Partial Replacement of Soybean Protein (30%) with Nannochloropsis oceanica in Broiler Diets: Effects on Growth Performance and Meat Quality

Abstract

The use of human-edible materials like soy in animal feed raises several concerns, as it contributes to high greenhouse gas emissions and requires significant land and water use for agriculture. For this reason, research is exploring alternative ingredients rich in proteins like microalgae, which offer potential nutritional and environmental benefits. Species like Nannochloropsis are promising since their use for human consumption is very limited, making them non-competitive with human food. This article aims to formulate a poultry feed in which 30% of the crude protein from soybean meal is replaced by Nannochloropsis oceanica single-cell ingredients. Growth parameters have been evaluated in comparison with a diet based on soy protein. Additionally, the effect on meat quality was assessed by evaluating nutritional, texture, stability, and sensory parameters. Results showed that the microalgae diet caused a slight reduction in animal growth due to lower digestibility of the feed. Considering the quality parameters of the meat, no differences were found in terms of shelf life and physicochemical parameters, except for the color. The microalgae diet significantly increased the content of n-3 fatty acids and carotenoids in the meat. Finally, regarding sensory properties, the only change detected was in the amount of exudate.

1. Introduction

Common ingredients in broiler diets typically include corn, soy, wheat, amino acids, vitamins, minerals, probiotics, and prebiotics to enhance growth efficiency and health [1]. However, the current use of human-edible crops like corn and soy in animal feed raises ethical questions about resource allocation, especially in regions facing food insecurity. From a sustainability perspective, the production of conventional feed ingredients like soybean meals is associated with deforestation, biodiversity loss, and high greenhouse gas (GHG) emissions [2,3,4]. The cultivation of these crops often involves significant land and water use, contributing to environmental degradation [5]. Over-reliance on a few key ingredients like soybean meal and corn can lead to limited dietary diversity and make the feed industry vulnerable to price volatility and supply chain disruptions [6,7].

Current research is focusing on the use of alternative ingredients such as insect flour, algae, and by-products from the food industry to reduce reliance on conventional feed ingredients.

Broilers have high nutrient requirements due to their rapid growth rates, and their diets are typically divided into starter, grower, and finisher phases, each with specific nutrient profiles to optimize growth and feed efficiency, and reduce nutrients released in feces. The use of high-quality protein sources, energy-dense ingredients, and balanced vitamins and minerals is crucial for achieving optimal performance. With this purpose, several innovative ingredients have been tested in recent years for the formulation of livestock feeds. Among these, microalgae have emerged as a promising ingredient due to their potential nutritional and environmental benefits. Microalgae are rich in proteins (content up to 50–70%), and essential fatty acids (like n-3); they also contain bioactive compounds like vitamins and antioxidants that can potentially enhance the growth, health, and overall productivity of poultry [8]. Moreover, their use in feed can contribute to sustainability efforts, as their cultivation requires less land and water compared to traditional crops like soybeans, and they can be grown using wastewater, industrial side streams, and carbon dioxide [9].

The production of microalgae for livestock consumption involves several key steps. Firstly, suitable strains of algae need to be selected based on their nutritional composition and ability to thrive in specific environmental conditions. Arthrospira platensis (Spirulina) and Chlorella spp. are the most common species of microalgae, already used as food ingredients, whose application is expanding also towards feed reformulation.

A study by Zampiga et al. [10] investigated the effects of substituting soybean meal with Spirulina in broiler diets during the early stages of rearing, indicating that dosages up to 5% showed comparable performance to commercial soy-based meal, while higher dosages (10–15%) significantly reduced body weight and daily weight gain compared to the control group. Moreover, they observed that high levels of microalgae inclusion reduce feed intake, suggesting an effect on feed palatability or digestibility. Further insights from Zampiga et al. [11] revealed that replacing soybean meal with Spirulina during the grower and finisher phases can be partially tolerated in quantities up to 3%. Higher dosages (6%) significantly reduced body weight, daily weight gain, and amino acid digestibility, compared to the control group. Moreover, Spirulina inclusion in the diet resulted in redder and more yellow meat with a slightly increased umami flavor but also higher lipid oxidation levels compared to the control diet [12]. Regarding Chlorella spp., findings from Bošković Cabrol et al. [13] indicated that inclusion levels of 15% and 20% significantly reduced body weight, weight gain, and feed intake compared to the control (soy-based diet) group, while a 10% inclusion level did not affect growth performance and recorded the highest score in the sensory evaluation. The fatty acid profile of the meat was also improved, with higher concentrations of EicosaPentaenoic Acid (EPA), DocosaHexaenoic Acid (DHA), n-3 polyunsaturated fatty acids, and a lower n-6/n-3 ratio. Moreover, C. vulgaris was found to increase breast muscle yield and improve meat quality parameters, such as water-holding capacity and cooking loss, resulting in yellower breast meat due to increased chlorophyll and carotenoid content [14]. Of course, the choice of microalgae species is important because the effect depends on the specific nutritional value, digestibility, and cell structure. Sun et al. [15] reported that feeding chickens with Chlorella, Tetraselmis, and Nannochloropsis oceanica microalgae and their combination induced different expressions of genes related to muscle hypertrophy or atrophy.

Microalgae from the genus Tetraselmis, Nannochloropsis, Isochrysis, and Dunaliella, that are already used in aquaculture or for the production of food supplements, do not produce toxins and represent good alternatives for feed reformulation [16,17,18]. Among these strains, the species of the marine genus Nannochloropsis deserve special attention, due to their suitability for intensive culture and high content of PUFAs (in particular EPA), antioxidants, and some vitamins. Marine microalgae from the Nannochloropsis sp. show a high capability to store lipids (21–28% DM) mainly in the form of triacylglycerols (TAG), where n-3 fatty acid EPA, crucial for brain and cardiovascular health, comprises up to 12% and 39% of total fatty acids (for dry matter) [19,20]. In the European Union, Nannochloropsis spp. biomass is currently authorized as a feed material but is not generally approved as a food ingredient for human consumption. Outside the EU, regulatory approval varies by jurisdiction and is typically limited to specific Nannochloropsis-derived products (e.g., purified oils or extracts), rather than whole biomass.

