A mixed animal and plant protein source replacing fishmeal affects bile acid metabolism and apoptosis in largemouth bass (Micropterus salmoides)

Abstract

Chicken meal, shrimp meal, blood meal, and soybean protein concentrate are common alternatives to fishmeal. This study used them to prepare three diets with different levels of fishmeal (FM48, FM40, and FM32) for largemouth bass (Micropterus salmoides). The results found no significant difference in the growth performance of largemouth bass fed different diets. Mixed protein increased the total cholesterol (T-CHO) content in plasma, and reduced the total superoxide dismutase (T-SOD) activity in plasma and liver. Targeted metabolomics analysis found that the low fishmeal diets affected the cholesterol and bile acid metabolism of largemouth bass. Mixed protein inhibited cyp7a1 and enhanced hmgcr and pparγ mRNA levels, as well as enhanced the expression levels of FXR in the liver. The fish-fed FM32 diet showed inhibited fxr, rxrα, and cyp7a1 mRNA levels in the intestine. The results of TUNEL fluorescence staining showed that mixed protein induced apoptosis in largemouth bass. The caspase 3 and caspase 9 mRNA levels in the fish-fed FM40 and FM32 diet significantly increased, as well as the expression levels of CASPASE 3. The experiment also found that it could induce oxidative stress and endoplasmic reticulum stress. In conclusion, the replacement of fishmeal with mixed animal and plant protein diets did not affect the growth performance, but the health and bile acid metabolism of largemouth bass was affected when the fishmeal level was reduced to 32%.

A mixed animal and plant protein source replaces fishmeal in largemouth bass diet, affecting cholesterol and bile acid metabolism, and inducing cell apoptosis.

Graphical Abstract

Graphical Abstract.

Introduction

Largemouth bass (Micropterus salmoides) is native to North America. Due to its delicious taste, it has become a popular economic fish in many countries and regions. In 2022, the production of largemouth bass reached 802, 500 tons in China, and will continue to increase in the future (Affairs, 2023). Fishmeal is an essential feed ingredient for largemouth bass. However, the price of fishmeal is increasing year by year, so it is urgent to find substitutes for fishmeal (Huang et al., 2022).

In our previous research (Chen et al., 2024), using a mixed animal protein source (chicken meal, blood meal, and shrimp meal) instead of fishmeal found no significant effect on the growth of largemouth bass, but excessive substitution can affect cholesterol and bile acid metabolism, which is not conducive to fish health. Therefore, based on the previous research results, the authors wanted to reduce the amount of animal protein and further explore the effect of mixed animal and plant protein sources on largemouth bass. Soybean protein concentrate (SPC) is a common source of plant protein in fish diets. SPC is made from soybean as raw material, and the antinutritional factors are mostly removed during the processing (Feng et al., 2017). It had shown that fishmeal with a high proportion of SPC replacement (more than 80%) did not inhibit the growth of Larimichthys crocea (Wang et al., 2017) and Acanthopagrus schlegelii (Kalhoro et al., 2018). However, in the experiment with Litopenaeus vannamei, when the substitution level of SPC was above 30%, the weight gain rate (WGR) significantly decreased (Zhu et al., 2021). According to the study by Li et al. (2018), when the substitution level of SPC for fishmeal did not exceed 20%, there was no significant change in the WGR of yellow catfish (Tachysurus fulvidraco). However, when the substitution level was increased to over 30%, the growth and the antioxidant capacity decreased. Chicken meal and blood meal are common alternatives to fishmeal. Shrimp meal and brewer’s yeast are good attractants in fish feed that promote growth (Zhou et al., 2018).

