Lipids are critical to maintain the homoeostasis of cellular energy and serve various functions in animals, including aquatic animals. Lipids are also considered a lower-cost ingredient compared with protein, and high lipid diets have been used for a protein-sparing effect in aquaculture in recent years(Reference Ding, Xu and Liu1). However, excess lipid content in diets can cause abnormal lipid deposition and oxidative stress, which would lead to metabolic disturbance and impair the growth, health and quality of aquatic animals(Reference Liao, Yan and Mai2,Reference Ling, Wu and Zhang3) . Therefore, it is necessary to develop an appropriate strategy to enhance the utilisation of dietary lipid to promote growth, which benefits cost-effective aquaculture production. Myo-inositol (MI) is the most abundant isomeric form of inositol in living cells and feed ingredients. It has beneficial effects on varieties of diseases including polycystic ovary syndrome, type 2 diabetes and some other metabolic syndromes(Reference Michell4,Reference Croze and Soulage5) , making this isomer as one of the best candidates as a functional ingredient in a feed(Reference Croze, Géloën and Soulage6). Moreover, the effect of MI as a major lipotropic factor has been well investigated in mammals(Reference Michell4,Reference Shimada, Hibino and Takeshita7–Reference Best, Ridout and Patterson10) . A previous study reported that MI treatment could effectively reduce the white adipose tissue accretion of mice compared with saline solution treatment(Reference Croze, Vella and Pillon11). In addition, dietary MI could reduce the total lipid and TAG contents in the liver by depressing the hepatic activities of glucose-6-phosphate dehydrogenase, malic enzyme, fatty acid synthetase and citrate cleavage enzyme(Reference Katayama12). In aquatic animals, previous studies have also demonstrated the function of MI in reducing lipid deposition(Reference Lee, Lee and Lim13–Reference Shiau and Su16). For example, a negative correlation between dietary MI and hepatic lipid content exists in parrot fish (Oplegnathus fasciatus)(Reference Khosravi, Lim and Rahimnejad15). In crustaceans, dietary MI is beneficial for reducing the midgut gland index and lowering lipid accumulation in the hepatopancreas of grass shrimp(Reference Shiau and Su16). However, to our best knowledge, although studies have investigated the effects of dietary MI level on lipid deposition in fish, crustaceans and mammals, the underlying molecular mechanisms remain unclear and deserve further investigation.
The adenosine 5’-monophosphate-activated protein kinase (AMPK), as a cellular ‘energy sensor’, is an essential regulator of cellular fatty acid metabolism. The activation of AMPK could inhibit several sterol regulatory element-binding protein 1 (SREBP1) enzymes involved in fatty acid synthesis(Reference Zhou, Rahimnejad and Tocher17). Moreover, the activation of AMPK can also phosphorylate acetyl-CoA carboxylase (ACC) leading to the decrease of malonyl-CoA, which could enhance lipid oxidation by regulating the expression of carnitine palmitoyltransferase I (CPT1)(Reference Merrill, Kurth and Hardie18,Reference Herzig and Shaw19) . Besides, knockout of the AMPK gene could cause fatty liver and obesity in mammals(Reference Viollet, Andreelli and Jørgensen20). Therefore, AMPK is considered an important target for regulating lipid metabolism in mammals(Reference Lin, Huang and Lin21). Calmodulin-dependent protein kinase kinase-β (CaMKKβ), a member of the calmodulin kinase family(Reference Vinet, Carra and Blom22), can phosphorylate AMPK at Thr172 in the catalytic α-subunit, and this process depends on the changes in intracellular Ca2+ concentration(Reference Carling23,Reference Hurley, Anderson and Franzone24) . Besides, inositol 1,4,5-trisphosphate (IP3), the most important metabolite of inositol, can bind to the IP3 receptor (IP3R) on the endoplasmic reticulum, which could activate Ca2+ channels to release Ca2+ from the endoplasmic reticulum to the cytoplasm(Reference Ando, Mizutani and Kiefer25,Reference Mikoshiba26) . Previous studies reported that there was a close link between the regulation of AMPK activity and IP3R-mediated Ca2+ signalling pathway(Reference Arias-del-Val, Santo-Domingo and García-Casas27,Reference Gu, Qi and Zhou28) . So far, many investigations have reported that MI plays a crucial part in the regulation of lipid metabolism in aquatic animals(Reference Lee, Lee and Lim13,Reference Khosravi, Lim and Rahimnejad15,Reference Bu, Lin and Liu29) . However, in contrast, no information has been reported on the specific molecular mechanism of MI regulating lipid metabolism.
