Dietary inulin modulates pork quality and systemic health via gut microbiome and metabolome changes in finishing pigs

Abstract

Background

Inulin is widely recognized for its ability to improve glucolipid metabolism and modulate the gut microbiome and metabolome. However, the potential to influence pork flavor development through gut environment changes in animal husbandry remains unexplored. This study investigated the relationships among systemic health, meat flavor, gut microbiome, and metabolome in pigs fed a diet supplemented with inulin. Thirty-six male Duroc × Landrace × Yorkshire pigs (75.0 ± 1.5 kg) were divided into 2 groups, and fed either a regular diet (CON group) or a diet containing 0.5% inulin (INU group) for 60 d.

Results

Inulin supplementation did not adversely affect production or slaughter performance (P > 0.05) but enhanced systemic health by improving serum biochemistry indicators (P < 0.05). Descriptive analysis of pork composition revealed numerically higher levels of key fatty acids (C16:0, C18:0, C18:1) and umami-related amino acids (e.g., glutamic acid) in the INU group compared to the CON group. Inulin supplementation also modulated the gut microbiome, reducing alpha-diversity and increase specific microbes including Lactobacillus, Lachnospiraceae_NK4A136_group, Romboutsia, Family_Xlll_AD3011_group, Roseburia, and Turicibacter. Furthermore, inulin supplementation significantly altered metabolic pathways, down-regulating arginine biosynthesis, linoleic acid metabolism, riboflavin metabolism, and alanine, aspartate, and glutamate metabolism. These microbial and metabolic changes strongly correlated with the observed alterations in pork quality.

Conclusion

Dietary inulin supplementation represents a promising strategy to modulate pork quality and systemic health without compromising productive performance. However, due to the exploratory nature of the amino acid and fatty acid analyses with limited sample size, further studies are needed to further confirm these findings.

Supplementary Information

The online version contains supplementary material available at 10.1186/s42523-026-00566-5.

Introduction

The increasing demand for healthier diets has heightened consumer interest in premium and flavorful meat. While rapid fattening techniques have successfully shortened pig growth cycles and boosted pork production, they have also led to declines in pork quality, including reduced sensory attributes, nutritional value, intramuscular fat content, and flavor [1, 2].

To address these challenges, animal husbandry has employed a diverse range of traditional medicinal herbs and naturally plant- derived products as nutrual solutions for enhancing the quality and flavor of animal products [3–9]. Inulin, a fructan linked by β-glycosidic bonds [10], is one such promising additive. It resists digestion in the stomach and small intestine but undergoes microbial fermentation in the gut [11, 12]. Extracted primarily from Jerusalem artichoke and chicory roots, inulin can be categorized by chain length into short, medium, and long chains, each with distinct health benefits [13]. The fractional precipitation method can be used to separate inulin of different chain lengths, which are divided into natural inulin (the degree of polymerization (DP) between 2 and 60), short-chain inulin (average DP ≤ 10), long-chain inulin (average DP ≥ 23), and those with a DP between 10 and 23 are typically referred to as medium-chain inulin [14]. Medium-chain inulin, for example, has demonstrated potential in mitigating high-fat diet-induced metabolic disorders, hepatic steatosis, and chronic inflammation in mice [15]. As a prebiotic, inulin enhances antioxidative capacity and improves pork juiciness without compromising other sensory attributes [2]. Based on our previous research, we identified a medium-chain inulin with a DP of 12 that effectively regulates glucolipid metabolism and modulates the gut microbiome (data not shown, under review).

The production of short-chain fatty acids (SCFAs) via degradation by the intestinal microbiota represents one of the pathways through which inulin exerts its biological effects [16]. As signaling molecules, alterations in SCFAs levels are closely correlated with key pathways in the host, including linoleic acid metabolism, histidine metabolism, fatty acid degradation, and glutamate metabolism [17]. Moreover, it has been reported that these alterations contribute to the enhancement of meat flavor [18]. Additionally, the enhancement of meat flavor can be directly accomplished through the interaction of the gut-muscle axis, primarily due to the fact that the intestinal microbiota can modulate the host’s amino acid and fatty acid metabolic pathways [19]. Thus, we hypothesized that the incorporating this inulin into the diets of finishing pigs could modify the intestinal microbiome and metabolic processes, thereby enhancing both pork quality and flavor.

