This study designed the normal diet (Con), high-lipid diet (HL), and 50 mg/kg berberine supplemented Con and HL diets (named as ConB and HLB, respectively). Fingerling black sea bream (Acanthopagrus schlegelii) (initial weight  g) were fed these four diets for 8 weeks, and the liver immune response and intestinal health were evaluated. Fish fed the HL diet had the lowest serum glutathione peroxidase and catalase activities, while berberine supplementation significantly improved serum catalase activity (). The expressions of tumor necrosis factor-alpha (tnfα), nuclear factor-kappa b (nf-κb), and interleukin 1 beta (il-1β) were significantly increased by the HL diet, whereas berberine supplementation significantly downregulated these proinflammation cytokine expressions (). Berberine supplementation in normal and high-lipid diets significantly promoted hepatic anti-inflammation cytokines interleukin-10 (il-10) and transforming growth factor beta-1 (tgfβ-1) expressions (). The intestinal expressions of tnfα, nf-κb, and tgfβ-1 were not significantly influenced by different treatments, while Cu/Zn-superoxide dismutase (Cu/Zn-sod) expression was significantly upregulated in the ConB group compared with the Con group. High-lipid diet reduced intestinal villus length. the thickness of muscularis mucosa, mucosal folds, and diversity and richness of intestinal microbiota but increased numbers of goblet cells and width of lamina propria. Berberine supplementation in high-lipid diet alleviated these pathological changes and rebalanced intestinal microbiota composition. At the genus level, fish fed the HL diet had significantly lower abundances of some probiotics such as Lactobacillus and Hydrogenophaga than that of the HLB group (). In conclusion, berberine supplementation in high-lipid diet reduced liver inflammation, improved the antioxidative status of fish, and alleviated intestinal damage and intestinal microbiota alteration of black sea bream.

1. Introduction

Berberine is an isoquinoline alkaloid that can be extracted from Coptis chinensis, Berberis thunbergii, and Hydrastis Canadensis, and it exerts various pharmacological functions, such as antitumor, antifungal, cardioprotective, and anti-inflammatory effects [1]. After oral administration, berberine enters the blood after structural transformation by intestinal microbiota. However, more than 90% of berberine remains in the intestine [2], where may be the main action site of berberine [3]. In traditional Chinese medicine, berberine is used for intestinal infections, especially bacterial diarrhea, mainly due to its antisecretory and antimicrobial characters [4]. In addition, berberine has a powerful free radical scavenging effect, and it can reduce the production of reactive oxygen species (ROS) via multiple ways, such as inhibiting ROS-producing oxidase enzymes, modulating the activities of several endogenous antioxidant enzymes, and quenching nitric oxide (NO), superoxide anion (O2-), and the precariously reactively molecules peroxynitrite (OONO-) directly [57]. Moreover, berberine can directly downregulate the expressions of proinflammatory cytokines, such as interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), and IL-1β, as well as indirectly affect the expressions of activator protein 1 (AP-1), nuclear factor-kappa B (NF-κB), the AMP-activated protein kinase (AMPK) signaling pathways, and the mitogen-activated protein kinase- (MAPK-) mediated pathways to decrease the inflammation responses [4, 8]. Thus, berberine is an effective antioxidant and anti-inflammation drug; it has the potential to be a functional feed additive for improving fish antioxidative and immune status.

The gut microbiota is a symbol representing health status ([9]); it can improve energy expenditure, nutrient absorption, and immune response, as well as inhibit proliferation of pathogens and keep the integrity of intestinal epithelium, which are beneficial for the host [10, 11]. It has been proved that berberine altered gut microbiota to influence the metabolic phenotypes in rats [2, 12, 13]. In mice, berberine exerts its hepatic lipid-lowering effect partly through regulating microbial bile acid metabolism and the microbiota-gut-brain axis [14]. And berberine administration could alter intestinal and fecal Bacteroidetes and Firmicutes richness in mice, which are related to intestinal energy harvest [15, 16]. Also, a study in grass carp (Ctenopharyngodon idella) revealed that administration of berberine reduced serum glucose, total cholesterol, and triglyceride contents by modulating the gut microbiota [3]. Yu et al. [17] observed that berberine improved the intestinal barrier stressed by a high-carbohydrate or high-fat diet by regulating the intestinal microbiota in blunt snout bream (Megalobrama amblycephala). Therefore, berberine exerts its multiple functions by modulating intestinal microbiota, and intestinal microbiota may be a target for berberine to maintain liver and intestine health threatened by high-lipid diet.