Besides identifying the most appropriate species, challenges remain in optimizing the dosage and blend of different algae species to maximize benefits, and the economic feasibility of large-scale microalgae production and its integration into commercial poultry feed needs further investigation.

This work aims to evaluate the feasibility of formulating a broiler feed in which 30% of the crude protein from soybean meal is replaced by Nannochloropsis oceanica (N. oceanica) single-cell ingredients. Animal growth parameters were evaluated through in vivo trials, comparing this formulation with a soy-based diet. Additionally, the impact on meat quality was assessed by evaluating color, nutritional composition (including fatty acid profile and carotenoid content), chemical and microbiological stability, and sensory attributes.

2. Materials and Methods

2.1. Feed Formulation, In Vitro Digestibility, and Characterization

Microalgae N. oceanica single-cell ingredients were provided by Necton S.A. (Olhão, Portugal). Nutritional and fatty acid composition of the ingredient are shown, respectively, in

Table A1. Composition and nutritional value of Nannochloropsis oceanica from ingredient technical sheet.

Parameter	Amount (g/100 g)
Total Carbohydrates	25.1
Protein	44.8
Lipids	16.2
Ashes	11.4
Moisture	2.50
Carotenoids	0.75

and

Table A2. Fatty acid composition of Nannochloropsis oceanica from ingredient technical sheet.

Parameter	% of Total Lipids
Saturated	35
MUFA	30
PUFA	35
n-3	30
n-6	7

MUFA = Monounsaturated Fatty Acid; PUFA = Polyunsaturated Fatty Acid.

. Raw materials, amino acids, and vitamin-mineral premix were acquired from commercial sources for livestock feeding. The complete list is reported in

Table 1. Feed formulations for control and N. oceanica (NOC) treatment and proximate composition.

Ingredient	Control Feed (g/100 g)	NOC Feed (g/100 g)
Wheat	64.46	62.28
Soybean meal	29.73	20.83
N. oceanica SCI	0.00	8.94
Soybean oil	1.93	1.42
L-lysine HCl	0.30	0.33
L-threonine	0.17	0.16
DL-methionine	0.29	0.30
L-Valine	0.06	0.08
Salt (NaCl)	0.26	0.25
Calcium carbonate	0.66	0.90
Dicalcium phosphate	1.34	1.21
Sodium bicarbonate	0.17	0.41
Vitamin-mineral premix 1	0.40	0.40
Choline Chloride	0.08	0.15
Noxyfeed® 2	0.020	0.020
Endofeed®DC 3	0.012	0.012
AXTRAPhy®5000 3	0.020	0.020
Proximate composition
Gross Energy (kcal/kg)	4016	3987
Dry Matter (g/100 g)	89.6	90.4
Crude Protein (g/100 g)	23.4	23.4
Crude Fat (g/100 g)	3.1	3.0
Ashes (g/100 g)	5.5	7.9

¹ Vitamin-mineral premix composition per kg feed was: vitamin A 10,000 UI; vitamin D3 4800 UI; vitamin E 45 mg; vitamin B1 3 mg; vitamin B2 9 mg; vitamin B6 4.5 mg; vitamin B12 40 µg; vitamin K3 3 mg; calcium D-pantothenate 16.5 mg; nicotinic acid 51 mg; folic acid 1.8 mg; biotin 150 µg; Fe 54 mg; I 1.2 mg; Cu 12 mg; Mn 90 mg; Zn 66 mg; Se 180 µg; calcium carbonate as carrier up to 4 g. ² Noxyfeed, (Industrial Tècnica Pecuaria, S.A. Barcelona, Spain) contained 56% BHT + Propyl gallate and 14% citric acid. ³ Endofeed^®DC and AxtraPhy^® 5000 were included following products’ recommendations to provide the following units of activity per kg of complete feed stuffs.

.

For feed validation, the Brill^TM Formulation program (version 2.9, Format Solutions Inc., Florham Park, NJ, USA) was used, with the Aviagen “Ross308 2022 all veg” requirements for feed formulation (settings: AMEn 2830 kg/kg, 21.5% CP, SID lys 12.9 g/kg, 8.8 g Ca/kg, 3.8 g npP/kg, and ideal aa profile ratio). As this was intended as an exploratory in vivo study, a single level of inclusion of the microalgae was used, replacing 30% of the crude protein coming from soybean meal, and a single formulation was provided from start to the end of the study (1 to 35 days). Complete feeds, 200 kg for each treatment, were prepared using a double ribbon 500 L horizontal mixer (Rosal Agroindustrial Facilities, Inc., Barcelona, Spain, 6 min mixing time). All ingredients, except for minerals, vitamins, amino acids, and fat, were ground in a 40 HP hammer mill (Rosal VRE-40, Rosal Agroindustrial Facilities Inc.) to pass through a 3 mm sieve. Minerals, amino acids, vitamins, and fat (through Mangra nozzles, Mangra S.A., Manlleu, Spain) were directly added to the mill. The whole process was automated and controlled by the software Autronic AJ-400 (version 1.1, Autronic s.r.l., Carpi, Italy). The composition of the control and the N. oceanica (NOC) feed is reported in Table 1. The diets were formulated to be iso-nitrogenous and iso-energetic. The difference in the inclusion level of certain ingredients between the control and NOC diets is a direct consequence of these formulation constraints to maintain the targeted metabolizable energy level, compensate for the different protein and ash contents of N. oceanica compared to soybean meal, and ensure correct pellet structure and feed manufacturability. According to this, although the diets were formulated to replace soybean meal-derived crude protein on a quantitative basis, the substitution effectively involved replacing a complex feed ingredient matrix. Therefore, the observed effects should be interpreted as the result of combined differences in digestibility, energy availability, and non-protein components of N. oceanica, and not only as protein replacement alone.