In a previous statistical study, the median of inulin in pig farming was 0.1-2% [20]. Supplementing weaned piglets’ diets with 2.5 and 5 g/kg of inulin is advantageous for intestinal development, reduces inflammation, and improves intestinal permeability [21]. Furthermore, 0.5% inulin diet can improve the growth performance and carcass traits of growing barrows [22]. However, higher doses may pose a potential risk of flatulence and osmotic diarrhoea in the host [23]. Thus, the study aims to explore the effects of an appropriate amount of dietary inulin supplementation on the physical phenotypes, pork quality and flavor, gut microbiome, and metabolome in finishing pigs.

Results

Production performance and slaughter traits

Dietary inulin supplementation did not apparently affect body weight (BW) over the 60-day experimental period (Table 1). However, significant differences were observed in slaughter traits between the two groups, with the INU group showing reduction in hoof, tail, liver indices, along with leather thickness, and an increased eye muscle area (EMA) index (P < 0.05).

Table 1.

Effects of inulin on production and slaughter performance of finishing pigs

Item1	Treatments2		SEM	P-value
	CON	INU
Initial body weight, kg	74.93	75.00	0.263	0.906
Final body weight, kg	120.83	120.33	0.593	0.694
Dressing percentage, %	71.18	70.91	0.483	0.797
Head index, %	5.22	4.68	0.113	0.008
Hooves index, %	2.23	2.02	0.039	0.002
Tail index, %	0.14	0.11	0.005	0.034
Liver index, %	1.40	1.12	0.049	< 0.001
Leather thick, mm	3.39	2.92	0.118	0.038
Back fat thick, mm	39.63	42.09	0.810	0.133
EMA, mm2	38.50	50.63	2.299	0.002

¹EMA, eye muscle area

²CON, control group; INU, the same diet supplemented with 0.5% inulin

Each group contains six samples for statistical analysis (n = 6)

Serum biochemistry

Inulin supplementation significantly improved serum biochemical indices as shown in Table 2. Levels of ALT, AST, UREA, GLU, TG, and TCHO were significantly lower in the INU group compared to the CON group (P < 0.05), with respective reduction proportions of 26.95%, 19.07%, 16.62%, 21.71%, 12.50%, and 31.36%. On the other hand, the HDL/LDL ratio increased in the INU group compared to the CON group (P < 0.01), primarily attributed to increased HDL concentrations (P < 0.01) and decreased LDL concentrations (P < 0.01). No significant differences were observed in total protein (TP) and albumin (ALB) levels.

Table 2.

Effects of inulin on serum biochemistry of finishing pigs

Item1	Treatments		SEM	P-value
	CON	INU
ALT, U/L	46.33	33.83	2.091	< 0.001
AST, U/L	39.33	31.83	1.515	0.005
TP, mmol/L	74.32	74.23	0.873	0.965
ALB, mmol/L	30.55	29.85	0.863	0.705
UREA, mmol/L	7.70	6.42	0.220	< 0.001
GLU, mmol/L	4.56	3.57	0.158	< 0.001
TG, mmol/L	0.32	0.28	0.007	< 0.001
TCHO, mmol/L	2.36	1.62	0.117	< 0.001
HDL, mmol/L	0.62	0.77	0.025	< 0.001
LDL, mmol/L	1.07	0.79	0.049	< 0.001
HDL/LDL	0.59	0.96	0.063	< 0.001

¹ALT, alanine aminotransferase; AST, aspartatea aminotransferase; TP, total protein; ALB, albumin; GLU, glucose; TG, triglycerides; TCHO, total cholesterol; HDL, high-density lipoprotein; LDL, low-density lipoprotein

Each group contains six samples for statistical analysis (n = 6)

Pork quality analysis

As shown in Table 3, inulin supplementation had no effect on dry matter, or ash content. Tenderness-related traits showed substantial improvements, including reductions in drip loss (29.66%, P < 0.001), cooking loss (6.05%, P < 0.05), and shear force (33.1%, P < 0.01). Flesh color was enhanced with decrease lightness (L*) and yellowness (b*) (P < 0.01) and increase redness (a*) (P < 0.05). Besides, dietary inulin promoted the contents of crude protein (P < 0.05) and ether extract (P < 0.05) in pork.