Black sea bream (Acanthopagrus schlegelii) is an important aquaculture species with many promising characteristics. Compared with carbohydrate, lipids are more important for black sea bream as an energy source, due to its high energy density and digestibility, and carnivorous fish has a limited ability in using dietary carbohydrate [18, 19]. For fish, the foregut is the main site for lipid digestion and uptake [20], making it become an important target for studying lipid utilization. Previously, Wang et al. [21] reported that suitable dietary lipid levels could spare protein utilization, thus reducing the dietary protein requirement. However, excessive dietary lipid causes growth retardation and massive lipid accumulation in the liver, impairs immune and antioxidative capacity, disrupts the intestinal structural integrity [17, 22], and alters gut microbiota diversity [10, 23]. Previous studies have observed that high-lipid diet-induced intestinal and/or liver inflammation altered intestinal microbiota composition in different aquaculture species [24]. Berberine is an effective feed additive to treat fatty liver caused by high-lipid diet in blunt snout bream [22, 25, 26] and zebrafish [27]. Previously, Wang et al. [19] reported that supplementation of 50 mg/kg berberine in high-lipid diet could effectively reduce hepatic lipid accumulation in black sea bream by upregulating lipolysis genes and downregulating lipogenesis gene expression. But berberine supplementation did not influence growth performance and feed utilization in the abovementioned studies. However, those studies seldom focused on the alteration of the gut microbiota influenced by berberine in fish. Thus, we assumed that berberine may have a beneficial effect in maintaining the intestinal barrier by modulating the intestinal microbiota stressed by high-lipid diet.

Therefore, the purpose of this experiment was to explore whether berberine can relieve fish intestinal and liver damage, as well as antioxidant and immune status caused by high-lipid diet and whether the change of intestinal microbiota is involved in its alleviation of high lipid stress.

2. Materials and Methods

2.1. Experimental Diets and Feeding Trial

Four isonitrogenic diets (42% crude protein) were designed, and the formulation and proximate compositions are shown in Table 1. The control diet (Con) contained 11.1% crude lipid, and the high-lipid diet (HL) had 20.3% crude lipid. 50 mg/kg berberine (C20H18ClNO4, HPLC ≥98%), which was recommended as the suitable dosage in fish feed reported by Zhou et al. [26], was purchased from the Spring and Autumn Biotechnology Company (Nanjing, China) and added to Con and HL diets and named ConB and HLB, respectively. The main protein and lipid ingredients were purchased from the Minghui Feed Company (Hangzhou, China). All the ingredients were thoroughly grounded and sieved through a 178 μm net. Then, the ingredients were mixed by a mixer, and water was added to increase viscidity. The diets were extruded through a 0.25 mm mold by a feed maker (Modle HKJ-218, HUARUI, Wuxi, China). All the diets were subsequently cooked by stream for ten minutes, then put on a clean floor, and dried by air conditioner and dehumidifier. Representative feed was collected for proximate composition analysis, and the remaining feed was sealed in plastic bags and put in −20°C until use.

Black sea bream fingerlings were obtained from a farm in Zhou Shan, China. After being acclimated for 14 days, 420 healthy fish (initial weight  g) were distributed into 12 tanks (400 L blue plastic tank, 35 fish per tank) with three replicates for each treatment. Seawater was obtained from the nearshore of Xixuan island (Marine Fisheries Research Institute of Zhejiang, Zhoushan, China). After sedimentation and filtration, water flowed into each tank at about 2 L/minute. The water condition was monitored every day and kept as temperature °C, pH , salinity , and dissolved mg/L. Then, the fish were fed experimental diets three times a day (7:00, 12:00, and 16:00) for 8 weeks.

