Growth performance, bile acid profile, fecal microbiome and serum metabolomics of growing-finishing pigs fed diets with bile acids supplementation

Abstract

The present experiment was conducted to determine the effect of bile acids (BAs) supplementation on growth performance, BAs profile, fecal microbiome, and serum metabolomics in growing-finishing pigs. A total of 60 pigs [Duroc × (Landrace × Yorkshire)] with an average body weight of 27.0 ± 1.5 kg were selected and allotted into one of 2 groups (castrated male to female ratio = 1:1), with 10 replicates per treatment and 3 pigs per replicate. The 2 treatments were the control group (control) and a porcine bile extract-supplemented group dosed at 0.5 g/kg feed (BA). After a 16-wk treatment, growth performance, BAs profiles in serum and feces, and fecal microbial composition were determined. An untargeted metabolomics approach using gas chromatography with a time-of-flight mass spectrometer was conducted to identify the metabolic pathways and associated metabolites in the serum of pigs. We found that BAs supplementation had no effect on the growth performance of the growing-finishing pig. However, it tended to increase the gain-to-feed ratio for the whole period (P = 0.07). BAs supplementation resulted in elevated serum concentrations of secondary bile acids, including hyodeoxycholic acid (HDCA), glycoursodeoxycholic acid, and tauro-hyodeoxycholic acid, as well as fecal concentration of HDCA (P < 0.05). Fecal microbiota analysis revealed no differences in alpha and beta diversity indices or the relative abundance of operational taxonomic units (OTUs) at both phylum and genus levels between groups. Metabolic pathway analysis revealed that the differential metabolites between control and BA groups are mainly involved in purine metabolism, ether lipid metabolism, glycerophospholipid metabolism, and amino sugar and nucleotide sugar metabolism, as well as primary bile acid biosynthesis. Our findings indicate that BAs supplementation tended to improve the feed efficiency, and significantly altered the BA profile in the serum and feces of growing-finished pigs, regardless of any changes in the gut microbial composition. The altered metabolic pathways could potentially play a vital role in improving the feed efficiency of growing-finished pigs with BAs supplementation.

Bile acids supplementation tended to improve the feed efficiency, and significantly altered the serum and fecal BAs profile and serum metabolomics profile of growing-finished pigs, regardless of any changes in the gut microbial composition. The altered metabolic pathways could potentially play a vital role in improving the feed efficiency of growing-finished pigs with BAs supplementation.

Introduction

Bile acids (BAs) are amphiphilic molecules of bile which are exclusively synthesized from cholesterol in the liver and secreted into the small intestine. In addition to their role in emulsification and absorption of dietary fat in the intestinal lumen, BAs have also become appreciated as important regulators of intestinal function, lipid, and energy metabolism. Due to their potential effects, dietary BAs have been extensively used in modern animal farming, particularly in broilers and fish. For broilers, special focus was given to the use of dietary BAs at their early stage during which limited endogenous bile was generated (Arshad et al., 2021). Experimentally, better feed conversion rate (Lai et al., 2018; Ge et al., 2019), carcass characteristics (Lai et al., 2018; Ge et al., 2019), or fat digestibility (Lammasak et al., 2019) were observed in response to individual or combinations of BAs supplementation during starter or finisher phases. For fish, high-fat diets commonly used in the industry resulted in excess hepatic lipid accumulation and impaired liver function (Liao et al., 2020). BAs have been identified to be beneficial for growth performance, digestive function, or hepatic lipid metabolism in various fish (Jiang et al., 2018; Liao et al., 2020). However, very few studies related to dietary BAs have been tested in pigs, and the few studies available are primarily focused on weaned piglets. Weaned piglets secrete only small amounts of BAs and have limited ability to emulsify dietary fats (Jones et al., 1991). Supplementation of choline and BAs mixtures improved average daily gain (ADG) and gain-to-feed ratio (G:F), associated with an elevation of lipase activity of weaned piglets (Qiu et al., 2021). Besides, individual BAs, hyodeoxycholic acid (HDCA), or chenodeoxycholic acid (CDCA) supplementation was identified to regulate enteroendocrine cell differentiation (Zong et al., 2019) or to enhance the protection and barrier function of the mucosa of the distal small intestine (de Diego-Cabero et al., 2015).

BAs are composed of primary bile acids (PBA) and secondary bile acids (SBA). The PBA, which are synthesized from cholesterol in the liver, are usually conjugated with glycine and taurine to make them more water-soluble before secretion. The majority of BAs are reabsorpted in the ileum, and only a small amount reaches the colon and is metabolized by intestinal microbiota into SBA via biochemical reactions, including deconjugation and dihydroxylation (Theriot et al., 2016). The microbial enzyme bile salt hydrolases can cause deconjugation of glycine and taurine from the conjugated BAs, and deconjugated glycine and taurine are further metabolized to carbon dioxide and ammonia, which serve as an energy source for certain bacteria (Long et al., 2017). Another kind of microbial enzyme, 7α-dehydratase, is expressed strictly in anaerobic bacteria (Clostridium species) (Mullish et al., 2018), and performs the removal of the hydroxyl group from the unconjugated BAs to synthesize SBA. All of the above imply the interaction between BAs and gut microbiota. Accordingly, we speculated that alterations in either of them would have an impact on host metabolism. However, only a few studies have been conducted to investigate the effect of dietary BAs supplementation on gut microbial changes (Islam et al., 2011; Qiu et al., 2021; Wan et al., 2020; Song et al., 2021).