Table 3.

Effects of inulin on pork quality traits of finishing pigs

Item1		Treatments		SEM	P-value
		CON	INU
DM, %		93.00	93.29	0.183	0.451
CP, %		81.96	82.48	0.122	0.024
EE, %		1.81	2.17	0.082	0.015
Ash, %		4.13	4.10	0.058	0.654
Drip loss, %		6.98	4.91	0.365	< 0.001
Cooking loss, %		29.58	27.79	0.403	0.017
Shear force, N		116.39	77.82	6.187	< 0.001
Longissimus pH 45 min		6.01	6.02	0.006	0.402
Chromaticity	Lightness, L*	38.26	35.10	0.564	0.001
	Redness, a*	7.57	8.75	0.256	0.012
	Yellowness, b*	7.47	6.19	0.233	0.001

¹ DM, dry matter; CP, crude protein; EE, ether extract

Each group contains six samples for statistical analysis (n = 6)

Fatty acid profile of pork

Fatty acids with concentrations greater than 0.01 µg/g were presented in Table 4 (Ranges & Median). Descriptive analysis indicated that pork from the INU group showed numerically higher concentrations of saturated fatty acids (SFAs), monounsaturated fatty acids (MUFAs), and polyunsaturated fatty acids (PUFAs) compared to the CON group, as reflected by the median values (Table 4). Due to the limited sample size (n = 3 per group), statistical comparisons were not performed.

Table 4.

Effects of inulin on pork free fatty acids of finishing pigs¹

Item2	Ranges of Values		Median
	CON	INU	CON	INU
C6:0, µg/g	0.048–0.049	0.048–0.049	0.049	0.048
C8:0, µg/g	0.057–0.058	0.058–0.058	0.058	0.058
C10:0, µg/g	0.055–0.060	0.059–0.064	0.056	0.064
C11:0, µg/g	0.058–0.060	0.060–0.066	0.058	0.064
C12:0, µg/g	0.056–0.063	0.063–0.071	0.060	0.068
C13:0, µg/g	0.056–0.061	0.061–0.070	0.059	0.064
C14:0, µg/g	0.069–0.088	0.116–0.139	0.076	0.138
C14:1, µg/g	0.062–0.070	0.070–0.078	0.068	0.078
C15:0, µg/g	0.050–0.057	0.056–0.066	0.055	0.060
C15:1, µg/g	0.060–0.064	0.065–0.072	0.062	0.072
C16:0, µg/g	0.923–1.192	1.661-2.00	1.057	1.892
C16:1, µg/g	0.101–0.140	0.305–0.312	0.120	0.308
C17:0, µg/g	0.060–0.072	0.068–0.079	0.064	0.072
C17:1, µg/g	0.054–0.066	0.062–0.073	0.058	0.066
C18:0, µg/g	0.543–0.687	0.911–1.039	0.615	1.024
C18:1n9t, µg/g	0.051–0.061	0.063–0.074	0.055	0.067
C18:1n9c, µg/g	1.020–1.533	2.964–3.037	1.092	3.000
C18:2n6t, µg/g	0.060–0.065	0.066–0.074	0.063	0.069
C18:2n6c, µg/g	0.456–0.497	0.625–0.720	0.476	0.663
C20:0, µg/g	0.053–0.064	0.065–0.078	0.060	0.071
C18:3n6, µg/g	0.063–0.068	0.067–0.076	0.065	0.073
C20:1, µg/g	0.062–0.087	0.091–0.113	0.072	0.104
C18:3n3, µg/g	0.053–0.068	0.068–0.080	0.060	0.075
C21:0, µg/g	0.044–0.050	0.051–0.061	0.048	0.054
C20:2, µg/g	0.067–0.079	0.083–0.097	0.073	0.090
C22:0, µg/g	0.054–0.060	0.061–0.069	0.058	0.065
C20:3n6, µg/g	0.071–0.077	0.079–0.087	0.073	0.087
C22:1n9, µg/g	0.177–0.216	0.224–0.245	0.191	0.238
C20:3n3, µg/g	0.053–0.058	0.062–0.073	0.056	0.066
C23:0, µg/g	0.051–0.055	0.057–0.068	0.054	0.060
C20:4n6, µg/g	0.156–0.169	0.216–0.229	0.160	0.216
C22:2, µg/g	0.060–0.063	0.066–0.075	0.062	0.069
C24:0, µg/g	0.061–0.065	0.067–0.075	0.063	0.069
C20:5n3, µg/g	0.056–0.060	0.062–0.070	0.059	0.065
C24:1, µg/g	0.069–0.072	0.075–0.083	0.070	0.077
C22:6n3, µg/g	0.098–0.101	0.103–0.109	0.099	0.105
SFA, µg/g	2.239–2.707	3.462–3.990	2.522	3.929
UFA, µg/g	2.952–3.310	5.399–5.590	3.003	5.508
MUFA, µg/g	1.762–2.172	3.973–4.046	1.807	3.985
PUFA, µg/g	1.138–1.196	1.426–1.544	1.190	1.523
PUFA/SFA	0.440–0.508	0.387–0.412	0.474	0.388
n-3/n-6	0.318–0.340	0.267–0.297	0.321	0.277