The Chinese mitten crab (Eriocheir sinensis), one of the most important economic freshwater crustacean species, is a popular food crustacean in China and other Asian countries due to delicious meat, unique flavour and high protein. Previous studies in our laboratory showed that dietary 13 % lipid caused excessive lipid accumulation and oxidative stress in the hepatopancreas of E. sinensis compared with those fed 7–8 % lipid level(Reference Xu, Li and Liu30,Reference Lin, Bu and Wang31) . This study aimed to use E. sinensis as a model species of crustacean to explore the influences of dietary MI on improvement of the negative effect of high lipid diet on the growth and antioxidant status in Chinese mitten crab. Particularly, lipid metabolism mediated by dietary MI was explored at different lipid levels to understand the role of MI in promotion of lipid utilisation in crustacean.
Materials and methods
Four experimental diets (crude protein, approximately 44·0 %) were formulated with different lipid levels (7·0 and 13·0 %) and MI levels (0 and 1600 mg/kg diet) with four replicates. Casein and gelatine were used as the main protein sources, and fish oil, soyabean oil, lecithin and cholesterol were the lipid source. The formulation and proximate composition of four experimental diets are shown in Table 1. The diets were prepared with the process described in detail in our previous study(Reference Bu, Lin and Liu29). The experimental diets were stored at –20°C until use.
* Vitamin premix (per 100 g premix): retinol acetate, 0·043 g; thiamine hydrochloride, 0·15 g; riboflavin, 0·0625 g; Ca pantothenate, 0·3 g; niacin, 0·3 g; pyridoxine hydrochloride, 0·225 g; para-aminobenzoic acid, 0·1 g; ascorbic acid, 0·5 g; biotin, 0·005 g; folic acid, 0·025 g; cholecalciferol, 0·0075 g; α-tocopherol acetate, 0·5 g; menadione, 0·05 g. All ingredients are filled with α-cellulose to 100 g.
† Mineral premix (per 100 g premix): KH2PO4, 21·5 g; NaH2PO4, 10·0 g; Ca(H2PO4)2, 26·5 g; CaCO3, 10·5 g; KCl, 2·8 g; MgSO4·7H2O, 10·0 g; AlCl3·6H2O, 0·024 g; ZnSO4·7H2O, 0·476 g; MnSO4 6H2O, 0·143 g; KI, 0·023 g; CuCl2·2H2O, 0 ·015 g; CoCl2·6H2O, 0 ·14 g; calcium lactate, 16·50 g; Fe-citrate, 1 g. All ingredients are diluted with α-cellulose to 100 g.
‡ Sangong Biotech, Ltd.
§ Shanghai Taiwei, Ltd.
Crab rearing and experimental conditions
The feeding trial was conducted at Zhejiang Freshwater Fisheries Research Institute (Huzhou, China). Similar sized, active and intact crabs were obtained from a local farm (Shanghai, China) and fed a mixture of equal proportions of the experimental diets for 5 d to acclimatise to the environmental conditions. A total of 640 juvenile crabs (4·58 (sem 0·05) g) were randomly distributed into sixteen tanks, with forty crabs per tank. Four arched tiles and four groups of corrugated plastic pipes were placed in each tank as shelters to reduce attacking behaviour. Crabs were fed three times daily (4 % of biomass) at 07.00, 16.30 and 22.30 for 8 weeks. After feeding, all tanks were cleaned daily by siphoning out the feed residue and faeces, and the water of 30 % tank volume was exchanged daily to maintain water quality. During the experiment, the water conditions were as follows: temperature 25 (sem 2) °C, dissolved oxygen >7·0 mg/l, pH 8·0 (sem 0·4) and ammonia and N <0·05 mg/l.
Sample collection and animal ethics
At the termination of the experiment, crabs were deprived of diet for 24 h, and counted and weighed to determine weight gain, specific growth rate, survival and feed conversion ratio. Four crabs per tank were collected and frozen (–20°C) until analysis of the proximate body composition. Haemolymph samples from five crabs per tank were collected immediately according to the method described by our previous study(Reference Han, Wang and Guo32). Haemolymph samples were placed into 1·5 ml Eppendorf tubes at 4°C for 24 h, followed by centrifugation at 4500 rpm for 10 min at 4°C. The supernatant was collected and stored at –80°C; until use. The hepatopancreas from three crabs in each tank was obtained to determine the hepatosomatic index, and hepatopancreas and muscle were quickly taken from these crabs, frozen in liquid N2 and stored at –80°C for the analysis of proximate composition. Three crabs from each tank were sampled for analyses of biochemical parameters and enzyme activities in the hepatopancreas. The hepatopancreas tissues from three other crabs in each tank were collected, frozen in liquid N2 and stored at –80°C to determine the expression of genes and proteins.
The study was conducted strictly according to the Guidance of the Care and Use of Laboratory Animals in China. The experimental protocol was approved by the Committee on the Ethics of Animal Experiments of East China Normal University (No. f20201001).
