Alpha-ketoglutaric acid mitigates the detrimental effects of soy antigenic protein on the intestinal health and growth performance of Mirror carp Cyprinus carpio

The study investigated the alleviated effects of Alpha-ketoglutaric acid (AKG) on the intestinal health of mirror carp (Cyprinus carpio Songpu) caused by soy antigenic protein. The diets were formulated from fishmeal (CON), 50% soybean meal (SBM), the mixture of glycinin and β-conglycinin (11 + 7S) and adding 1% AKG in the 11 + 7S (AKG). Carp (~ 4 g) in triplicate (30 fish per tank) was fed to apparent satiation thrice a day for six weeks. Compared with CON, SBM treatment resulted in significantly poor growth performance (P < 0.05), whereas 11 + 7S and AKG treatments were not significantly different from CON (P > 0.05). Gene expression of tumor necrosis factor (TNF-α) and interleukin-1 β (IL-1β) in proximal intestines (PI) and distal intestines (DI) were increased (P < 0.05), and transforming growth factor (TGF-β) in PI and middle intestines (MI) was decreased (P < 0.05) in both SBM and 11 + 7S. The caspase-3 in DI increased in SBM (P < 0.05) and the caspase-3 and caspase-9 in DI increased in 11 + 7S (P < 0.05); conversely, TGF-β in PI and MI was increased, TNF-α and IL-1β in the MI, caspase-3, and caspase-9 in DI was decreased in AKG (P < 0.05). The TOR (target of rapamycin) in PI and MI, ACC in PI, MI and DI was decreased in SBM (P < 0.05), the AMPK in the PI and DI, TOR in PI, MI and DI, ACC in PI and DI, 4E-BP in DI was reduced in 11 + 7S (P < 0.05). AMPK in the PI and DI, ACC in the PI and MI, TOR in PI, MI, and DI, 4E-BP in PI and DI was recovered by AKG supplementation (P < 0.05). Lipids and lipid-like metabolism, organic acids and derivatives metabolism increased in AKG dietary treatment. In conclusion, AKG reduces the expression of intestinal inflammation and apoptosis pathway and changes glycerophospholipid metabolism and sphingolipid metabolism in the intestine of fish.


Introduction
Soybean meal is frequently substituted for fish meal in aquafeed; however, substituting soybean meal for fish meal in excess of a certain proportion may negatively impact fish intestinal health and growth performance (Zhang et al. 2018;Hossain et al. 2018;Lin et al. 2011).Soybean meal influences intestinal mucosa metabolic processes, including lipid, amino acid, sugar, apoptosis, and oxidative injury (Zhao and Xu. 2022;Zhao et al 2023).Zhang et al. (2014) reported that replacing fish meal more than 60% with soy protein isolate (SPI) of carp Cyprinus carpio diet resulted in poor intestinal tract growth and development and the integrity of epithelium was damaged.Soybean meal contains a lot of anti-nutritional factors, such as trypsin inhibitor, plant lectin, and soybean antigen protein.The soybean antigen protein includes four kinds; among them, glycinin (7S) and β-conglycinin (11S) are the main anti-nutritional factors, which account for 70% of the soybean protein (Wilson et al. 2005).Previous studies showed that soybean meal and antigen proteins like glycinin or β-conglycinin could destroy intestinal structure (Krogdahl et al. 2015;Han et al. 2018;Luo et al. 2023a, b), decrease mucosal thickness (Zhang et al. 2014) and mucosa width of the fish intestine, induced nonspecific immunity (eosinophil, neutrophil, macrophages, and lymphocytes increased) (Li et al. 2017a, b;Krogdahl et al. 2015).Either glycinin or β-conglycinin in the diet could decrease intestinal digestive enzymes in fish and reduce growth performance (Li et al. 2017a, b;Zhang et al. 2014).
The intestine of fish is the largest contact area between internal organs and the external environment which has the function of digesting and absorbing nutrients; the health condition of the intestine can affect the health of fish (Lygren et al. 2001).Consequently, it is essential to protect the intestinal health of fish, which would optimize fish production output.Previous research shows that the supplementation of Alpha-ketoglutaric acid (AKG) in diet can reduce the intestinal damage caused by soybean protein in fish (Zhao and Xu 2022).Previous study reported that AKG is a fundamental component of the tricarboxylic acid cycle and contributes significantly to the regeneration and metabolism of intestinal tissues (Guo et al. 2017).Diet supplemented with AKG was beneficial to the growth performance and intestinal health of common carp (Ai et al. 2019;Wu et al. 2022).
Previous studies have shown that, supplementation of glycinin or β-conglycinin alone caused significant adverse effects on fish growth and gut health (Han et al. 2018;Duan et al. 2019;Luo et al. 2023a, b); however, soybean meal contained both the glycinin and β-conglycinin together, the effects of the simultaneous presence of glycinin or β-conglycinin are still unclear.Therefore, this study explored the detrimental effects of glycinin and β-conglycinin on the intestinal health, signal route, and metabolism of carp.Whether AKG supplementation might mitigate the damage caused by soybean antigen protein from the standpoint of metabolome and gene pathways.