¹Each group comprised three individual pig samples (n = 3 per treatment). Data are presented as the range (minimum–maximum) and median. No statistical comparisons were performed due to the limited sample size

² SFA, saturated fatty acid; UFA, unsaturated fatty acid; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid; n-3, sum of the C18:3n3, C20:3n3, C20:5n3 and C22:6n3; n-6, sum of the C18:2n6t, C18:2n6c, C18:3n6, C20:3n6, C20:4n6

Briefly, the primary saturated fatty acids were palmitic acid (C16:0), stearic acid (C18:0), and myristic acid (C14:0). Monounsaturated fatty acids included oleic acid (C18:1), palmitoleic acid (C16:1), erucic acid (C22:1), and eicosenoic acid (C20:1). Polyunsaturated fatty acids included linoleic acid (C18:2), arachidonic acid (C20:4), eicosapentaenoic acid (C20:5n3), and docosahexaenoic acid (C22:6n3). However, the PUFA/SFA and n-3/n-6 ratios were relatively lower compared to the CON group.

The amino acids profile of the pork

The amino acid composition of longissimus dorsi muscle from pigs fed control (CON) or inulin-supplemented (INU) diets is presented in Table 5. Descriptive analysis revealed that the INU group exhibited numerically higher median values for total amino acids (TAAs) (1592.67 µg/g vs. 1447.49 µg/g in CON), representing a 10.03% increase. Umami and sweet amino acids (UAAs & SAAs) also showed a higher median in the INU group (1434.74 µg/g vs. 1293.45 µg/g in CON), corresponding to a 10.92% increase. Notably, glutamic acid (a key umami-contributing amino acid) displayed a markedly higher median in the INU group (22.16 µg/g) compared to the CON group (14.09 µg/g). Other amino acids with increased median values in the INU group included arginine (21.46 vs. 17.37 µg/g), lysine (14.14 vs. 9.84 µg/g), valine (21.18 vs. 18.02 µg/g), and isoleucine (14.26 vs. 12.59 µg/g). In contrast, several amino acids showed lower median values in the INU group, including serine (15.67 vs. 17.08 µg/g), threonine (18.62 vs. 21.36 µg/g), alanine (73.06 vs. 80.31 µg/g), methionine (6.98 vs. 8.73 µg/g), and phenylalanine (13.27 vs. 18.54 µg/g).