The nutritional evaluation of feeds was performed, assessing the contents of dry matter, crude protein, ether extract, ash, and gross energy, using the corresponding AOAC methods [21].

The in vitro digestibility was measured using the 2-phase method described by Yegani et al. [22]. Digestibility coefficients of dry matter (DM_d), crude protein (CP_d), organic matter (OM_d), and gross energy (GE_d) were calculated.

2.2. In Vivo Trials

The experiments followed European Union principles for animal care and experimentation (EC Directive 86/609/EEC) and the Spanish guidelines for experimental animal protection (Royal Decree 53/2013 of 1 February on the protection of animals used for experimentation or other scientific purposes).

Animals were reared in the experimental farm at IRTA (Institute of Agrifood Research and Technology, Constantí, Spain). Male Ross 308 broiler chickens used in the trials were obtained from commercial hatcheries. A total of 90 animals were used for the in vivo trials. A Randomized Complete Block Design was used, distributing the animals in two treatments: the control group, fed with conventional feed, and the “NOC” group, raised with N. oceanica-enriched feed. Each group was randomly assigned to 9 cages (1 m²) with 5 chickens each. Experimental diets were provided continuously from day 0 (arrival from the hatchery) until day 35 of age. No phase-specific reformulation (starter, grower, finisher) was applied; both treatments followed the same feeding strategy throughout the entire rearing period.

Performance (body weight and feed consumption) was assessed per triplicate at 10 and 35 days, and also before and after the feces collection period. Body Weight (BW) was directly measured on the animal, while Average Daily Gain (ADG), Average Daily Feed Intake (ADFI), Feed Conversion Ratio (FCR), and European Production Efficiency Factor (EPEF) were calculated.

To determine the in vivo digestibility of feed, a 3-day collection of feces was implemented from day 15 to 18. For this purpose, excreta were kept frozen until analysis; samples were freeze-dried, and then analyzed for dry matter (DM), crude protein (CP), organic matter (OM), ashes (ASH), and gross energy (GE), and total tract digestibility coefficients (DM_d, CP_d, OM_d, ASH_d, GE_d) were calculated using Equation (1) [23].

$$\mathrm{D}\mathrm{i}\mathrm{g}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}\mathrm{c}\mathrm{o}\mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{t}(\mathrm{\%})=\left({\displaystyle \frac{Nutrientintake-Nutrientinfeces}{Nutrientintake}}\right)\times 100$$

(1)

Moreover, the apparent metabolizable energy (AME) coefficient was calculated using Equation (2) [24]:

$$\mathrm{A}\mathrm{M}\mathrm{E}(\mathrm{k}\mathrm{c}\mathrm{a}\mathrm{l}/\mathrm{k}\mathrm{g})=\left({\displaystyle \frac{GEintake-GEinfeces}{Feedintake\left(kg\right)}}\right)$$

(2)

2.3. Meat Quality Analysis

A total of 36 chickens (2 birds/replicate) were selected for meat quality analysis, distributed as follows: control (n = 18) and NOC treatment (n = 18). The animals were slaughtered in a commercial abattoir (Granja Gaià, Tarragona, Spain) and the cooled carcasses were delivered to IRTA facilities (Monells, Spain) at 4 °C. For each treatment, the animals were split into 2 groups: 9 animals were destined for meat physicochemical and sensory analysis, while the other 9 were used for shelf life and composition analysis.

2.3.1. Breast and Carcass Performance

Performance parameters, namely carcass and both breast sides, were weighed, and their yield was calculated. To calculate the carcass yield, the body weight of the animal at day 36 was considered.

2.3.2. Physicochemical Analysis

For meat physicochemical analysis, one of the two breasts of each animal was taken into account and divided as shown in Figure 1.

Specifically, pH, electrical conductivity, color (also measured on the skin), drip loss, and texture were evaluated. The pH was measured with a pH penetration electrode (Crison 52-32) on a portable pH-meter (CRISON PH 25, Crison Instruments S.A., Alella, Spain).

Electrical conductivity was measured with a conductometer (VWR^® EC30, VWR International, Eurolab LLC, Llinars del Vallès, Spain). This measurement is positively correlated to the extracellular water susceptible to exudation, offering an additional parameter to evaluate meat quality.

Color was determined on the skin and breast with a Minolta CR-600d colorimeter (Minolta Co., Osaka, Japan) by measuring CIE L*a*b parameters L* (lightness), a* (redness), and b* (yellowness) with standard illuminant D65 and 10° viewing angle.

Drip loss was analyzed by the EZ drip method, described by Rasmussen and Andersson [25], obtaining the weight of the exudates from a cylinder of breast meat, in duplicate. Briefly, one day post-slaughter, the muscles of interest were removed from the carcass. Within one hour, a 25 mm slice was cut perpendicular to the muscle fibers and immediately cored in the fiber direction using a 25 mm steel corer. The resulting sample was placed in a sealed container to prevent evaporation and exudate loss and stored at 4–6 °C for 24 h. Drip loss was then assessed by weighing.

Meat texture was evaluated instrumentally by determining the shear force needed to cut it into two pieces, according to the Warner–Bratzler method [26,27] using the Texture Analyzer TA HD-Plus and software Exponent 32 (Stable Micro Systems Ltd., Godalming, United Kingdom). Previously to this test, the samples were cooked in a convection oven preheated at 200 °C (SCC 101G, Rational AG, Lech, Germany) and 100% humidity level until they reached a core temperature of 74 °C. This treatment has been designed according to the literature data in order to eliminate any pathogens that may be present [28,29,30].