Replacing fishmeal with other protein sources usually induces negative effects, such as oxidative stress. The replacement of fishmeal with chicken meal significantly affected the antioxidant capacity of Trachinotus ovatus (Hu et al., 2019; Zhang et al., 2019a) and Monopterus albus (Cao et al., 2020). Excessive addition of SPC in feed also can lead to oxidative stress in fish. In the substitution experiments of pearl gentian groupers (Zhang et al., 2023b), Pelteobagrus fulvidraco (Yu, 2020), and Penaeus vannamei (Yu, 2020), the superoxide dismutase (T-SOD) activity showed a decreasing trend with the increase of SPC substitution rate. But Ma et al. (Ma et al., 2021) found that replacing fishmeal with chicken meal had no significant effect on the antioxidant capacity of largemouth bass. Breeding varieties and feed formula may be the main factors causing this difference. High plant protein to replace fishmeal also causes other negative effects. Zhang (Zhang, 2019) replaced all fishmeal with mixed plant protein (SPC: cottonseed protein concentrate (CPC) = 1: 1.66) of Lateolabrax japonicas, and found that the caspase 3 mRNA levels were significantly upregulated and apoptosis was occurring in the liver. Whole plant protein diet caused oxidative stress and intestinal damage in Acipenser schrenckii, and inhibited antioxidant, apoptosis, and autophagy, leading to weakened immunity and reduced growth performance (Wei, 2020). SPC replacing some fishmeal could improve the intestinal digestion capacity, but replacing many fishmeal could destroy the intestinal structure of Epinephelusspp (Wang, 2020). Studies had shown that replacing fishmeal with SPC could break the intestinal morphology and microbiota of Larimichthys crocea (Wang et al., 2023b), Totoaba macdonaldi (Ernesto et al., 2021), and Lateolabrax japonicus (Xiang, 2017), affecting their digestive ability. Replacing fishmeal with SPC significantly reduced the total cholesterol (T-CHO) content of juvenile shrimp (Litopenaeus vannamei;Ray et al., 2020) and Monopterus albus (Tang et al., 2019).

With the development of science and technology, metabolomics, which was born at the end of the last century, is becoming more and more important in biological research. It can more directly and accurately reflect the physiological state of organisms. Metabolomics is mainly divided into two categories: untargeted and targeted analysis. Metabolomics analysis is rare in fishmeal replacement studies. Chen et al. (2024) used untargeted metabolomics to analyze the effects of an animal mixed protein diet on largemouth bass and found it significantly affected the cholesterol and bile acid metabolism. This experiment used H650 high-throughput targeted metabolomics, which could conduct accurate and absolute quantitative analysis of about 650 functional small molecule metabolites, including amino acids, lipids, purine nucleotides, carbohydrates, organic acids, bile acids, etc. Its unverifiable results were widely welcomed among researchers. Targeted metabolomics played an important role in the study of CPC (Xu et al., 2024) and clostridium autoethanogenum protein (Yang et al., 2022) as substitutes for largemouth bass meal.

This experiment used a mixed animal and plant protein source (chicken meal, shrimp meal, blood meal, and SPC) to replace fishmeal and to study the effects on the growth performance, health, and metabolic pathway of largemouth bass.

Materials and Methods

Statement

All animals in this experiment were used in a background that complied with the ethical guidelines of the Institutional Animal Care and Use Committee of Guangdong Ocean University.

Diet preparation

This experiment set up three diets with different levels of fishmeal (48%, 40%, and 32%; Table 1), chicken meal, shrimp meal, blood meal, and SPC were used to replace fishmeal. The proportion of each raw material was adjusted in the low fishmeal diets to obtain nutritional characteristics similar to the 48% fishmeal diet (Supplementary Table S1). Lysine, threonine, methionine, taurine, and choline chloride were added to the diet to meet the nutritional growth requirements of the largemouth bass.

Table 1.

Diets composition (dry basis, %)

Ingredient	FM48	FM40	FM32
Fish meal	48	40	32
Chicken meal	10	12	14
Shrimp meal	3	4	5
Blood meal	0	0.5	1
Soybean protein concentrate	3	7	11
Brewer’s yeast	1	1.5	2
Fish oil	0	0.8	1.6
Soybean oil	3.5	3.1	2.7
Phospholipid oil	2	2	2
Wheat flour	8	8	8
Cassava starch	4	4	4
Peeled soybean meal	4	4	4
Cottonseed protein concentrate	3	3	3
Fungicide	0.1	0.1	0.1
Calcium phosphate primary	1.5	2	2.5
l-lysine hydrochloride	0.38	0.44	0.51
l-threonine	0.15	0.14	0.14
dl-methionine	0.14	0.2	0.26
Taurine	0	0.03	0.06
Choline chloride	0.5	0.5	0.5
Degummed bone meal	7.73	6.69	5.63
Total	100	100	100
Nutrition level
Crude protein	48.17	48.08	48.02
Crude lipid	11.63	11.64	11.65
Moisture	9.09	9.02	9.00

All raw materials are provided by Guangzhou Chengyi Aquaculture Co. LTD (Guangzhou, China). Fish meal: crude protein 66.74%, crude lipid 10.03%; chicken meal: crude protein 62.71%, crude lipid 12.57%; shrimp meal: crude protein 62.41%, crude lipid 7.75%; blood meal: crude protein 92.23%, crude lipid 0.18%; soybean protein concentrate: crude protein 63.98%, crude lipid 1.78%.