Table 5.

Effects of inulin on pork amino acids of finishing pigs¹

Item2	Ranges of Values		Median
	CON	INU	CON	INU
Histidine, µg/g	10.71–13.33	11.87–16.19	12.26	14.03
4-Hydroxy-L-Proline, µg/g	5.22–5.78	4.13–6.14	5.39	5.11
Arginine, µg/g	15.41–18.61	20.54–22.39	17.37	21.46
Asparagine, µg/g	11.34–11.60	9.20-10.66	11.47	9.39
Glutamine, µg/g	1002.31-1088.61	1150.12-1231.54	1045.46	1190.83
Serine, µg/g	16.34–17.81	14.86–15.90	17.08	15.67
Glycine, µg/g	70.49–75.39	72.93–74.69	75.31	74.09
Aspartic acid, µg/g	3.35–4.15	3.58–3.80	3.49	3.63
Glutamic acid, µg/g	13.91–14.28	20.76–23.56	14.09	22.16
Threonine, µg/g	20.24–22.39	18.42–19.88	21.36	18.62
Alanine, µg/g	80.22–80.40	70.65–75.47	80.31	73.06
γ-Aminobutyric acid, µg/g	0.22–0.30	0.17–0.21	0.26	0.19
Proline, µg/g	23.57–25.04	21.22–22.45	24.29	22.34
(R)-2-Aminobutyric acid, µg/g	0.92–1.28	1.37–2.47	1.23	1.75
Lysine, µg/g	9.36–10.31	13.63–14.58	9.84	14.14
Methionine, µg/g	8.67–8.74	6.77–7.54	8.73	6.98
Tyrosine, µg/g	19.01–19.66	16.06–17.72	19.57	16.66
Valine, µg/g	17.81–18.55	19.28–21.93	18.02	21.18
Isoleucine, µg/g	12.55–13.34	12.89–14.36	12.59	14.26
Leucine, µg/g	18.94–20.60	20.07–22.89	19.72	21.11
Phenylalanine, µg/g	17.69–19.74	12.11–13.67	18.54	13.27
Tryptophan, µg/g	6.44–7.46	7.51–9.95	7.38	8.70
TAAs, µg/g	1398.70-1479.67	1539.18-1635.42	1447.49	1592.67
EAAs, µg/g	114.78-117.45	116.31-120.14	116.78	117.31
UAAs & SAAs, µg/g	1248.73-1332.11	1384.14-1470.59	1293.45	1434.74

¹Each group comprised three individual pig samples (n = 3 per treatment). Data are presented as the range (minimum–maximum) and median. No statistical comparisons were performed due to the limited sample size

² TAAs, total amino acids; EAAs, essential amino acids; UAAs & SAAs, umami amino acids (asparagine, glutamine, aspartic acid and glutamic acid) and sweet amino acids (serine, glycine, threonine, alanine and proline)

These descriptive trends suggest that inulin supplementation may modulate the amino acid profile of pork, particularly by increasing umami-related amino acids, although these findings should be interpreted with caution due to the exploratory nature of the analysis and the limited sample size.

Effects of INU treatment on cecum microbiome

Gut microbiome composition

Based on 97% sequence similarity, a total of 3,846 Operational Taxonomic Unit (OTUs) were identified in the INU group, which were then assigned to 39 phyla, 92 classes, 185 orders, 267 families, and 466 genera. In the CON group, 4,118 OTUs were obtained and were clustered into 39 phyla, 98 classes, 181 orders, 271 families, and 481 genera. There were 2,847 OTUs that were common across the two experimental groups (Fig. 1A). Inulin supplementation significantly reduced alpha-diversity in the gut microbiome, as indicated by lower Shannon, Simpson, Chao1, and observed species indices in the INU group compared to the CON group (P < 0.01, Table 6).

Fig. 1.

