Blended Oil With an Optimized Fatty Acid Profile Improves Fish Meal Substitution Efficacy in Carnivorous Teleost Largemouth Bass Diet

Abstract

Several factors are known to affect the substitution of fish meal (FM) in aquafeeds, yet the influence of dietary fatty acid (FA) composition remains unclear. To investigate this, fish oil (FO), an FA‐optimized blended oil (BO1) designed to meet the essential FA (EFA) requirements of largemouth bass (Micropterus salmoides), and a blend rich in n‐6 polyunsaturated FAs (PUFAs) (BO2, a 2:3 mixture of FO and soybean oil) were used as dietary lipid sources. Three isoproteic (50%) and isolipidic (9%) diets with distinct FA profiles were formulated at either 24% (24FO, 24BO1, and 24BO2) or 16% (16FO, 16BO1, and 16BO2) FM inclusion levels. Juvenile fish (initial weight about 12.50 g) were fed the diets for 10 weeks. Results showed no significant differences in growth performance among the 24% FM groups. At the 16% FM level, the 16BO1 group exhibited growth comparable to the 16FO group and achieved significantly higher final body weight (FBW), weight gain rate (WGR), and specific growth rate (SGR) than the 16BO2 group (p < 0.05). Moreover, compared to 16BO2, the 16BO1 group demonstrated improved lipid metabolism (indicated by reduced hepatosomatic index [HSI], viscerosomatic index [VSI], triglycerides [TGs], nonesterified FAs [NEFAs], and blood urea nitrogen [BUN]), enhanced protein synthesis (reflected in increased total amino acids [TAAs], alanine transaminase [ALT], and aspartate transaminase [AST]), elevated antioxidant capacity (total antioxidant capacity [T‐AOC] and catalase [CAT]), and upregulated mRNA expression of genes related to lipid oxidation (pparα, atgl, and acsl4) and protein synthesis (akt2 and eif4g). These findings demonstrate that optimizing dietary FA composition enhances FM substitution efficacy by promoting lipid‐based energy supply, improving protein synthesis, and strengthening antioxidant responses. This study is the first to reveal that dietary FA profiles modulate FM replacement efficiency in aquatic feeds, providing new insights and viable strategy for developing low‐FM diets to promote sustainable largemouth bass aquaculture.

1. Introduction

The escalating demand for aquaculture products has intensified reliance on fish meal (FM) as a primary protein source in aquafeeds due to its balanced amino acid profile and high digestibility [1–3]. However, FM scarcity and rising costs have spurred exploration of sustainable alternatives, such as plant proteins and animal by‐products [4–6]. Current research has identified amino acid imbalance, antinutritional factors, and micronutrient deficiencies as primary constraints affecting FM substitution efficacy [7–9], while the effects and regulatory mechanisms of dietary fatty acid (FA) composition—a critical nutritional component in aquatic feeds—on FM replacement efficiency remain unknown to date.

Lipids play a pivotal role in fish nutrition, serving as energy sources and structural components of cell membranes while modulating metabolic pathways linked to protein turnover [10–12]. Notably, the FA composition of lipid sources has been shown to significantly influence protein synthesis and degradation in trout, mice, and cattle [13–15]. For instance, in livestock studies, stearic acid (SA) activates the PI3K‐mTOR‐4EBP1/S6K and mTOR‐SREBP‐1 pathways by upregulating cyclin‐dependent kinase 1 (CDK1) to enhance milk synthesis [15]. Similarly, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) significantly affect protein metabolism in mouse C2C12 cells, with EPA demonstrating greater capacity to increase skeletal muscle protein content [14]. In fish, EPA and DHA promote the proliferation and protein synthesis of primary turbot muscle cells through the Akt‐TOR‐S6K pathway [13]. Moreover, our recent studies in the carnivorous marine teleost golden pompano (Trachinotus ovatus) showed that fish fed diets containing 6% or 12% FM achieved growth performance comparable to those fed 30% FM [16, 17]. These results were achieved using blended oils with optimized FA profiles specifically designed to meet the essential FA (EFA) requirements of T. ovatus, providing 1.24%–1.73% n‐3 long‐chain polyunsaturated FAs (n‐3 LC‐PUFAs) with a DHA:EPA ratio of ~1.4 [18, 19]. These findings underscore the regulatory role of FAs on protein metabolism, suggesting that dietary FA composition may influence FM substitution efficacy.

Largemouth bass, an important carnivorous freshwater fish widely cultivated in China, traditionally requires diets containing over 35% FM for optimal growth and metabolism [20]. Extensive research has been conducted on FM substitution for this species, with alternative protein sources including plant‐based, microbial, and insect proteins [21]. We recently established that largemouth bass juveniles lack LC‐PUFA biosynthetic capacity and require about 0.76% n‐3 LC‐PUFA with a DHA:EPA ratio of 1.06 in diets [22]. To evaluate whether dietary FA profiles influence FM replacement efficacy and the underlying mechanisms in largemouth bass, three diets with different FA composition were prepared at both 24% and 16% FM levels by using fish oil (FO), an FA‐optimized blend (BO1) designed on the base of EFA requirements of largemouth bass, and BO2 consisting of FO and soybean oil at 2:3 rich in n‐6 PUFA as dietary lipids, respectively. After a 10‐week feeding trial with the six diets, growth performance, lipid metabolism, protein synthesis, and antioxidant capacity among the three dietary groups at identical FM level were compared, and the influence of dietary FA profiles on FM substitution efficacy and the primary mechanisms were investigated. The findings will provide novel insights into the interplay between FA composition and FM substitution, offering practical strategies for utilizing FA‐optimized lipid sources to enhance FM substitution in largemouth bass and other teleost cultivation.

2. Materials and Methods

2.1. Ethical Statement

The experimental procedure was approved by the Ethics of the Institutional Animal Care and Use Committee (IACUC) of Laboratory Animals of South China Agricultural University (Approval number: SYXK‐2019‐0136).

2.2. Experimental Diets

Using FO, blend oil 1 (BO1, consisting of FO, olive oil, palm oil, and soybean oil at 4.0:2.0:2.5:1.5, designed on the base of the demand of largemouth bass for EFA) [22] and blend oil 2 (BO2, consisting of FO and soybean oil at 2:3 commonly used in feed companies) as the dietary lipid sources. And using FM, basic protein (consisting of soy protein concentrate, chicken powder, and corn protein powder at 1.2:1.5:1.6) and compound protein (consisting of meat and bone meal and fermented soybean meal at 2:1) as the dietary protein sources. Six isonitrogenous (50% crude protein) and isolipid (9% crude lipid) diets were formulated using a 2 × 3 factorial design: two FM levels (24% and 16%) × three lipid sources (FO, BO1, and BO2). FM was replaced by compound protein. Diets were labeled 24FO, 24BO1, 24BO2, 16FO, 16BO1, and 16BO2. Ingredients were ground (60 mesh), mixed, extruded (2.5 mm pellets; Yanggong Machinery, Beijing), air‐dried, vacuum‐packed, and stored at −20°C. Formulation, nutrient composition, FA profiles, and amino acid profiles are detailed in Tables 1–3, respectively.

Table 1.

Feed formula and nutritional composition (% dry weight).

Component	Diets
	24FO	24BO1	24BO2	16FO	16BO1	16BO2
Fish meal	24.00	24.00	24.00	16.00	16.00	16.00
Basic proteina	43.00	43.00	43.00	43.00	43.00	43.00
Compound proteinb	—	—	—	10.50	10.50	10.50
Fish oil	8.00	—	—	8.00	—	—
Blend oil 1	—	8.00	—	—	8.00	—
Blend oil 2	—	—	8.00	—	—	8.00
Flour	13.17	13.17	13.17	10.50	10.50	10.50
ɑ‐Starch	5.00	5.00	5.00	5.00	5.00	5.00
Beer yeast powder	3.00	3.00	3.00	3.00	3.00	3.00
DL methionine	0.41	0.41	0.41	0.5	0.5	0.5
L‐lysine	0.42	0.42	0.42	0.5	0.5	0.5
Micronutrientsc	3.00	3.00	3.00	3.00	3.00	3.00
Total	100	100	100	100	100	100
Proximate composition	—	—	—	—	—	—
Moisture (%)	12.65	11.67	11.83	10.86	12.19	11.12
Crude protein (%)	50.64	50.68	50.54	50.78	50.71	50.13
Crude lipid (%)	9.31	9.35	9.34	9.38	9.18	9.32
Ash (%)	9.35	9.33	9.54	10.81	11.43	10.77

^aBasic protein is consisted of soy protein concentrate, chicken powder, and corn protein powder at 1.2:1.5:1.6.

^bCompound protein is consisted of meat and bone meal, fermented soybean meal at 2:1.