2.3.3. Composition and Nutritional Analysis

Proximate composition, fatty acid profile, and carotenoid content were determined for the chicken breast.

Proximate Composition

The percentual moisture content (MC) of the samples on a dry matter basis was determined by drying at 103 ± 2 °C until constant weight [31]. Total fat was determined by Soxhlet extraction [32] and protein content by Kjeldahl [33]. Ashes were determined by weighing approximately 5–10 g of meat and placing the samples at 600 °C for 2 h according to the AOAC official method [34].

Fatty Acids Profile

Fatty acids quantification was carried out by gas chromatographic analysis. First, fatty acids of approximately 1.5 g of minced chicken breast were extracted following the procedure described by Folch et al. [35] using 15 mL of a chloroform/methanol (2:1, v/v). Extracted lipids were converted to fatty acid methyl esters (FAMEs) using 3 mL of 0.5 N CH₃ONa–methanol solution following the standardized procedure ISO 12699-2:2017 [36]. FAMEs were analyzed and quantified employing a gas chromatograph with flame ionization detector (GC-FID Agilent 8860, 2026 Agilent Technologies Inc., Santa Clara, CA, USA) with a Zebron ZB-FAME capillary column (30 m, 0.25 mm i.d., 0.20 μm; Phenomenex Inc., Torrance, CA, USA). The chromatographic conditions were: H₂ as carrier gas at 1.7 mL/min (constant flow), injector temperature 240 °C (split mode, 50:1), and detector temperature 260 °C. The oven temperature was initially set at 100 °C for 2 min, followed by an increase of 10 °C/min until 140 °C, then 3 °C/min until 190 °C, and finally an increment of 10 °C/min until 260 °C, after which it was held for 5 min (34.67 min total run). Identification of the single methyl esters was done by comparing the retention times of the peaks with a commercial standard of FAME Mix C4-C24 (Supelco Analytical, Merck Life Science S.L.U., Madrid, Spain). Samples were analyzed in duplicate using tripentadecanoin as internal standard (T4257, Merck Life Science S.L.U.). Results were expressed as % of total FA (qualitative evaluation). The average amount of each fatty acid was used to calculate the % of total Monounsaturated and Polyunsaturated Fatty Acids (MUFAs and PUFAs, respectively) and ω-3 and ω-6 series.

Carotenoid Content

Carotenoids were extracted following Biehler et al. [37] with minor modifications. Approximately 1.5 g of sample were mixed with 8 mL of hexane:acetone (1:1, v/v) and homogenized using an Ultra-Turrax T18, (IKA, Barcelona, Spain) at 13,500 rpm for 30 s, keeping the sample on ice. The mixture was then stirred for 15 min using a magnetic stirrer and centrifuged at 2500× g for 5 min. The upper (organic) phase was transferred to a beaker, while the lower (aqueous) phase was returned to the separation funnel and re-extracted in the same way. The extracted organic phases were then combined, and 25 mL of a 30% NaCl aqueous solution was added. After mixing and allowing phase separation for 5 min, the upper phase was collected into a tared 50 mL Falcon tube. The aqueous phase was extracted again with 8 mL hexane, and this final organic layer was combined with the previous extract. The total organic phase (O) was weighed. A 5 mL aliquot was transferred to an amber tube and weighed (A), then evaporated to dryness under N₂. The residue was dissolved in a known volume V (e.g., 1 mL) of acetone, sonicated for 2 min, and centrifuged again (2500× g, 5 min).

The absorbance of the extract (diluted if needed) was measured at 450 nm using a spectrophotometer. Total carotenoids were calculated using Equation (3) [37].

$$\mathrm{C}\mathrm{a}\mathrm{r}\mathrm{o}\mathrm{t}\mathrm{e}\mathrm{n}\mathrm{o}\mathrm{i}\mathrm{d}\mathrm{s}(\mathrm{g}/\mathrm{k}\mathrm{g})=\mathrm{A}\mathrm{b}\mathrm{s}\left(450\mathrm{n}\mathrm{m}\right)·{\displaystyle \frac{1}{{\mathrm{e}}_{avg}·l}}·\mathrm{m}·\mathrm{F}·\mathrm{V}·{\displaystyle \frac{\mathrm{O}}{\mathrm{A}·\mathrm{W}}}·1000$$

(3)

where

• ε_avg = 135,310 L/(mol·cm), average absorption coefficient of carotenoids.
• l = 1 cm (cuvette length).
• m = 548 g/mol, average molar weight of carotenoids.
• F, dilution factor during the sample measurement.
• V (l), volume of acetone used to redissolve the dried sample.
• O (g), weight obtained of final organic phase after extraction.
• A (g), weight of organic phase aliquot brought to dryness.
• W (g), weight of initial sample.

2.3.4. Shelf Life and Stability Studies

Shelf life was evaluated in sliced breast samples of 2.5 cm width disposed in modified atmosphere packaging (MAP), with an initial gas composition of 75% N₂ and 25% CO₂. From each animal (n = 9/treatment), one slice from each of two breasts was placed in the tray, and 3 trays per animal were obtained, corresponding to days 1, 6, and 9 of post-packaging analysis, sampling each breast from the cranial to the caudal region (Figure 1). The samples were stored in a cold room (4 °C) applying cycles of 12 h of light and 12 h of darkness. In each of the evaluation days, a tray from each animal was drawn, and gas composition, pH, instrumental color, Thiobarbituric acid reactive substances (TBARSs), and microbiological counts were taken into account.