Experimental diets and fish feeding
Purified β-conglycinin (11S, 90.2%) and glycinin (7S, 67.9%) were provided by Prof. Yeming Chen at Jiangnan University.AKG (Alpha-ketoglutaric acid) was purchased with a purity of over 98.0% (Sigma-Aldrich, Missouri, USA).All the formulated experimental diets were isonitrogenous and isolipid (Table 1) and met or exceeded all established nutrient requirements of Cyprinus carpio (NRC 2011).The control diet (CON) was formulated principally from fishmeal to contain approximately 34% crude protein and 7% lipid, the second dietary treatment evaluated was a diet with an inclusion level of 50% soybean meal (SBM) as the primary protein source protein.In addition, the content Vol.: (0123456789) of glycinin and β-conglycininin from soybean meal in SBM dietary treatment were analyzed by Jiangnan University value at 21.4% and 56.4%, respectively, the dietary treatment 11 + 7S was formulated by adding the mixture of glycinin and β-conglycinin at the same concentration analyzed in SBM dietary treatment.Likewise, the AKG dietary treatment was formulated by adding 1% AKG to the 11 + 7S dietary treatment.The diet preparation procedures were similar to those previously described in our lab (Ai et al. 2019), dietary ingredients were ground into fines to pass through a 60 mesh sieve, all ingredients in each group were thoroughly mixed in a mixer and prepared particles of 2 mm in diameter.The diets were stored at -20 ℃ until feeding.The carp (Cyprinus carpio Songpu) was provided by Heilongjiang Fisheries Research Institute, Chinese Academy of Fishery Sciences.Fish were acclimatized to the recirculating system for two weeks.Then, three hundred and sixty carps (4 ± 0.5 g) were randomly divided into 12 tanks (the radius of tank was 0.5 m and the height was 0.6 m, four tanks was a circulating water system).Each dietary treatment was randomly assigned to three replicate tanks (n = 3) and 30 fish were stocked per tank.The fish were fed to apparent satiation thrice daily (09:00, 13:00 and 17:00 h) for six weeks.During the feeding period, water temperature were maintained at 23 ℃ to 27 ℃, dissolved oxygen concentration was maintained over 5.0 mg L −1 , and ammonia nitrogen was kept below 0.5 mg L −1 .
Three fish were randomly selected from each tank and euthanized with an overdose (> 300 mg L −1 ) of Tricane methanesulphonate .Half of the proximal (PI), middle (MI), and distal intestines (DI) were stored at -20 ℃ for enzyme analysis.The other half was stored in liquid nitrogen and transferred to -80 ℃ for gene expression analysis (3 fish were pooled as a composite samples).In addition, the intestinal mucosa of 6 composite samples of each dietary treatment (3 fish were pooled as a composite samples, 2 composite samples per replicate tank) Table 1 Ingredients and analyzed composition of the test diet (dry basis) 1 The vitamin premix provided the following per kg of the diet: VA 8 000 IU, VC 500 mg, VD 3 3 000 IU, VE 60 mg, VK 3 5 mg, VB 2 30 mg, VB 6 15 mg, VB 12 0.5 mg, choline chloride 5 000 mg, nicotinic acid 175 mg, D-biotin 2.5 mg, inositol 1 000 mg, folic acid 5 mg, pantothenic acid 50 mg 2 The mineral premix provided the following per kg of the diet: Zn 25 mg, Cu 3 mg, Fe 25 mg, Mn 15 mg, I 0.6 mg, Co 0.1 mg, Se 0.4 mg 3 Abbreviationsg: CON = control; SBM = soybean meal; AKG: Alpha-ketoglutaric acid, 98. from CON, 11 + 7S, and AKG dietary treatment was scraped and stored at -80 ℃ for untargeted metabolomics analysis.