^cMicronutrients are composed of 1% vitamin premix, 1% mineral premix, 0.5% choline chloride, and 0.5% calcium dihydrogen phosphate. The vitamin premix (per kilogram of premix) contained vitamins A 900,000 IU, B1 320 mg, B2 1090 mg, B6 360 mg, B12 8 mg, D3 200,000 IU, E 3000 IU, and K3 220 mg; calcium pantothenate 80 mg, nicotinamide 780 mg, folic acid 165 mg, and inositol 8 g. The mineral premix (per kilogram of premix) contained K 100 g, Mg 30 g, Fe 8 g, Mo 1 g, Zn 30 g, Cu 3 g, Mn 2 g, Co 1 g, I 500 mg, Se 40 mg, NaCl 100 mg, and zeolite powder 5 g. Blend oil 1 (BO1) consisting of fish oil, olive oil, palm oil, and soybean oil at 4.0:2.0:2.5:1.5; BO2 consisting of fish oil and soybean oil at 2:3.

Table 3.

Amino acid composition of experimental feed (g/100 g dry weight).

Amino acid	Diets
	24FO	24BO1	24BO2	16FO	16BO1	16BO2
Histidine	1.29	1.13	1.33	1.18	1.05	1.23
Threonine	1.93	1.90	1.88	1.85	1.81	1.83
Arginine	2.82	3.28	2.79	2.99	3.42	3.21
Valine	2.03	2.11	1.97	1.97	2.02	1.97
Methionine	0.96	1.03	0.94	0.88	0.93	0.89
Phenylalanine	2.73	2.68	2.68	2.68	2.67	2.75
Isoleucine	2.05	2.06	1.99	1.97	1.99	1.96
Leucine	4.75	4.71	4.69	4.66	4.54	4.64
Lysine	2.89	2.82	2.85	2.78	2.73	2.84
EAA	21.47	21.73	21.12	20.95	21.21	21.32
Aspartic acid	4.26	4.12	4.20	4.18	4.04	4.14
Serine	2.28	2.21	2.26	2.27	2.19	2.24
Glutamic acid	9.64	9.63	9.55	9.67	9.62	9.70
Glycine	2.87	2.73	2.86	3.06	2.86	3.02
Alanine	3.44	3.77	3.38	3.38	3.71	3.62
Tyrosine	1.69	1.70	1.67	1.67	1.65	1.67
NEAA	24.17	24.16	23.91	24.22	24.07	24.39
TAA	45.64	45.89	45.04	45.17	45.28	45.71

Abbreviations: EAA, essential amino acids; NEAA, no essential amino acids.

Table 2.

Fatty acid composition of experimental feed (mg/g).

Fatty acid	Diets
	24FO	24BO1	24BO2	16FO	16BO1	16BO2
14:0	4.67	2.98	3.63	4.10	2.87	3.16
16:0	21.86	26.62	20.40	20.23	28.09	18.98
16:1	5.82	3.62	3.76	5.27	3.57	3.38
18:0	5.05	5.51	5.38	5.12	6.12	5.41
18:1n‐9	22.53	30.95	25.51	20.59	33.86	24.60
18:2n‐6	12.28	17.60	30.36	11.45	18.04	28.34
20:1n‐9	0.76	1.56	0.40	0.66	1.13	0.35
18:3n‐3	1.83	1.94	3.52	1.68	2.54	3.21
20:4n‐6	0.92	0.55	0.52	0.83	0.56	0.49
20:5n‐3	6.71	3.84	3.98	5.57	3.30	3.02
22:6n‐3	9.61	5.48	5.58	8.08	4.87	4.41
SFA	31.98	35.73	31.63	29.45	37.63	29.52
MUFA	29.10	36.13	29.68	26.52	38.56	28.33
n‐6 PUFA	13.20	18.16	30.36	12.28	18.60	28.34
n‐3 PUFA	19.45	11.93	13.83	16.21	11.29	11.24
n‐3/n‐6	1.47	0.66	0.46	1.32	0.61	0.40
n‐3 LC‐PUFA	17.61	10.00	10.31	14.53	8.75	8.03

Note: SFA, saturated fatty acids, including 14:0, 16:0, and 18:0; MUFA, monounsaturated fatty acids, including 16:1, 18:1n−9, and 20:1n−9; n−6 PUFA, n−6 polyunsaturated fatty acids, including 18:2n−6 and 20:4n−6; n‐3 PUFA, n−3 polyunsaturated fatty acids, including 18:3n−3, 20:3n−3, 20:5n−3, and 22:6n−3; n−3/n−6, n−3 PUFA/n−6 PUFA; n−3 LC‐PUFA, n−3 long‐chain PUFA, including 20:3n−3, 20:5n−3, and 22:6n−3.

2.3. Animals and Feeding

Juvenile largemouth bass were purchased from a local hatchery (Foshan, China) and acclimated in floating cages within reservoirs (Zengcheng Experimental Base, South China Agricultural University) for 2 weeks. During acclimation, fish were fed a mixture of all six experimental diets in equal quantities. Subsequently, 720 healthy fish of similar size (average weight: about 12.50 g) were randomly allocated to 18 floating cages (1.5 m × 1.5 m × 2 m), with triplicates per dietary treatment, and cultivated for 10 weeks. Fish were hand‐fed their respective diets to visual satiation twice daily (08:00 and 17:00). Water temperature and dissolved oxygen were maintained at 28–34°C and above 6 mg/L, respectively.

2.4. Sample Collection

Fish were fasted for 24 h prior to sampling. Fish were anesthetized with 0.01% 2‐phenoxyethanol. Then, two fish from each cage were dissected to obtain visceral and liver tissues for calculating VSI, HSI, and condition factor (CF). And two fish from each cage were used for blood, liver, and muscle sample collection. Blood samples were stored at 4°C overnight and then centrifuged (3500 × g at 4°C for 10 min) to obtain serum. All serum, liver, and muscle samples were immediately frozen in liquid nitrogen and stored at −80°C until analysis [22].

2.5. Proximate Composition of Muscle, FA Analysis, and Amino Acid Analysis

Proximate analysis was conducted following established methods [23]. Moisture content was determined by oven‐drying samples at 105°C to constant weight. Crude fat content was measured via Soxhlet extraction (ST 255 Soxtec, FOSS, China). Crude protein was quantified using the Kjeldahl method with an automatic analyzer (KDN‐19K, Shanghai) after acid digestion. Ash content was determined by incinerating samples in a muffle furnace at 550°C for 4 h.

The method for FA determination was following the method previously described [23]. In brief, total lipids were extracted from each sample with chloroform/methanol (2:1 v/v), then converted to FA methyl esters using 0.5 mol KOH‐methanol and 15% BF3, and finally measured by gas chromatography (GC; Model 7890B, Agilent, USA). The diet FA contents were determined using the 13:0 (mg/mL) as an internal standard, and results were expressed as absolute concentration (mg/g).

Amino acid content in feed and muscle was determined using an amino acid analyzer (Hitachi L‐8900) after hydrolysis with 6 M HCl at 110°C for 22 h under nitrogen. Serum amino acids were analyzed by mixing serum with 5% sulfosalicylic acid, incubating at 4°C overnight, centrifuging at 12,000 rpm for 10 min, and filtering through a 0.22 μm membrane prior to analysis [17].

2.6. Analysis of Biochemical Indicators and Enzyme Activities

Serum and liver biochemical parameters—including triglycerides (TGs), cholesterol (TCHO), high‐density lipoprotein TCHO (HDL‐C), low‐density lipoprotein TCHO (LDL‐C), nonesterified FAs (NEFAs), albumin (ALB), blood urea nitrogen (BUN), total amino acids (TAAs), malondialdehyde (MDA), acid phosphatase (ACP), alkaline phosphatase (AKP), alanine transaminase (ALT), aspartate transaminase (AST), total antioxidant capacity (T‐AOC), catalase (CAT), and superoxide dismutase (SOD)—were measured using commercial kits (Nanjing Jiancheng Bioengineering Institute) according to manufacturer protocols. All analyses were performed with a microplate reader (BioTek Instruments Inc., USA).

2.7. Real‐Time Quantitative PCR Analysis

Total RNA was extracted from muscle using TRIzol (Takara, Japan). cDNA was synthesized from 1000 ng DNase‐treated RNA with PrimeScript RT reagent (Takara). Real‐time PCR was performed using SYBR qPCR Mix (TOYOBO) on a CFX96 system (Bio‐Rad). Reaction mixtures contained 1 μL cDNA, 0.4 μL primers (10 μM), 3.2 μL ddH2O, and 5 μL mix. Cycling conditions were 95°C for 30 s, followed by 40 cycles of 95°C for 10 s and 60°C for 30 s [22]. Primers are listed in Table 4. EF1‐α served as the reference gene, and relative expression was calculated using the 2^−ΔΔCT method [24].

Table 4.