Fish rearing

The experimental fish were divided into three groups, each with four breeding tanks. Thirty healthy largemouth bass (12.68 ± 0.14 g) of similar size were fed in each tank. Feeding the fish twice daily (8:00 and 16:00) to satiation. The feeding trial lasted for 8 wk. Maintained sufficient oxygen content in the water during the experiment. Changed the water at least once a week to maintain a clean environment.

Sample collection and analyses

Collected the samples in largemouth bass after 24 h of starvation. Three fish of similar size were taken from each tank and immediately stored at −20 °C. The tail vein blood was drawn using small syringes. Blood was centrifuged at 4 °C 1788.8 g for 10 min, and plasma was collected from the upper part of the centrifuge tube. Plasma was stored at −80 °C for metabolomics analysis. The liver and intestinal samples used in western blot and qPCR experiments were placed at −80 °C. The intestines used for HE and TUNEL staining were placed in a 4% formaldehyde solution. The moisture, crude protein, and crude lipid content of the whole fish were determined by Chen et al. (2024).

The contents of the malondialdehyde (MDA), total protein, triglyceride (TG), total cholesterol (T-CHO), total bile acid (TBA), and the activity of catalase (CAT) and total superoxide dismutase (T-SOD) were determined according to the method provided by Nanjing Jiancheng Bioengineering Institute (Nanjing, China).

qPCR analysis

Total RNA was extracted from the liver and the intestine using a commercial kit from Beyotime Biotechnology Co., Ltd (Shanghai, China). The cDNA synthesis and qPCR analysis were performed using kits from Accurate Biotechnology Co., Ltd (Changsha, China). The amplification experiment required 1 μL cDNA template, 5 μL premix II, 0.5 μL primer F and R, and 3 μL RNase free water. qPCR analysis was performed using the LightCycler480 (Roche Applied Science, Switzerland, Basel).

ef1α is the reference gene of this experiment, which with high stability. The primers for all genes were synthesized by Sangon Biotech Co., Ltd (Shanghai, China; Supplementary Table S2; Yu et al., 2018, 2019; Xie et al., 2020a; Yin et al., 2021; Zhao et al., 2021; Wang et al., 2022).

Western blot analysis

The method of western blot analysis can be found in Supplementary Material (Xu et al., 2022a, 2022b). The antibodies used in this experiment: FXR (1:800, bs-12867R, Bioss), PPARA (1:800, 66,836-1-Ig, Proteintech), P-PPARA (1:800, ab3484, Abcam), CASPASE 3 (1:1000, ab184787, Abcam), CASPASE 9 (1:1500, ab184786, Abcam), CASPASE 12 (1:1500, ab315271, Abcam), and GAPDH (1:800, 2118S, Cell Signaling Technology).

Histological analysis

The fully fixed intestines were placed into paraffin solution, they were completely soaked and cut into small wax pieces. The paraffin blocks were sectioned using a slicing machine. Sections were stained with hematoxylin and eosin (HE staining). The TUNEL staining was performed using a kit from Servicebio Technology Co., Ltd (Wuhan, China). The results were observed and calculated under a microscope (Nikon, Tokyo, Japan).

Targeted metabolomics analysis

Eight plasma samples from each of FM48 and FM32 were selected for targeted metabolomics analysis. For samples and specific operational information, please refer to Supplementary Material.

Data analysis

$${\displaystyle \begin{array}{l}{\displaystyle WGR(\mathrm{\%})=100\times (W_b-W_a)/W_a,}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{S}\mathrm{G}\mathrm{R}(\mathrm{\%}/d)=100\times (Ln{W}_b-Ln{W}_a)/b,}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{F}\mathrm{e}\mathrm{e}\mathrm{d}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}(\mathrm{F}\mathrm{C}\mathrm{R})=\mathrm{d}\mathrm{r}\mathrm{y}\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}\mathrm{f}\mathrm{e}\mathrm{d}(g)/(W_b-W_a),}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{F}\mathrm{e}\mathrm{e}\mathrm{d}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}(FR,\mathrm{\%})=100\times \mathrm{d}\mathrm{r}\mathrm{y}\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}\mathrm{f}\mathrm{e}\mathrm{d}(g)/[(W_b+W_a)/2\times b],}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{S}\mathrm{u}\mathrm{r}\mathrm{v}\mathrm{i}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}(SR,\mathrm{\%})=100\times (\mathrm{f}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{l}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{t}\mathrm{y})/(\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{t}\mathrm{y}),}\end{array}}$$

W_a is the initial body weight (g), W_b is the final body weight (g), and b is the number of experimental days.

The data are first tested for normal distribution and homogeneity of variance test, followed by a one-way analysis of variance. If there are significant differences (P < 0.05), the Duncan method will be used for multiple comparisons. All data are processed using SPSS Statistics 21 (IBM, USA). The results are presented as the means ± standard error (SE).