2.2. Sample Collection

At 24 h after the last feeding, all fish were anesthetized by MS-222 (60 mg/L), the foregut and liver samples were randomly collected from 12 fish of each tank. The blood sample was obtained from the caudal vein, and pooled blood samples were centrifuged at 10,000 g after a 2 h settlement at 4°C. Three intestine and liver samples were put into 4% paraformaldehyde and 0.5% glutaraldehyde solution for histological observation. The other samples were put into RNase-free tubes and stored in liquid nitrogen for gene expression analysis and enzyme activity determination. The whole intestine from 12 fish (three fish per tank) was carefully removed then transferred to 2 mL sterile tubes and put into liquid nitrogen immediately for intestinal microbiota composition analysis.

All the experimental processes followed the Committee on the Ethics of Animal Experiments of Zhejiang University.

2.3. Sample Analysis

The liver and foregut supernatant was obtained according to Wang et al. [21]. The activities of superoxide dismutase (SOD), glutathione peroxidase (GSH-Px), catalase (CAT), and malondialdehyde (MDA) contents in the serum and liver were determined by the diagnostic reagent kits (catalog number: SOD, A001-3-2; GSH-Px, A005-1-2; CAT, A007-1-1; MDA, A003-1-2) for fish (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) following the manufacture’s instruction.

2.4. Histological Observation

The liver and foregut samples for microscope observation were dehydrated by grades of ethanol and fixed in paraffin wax and then stained with hematoxylin and eosin (H&E) after drying. Pictures were taken by the Olympus CX21 microscope. Foregut samples for scanning electron microscope (SEM) observation were washed in graded levels of ethanol and fixed in 1% OsO4 for around two hours, before using different levels of ethanol to wash the samples gradually. The samples were subsequently dehydrated by liquid CO2 (Hitachi Model HCP-2), after which they were coated with gold-palladium (Hitachi Model E-1010 ion sputter). The SEM pictures were observed by Hitachi Model SU-8010.

2.5. RNA Extraction and qPCR

The total RNA of the liver and foregut samples (four replicates per treatment) were extracted by TaKaRa MiniBEST Universal RNA Extraction Kit (Takara, Japan). Then, cDNA was reversely transcribed by PrimeScript™ RT reagent Kit with gDNA Eraser (Perfect Real Time, Takara, Japan). The cDNA product was diluted with sterilized double-distilled water. The PCR primers of tnfα (tumor necrosis factor-alpha), il-1β (interleukin 1 beta), nf-κb (nuclear factor-kappa b), il-10 (interleukin-10), tgfβ-1 (transforming growth factor beta-1), Cu/Zn-sod (Cu/Zn-superoxide dismutase), and gpx (glutathione peroxidase) were designed by Primer Premier 5.0 (Table 2). The qPCR process was set as 95°C for 30 s and 40 cycles of 95°C for 5 s followed by 60°C for 30s. β-Actin was used as the reference gene. SYBR Premix Ex Taq™II Kit (Takara, Japan) was used for real-time PCR. The relative gene expression level was calculated by the 2ΔΔCT method [28].

2.6. DNA Extraction and Sequencing of Intestinal Microbiota

The E.Z.N.A.®Stool DNA Kit (D4015, Omega, Inc., USA) was used for the extraction of total DNA from the intestinal mucosa layer. The DNA was amplified using 16S rDNA bacteria domain-specific primers for V4: 515F-806R region in a 25 μL reaction system, with 12.5 μL PCR premix, 25 ng DNA template, 5 μL primers, and PCR-grade water. The PCR process was set as follows: 98°C for 30 s, 30 cycles of denaturation at 98°C for 10 s, and 50°C for 30 s for annealing, followed by a final extension at 72°C for 10 min. Then, 2% agarose gel electrophoresis was used to verify the PCR product. The PCR products were purified and quantified by LC-Bio Technology Co., Ltd., Hang Zhou, Zhejiang Province, China. Sequencing was performed on the NovaSeq PE250 platform.