Gene expression analysis
According to the manufacturer's instructions, total RNA was extracted from the intestine with the RNAiso Plus (9109; Takara Biotech, Dalian, China).
The concentration and purity of RNA extracts were detected using the absorbance at 260 and 280 nm on a NanoDrop 2000 spectrophotometer (Thermo, New York, USA).Subsequently, purified 1 μl total RNA was reversely transcribed to cDNA using Prime-ScriptTM RT reagent kit gDNA Eraser (RR047A; Takara Biotech, Dalian, China), according to the manufacturer's instructions.The specific primers for the genes were designed by Primer Premier 5 Software and NCBI (https:// www.ncbi.nlm.nih.gov/), based on the carp's sequence and synthesized by Sangon (Sangon Biotec, Shanghai, China) (Table 2).In addition, β-actin was used as reference genes because of their stable expression.After the initial cDNA product was diluted ten times with nuclease-free water (Takara Biotech, Dalian, China), the real-time quantitative PCR was carried out using a StepOne Plus real-time PCR system (Applied Biosystems, Foster City, CA, USA).The gene expression levels were calculated using the 2 −ΔΔCT method, and the relative expression level of the gene in the control dietary treatment was used as a calibrator.

Metabolic statistical analysis and pathway analysis
The untargeted metabolomics of intestinal mucosa were analyzed by Majorbio Corporation, Shanghai, China.The MS and MS/MS mass spectra were matched with the metabolic database; the main databases were public databases (http://www.hm11+7S.ca/; https://metlin.scripps.edu/)and Majorbio selfbuilt databases (Majorbio Corporation, Shanghai, China).The positive and negative data were combined to get normalized data which was imported into the ropls R package (version 1.6.2) for principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA).
Principal component analysis (PCA) was used to observe the overall distribution and the degree of dispersion among the samples.The orthogonal partial least squares discriminant analysis (OPLS-DA) was used to distinguish the metabolic entirety among the differential metabolism between dietary treatments.
In the OPLS-DA analysis, variables with variable weight in the project (VIP) more than 1 and P value less than 0.05 were considered differential variables.Furthermore, to prevent the model from overfitting, 200 permutation tests are used to examine the fitting effect of the model.

Statistical analysis
All results were expressed as the mean ± standard deviation (SD).The data were subjected to a one-way analysis of variance (ANOVA).When a significant difference was observed (P < 0.05), Duncan's multiple range significance tests were used to resolve the discrepancy.All statistical analyses were performed using the SPSS 24.0 for Mac (SPSS Inc., Illinois, USA).