Nucleotide sequences of the primers used to assay gene expressions using real‐time PCR.

Gene name	Forward primer (5´–3´)	Reverse primer (3´–5´)	Accession number
pi3k	GGATGAGACACAGAAGATGCGA	CCTCAGGTTTCCCAGTTGGT	XM_038715187.1
akt	AGCGAGATTCTATGGTGC	TGCCGTAGTCATTGTCCT	XM_038692874.1
mtor	TCAGGACCTCTTCTCATTGGC	CCTCTCCCACCATGTTTCTCT	XM_038702468.1
s6k1	AGGCTTATTCCTTCTGCG	ATGGTCTCATTGCGGTCT	XM_038698962.1
4ebp	ATATCCGATACAGGCGCGTT	TGGTCTTCTGGCAGTCAGTG	XM_038703877.1
eif4g	CAATCTTATCCACGAGTCTATC	GAGGCATTCCAAATCTTCAC	XM_038701911.1
pparα	CCACCGCAATGGTCGATATG	TGCTGTTGATGGACTGGGAAA	XM_038705496.1
atgl	CCATGATGCTCCCCTACACT	GGCAGATACACTTCGGGAAA	XM_038705351.1
lpl	TTCCTCGACCCTCTGAAAGA	GGAGTCAAGTTTGCCAGGAA	XM_038715977.1
hsl	CCTGGAGGAGTGTTTCTACG	CATAAGTGTTAGCAGACGGGA	XM_038710963.1
cpt1a	ACTTGTGCCTTTGTTCGTGC	GGTGAGTCTTCTCCCAGGTATT	XM_038705335.1
acsl4	GTAAAGACAAGCCAAACCCAAG	GGTCGTTGTTATCATTCCCGTA	XM_038699900.1
fas	ATGGTCGTGTCCTGTATCCG	CTACGGAATGAGTCAAGGGC	XM_038735140.1
acc	ATCCCTCTTTGCCACTGTTG	GAGGTGATGTTGCTCGCATA	XM_038709732.1
srebp	GCTCTGAGTGCCGTGAA	GCTGCGAATAGCCCAATC	XM_038699585.1
pparγ	CCTGTGAGGGCTGTAAGGGTTT	TTGTTGCGGGACTTCTTGTGA	XM_046076317.1
dgat	CACGCCTCTTCTTGGAGAAC	AATGGTACCCACAGCCAGAC	XM_038724648.1
lpin	ACCTTTCTACGCTGCTTTTG	CTTCTTCCTGGTTGGTGCTAT	XM_038701581.1
fatp1	GACCGCATTGGAGACACTTT	TGGGGAGGTAGTTTTTGACG	XM_038702199.1
fatp6	CACCTCCCTCAAATCCCCCA	GCCCCAAATGCCCAAAAAC	XM_038733305.1
fabp3	CCTCAAGGAGAGCCAGAAA	CACCGTCCACCGAGATAAT	XM_038725022.1
mtp	CATCCTCAGTAGCAACCCCTC	GCAGACGATGACCCGACTT	XM_046048380.1
ef1α	TGCTGCTGGTGTTGGTGAGTT	TTCTGGCTGTAAGGGGGCTC	XM_038724777.1

Note: akt, protein kinase B; s6k1, ribosomal protein S6 kinase 1; 4ebp, eukaryotic translation initiation factor 4E‐binding protein 1; eif4g, eukaryotic translation initiation factor 4G; hsl, hormone‐sensitive triglyceride lipase; acsl4, long‐chain acyl‐coenzyme A synthase 4.

Abbreviations: acc, acetyl‐CoA carboxylase; atgl, adipose triglyceride lipase; cpt1a, carnitine palmitoyltransferase 1A; dgat, diacylglycerolacyltransferase; ef1α, elongation factor 1 alpha; fabp3, fatty acid‐binding protein 3; fas, fatty acid synthase; fatp1, fatty acid transport protein 1; fatp6, fatty acid transport protein 6; lpin, lipoprotein; lpl, lipoprotein lipase; mtor, mammalian target of rapamycin; mtp, mitochondrial trifunctional protein; pi3k, phosphatidylinositol3 kinase; pparα, peroxisome proliferator‐activated receptor alpha; pparγ, peroxisome proliferator‐activated receptor γ; srebp, sterol‐regulatory element binding protein.

2.8. Statistical Analysis

The data were analyzed using SPSS 24.0 and expressed as the mean ± standard error of the mean (SEM). Two‐way ANOVA was used to analyze the effects of FM level, lipid source, and their interaction, and statistically significant difference was determined at p  < 0.05. One‐way ANOVA was also used to compare the means of three lipid sources within different FM levels, and Tukey’s test was selected when p  < 0.05. And T‐test was used to compare the means of two FM levels at each lipid sources [25].

3. Results

3.1. Growth Performance and Morphological Indices

As shown in Table 5, at 24% FM level, no significant differences in the FBW, WGR, and SGR were observed among the three lipid source groups (p > 0.05). However, at 16% FM level, the 16BO1 group exhibited significantly higher FBW, WGR, and SGR compared to 16BO2 group (p < 0.05), and achieving performance comparable to 16FO group (p > 0.05). 16BO1 group showed a significantly higher CF compared to 16FO group (p < 0.05). In addition, BO1‐fed fish displayed reduced HSI and VSI compared to BO2‐fed fish at both FM levels (p < 0.05). At same lipid source, the 16FO group showed significantly greater WG and SGR but reduced HSI and VSI compared to 24FO (p < 0.05). Concurrently, the 16BO1 group displayed elevated final body weight (FBW), WG, SGR, daily feeding rate (DFR), and CF versus 24BO1 (p < 0.05), whereas 16BO2 exhibited lower FCR and VSI relative to 24BO2 (p < 0.05). The significant interactive effects on FM levels and lipid sources in diets were observed on the FBW, WGR, HSI, and VSI (p < 0.05).

Table 5.

Growth performance, feed utilization, and morphometric parameters of juvenile largemouth bass fed different diets for 10 weeks.

Items	Fish meal level		Lipid source			Two‐way ANOVA
			FO	BO1	BO2	Fish meal level	Lipid source	Fish meal level ∗Lipid source
IBW	24%	—	12.59 ± 0.13	12.42 ± 0.08	12.59 ± 0.13	ns	ns	ns
	—	16%	12.51 ± 0.14	12.51 ± 0.11	12.50 ± 0.08	—	—	—
FBW	24%	—	98.28 ± 2.68	94.97 ± 2.38	92.91 ± 1.50	∗∗	∗∗	∗
	—	16%	104.2 ± 1.42ab	109.58 ± 2.55a  ∗	92.64 ± 3.63b	—	—	—
WG1	24%	—	680.33 ± 16.21	654.47 ± 22.67	643.2 ± 20.30	∗∗	∗∗	∗
	—	16%	734.06 ± 6.23ab  ∗	776.43 ± 26.59a  ∗	645.86 ± 25.93b	—	—	—
SGR2	24%	—	2.93 ± 0.03	2.89 ± 0.04	2.86 ± 0.04	∗∗	∗	ns
	—	16%	3.03 ± 0.01ab  ∗	3.10 ± 0.04a  ∗	2.87 ± 0.05b	—	—	—
SR3	24%	—	93.33 ± 2.20	95.00 ± 3.82	93.33 ± 1.67	ns	ns	ns
	—	16%	93.33 ± 2.20	97.50 ± 1.44	90.00 ± 2.89	—	—	—
FCR4	24%	—	0.79 ± 0.02	0.79 ± 0.01	0.78 ± 0.02	∗	ns	ns
	—	16%	0.82 ± 0.01	0.79 ± 0.01	0.83 ± 0.01 ∗	—	—	—
DFR5	24%	—	1.98 ± 0.03	1.93 ± 0.01	1.90 ± 0.05	∗∗	ns	ns
	—	16%	2.06 ± 0.03	2.03 ± 0.03 ∗	1.99 ± 0.02	—	—	—
HSI6	24%	—	3.56 ± 0.21ab	3.17 ± 0.17b	3.90 ± 0.14a	∗	∗∗∗	∗∗
	—	16%	2.60 ± 0.14b  ∗	2.97 ± 0.15b	4.19 ± 0.15a	—	—	—
VSI7	24%	—	10.12 ± 0.25a	9.09 ± 0.26b	10.23 ± 0.26a	ns	∗∗∗	∗
	—	16%	9.43 ± 0.16b  ∗	9.50 ± 0.30b	11.04 ± 0.23a  ∗	—	—	—
CF8	24%	—	2.34 ± 0.05	2.36 ± 0.04	2.50 ± 0.11	ns	∗	ns
	—	16%	2.36 ± 0.06b	2.62 ± 0.04a  ∗  ∗	2.52 ± 0.06ab	—	—	—

Note: Values are mean ± standard error, where n = 3 for IBW, FBW, WG, SGR, SR, FCR, and DFR, and n = 6 for HSI, VSI, and CF. Values in each row without sharing a common superscript lowercase letters indicate significantly different (p < 0.05), those sharing a common superscript lowercase letter or without superscript letters indicate no difference (p > 0.05). For values between 24% and 16% FM levels of each lipid source,  ^∗ indicates a significant difference (p < 0.05). For Two‐way ANOVA, ns: not significant;  ^∗ p < 0.05;  ^∗∗ p < 0.01;  ^∗∗∗ p < 0.001. FM, Fish meal; IBW, initial mean body weight, g fish⁻¹; FBW, final mean body weight, g fish⁻¹.

