Carbohydrates are relatively inexpensive energy substance among the three major nutrients, which are essential and important nutrient in aquatic animal. An appropriate level of carbohydrate to the diet contributes to the reduction of protein and lipids consumed by fish for energy supply, improves the anabolism of protein and lipids by fish and diminishes the excretion of ammonia nitrogen into environment(Reference Beamish, Hilton and Niimi1). However, the inefficiency of carbohydrate utilisation by fish is comparatively weak and excess carbohydrate in the diet often induces persistent hyper glycaemia, excessive glycogen and lipid deposition and a decline in immunity, leading to slow growth and even mortality in fish(Reference Dixon and Hilton2). The appropriate carbohydrate level in the diet of most carnivorous fish is no more than 20 %, while for herbivorous and omnivorous fish it ranges between 30 % and 50 %(Reference Kamalam, Medale and Panserat3). For the most part, herbivorous and omnivorous fish make greater use of carbohydrate in their diet than carnivorous fish(Reference Li, Zhu and Han4,Reference Shimeno, Kheyyali and Shikata5) , which is related to the high or the low carbohydrate content of the diet available to fish in nature. In a study by Legate et al. (Reference Legate, Bonen and Moon6) on glucose tolerance in rainbow trout (Oncorhynchus mykiss) and black bullhead catfish (Ameiurus melas), rainbow trout showed high blood glucose level for up to 360 min after ingestion or injection of glucose, while black bullhead catfish maintained the level for only 60 min before returning to pre-feeding, and similar results were seen in studies in herbivorous and omnivorous fish. Previous hypotheses suggest that the causes of poor carbohydrate utilisation in fish may share similarities with mammalian type 2 diabetes(Reference Yan, Qin and Deng7). In clinical practice, the use of oral medication may be a potential solution to remedy this non-routine phenotype, which gives a perspective to relax the restrictions on carbohydrate application to carnivorous fish(Reference Wang8). Although herbivorous fish are inherently high in carbohydrate utilisation, studies(Reference He, Jia and Zhang9,Reference Yue, Huang and Yang10) have also shown that oral administration of some drugs can be effective in alleviating high sugar stress and further promoting growth performance. In light of extensive follow-up investigations, some particular amino acids like taurine might be a suitable option(Reference Kamalam, Medale and Panserat3).
Taurine (2-amino ethane sulfonic acid) is considered to be a conditionally essential amino acid in animals and was first identified in 1827 by the German scientist Tiedemann from bovine bile(Reference Tiedemann and Gmelin11). In nature, taurine is mainly found in the tissues of marine animals and mammals, while most plants and bacteria contain very little or even no taurine(Reference Liu12). Earlier research reported that taurine deficiency led to the development of ‘green liver syndrome’ and reduced growth performance of sea bream (Pagrus Major) and yellowtail (Seriola quinqueradiata) fed fishmeal-free diets(Reference Takagi, Murata and Goto13). However, the taurine supplementation in the diet restored growth performance and improved feed utilisation, body protein and fat deposition in rainbow trout(Reference Gaylord, Teague and Barrows14). Meanwhile, in mammals, taurine has also been employed in diabetes control due to its insulin-like structure, which enhances insulin sensitivity and reduces insulin resistance(Reference Mccarty15). However, scholars rarely focused on how taurine-regulated nutrient metabolism, particularly carbohydrates in early studies on aquatic animals(Reference Coutinho, Simoes and Monge-Ortiz16).
Grass carp (Ctenopharyngodon idellus) is a typical herbivorous fish that makes greater use of the carbohydrates in its diet compared with carnivorous fish, which is an economically important freshwater fish all over the world(Reference Li, Wang and Liu17). Since fish have a weak ability to synthesise taurine on their own(Reference Goto, Tiba and Sakurada18), they are prone to poor growth and even green liver syndrome in some fish when deficient in taurine, but when the feed is supplemented with taurine, the growth performance of these fish is significantly improved and the incidence of green liver syndrome is greatly reduced(Reference Takagi, Murata and Goto19). There are few studies on the nutritional requirements for taurine in freshwater fish, especially on the appropriate supplemental dose of taurine in high-carbohydrate diets. Therefore, an 8-week experiment was performed to investigate implications on growth performance, glucose metabolism and insulin pathway-induced taurine supplementation in a high-carbohydrate diet for grass carp.
Material and methods
Ethics statement
This study was carried out following the recommendations of Care and Use of Laboratory Animals in China, Animal Ethical and Welfare Committee of China Experimental Animal Society. The protocol was approved by the Animal Ethical and Welfare Committee of Guangdong Ocean University (Guangdong, China).