The oxygen and carbon dioxide composition inside the tray was determined using a Dansensor Check Mate II gas analyzer (Ametek Mocon, Brooklyn Park, MN, USA).

pH and instrumental color were measured as previously described in the “Physicochemical analysis” section.

Lipid oxidation is determined by the TBARS method following the procedure of Bou et al. [38] with some modifications. Homogenization of 5 g of sample in 20 mL of acidic solution of trichloroacetic acid 15% for 30 s (Ultra-Turrax IKA T18, IKA-Werke GmbH & Co. KG., Staufen, Germany). The resulting homogenate is then centrifuged for 15 min at 10.000 g (Beckman Avanti JXN-30, Beckman Coulter Inc., Brea, CA, USA). An aliquot of supernatant is filtered and reacts with the same volume of a 20 mM TBA aqueous solution at 70˚C for 30 min. The sample absorbance is measured at 530 nm (pink coloration). The amount of TBARS is calculated using a tetraethoxy-propane standard (TEP) calibration line from 0.012 to 1.2 ppm. The results are expressed in μg malondialdehyde (MDA) per g of sample.

For microbiological analysis, total aerobic bacteria (TAB), Lactic acid Bacteria (LAB), Enterobacteriaceae (EB), and Escherichia coli (E. coli) enumeration were performed. The preparations of samples, initial suspension, and decimal dilutions for microbiological analyses were performed according to the International Organization for Standardization (ISO) protocol [39]. To determine TAB, incubation in Plate Count Agar (PCA) (Merck, Darmstadt, Germany) at 30 ± 1 °C for 72 h was used, according to the ISO method 4833-1 [40]. LB colony enumeration was carried out by Man, Rogosa & Sharpe agar (MRS) (Merck, Darmstadt, Germany), after anaerobic incubation at 30 ± 1 °C for 72 h, following ISO protocol 15214 [41]. Total EB and E. coli count were determined in a REBECCA^® EB medium (bioMérieux Espãna S.A, Madrid, Spain) using pour plating and incubation at 37 °C ± 1 °C for 24 h [42].

All determinations were performed in triplicate. Counts were expressed in log10 (cfu/g). The detection limit for all the count methods was 1 log10 cfu/g.

In addition to physicochemical and microbiological analysis, a group of 5 trained panelists evaluated daily color/visual and odor perception of raw breast meat until day 9 post packaging from Monday to Friday. The subjective perceptions were recorded using a 5-point scale: (1), highly undesirable; (2), moderately undesirable; (3), slightly desirable; (4), moderately desirable; and (5), highly desirable [43].

Finally, for the study of the visual evolution of the samples, photographs were taken of the packaged samples under a LED 99 CRI standardized lighting system (Waveform Lighting LLC., Vancouver, WA, USA), using a Canon EOS RP camera and a Canon EF 24–105 mm zoom lens (Canon Inc., Tokyo, Japan) set to 85 mm, with an exposure time of 1 s and an aperture of f/16.

2.3.5. Sensory Analysis Methodology

Sensory analysis was carried out by a trained panel made up of 6 tasters. All of them had more than 2 years of experience in the descriptive analysis of different foods. Initially, two open discussion sessions were held to select by consensus the sensory attributes to be evaluated. The evaluated sensory parameters included 3 odor attributes (general intensity, red meat, and cooked egg white), 3 visual attributes (exudate amount, color intensity (from white to yellow), and protein remnants amount), 3 taste attributes (chicken, metallic, and bitter), and 3 texture attributes (hardness, juiciness, and adherence to the teeth). The intensity of these attributes was evaluated on a scale from 0 (absence of attribute) to 10 (maximum intensity). The evaluation was performed using a complete design. A total of 4 sessions were held. Samples were presented in different orders for tasters and sessions following a Williams Latin square design (balanced for residual or first-order carryover effects) [44]. The effect of the anatomic part was also blocked among different panelists and sessions. All samples, coded with 3-digit random numbers, were analyzed in a standardized tasting room [45]. The tasters were provided with mineral water and apple slices to clean their palates between samples. The performance and reliability of the panel were verified using the standard methodology [46].

2.4. Statistical Analysis

Data were analyzed by one-way analysis of variance (ANOVA), and significant differences between mean values were assessed. A 95% level of confidence (p < 0.05) was used as the threshold for significance.

All experimental data except from sensory tests were statistically analyzed using either SAS 9.4 (SAS System for Windows V.9.4. Cary, NC, USA.) or JMP Pro 17 software (JMP Statistical Discovery LLC, Cary NC, USA).

Sensory test data analysis was performed using XLSTAT^® software (version 2021.1.1.1110, Lumivero, Denver, CO, USA). In this case, the effect of the session and taster were considered as random factors.

The standard error of the mean (SEM) was calculated for all datasets, and is presented in the tables in Section 3.

3. Results

The results of the study are reported in the following, divided by category, starting from feed characterization and growth performance evaluation (Section 3.1 and Section 3.2, respectively) and then moving to meat nutritional (Section 3.3) and quality (Section 3.4, Section 3.5 and Section 3.6) analysis.

3.1. Feed Proximate Composition and In Vitro Digestibility

Table 1 presents the proximate composition of the feed formulations. As expected, they were well balanced in energy, dry matter, crude protein, and fat. However, ash content was notably higher, likely due to the mineral and salt-rich nature of N. oceanica. Soybean meal ash average content is 7.5% [47], while the technical sheet of the N. oceanica ingredient used in this study (Table A1) revealed a content of 11.4%.

Results on in vitro digestibility are reported in

Table 2. Poultry feeds in vitro digestibility coefficients (n = 6) of dry matter (DM_d), crude protein (CP_d), gross energy (GE_d), and organic matter (OM_d) are shown.