Illumina NovaSeq platform (LC-Bio) was used for sample sequencing under the manufacturer’s instruction. Paired-end reads were assigned to samples based on their unique barcode and truncated by cutting off the barcode and primer sequence. To obtain high-quality clean tags, fqtrim (v0.94) was used for filtering the raw reads. Chimera sequences were removed by Vsearch software (v2.3.4). Feature abundance was normalized by SILVA classifier. Chao1, observed species, Shannon, and Simpson are alpha diversity indices for analyzing the complexity of species diversity of all the samples calculated by QIIME2. Beta diversity was also calculated by QIIME2. Sequence alignment was performed by Blast, and feature sequences were annotated in the SILVA database. The diagrams were illustrated by the R package (v3.5.2).

2.7. Data Statistical Analysis

The data for antioxidative parameters, alpha diversity of the intestinal microbiota, and intestinal histological indices were tested for normality and homogeneity of variances by Kolmogorov–Smirnov and Levene’s tests, respectively. Then, the data were subjected to one-way analysis of variance (ANOVA) by IBM SPSS Statistics 20.0 (SPSS) using Duncan’s multiple range tests, and different superscripts of mean values indicate the significant difference among different treatments based on one-way ANOVA analysis. Two-way ANOVA was used to determine the effects of berberine supplementation, dietary lipid level, and their interaction on these parameters. is considered significant. All data were expressed as mean values and standard deviations (SD).

3. Result

3.1. Antioxidative Parameters in the Serum and Liver

As shown in Table 3, fish fed the low-lipid diets (Con and ConB) had significantly higher serum GSH-Px activity than that of the HL and HLB groups (). Berberine supplementation in the control and high-lipid diets significantly improved serum CAT activity (). The highest serum MDA content was found in fish fed the HL diet. In the liver, the highest CAT activity was in fish fed the ConB diet, while SOD and GSH-Px activities, as well as MDA content in the liver, showed no significant differences among the different groups (). Based on two-way ANOVA analysis, serum GSH-Px activity was significantly influenced by dietary lipid level, and serum CAT activity was significantly affected by berberine supplementation and dietary lipid level ().

3.2. Gene Expression in the Liver and Foregut

As shown in Figure 1, the relative expression levels of nf-κb, il-1β, tnfα, il-10, and tgfβ-1 were significantly upregulated in the liver of fish fed HL diet compared with that of the Con group, while berberine supplementation in high-lipid diet significantly downregulated tnfα, nf-κb, and il-1β expressions (). Berberine supplementation in normal and high-lipid diets significantly upregulated il-10 and tgfβ-1 expressions in the liver (). Based on the two-way analysis (Table 4), all these genes were affected by both dietary lipid level and berberine supplementation significantly, and significant interactions between berberine and dietary lipid level were observed in all these gene expression levels (except for il-10).

In the foregut (Figure 2), nf-κb, tnfα, and tgfβ-1 expression levels in the different groups showed no significant differences (). Fish fed the ConB diet had a significantly upregulated expression level of Cu/Zn-sod than that of the other three groups (). Based on the two-way analysis (Table 4), these genes were not significantly influenced by berberine supplementation or dietary lipid level, and no significant interaction was observed.

3.3. Intestinal Morphology

As shown in Figure 3, the intestinal villus of the fish fed Con, ConB, and HLB diets were intact and neat, whereas fish fed the HL diet had relatively sparse intestinal villus, fewer mucosal folds along with expanded lamina propria, and decreased muscular mucosae. The HL group showed the shortest villus length but the highest number of goblet cells among the treatments (Table 5). Based on two-way ANOVA, villus length was significantly influenced by the dietary lipid level, and significant interaction between dietary lipid level and berberine supplementation was observed in villus length ().