¹Weight gain (WG, %) = 100 × (Final mean body weight [g] − initial weight [g])/initial weight (g).

²Specific growth ratio (SGR, %/d) = 100 × (Ln final mean body weight – Ln initial mean body weight)/feeding days.

³Survival rate (SR, %) = 100 × (Final number of fish)/(initial number of fish).

⁴Feed conversion ratio (FCR) = Feed consumed (g)/(wet weight [g] − initial weight [g]).

⁵Daily feeding rate (DFR, %) = 100 × Feed consumed (g)/([final mean body weight (g) + initial weight (g)]/2)/experimental days.

⁶Hepatosomatic index (HSI, %) = 100 × Liver wet weight (g)/final body weight (g).

⁷Viscerosomatic index (VSI, %) = 100 × Viscera wet weight (g)/final body weight (g).

⁸Condition factor (CF) = 100 × Final body weight (g)/body length (cm)³.

3.2. Muscle and Liver Composition

As shown in Table 6, at 24% FM level, the 24FO group showed significantly higher contents of moisture, crude protein, and ash while lower content of crude lipid in fish muscle compared to 24BO1 and 24BO2 groups (p < 0.05). The hepatic crude lipid content of fish in 24FO and 24BO1 groups was significantly lower than that in 24BO2 group (p < 0.05). At 16% FM level, muscle ash content in the 16BO1 group was significantly higher than that in 16FO (p < 0.05). At same lipid source, the 24FO group showed significantly increased muscle moisture and ash levels but reduced crude lipid content relative to 16FO (p < 0.05). Moreover, muscle moisture was elevated in 16BO1 compared to 24BO1 (p < 0.05), whereas hepatic crude lipid decreased in 16BO2 versus 24BO2 (p < 0.05). The significant interactive effects on FM levels and lipid sources in diets were observed on muscle moisture, muscle crude lipid, and hepatic crude lipid (p < 0.05).

Table 6.

Muscle and liver composition of juvenile largemouth bass fed different diets for 10 weeks (%).

Items	Fish meal level		Lipid source			Two‐way ANOVA
			FO	BO1	BO2	Fish meal level	Lipid source	Fish meal level ∗Lipid source
Muscle
Moisture	24%	—	78.97 ± 0.13a  ∗∗	76.40 ± 0.40c	77.97 ± 0.18b	ns	∗∗∗	∗∗∗
	—	16%	77.84 ± 0.20	77.64 ± 0.08 ∗	78.19 ± 0.29	—	—	—
Crude protein	24%	—	86.64 ± 1.10a	82.15 ± 0.73b	83.14 ± 0.84b	ns	∗	ns
	—	16%	84.08 ± 1.13	83.58 ± 0.51	82.17 ± 1.44	—	—	—
Crude lipid	24%	—	5.26 ± 0.51b	11.42 ± 0.84a	9.93 ± 1.00a	ns	∗∗	∗∗
	—	16%	9.56 ± 0.66 ∗∗	9.88 ± 0.44	9.07 ± 0.96	—	—	—
Ash	24%	—	6.37 ± 0.15a  ∗∗	5.30 ± 0.13c	5.79 ± 0.10b	∗∗	∗∗∗	ns
	—	16%	5.72 ± 0.13a	5.19 ± 0.12b	5.49 ± 0.10a,b	—	—	—
Liver
Moisture	24%	—	73.54 ± 1.08	72.60 ± 1.56	73.11 ± 2.16	ns	ns	ns
	—	16%	72.00 ± 1.07	69.85 ± 1.76	74.93 ± 1.87	—	—	—
Crude lipid	24%	—	10.46 ± 0.56b	13.13 ± 1.49b	17.45 ± 0.45a ∗	ns	∗∗∗	∗
	—	16%	11.83 ± 1.04	13.71 ± 0.49	14.07 ± 0.95	—	—	—

Note: Data are expressed as mean ± standard error (n = 6). Significant differences (p < 0.05) among three lipid sources within different fish meal levels are indicated by different letters.  ^∗Indicates a significant difference (p < 0.05) between the fish meal levels at each lipid sources; ns, not significant;  ^∗ p < 0.05;  ^∗∗ p < 0.01;  ^∗∗∗ p < 0.001.

3.3. Serum and Hepatic Lipid Metabolites

At 24% FM level, no significant difference in fish serum lipid metabolites was found among three lipid source groups (p > 0.05) (Figure 1). But at 16% FM level, the 16BO1 group exhibited significantly lower serum TG and NEFA levels compared to the 16BO2 group (p < 0.05), aligning with 16FO group (p > 0.05). At same lipid source, the NEFA level of 16BO2 group was significantly higher than 24BO2 group (p < 0.05). A significant interaction between FM levels and lipid sources was observed for serum NEFA (p < 0.05).

Figure 1.

(a–e) Serum triglycerides (TG), cholesterol (TCHO), high‐density lipoprotein cholesterol (HDL‐C), low‐density lipoprotein cholesterol (LDL‐C), non‐esterified fatty acids (NEFA) levels, and (f–i) hepatic TG, TCHO, LDL, NEFA levels of juvenile largemouth bass fed different diets for 10 weeks. Data are expressed as mean ± standard error (n = 6). Significant differences (p < 0.05) among three lipid sources within different fish meal levels are indicated by different letters.  ^∗Indicates a significant difference (p < 0.05) between the fish meal levels at each lipid source; ns, not significant;  ^∗ p < 0.05;  ^∗∗ p < 0.01;  ^∗∗∗ p < 0.001. LS, lipid source.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

In the liver, at 24% FM level, hepatic LDL‐C levels were significantly reduced in both the 24FO and 24BO1 groups compared with the 24BO2 group (p < 0.05). And the hepatic TG level in 24BO1 group showed no significant difference compared to the 24BO2 group but was significantly higher than 24FO group (p < 0.05). At 16% FM level, both 16FO and 16BO1 groups exhibited significantly lower hepatic TG and LDL level relative to 16BO2 (p < 0.05). At same lipid source, the hepatic TG content of 16FO group was significantly higher than 24FO group (p < 0.05). No significant interaction between FM levels and lipid sources was observed in fish hepatic lipid metabolites (p > 0.05).

3.4. Hepatic and Muscle FA Composition

As shown in Table 7, BO1‐fed fish exhibited significantly elevated muscle 18:1n‐9 content relative to both BO2‐fed and FO‐fed fish at all FM levels (p < 0.05). Conversely, muscle 18:2n‐6 and n‐6 PUFA content was reduced in BO1‐fed fish compared to BO2‐fed fish but elevated versus FO‐fed fish at all FM levels (p < 0.05). MUFA content in the BO1‐fed group was significantly higher than in the BO2‐fed group (p < 0.05) but showed no difference compared to the FO‐fed group at all FM levels (p > 0.05). Critically, n‐3 PUFA profiles (20:5n‐3, 22:6n‐3, and n‐3 LC‐PUFA) and the n‐3/n‐6 ratio exhibited no divergence between BO1‐ and BO2‐fed groups (p > 0.05) but were markedly reduced relative to the FO‐fed group at all FM levels (p < 0.05). At same lipid source, 20:5n‐3, 22:5n‐3, and 22:6n‐3 content were markedly increased in the 24FO group relative to 16FO (p < 0.05). Correspondingly, 22:5n‐3 content showed significant elevation in both the 24BO1 group versus 16BO1 and the 24BO2 group versus 16BO2 (p < 0.05). A significant interaction between FM levels and lipid sources was observed in 22:5n‐3 content in fish muscle (p < 0.05).

Table 7.

Muscle fatty acid composition of juvenile largemouth bass fed different diets (%).