Proximate composition analysis
Proximate analyses of diets, whole body, hepatopancreas and muscle were determined by the standard methods(33). Ash was measured in a muffle furnace at 550°C for 6 h. Crude protein was measured by the Kjeldahl method (using Kjeltec™ 8200, Foss) and total lipid was measured by chloroform/methanol method(Reference Folch, Ascoli and Lees34). Samples were dried to constant weight at 105°C to measure the moisture content. The MI content in hepatopancreas was measured by the enzymatic assay(Reference Gonzalez-Uarquin, Molano and Heinrich35). Briefly, the assay reaction was to convert MI into stable Iodonitrotetrazolium-formazan which was measured spectrophotometrically at 492 nm wavelength.
Hepatopancreas samples were weighed and homogenised on ice in nine volumes (v/w) of ice-cold saline 8·9 g/ml, and then centrifuged at 2500 g at 4°C for 30 min. The supernatant was collected and stored at −80°C until analysis. The total antioxidant capacity (2, 2-azino-bis (3-ethylbenzo-thiazoline-6-sulfonic acid diammonium salt) method, 405 nm), the activities of superoxide dismutase (SOD, hydroxylamine method, 550 nm) and glutathione peroxidase (5, 50-dithiobis-(2-nitrobenzoic acid) method, 412 nm), malondialdehyde (MDA, thiobarbituric acid method, 532 nm) content and TAG (glycerophosphate oxidase – peroxidase method, 510 nm), cholesterol (T-CHO, cholesterol oxidase - peroxidase) method, 510 nm), NEFA (acetyl CoA synthase - acetyl CoA oxidase - peroxidase method, 546 nm) and total protein content (bicinchoninic acid method, 562 nm) in hepatopancreas were measured using the diagnostic reagent kits (Cat. No. A015-2, A001-3, A005-1, A003-1, A110-1, A111-1, A042-2 and A045-4, Nanjing Jiancheng Bioengineering Institute).
The TAG (glycerophosphate oxidase – peroxidase method, 510 nm), T-CHO (cholesterol oxidase - peroxidase method, 510 nm), NEFA (acetyl CoA synthase - acetyl CoA oxidase - peroxidase method, 546 nm), LDL-cholesterol (enzymatic method, 546 nm) and HDL-cholesterol (enzymatic method, 546 nm) contents in haemolymph were measured using the diagnostic reagent kits (Cat. No. A110-1, A111-1, A042-2, A113-1, A112-1, Nanjing Jiancheng Bioengineering Institute).
RNA extraction and quantitative real-time PCR analysis
The total RNA was isolated from hepatopancreas using RNAiso Plus (Takara) and was reverse-transcribed into cDNA using the PrimeScriptTM RT Reagent kit (Takara). The integrity and quality of RNA were determined by agarose gel electrophoresis and spectrophotometry. The quantitative real-time (RT)-PCR was carried out in the CFX96 Real-Time PCR system (Bio-rad, Richmond) and conducted in a final reaction volume of 20 μl containing 10 μl of SYBR qPCR reagent (Q711-02/03, Vazyme Biotech Co., Ltd), 0·4 μl (10 μM) of each forward and reverse primers, 8·2 μl of DEPC-H2O and 1 μl cDNA template. The programme of RT-PCR reactions was the same as in our previous study(Reference Bu, Lin and Liu29). The amplification efficiency of genes was analysed according to the eq. E = 10(–1/Slope) - 1 and the E-values ranged from 90·1 to 102·6 %. The β-actin and S27 genes were used as an internal control, and the quantitative RT-PCR was quantified by the 2−ΔΔCt method(Reference Livak and Schmittgen36). The primers used are shown in Table 2.
srebp1, sterol regulatory element-binding protein 1; fas, fatty acid synthase; dgat1, diacylglycerol O-acyltransferase 1; Δ9 fad, Δ9 fatty acyl desaturase; cpt, carnitine palmitoyltransferase; fabp, fatty acid binding protein; mttp, microsomal TAG transfer protein; camkkβ, calmodulin-dependent protein kinase kinase-β; S27, ubiquitin/ribosomal S27 fusion protein.