In view of the limited knowledge about the effect of dietary BAs on growth performance and microbial changes, particularly in young pigs, the objective of this study was to determine the impact of BAs supplementation on the growth performance, BAs profile, fecal microbial composition, and serum metabolomics of growing-finishing pigs.

Materials and Methods

Animal ethics statement

The Animal Welfare Committee of Southwest University of Science and Technology approved all research protocols used for this experiment (Reference number: L2022011).

Preparation of BAs

The supplemented BAs were supplied by Shandong Longchang Animal Health Care Co. Ltd. (Shandong, China), and extracted from porcine bile using a process involving saponification, decolorization, acidification, purification, and desiccation. The analyzed profiles of BAs are summarized in Table 1.

Table 1.

Analyzed profiles of the bile acids supplements¹

Components	Concentration, ng/mg	Components	Concentration, ng/mg
Primary bile acids		Secondary bile acids
HCA	65.8497	THDCA	0.9477
CDCA	2.3611	GUDCA	1.6422
GCDCA	2.0147	TUDCA	3.6520
TCDCA	1.9281	TLCA	0.6454
CA	0.4093	DCA	0.0833
TCA	0.4085	TDCA	7.9003
Total	72.9714	Total	14.8709

¹HCA, hyocholic acid; CDCA, chenodeoxycholic acid; GCDCA, glycochenodeoxycholic acid; TCDCA, taurochenodeoxycholic acid; CA, cholic acid; TCA, taurocholic acid; THDCA, auro-hyodeoxycholic acid; GUDCA, glycoursodeoxycholic acid; TUDCA, tauroursodeoxycholic acid; TLCA, tauro-lithocholic acid; DCA, deoxycholic acid; TDCA, tauro-deoxycholic acid.

Animals and experimental design

A total of 60 pigs [Duroc × (Landrace × Yorkshire)] with an average body weight of 27.0 ± 1.5 kg were selected and allotted into 1 of 2 groups (castrated male-to-female ratio = 1:1), with 10 replicates per treatment and 3 pigs per replicate. The two treatments were the control group (control) and the BAs supplemented group (BA).

Diet and feeding

All diets were formulated according to the nutrient requirements of growing-finishing pigs as recommended by the NRC (2012). The ingredient composition of the basal diets is shown in Table 2. The BA diet was made by premixing 0.5 g BAs per kg of feed into corn flour and then added to the basal diet. All pigs had ad libitum access to feed and water and were housed in an environmentally controlled room with a temperature in the range of 18 to 22 °C during the 16-wk experimental period.

Table 2.

The ingredient composition of basal diets (as fed basis)

Ingredients, g/kg	Weeks 1 to 8	Weeks 9 to 12	Weeks 13 to 16
Corn	685	699	730
Soybean meal	200	180	150
Wheat bran	70	50	50
Corn germ meal	-	30	30
Soybean oil	10	10	10
Dicalcium phosphate	8.0	7.5	7.5
Calcium carbonate	6.6	5.2	4.7
L-Lysine HCl, 70%	7.5	6.5	6.0
L-Threonine, 99%	1.5	1.0	1.0
DL- Methionine, 99%	1.0	0.5	0.5
L-Tryptophan, 98.5%	0.4	0.3	0.3
Sodium chloride	4.0	4.0	4.0
Choline chloride (50%)	1.0	1.0	1.0
Premix1	5.0	5.0	5.0
Total	1,000	1,000	1,000
Calculated nutrient level, g/kg
DE, MJ/kg	13.43	13.59	13.63
SID-Lys (%)	1.08	0.96	0.87
SID-Met (%)	0.46	0.39	0.37
SID-Trp(%)	0.17	0.15	0.14
SID-Thr (%)	0.55	0.48	0.44
Analyzed nutrient level, g/kg DM
GE, MJ/kg energyy	18.83	18.79	18.76
Dry matter, g/kg as feed	857	864	861
Ash	25	24	23
Crude protein	157	152	141
Crude fat	39	38	39
Crude fiber	29	27	28
Calcium	5.6	4.9	4.6
Total phosphorus	4.5	4.0	3.7

¹Supplied per kilogram of feed: 8,000 IU Vitamin A, 2,000 IU Vitamin D₃, 12 IU Vitamin E, 1.5 mg B₁ Vitamin, 4 mg B₂ Vitamin, 2 mg B₆ Vitamin, 0.02 mg B₁₂ Vitamin, 1.2 mg K₃ Vitamin, 12 mg pantothenic acid, 20 mg niacin, 0.08 mg biotin, 0.5 mg folic acid, 120 mg FeSO₄·7H₂O, 20 mg CuSO₄, 30 mg MnSO₄·H₂O, 120 mg ZnSO₄·H₂O, 0.5 mg Na₂SeO₃, 0.5 mg KI.