Main fatty acids	Fish meal level		Lipid source			Two‐way ANOVA
			FO	BO1	BO2	Fish meal level	Lipid source	Fish meal level ∗Lipid source
14 : 0	24%	—	2.87 ± 0.44a	1.54 ± 0.16b	1.98 ± 0.26a,b	ns	∗∗∗	ns
	—	16%	2.62 ± 0.27a	1.54 ± 0.14b	1.78 ± 0.18b	—	—	—
16 : 0	24%	—	23.02 ± 0.57	23.26 ± 0.17	21.67 ± 0.46	ns	∗∗∗	ns
	—	16%	23.76 ± 0.7a	23.85 ± 0.18a ∗	20.83 ± 0.34b	—	—	—
16.1	24%	—	5.34 ± 0.65a	2.93 ± 0.57b	3.05 ± 0.36b	ns	∗∗∗	ns
	—	16%	4.93 ± 0.38a	3.6 ± 0.24a,b	2.75 ± 0.53b	—	—	—
18 : 0	24%	—	4.58 ± 0.40	5.23 ± 0.39	5.49 ± 0.42	ns	ns	ns
	—	16%	4.94 ± 0.45	5.03 ± 0.33	5.36 ± 0.32	—	—	—
18.1	24%	—	22.62 ± 0.9b	29.04 ± 0.57a	22.37 ± 1.16b	ns	∗∗∗	ns
	—	16%	24.72 ± 0.97b	29.66 ± 0.96a	23.89 ± 0.66b	—	—	—
18.2 n‐6	24%	—	10.63 ± 0.53c	13.66 ± 0.27b	22.62 ± 0.83a	ns	∗∗∗	ns
	—	16%	10.36 ± 0.24c	14.00 ± 0.51b	21.38 ± 0.8a	—	—	—
18.3 n‐6	24%	—	0.89 ± 0.18b	1.15 ± 0.08a,b	1.48 ± 0.16a	ns	∗∗∗	ns
	—	16%	0.85 ± 0.06c	1.19 ± 0.06b	1.68 ± 0.09a	—	—	—
18.3 n‐3	24%	—	1.01 ± 0.11	1.01 ± 0.07	1.12 ± 0.19	ns	ns	ns
	—	16%	0.92 ± 0.03	1.02 ± 0.08	0.95 ± 0.04	—	—	—
20.3 n‐3	24%	—	2.22 ± 0.3a	1.12 ± 0.24b	1.68 ± 0.1a,b	ns	∗∗	ns
	—	16%	2.3 ± 0.24a	1.44 ± 0.34a,b	1.32 ± 0.16b	—	—	—
20.5 n‐3	24%	—	3.17 ± 0.14a ∗	1.7 ± 0.14b	1.54 ± 0.1b	∗∗	∗∗∗	ns
	—	16%	2.68 ± 0.12a	1.47 ± 0.13b	1.29 ± 0.13b	—	—	—
22.5 n‐3	24%	—	2.64 ± 0.12a  ∗∗	1.49 ± 0.07b  ∗∗	1.51 ± 0.05b ∗	∗∗∗	∗∗∗	∗∗
	—	16%	1.92 ± 0.04a	1.18 ± 0.06b	1.23 ± 0.02b	—	—	—
22.6 n‐3	24%	—	20.24 ± 1.28a ∗	13.46 ± 0.93b	12.98 ± 0.68b	∗	∗∗∗	ns
	—	16%	16.88 ± 0.83a	12.33 ± 0.81b	12.34 ± 0.99b	—	—	—
SFA	24%	—	30.56 ± 0.63	30.03 ± 0.39	28.73 ± 0.35	ns	∗∗∗	ns
	—	16%	31.32 ± 0.73a	30.52 ± 0.43a	27.72 ± 0.39b	—	—	—
MUFA	24%	—	27.96 ± 1.53a,b	32.01 ± 0.9a	24.61 ± 1.55b	ns	∗∗∗	ns
	—	16%	30.59 ± 1.53a,b	33.4 ± 1.17a	26.34 ± 1.01b	—	—	—
n‐6 PUFA	24%	—	11.07 ± 0.44c	14.81 ± 0.3b	23.36 ± 0.69a	ns	∗∗∗	ns
	—	16%	11.06 ± 0.24c	15.18 ± 0.54b	23.07 ± 0.88a	—	—	—
n‐3 PUFA	24%	—	26.19 ± 1.41a	17.56 ± 0.75b	18.25 ± 0.74b	ns	∗∗∗	ns
	—	16%	24.67 ± 1.27a	17.4 ± 1.24b	17.13 ± 1.28b	—	—	—
n‐3/n‐6	24%	—	2.41 ± 0.22a	1.16 ± 0.06b	0.78 ± 0.05b	ns	∗∗∗	ns
	—	16%	2.25 ± 0.17a	1.16 ± 0.13b	0.76 ± 0.07b	—	—	—
n‐3 LC‐PUFA	24%	—	24.24 ± 1.17a	15.68 ± 0.83b	16.22 ± 0.97b	ns	∗∗∗	ns
	—	16%	21.81 ± 1.05a	14.88 ± 0.93b	14.86 ± 1.12b	—	—	—

Note: Notes are the same with Table 6.

Table 8 demonstrates that BO1‐fed fish had enriched hepatic 16:0, 18:1n‐9, and MUFA levels versus BO2‐fed counterparts at both FM levels (p < 0.05) while showing suppressed 18:2n‐6 and n‐6 PUFA accumulation (p < 0.05). Notably, at 24% FM level, the 24FO group displayed significantly elevated 20:5n‐3, 22:5n‐3, 22:6n‐3, n‐3 PUFA, and n‐3 LC‐PUFA levels relative to 16FO (p < 0.05), contrasting with null effects at 16% FM (p > 0.05). At same lipid source, n‐3 PUFA profiles (22n:5n‐3, 22n:6n‐3, n‐3 PUFA, and n‐3 LC‐PUFA) in 24FO group were significantly higher than 16FO group. The significant interactive effects on FM levels and lipid sources in diets were observed on the 18:2n‐6, 20:3n‐3 content, and the radio of n‐3/n‐6 in fish liver (p < 0.05).

Table 8.

Hepatic fatty acid composition of juvenile largemouth bass fed different diets (%).

Main fatty acids	Fish meal level		Lipid source			Two‐way ANOVA
			FO	BO1	BO2	Fish meal level	Lipid source	Fish meal level ∗Lipid source
14:0	24%	—	1.58 ± 0.16	1.47 ± 0.1	1.38 ± 0.08	ns	∗	ns
	—	16%	2.00 ± 0.24	1.47 ± 0.12	1.44 ± 0.14	—	—	—
16:0	24%	—	20.22 ± 0.7a,b	21.77 ± 0.53a	19.05 ± 0.48b	ns	∗∗∗	ns
	—	16%	20.96 ± 0.71a,b	21.89 ± 0.59a	19.39 ± 0.32b	—	—	—
16.1	24%	—	6.22 ± 0.87	4.91 ± 0.47	5.29 ± 0.72	ns	ns	ns
	—	16%	6.94 ± 0.41	6.36 ± 0.69	4.76 ± 1.02	—	—	—
18:0	24%	—	6.3 ± 0.76	5.97 ± 0.54	5.86 ± 0.91	ns	ns	ns
	—	16%	5.03 ± 0.3	5.36 ± 0.4	6.18 ± 1.08	—	—	—
18.1	24%	—	27.9 ± 2.61b	38 ± 1.55a	27.52 ± 2.38b	ns	∗∗∗	ns
	—	16%	31.53 ± 1.38b	40.72 ± 0.46a	26.6 ± 3.45b	—	—	—
18.2n‐6	24%	—	6.73 ± 0.41c	9.81 ± 0.60b ∗	16.01 ± 0.69a	ns	∗∗∗	∗
	—	16%	7.44 ± 0.40b	7.48 ± 0.51b	16.63 ± 0.84a	—	—	—
18.3n‐6	24%	—	0.48 ± 0.07b	0.78 ± 0.06a	0.83 ± 0.07a	ns	∗	ns
	—	16%	0.75 ± 0.14	0.61 ± 0.06	0.94 ± 0.12	—	—	—
18.3n‐3	24%	—	2.2 ± 0.03a	1.77 ± 0.05b	1.84 ± 0.1b	ns	∗	ns
	—	16%	1.93 ± 0.16	1.78 ± 0.12	1.66 ± 0.12	—	—	—
20.3n‐3	24%	—	2.55 ± 0.14a∗	1.23 ± 0.23b	1.31 ± 0.3b	ns	ns	ns
	—	16%	1.35 ± 0.32	1.18 ± 0.28	2.22 ± 0.55	—	—	—
20.5n‐3	24%	—	1.41 ± 0.08a	0.93 ± 0.13b	0.85 ± 0.12b	ns	∗∗	ns
	—	16%	1.3 ± 0.12a	0.68 ± 0.13b	0.97 ± 0.2a,b	—	—	—
22.5n‐3	24%	—	1.98 ± 0.09a∗	0.87 ± 0.08b	0.9 ± 0.1b	∗	∗∗∗	ns
	—	16%	1.32 ± 0.21	0.66 ± 0.18	0.95 ± 0.16	—	—	—
22.6n‐3	24%	—	13.92 ± 0.72a∗	7.93 ± 1.07b	8.81 ± 0.73b	ns	∗∗∗	ns
	—	16%	10.93 ± 0.55	8.16 ± 0.92	8.57 ± 1.43	—	—	—
SFA	24%	—	27.05 ± 0.66	29.44 ± 0.81	25.96 ± 1.25	ns	∗	ns
	—	16%	27.75 ± 0.87	28.96 ± 0.81	26.73 ± 1.27	—	—	—
MUFA	24%	—	33.65 ± 3.45a,b	43.54 ± 1.69a	31.83 ± 2.83b	ns	∗∗∗	ns
	—	16%	37.96 ± 1.25a,b	47.59 ± 0.84a	30.97 ± 4.49b	—	—	—
n‐6 PUFA	24%	—	7.21 ± 0.43b	10.23 ± 0.64b	15.55 ± 1.29a	ns	∗∗∗	ns
	—	16%	8.15 ± 0.49b	8.8 ± 0.82b	17.57 ± 0.92a	—	—	—
n‐3 PUFA	24%	—	21.61 ± 0.62a  ∗∗	12.10 ± 1.43b	13.28 ± 1.11b	ns	∗∗∗	ns
	—	16%	16.39 ± 1.13	11.01 ± 0.22	13.41 ± 1.84	—	—	—
n‐3/n‐6	24%	—	2.92 ± 0.08a∗	1.3 ± 0.08b	1.05 ± 0.04b	∗	∗∗∗	∗∗∗
	—	16%	2.23 ± 0.19a	1.42 ± 0.09b	1.05 ± 0.14b	—	—	—
n‐3 LC‐PUFA	24%	—	17.31 ± 0.86a∗	10.87 ± 1.2b	11.61 ± 1.02b	ns	∗∗	ns
	—	16%	13.74 ± 0.75	9.49 ± 1.23	12.01 ± 1.59	—	—	—