Growth performances
The growth performance data were showed in Table 3. FBW, WG, SGR, and PER for SBM dietary treatment significantly reduced compared to the CON, 11 + 7S, and AKG dietary treatments (P < 0.05).
It was also noted that FBW, WG, SGR, and PER of 11 + 7S and AKG dietary treatment were not significantly different from CON dietary treatment.Besides that, FCR values in SBM dietary treatment significantly increased (P < 0.05) compared to the other dietary treatments.Furthermore, SR and CF are not significantly different in all dietary treatments (P > 0.05).

Gene expression of AMPK and TOR signal pathway in intestines
The mRNA levels of AMPK, ACC, TOR, and 4E-BP genes in the PI, MI, and DI of fish were presented in Fig. 1 Compared with CON, for AMPK, the result indicated that the mRNA levels in the MI increased in SBM (P < 0.05), whereas it caused no significant change in the PI and DI of the fish (P > 0.05).
Furthermore, the mRNA level in the PI and DI was decreased in 11 + 7S (P < 0.05), whereas it caused no significant change in the MI of the fish (P > 0.05).
Upon treatment with AKG, the AMPK recovered to the control levels in the PI and DI (P > 0.05); For ACC, significant reductions in ACC caused by SBM were observed in the PI, MI, and DI (P < 0.05).A significant reduction caused by 11 + 7S was observed in the PI and DI (P < 0.05), and the decrease in the PI and MI was completely blocked by AKG supplementation (P < 0.05).However, AKG failed to block the reduction of ACC by soybean antigenic protein exposure in the DI of fish; For TOR, the mRNA levels of TOR in the PI and MI of the carp was significantly decreased in SBM dietary treatment (P < 0.05), the TOR in the PI, MI, and DI was decreased in 11 + 7S dietary treatment (P < 0.05).AKG successfully block the reduction in the PI, MI, and DI, and even significantly higher than the control (P < 0.05); For 4E-BP, SBM significantly increased the mRNA levels in the PI (P < 0.05).In contrast, it caused no significant change in the MI and DI of the fish (P > 0.05), and the mRNA level in the DI was decreased by 11 + 7S, whereas it caused no significant change in the PI and MI of the fish (P > 0.05).Upon treatment with AKG, the 4E-BP mRNA levels recovered to the control levels in the PI and DI (P > 0.05), and the 4E-BP mRNA levels increased even higher than the control levels in the MI (P < 0.05).

Gene expression of the apoptosis signaling pathway in intestines
The mRNA levels of caspase-3 and caspase-9 in the PI, MI, DI were presented in Fig. 2 Compared with CON, the mRNA levels of caspase-3 in the DI was increased significantly (P < 0.05), and caspase-9 in the MI was decreased in SBM dietary treatment (P < 0.05), caspase-9 in the MI was reduced in 11 + 7S dietary treatment (P < 0.05) and caspase-3 and caspase-9 in the DI was increased.Co-administration of the mixture of antigen protein and AKG, caspase-3 in the DI recovered to the control levels (P < 0.05), furthermore, the caspase-9 in the PI and MI increased and even higher than the control.It significantly decreased caspase-9 in the DI but higher than the control dietary treatment.

Gene expression of inflammatory cytokines in intestines
The mRNA levels of inflammatory cytokines in the PI, MI, DI were presented in Fig. 3

Gene expression of junction protein genes in intestines
The mRNA levels of claudin-3c, claudin-7, and claudin-11 in the PI, MI, DI were presented in Fig. 4. Compared with the CON, claudin-3c in MI and DI, claudin-7 in MI and occludin in PI were significantly down-regulated by SBM (P < 0.05), and claudin-7 in PI, claudin-11 in PI and DI were upregulated (P < 0.05).In contrast, there was not significantly changed in claudin-3c in PI, claudin-7 in DI, claudin-11 in MI, and occludin in MI and DI (P > 0.05).The claudin-7 in MI, occludin in PI were significantly down-regulated by 11 + 7S, and none of tight junction protein transcript genes was significantly down-regulated in all the intestines of fish.claudin-3c in PI and MI, claudin-7 in PI occludin in MI were significant increased in AKG dietary treatment, and none of tight junction protein transcript genes was significantly down-regulated in all the intestines of fish.