Parameter	Control	NOC	SEM	p-Value
DMd	80.2	79.0	1.10	0.600
CPd	92.7	85.1	0.50	<0.001
GEd	77.0	74.6	0.49	0.035
OMd	77.3	76.4	0.59	0.470

. All the coefficients are lower for NOC feed, suggesting a reduction in digestibility, although the only significant differences were observed regarding crude protein and gross energy. The reason is likely due to the microalgae’s thick and rigid cell wall. These characteristics make cells highly resistant to enzymatic degradation, thus limiting the release and absorption of intracellular nutrients such as proteins [48]. Additionally, N. oceanica contains non-starch polysaccharides, which are known to reduce nutrient digestibility by increasing digesta viscosity, interfering with enzyme activity, and altering intestinal morphology [49]. Certain pretreatments of microalgae biomass, like freeze–thawing, enzyme hydrolysis, and pressure homogenization, could help in breaking the cells and releasing nutritional components and protein, enhancing digestibility. This is something to be considered in future studies.

3.2. Feed Performance and In Vivo Digestibility

Results about in vivo digestibility of feed formulations are reported in

Table 3. Poultry feed in vivo digestibility mean values (n = 6). Different letters in the same row, when reported, indicate significant differences between control and NOC treatments.

Parameter	Control	NOC	SEM	p-Value
DMd	70.6	68.2	0.52	0.032
CPd	65.6	63.3	0.87	0.137
Ashesd	52.3	47.7	1.87	0.159
OMd	71.8	70.2	0.60	0.132
GEd	73.3	70.5	0.45	0.012
AME (kcal/kg)	2944	2812	17.8	0.006

Dry matter (DM_d), crude protein (CP_d), organic matter (OM_d), ashes (ASH_d), gross energy (GE_d), and apparent metabolizable energy (AME).

. The NOC formulation presents a slightly lower digestibility of DM and GE. Despite the diet being designed to be isoenergetic, the lower AME measured for the NOC diet suggests that the microalgal inclusion affected energy availability, which may have contributed to reduced feed efficiency and growth independently of protein quality. The reduced digestibility, as previously commented, could be connected to the high fiber and anti-nutritional compounds in microalgae (e.g., polysaccharides, phenolics). As reported in the literature, these compounds may interfere with nutrient absorption or gut comfort [50].

These results confirm only partially the ones observed for in vitro studies: DM digestibility is lower in both cases, but, positively, the lower CP digestibility observed in vitro is not confirmed in vivo.

Digestibility results were also reflected in the feed performance. The data reported in

Table 4. Feed in vivo performance, mean values evaluated at 10 and 35 days (n = 9). Different letters in the same row, when reported, indicate significant differences between control and NOC treatments.

Parameter	Control	NOC	SEM	p-Value
BW 1 to 10 days (g)	241	242	6.90	0.881
ADG 1 to 10 days (g/day)	19.6	19.7	0.69	0.881
ADFI 1 to 10 days (g/day)	28.8	28.5	0.67	0.737
FCR 1 to 10 days	1.482	1.451	0.02	0.253
BW 1 to 35 days (g)	2108	1816	44.4	<0.001
ADG 1 to 35 days (g/day)	58.9	50.6	1.27	<0.001
ADFI 1 to 35 days (g/day)	80.2	74.6	1.77	0.010
FCR 1 to 35 days	1.362	1.474	0.01	<0.001
EPEF 1 to 35 days	361	286	8.40	<0.001

Body weight (BW), Average Daily Gain (ADG), Average Daily Feed Intake (ADFI), Feed Conversion Ratio (FCR), and European Production Efficiency Factor (EPEF).

showed significant differences for all the parameters after 35 days. All performance variables (BW, ADG, ADFI, FCR, and EPEF) were impaired in broilers fed with NOC diets. However, the absence of significant differences after 10 days suggested a possible use of the feed in the first stages of the feeding phase without compromising the performance. After 10 days, the animals apparently started eating less, impairing body weight gain and efficiency. The explanation, according to the literature, could be related to the fact that early-stage broilers have less developed sensory and metabolic feedback systems, so they may not detect or respond to subtle differences in feed composition [51]; however, these aspects were not directly assessed in the present study. As a consequence, digestibility issues or palatability changes become more impactful with growth. Additionally, it is possible to hypothesize that the higher content of fiber, unsaturated fatty acids, and anti-nutritional factors in microalgae, compared to soy, may induce earlier satiety, thus reducing the feed intake as observed in the present study [52].

The previous results found further confirmation on the data of the body parameters measured after slaughter, reported in

Table 5. Carcass and breast weight and yield parameters (mean values, n = 18). Different letters in the same row, when reported, indicate significant differences between control and NOC treatments.

Parameter	Control	NOC	SEM	p-Value
Carcass weight (kg)	1.666	1.453	0.027	<0.001
Left Breast Weight (kg)	0.199	0.161	0.005	0.003
Right Breast Weight (kg)	0.193	0.161	0.005	0.003
Carcass yield (%)	80.21	78.83	0.141	<0.001
Breast yield (%)	23.48	22.10	0.289	0.022

. The broilers who followed the NOC diet showed significantly lower weight and yield of carcass and breasts.

Similar results are reported in the literature. Sun et al. [15] observed that feeding broilers chicken with Chlorella sp., Tetraselmis sp., and N. oceanica, alone or in combination to replace soybean meal in an amount of 10%, decreased the breast weight percentages. Spinola et al. [53] reported that the inclusion of 15% of Spirulina in broiler diets reduces the daily feed intake, lowering the final body weight and carcass yield. Pestana et al. [54] attribute these negative effects to the high incorporation levels of microalgae that may promote biomass gelation and escalate digesta viscosity, thereby potentially hindering nutrient digestibility and bioaccessibility. On the other hand, other studies reported no influence on body parameters [55,56] or even a positive influence but with lower percentages of microalgae inclusion [57,58].