Note: Notes are the same with Table 6.

3.5. Serum and Liver Protein Metabolites

At 24% FM level, the 24FO group had higher serum ALB level than 24BO1 and 24BO2 groups (p < 0.05) (Figure 2). At 16% FM level, serum ALT and AST activities and BUN content were significantly lower in the 16BO1 group than the 16BO2 group (p < 0.05), whereas TAA content was significantly higher. At same lipid source, the 24FO group exhibited elevated serum AKP activity, AST activity, ALB, and BUN levels relative to the 16FO group (p < 0.05). And the serum BUN content was significantly reduced in the 24BO2 group relative to the 16BO2 group, whereas TAA content was significantly increased (p < 0.05). The significant interactive effects on FM levels and lipid sources in diets were observed on ALT, AST, ALB, BUN, and TAA of fish serum (p < 0.05).

Figure 2.

(a–g) Serum acid phosphatase (ACP), alkaline phosphatase (AKP), alanine transaminase (ALT), aspartate transaminase (AST), albumin (ALB), blood urea nitrogen (BUN), total amino acids (TAA) contents of juvenile largemouth bass fed different diets for 10 weeks. Notes are the same as Figure 1.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

As shown in Figure 3, in the liver, at 24% FM level, no differences in protein metabolites were observed among groups (p > 0.05). At 16% FM level, the 16BO1 group demonstrated elevated hepatic ALT and AST activities than both the 16FO and 16BO2 groups (p < 0.05), while its BUN content was significantly lower than that in the 16BO2 group (p < 0.05). At same lipid source, the 16BO1 group exhibited significantly higher hepatic ALT and AST activities than the 24BO1 group (p < 0.05), while the 16BO2 group showed elevated hepatic TAA content relative to the 24BO2 group (p < 0.05). No significant interaction between FM levels and lipid sources was observed in fish hepatic protein metabolites (p > 0.05).

Figure 3.

(a–e) Hepatic alanine transaminase (ALT), aspartate transaminase (AST), albumin (ALB), blood urea nitrogen (BUN), total amino acids (TAA) contents of juvenile largemouth bass fed different diets for 10 weeks. Notes are the same as Figure 1.

(a)

(b)

(c)

(d)

(e)

3.6. Serum Free Amino Acid

As shown in Table 9, at 24% FM level, the 24BO1 group exhibited significantly lower serum serine and threonine levels than the 24FO group (p < 0.05) but comparable levels to 24BO2 (p > 0.05). Concurrently, serum methionine was elevated in 24BO1 relative to both 24FO and 24BO2 (p < 0.05), whereas glutamic acid content was diminished versus 24BO2 (p < 0.05). At 16% FM level, serum serine, valine, isoleucine, leucine, EAA, and TAA were significantly higher in the 16BO1 group than in 16BO2 (p < 0.05), with threonine and methionine also elevated in 16BO1 relative to both 16FO and 16BO2 (p < 0.05). At the same lipid source, serum arginine content was significantly increased in 16FO versus 24FO (p < 0.05). Notably, glutamic acid levels were augmented in 16BO1 relative to 24BO1 (p < 0.05), while 24BO2 demonstrated elevated serum glutamic acid, histidine, glycine, NEAA, and TAA contents compared to 16BO2 (p < 0.05). The significant interactive effects on FM levels and lipid sources in diets were observed on the glutamate, threonine, arginine, isoleucine, leucine, EAA, NEAA, and TAA content of fish serum (p < 0.05).

Table 9.

Serum free amino acid of juvenile largemouth bass fed different diets for 10 weeks (%).

Amino acids	Fish meal level		Lipid source			Two‐way ANOVA
			FO	BO1	BO2	Fish meal level	Lipid source	Fish meal level ∗Lipid source
Aspartic acid	24%	—	1.46 ± 0.07	1.26 ± 0.15	1.38 ± 0.12	ns	ns	ns
	—	16%	1.63 ± 0.16	1.43 ± 0.19	1.19 ± 0.23	—	—	—
Glutamic acid	24%	—	4.51 ± 0.23a,b	3.79 ± 0.14b	5.90 ± 0.30a ∗	ns	∗	∗∗∗
	—	16%	5.08 ± 0.12	4.97 ± 0.31 ∗∗	4.45 ± 0.45	—	—	—
Serine	24%	—	0.88 ± 0.14b	1.57 ± 0.15a	1.14 ± 0.08a,b	ns	∗∗∗	ns
	—	16%	1.26 ± 0.17a,b	1.81 ± 0.20a	1.01 ± 0.15b	—	—	—
Histidine	24%	—	2.40 ± 0.31	2.29 ± 0.21	2.24 ± 0.10 ∗	ns	ns	ns
	—	16%	2.41 ± 0.11a	2.36 ± 0.16a	1.80 ± 0.11b	—	—	—
Glycine	24%	—	8.83 ± 1.31	8.51 ± 0.88	8.95 ± 0.44 ∗	ns	ns	ns
	—	16%	10.55 ± 1.39	8.14 ± 0.46	6.37 ± 0.49	—	—	—
Threonine	24%	—	1.89 ± 0.30b	3.88 ± 0.16a	3.14 ± 0.21a,b  ∗	ns	∗∗∗	∗
	—	16%	2.58 ± 0.28b	3.72 ± 0.29a	2.45 ± 0.18b	—	—	—
Arginine	24%	—	15.05 ± 1.51	17.74 ± 1.12	18.42 ± 1.47	ns	ns	∗∗
	—	16%	21.33 ± 1.24a  ∗∗	19.04 ± 1.06a,b	15.57 ± 0.86b	—	—	—
Alanine	24%	—	11.63 ± 0.86	9.52 ± 0.91	10.61 ± 0.52	ns	ns	ns
	—	16%	9.62 ± 0.92	11.32 ± 0.94	8.45 ± 1.04	—	—	—
Tyrosine	24%	—	3.97 ± 0.21	3.74 ± 0.41	3.50 ± 0.43	ns	ns	ns
	—	16%	3.10 ± 0.36	4.12 ± 0.30	3.01 ± 0.35	—	—	—
Valine	24%	—	3.22 ± 0.30	3.49 ± 0.11	3.57 ± 0.26	ns	ns	ns
	—	16%	3.55 ± 0.13a,b	4.08 ± 0.32a	3.05 ± 0.16b	—	—	—
Methionine	24%	—	1.55 ± 0.10b	2.04 ± 0.13a	1.50 ± 0.07b	ns	∗∗∗	ns
	—	16%	1.57 ± 0.08b	2.16 ± 0.11a	1.63 ± 0.05b	—	—	—
Phenylalanine	24%	—	2.51 ± 0.14	2.91 ± 0.20	2.73 ± 0.31	∗∗	ns	ns
	—	16%	2.4 ± 0.13	2.43 ± 0.29	1.85 ± 0.13	—	—	—
Isoleucine	24%	—	2.13 ± 0.16	2.33 ± 0.07	2.48 ± 0.21	ns	ns	∗
	—	16%	2.32 ± 0.05a,b	2.78 ± 0.23a	2.02 ± 0.13b	—	—	—
Leucine	24%	—	3.13 ± 0.24	3.46 ± 0.11	3.38 ± 0.09	ns	∗	∗
	—	16%	3.50 ± 0.12a,b	4.05 ± 0.27a	3.02 ± 0.17b	—	—	—
lysine	24%	—	1.23 ± 0.3	0.38 ± 0.11	0.47 ± 0.28	ns	ns	ns
	—	16%	0.40 ± 0.26	0.49 ± 0.16	0.32 ± 0.03	—	—	—
EAA	24%	—	34.11 ± 2.83	39.49 ± 2.09	38.66 ± 1.98	ns	∗	∗
	—	16%	40.06 ± 1.26a	42.35 ± 0.96a	32.24 ± 1.64b	—	—	—
NEAA	24%	—	31.28 ± 2.61	28.39 ± 1.23	35.16 ± 3.33 ∗	ns	ns	∗
	—	16%	37.53 ± 3.78	36.64 ± 5.32	24.48 ± 1.52	—	—	—
TAA	24%	—	65.38 ± 5.32	67.88 ± 2.64	73.83 ± 3.91 ∗	ns	ns	∗∗
	—	16%	77.59 ± 4.78a	78.98 ± 5.18a	56.71 ± 2.53b	—	—	—