Immunoblot analyses were performed according to the procedure in our laboratory(Reference Lu, Ma and Wang37). Briefly, protein homogenates were prepared from liver tissues in the cell lysis buffer (Beyotime Biotechnology) containing 1 mM phenylmethanesulfonyl fluoride (Beyotime Biotechnology). After 10 400 g centrifugation for 10 min, the supernatant was mixed with 5 × SDS loading buffer and boiled at 95°C for 5 min. Protein concentrations were determined using the Pierce Bicinchoninic Acid Protein Assay Kit (Thermo, Scientific). Proteins (20–50 μg) were separated on SDS-PAGE and transferred electrophoretically to nitrocellulose filter membranes. The membranes were blocked with 5 % non-fat milk in tris-buffered saline (TBS) with 0.05% Tween 20 and were incubated overnight at 4°C with primary antibodies against phospho-IP3R (DF2999, Affinity), phospho-AMPKα (AF3423, Affinity), phospho-ACC1 (AF3421, Affinity) and β-actin (AB0061, Abways). After washing, membranes were incubated with anti-rabbit or mouse horseradish peroxidase-conjugated secondary antibody. The images of Western blotting were determined using the Odyssey CLx Imager (Licor), and the target proteins were quantified using the ImageJ 1.44p software (US National Institutes of Health).
Calculation and statistical analysis
The calculation formulas can be found in online Supplementary File. Results are presented as mean values with their standard error of the mean. Data were checked for normality and homogeneity of variances, and they were normalised when appropriate. Data were analysed by two-way ANOVA to determine if there was any interaction between dietary lipid level and MI level. At the same dietary lipid level, independent-samples t test was used to analyse the significant differences between crabs fed diets with 0 and 1600 mg/kg MI. At the same dietary MI level, independent-samples t test was used to determine significant differences between crabs fed diets with 7 and 13 % lipid. The level of significance was set at P < 0·05. All statistical analyses were conducted using the SPSS 20.0 software package for Windows (SPSS).
Growth performance, feed utilisation and hepatosomatic index
The weight gain and specific growth rate were significantly influenced by MI level (P < 0·05, online Supplementary Table S1). The H + MI group (13 % lipid and 1600 mg/kg MI) showed significantly higher weight gain (Fig. 1(a)) and specific growth rate (Fig. 1(b)) than the H group (13 % lipid and 0 mg/kg MI, P < 0·05). Dietary MI supplementation also had a positive impact on weight gain (Fig. 1(a)) and specific growth rate (Fig. 1(b)) when the lipid level was 7 %, though there were no significant differences between the N (7 % lipid and 0 mg/kg MI) and N + MI groups (7 % lipid and 1600 mg/kg MI, P > 0·05). There was no significant difference in survival among the groups (P > 0·05) (Fig. 1(c)). Crabs fed the N (7 % lipid and 0 mg/kg MI) diet had a markedly lower FCR than those fed the H (13 % lipid and 0 mg/kg MI) diet (P < 0·05) (Fig. 1(d)). The hepatosomatic index declined in N + MI (7 % lipid and 1600 mg/kg MI) and H + MI (13 % lipid and 1600 mg/kg MI) groups when compared with the N (7 % lipid and 0 mg/kg MI) and H (13 % lipid and 0 mg/kg MI) groups, respectively (P < 0·05) (Fig. 1(e)).
Proximate compositions of tissues and myo-inositol content in hepatopancreas
The whole-body lipid content was markedly affected by dietary MI level (P < 0·05), and the total lipid content in hepatopancreas and muscle, whole-body protein content and hepatopancreas MI content were markedly affected by dietary lipid and MI levels (P < 0·05, online Supplementary Table S2). No significant difference was observed in whole-body moisture content among the groups (P > 0·05) (Fig. 2(a)). Crabs fed 13 % lipid showed a significantly lower crude protein level in the whole body than those fed 7 % lipid regardless of dietary MI (P < 0·05) (Fig. 2(b)). Moreover, the whole-body protein content significantly increased in the N + MI (7 % lipid and 1600 mg/kg) and H + MI (13 % lipid and 1600 mg/kg MI) groups when compared with the N (7 % lipid and 0 mg/kg MI) and the H (13 % lipid and 0 mg/kg MI) groups, respectively (P < 0·05) (Fig. 2(b)). The whole-body lipid content decreased in N + MI (7 % lipid and 1600 mg/kg MI) and H + MI (13 % lipid and 1600 mg/kg MI) groups when compared with the N (7 % lipid and 0 mg/kg MI) and H (13 % lipid and 0 mg/kg MI) groups, respectively (P < 0·05) (Fig. 2(c)). Dietary MI supplementation markedly increased whole-body ash content when the lipid level was 7 % (P < 0·05) (Fig. 2(d)). Crabs fed 13 % lipid showed a significantly lower crude protein level in the hepatopancreas than those fed 7 % lipid regardless of dietary MI (P < 0·05) (Fig. 3(a)). No significant difference was observed in the crude protein level in muscle among the groups (P > 0·05) (Fig. 3(b)). Total lipid contents in hepatopancreas (Fig. 3(c)) and muscle (Fig. 3(d)) in the H (13 % lipid and 0 mg/kg MI) group were markedly higher than those in the N group (7 % lipid and 0 mg/kg MI, P < 0·05). However, total lipid contents in hepatopancreas (Fig. 3(c)) and muscle (Fig. 3(d)) were significantly reduced in the H + MI (13 % lipid and 1600 mg/kg MI) group compared with those in the H (13 % lipid and 0 mg/kg MI) group (P < 0·05). Crabs fed the H (13 % lipid and 0 mg/kg MI) diet had a markedly lower MI level in hepatopancreas than those fed the N (7 % lipid and 0 mg/kg MI) diet (P < 0·05) (Fig. 3(e)). Dietary MI supplementation markedly increased MI level in hepatopancreas regardless of dietary lipid (P < 0·05) (Fig. 3(e)).