Results

Growth performance

There were no significant differences in percent weight gain (WGR), specific growth rate (SGR), FCR, FR, and survival (SR) among the three groups (P > 0.05). There were no significant differences in moisture, crude lipid, and crude protein among the groups (P > 0.05; Table 2).

Table 2.

Effects of three diets on the growth performance of largemouth bass

Items	FM48	FM40	FM32
WGR, %	410.09 ± 7.13	426.13 ± 7.74	411.54 ± 3.38
SGR, %/d	2.91 ± 0.03	2.97 ± 0.03	2.92 ± 0.01
FCR	0.83 ± 0.00	0.82 ± 0.00	0.82 ± 0.01
FR, %	2.06 ± 0.03	2.02 ± 0.02	2.01 ± 0.02
SR, %	89.17 ± 1.59	88.33 ± 6.16	83.33 ± 4.08
Moisture, %	70.84 ± 0.46	70.80 ± 0.28	70.08 ± 0.10
Crude lipid, %	7.08 ± 0.38	7.14 ± 0.39	8.05 ± 0.19
Crude protein, %	17.41 ± 0.28	17.34 ± 0.26	17.37 ± 0.28

Data represent means ± SE of four replicates (n 4). The absence of superscript letters indicates no significant differences among the three groups (P > 0.05).

Biochemical responses

The T-CHO content in the plasma of the fish fed the FM32 diet was significantly higher than those fed the other diets (P < 0.05). The T-SOD activity in plasma was significantly higher in the fish fed the FM48 diet than those fed the FM40 and FM32 diets (P < 0.05; Table 3).

Table 3.

Effects of three diets on biochemical responses of largemouth bass

Items	FM48	FM40	FM32
Plasma
MDA, mmol/mL	52.91 ± 2.64	60.02 ± 2.94	59.36 ± 2.88
TP, mg/mL	47.30 ± 2.87	45.49 ± 2.13	47.31 ± 2.28
TG, mmol/L	3.90 ± 0.41	4.37 ± 0.88	4.10 ± 0.20
T-CHO, mmol/L	18.46 ± 0.71a	22.98 ± 3.58a	32.01 ± 1.42b
TBA, μmol/L	3.62 ± 0.18	3.44 ± 0.18	3.53 ± 0.16
CAT, U/mL	6.32 ± 0.35	5.52 ± 0.34	5.03 ± 0.02
T-SOD, U/mL	487.52 ± 9.84b	414.91 ± 5.74a	411.61 ± 15.56a
Liver
MDA, mmol/mgprot	0.29 ± 0.02	0.30 ± 0.04	0.30 ± 0.02
TP, mg/mL	8.86 ± 0.70	8.22 ± 0.97	7.94 ± 0.20
TG, mmol/L	0.11 ± 0.03	0.12 ± 0.03	0.14 ± 0.02
T-CHO, mmol/gprot	0.08 ± 0.01	0.08 ± 0.02	0.08 ± 0.01
TBA, μmol/L	8.38 ± 1.12	8.24 ± 1.42	8.53 ± 1.42
CAT, U/mgprot	3.00 ± 0.21	2.93 ± 0.33	2.80 ± 0.06
T-SOD, U/mgprot	959.66 ± 50.71b	767.07 ± 61.42ab	729.61 ± 55.14a

Data represent means ± SE of four replicates (n 4). The absence of superscript letters indicates no significant differences among the three groups (P > 0.05). a, b indicates significant differences among the three groups (P < 0.05).

The fish fed an FM48 diet had the highest T-SOD activity in the liver, which was significantly higher than those fed an FM32 diet (P < 0.05; Table 3).

Lipid metabolism

The hmgcr and pparγ mRNA levels in the liver were significantly higher in the fish-fed FM32 diet than that fed FM48 diet (P < 0.05). The cyp7a1 mRNA levels in the liver were significantly higher in the fish-fed FM48 diet than that fed FM32 diet (P < 0.05). The fxr mRNA levels in the intestine were significantly higher in the fish fed the FM48 diet than in those fed the other diets (P < 0.05). Fish-fed FM48 diet had the highest rxrα mRNA levels in the intestine, while fish-fed FM32 diet had the lowest rxrα mRNA levels (P < 0.05). The cyp7a1 mRNA levels in the intestine were significantly lower in the fish-fed FM32 diet than those fed the other diets (P < 0.05; Figure 1).

Figure 1.