Data collection and sampling

Feed consumption per replicate was measured daily, and growth performance data, including ADG, average daily feed intake (ADFI), and G:F, were calculated per 4 wk.

At the end of the trial, 1 pig per replicate with 10 replicates per treatment was randomly selected to collect blood and fecal samples. Fasting blood samples were collected in duplicate into two 5 mL tubes without anticoagulant and left at room temperature for 2 h, followed by centrifugation at 2,550 × g at 4 °C for 10 min. Serum samples were harvested and stored at −20 °C until analysis. The fresh feces of each pig were collected into sterile tubes and kept on ice until transferred to a freezer at −20 °C.

Feed chemical analyses

Gross energy was determined using an adiabatic bomb calorimeter (Parr Instrument Company, Moline, IL, USA). Dry matter (DM), crude fat, crude protein, Ca, and P were analyzed according to the AOAC method (AOAC, 2007). Crude fiber was determined using a fiber analyzer (ANKOM Technology, Macedon, NY, USA). All chemical analyses were performed in duplicate.

BAs extraction and quantification

The extraction of BAs from serum followed procedures described by Fang et al. (2019). For the extraction procedure, 100 μL of serum was mixed thoroughly with 100 μL of cold sodium acetate buffer (50 mM, pH 5.6) and 300 μL of ethanol, and incubated for 30 min at 4 °C followed by centrifugation at 20,000 × g for 20 min. The supernatant was collected and diluted 4 times with sodium acetate buffer before being applied to a Bond Elute C18 cartridge (Agilent 12102161, Harbor City, CA, USA) which was pre-activated with 5 mL of methanol. After washing with 25% ethanol and eluting with 5 mL of methanol, the solvent was removed under nitrogen gas, and the residue was resuspended with 1 mL of methanol, followed by filtrating with 0.45 μm nylon syringe filters. The filtrate was collected for further BAs quantification.

The extraction of BAs from feces followed the procedures described by Fang et al. (2018). Feces were suspended in cold sodium acetate buffer (50 mM, pH 5.6) and then refluxed with ethanol for 1 h, followed by centrifugation at 20,000 × g for 15 min. The supernatant was collected and diluted 4 times with ultrapure water before being applied to a Bond Elute C18 cartridge (Agilent 12102161). After washing with 20% ethanol and eluting with 5 mL of methanol, the solvent was treated using the same method as described above.

Finally, BAs quantification was performed using Ultimate 3000 high-performance liquid chromatography coupled with Waters Xevo TQ LC/MS mass spectrometer with an ESI source as reported by Fang et al. (2018).

Untargeted serum metabolomics analysis

Untargeted serum metabolomics analyses were carried out using gas chromatography with a time-of-flight mass spectrometer (GC-TOF-MS). Firstly, metabolite extraction was performed using the liquid-liquid extraction method, and ribitol was selected as an internal standard (20 μL stock solution of 20 mg/mL in H₂O). After the procedure of extraction and derivatization, GC-TOF-MS was applied to analyze the metabolites in all samples. The raw data were processed using MassLynx software (Waters Co., Milford, MA, USA) for data alignment, normalization, and peak picking. Candidate metabolites were identified by searching the exact molecular mass data from the NIST library and the Human Metabolome Database (HMDB). (http://www.hmdb.ca/).

DNA extraction and 16S rRNA gene sequencing

Microbial DNA was extracted from 0.25 g of thawed stool samples using the Mo Bio PowerFecalTM DNA Isolation Kit (MOBIOLaboratories, Carlsbad, CA, USA) according to the manufacturer’s protocol. Before sequencing, the concentration and purity of the extracted genomic DNA were measured. Extracted fecal DNA samples were sent to Novogene Co., Ltd. (Beijing, China) to perform amplicon pyrosequencing on the Illumina NovaSeq platforms. The V4 hypervariable region of the 16S rRNA gene was amplified using 515F and 806R primers. The effective tags were mapped to operational taxonomic units (OTUs) using Uparse v7.0.1001 at 97% sequence similarity. Representative sequences for each OTU were selected. The Ribosomal Database Project (RDP) classifier Version 2.6 was used for taxonomic classification. The relative abundance of each OTU was examined at different taxonomic levels. The alpha and beta diversity calculations, as well as taxonomic community assessments, were performed by Qiime 1.9.1. The principal coordinate analysis (PCoA) was displayed with R software v2.15.3 to assess the clustering of fecal samples.

Statistics

The statistical analysis was performed using Student’s t-tests with SAS software (Version 9.3; SAS Institute Inc., Cary, NC, USA). The pen was recognized as a statistical unit for the performance data including ADG, ADFI, and G:F. The selected pig in each replicate for blood and fecal sampling was taken as an experimental unit for BA profile and bacterial alpha-diversity indexes. To compare data of relative abundance at different taxonomic levels between groups, the Wilcoxon rank-sum test was used. Values were expressed as mean ± SEM unless otherwise specified. All variables were considered significant when P ≤ 0.05, whereas 0.05 < P < 0.1 was considered as showing a trend.