Biochemical indicators in haemolymph and hepatopancreas
As shown in Table 3, there were significant main effects of dietary MI on the TAG, T-CHO and NEFA contents in hepatopancreas and T-CHO and LDL-cholesterol contents in haemolymph (P < 0·05). The TAG and T-CHO contents were also affected by the interaction between dietary MI and lipid levels (P < 0·05). Crabs in the H (13 % lipid and 0 mg/kg MI) group had significantly higher TAG, T-CHO and NEFA levels in haemolymph than those in the N (7 % lipid and 0 mg/kg MI) group (P < 0·05). Dietary 1600 mg/kg MI significantly reduced the TAG, T-CHO and LDL-cholesterol levels in haemolymph compared with those fed diets without MI supplementation when the lipid level was 13 % (P < 0·05). No significant difference was observed in the HDL-cholesterol level in haemolymph among the groups (P > 0·05). Crabs fed 1600 mg/kg MI showed a significantly lower TAG content in hepatopancreas than those fed 0 mg/kg MI regardless of dietary lipid (P < 0·05). The T-CHO and NEFA contents of the hepatopancreas in the H + MI (13 % lipid and 1600 mg/kg MI) group were significantly lower than those in the H (13 % lipid and 0 mg/kg MI) group (P < 0·05).
T-CHO, cholesterol; SOD, superoxide dismutase; T-AOC, total antioxidant capacity; GSH-Px, glutathione peroxidase; MDA, malondialdehyde; NS, no significant difference; LL, lipid levels; ML, myo-inositol levels; LL × ML, lipid levels × myo-inositol levels.
* Values in the same row with different superscripts are significantly different (P < 0·05).
A,BMeans significant difference between myo-inositol levels within the same lipid level (P < 0·05).
a,bMeans significant difference between lipid levels within the same myo-inositol level (P < 0·05).
Antioxidant indicators in hepatopancreas
The SOD activity and MDA content were significantly affected by dietary MI and the interaction between MI and lipid (P < 0·05). Crabs in the H (13 % lipid and 0 mg/kg MI) group had significantly lower total antioxidant capacity and the activities of SOD and glutathione peroxidase, and a higher MDA content in hepatopancreas than those in the N (7 % lipid and 0 mg/kg MI) group (P < 0·05). The increased total antioxidant capacity and the activities of SOD and glutathione peroxidase, and the decreased MDA content in hepatopancreas were found in the H + MI (13 % lipid and 1600 mg/kg MI) group compared with those in the H (13 % lipid and 0 mg/kg MI) group (Table 3, P < 0·05).
As shown in Fig. 4 and online Supplementary Table S3, the mRNA abundances of srebp1, dgat1, cpt2, camkkβ, fabp10 and mttp were markedly influenced by dietary lipid and MI (P < 0·05), and the mRNA abundances of fas, Δ9 fad and fabp3 were dramatically influenced by dietary lipid, MI and their interaction (P < 0·05). Specifically, the H (13 % lipid and 0 mg/kg MI) group showed a markedly higher relative expression of genes involved in lipid synthesis (srebp1, fas and Δ9 fad) than the N (7 % lipid and 0 mg/kg MI) group (P < 0·05), but no significant difference was found in dgat1 between the two groups (P > 0·05). Crabs fed 1600 mg/kg MI showed significantly lower mRNA abundance of srebp1, fas and Δ9 fad than those fed 0 mg/kg MI regardless of dietary lipid (P < 0·05). Moreover, crabs in the H + MI (13 % lipid and 1600 mg/kg MI) group had a significantly lower mRNA abundances of dgat1 than those in the H (13 % lipid and 0 mg/kg MI) group (P < 0·05). The H + MI (13 % lipid and 1600 mg/kg MI) group showed significantly higher transcriptional levels of cpt1a, cpt1b and cpt2 than the H group (P < 0·05). Crabs fed 13 % lipid had a significantly higher transcriptional level of cpt2 than those fed 7 % lipid regardless of dietary MI (P < 0·05). In addition, crabs fed 1600 mg/kg MI had a markedly higher camkkβ mRNA level than those fed 0 mg/kg MI regardless of dietary lipid (P < 0·05). The mRNA abundances of genes involved in fatty acid uptake (fabp3 and fabp10) in the H group were significantly higher than those in the N group (P < 0·05), but crabs fed the H + MI (13 % lipid and 1600 mg/kg MI) diet significantly down-regulated the transcriptional levels of these genes compared with those fed the H (13 % lipid and 0 mg/kg MI) diet (P < 0·05). No significant difference was observed in the transcriptional level of fabp9 among the groups (P > 0·05). Crabs in the H (13 % lipid and 0 mg/kg MI) and H + MI (13 % lipid and 1600 mg/kg MI) groups had a significantly higher transcriptional level of mttp than those in the N (7 % lipid and 0 mg/kg MI) and N + MI (7 % lipid and 1600 mg/kg MI) groups, respectively (P < 0·05). Furthermore, crabs fed 1600 mg/kg MI had a significantly higher transcriptional level of mttp than those fed 0 mg/kg MI regardless of dietary lipid (P < 0·05).