3.3. Meat Composition and Nutritional Values

Table 6. Distribution of fat, protein, moisture, and ash percentage determined in chicken breast samples (mean values, n = 9), according to treatment. Different letters in the same row, when reported, indicate significant differences between control and NOC treatments.

Parameter	Control	NOC	SEM	p-Value
Protein (g/100 g)	22.47	22.78	0.111	0.181
Fat (g/100 g)	1.024	0.776	0.079	0.137
Moisture (g/100 g)	75.32	75.65	0.195	0.425
Ashes (g/100 g)	1.344	1.335	0.008	0.590
Carotenoids (mg/kg)	0.191	1.254	0.031	<0.001

shows the proximate composition of breasts, while in

Table 7. Fatty acids composition (mean values, n = 9) in chicken breast samples, according to treatment. Different letters in the same row, when reported, indicate significant differences between control and NOC treatments.

Fatty Acids Composition % (g/100 g of Fat)	Control	NOC	SEM	p-Value
Saturated	38.77	38.87	0.926	0.977
Unsaturated	61.85	61.65	0.879	0.950
MUFA	26.37	29.00	1.847	0.560
PUFA	35.48	32.65	0.981	0.433
n-3	2.69	4.05	0.058	0.010
ALA	1.10	1.22	0.171	0.805
EPA	0.42	1.63	0.068	0.021
DHA	1.17	1.20	0.079	0.914
n-6	32.79	28.6	0.912	0.270

MUFA = Monounsaturated Fatty Acid; PUFA = Polyunsaturated Fatty Acid; ALA = Alpha-Linolenic Acid; EPA = EicosaPentaenoic Acid; DHA = DocosaHexaenoic Acid.

, the fatty acid composition is reported. The only significant differences detected in composition and in nutritional value linked with microalgae diets were: (i) an increase in carotenoid content, due to the high content of this pigment in the Nannochloropsis biomass and (ii) a significant increase in n-3 fatty acids, especially linked to an increase in EicosaPentaenoic Acid (EPA) due to its higher content in microalgae compared to soy. The increase in n-3 fatty acids is particularly valuable because poultry has a limited ability to convert Alpha-Linolenic Acid (ALA) from plant-based feeds like soybean meal into EPA and DocosaHexaenoic Acid (DHA) [59]. Soybean is a good source of ALA but does not naturally contain EPA and would require genetic modification to produce them [60], while N. oceanica offers a direct and effective source.

On the other hand, the increase in carotenoids, due to their important antioxidant properties, also contributed to enhancing meat quality [61]. These findings are consistent with those of other authors who confirmed that microalgae such as Chlorella and Spirulina inclusion in poultry diets reliably enhances meat quality through increased polyunsaturated fat levels, especially EPA and DHA, and increased carotenoid deposition [13,14,62,63,64].

3.4. Meat Physicochemical Parameters

Figure 2 shows the color intensity expressed as CIELab coordinates measured respectively on the chicken skin and on the chicken breast. As can be seen, the differences between samples with a conventional diet and microalgae one especially involve the intensity of yellow (b* parameter), which is higher for microalgae treatments both in skin and breasts (see also Figure A1, where differences in feet color are also visible). Regarding the breast color, a significant increase in redness (a* parameter) was also observed. Since carotenoids have been demonstrated to contribute to pigmentation in the meat, skin, legs, feet, and beak of birds [65], the color changes observed as a consequence of NOC diets are linked with the increase in carotenoid content already discussed in Section 3.3. Similar results have been observed for comparable microalgae dietary inclusion in the literature, especially regarding the increase in yellow color [10,66]. A previous study by Van Nerom et al. [67] found that consumers are generally open to purchasing chicken meat from algae-fed animals, and a yellower color does not pose a significant concern.

All the other physicochemical parameters (reported in

Table 8. Distribution of pH, electrical conductivity, drip loss, and shear force values (mean values, n = 9) determined in chicken breast samples, according to treatment. No significant differences were observed for any of the parameters.

Parameter	Control	NOC	SEM	p-Value
pH	5.758	5.767	0.022	0.840
Electrical conductivity (mS)	10.70	10.68	0.511	0.991
Drip loss (%)	0.930	0.878	0.193	0.894
Shear force (kg)	1.696	1.753	0.078	0.719

), pH, electrical conductivity, drip loss, and texture, showed no significant differences, according to the results from other studies [15].

3.5. Sensory Analysis Results

Table 9. Intensity values of the sensory parameters of cooked breast samples evaluated by the panel tests (mean values, n = 44). Different letters in the same row, when reported, indicate significant differences between control and NOC treatments.

	Odor			Visual Aspect
Treatment	Overall Intensity	Red Meat	Cooked Egg White	Exudate Amount	Color Intensity	Protein Remnants
Control	5.84	2.92	3.00	6.32	4.91	4.14
NOC	5.96	2.63	2.94	5.65	5.20	3.91
SEM	0.137	0.237	0.255	0.194	0.134	0.210
p-value	0.560	0.248	0.814	0.009	0.219	0.351
	Taste			Texture
Treatment	Chicken	Metallic	Bitter	Hardness	Juiciness	Adherence (to teeth)
Control	5.97	3.93	1.77	4.95	4.26	5.75
NOC	5.80	4.08	1.84	5.00	4.19	5.61
SEM	0.146	0.222	0.185	0.114	0.162	0.149
p-value	0.467	0.549	0.756	0.727	0.721	0.496

presents the results of the sensory panel test, evaluating the intensity of the sensory attributes selected. The data shows no differences in odor, taste, and texture attributes. As reported in the literature, microalgae can have an impact on meat sensory properties, but this was observed especially for animals whose meat has higher fat content, like swine and beef [68,69]. In poultry, most of the results reported no differences in taste, texture, and odor [14,70,71]. However, with higher microalgae inclusion, sensory traits can change. In the study of Altmann et al. [12] for 50% soymeal replacement with Spirulina, a slight increase in adhesiveness, umami, and chicken overall flavor, with a contemporary reduction in barn odor, was reported. Spínola et al. [72] also reported changes in texture and flavor of chicken breast of poultry fed with a Spirulina-enriched diet (15% inclusion), suggesting that the effect depends not only on the amount but also on the microalgae considered.