For the metabolomics data, statistical analysis was performed in MetaboAnalyst (http://www.metaboanalyst.ca), including principal component analysis (PCA) and supervised orthogonal partial least-squares discriminant analysis (OPLS-DA). Variable importance in the projection (VIP) ranks the overall contribution of each variable to the OPLS-DA model, so variables with VIP value > 1, P value < 0.05, and fold change (FC) ≥ 2 or FC ≤ 0.5 were considered as differential metabolites between groups. The pathway enrichment and topological analysis were finally conducted in MetaboAnalyst.

Results

Analyzed profiles of the BAs supplements

As shown in Table 1, the concentration of PBA was much greater than that of SBA (72.9714 vs. 14.8709 ng/mg). Among the BAs profiles, hyocholic acid (HCA) had the greatest concentration with a value of 65.8497 ng/mg, which was more than 18 times greater than the concentration of the other BAs profiles. In comparison, the lowest concentration of 0.0833 ng/mg was detected for deoxycholic acid (DCA).

Growth performance of pigs

The growth performance of pigs is shown in Table 3. The initial body weight (BW) of growing pigs in both groups was comparable, and ADG, ADFI, and G:F for each growth period and the entire period were similar between the two groups (P > 0.05). However, it is worth noting that G:F for the entire period (weeks 1 to 16) tended to be greater in the BA group compared to the control group (0.360 vs. 0.347, P = 0.07).

Table 3.

Growth performance of growing-finishing pigs¹

Item	Control	BA	SEM	P values
Initial BW, kg	27.81	27.86	1.08	0.978
Weeks 1 to 4
ADG, kg/d	0.741	0.746	0.028	0.907
ADFI, kg/d	1.618	1.559	0.069	0.560
G:F	0.458	0.479	0.008	0.110
Week 5 to 8
ADG, kg/d	0.912	0.908	0.022	0.878
ADFI, kg/d	2.361	2.336	0.073	0.815
G:F	0.386	0.389	0.009	0.815
Week 9 to 12
ADG, kg/d	0.851	0.876	0.038	0.646
ADFI, kg/d	2.842	2.810	0.123	0.858
G:F	0.299	0.312	0.010	0.404
Week 13 to 16
ADG, kg/d	0.841	0.854	0.022	0.674
ADFI, kg/d	2.719	2.726	0.198	0.980
G:F	0.309	0.313	0.019	0.805
Week 1 to 16
ADG, kg/d	0.836	0.846	0.022	0.759
ADFI, kg/d	2.410	2.358	0.099	0.716
G:F	0.347	0.360	0.005	0.072

¹ADG, average daily gain; ADFI, average daily feed intake; G:F, gain:feed; control, pigs fed with basal diet; BA, pigs fed with diet which was made by adding 0.5 g bile acids per kg of feed to the basal diet; n = 10 for each group.

Serum and fecal BAs profiles

The serum BAs profile is presented in Figure 1. BAs supplementation increased the concentration of SBA (P = 0.03; Figure 1A). However, the concentration of total BA (TBA), PBA, glycine-conjugated BA (GCBA), and taurine-conjugated BA (TCBA) did not differ between the groups (P > 0.05). The significant increase in SBA concentration with BAs supplementation was mainly due to the increase in the levels of HDCA, glycoursodeoxycholic acid (GUDCA), and tauro-hyodeoxycholic acids (THDCA) (P < 0.05; Figure 1B). Furthermore, the unconjugated BAs, including HDCA and CDCA were found to be most abundant in serum.

Figure 1.

Alteration of BAs profiles in the serum after BAs supplementation. (A) Concentrations of total (TBA), primary (PBA), secondary (SBA), glycine-conjugated (GCBA), and taurine-conjugated bile acids (TCBA) in the serum of pigs. (B) Concentrations of individual bile acids in the serum of pigs. Control, pigs fed with basal diet; BA, pigs fed with diet which was made by adding 0.5g bile acids per kg of feed to the basal diet; n = 10 for each group. Each column and error bar represents the mean and SEM. *P < 0.05 were considered significantly different. CA, cholic acid; CDCA, chenodeoxycholic acid; TCA, taurocholic acid; GCA, glycocholic acid; TCDCA, taurochenodeoxycholic acid; GCDCA, glycochenodeoxycholic acid; THCA, taurohyocholic acid; DCA, deoxycholic acid; HCA, hyocholic acid; HDCA, hyodeoxycholic acid; TDCA, tauro-deoxycholic acid; TLCA, tauro-lithocholic acid; TUDCA, tauroursodeoxycholic acid; GUDCA, glycoursodeoxycholic acid; THDCA, tauro-hyodeoxycholic acid.