Note: Notes are the same with Table 6.

3.7. Serum and Liver Antioxidative Enzymes

At 24% FM level, serum T‐AOC content was significantly lower in the 24BO2 group compared to the other two groups (p < 0.05; Figure 4), while serum MDA content was elevated in the 24FO group (p < 0.05). Hepatic SOD activity was significantly higher in the 24BO2 group than in the 24FO group (p < 0.05). At 16% FM level, the 16BO1 group exhibited higher serum and hepatic T‐AOC content along with elevated hepatic CAT activity relative to the 16BO2 group (p < 0.05). Moreover, serum SOD activity and serum/hepatic MDA content were significantly elevated in the 16FO group compared to the 16BO1 group (p < 0.05). At same lipid source, serum SOD activity was significantly elevated in the 16FO group versus the 24FO group (p < 0.05), whereas the 16BO2 group demonstrated higher serum T‐AOC content relative to the 24BO2 group (p < 0.05). Concomitantly, the 16BO1 group exhibited elevated serum T‐AOC levels with enhanced hepatic CAT and SOD activities compared to 24BO1 (p < 0.05). The significant interactive effects on FM levels and lipid sources in diets were observed on the serum T‐AOC and MDA content and the serum SOD activities and the hepatic CAT activities (p < 0.05).

Figure 4.

(a–d) Serum total antioxidant capacity (T‐AOC), catalase (CAT), superoxide dismutase (SOD), malondialdehyde (MDA) and (e–h) hepatic T‐AOC, CAT, SOD, MDA contents of juvenile largemouth bass fed different diets for 10 weeks. Notes are the same as Figure 1.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

3.8. Gene Expression Related to Muscle Lipid and Protein Metabolism

At 24% FM level, muscle dgat transcript levels peaked in the 24FO group, whereas acsl4 and atgl expression were minimized versus 24BO1/24BO2 group (p < 0.05; Figures 5–8). At 16% FM level, the 16BO1 group displayed significant upregulation of atgl, acsl4, pparα, and akt compared to the 16BO2 group (p < 0.05) with concomitant elevation of dgat and eif4g expression relative to the 16FO and 16BO2 group (p < 0.05). At same lipid source, the 16FO group demonstrated elevated muscle atgl, pparγ, and acsl4 transcripts but reduced dgat expression versus 24FO (p < 0.05). Simultaneously, pparα and acsl4 upregulation was observed in 16BO1 relative to 24BO1 (p < 0.05), whereas 24BO2 exhibited enhanced pi3k, akt, 4ebp, and atgl expression compared to 16BO2 (p < 0.05). The significant interactive effects on FM levels and lipid sources in diets were observed on the mRNA levels of ppara, atgl, ppary, and dgat in fish muscle (p < 0.05).

Figure 5.

Relative expression of genes related protein synthesis of the largemouth bass fed different diets. Notes are the same as Figure 1.

Figure 8.

Relative expression of genes related lipid transport of the largemouth bass fed different diets. Notes are the same as Figure 1.

Figure 6.

Relative expression of genes related lipolysis of the largemouth bass fed different diets. Notes are the same as Figure 1.

Figure 7.

Relative expression of genes related lipid synthesis of the largemouth bass fed different diets. Notes are the same as Figure 1.

4. Discussion

The present findings reveal that the dietary FA profiles can produce obvious effects on FM substitution efficacy in diets of largemouth bass. Under the 24% FM level, no significant differences in growth performance (final weight, weight gain rate (WGR), and specific growth rate (SGR)) were observed among the 24FO, 24BO1, and 24BO2 groups, suggesting that FA sources may exert limited regulatory effects on growth under high FM conditions. This is similar to the research results of previous studies on golden pompano [26]. However, when FM was reduced to 16%, the 16BO1 group exhibited significantly higher growth performance (FBW and WG) than the 16BO2 group, with values comparable to the FO group (16FO). This indicates that BO1 with optimized FA profiles can effectively mitigates growth inhibition caused by reduced FM inclusion, highlighting the critical role of FA optimization in maintaining growth under low‐FM conditions, which aligns with our previous observation in golden pompano [27]. Furthermore, our deeper analysis demonstrated that the blended oil with optimized FA composition (BO1) facilitated this improvement by promoting lipid oxidation for energy, enhancing protein synthesis, and bolstering antioxidant capacity, thereby providing an effective nutritional strategy for sustainable aquaculture.

4.1. Lipid Metabolism Modulation by FA Composition: Alleviating Low‐FM‐Induced Lipid Nutrient Deficits

FM is not only a high‐quality protein source but also a significant lipid source, as it contains intrinsic FO. Reducing FM from 24% to 16% inherently decreases the intake of FO‐derived FAs (e.g., n‐3 LC‐PUFA and MUFA), creating a “lipid nutrient gap” [28]. The BO1 with optimized FA composition fills this gap by balancing EFA supply with metabolically efficient FAs, while the conventional BO2 exacerbates lipid metabolic disorders due to its high n‐6 PUFA content. First, the consistent reduction in HSI and VSI in BO1‐fed fish across both FM levels indicates that BO1 mitigates ectopic lipid deposition—a common issue in low‐FM diets caused by insufficient energy supply from lipids, which triggers excessive lipid storage [29, 30]. The hepatosomatic index (HSI) and viscerosomatic index (VSI) reflect hepatic lipid accumulation and visceral fat deposition, respectively [31, 32]. Elevated HSI and VSI are typically associated with metabolic stress and impaired nutrient utilization in fish [33, 34]. The reduced HSI and VSI in BO1‐fed fish suggest that its FA profile may enhance lipid catabolism, thereby alleviating hepatic steatosis and visceral adiposity. This aligns with studies showing that optimal FA compositions improve energy partitioning and reduce ectopic lipid deposition in teleosts [29, 35].

Additionally, the 16% FM level further highlights BO1’s role in lipid‐energy homeostasis. TG and NEFA are critical indicators of lipid mobilization and energy utilization, with elevated levels often reflecting inefficient lipid catabolism or excessive lipid mobilization due to energy deficits [36, 37]. At this low‐FM level, 16BO1 exhibited serum TG and NEFA levels equivalent to 16FO but significantly lower than 16BO2. This suggests that BO1 compensates for the loss of FM‐derived FO by enhancing lipid catabolism: The upregulation of lipid oxidation‐related genes (ppara, atgl, and acsl4) in 16BO1 confirms that BO1 activates PPARa‐mediated β‐oxidation (PPARa as a master regulator of FA oxidation), promotes TG hydrolysis (ATGL), and accelerates long‐chain FA activation (ACSL4) [38–40]. And the hepatic TG content in BO1‐fed fish was reduced relative to BO2‐fed fish, particularly at the 16% FM level. The reduced TG and NEFA in 16BO1 groups suggest enhanced lipid utilization, which likely spared dietary proteins from being catabolized for energy, thereby improving growth performance after FM replacement. The lower TG and NEFA levels in 16BO1 groups aligns with observations in largemouth bass and large yellow croaker (Larimichthys crocea) [22, 41]. Furthermore, BO1‐fed fish displayed lower hepatic LDL levels compared to BO2‐fed fish at two FM levels. LDL is a major transporter of TCHO and TGs to peripheral tissues, and its elevation is often associated with lipid accumulation and metabolic stress [42]. The lower LDL levels in BO1 groups imply improved lipid homeostasis and align with studies on hybrid grouper [43]. Conversely, BO2‐fed fish exhibited higher hepatic TG and LDL, suggesting lipid overload—a metabolic state that may divert energy toward storage rather than growth, ultimately undermining FM replacement efficacy. This is similar to the research results of previous studies on golden pompano [44]. The FA composition of muscle and liver tissues, which reflects the FA profile of the feed, provided mechanistic insights. BO1‐fed fish demonstrated elevated deposition of SFA and MUFA in both muscle and liver tissues—lipid species that are structurally amenable to efficient β‐oxidation pathways [45, 46]. In teleosts, SFA and MUFA are readily catabolized to generate ATP, thereby sparing amino acids for protein synthesis—a critical adaptation in low‐FM diets [45]. Conversely, BO2‐fed fish exhibited elevated n‐6 PUFA deposition, which is less energetically favorable for β‐oxidation and may promote lipid droplet formation [47]. This inefficiency likely exacerbated energy deficits, as lipid‐derived energy could not fully compensate for reduced dietary protein, mirroring observations in large yellow croaker, where high n‐6 PUFA diets impaired growth despite adequate protein intake [48]. These studies suggest that the superior FM replacement efficacy of BO1 likely arises from its optimized FA profile, which enhanced lipid catabolism and energy provision, thereby reducing reliance on protein degradation.