According to the SEM result in Figure 4 and Table 5, berberine supplementation in the normal diet significantly increased intestinal microvillus density (). Besides, dietary lipid level and berberine supplementation influenced intestinal microvillus density significantly based on two-way ANOVA analysis ().

3.4. Intestinal Microbiota

As shown in Table 6, observed species and Chao1 were used to estimate community richness, while Simpson’s and Shannon’s indices were used to determine community diversity. Shannon’s indices of the HL group were significantly lower than that of the HLB group, indicating berberine supplementation in the HL diet improved the diversity of intestinal microbiota significantly, whereas Simpson’s indices showed no significant difference among different treatments. Observed species and Chao1 in the HL group were significantly lower than those of the Con group, indicating that high-lipid treatment reduced richness in the intestinal microbiota of black sea bream.

A sum of 144 operational taxonomic units (OTUs) was shared among the four treatments. Fish fed the Con diet had the highest unique OTUs (195), followed by those from the fish fed the HLB (175), ConB (134), and HL (129) diets (Figure 5).

3.5. Community Composition of Intestinal Microbiota

Proteobacteria was the most abundant phylum in all treatments, followed by Cyanobacteria, Firmicutes, Bacteroidetes, and Acidobacteria (Table 7). The fish fed HL diet had a significantly higher abundance of Bacteroidetes than that of the Con and HLB groups. Fish fed HLB and Con diets had similar intestinal microbiota composition in phylum (Figure 6(a)). Pseudomonas, Ralstonia, Chloroplast_unclassified, Brevundimonas, Weissella, Hydrogenophaga, Caulobacter, Sphingomonas, and Novosphingobium were the top 10 genera in the intestinal microbiota of black sea bream fed the different diets (Table 7). At the genus level, fish fed Con and HLB diets had a similar intestinal composition, while the ConB and HL groups shared the close bacterial genera composition (Figure 7). Fish fed the HL diet had a significantly lower abundance of Pseudomonas and Hydrogenophaga than that of fish fed the HLB diet () (Table 7). And significant interaction between berberine supplementation and diet was observed in the abundances of Pseudomonas, Ralstonia, and Hydrogenophaga.

Linear discriminant analysis (LDA) effect size (LEfSe) analysis was used to estimate the relative bacterial abundance in the taxon of black sea bream fed different diets. Fish fed Con, ConB, HL, and HLB diets had 16, 2, 9, and 16 significantly enriched taxa than the other treatments, respectively (Figure 8). These 43 taxa were classified from 5 orders, 7 families, 13 genera, and 18 species. Biomarkers of the Con group included order Bacillales, family Bacillacease, and genus Anaerotruncus, Pseudoxanthomonas, Eisenbergiella, Blautia, Oceanobacillus, and Clostridium. The biomarker of the ConB group included genus Comamonas. Biomarkers of the HL group included order Flavobacteriales, Actinobacteria-unclassified, family Weeksellaceae and Actinobacteria-unclassified, genus Actinobacteria-unclassified and Elizabethkingia. Biomarkers of the HL group included order Pseudomonadales and Enterobacterales, family Pseudomonadaseae, Moraxellaceae, Planococcaceae, and Enterobacteriaceae, and genus Pseudomonas, Psychrobacter, Ruegeria, and Lysinibacillus.