In feces, there was no difference in TBA, PBA, SBA, GCBA, and GCBA concentrations between groups (P > 0.05; Figure 2A). However, the concentration of HCA, one of the PBA, was significantly greater in the feces of pigs in the BA group compared to the control pigs (P < 0.05; Figure 2B). It is worth noting that the concentration of HDCA was orders of magnitude greater than the other BA profiles in the feces.

Figure 2.

Alteration of BAs profiles in the feces after BAs supplementation. (A) Concentrations of total (TBA), primary (PBA), secondary (SBA), glycine-conjugated (GCBA), and taurine-conjugated bile acids (TCBA) in the feces of pigs. (B) Concentrations of individual bile acids in the feces of pigs. Control, pigs fed with basal diet; BA, pigs fed with diet which was made by adding 0.5 g bile acids per kg of feed to the basal diet; n = 10 for each group. Each column and error bar represents the mean and SEM. *P < 0.05 were considered significantly different. CA, cholic acid; CDCA, chenodeoxycholic acid; TCA, taurocholic acid; GCA, glycocholic acid; TCDCA, taurochenodeoxycholic acid; GCDCA, glycochenodeoxycholic acid; THCA, taurohyocholic acid; DCA, deoxycholic acid; HCA, hyocholic acid; HDCA, hyodeoxycholic acid; TDCA, tauro-deoxycholic acid; TLCA, tauro-lithocholic acid; TUDCA, tauroursodeoxycholic acid; GUDCA, glycoursodeoxycholic acid; THDCA, tauro-hyodeoxycholic acid.

Fecal microbiota

As shown in the rarefaction curve, all sample analyses reached a stable plateau at 40,000 sequences and 900 OTUs (Supplementary Figure S1), indicating sufficient sequencing depth to capture the species richness of the fecal samples. At the phylum level, all fecal samples were dominated by four phyla: Bacteroidetes, Firmicutes, Spirochaetes, and Euryarchaeota (Supplementary Figure S2A). The top 10 phyla were selected for significance analysis, which indicated that all top 10 phyla were comparable between the two groups (P > 0.05; Supplementary Table S1). At the genus level, the top four dominant genera were Methanobrevibacter, Megasphaera, Lactobacillus, and unidentified_Prevotellacea (Supplementary Figure S2B). The top 20 genera were selected for significance analysis, and no significant differences in relative abundance were found between the groups (P > 0.05; Supplementary Table S2). However, at the species level, BAs supplementation increased the relative abundance of Alloprevotella_sp_feline_oral_taxon_309 (P = 0.03) among all top 20 species (Supplementary Table S3). The alpha diversity of bacteria communities was not significantly different between the groups (P > 0.05; Supplementary Figure S3).

The PCoA based on weighted Unifrac distance metrics was performed to measure the similarity of fecal bacterial composition between the control and BA groups. The result showed no clear separation and clustering for the two groups (Figure 3).

Figure 3.

Principal coordinate analysis (PCoA) of bacterial communities obtained from control (C) and BAs supplemented (BA) pigs. Control, pigs fed with basal diet; BA, pigs fed with diet which was made by adding 0.5 g bile acids per kg of feed to the basal diet.

Serum metabolomics

Metabolomics analyses were carried out using the GC-TOF-MS approach to determine the metabolites of serum samples obtained from control and BA pigs. The OPLS-DA plots of the metabolomic data showed a clear separation between the control and the BA groups without any overlap (Figure 4), indicating significant changes in serum metabolic profile with BAs supplementation. The major metabolites contributing to the sample separation were further identified by structural and quantitative analyses. Forty-one metabolites were identified as significantly different (VIP > 1, P < 0.05, and FC) between the 2 groups, of which 23 metabolites were up-regulated and 18 metabolites were down-regulated in BA pigs (Table 4). The up-regulated metabolites in BA pigs included 29-demethylgeodisterol-O-sulfite, glycocholic acid (GCA), and codonocarpine, while the down-regulated ones included cytidine monophosphate (CMP)-N-glycoloylneuraminate, bromocriptine, 1-phenyl-1-pentanone, and frangulin A compared to the control pigs. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis revealed that these differently enriched metabolites were mainly involved in purine metabolism, ether lipid metabolism, glycerophospholipid metabolism, and amino sugar and nucleotide sugar metabolism, as well as primary bile acid biosynthesis (Figure 5).

Figure 4.

The orthogonal partial least-squares discriminate analysis (OPLS-DA) score plots of serum metabolites obtained from control (C) and BAs supplemented (BA) pigs. Control, pigs fed with basal diet; BA, pigs fed with diet which was made by adding 0.5 g bile acids per kg of feed to the basal diet. Individual pigs are shown as circles.

Table 4.