Experimental diets
Six iso-nitrogenous (30·37 %) and iso-lipidic (2·37 %) diets were designed for this experiment, of which two diets with different carbohydrate level (31·49 %, 38·61 %) were used as positive control (31·49 %, PC) and negative control (38·61 %, T00), respectively, and then supplemented with 0·05 % (T05), 0·10 % (T10), 0·15 % (T15) and 0·20 % (T20) taurine in the negative control, respectively, as experimental groups, and the composition and chemical analysis of this experimental diets are shown in Table 1. After the dry ingredients have been ground and passed through 60 mesh sieve, the ingredients were accurately weighted according to the formation proportions and mixed thoroughly with a V-type (Zhejiang Chint Electric Co., Ltd., JS-14S type) mixer and finally with fish oil, soybean lecithin, which must finally pass through an 80 mesh sieve again to prevent lumps and uneven mixing and then added 300ml/kg diet of water to form a moist dough. The dough was extruded in 2·5 mm pellets by F-26 twin-screw extruder (South China University of Technology, F-75 type), and the pellets were air dried to about 10 % moisture. Finally, the product was sealed in bags and stored at −20°C freezer until required.
Formulation and proximate composition of experimental diets (g/kg dry matter)
* Premix (g /kg): vitamin A 120 000 μg; vitamin D3 40 000 μg; vitamin E 480 mg; vitamin K3 200 mg; vitamin B1 200 mg; vitamin B2 280 mg; vitamin B8 240 mg; vitamin B12 0·6 mg; calcium pantothenate 720 mg; nicotinic acid 1000 mg; folic acid 60 mg; biotin 1·2 mg; VC phosphatase 6850 mg; creatine 3200 mg; magnesium 4000 mg; iron 4800 mg; zinc 2000 mg; manganese 800 mg; copper 160 mg; cobalt 12 mg; selenium 4 mg; iodine 40 mg (obtained from Zhanjiang Yue Hai Feed Co. Ltd.).
† Moisture, crude protein, crude lipid, carbohydrate and taurine contents were measured value.
Fish rearing and experimental conditions
Grass carp was purchased from Xiangyin County Fine Seed Factory of Hunan Province, temporarily raised in chetianjiang Reservoir cage (5 m × 4 m × 5 m), and fed with ordinary commercial feeds for a 1 week to adapt to the experimental conditions before the start of the formal rearing experiment. After starvation for 24 h, grass carp with healthy body and uniform size, with an average weight 100 g, were randomly selected and placed into 18 cages (2 m × 2 m × 2 m), three replicate groups per treatment with 30 fish in each cage. Water temperature during the feeding trial was 28–32°C, pH 7·2–7·4 and dissolved oxygen over 5·0 mg/l. Each experimental feed was given in triplicate to the fish twice daily at 07.00 and 17.30 until apparent satiety for 8 weeks, and the amount of feed consumed was recorded.
Sample collection
When the 8-week breeding experiment was over, the fish were starved for 24 h. After all fish in each cage were anaesthetised with diluted eugenol (1:10 000; Shanghai Reagent Corp.), and then the total weight and number of all fish were counted to facilitate the calculation of weight gain rate, specific growth rate, survival rate, feed coefficient rate and protein efficiency rate. After weighting, six fish in each cage were randomly sampled in order to collect blood in 1·5 ml Eppendorf tubes with 1 ml sterile syringe, and then stored at 4°C overnight to allow for serum precipitation(Reference Yin, Liu and Tan20). The following day, the resultant mixture was finally centrifuged at 4000 rpm for 10 min at 0–4°C, and the serum was collected and stored at –80°C for later analysis of serum biochemical parameters(Reference Qian, Yin and Liu21). Three fish from each cage were collected and quickly stored at –20°C refrigerator for analysis of whole-body composition(Reference Liu, Yang and Dong22). The livers of fish were then dissected and kept in a 4 % paraformaldehyde solution for histological analysis, as described by Sahlmann et al. (Reference Sahlmann, Sutherland and Kortner23) Livers and muscles were quickly collected from four fish in each cage to analyze enzyme activity, amino acid composition and glycogen content, while intestines were collected to analyse to digestive enzymes and quickly preserved in liquid nitrogen. The livers of three fish from each cage were placed in RNA later to extract mRNA for genes related to insulin signalling pathway. After the end of sampling, all samples were immediately stored at –80°C until analysis.
Nutritional composition of feed and whole fish
The composition of diets and fish was analysed by standard procedures (Association of Official Analytical chemists, AOAC, 2005). Moisture content to constant weight was measured by an oven at 105°C, crude protein content was determined by an Automatic Kjeldahl system (Kjeletc 8400, FOSS Technology and Trade Co., Ltd.) and crude lipid content was measured by an Automatic Fat Extractor (ANKOM XT15I Extractor, ANKOM Technology Co., Ltd.). Amino acids profiles in the liver were determined using an automated amino acid analyzer (Hitachi L-8500A; Hitachi).
Serum biochemistry, glycogen content and enzyme activity assay
Blood glucose (F006–1–1), total cholesterol (T-CHO, A111–1–1), TAG (A110–1–1), HDL-cholesterol and LDL-cholesterol (HDL, A112–1–1; LDL, A113–1–1) and liver/muscle glycogen were measured by commercial kits from Nanjing Jiancheng Institute of Biological Engineering, China. Serum insulin (ml022831) and insulin-like growth factor-1 (ml022803), the levels of intestinal α-amylase (ml036449), trypsin (ml036384), lipase (ml036371), hepatic insulin-like growth factor-1 receptor (ml028391), glucokinase (ml024808), hexokinase (HK, ml076603), pyruvate kinase (PK, ml037329), glucose-6-phosphatase (G6Pase, ml076993), phosphoenolpyruvate carboxykinase (ml036430) and fatty acid synthetase (ml036370) were tested according to the instructions of commercial kits (Shanghai Elisa Biotech Co., Ltd.) from ELISA.