There was a significant main effect of dietary MI on the p-IP3R, p-AMPK and p-ACC1 protein levels in hepatopancreas (online Supplementary Table S4, P < 0·05). The p-IP3R (Fig. 5(a)), p-AMPK (Fig. 5(b)) and p-ACC1 (Fig. 5(c)) protein levels in crabs fed the N + MI (7 % lipid and 1600 mg/kg MI) and H + MI (13 % lipid and 1600 mg/kg MI) diets were significantly increased compared with those fed the N (7 % lipid and 0 mg/kg MI) and H (13 % lipid and 0 mg/kg MI) diets, respectively (P < 0·05).
In the present study, no significant differences were found in the survival and growth performance of Chinese mitten crab between the N and H groups. This result is consistent with the results of a previous study(Reference Xu, Li and Liu30), which demonstrated that high lipid diets did not significantly impair the growth of crustaceans. However, excessive energy in diets may affect energy metabolism, leading to lipid accumulation in tissues, which further impairs organelle integrity by increasing the production of reactive oxygen species and the risk of oxidative stress in aquatic animals and mammals(Reference Chen, Cao and Yan38–Reference Yin, Xie and Zhuang40). The MDA, a stable end product of fatty acid peroxidation, is a biomarker of oxidative stress(Reference Koruk, Taysi and Savas41), and the elevation of antioxidant enzymes, such as SOD and glutathione peroxidase, is one of the main approaches against the toxicity of reactive oxygen species(Reference Lesser42). Indeed, although the growth performance of crab was not markedly affected by dietary 13 % lipid, oxidative stress was induced by excess lipid deposition through increasing the risk of lipid peroxidation and disturbing the enzymatic–antioxidant system of E. sinensis, which is in line with the results in the previous studies on the same species(Reference Lin, Bu and Wang31) and other aquatic animals(Reference Sun, Jin and Jiao43,Reference Guo, Zhou and Zhao44) . In contrast, dietary MI supplementation enhanced the growth performance and antioxidant capacity of E. sinensis under a high lipid diet, which is supported by the findings in other studies on fish and crustacean(Reference Khosravi, Lim and Rahimnejad15,Reference Shiau and Su16,Reference Bu, Lin and Liu29,Reference Chen, Guo and Espe45) . This result seemingly suggests that MI, as a cyclitol, can donate hydrogen and have a chelating character to reduce oxidative stress by scavenging hydroxyl radicals(Reference Hu, Chen and Lin46). Similarly, dietary MI could prevent lipid peroxidation and protein oxidation induced by waterborne copper in muscle via NF-E2-related factor 2/antioxidant response element signalling pathway in Jian carp (Cyprinus carpio var. Jian)(Reference Jiang, Liu and Jiang47). Collectively, dietary MI could reverse the lipid peroxidation and down-regulation of antioxidation enzyme activities caused by dietary high lipid and improve the health of E. sinensis. On the other hand, dietary MI could facilitate lipid utilisation in animals fed high lipid diet by optimising lipid metabolism, leading to fast growth(Reference Michell4,Reference Khosravi, Lim and Rahimnejad15,Reference Bu, Lin and Liu29) .