Modulated metabolite analysis
The OPLS-DA score plot was presented in Fig. 5 To directly evaluate the induced metabolic alterations and establish whether or not differences in metabolic profiles exist among the dietary treatment, pattern recognition techniques such as OPLS-DA, OPLS-DA permutation testing, and PCA-DA were applied.The OPLS-DA score plots of 11 + 7S-CON and AKG-11 + 7S had the cumulative values of R 2 Y being 99.4% and 99.7%, and Q 2 being 86.4% and 57.6%, respectively, suggesting that the model derived by OPLS-DA had a good fit and high predictability, and could be exploited in the subsequent analysis.
The identified metabolites was presented in Fig. 6.Univariate statistical results revealed 1338 signals (differing between CON and 11 + 7S dietary treatment) with P < 0.05 and VIP > 1 in ESI-, and 632 signals in ESI + .57 and 46 identified metabolites were up-regulated and down-regulated in the 11 + 7S dietary treatment compared to the control dietary treatment.Furthermore, compared to the 11 + 7S dietary treatment, 41 and 10 identified metabolites were upregulated and down-regulated in the AKG dietary treatment.But according to Fig. 6 (A), many of them were lipids and lipid-like molecules (54.22% of total components), organic acids, and derivatives (21.69% of complete components).According to Fig. 6 (A), main differential metabolites belong to lipids, lipidlike molecules, phenylpropanoids, and polyketides.Compared with the 11 + 7S dietary treatment and AKG dietary treatment, 37.78% were lipids and lipidlike molecules, and 28.89% were organic acids and derivatives (Fig. 6, B).

Pathway analysis
The metabolic pathway was presented in Fig. 7. Follow the identification of the significant differential metabolites, a targeted data analysis approach was employed.Pathway enrichment analysis combined with the topology analysis was conducted using Metaboanalyst to reveal the most relevant pathways influenced between CON and 11 + 7S dietary treatment, 11 + 7S and AKG dietary treatment.Based on the (2) sphingolipid metabolism.Six important metabolic pathways changed after the AKG treatment compared to 11 + 7S dietary treatment, (1) alanine, aspartate,

