independent samples were applied to verify significant differences between the isotopic values of experimental ingredients and the conditioning diet. The mean values of the production parameters obtained from every treatment, were compared by one-way ANOVA followed by pair-wise comparisons (Tukey tests). The expected proportions of dietary nitrogen (i.e., established in the formulation of the different diets) and the observed, respective proportions of nutrients assimilated in tilapia muscle tissue were compared by means of Chi-square goodness of fit tests (X2). All statistical tests were conducted using SPSS 17.0 software (SPSS Inc.) at a significance level of P < 0.05.