The role of lipid reduction by MI has been reported in human, mammals and aquatic animals. Extra dietary MI can be beneficial in adiposity, hyperglycaemia and insulin resistance for human(Reference Michell4). Dietary MI supplementation could reduce TAG accumulation in high fructose-induced fatty liver in rats(Reference Shimada, Hibino and Takeshita7). In contrast, dietary MI deficiency could result in lipid accumulation in the liver of rats(Reference Hayashi, Maeda and Tomita48), fish(Reference Lee, Lee and Lim13–Reference Khosravi, Lim and Rahimnejad15) and crustaceans(Reference Shiau and Su16,Reference Bu, Lin and Liu29) . Furthermore, the increase of dietary lipid levels may result in the increase of body lipid and the decrease of body protein in aquatic animals(Reference Ding, Xu and Liu1,Reference Han, Li and Wang49) . The present study found that 13 % dietary lipid increased the hepatosomatic index and lipid contents in hepatopancreas and muscle of crabs compared with 7 % dietary lipid. This result is consistent with the results of the same species, which illustrates a positive relationship between dietary lipid and body lipid(Reference Xu, Li and Liu30,Reference Lin, Bu and Wang31) . Moreover, dietary MI could reverse the reduction of protein in the whole body and lipid deposition in the whole body, hepatopancreas and muscle of E. sinensis fed the high (13 %) lipid diet, which is similar to the results of a previous study(Reference Gong, Lei and Zhu50). These results indicate that MI may help achieve the protein-sparing effect in crab feed. Intriguingly, 13 % dietary lipid reduced the MI content in the hepatopancreas compared with 7 % dietary lipid, but dietary MI supplementation markedly enhanced the MI content in the hepatopancreas of crabs. This finding further reflects that the reduction of MI in hepatopancreas may be used to resist the negative effects of excessive lipid deposition induced by high lipid diet(Reference Croze, Géloën and Soulage6). In other words, MI may play a vital role in reducing lipid accumulation of crabs and improving lipid utilisation.
In the present study, crabs fed the high (13 %) lipid diet showed increased TAG and cholesterol contents in haemolymph and TAG content in hepatopancreas when MI was not added to the diet, indicating an active endogenous lipid transport. This result is in agreement with some other reports on crustaceans and fish(Reference Yin, Xie and Zhuang40,Reference Guo, Zhou and Zhao44,Reference Hamidoghli, Won and Aya51,Reference Xu, Li and Liu52) . On the contrary, the reduced TAG, cholesterol and LDL-cholesterol in haemolymph and low TAG, cholesterol and NEFA in the hepatopancreas of crabs fed the high lipid diet with MI supplementation suggest that MI may accelerate the overall process of endogenous lipid consumption. In other words, MI was involved in lipid catabolism to supply the energy need of crabs for rapid growth. In addition, LDL carries cholesterol from the liver to peripheral tissues, while HDL carries cholesterol from peripheral tissues to the liver(Reference Xiao, Ji and Ye53). Therefore, the reduced cholesterol in the haemolymph of crabs fed the high lipid diet with MI supplementation is possibly attributed to the decreased LDL-cholesterol(Reference Jain, Vargas and Gotzkowsky54).
So far, the specific molecular mechanism of MI regulating lipid metabolism remains unclear in crustacean. The present study found that the mRNA abundances of genes involved in lipid uptake were up-regulated in the H group compared with the N group, suggesting that high dietary lipid could enhance lipid absorption in the hepatopancreas, resulting in an elevation of lipid accumulation in the hepatopancreas of E. sinensis, which is supported by the previous studies(Reference Lin, Bu and Wang31,Reference Cai, Mai and Ai55) . The present study also demonstrated that dietary MI effectively reversed the up-regulation of gene expressions in lipid uptake induced by high dietary lipid, resulting in slow lipid deposition in the hepatopancreas of crabs. Moreover, mttp is vital in the formation of VLDL in aquatic animals and mammals(Reference Hussain, Nijstad and Franceschini56,Reference Lin, Han and Lu57) , and VLDL is responsible for transporting TAG from the hepatopancreas to perihepatic tissues(Reference Jacobs, Lingrell and Zhao58). Thus, the present result indicates that dietary MI might promote the secretion of VLDL to avoid excessive lipid accumulation. In addition to lipid uptake and transport, lipogenesis and lipid catabolism were also affected by dietary lipid and MI. The Srebp1, as a transcription factor, plays a crucial role in lipogenesis, which can activate downstream genes in different steps of fatty acid synthesis de novo and TAG synthesis including fas, dgat1 and Δ9 fad (Reference Amemiya-Kudo, Shimano and Hasty59–Reference Che, Xu and Gao61). In the present study, dietary MI decreased the expression of genes involved in fatty acid synthesis de novo for which the expression was increased by high dietary lipid, suggesting that dietary MI might reduce the lipid deposition in the hepatopancreas of crabs by down-regulating the transcriptional level of srebp1 under high lipid diet. This result is similar to the results on mammals, where dietary supplementation with MI could reduce the expression of lipogenic genes and TAG accumulation in the high fructose-induced fatty liver in rats(Reference Shimada, Hibino and Takeshita7,Reference Shimada, Ichigo and Shirouchi62) . On the other hand, β-oxidation is a vital approach to lipid catabolism in crabs, fish and mammals(Reference Lin, Bu and Wang31,Reference Tabata, Rodgers and Hall63,Reference Li, Li and Ning64) . CPT1 is involved in long-chain fatty acid oxidation, catalysing the conversion of fatty acid CoA to fatty acid carnitines for accessing the mitochondrial matrix, and the fatty acyl group is transferred back to CoA by CPT2(Reference Kerner and Hoppel65). The present study showed that dietary MI markedly up-regulated the mRNA abundances of cpt1a, cpt1b and cpt2 in the hepatopancreas of crabs under high lipid diets, indicating the enhanced activity of fatty acid β-oxidation, which is consistent with the change of lipid content in the hepatopancreas of crabs. Therefore, dietary MI could lessen the lipid accumulation in the hepatopancreas of crabs induced by high lipid diet through inhibiting lipid absorption and synthesis and facilitating fatty acid β-oxidation.