Effects of diets with inulin-supplementation on the profile and relative abundance of the gut microbial community at the phylum, family, and genus levels in pigs. (A) The mutual and specific OTUs in two different diet groups, (B) Beta-diversity, principal coordinate analysis (PCoA) plot. (C) The ratio of Firmicutes to Bacteroidetes. (D-F) The relative abundances at the phylum, family, and genus levels of the gut microbiota of pigs fed either CON or INU diet. * P < 0.05, ** P < 0.01

Table 6.

Alpha-diversity of the cecum microbiota of pigs feeding inulin

Item		Treatments		SEM	P-value
		CON	INU
Coverage percentage, %		> 99	> 99
Richness Estimators	Observed species	1075.83	935.83	27.264	0.003
	Chao1	1174.87	1068.72	19.109	0.001
	ACE	1185.72	1073.22	22.576	0005
Alpha-diversity Indexes	PD_whole_tree	89.56	90.86	5.043	0.905
	Shannon	7.57	6.79	0.136	< 0.001
	Simpson	0.99	0.97	0.003	< 0.001

Each group contains six samples for statistical analysis (n = 6)

Beta-diversity analysis, using PCoA, revealed distinct microbiome community structures between the two groups (Fig. 1B). The INU group exhibited an increased relative abundance of Firmicutes and Proteobacteria and a reduced abundance of Bacteroidetes, resulting in a significantly higher Firmicutes/Bacteroidetes (F/B) ratio (P < 0.05, Fig. 1C). The relative abundance of microbes exceeding 0.5% were displayed at the phylum, family, and genus levels (Fig. 1D-F). Concretely, within the Firmicutes phylum, the dominant families included in both groups were Lachnospiraceae, Oscillospiraceae, Clostridiaceae, and Peptostreptococcaceae, while in phyla of Bacteroidetes were Prevotellaceae, Rikenellaceae, Muribaculaceae, and p-251-o5.

Gut microbiome biomarker

At the genus level, inulin supplementation enriched beneficial microbes, including Lactobacillus, Lachnospiraceae_NK4A136_group, Roseburia, Turicibacter, Streptococcus, Terrisporobacter, Romboutsia, and UCG-005, while decreasing potentially harmful genera such as Treponema, Rikenellaceae_RC9_gut_group, and Methanobrevibacter (Fig. 2A). Linear discriminant analysis effect size (LefSe, with LDA > 4.0) further identified significant biomarkers differentiating the two dietary groups (Fig. 2B). Additionally, the cladograms illustrated the phylogenetic distribution of discrepant bacteria, as shown in Fig. 2C. By integrating these two logical analyses, we identified and selected 15 microbial genera for further investigation.

Fig. 2.

Demonstration of differential microbes in the gut of inulin-fed pigs. (A) The differential microbes among groups at the genus levels. Only microbes that had a relative abundance of more than 0.5% were compared, (B) Histogram of the linear discriminant analysis (LDA) value distribution of different bacteria among groups (LDA score > 4.0), (C) Cladogram constructed to visualize the microbial community relative abundance data among the gut samples from the groups. Different color nodes indicate different groups, and the species classifications at the phylum, class, order, family, and genus levels are shown from the inside to the outside. Differences were declared to be statistically significant when * P < 0.05, ** P < 0.01

Gut metabolome

Composition and biomarker

Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis of gut metabolome identified 854 metabolites (582 positive and 272 negative ions) in the treatment groups. To compare the distribution of cecum metabolites, orthogonal partial least squares discriminant analysis (OPLS-DA) was conducted. This revealed a distinct clustering pattern between the CON and INU groups, suggesting that inulin supplementation significantly altered gut metabolites. As shown in Fig. 3A, the OPLS-DA model (R²X = 0.735, R²Y = 1, and Q² = 0.996) score plot revealed that the first principal component explained 70.2% of the features between the groups, while the second principal component explained 3.29%.

Fig. 3.