(A-B) Lipid metabolism-related gene expression in the liver. (C-D) Lipid metabolism-related gene expression in the intestine. (E-F) The western blot analysis of FXR, PPARA, P-PPARA, and GAPDH in the liver. (fxr, farnesoid X receptor; rxrα, retinoid X receptor α; hmgcr, 3-hy-droxy-3-methylglutaryl-coenzyme A reductase; cyp7a1, cholesterol 7 alpha-hydroxylase; cyp8b1, 12a-hydroxylase; cyp27a1, sterol 26-hydroxylase; ppar, peroxisome proliferator activating receptor). Values are means, and vertical bars indicate SE. a, b, c indicate significant differences among the three groups (P < 0.05). The following figures are the same.

The FXR protein expression levels in the liver were significantly higher in fish the fed FM32 diet than that fed the FM40 diet (P < 0.05). There were no significant differences in the protein expression levels of PPARA (P > 0.05; Figure 1).

Antioxidant capacity

The sod mRNA levels in the intestine were significantly lower in the fish fed the FM48 diet than in those fed the other diets (P < 0.05). There were no significant differences in cat and Cu-Zn sod mRNA levels among the three groups (P > 0.05; Figure 2).

Figure 2.

(A) Antioxidant-related gene expression in the liver. (B) Antioxidant-related gene expression in the intestine. (cat, catalase; sod, superoxide dismutase).

Endoplasmic reticulum stress

The grp78 mRNA levels in the liver were significantly higher in fish fed an FM32 diet than in those fed the other diets (P < 0.05). There were no significant differences in ire1, atf6, and eif2α mRNA levels among the three groups (P > 0.05; Figure 3).

Figure 3.

(A) ERS-related gene expression in the liver. (B-C) The western blot analysis of CASPASE 12 and GAPDH in the liver. (ire1, inositol-requiring enzyme 1; atf6, activating transcription factor 6; eif2α, eukaryotic initiation factor 2α; grp78, glucose-regulated protein 78).

The CASPASE 12 protein expression levels in the liver were significantly lower in the fish fed an FM48 diet than those fed an FM40 and FM32 diet (P < 0.05; Figure 3).

Inflammation response

The il-10 mRNA levels in the liver were significantly higher in the fish fed an FM48 diet than those fed the FM40 and FM32 diet (P < 0.05). There were no significant differences in mRNA levels of other immune-related genes in the liver and intestine of each group, such as tcrα, tcrβ, tgf-β1, il-1β, il-8, il-15, and il-18 (P > 0.05; Figure 4).

Figure 4.

(A-B) Immune-related gene expression in the liver. (C-D) Immune-related gene expression in the intestine. (tcr, T-cell receptor; tgf-β1, transforming growth factor-β; il, interleukin).

Apoptosis

The caspase 3 and caspase 9 mRNA levels in the liver were significantly lower in the fish fed FM48 diet than those fed FM40 and FM32 diet (P < 0.05). The caspase 9 mRNA levels in the intestine were significantly lower in the fish fed an FM48 diet than in those fed the other diets (P < 0.05; Figure 5).

Figure 5.

(A) Apoptosis-elated gene expression in the liver. (B) Apoptosis-related gene expression in the intestine. (C-D) The western blot analysis of CASPASE 3, CASPASE 9, and GAPDH in the liver. (E-F) The results of TUNEL staining in the intestine. (caspase, cysteine-aspartic proteases).

The protein expression levels of CASPASE 3 in the intestine increased with the decreased dietary fishmeal levels (P < 0.05). The cell apoptosis rate was significantly higher in the fish-fed FM40 and FM32 diet than that fed the FM48 diet (P < 0.05; Figure 5).

Intestinal histology

The fold of fish fed with FM48 was significantly higher than that fed with FM32 (P < 0.05). The muscle layer fed FM32 was the thinnest, significantly less than those fed with FM48 and FM40 (P < 0.05; Figure 6).

Figure 6.

The results of HE staining in the intestine. Blue arrow: fold height; red arrow: fold width; black arrow: muscle layer thickness.

Differential metabolites

A total of 336 metabolites were detected in plasma. With the standard of P value < 0.05, it was found that a mixed protein diet had significant effects on the metabolism of 31 substances in largemouth bass (Figure 7, Supplementary Table S3). From the volcano map, 12 differential metabolites were upregulated and nineteen were downregulated. From the hierarchical clustering heatmap, cholesterol sulfate, taurohyodeoxycholic acid and acetylcholine content decreased, cholesterol, lithocholic acid (LCA), isolithocholic acid (IsoLCA), and 3-dehydrocholic acid (3-DHCA) content increased.

Figure 7.

(A) OPLS-DA score of FM48 and FM32. (B) The volcano map of FM48 and FM32. (C) The hierarchical clustering heatmap of FM48 and FM32.