AMPK can regulate lipid synthesis and decomposition(Reference Zhou, Rahimnejad and Tocher17). For instance, dietary Zn can inhibit SREBP1-mediated fatty acid synthesis and promote fatty acid β-oxidation through the Ca2+/CaMKKβ/AMPK pathway(Reference Shi, Jin and Jiao66). Meanwhile, AMPK can phosphorylate and inactivate ACC, resulting in the reduction of malonyl-CoA, which is an inhibitor of CPT1(Reference McGarry, Takabayashi and Foster67). The present results exhibit that dietary MI can inhibit the fatty acid synthesis and promote fatty acid entry into mitochondria for β-oxidation in the hepatopancreas of crab through AMPK activation. In addition, the present study shows that the IP3R was activated and camkkβ was up-regulated by dietary MI supplementation. A previous study reported that the activation of IP3R could cause an elevation of cytosolic Ca2+ and up-regulate CaMKKβ with subsequent activation of its downstream molecule AMPK(Reference Gu, Qi and Zhou28). Moreover, IP3, as a second messenger, is the most crucial metabolite of inositol produced through phosphoinositide turnover and can bind to the IP3R on the endoplasmic reticulum to regulate cytosolic Ca2+ homoeostasis(Reference Mikoshiba26). Thus, according to the metabolic fate of MI, MI may regulate lipid metabolism of E. sinensis through activating the IP3R/CaMKKβ/AMPK pathway. It is worth noting that the Ca2+ pathway-mediated AMPK activity regulated by IP3R is still not completely clear. For example, some studies have shown that when IP3R is blocked, low cytoplasmic Ca2+ can cause the mitochondria to fail to produce energy normally, resulting in insufficient ATP, which in turn activates the AMPK pathway(Reference Decuypere, Chandra Paudel and Parys68,Reference Cardenas, Miller and Smith69) . Others believe that the increased intracellular Ca2+ could activate the CaMKKβ and its substrate molecule AMPK(Reference Pfisterer, Mauthe and Codogno70,Reference Grotemeier, Alers and Pfisterer71) . Further study is warranted to verify if MI interacts directly or indirectly with IP3R and the complete crosstalk between IP3R and lipid homoeostasis in E. sinensis.
In summary, dietary MI could promote growth performance and increase the MI deposition in hepatopancreas and the lipid utilisation of crabs fed high (13 %) lipid through optimising lipid metabolism. Dietary MI could also reduce oxidative damage induced by excessive lipid accumulation in the hepatopancreas of E. sinensis fed high lipid diet. Moreover, MI may be an important factor in inhibiting the fatty acid synthesis and promoting fatty acid β-oxidation, which may be partly involved in the up-regulation of the IP3R/CaMKKβ/AMPK pathway in E. sinensis fed 13 % lipid (Fig. 6). These results can explain how MI improves the lipid utilisation in crabs fed high lipid diet and provide an in-depth understanding of the application for using MI in crabs fed high lipid diet.
This work was supported by grants from National Key R&D Program of China (2018YFD0900400), the China Agriculture Research System-48 (CARS-48), the National Natural Science Foundation of China (32072986), Agriculture Research System of Shanghai, China (202104), the Research and Development Project in Key Areas of Guangdong Province (2020B0202010001) and The Opening Fund of Key Laboratory of Sichuan Province for Fishes Conservation and Utilization in the Upper Reaches of the Yangtze River (NJTCCJSYSYS01).
X. B., X. W. and L. C. conceived the experiment; X. B., Z. L. and C. W. formulated the experimental diets; X. B., L. L. and S. L. conducted the experimental trials and laboratory analysis; Q. S. provided experimental materials and analysis tools. X. B. wrote and revised the manuscript; X. W., J. G. Q. and L. C. conducted polish work of this manuscript. All authors contributed to and approved the manuscript.
No potential conflict of interest.
For supplementary material referred to in this article, please visit https://doi.org/10.1017/S0007114521001409