Real-time quantitative RT-PCR analysis of gene expression in the liver
Total RNA was extracted from the livers of three fish in each treatment with a Trans Gen Biotech (Beijing, China) RNA kit and total RNA quality and concentration (OD260/OD280, 1·8–2·1) were measured by UV spectrophotometer Nanodrop 2000c (Thermo) and electrophoresed on a 1 % agarose gel (28s, 18s ribosomal RNA bands were clear, and the brightness of the 28s band was about twice that of the 18s band) for integrity testing, as described by Luo et al. (Reference Luo, Feng and Jiang24).
After passing the test, the cDNA first strand was synthesised by reverse transcription using Accurate Biology Evo M-MLV kit (Hunan, China), as described by Gan et al. (Reference Gan, Jiang and Wu25), and the obtained cDNA was stored at 20°C refrigerator for the analysis of real-time quantitative RT-PCR. In this experiment, specific upstream and downstream primer sequences were designed as shown in Table 2. The volume of all RT-PCR reactions was 10 µl (1µl cDNA, 0·8µl primer, 3·2µlRNseFree dH2O and 5µl SYBR® Green Real-Time PCR Master Mix) on a Light Cycler 480 Biosystems Real-time PCR system (Shanghai, China). Relative gene expression was calculated by the 2−ΔΔCT method using β-actin as the reference gene(Reference Livak and Schmittgen26).
List of primer pairs used for the real-time PCR analysis of grass carp hepatic gene expression
β-actin, reference gene; Igf-1, insulin-like growth factor 1; Igf-1R, insulin-like growth factor 1 receptor; Ir, insulin receptor; Irs1, insulin receptor substrate; Pi 3 k, phosphatidylinositol 3-kinase; Akt1, protein kinase B; Gs3kβ, glycogen synthase kinase 3 β; Foxo1, Fork head transcription factor 1; Hif-1, hypoxia inducible factor 1 α.
Intestinal microflora analysis
Microbial DNA was extracted using the Hi Pure Soil DNA Kits (or Hi Pure Stool DNA Kits) (Magen) according to manufacturer’s protocols. The full-length 16S rDNA was amplified by PCR (95°C for 2 min, followed by 35 cycles of 95°C for 30 s, 60°C for 45 s and 72°C for 90 s, with a final extension 72°C for 10 min) using primers 27F: AGRGTTYGATYMTGGCTCAG; 1492 R: RGYTACCTTGTTACGACTT(Reference Singer, Bushnell and Coleman-Derr27). The PCR reaction was carried out in a 50 μl reaction volume with Trans Gen High-Fidelity PCR Super Mix (Trans Gen Biotech), 0·2 μM forward and reverse primers and 5 ng template DNA. For more information, see this reference(Reference Yin, Liu and Tan28).
Statistical analysis
Homogeneity tests for normality and variance were performed on all data using Social Science Software Report (vision 26, SPSS Inc.). The following are the analysis methods used in this experiment: (1) we performed Dunnett’s test for the group PC and each group T (T00, T05, T10, T15 and T20, respectively) and (2) the orthogonal polynomial comparison method was applied to identify whether the effects were linear or quadratic. Our data analysis methods are specifically referenced in this literature(Reference Yin, Liu and Tan20). The * in Tables 3–6, and in the tables to the right of the figures, illustrates the magnitude of any significant differences seen between each treatment group and the positive control by Dunnett’s test. ‘*’ denotes P < 0·05; ‘**’ denotes P < 0·01; ‘***’ denotes P < 0·001; no ‘*’ denotes no significant difference between this group and the Group PC.
Growth performance of grass carp fed with different diets
The * in the upper right corner of the figures means a significant difference between each Group T treatment and Group PC was found by Dunnett’s test. ‘*’ denotes P < 0·05; ‘**’ denotes P < 0·01; ‘***’ denotes P < 0·001; no ‘*’ denotes no significant difference between this group and the Group PC.
Growth performance was calculated using the following formula:
Weight gain rate (WGR, %) = 100 × (final body weight (g) − initial body weight (g))/initial body weight (g).
Specific growth rate (SGR, %) = 100 × (ln(final body weight (g)) – ln(initial body weight (g))/days of trial.
Feed conversion ratio (FCR) = feed intake (g)/ (final body weight (g) − initial body weight (g)).
Survival rate (SR, %) = 100 × (final fish number)/initial fish number.
Protein efficiency rate (PER, %) = (final body weight (g) − initial body weight (g))/protein intake.
Condition factor (CF, g/cm3) = 100 × body wet weight (g)/body length (cm)3.
Hepatosomatic indices (HSI, %) = 100 × (liver wet weight (g)/body wet weight (g).
Viscerosomatic index (VSI, %) = 100 × viscera wet weight (g)/body wet weight (g).