Differential metabolic pathways

Based on the KEGG database, 58 differential metabolic pathways were found between FM48 and FM32. The most affected metabolic pathways were lipid metabolism, bile acid metabolism, biosynthesis of unsaturated fatty acids, cholinergic synapse, amino acid metabolism, cAMP signaling pathway, etc. (Figure 8).

Figure 8.

(A) Metabolic pathway influence factor bubble fig between FM48 and FM32. (B) Differential abundance scores of the metabolic pathways between FM48 and FM32.

Discussion

In recent years, there has been increasing research on the replacement of fishmeal for largemouth bass. The types of protein sources are divided into animal protein sources and plant protein sources. Experiments showed that a single protein source readily inhibited the growth of animals, mixed protein sources could ameliorate this phenomenon (Wang et al., 2023a). Cui et al. (2021) found that using SPC and CPC alone inhibited the growth of largemouth bass, but using a mixed diet had no negative effect. Millamena (2002) replaced fishmeal with meat and blood meal (4: 1) and found no effect on growth, but the whole meat meal group and whole blood meal group inhibited the growth of grouper (Epinephelus coioides). Chen et al. (2024) replaced fishmeal with a mixed animal protein source (chicken meal, blood meal, and shrimp meal) and found no significant effect on the growth of largemouth bass. SPC is often mixed with other substances to replace fishmeal. Complete replacement of fishmeal by SPC and CPC reduced WGR and SR of Acipenser schrenckii (Wei et al., 2019). Replacing fishmeal with SPC and soybean meal mixture also inhibited the growth of Litopenaeus vannamei (Xie et al., 2016). In actual production, the cost of the whole animal protein diet is more expensive, and the whole plant protein diet is easy to cause nutritional imbalance and induces negative reactions. So the most common method is to replace fishmeal with a mixture of plant and animal protein sources. Tidwell et al. (2005) found that a mixture of blood meal and corn protein meal did not affect the growth rate of largemouth bass. Cao et al. (2017) and Li et al. (2021) used shrimp hydrolysate, fermented soybean meal, and corn protein meal as fishmeal substitutes, which did not have a negative impact on the growth of largemouth bass. Niu et al. (2022) replaced 40% fishmeal with composite animal and plant proteins and found that it could improve the growth performance of largemouth bass. Fumiaki et al. (2023) and Rawles et al. (2022) partially replaced fishmeal with a commercial plant and animal protein mixture, which did not affect the growth of red sea bream (Pagrus major) and Morone chrysops. In the present study, a mixture of chicken meal, shrimp meal, blood meal, and SPC was used instead of fishmeal. There were no significant differences in WGR, SGR, FCR, FR, and SR between the groups. From the results of this experiment, it is feasible to replace fishmeal with a mixed animal and plant protein source.

In the present study, the T-CHO content in the plasma of the fish fed the FM32 diet was significantly higher than those fed the other diets. This is consistent with the experimental results of Fumiaki et al. (2023) and Ray et al. (2020). It was found that chicken meal replaced fishmeal and the T-CHO content of juvenile yellow catfish first increased and then decreased (Luo et al., 2017). However, Ma et al. (2021) used chicken meal to replace fishmeal, showing no significant changes in T-CHO content in largemouth bass. The same results were obtained with chicken meal replacing fishmeal in Takifugu obscurus (Cui, 2022) and Sparus aurata (Karapanagiotidis et al., 2019). Cholesterol belongs to the class of lipids and plays an important role in activities, such as lipid metabolism. Bonaldo et al. (2015) found that as the level of plant protein in feed increased, the lipid utilization efficiency of turbot decreased. Adding cholesterol to low fishmeal feed did not affect lipid metabolism in Litopenaeus vannamei (Li et al., 2022). Hmgcr is the rate-limiting enzyme for de novo cholesterol synthesis (Laura and Andrew, 2013). In the present study, the hmgcr mRNA levels were significantly higher in the fish fed the FM32 diet than that fed the FM48 diet, which is consistent with the trend of changes in cholesterol content in the blood. These results suggested that the use of mixed animal and plant protein sources to replace fishmeal may affect the cholesterol metabolism of largemouth bass.