Discussions
In the present study, the survival rate of fish in each dietary treatment group did not differ significantly, indicating that a high dose of soy antigen protein did not kill carp.However, the growth performance (SGR, WG, FCR, and PER) of fish was significantly impaired by soybean meal.Interestingly, the WG decreased by 40% in soybean meal dietary treatment, while the 11 + 7S dietary treatment only decreased by 8.9% compared to dietary CON.In addition, the palatability of the feed containing soybean meal as a protein source is significantly lower than that of the two test diets containing purified soybean proteins.In the experiment, it was observed that the appetite of the 11 + 7S dietary treatment was much higher than that of the soybean meal dietary treatment.Secondly, soybean meal contains other anti-nutritional factors, such as proteinase inhibitors, lectin, saponin, and oligosaccharide that inhibits the growth performance.Proteinase inhibitors can complex with the corresponding protease and inhibit its activity in the intestine of fish (Santigosa et al. 2010).Lectin can cause histological changes in the intestine of fish, which leads to a decrease in the growth performance of fish (Iwashita et al. 2009), and saponin is a crucial factor causing inflammation of the intestine (Kortner et al. 2012).The disparity in growth performance between the 11 + 7S and SBM dietary treatments suggested that the effect of other anti-nutritional factors in soybean meal on the growth performance of carp may be more pronounced.Moreover, this difference may be due to the body's adaptation to antigens.Carp developed a tolerance to β-conglycinin, glycinin, and other anti-nutrient factors of soybean meal, so carp could tolerate high levels of anti-nutrient factors in the diet.The previous study by Uran (2009) showed that the gene expression levels of IL-1β and TNF-1β significantly increased when fed with soybean meal to carps in the early stage of intestinal inflammation, which began to recover after five weeks (Urán et al. 2009).The tolerance effect of the fish immune system to anti-nutritional factors in soybean meal requires further investigation.
The intestinal barrier can prevent the invasion of pathogenic bacteria and protect the health of the fish, while a tight junction is an integral part of the intestinal barrier.Several studies indicate that occludin, claudin-3c, and claudin-11 play a barrier-forming role in fish, whereas claudin-7 appears to have poreforming characteristic (Chasiotis et al. 2011;Chasiotis et al. 2012;Krause et al. 2010).Previous studies on mammals found that β-conglycinin reduces the mRNA expression of tight junction: occludin and ZO-1, and increases the permeability of intestinal epithelial cells (Zhao et al. 2014), which shows that β-conglycinin directly damages intestinal epithelial cells and destroy it's integrity.Furthermore, previous studies reported that addition of glycinin in the fish diet significantly reduced the expression of tight junction proteins like occludin, claudin3c, and claudin7 in the intestine (Jiang et al. 2015;Zhao et al. 2017).In the present study, both soybean meal and the mixture of β-conglycinin and glycinin significantly reduced the gene expression level of occludin and claudin-3c in the MI and DI of carps and increased the mRNA levels of claudin-11 in PI and DI.These findings suggest that soy meal and soy antigen protein may damage tight junction components.
Pre-inflammatory cytokines such as IL-1β, TNF-α and anti-inflammatory cytokines such as TGF-β1 play important roles in regulating intestinal inflammation (Fast et al. 2007).Although the anti-nutritional factors in soybean protein cause inflammation in the intestine of fish, some results showed that glycinin and β-conglycinin were the cause of intestinal inflammation, which increased the gene expression of tumor necrosis factor (TNF-α) and interleukin-1 β (IL-1β) (Li et al. 2017a, b;Jiang et al. 2015).In our experiment, both soybean meal and the mixture of β-conglycinin and glycinin increased the expression of TNF-α and IL-1β in PI and DI.Furthermore, both soybean meal and the mixture of β-conglycinin and glycinin decreased the expression of TGF-β1 in PI and MI.These results indicated that whether it was soybean meal or these two antigen proteins to replace fish meal induced intestinal inflammation and the most severe place was at DI. Interestingly, although soybean meal has more anti-nutritional factors, the gene expression of a pro-inflammatory factor in the soybean meal dietary treatment is less than that of the 11 + 7S dietary treatment, indicating that soy antigen protein is the main factor causing intestinal inflammation.
The intestinal mucosal system is in a dynamic equilibrium between cell proliferation and apoptosis and both inflammation and oxidative stress can induce cell apoptosis and ultimately affect health (Fuchs et al. 2011;Buttke et al. 1994).Although the three main apoptotic pathways are the mitochondrial pathway, death-ligand pathway, and endoplasmic reticulum pathway, caspase plays a vital role in the activation and apoptotic cascade reaction, which eventually leads to apoptosis, caspase-3 is considered to be the essential enzyme in the process of apoptosis.Previous experiments on Atlantic salmon Salmo salar have shown that dietary soybean meal increased the mRNA levels of apoptosis in the intestine (Sperstad et al. 2007;Sahlmann et al. 2013), purified β-Conglycinin increased the mRNA levels of apoptosis in the intestine (Luo et al. 2023a, b), both the purified glycinin and β-Conglycinin also induced apoptosis in the intestine of fish (Jiang et al. 2015;Duan et al. 2019).