Body composition of grass carp (%, wet matter)
The * in the upper right corner of the figures means a significant difference between each Group T treatment and Group PC was found by Dunnett’s test. ‘*’ denotes P < 0·05; ‘**’ denotes P < 0·01; ‘***’ denotes P < 0·001; no ‘*’ denotes no significant difference between this group and the Group PC.
Plasma parameters and hepatic IGF-1R quantity of grass carp
IGF-1, insulin-like growth factor 1; IGF-1R, insulin-like growth factor 1 receptor; TG, triglycerides; T-CHO, total cholesterol; LDL, low-density lipoprotein; HDL, high-density lipoprotein.
The * in the upper right corner of the figures means a significant difference between each Group T treatment and Group PC was found by Dunnett’s test. ‘*’ denotes P < 0·05; ‘**’ denotes P < 0·01; ‘***’ denotes P < 0·001; no ‘*’ denotes no significant difference between this group and the Group PC.
Hepatic amino acids profile (mg/g dry matter) of grass carp
His, histidine; Tau, taurine; Ser, serine; Arg, arginine; Gly, glycine; Asp, aspartic; Glu, glutamic; Thr, threonine; Ala, alanine; Pro, proline; Cys, cysteine; Lys, lysine; Tyr, tyrosine; Met, methionine; Val, valine; Ile, isoleucine; Leu, leucine; Phe, phenylalanine.
The * in the upper right corner of the figures means a significant difference between each Group T treatment and Group PC was found by Dunnett’s test. ‘*’ denotes P < 0·05; ‘**’ denotes P < 0·01; ‘***’ denotes P < 0·001; no ‘*’ denotes no significant difference between this group and the Group PC.
Results
Growth performance, survival, feed utilisation and protein efficiency rate
Table 3 shows the effects of high-carbohydrate diet and dietary taurine level on the growth performance, feed utilisation, protein efficiency rate and morphological indices for grass carp. There was no significant difference in survival rate, protein efficiency rate and feed conversion ratio in group T00 with higher carbohydrates diet compared with group PC, and neither were any of these parameters significantly affected by supplementation of taurine at any level. However, final body weight (FBW), weight gain rate and specific growth rate displayed a quadratic model with increasing of dietary taurine level and attained their extreme in group T10. Weight gain rate value was used in the quadratic function model for the estimation of optimal dietary taurine level (Fig. 1), and the analysis results showed that optimal dietary taurine supplementing level for maximum weight gain rate of grass carp was estimated to be 0·08 %. Group T00 in higher carbohydrate diet improved significantly viscerosomatic index, hepatosomatic indices, condition factor indexes when compared with the group PC. Meanwhile, viscerosomatic index, hepatosomatic indices and condition factor showed a quadratic model with increasing dietary taurine level, viscerosomatic index and HSI reached their lowest values in group T10, condition factor in group T20.
Relationship between grass carp WGR and dietary taurine level based on the analysis of a quadratic function model. WGR, weight gain rate.
Whole-body components
Table 4 shows the effects of high-carbohydrate diet and dietary taurine level on the body composition of grass carp. There were no significant differences in moisture, crude protein and crude lipid content among all treatments. In addition, the crude protein content tended to first increase and then decrease with the increase of dietary taurine level, reaching their maximum value in group T10.
Plasma biochemical indices and hepatic IGF-1 receptors in grass carp
Table 5 shows the effects on plasma biochemical parameters as well as hepatic insulin-like growth factor receptors induced by taurine in a high-carbohydrate diet. Compared with the group PC, plasma TAG, triglyceride (TC) level increased significantly with increasing dietary taurine level high-carbohydrate diet where the group T00 was located, while plasma glucose, insulin, IGF-1 and hepatic IGF-1R level decreased significantly. In addition, plasma insulin, TG, LDL, HDL and hepatic IGF-1R level showed a quadratic function model with increasing dietary taurine level, with insulin and TG level reaching a minimum in group T00 and IGF-1R level reaching a maximum in group T00. While plasma TC and IGF-1 presented a linear trend, reaching extreme value in group T15.
Hepatic amino acid profile of grass carp
Table 6 shows the effect on hepatic amino acid composition of grass carp caused by taurine in a high-carbohydrate diet. The results revealed that the content of hepatic Glu, Pro, Cys and Lys in grass carp fed a high-carbohydrate diet decreased significantly, while the remaining amino acids were not significantly different. However, the composition of each amino acid in liver showed a trend of first increasing and then decreasing with increasing dietary taurine level and then reached the maximum in group T05 and T10, while Tau content reached the maximum in group T20.
Intestinal digestive enzymes analysis
The results in Fig. 2 show the effect on intestinal digestive enzymes caused by taurine in a high-carbohydrate diet. Compared with the group PC, trypsin and amylase activities were significantly lower in the high carbohydrate of the group T00, and there was no significant difference in lipase activity. Meanwhile, trypsin and lipase activities showed a quadratic function model with increasing dietary taurine level, reaching their maximum value in group T15 (Fig. 2(a) and (b)); amylase activity exhibited a linear trend with increasing dietary taurine level, reaching its maximum value in group T15 (Fig. 2(c)).