Main significant metabolites in serum of C and BA pigs^1,2

Metabolites	Fold change (BA/C)	Regulated	VIP value	P value
29-Demethylgeodisterol-O-sulfite	7.4399	Up	1.9746	0.0130
GCA	5.3875	Up	1.7605	0.0300
Codonocarpine	4.6407	Up	1.8916	0.0183
Inosine	3.9745	Up	2.1327	0.0064
Famotidine	3.9572	Up	2.3454	0.0021
Isoprothiolane	3.5394	Up	2.2327	0.0038
Allopurinol-1-ribonucleoside	3.5095	Up	2.2254	0.0040
Guanine	3.3116	Up	2.1826	0.0050
Docosa-4,7,10,13,16-pentaenoyl carnitine	2.8795	Up	1.8889	0.0185
Hypoxanthine	2.8746	Up	2.2235	0.0040
LysoPC (22:5)	2.6118	Up	2.2624	0.0033
Arachidonoyl dopamine	2.6081	Up	1.8078	0.0252
Dioxibrassinin	2.5694	Up	2.1228	0.0067
PS (18:1/0:0)	2.4728	Up	1.7323	0.0331
Callystatin A	2.4394	Up	2.0547	0.0092
PI (20:4/0:0)	2.3574	Up	2.0213	0.0107
Glycerophosphocholine	2.3365	Up	2.3033	0.0026
PS (18:1)	2.2718	Up	2.0508	0.0093
Methyl methylthio selenide	2.1987	Up	2.2026	0.0045
Chenodeoxycholic acid 3-sulfate	2.1465	Up	2.0045	0.0115
2-(Methylthio)-3H-phenoxazin-3-one	2.0372	Up	1.9678	0.0134
Nap-His-OH	0.3889	Down	2.4979	0.0008
Glycineamideribotide	0.3651	Down	1.6834	0.0392
Cyclochlorotine	0.3276	Down	1.8482	0.0217
Glucosyloxyanthraquinone	0.2553	Down	2.4352	0.0012
Acetylcarnitine	0.2237	Down	2.5564	0.0005
Isoeugenitol	0.2198	Down	2.1212	0.0067
Dihydrozeatin riboside monophosphate	0.2070	Down	2.4658	0.0010
4-Amino-2-methyl-5-phosphomethylpyrimidine	0.1923	Down	2.4261	0.0013
Auramycinone	0.1893	Down	2.4482	0.0011
Frangulin A	0.1773	Down	2.4863	0.0008
1-Phenyl-1-pentanone	0.0907	Down	1.6360	0.0458
Bromocriptine	0.0726	Down	1.8866	0.0187
CMP-N-glycoloylneuraminate	0.0622	Down	1.7678	0.0292

¹The metabolites with VIP value > 1, P value < 0.05, and fold change (FC) ≥ 2 or FC ≤ 0.5 were listed as significant differential metabolites between groups.

²GCA, glycocholic acid; LysoPC, lysophosphatidylcholine; PI, glycerophosphoinositols; PS, phosphatidylserine; CMP, cytidine monophosphate; C, pigs fed with basal diet; BA, pigs fed with diet which was made by adding 0.5 g bile acids per kilogram of feed to the basal diet; n = 10 for each group.

Figure 5.

Kyoto Encyclopedia of Genes and Genomes (KEGG) metabolic pathway enrichment analysis based on significant differential metabolites in serum from control and BA pigs. Larger sizes and darker colors represent greater pathway enrichment and higher pathway impact values, respectively. Control, pigs fed with basal diet; BA, pigs fed with diet which was made by adding 0.5 g bile acids per kilogram of feed to the basal diet.

Discussion

To our knowledge, little research so far has investigated the effect of BAs on the growth performance of pigs, particularly in growing-finishing pigs. In the present study, the supplementation of BAs complex showed no significant effects on the growth performance of growing-finishing pigs. Our results are in agreement with previous studies that tested BAs sources in weaned piglets, such as BAs complex (Qiu et al., 2021), individual HDCA (Zong et al., 2019), and CDCA (de Diego-Cabero et al., 2015). However, it is somewhat surprising that we only observed a tendency to increase the G:F of pigs with such a long experiment duration of 16 wk, compared to those of previous studies mentioned above, with durations ranging from 6 to 28 d. The most reasonable explanation could be the more mature stage of growth at which the pigs in our study were, as opposed to the weaning stage of piglets. Weaned piglets secrete only small amounts of BAs and have limited ability to emulsify dietary fats (Jones et al., 1991). This could explain the result obtained by Song et al. (2021) who found that CDCA supplementation for 30 d increased the final BW and ADG, and increased G:F in weaned piglets. By contrast, dietary BAs supplementation has shown promising effects in broilers at different rearing phases. Ge et al. (2019) found that combinations of BAs (mainly containing HDCA, CDCA, and HCA) improved the ADG and G:F of broilers during the starter phase. Similar results were reported by Lammasak et al. (2019) who fed broilers with lyophilized pig bile powder. However, other studies reported the beneficial effects of BAs on growth performance only during the finisher phase but not the starter phase of broilers (Lai et al., 2018). According to a review by Arshad et al. (2021), combinations of BAs have shown a greater potential to improve feed efficiency even at low levels (0.008%), compared to individual BAs. Based on these, we speculate that, in addition to the specific form of bile acid and its dosage, the growth stage of the tested animals and experimental duration also account for the influence of dietary BAs on the growth performance of animals.