In the present study, although the gene expression of caspase-3 in the PI and MI did not change significantly, soybean meal and the mixture of β-conglycinin and glycinin increased the gene expression of caspase in the DI.In addition, as the key promoter in the death ligand apoptosis pathway, the expression levels of caspase-8 increased significantly in both the SBM and 11 + 7S dietary treatment, whereas it had not changed in PI and MI.This data suggests that the apoptosis triggered by soybean meal and the combination of β-conglycinin and glycinin may have occurred predominantly in DI cells.Furthermore, soybean meal did not raise the gene expression of caspase-9 in DI of carp, although the mixture of β-conglycinin and glycinin increased the gene expression of caspase-9 in DI.The previous studies showed that caspase-9 is an important initiating factor in mitochondrion dependent apoptosis pathway (Sharifi et al. 2009).This suggests that the soy antigen protein is the primary agent responsible for triggering apoptosis in intestinal cells; nevertheless, this hypothesis has to be investigated further.Such studies show that anti-nutritional factors in soybean protein can destroy the tight junction of the intestine, induce intestinal apoptosis and cause oxidative damage to the intestine, which eventually results in poor growth performance.
AMPK is a highly conserved protein kinase that widely exists in eukaryotes.It can regulate the metabolism of fatty acids, cholesterol, and glucose and affect nutrients' metabolism through downstream proteins (Hayashi et al. 2000;Barnes et al. 2004).The expression of AMPK increased in the body under a low energy state and phosphorylated to provide energy.In the present study, the expression of AMPK in the intestine shows different results of soybean meal and the mixture of β-conglycinin and glycinin.Soybean meal significantly increased the expression level of AMPK in MI, whereas the combination of β-conglycinin and glycinin significantly decreased it in PI and DI; however, the mechanisms underlying these results require further investigation.ACC is one of the target proteins regulated by AMPK and is the rate-limiting enzyme of fatty acid synthesis (Abu-Elheiga et al. 1995).ACC is phosphorylated and its content1 declines when fatty acids are delivered into mitochondria to create ATP (Hardie et al. 2006).Based on the findings of the current study, which show that both soybean meal and the combination of β-conglycinin and glycinin decreased the expression level of ACC in PI, MI, and DI, and taking into account the results of the metabolic changes, it is possible that both soybean and the purified soy antigen protein significantly reduce the restriction of ACC on the synthesis of fatty acids, allowing for their consumption.mTOR is considered a direct receptor of ATP level, usually inhibited under low nutritional conditions, such as low amino acid and cellular ATP content.4E-BP is the downstream effector of TOR, and the improvement of its expression level would reduce the expression of TOR (Azar et al. 2006;Tain et al. 2009).In the present study, soybean meal decreased the expression level of TOR in PI and MI and increased the level of 4E-BP in PI, while the mixture of β-conglycinin and glycinin decreased in PI, MI, and DI and increased the level of 4E-BP in PI, MI, and DI.The differential expression of four energy control genes indicates that soybean meal and the purified soy antigen protein may influence the energy metabolism of fish intestines, particularly DI.
Metabolomics is commonly used to study the end products of cell metabolism (Wang et al. 2011).Taurine and cysteine were decreased in the muscle of turbot fed a plant-based diet, whereas glucose and tyrosine were increased, and dimethylamine and threonine were decreased in the liver metabolome (Wei et al. 2017).Based on the pathway analytic result in our study, glycerophospholipid metabolism and sphingolipid metabolism in intestines had been changed by the mixture of β-conglycinin and glycinin.Furthermore, lipids and lipid-like molecules in the intestines significantly increased.The changes in lipid substances and lipid metabolism pathway in the intestine indicated that soybean meal significantly increased the consumption of lipids in the intestine of fish to supplement the intestine with energy consumed by oxidative damage and cell regeneration, which was similar to the decrease of the expression level of ACC.Moreover, the fold change of diosmetin and genistein between the 11 + 7S and Control dietary treatments was 8.6 and 0.6, respectively, which means that the mixture of β-conglycinin and glycinin increased the content of diosmetin 8.6 and 0.6 times.Diosmetin, one flavonoid complex distributed widely in each plantation, has a strong antioxidant effect and anti-bacterial functions (Boutin et al. 1993).The previous studies showed that glycinin and β-conglycinin could cause oxidative damage to the fish intestine and reduce the activity of intestinal antioxidant enzymes (Zhang et al. 2014;Jiang et al. 2015;Luo et al. 2023a, b).Therefore, carp seems to absorb and utilize genistein and diosmetin from the diet to compensate for decreased antioxidant capacity in the intestine.The above data analysis suggests that the combination of β-conglycinin and glycinin can alter the intestinal energy consumption.
Although AKG did not influence feed utilization rate and weight gain in the present study, it effectively prevented the negative influences, such as reducing the expression of inflammatory cytokines, including IL-1 and TNF, in the DI.Furthermore, 1% AKG improved the expression of anti-inflammatory cytokines in the DI and reduced the injury caused by β-conglycinin and glycinin on the apoptosis and tight junction of intestinal epithelial cells.Additionally, according to the metabolic pathway analysis, 1% AKG can influence alanine, aspartate, glutamate metabolism, sphingolipid metabolism, pyrimidine metabolism, arginine biosynthesis, pantothenate and COA biosynthesis, and glycophorophospholipid metabolism.As a result, the availability of 1% AKG has the potential to prevent the damage that soybean protein would otherwise cause to the intestinal health of fish.