4. Discussion

Massive hepatic lipid accumulation and increased liver nf-κb, il-1β, and tnfα expression levels caused by high dietary lipid have been reported in black sea bream [29, 30]. Normally, too much lipid deposition in the liver results in endoplasmic reticulum stress, which may promote proinflammatory factor (like il-1β, and tnfα) expressions [2931]. High-lipid diet causing liver inflammation has been reported in many fish species, like blunt snout bream [22] and turbot (Scophthalmus maximus) [32]. In the present study, berberine supplementation in high-lipid diet significantly downregulated the expression levels of these proinflammatory factors, and berberine supplementation in normal and high-lipid diets also promoted anti-inflammatory cytokine (il-10 and tgfβ-1) expression levels. These results indicate that berberine could reduce liver inflammation induced by high-lipid diet. The anti-inflammation function may be attributed to the ability of berberine to alleviate liver lipid accumulation as reported by Wang et al. [19]. Similarly, other hepatic lipid-lowering additives also showed an anti-inflammation effect in black sea bream, such as L-carnitine and bile acid [29], choline [30] and fenofibrate [31]. Studies in mice found that berberine alleviated nonalcoholic steatohepatitis by inhibiting phosphoinositide 3-kinase (PI3K), NF-κB, and IL-8 in the cysteine X cysteine ligand 12/cysteine X cysteine receptor 4 (CXCL12/CXCR4) inflammatory signaling pathway [33], or by activating AMPK signaling and decreasing the TLR4/NF-κB-p65 signaling pathway [34]. In addition, the anti-inflammatory effect of berberine has been reported in 3T3-L1 adipocytes [35] and visceral adipose tissue [36]. However, in the present study, tnfα, nf-κb, and tgfβ-1 expression levels in the foregut were not significantly influenced by high-lipid diet or berberine supplementation. The rebalanced intestinal microbiota by berberine supplementation might be a contributing factor since the intestinal immune response is highly related to microbiota dysbiosis [24]. However, high-lipid diet upregulated intestinal nf-κb [30], tnfα [29], and il-1β [31] expressions in black sea bream in the previous studies. A study in blunt snout bream observed that 50 mg/kg berberine supplementation in high-lipid and high-carbohydrate diets significantly inhibited intestinal tnfα and il-6 expressions [17]. These inconsistencies could be due to the differences in basal diet, fish species and size, the segment of intestine for analysis, and environmental conditions.

In the present study, high-lipid diets (HL and HLB) reduced serum GSH-Px activity significantly, while berberine supplementation in normal and high-lipid diets improved serum CAT activity significantly. Berberine was proven to quench superoxide anions and nitric oxide effectively [6], and it can reduce ROS by the p38 MAPK pathway [5]. Previously, berberine supplementation in high-starch diet enhanced serum GSH-Px activity before and after ammonia stress challenge in black sea bream [37, 38]. Chen et al. [25] observed that 50 mg/kg berberine supplementation in a HL diet reduced hepatic MDA, lipid peroxide, and protein carbonyl contents of blunt snout bream through elevating SOD activity and total sulfhydryl levels. Another study in blunt snout bream found that the HF (15% fat) diet damaged mitochondria normal structure and induced more ROS production in the liver, while berberine supplementation reduced hepatic MDA content along with increased glutathione (GSH) activity via upregulating Sirtuin 3 expression and increasing the complex I and II activities [22]. Xu et al. [39] also reported that berberine (50 mg/kg) supplementation in normal and high-fat diets improved hepatic CAT and SOD activities with a decreased MDA content regardless of different feeding modes. Chen et al. [27] demonstrated that berberine could eliminate hepatic ROS production and reduce DNA damage caused by the high-cholesterol diet of zebrafish. However, these studies did not determine the antioxidative parameters in serum, whereas the present study found that berberine supplementation enhanced serum antioxidative status more significantly than that of the liver. This could be because berberine was widely distributed in the organs and tissues after oral administration, and it enhanced the whole-body antioxidative status. Interestingly, the significantly upregulated expression of Cu/Zn-sod in the foregut was only observed in the ConB group, and hepatic CAT activity of the ConB group was significantly higher than the other groups, indicating that the antioxidative effect of berberine might be influenced by dietary lipid level.