Differential metabolites and their classification in the gut of inulin-fed pigs. (A) The orthogonal projections to latent structures discriminant analysis (OPLS-DA) presenting the distribution and separation of the gut metabolites in the different treatments, (B) Volcano plot displaying the discrepant metabolites between the CON group and the INU group. Taking variable importance in the projection (VIP > 1.0), fold change (FC > 2.0 or FC < 0.5, INU vs. CON) and P-value (P < 0.05) of metabolites, 379 differentiated metabolites were identified. Of them, 71 were observed to be up-regulated, indicated by red dots, and 308 were downregulated, displayed by green dots, (C) Heatmap used to show differential metabolite levels in individual samples. (D) Significantly modulated metabolic pathways following the analysis of differentiated metabolites between two groups. (E) Ascertained classifications of discrepant metabolites (P < 0.05) that were up-regulated or downregulated in the INU group than those in the CON group

The variable importance in projection (VIP) value of the OPLS-DA model was used to measure the contribution of metabolites to distinguishing characteristics between groups. Univariate statistical analysis was combined to calculate the P-value and fold change (FC) between the INU and CON groups, leading to the identification of 379 differentiated metabolites with 71 upregulated and 308 downregulated in the INU group (Fig. 3C, VIP > 1.0, FC > 2.0 and FC < 0.5, P < 0.05). In addition, the heatmap shows that the samples from the CON and INU groups cluster well, indicating a strong correlation with the additive treatment (Fig. 3C).

The KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway enrichment analysis of differential metabolites revealed that the highlighted metabolic pathways primarily involved amino and fatty acid metabolism (Fig. 3D). Based on their impact, the top five differential metabolic pathways were identified as linoleic acid metabolism, arginine biosynthesis, riboflavin metabolism, alanine, aspartate, and glutamate metabolism, and taurine and hypotaurine metabolism (the profile of metabolic pathways modulated by inulin supplementation can be referred to in Table 7). Among them, the taurine and hypotaurine metabolism pathway was upregulated in the inulin dietary supplementation, while the other four pathways were downregulated. Classification analysis of significantly modulated metabolites within the top five pathways revealed that those regulated by inulin primarily included lipids and lipid-like molecules, organic acids and derivatives, as well as organoheterocyclic compounds. Thirteen primary differential metabolites and their respective categories are detailed in Fig. 3E.

Table 7.

Profile of metabolic pathways modulated by inulin supplementation

Names of pathway	Total	Hits	Raw P	− log 10 (P)	Impact
Linoleic acid metabolism	5	2	0.02	1.65	1.00
Arginine biosynthesis	14	8	< 0.01	7.19	0.52
Riboflavin metabolism	4	1	0.19	0.73	0.50
Alanine, aspartate and glutamate metabolism	28	4	0.05	1.32	0.49
Taurine and hypotaurine metabolism	8	2	0.06	1.25	0.43

Pathways correlation

The interconnection of metabolic pathways suggested inulin supplementation reshaped the gut metabolome, promoting beneficial changes linked to pork flavor and host health (Fig. 4).

Fig. 4.

Correlations among the metabolic pathways significantly modified by inulin supplementation based on the differential metabolites and KEGG pathways. The metabolites linked with solid lines indicate a direct conversion, while those connected with dashed lines mean an indirect conversion through a biochemical process. The metabolites (black fonts) with light-red background indicated an increase in concentration, whereas the light-blue indicated a decrease in concentration due to the dietary inulin supplementation. The metabolic pathways (white fonts) marked with deep-red background was the up-regulated metabolic pathways, and those marked with deep-blue were the downregulated pathways. The yellow rectangles represented the physiological and/or biochemical processes affected by the inulin-modulated metabolic pathways

Correlation among the phenotypes, differentiated microbes and metabolites

Spearman correlation analysis revealed strong association among health parameters, gut microbiota, metabolites, and pork quality traits. Beneficial genera enriched in the INU group, such as Escherichia-Shigella, Treponema, UCG-005, Streptococcus, Rikenellaceae_RC9_gut_group, Terrisporobacter, Lactobacillus, Romboutsia, Lachnospiraceae_NK4A136_group, Ruminococcus, Methanobrevibacter, NK4A214_group, Roseburia, Family_XIII_AD3011_group, Turicibacter, were positively correlated with improved phenotypes, including higher levels of HDL, crude protein (CP), ether extract (EE) and redness (a*) (P < 0.05). These genera were also associated with reduced levels of ALT, AST, glucose, triglycerides (TG), total cholesterol (TCHO), and shear force, indicating enhanced systemic health and meat tenderness (Fig. 5B).