Conclusion
In conclusion, (1) the gene expression of intestinal inflammation and the apoptosis pathway is increased by both soybean meal and the soy antigen protein, with DI being the most severely affected location.
(2) The presence of soy antigen protein increases the rate of energy metabolism in the intestine of fish, in particular the rate at which lipids are metabolised in the intestinal mucosa.(3) In the intestine of fish, 1% AKG causes changes in both the glycerophospholipid metabolism and the sphingolipid metabolism, and this causes a reduction in the expression of intestinal inflammation and the apoptosis pathway.
. Compared with CON, the mRNA levels of tumor necrosis factor (TNF-α) in the MI, transforming growth factor (TGF-β1) in the PI and MI, interleukin-1 β (IL-1β) in the MI were significantly down-regulated (P < 0.05), and TNF-α in the PI, IL-1β in the PI and DI were upregulated by SBM (P < 0.05).In contrast, it caused no significant change in TNF-α in the DI and TGF-β1 in the DI (P > 0.05).The TNF-α in the PI and MI, TGF-β1 in the PI and MI, IL-1β in the MI were down-regulated by 11 + 7S (P < 0.05), and TNF-α in the PI and DI, TGF-β1 in the DI and IL-1β in the DI were up-regulated.Co-administration of the mixture

Fig. 1
Fig. 1 Effect of different treatments on the mRNA levels of ACC, AMPK, 4E-BP and TOR in the proximal intestine (PI), mid intestine (MI) and distal intestine (DI).Values are means

Fig. 2
Fig. 2 Effect of different treatments on the mRNA levels of caspase 3 and caspase 9 in the proximal intestine (PI), mid intestine (MI) and distal intestine (DI).Values are means of

Fig.
Fig. Effect of different treatments on the mRNA levels of TNF-α, TGFβ1 and IL-1β in the proximal intestine (PI), mid intestine (MI) and distal intestine (DI).Values are means of

Fig. 4
Fig. 4 Effect of different treatments on the mRNA levels of in the proximal intestine (PI), mid intestine (MI) and distal intestine (DI).Values are means of three replicates, with standard

Table 3
Growth performances of Songpu mirror carp fed with different dietary treatments for 6 weeks 1 In the same row, values with different small letter superscripts mean significant difference (P < 0.05), while with the same or no letter superscripts mean no significant difference (P > 0.05). 2 Abbreviations: IBM, Initial body weight; FBW, final body weight; SR, Survival rate; WG, Weight gain; SGR, specific growth rate; PER, protein efficiency ratio; FCR, feed conversion ratio; and CF, condition factor