The authors drew Figure 9 based on the results of the H650 targeted metabolomics analysis. The metabolomic analysis found that mixed protein sources reduced cholesterol sulfate content (Figure 9). Cholesterol sulfate synthesizes cholesterol in the body through a series of reactions and participates in bile acid biosynthesis. Metabolomics analysis found that mixed protein sources could reduce the content of palmitic acid, 11Z-eicosenoic acid, and erucic acid while increasing the content of alpha-linolenic acid and 13Z,16Z-docosadienoic acid. And it significantly affected the biosynthesis pathways of unsaturated fatty acids (Figure 9). The metabolism of fatty acids may affect the lipid metabolism of largemouth bass. Lipid metabolism is the process of breaking down lipid substances into small molecules, such as fatty acids. Under sufficient oxygen conditions, fatty acids can be β-oxidized to acetylCoA and participate in the synthesis of bile acids.

Figure 9.

The effects of mRNA levels, differential metabolites, and metabolic pathways on cholesterol and bile acid metabolism. Red arrows indicate promoted responses, blue arrows indicate inhibited responses.

Cholesterol content is closely associated with bile acid metabolism. Cholesterol undergoes a series of reactions to synthesize bile acids (Javitt, 1990; Norlin et al., 2000). In the present study, there were no significant differences in TBA and TG content among the three groups. The genes involved in bile acid synthesis are cyp7a1, cyp8b1, and cyp27a1. Cyp7a1 and cyp8b1 are two key enzymes in the classical pathway, of which cyp7a1 is the only rate-limiting enzyme. Cyp27a1 is a key enzyme in the alternative pathway. The cyp7a1 had lower expression in low fishmeal feed, indicating that mixed animal plant protein sources inhibit the synthesis of bile acids in largemouth bass. In the experiment of Fumiaki et al. (2023), gene expression levels of hepatic cyp7a1 were lower in the no-fishmeal group than in the control group, and the same trend was observed for bile acid content. The mRNA level of hmgcr is significantly upregulated, while the mRNA level of cyp7a1 is significantly downregulated, which can cause cholesterol accumulation (Zhang et al., 2019b) and affect bile acid metabolism. In Figure 9, the classical pathway of bile acid biosynthesis is inhibited and more cholesterol is decomposed into bile acid by the alternative pathway. Metabolomics analysis found that mixed protein sources could reduce the content of taurohyodeoxycholic acid while increasing the content of LCA, IsoLCA, and 3-DHCA (Figure 9). LCA is an intermediate product of alternative pathways, and IsoLCA can be converted into LCA. The increase in their content in this experiment could also indicate that bile acids were mainly synthesized through the alternative pathway.

The replacement of fishmeal with mixed protein sources inhibited the expression of fxr and rxrα in the intestine. And, it enhanced the protein expression of FXR in the liver. fxr plays an important role in bile acid and lipid metabolism (Wang et al., 2008). Fxr could bind to the rxrα to jointly promote the metabolism of bile acids (Thompson et al., 2018). So the fxr mRNA levels were highly expressed in the liver and low in the intestine, possibly related to the activity of rxrα. In Figure 9, the increase in fxr and ppar mRNA levels actively regulates the synthesis, transport, and detoxification of BA, affecting bile secretion.

SOD has antioxidant properties and removes harmful O²⁻ radicals from the body (Sakai et al., 1992). In the present study, the T-SOD activity was significantly lower in fish fed the FM32 diet than those fed an FM48 diet. This indicated that the antioxidant capacity of the fish body weakens as the dietary fishmeal is replaced. In the present study, the sod mRNA levels in the intestine were significantly higher in the fish fed FM40 and FM32 diets than that fed FM48 diet. This suggested that the fish’s body was undergoing oxidative stress. In the feed of Yellow Catfish (Pelteobagrus fulvidraco), when the level of SPC replacing fishmeal was 20% and 30%, the SOD activity in the liver was significantly increased (Li et al., 2018). This may be related to the presence of soy isoflavones in SPC, which have antioxidant activity.

In this study, the mixed animal and plant diet induced endoplasmic reticulum stress (ERS). The endoplasmic reticulum is an important organelle in eukaryotic cells, where protein synthesis and processing are performed (Ozcan and Tabas, 2012; Tam et al., 2017). ERS is caused by the accumulation of unfolded and misfolded proteins (Li et al., 2013; Liang et al., 2016; Maja et al., 2020). The body can alleviate ERS by unfolding protein responses (Lukas and Arthur, 2011). grp78 is a key factor in responses, which can help peptide chains fold correctly in the endoplasmic reticulum (Alhusaini et al., 2010). In the present study, the grp78 mRNA levels in the liver were significantly higher in the fish fed an FM32 diet than in those fed the other diets. Similarly, the elevated level of CASPASE 12 protein expression also indicated that mixed animal and plant diet-induced ERS in largemouth bass. There was a research report that low fishmeal feed could cause ERS in Litopenaeus vannamei (Xie et al., 2020b). High-fat diet-induced ERS in yellow catfish (Ling et al., 2019). High carbohydrate diet-induced ERS in largemouth bass (Zhao et al., 2021). Heat stress (Zhao et al., 2022) and a hypoxic environment (Liu et al., 2023; Zhang et al., 2023a) could also induce ERS in largemouth bass. The occurrence of ERS is caused by multiple factors. In this study, it is possible that the presence of antitrophic factors in plant proteins or other causes induced ERS.