However, intestinal morphology was affected by the high-lipid diet in the present study, such as a reduced muscular mucosae thickness, villus length, and expanded lamina propria. The intestinal physical barrier directly determines the absorption capacity of nutrients, and its integrity is critical for intestinal health [40]. A high-fat diet may disrupt the gut barrier function [41]. Endotoxin, mainly are lipopolysaccharides, can damage the intestinal structure and increase intestinal permeability, as well as trigger inflammation response and change intestinal microbial community [37, 38]. Thus, the high-fat diet might induce more endotoxin and cause intestinal structural changes in the present study, whereas berberine supplementation in high-lipid diet alleviated the pathological changes caused by the high-lipid diet. It has been proved that berberine inhibits the intestinal secretory response of bacterial enterotoxin, regulates intestinal motility, and restores intestinal barrier function [42]. Berberine alleviated sparsely arranged intestinal mucosal villi and decreased gap junctions in nonalcoholic fatty liver disease (NAFLD) rats [37, 38]. In the present study, berberine supplementation in the normal diet increased intestinal microvillus density significantly, and both dietary lipid level and berberine were contributing factors that influenced microvillus density. Likewise, blunt snout bream fed high-lipid or high-carbohydrate diet caused a decrease of microvilli length, but berberine supplementation increased it significantly [17]. In this study, black sea bream fed the HL diet exhibited increased goblet cells count, which might be attributed to the alteration in mucin dynamic [43]. Generally, high-lipid diet triggers more liberation of mucin in the intestine to relieve the intestinal inflammation process, which induces a vast secretion of molecules by the host to maintain the intestinal barrier function and a viscoelastic protective layer composed of mucins secreted by goblet cells [24]. Thus, the increased goblet cells in HL treatment might be compensation for impaired intestinal immunity. In accordance with the present study, increased mucus production and goblet cells in the intestine induced by a high-fat diet was reported in zebrafish [24]. Studies in Atlantic salmon (Salmo salar L.) and grass carp also demonstrated that intestinal inflammation is correlated to increased goblet cell numbers [44, 45]. In addition, the intestinal morphology alteration might be related to the changes in the gut microbiota. Germ-free rodents had fewer goblet cells with a smaller size compared with conventionally raised mice. Besides, the mucus layer in conventionally raised mice was twice in thickness as that of germ-free rodents [43].

Host gut microbiota is influenced by the proportion and composition of dietary lipids [46]. In the present study, high-lipid diet reduced the richness and diversity of gut microbiota. A similar result was also observed in rice field eel (Monopterus albus) [23]. Li et al. [12] pointed out that healthy fish has a relatively higher bacterial diversity than diseased fish and reduced microbiota richness is a sign of some harmful conditions. For example, germ-free fish suffered impaired nutrient absorption, immunosuppression, intestinal epithelial cell dysfunction, altered metabolism, and weakened barriers against pathogens [47]. Studies in humans found that reduced microbial diversity might link to some diseases, such as type-2 diabetes and obesity [48]. Similarly, a high soybean meal diet also reduced intestinal microbiota in yellow drum (Nibea albiflora) [49]. In the present study, berberine supplementation in high-lipid diet improved the richness and diversity of the intestinal microbiota, indicating that berberine could rebalance intestinal microbiota impaired by high-lipid diet. In addition, Pan et al. [3] reported that compared with the control group, alpha diversity was decreased during berberine feeding, while an opposite trend showed after berberine feeding stopped in the berberine group, indicating that berberine might maintain the stability of gut microbiota of grass carp.

In the only study performed on black sea bream, Firmicutes and Proteobacteria were the dominant phyla in the intestine [50]. But the present study found that the main intestinal phyla were Proteobacteria and Cyanobacteria in black sea bream. Intestinal microbiota can be affected by certain factors, including genetic background, living environment, and diet ([51]). Proteobacteria being the main phylum in the intestine is inconsistent with the results in other aquatic animals, such as swimming carb [10], zebrafish [24], white leg shrimp (Penaeus vannamei) [11], channel catfish (Ictalurus punctatus) [52], and blunt snout bream [17].