Intestinal digestive enzyme activity of grass carp fed varying levels of dietary taurine in a high-carbohydrate diet. Notes: Red spots: the PC; black spots: the T00-T20. The * in the table to the right of the figures signifies the magnitude of any significant differences between each group T treatment and Group PC found by Dunnett’s test. ‘*’ denotes P < 0·05; ‘**’ denotes P < 0·01; ‘***’ denotes P < 0·001; no ‘*’ denotes no significant difference between this group and the Group PC.
Glycogen content analysis
The results in Fig. 3 show the effect on glycogen content caused by taurine in a high-carbohydrate diet. There was no significant difference in hepatic glycogen content among all treatments (Fig. 3(a)). Muscle glycogen content was significantly lower in group T00 when compared with the group PC. Meanwhile, muscle glycogen content showed a quadratic curve model with the increasing dietary taurine level, reaching its lowest value in group T10 (Fig. 3(b)).
Hepatic (a) and muscle glycogen (b) content of grass carp fed varying levels of dietary taurine in a high-carbohydrate diet. Red spots: the PC; black spots: the T00-T20. The * in the table to the right of the figures signifies the magnitude of any significant differences between each group T treatment and Group PC found by Dunnett’s test. ‘*’ denotes P < 0·05; ‘**’ denotes P < 0·01; ‘***’ denotes P < 0·001; no ‘*’ denotes no significant difference between this group and the Group PC.
Hepatic enzyme activities associated with glycolipid metabolism
Figure 4 shows the effects on enzyme activities related to glucolipid metabolism induced by taurine in a high-carbohydrate diet. Compared with the group PC, the activities of GK, PK, phosphoenolpyruvate carboxykinase and fatty acid synthetase enzymes were significantly lower in group T00 in high-carbohydrate diet, while the activities of HK and G6P enzymes were not significantly different. Meanwhile, GK, PK and phosphoenolpyruvate carboxykinase activities showed a quadratic function model with increasing dietary taurine level, while HK, G6P and fatty acid synthetase activities exhibited a linear trend. Among them, GK and PK activities reached extreme values in group T10, while phosphoenolpyruvate carboxykinase and fatty acid synthetase activity in group T15. Dietary taurine level did not significantly affect the HK and G6P activities.
Hepatic enzyme activities associated with glycolipid metabolism in grass carp fed varying levels of dietary taurine in a high-carbohydrate diet. GK, glucokinase; HK, hexokinase; PK, pyruvate kinase; G6Pase, glucose-6-phosphatase; PEPCK, phosphoenolpyruvate; FAS, fatty acid synthetase. Red spots: the PC; black spots: the T00-T20. The * in the table to the right of the figures signifies the magnitude of any significant differences between each group T treatment and Group PC found by Dunnett’s test. ‘*’ denotes P < 0·05; ‘**’ denotes P < 0·01; ‘***’ denotes P < 0·001; no ‘*’ denotes no significant difference between this group and the Group PC.
PI3K/AKT pathway-related gene expression in the liver
Figure 5 shows the effect on phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT) pathway-related gene expression in the liver induced by taurine in a high-carbohydrate diet. Group T00 in high-carbohydrate diet downregulated significantly the relative expression of Ir, Irs1, pi 3 k, akt and gs3kβ in the liver of grass carp, upregulated significantly the relative expression of igf-1R and foxo1 but had no significant effect on the relative expression of igf-1. After different taurine-level supplementation in a high-carbohydrate diet, the relative expression of Ir, Irs1, pi 3 k, akt, gs3kβ and hif1α showed a quadratic pattern of first increasing and then decreasing, reaching a maximum value in group T10. While igf-1 expression showed a linear trend, decreasing from a maximum value in group T05. The relative expression of igf-1R and foxo1 exhibited a quadratic function model with increasing dietary taurine level, reaching a minimum value in group T20.
Hepatic PI3K/AKT signaling pathway-related gene expression of grass carp fed varying levels of dietary taurine in a high-carbohydrate diet. Notes: Red spots: the PC; black spots: the T00-T20. The * in the table to the right of the figures signifies the magnitude of any significant differences between each group T treatment and Group PC found by Dunnett’s test. ‘*’ denotes P < 0·05; ‘**’ denotes P < 0·01; ‘***’ denotes P < 0·001; no ‘*’ denotes no significant difference between this group and the Group PC.
The impact of different diets on intestinal microflora
Richness and diversity analysis
The effects of carbohydrate and taurine levels on the diversity and abundance of intestinal flora in grass crap are shown in Table 7. There was no significant difference in Seq, operational taxonomic unit (OTU), Shannon and Simpson indexes among all treatments. Compared with the group PC, group T00 in high-carbohydrate diet did not differ significantly in Chao and abundance-based coverage estimator (ACE) indexes. Meanwhile, Chao and ACE indexes showed a quadratic function model with increasing dietary taurine level, reaching a maximum value in group T05 and T10.