4.2. Protein‐Sparing Mechanisms Driven by Optimized FA Composition: Synergizing Lipid Energy and Protein Utilization

A core challenge of FM substitution is maintaining protein utilization efficiency when replacing FM with lower‐quality alternative proteins. The present results reveal that BO1’s FA profile enhances protein retention by improving lipid‐derived energy supply—a “protein‐sparing effect” that is particularly critical for low‐FM diets. At 16% FM, 16BO1 group fish maintained serum TAAs at 37.42 μmol/mL—14% higher than 16BO2 group (32.99 μmol/mL) while reducing urea nitrogen by 19% (6.03 vs. 7.45 mmol/L). Elevated TAA reflects enhanced amino acid retention, while reduced BUN, a marker of protein catabolism, indicates diminished nitrogen excretion [49, 50]. The changes in these two parameters collectively suggest that 16BO1 group reduced protein degradation and improved the utilization efficiency of dietary amino acids for anabolism, which aligns with observations in yellow catfish and turbot [51, 52]. Notably, 1.3‐fold increase serum leucine (4.08 μmol/mL in 16BO1 vs. 3.05 μmol/mL in 16BO2) and other elevated branched‐chain amino acids (such as valine and isoleucine) in BO1 may activate the mTORC1 pathway, a critical regulator of protein synthesis. Leucine, in particular, is a well‐documented mTOR agonist that stimulates ribosomal biogenesis and translation initiation [53]. This aligns with observations in mammals, where BCAA‐enriched diets enhance muscle protein accretion via mTOR signaling [54, 55]. Muscle gene expression analysis provides evidence of mTOR pathway activation. The expression of eif4g (eukaryotic translation initiation factor 4G), a critical component of the eIF4E complex that recruit ribosomes to mRNA [56], was significantly higher in the 16BO1 group than in the 16BO2 group. Similarly, akt (a serine/threonine kinase in the PI3K/Akt/mTOR pathway) was upregulated in 16BO1 group fish. Akt phosphorylates downstream targets like mTOR to promote translation initiation and inhibit proteolysis [57]. These findings parallel studies in trout and golden pompano [16, 58]. The coordinated upregulation of these genes suggests that BO1’s FA profile enhances translational efficiency, thereby compensating for the reduced of FM.

Notably, the interaction between FM level and FA composition is evident in hepatic transaminase activity: 16BO1 exhibited higher ALT and AST than 24BO1. This suggests that under low‐FM conditions, BO1 enhances transamination—facilitating the conversion of nonessential amino acids to essential ones—thereby adapting to the reduced supply of essential amino acids from FM [59]. Such metabolic shifts are consistent with studies showing that balanced FA profiles enhance hepatic nitrogen efficiency in carnivorous fish, as observed in European sea bass [60]. The results suggest that the improved protein metabolism in the BO1 group may be a key mechanism through which dietary FA composition affects FM substitution efficiency in largemouth bass. This indicates that FM substitution should not rely solely on alternative proteins; reducing protein content must be paired with optimized lipid sources to maintain protein utilization efficiency.

4.3. Antioxidant Enhancement and Health Maintenance: Protecting Low‐FM Diet Adaptation

The addition of different fat sources to a low‐FM diet significantly affected the antioxidant capacity and health status of M. salmoides, indicating that FA composition plays a critical role in regulating FM substitution efficacy. In the 16% FM groups, BO1‐fed fish exhibited higher serum T‐AOC compared to BO2 and FO groups (p < 0.05), while the lowest serum MDA levels were found in 16BO1 group. Elevated T‐AOC reflects enhanced systemic antioxidant defense, whereas reduced MDA, a lipid peroxidation byproduct, indicates attenuated oxidative stress [61, 62]. These improvements in antioxidant status may be attributed to BO1’s unique FA profile, which includes olive oil‐derived monounsaturated FAs and palmitic acid. MUFA, such as oleic acid, have been reported to enhance cellular antioxidant enzyme activities by reducing reactive oxygen species (ROS) generation and stabilizing membrane integrity [63]. Conversely, BO2, rich in soybean oil‐derived n‐6 PUFAs, likely increased susceptibility to lipid peroxidation due to the higher oxidative instability of n‐6 PUFA [64]. Notably, BO1 also improved liver health by enhancing antioxidant capacity, as evidenced by significantly lower serum ALT and AST activities in the 16% FM groups (p < 0.05). ALT and AST are biomarkers of hepatic damage, and their reduced activities suggest mitigated hepatocyte injury [65]. This aligns with the observed upregulation of hepatic CAT activity in the 16BO1 group compared to the 16BO2 group (p < 0.05). CAT, a key enzyme neutralizing hydrogen peroxide, is critical for protecting hepatic tissues from oxidative damage [66]. Intriguingly, 16BO1 group exhibited higher hepatic CAT activity than 24BO1 group, implying that reducing FM levels did not compromise antioxidant capacity when BO1 was used. Instead, the balanced FA composition in BO1 likely optimized lipid metabolism, reducing oxidative burden. Furthermore, the superior antioxidant performance of BO1 may explain its positive effects on growth and lipid metabolism. Enhanced antioxidant defenses reduce protein oxidation and cellular damage, preserving energy allocation for growth [67]. The above results indicate that the antioxidant enhancement properties of BO1, driven by its unique FA characteristics, contribute to improving fish health and FM replacement efficacy.

This study highlights the importance of considering dietary FA composition when formulating low‐FM aquafeeds. The successful application of a blended oil with optimized FA composition in largemouth bass offers a practical model for other carnivorous species. Future research should focus on (1) defining species‐specific FA requirements more precisely; (2) exploring interactions between novel lipid blends and alternative protein sources; (3) assessing long‐term effects of optimized diets on fish health, product quality, and environmental impact.

5. Conclusions

This study demonstrates that dietary FA composition can critically modulate FM substitution efficacy in largemouth bass, which is the first report in aquatic feeds to our knowledge. Blended oil (BO1) with optimized FA composition can maintain fish growth at 16% dietary FM level by promoting fat decomposition, enhancing protein synthesis and retention, and improving antioxidant capacity, while BO2 rich in n‐6 PUFA leads to decreased growth performance, accompanied by excessive fat accumulation and decreased protein synthesis. These findings indicate the importance of FA composition and lipid source selection in low‐FM feed formula and provide new insights and strategies for improving FM substitution effect and developing sustainable carnivorous fish feed.

Author Contributions

Junfeng Guan: investigation, data curation, formal analysis, writing – original draft preparation. Jianzhao Xu: investigation, data curation, writing – original draft preparation. Xin Gao and Zekui Huang: resources. Chao Xu and Ermeng Yu: investigation. Dizhi Xie: conceptualization, methodology, formal analysis, writing – review and editing. Yuanyou Li: conceptualization, writing – review and editing, project administration, funding acquisition.

Funding

This work was financially supported by the National Natural Science Foundation of China (Grant 32273148) and National Key Research and Development Program of China (Grant 2024YFD2402000).

Disclosure

After using DeepSeek and ChatGPT, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Conflicts of Interest

The authors declare no conflicts of interest.

Guan, Junfeng , Xu, Jianzhao , Gao, Xin , Huang, Zekui , Xu, Chao , Yu, Ermeng , Xie, Dizhi , Li, Yuanyou , Blended Oil With an Optimized Fatty Acid Profile Improves Fish Meal Substitution Efficacy in Carnivorous Teleost Largemouth Bass Diet, Aquaculture Nutrition, 2026, 8841385, 25 pages, 2026. 10.1155/anu/8841385

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.