In the present study, fish fed the HL diet had a significantly higher community of Bacteroidetes in the intestine, while berberine supplementation in high-lipid diet reduced the abundance of Bacteroidetes significantly. Similarly, phylum Bacteroidetes abundance was increased by a high-fat diet in zebrafish [24]. However, grass carp fed a berberine-supplemented diet for 28 days had a relatively increased abundance of Firmicutes, Bacteroidetes, and Proteobacteria in the intestine [3]. A study in hypercholesteraemic hamsters found that berberine treatment could increase the Firmicutes-to-Bacteroidete (F/B) ratio [53], which is associated with intestinal energy absorption, and a higher ratio could be an indication of more intestinal energy harvest [15, 54]. In the present study, berberine supplementation in the normal diet decreased intestinal F/B ratio but increased it when berberine was supplemented in the high-lipid diet. Decreased intestinal or fecal F/B ratio by berberine treatment was reported in high-fat diet-fed rats [55] and mice [16]. However, berberine treatment improved the intestinal F/B ratio in diabetes rats [56]. The variances of the F/B ratio by berberine treatment deserve further study.

In the present study, berberine supplementation in the high-lipid diet restored intestinal Pseudomonas richness decreased by high-lipid diet (Table 7). A study in Pacific white shrimp (Litopenaeus vannamei) found that supplementation of probiotic Bacillus coagulans increased Pseudomonas community in the intestine [57]. Besides, berberine was found to inhibit some opportunistic pathogens’ growth, such as Prevotella and Proteus, but increased the abundance of some probiotics, like Lactobacillus and Akkermansia [58]. In this study, Lactobacillus and Weissella abundances in the HLB and Con groups were higher than that of the HL group. These two bacteria are probiotic candidates belonging to lactic acid bacteria and can retard intestinal pathogenic colonization in the epithelial layer and enhance fish resistance to infection [59]. Increased Lactobacillus and Weissella abundance was reported in turbot-fed fructooligosaccharide-supplemented diet [59]. Besides, administration of Lactobacillus improved growth performance has been reported in Nile tilapia (Oreochromis niloticus) [60], grass carp [61], and black sea bream [62]. And replacement of fishmeal with meat and bone meal reduced intestinal Weissella abundance of gilthead sea bream (Sparus aurata) [63]. According to the LEfSe result, Actinobacteria abundance was higher in the HL group than that of the HLB group; a similar result was also found in obese rats treated with berberine [13]. In addition, the abundances of Ruegeria, Devosia, and Rhodobacteraceae were higher in the HLB group than that of the other groups. Ruegeria is a Gram-negative bacteria and belongs to the class Alphaproteobacteria [64]. It is used as a probiotic and symbiotic with unicellular eukaryotic phytoplankton [65]. Dietary tributyrin supplementation in a high soya bean meal diet increased intestinal Devosia abundance in yellow drum [49]. Rhodobacteraceae belongs to photosynthetic bacteria that contain abundant functional factors and nutrients, indicating it could be a potential probiotic [66, 67]. However, it is unclear why berberine supplementation in the normal diet modulated intestinal microbiota composition similar to that of the HL group, which deserves further study.

5. Conclusion

In conclusion, berberine supplementation in high-lipid diet downregulated liver proinflammatory cytokine expressions but promoted anti-inflammatory cytokine expressions, improved intestinal villus length and muscularis mucosae thickness, rebalanced intestinal microbiota diversity, and promoted some intestinal probiotic growth of black sea bream. 50 mg/kg dietary berberine was a safe and effective dosage for alleviating high dietary lipid stress in the liver and intestine. However, the alleviation effect of berberine may vary with dietary lipid and berberine levels, which deserves further study.

Data Availability

Derived data supporting this study are available from the corresponding author on reasonable request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Authors’ Contributions

Lei Wang and Qingjun Shao designed this experiment. Gladstone Sagada, Bingying Xu, and Jinzhi Zhang provided help during the feeding trial and sampling. Lei Wang wrote this manuscript, and Gladstone Sagada helped revise it. All authors read and approved the final manuscript.


This work was funded by the Ministry of Science and Technology, China (Project No: 2020YFD0900801) and the Natural Science Foundation of Universities of Anhui Province (Project No: KJ2021A0105).