Sequence, OTUs, alpha diversity and richness metrics for the intestinal bacterial analysis
Taxonomic analysis of species
At the phylum level, Fig. 6(a) shows the effect of taurine in a high-carbohydrate diet on the relative abundance of intestinal flora. Among the top three phylum levels, Bacteroidetes, Fusobacteria and Firmicutes, were relatively more abundant and dominant. At the genus level, Fig. 6(b) shows the effect of taurine in high-carbohydrate diet on the relative abundance of intestinal flora. Among the top three genus level, Bacteroides, Cetobacterium and Erysipelatoclostridium were relatively more abundant and dominant genera.
Flora abundance at the phylum (a) and genus (b) level, in the gut of grass carp fed varying levels of dietary taurine in a high-carbohydrate diet.
Linear discriminant analysis
Figure 7 showed the differences in intestinal floral of grass carp among all treatments using linear discriminant analysis. The results of Lef se analysis showed that the abundance of Stenoxybacter, Lachnospiraceae and Mannheimia was significantly higher in group T05 than that in group PC, the abundance of Chloroflexi, Caldilineaceae, Anaerolineae, Limnobacter and Caldilineales in group T10 was significantly higher than that in group PC and the abundance of Flavobacteriales, Obscuribacterales, Breznakia, Melainabacteria in group T20 was significantly higher than that in group PC.
Lef se analysis of differences in intestinal microbial compositions between grass carp fed varying levels of dietary taurine in a high-carbohydrate diet (LDA score ≥ 3, P < 0·05).
Discussion
Among the different dietary fishes, herbivorous fishes can make the most use of carbohydrate in the feed. Jiang et al. (Reference Jiang, Li and Liu29) concluded that the appropriate carbohydrate level was 31·94 % in the diet of blunt snout bream (Megalobrama amblycephala); in the studies of Li et al. and Tian et al. (Reference Li, Zhu and Han4,Reference Tian, Liu and Hung30) , it was noted that the appropriate carbohydrate level in the diet of grass carp was 38 % and 33 %, respectively, which may be due to the difference in carbohydrate source of diets, with grass carp utilising corn starch more than wheat starch,(Reference Kamalam, Medale and Panserat3) and this similar findings were also found in the Labeo rohita (Reference Mohapatra, Sahu and Chaudhari31). In this study, the intestinal trypsin, α-amylase and lipase activities in high-carbohydrate diet (38·61 %) of grass carp were significantly inhibited, while weight gain rate and specific growth rate showed a decreasing trend, but the difference was not significant, which demonstrated that the grass carp was able to tolerate high carbohydrate level without significant effects on growth. In contrast, in a study by Hu et al. (Reference Hu, Chen and Zhang32), it was concluded that grass carp fed 31–37 % carbohydrate level diet would promote compensatory secretion of trypsin to enhance the absorption of dietary protein, although it would lead to a decrease in weight gain rate, intestinal amylase and lipase activities. In study of Ren et al. (Reference Ren, Jia and Ge33), although isonitrogenous and isoenergetic diets with dietary carbohydrate levels of 32 % to 42 % did not cause significant inhibition of the growth and intestinal protease activity of blunt snout bream, there was a tendency for a decrease in intestinal protease activity, which was consistent with the results of this experiment. After taurine supplementation in high-carbohydrate diet, weight gain rate, specific growth rate, trypsin, α-amylase and lipase activities of grass carp were promoted and showed significant maxima at taurine levels of 0·07–0·13 %, in agreement with the results of Luo et al. and Xu et al. (Reference Luo, Wen and Wang34,Reference Xu, Ming and Zhang35) The appropriate taurine level in grass carp fed 38·61 % carbohydrate level feed was 0·08 % as judged by weight gain rate.
Grass carp, as a typical herbivorous fish, is believed to be effective in converting sugars from feed into glycogen and lipids for storage. Studies by Guo et al. (Reference Guo, Liang and Fang36) indicated that the crude fat of whole body composition and liver, hepatic glycogen and hepatosomatic indices were significantly higher with the increase of dietary carbohydrate for grass carp, and the reason for the difference could be due to fat and glycogen deposition in the liver of grass carp. In contrast, in our research, grass carp fed a 38 % carbohydrate diet showed a decrease in crude protein and muscle glycogen, an increase in viscerosomatic index and condition factor and no significant difference in hepatic glycogen and crude fat, a discrepancy that excess carbohydrates were stored preferentially as fat rather than glycogen in grass carp. After taurine supplementation in high-carbohydrate diet, crude protein content increased and crude fat content of body composition decreased in T05 and T10 groups compared with group T00, which was consistent with the results of studies on Persian Sturgeon (Acipenser persicus)(Reference Hoseini, Hosseini and Eskandari37), rainbow trout (Oncorhynchus mykiss)(Reference Xu, Xu and Zheng38) and Atlantic salmon (Gadus morhua)(Reference Marit, Kari and ElMowafi39).