The composition of the BA profile varies markedly among mammalian species. CA and DCA are the predominant BAs in bovines, whereas CDCA and CA are most abundant in avians. In pigs, HCA and its species, including HDCA, GHCA, THCA, GHDCA, and THDCA, constitute ~76% of the BA blood pool in pigs compared to the trace amounts in human blood (Zheng et al., 2021). The analyzed profiles of the BAs supplements used in our study, which were extracted from porcine bile, are in good agreement with the findings obtained by Zheng et al. (2021), with HCA being the most abundant one. Concomitantly, the HCA excretion in feces was much greater in response to BAs supplementation in the present study. In addition, earlier research has established that HDCA is formed secondarily from HCA by dehydroxylation during the enterohepatic circulation (Bergström et al., 1959). Therefore, the greater HDCA concentration in the serum of the BA group could be mainly ascribed to the high HCA concentration in the BAs supplements. HDCA has been shown to prevent gallstone formation and atherosclerotic lesion formation in mammals (Shih et al., 2013). THDCA, a taurine-conjugated derivatives of HDCA, has been demonstrated to increase biliary lipid secretion and is therefore cytoprotective for host hepatocytes (Puglielli et al., 1994). Both HDCA and THDCA, which are also known as HCA species, have been found recently to improve glucose homeostasis through a distinct TGR5 and FXR signaling mechanism (Zheng et al., 2021). Similarly, UDCA has been administrated as a therapeutic BA to protect against gallstones or nonalcoholic steatohepatitis (Gérard, 2014). Accordingly, we speculate that the greater concentration of serum HDCA and THDCA, as we observed in the present study, would be beneficial to the health of the finishing pigs, and further research is needed to investigate this hypothesis more thoroughly. It should also be noted that the HDCA concentration in feces was greater by orders of magnitude than the other BA profiles in each group, but was not even detected in the analyzed BA supplements. Generally, SBA, the more hydrophobic bile acid species, especially its unconjugated forms, have greater cytotoxic effects. For the tested pigs, the great excretion of HDCA in feces might be an adaptive adjustment and feedback regulation to the high serum BAs concentration.

Once upon being synthesized from cholesterol in the liver, the PBA are conjugated with glycine or taurine. Almost 95% of the BAs are reabsorbed into the liver from the distal ileum, while the remaining 5% reaches to the colon and is metabolized by intestinal bacteria. Hence, CA and CDCA are converted by the intestinal microbiota into DCA and LCA, respectively (Honda et al., 2020). Therefore, alterations in gut microbiota can have significant effects on BAs metabolism. Besides, BAs can affect the gut microbiota directly as an antibacterial substance (Kurdi et al., 2006), or indirectly by stimulating the expression of antimicrobial peptides through nuclear receptors (D’Aldebert et al., 2009). Given the strong link between BA metabolism and the gut microbiota, we analyzed the composition of fecal microbiota using 16S rRNA gene sequencing. Surprisingly, we did not observe dramatic changes in the fecal microbial composition, except for the increased relative abundance of the species Alloprevotella_sp_feline_oral_taxon_309 with BAs supplementation. Alloprevotella_sp_feline_oral_taxon_309, an acetic and succinic acid producer, has been demonstrated to exhibit anti-inflammatory effects in a mouse model of inflammatory bowel disease (Li et al., 2020). In the latest study, the relative abundance of colonic Alloprevotella_sp_feline_oral_taxon_309 significantly increased with the addition of dietary porcine intestinal antimicrobial peptide and showed a significant positive correlation with secretory immunoglobulin A in the jejunal mucosa of piglets (Ji et al., 2023). The above studies suggest that the rise in Alloprevotella_sp_feline_oral_taxon_309, induced by the addition of BAs in the current research, may play a crucial role in maintaining the intestinal immune homeostasis and enhancing resistance to intestinal infections in pigs. However, further research is needed to confirm these findings. There is little understanding of the effects of BAs mixture supplementation on gut microbiota in animals or humans. Conversely, changes in gut microbial composition in response to specific BAs have been found in previous studies. Feeding rats with CA induced phylum-level alterations, resulting in an increase in the Firmicutes at the expense of Bacteroidetes (Islam et al., 2011). Tauroursodeoxycholic acid (TUDCA) supplementation was found to improve murine pancreatitis in association with Lactobacillus (Wan et al., 2020). Moreover, Song et al. (2021) recently reported that supplementing the diet of weaned piglets with CDCA led to an increase in two beneficial gut bacteria, Prevotella 9 and Prevotellaceae TCG-001, and a decrease in a harmful bacterium, Dorea. The reason for the contradictions may be related to specific forms of the supplemented BAs, considering that CA, TUDCA, or CDCA constitute only a tiny fraction of the BAs supplements used in our study. However, the reason for the lack of effect, especially in terms of HCA, the dominant BA in the BAs supplements, warrants further greater scrutiny.