ERS is usually accompanied by inflammation and apoptosis (Tam et al., 2017). Il-10 is an anti-inflammatory cytokine, and its mRNA level decreased, indicating that fish in the FM40 and FM32 groups were undergoing liver self-repair. Apoptosis is an automatic cell death controlled by genes, mainly regulated by CASPASE family proteins (Nuñez et al., 1998; Budihardjo et al., 1999; Espe et al., 2015). In this experiment, using a mixed protein source instead of fishmeal promoted the expression of caspase 3, caspase 9 mRNA, and CASPASE 3 protein. The results showed that the experiment promoted apoptosis through the mitochondrial pathway of the organism. The results of TUNEL fluorescence staining also showed that the experiment promoted apoptosis. Oxidative stress, ERS, and inflammatory response ultimately cause intestinal damage. In the present study, the increase of mixed protein source substitution decreased the fold height and muscle layer thickness in the hindgut of largemouth bass. Ma et al. (2021) study found that chicken meal caused intestinal damage in largemouth bass. Puffed blood meal caused intestinal damage to Yellow River carp (Yuan et al., 2017). This may be caused by antinutritional factors or nutritional imbalance (Navarrete et al., 2013; Bansemer et al., 2015).

Targeted metabolomics analysis found that acetylcholine is a differential metabolite of FM48 and FM32, which can be downregulated by animal and plant proteins of largemouth bass. Acetylcholine mainly affects cholinergic synapses. Acetylcholine is a neurotransmitter that can act on various types of choline receptors. The decrease in acetylcholine content can lead to slower information transmission and slower various physiological activities. In the later stage of this experiment, the feeding behavior of largemouth bass was slow and the feeding amount was low, which may be related to the synthesis rate of acetylcholine. Acetylcholine is hydrolyzed by cholinesterase into choline and acetic acid, a process known as inactivation (Wu et al., 2020). Wu et al. (2020) injected cholinesterase inhibitors into grass carp, which can have a protective effect on foodborne enteritis. Gioele et al. (2021) found that acetylcholine can regulate the immune function of butterfly fish (Pantodon buchholzi). In this experiment, the reason for liver and intestinal damage caused by animal and plant mixed diets may be related to the metabolic pathway of acetylcholine.

Targeted metabolomics analysis found that adenosine-3ʹ,5ʹ-cyclic monophosphate (cAMP) is a differential metabolite of FM48 and FM32, which can be downregulated by animal and plant proteins of largemouth bass. cAMP could also affect multiple metabolic pathways, such as cholinergic synapse, hedgehog signaling pathway, and cAMP signaling pathway. CAMP directly or indirectly regulates physiological metabolism. Huang’s (Huang et al., 2023) experiment showed that lipolysis of adipocytes can be induced by activating the cAMP/PKA signaling pathway in grass carp (Ctenopharyngodon idella). The experiments of Montero et al. (2016) showed that it can promote macrophage M2 polarization and high il-10 mRNA levels by activating the cAMP/CREB signaling pathway. In this experiment, the il-10 mRNA levels were significantly higher in the fish fed an FM48 diet than those fed the FM40 and FM32 diets. This may be the result of animal and plant proteins inhibiting the cAMP/CREB signaling pathway.

Conclusion

The substitution of fishmeal with a mixed animal and plant protein source (chicken meal, shrimp meal, blood meal, and SPC) had no significant impacts on the growth performance of largemouth bass. However, the excessive replacement ratio (FM32) would affect the cholesterol and bile acid metabolism, cause oxidative stress, ERS and apoptosis, and damage the health of largemouth bass.

Glossary

Abbreviations

3-DHCA

3-dehydrocholic acid

CAT

catalase

CPC

cottonseed protein concentrate

ERS

endoplasmic reticulum stress

FCR

feed conversion rate

FR

feeding rate

IsoLCA

isolithocholic acid

LCA

lithocholic acid

MDA

malondialdehyde

SGR

specific growth rate

SR

survival rate

SPC

soybean protein concentrate

TP

total protein

TG

triglyceride

T-CHO

total cholesterol

TBA

total bile acid

T-SOD

total superoxide dismutase

WGR

weight gain rate

Conflict of interest statement

The authors have no financial or personal conflicts of interest to declare.