In a comparative study of glucose tolerance in fish of different diets(Reference Li, Liu and Chen40), it was found that herbivorous fish tended to exhibit excellent glycaemic clearance. In the present study, a decreasing trend in serum glucose, insulin and IGF-1 levels was observed in grass carp fed a high-glucose, similar to the findings of Hu et al. (Reference Hu, Chen and Zhang32) However, previous studies(Reference Cai, Liang and Yuan41) indicated a significant increase in serum glucose, insulin and insulin receptor levels in grass carp fed on high dextrin (42 %) levels compared with the low dextrin group (25 %), which difference could be attributed to the different carbohydrate level and source in the study. Furthermore, in a study about blunt snout bream(Reference Zhou, Liu and Ge42), no significant differences in serum glucose levels were observed in blunt snout bream fed with 0–47 % carbohydrate levels, but lowered complement C3 and C4 levels, which may inhibit IGF-1 and insulin expression by decreasing GH expression(Reference Roith43,Reference Gutiérrez, Párrizas and Carneiro44) . No significant changes in serum glucose levels and significant increases in serum insulin and IGF-1 levels were observed in grass carp after taurine supplementation in high-carbohydrate diet, which was attributed to the better pro-islet effect of taurine relative to carbohydrates(Reference Kamalam, Medale and Panserat3). In the present study, serum insulin and IGF-1 in grass carp increased with the increase of taurine level in the feed, which was consistent with the above results. In recent years, dietary control of micronutrients to treat insulin resistance has attracted increasing attention, and essential amino acid deficiency or excess plays an important role in maintaining homoeostatic balance in the body. Zhou et al. (Reference Zhou, Huang and Lin45) found that the PI3K/AKT/target of rapamycin insulin signaling pathway mediates chronic leucine supplementation induced hepatic glucose metabolism and lipogenesis in juvenile golden pompano (Trachinotus ovatus). Studies have shown that a methionine-restricted diet can improve overall insulin sensitivity, primarily by inhibiting hepatic gluconeogenesis and enhancing hepatic akt phosphorylation(Reference Newgard, An and Bain46). Studies in grass carp showed a significant increase in growth performance at a lysine supplementation of 2·39 %(Reference Huang, Liang and Ren47). Our results showed that hepatic amino acid content in grass carp fed a high-carbohydrate diet decreased significantly, and amino acids first rose and then fell in the liver through taurine supplementation, in correspondence with the growth of this experiment. It may be that high levels of taurine put the intestine in a highly acidic environment, which is not conducive to growth. All free amino acid contents in liver and muscle of Atlantic salmon (Salmo salar) were increased when dietary taurine levels were appropriate(Reference Espe, Ruohonen and El-Mowafi48). Similarly, taurine affected most of the free amino acid content in both liver and muscle of Japanese Seabass (Lateolabrax japonicus)(Reference Liu, Wang and Liang49), which facilitated amino acid deposition. It played a very important role in the growth, development and metabolic regulation of organisms.
The liver is the major metabolic organ in fish and the regulation of liver metabolism by insulin and IGF-1 relies on their interaction with specific receptors to generate a cascade response that results in transcriptional-level regulation of genes related to glucose metabolism. Several studies(Reference Gutiérrez, Párrizas and Carneiro44,Reference Baños, Moon and Castejón50) have concluded that igf-1 receptors are higher in number in fish tissues compared with insulin receptors and have an important role in glucose metabolism for fish. PI3K/AKT is a signalling pathway closely related to metabolism(Reference Huang, Liu and Guo51). A recent study(Reference Sehat, Tofigh and Lin52) indicated that apart from regulating cell growth, proliferation and metabolism by relying on PI3K/AKT and MAPK signalling pathways, IGF-1R can also undergo nuclear translocation after SUMOylation modification to establish a direct link between the cell membrane and the nucleus to achieve regulation of specific gene transcriptional expression. Many downstream targets can be phosphorylated by akt1, among them are inhibitors of macromolecular synthesis, e.g. glycogen synthase kinase 3β (Gsk3β) is inhibited by akt1 phosphorylation, while glycogen synthase is activated after akt1 dephosphorylation. Foxo1's S256 site is phosphorylated by akt1, its transcriptional activity is inhibited, while it promotes cell survival in the heart and inhibits glucose production in the liver(Reference Zhang, Li and Qi53–Reference Evans-Anderson, Alfieri and Yutzey55). In addition, the pi3k/hif-1 pathway is also able to regulate glucose metabolism. Increased transport of oxygen and nutrients through enhanced erythropoiesis and angiogenesis(Reference Manalo, Rowan and Lavoie56), improved oxygen utilisation in metabolism(Reference Semenza57), all of which were found to be associated with hif-1 activation. Furthermore, hif-1α levels in cells can be promoted by signalling molecules, such as epidermal growth factor, insulin-like growth factor 1 and insulin(Reference Lu, Forbes and Verma58). Studies have demonstrated that the expression level of hk1 (Hexokinase1) and hk2 (Hexokinase2) can be upregulated by hif-1α activation(Reference Semenza59). In fact, many other glycolytic enzymes can be effectively upregulated by the activation of hif-1α, resulting in elevated glycolysis. In the present study, grass carp fed a high-carbohydrate diet down-regulated the expression levels of hepatic Ir, pi 3 k, akt1 and hif-1α, leading to a decrease in GK and PK activities related to glycolytic enzymes, which inhibits the transcriptional activity of the related genes(Reference Aukrust, Bjorkhaug and Negahdar60–Reference Lin, Liu and Lee62). After taurine supplementation in high-carbohydrate diet, the hepatic hif-1α and foxo1 gene expressions, indicating that the transcriptional activities of HK, GK, PK and G6Pase may be enhanced by taurine through the modification of hif-1α and foxo1 to improve their enzymatic activities.