The differentially enriched metabolites in the serum of control and BA pigs were identified. The concentrations of 29-demethylgeodisterol-O-sulfite, GCA, codonocarpine, and inosine were found to be most up-regulated, while CMP-N-glycoloylneuraminate, bromocriptine, 1-phenyl-1-pentanone, and frangulin A were the most down-regulated ones. Codonocarpine is an important type of spermidine alkaloids, which have diverse biological and physiological functions and can be utilized as medications based on their antiviral activities (Faisal et al., 2023), as well as their anti-hyperglycemic and anti-hyperlipidemia effects (Zhang et al., 2018). Various findings have demonstrated that inosine plays a crucial role in numerous physiological and pathophysiological processes, including energy expenditure (Niemann et al., 2022), antitumor immunity (Luu and Visekruna, 2021), and colon mucosal barrier function (Li et al., 2021). In particular, Watkins et al. (2013) found that the presence of free amino acids, unsaturated fatty acids, and nucleotides, such as inosine, contributed to the distinct meat flavors of mutton. According to Zheng et al. (2022), CMP-N-glycoloylneuraminate showed a significantly negative correlation with high density lipoprotein cholesterol (HDL-C), suggesting a potential negative impact on lipid metabolism. Additionally, CMP-N-glycoloylneuraminate was found to be greater in malaria-infected mice groups (Das et al., 2022). Furthermore, we enriched the metabolic pathways of the above differential metabolites by KEGG. Most of these metabolites were accounted for in purine metabolism, ether lipid metabolism, glycerophospholipid metabolism, and amino sugar and nucleotide sugar metabolism, as well as primary bile acid biosynthesis. Among them, the significantly changed inosine, guanine, hypoxanthine, and glycineamideribotide, as important metabolites in purine metabolism, are the main contributors to the enriched purine metabolism pathway observed in this experiment. Our findings align with those of Alemán et al. (2018), who identified a positive correlation between purine metabolic pathways and the contents of fecal ursocholic and cholic acids, as well as specific genera, such as Roseburia and Ruminococcus, in women. Furthermore, Zhang et al. (2022) revealed an intimate relationship between guanosine and inosine, both involved in purine metabolism, and certain genera, including Lachnoclostridium, Enterorhabdus, Roseburia, Lachnospiraceae_UCG-006, and Colidextribacter. According to James et al. (2023), SBA converted from BAs, which are metabolites generated from probiotics, could impact purine metabolism. Purine metabolism was found to be closely related to nucleic acid synthesis, energy carriers, and amino acid biosynthesis (Ma et al., 2019). BAs, as signaling molecules, play a crucial role in maintaining cholesterol homeostasis and regulating glucose and lipid metabolism (Watanabe et al., 2006). Therefore, our findings suggested that BAs supplementation may have positive effects on glucose and lipid metabolism, intestinal function, and immunity. However, further studies are needed to confirm and fully understand the extent of these potential effects.

Conclusion

In conclusion, our results in the present study indicated that BAs supplementation tended to improve the feed efficiency, and significantly altered the BA profile in the serum and feces of growing-finished pigs, regardless of dramatic changes in the gut microbial composition. Metabolic pathway analysis revealed that the differently enriched metabolites were mainly involved in purine metabolism, ether lipid metabolism, glycerophospholipid metabolism, and amino sugar and nucleotide sugar metabolism, as well as primary bile acid biosynthesis. These metabolic pathways could potentially play a vital role in improving the feed efficiency of growing-finished pigs with BAs supplementation. Moreover, our findings suggested that BAs supplementation may have positive effects on glucose and lipid metabolism, intestinal function, and immunity. However, further studies are needed to confirm and fully understand the extent of these potential effects.

Glossary

Abbreviations

ADFI

average daily feed intake

ADG

average daily gain

BA

bile acid

BW

body weight

CA

cholic acid

CDCA

chenodeoxycholic acid

DCA

deoxycholic acid

DM

dry matter

FC

fold change

G:F

gain-to-feed ratio

GCA

glycocholic acid

GCBA

glycine-conjugated bile acid

GCDCA

glycochenodeoxycholic acid

GUDCA

glycoursodeoxycholic acid

HCA

hyocholic acid

HDCA

hyodeoxycholic acid

LysoPC

Lysophosphatidylcholine

OPLS-DA

supervised orthogonal partial least-squares discriminant analysis

PBA

primary bile acids

PCoA

principal coordinate analysis

PI

glycerophosphoinositols

PS

phosphatidylserine

SBA

secondary bile acids

TBA

total bile acids

TCA

taurocholic acid

TCBA

taurine-conjugated bile acid

TCDCA

taurochenodeoxycholic acid

TDCA

tauro-deoxycholic acid

THCA

taurohyocholic acid

THDCA

tauro-hyodeoxycholic acids

TLCA

tauro-lithocholic acid

TUDCA

tauroursodeoxycholic acid

OTUs

, operational taxonomic units

VIP

variable importance in the projection

Conflict of interest statement.

The authors declare no conflict of interest.