Once more, looking at Table 9, we can see that, regarding the visual aspect, the only parameter significantly affected by the microalgae diet was the amount of exudate, which was lower compared to the value of the control sample. Surprisingly, the color differences observed in the raw meat were not perceived anymore after cooking. Regarding the observed change in exudate amount, since no differences in breast moisture content were detected (Table 6), it is possible to assume that the incorporation of microalgae in the diet contributed to better water retention in the muscles and reduced cooking loss. The phenomenon could be linked to the presence of natural antioxidants and bioactive compounds in microalgae (as seen previously), which help maintain muscle integrity and improve water retention during cooking [55]. These results are in line with other literature findings. For example, El-Bahr et al. [63] reported that the inclusion of microalgae as additives (1 g/kg) in broiler feed led to a reduction in breast cooking loss. Altmann et al. [12] verified increased water-holding capacity for 9.7–11.8% inclusion of Spirulina in broiler diets.

3.6. Shelf Life/Stability Studies

Shelf life parameters (instrumental color stability, gas composition inside the container, visual and olfactory perception after opening the container, pH, microbiological growth, TBARS) showed no significant changes as a consequence of the microalgae inclusion; see Figure A2, Figure A3, Figure A4, Figure A5, Figure A6 and Figure A7 (Appendix A).

Figure A2 shows that the evolution of O₂ and CO₂ concentrations within the packaging followed comparable trends in control and NOC samples, indicating that the microalgae-based diet did not alter package atmosphere dynamics or respiration-related processes. Consistently, the visual and olfactory assessments reported in Figure A3 revealed no diet-related differences in perceived freshness after opening the packages, suggesting the absence of off-odors or abnormal visual deterioration associated with the NOC treatment. As shown in Figure A4, lipid oxidation levels (TBARS) increased slightly over time in both treatments but remained low and statistically similar, demonstrating that the higher n-3 fatty acid content of NOC meat did not lead to enhanced oxidative instability during storage. Instrumental color measurements reported in Figure A5 further confirmed that color evolution over time was comparable between treatments, indicating that the increased yellowness associated with carotenoid deposition in NOC samples did not negatively affect color stability. Microbiological results (Figure A6) showed similar growth patterns of total aerobic bacteria, lactic acid bacteria, Enterobacteriaceae, and E. coli, as well as comparable pH evolution, confirming that the dietary treatment did not influence microbial safety or spoilage kinetics. Finally, the standardized photographic documentation presented in Figure A7 visually supports these findings, showing no evident differences in appearance or surface quality between treatments throughout storage.

These results collectively indicate that the inclusion of Nannochloropsis oceanica in the broiler diet did not compromise the technological or microbiological quality of the meat during refrigerated storage under modified atmosphere packaging.

A recent review by Prates [73], stated that meat from animals fed with diets enriched with algae shows significantly lower levels of malondialdehyde (MDA), compared to conventional meat. This reduction in oxidation helps retain the meat’s flavor, color, and texture during storage, extending its marketable life and reducing food waste. Improved oxidative stability also reduces the formation of off-flavors, ensuring that meat maintains its sensory qualities for a longer period.

In the present study, the shelf life evaluation of breasts did not reveal any effect of microalgae inclusion in the diet, probably because the low fat content did not lead to shelf life alteration for the observation time considered. Other authors reported a reduction in lipid oxidation and microbial count, but for longer studies [63] (32 days in breasts) or considering different cuts like chicken thigh and liver [64,74]. Regarding the other parameters studied, there are no other studies taking into account analysis as a function of time to evaluate the product stability, so it is difficult to compare.

4. Conclusions

The partial replacement of soybean meal with Nannochloropsis oceanica (30% protein substitution) in poultry feed presents a promising yet complex alternative, offering potential environmental and nutritional advantages while posing challenges to growth performance. The microalgae diet led to a slight reduction in animal growth and feed intake after the initial feeding phase, due to lower digestibility and the hypothetical presence of anti-nutritional compounds (not directly analyzed in this study). However, it should be underlined that since the diets were formulated on a crude protein basis rather than on standardized ileal digestible amino acids, the growth depression observed in broilers fed the N. oceanica diet may be attributed to essential amino acid inadequacy, particularly sulfur amino acids, rather than to intrinsic limitations of the microalgal protein. This conclusion is consistent with the reduced digestibility values measured in the algae-based diet and underscores the need for formulation strategies based on digestible amino acid supply in future studies.

At the same time, microalgae inclusion significantly enhanced the n-3 fatty acids, particularly EPA, and carotenoids, which are valuable for human health and meat quality. These improvements translated into more intense meat coloration and reduced cooking exudate, suggesting better water retention and antioxidant protection, without negatively compromising sensory attributes and shelf life stability.

Importantly, comparing microalgae to soy in terms of cost is currently unfair, as the production scale of microalgae is far smaller. To enhance the competitiveness of microalgae ingredients and reduce their costs, substantial infrastructure investments and technological advancements are required.

Future research should focus on optimizing dietary inclusion by studying the dose–response effect, considering different levels of inclusion. Improving digestibility through microalgae pre-treatment and exploring phase-feeding strategies to mitigate performance drawbacks are other points that need to be deepened in future works. Additionally, broader studies on different poultry breeds, meat cuts, and longer storage periods could further validate the commercial viability and scalability of microalgae-based feed formulations.