Similar to mammals, carbohydrate in feed is broken down into small molecules of sugar in the intestine for fish and then metabolised into pyruvate by the glycolytic pathway, which is oxidatively decarboxylated to produce acetyl coenzyme A, providing raw materials for cholesterol and TAG synthesis. In this study, serum TAG, total cholesterol and low-density cholesterol lipoprotein levels of grass carp fed 38·64 % carbohydrate level diet showed a decreasing trend and high-density cholesterol lipoprotein increased significantly compared with the group PC, indicating that there was no burden on lipid metabolism in grass carp fed 38·64 % carbohydrate level diet, which was related to the high sugar tolerance of grass carp. After taurine supplementation in high-sugar diet, serum TAG, total cholesterol and HDL levels of grass carp showed a trend of first increase and then decrease, while LDL levels increased significantly. A study suggested(Reference He, Song and Liang63) that the growth of rat cancer cells was inhibited by taurine through the inhibition of the expression of energy metabolism-related genes such as akt1, igf-1, hif-1α and hk1. But also some studies(Reference Zhang, Wei and Liu64) considered that taurine supplementation facilitated the glycolytic pathway by promoting the gene expression of GK and PK. The reason why taurine increased serum TAG and total cholesterol in this study may be because taurine-enhanced glycolysis and provided more raw materials for lipid synthesis in grass carp.
About 105–108 microflora are present in the intestine of freshwater fish, and their main flora composition includes Bacteroidetes, Fusobacteria, Firmicutes and Proteobacteria (Reference Dimitroglou, Merrifield and Moate65), which was similar to the flora composition in this study. Moreover, there was no significant effect of dietary carbohydrate on the diversity and abundance of intestinal flora of grass carp under the present experimental conditions, and the diversity of intestinal flora was significantly increased after taurine supplementation in high-carbohydrate diet. The abundance of Stenoxybacter, Lachnospiraceae, Mannheimia was significantly higher in group T05 than that in group PC. Compared with the group NC, the group T05, T10 and T15 had a higher abundance in Bacteroidetes and a lower abundance in Fusobacteria, which was consistent with previous studies(Reference Ma, Guo and Liu66). Earlier studies have shown that starch and sucrose metabolic pathways and carbohydrate degradation pathways of the intestinal microbiota were enriched in Nile tilapia (Oreochromis niloticus) supplemented with mannan oligosaccharides on a high-carbohydrate diet, which may be a reason for improved carbohydrate utilisation(Reference Wang, Wu and Li67). Lachnospiraceae was considered as the core symbiotic microbiota, which can exert anti-inflammatory functions(Reference Sokol, Pigneur and Watterlot68). Taurine plays an important role in the metabolism of the three major substances, and the phylum Bacteroidetes is an important participant in the metabolism of polysaccharides, cholesterol and carbohydrate fermentation(Reference Hooper, Wong and Thelin69); however, high levels of taurine inhibit its growth, probably because excessive taurine causes the intestine to be in a highly acidic environment, and it is not suitable for growth under such acidic conditions(Reference Gaylord, Teague and Barrows70). Phylum Fusobacteria is a Gram-negative anaerobic bacterium that can attack the liver, intestines and other tissues in humans to cause inflammation(Reference Kelly, Yang and Pei71). This may be the reason for the better growth of groups T05 and T10.
In this study, grass carp were found to convert excess carbohydrate from feeds into fat for storage without significant effect on growth performance, while a high-carbohydrate diet caused an increase in body fat and fatness, reduced digestibility of protein, fat and starch; the growth performance of grass carp was significantly improved after taurine supplementation in high-carbohydrate diet, the growth performance, digestive enzymes, as well as the activities of digestive enzymes, glycolytic and lipid synthase enzymes were significantly promoted, with an enhanced ability to utilise carbohydrates (Fig. 8). Based on specific growth rate, dietary optimal tributyrin taurine supplementation in grass carp was estimated to be 0·08 %.
A summary of the effects of a high-carbohydrate diet and dietary taurine on the measured indexes.
Acknowledgements
We thank all those who helped in fish culture, sampling and testing.
This work was supported by the National Key R&D Program of China (2019YFD0900200) and the National Natural Science Foundation of China (No. 31772864), Natural Science Foundation of Guangdong Province (2018A030313154 and 2020A1515011129).
L. P. and J. Q. designed the study, carried out the study, analysed the data, draft and revised the manuscript. H. L. designed the study and developed the questions, revised the manuscript critically for important intellectual content and approved the final version to be published. B. T. approved the final version to be published and agreed to be accountable for all aspects of the work. X. D. designed the experiment and assisted in the correction. Q. Y. designed the experiment, assisted in the correction and developed the questions. S. C and S. Z. assisted in the correction, developed the questions and purchased the experiment-related consumables. All authors have read and agreed to the published version of the manuscript.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