Fig. 5.

The cause-and-effect process of dietary affecting the phenotypes of pigs fed CON or INU diet. Correlations among the differential phenotypes, microbes and metabolites altered by inulin supplementation. The red font indicated the microbes or metabolites and their positive correlation parameters significantly up-regulated by inulin supplementation. (A) The dietary treatment specially impacted the gut microbes, (B) Correlations between the health parameters and differ enterally microbes, (C) Associations between health parameters and screening metabolites, (D) Correlations between the biomarker microbes and metabolites

In addition to their correlations with systemic health and pork quality, these beneficial microbes positively influenced the levels of taurine, taurocholic acid in the gut, and fatty acids like SFAs and MUFAs in meat, which serve as flavor precursors in pork. In contrast, genera with more abundant in the CON group, such as NK4A214_group, Methanobrevibacter, Treponema, Rikenellaceae_RC9_gut_group, and Ruminococcus, were associated with less favorable phenotypes including higher levels of drip loss, LDL and yellowness (b*) (Fig. 5B).

Correlation analysis also identified 13 key metabolites significantly linked to both host phenotypes and pork flavor traits. Inulin-modulated metabolic pathways, particularly those involving amino acids like alanine, aspartate, and glutamate, as well as fatty acids, were strongly connected to these observed improvements (Fig. 5C&D). For example, higher levels of umami-related amino acids such as glutamic acid, and beneficial fatty acids in pork from the INU group were closely linked to the reshaped gut microbiota and metabolome. The integration of microbiome, metabolome, and phenotype data highlights the pivotal role of inulin in optimizing systemic health and pork quality through gut microbiome modulation.

Discussion

This study demonstrated that dietary inulin supplementation improved systemic health, pork quality, and flavor while having no adverse effects on production performance in finishing pigs. These findings align with previous studies showing that polysaccharides in pig diets do not directly enhance production performance but can significantly improve the quality of animal by-products [7, 24]. The observed reduction in by-products, such as liver, tail, head and hooves indices, also suggest potential economic benefits from inulin supplementation.

Inulin’s role as a non-digestible carbohydrate likely contributed to the observed improvements in systemic health by balancing carbohydrate and protein metabolism. Reduced fasting blood glucose levels and improved serum indicators, such as lower ALT, AST, and TG levels, reflect enhanced hepatic function and metabolic stability [25, 26]. The increase in HDL/LDL ratio further supports the hypothesis that inulin promotes systemic health by regulating lipid metabolism, as seen in previous studies [27]. Our findings indicated that a diet supplemented with inulin balanced the metabolism of carbohydrates and protein, maintaining energy metabolism and homeostasis, thus contributing to enhanced systemic health in the host.

The enhancements in pork quality, including reduced drip loss, cooking loss, and shear force, suggest an improvement in water-holding capacity and tenderness. Increased crude protein and fatty acid content also highlight the nutritional benefits of inulin supplementation. These improvements can be attributed to the modulation of oxidative stress and increase levels of essential nutrients in the meat, consistent with earlier findings [2, 28]. Fatty acids play a crucial role in meat flavor development, as their degradation during cooking produces volatile compounds such as aldehydes, ketones, and alcohols that contribute to the characteristic aroma of pork [29, 30]. In the present study, descriptive analysis of fatty acid profiles (Table 4) revealed numerically higher levels of various fatty acids in the INU group compared to the CON group, including key saturated fatty acids (e.g., palmitic acid C16:0, stearic acid C18:0), monounsaturated fatty acids (e.g., oleic acid C18:1), and polyunsaturated fatty acids (e.g., linoleic acid C18:2). These observations suggest that inulin supplementation may promote fatty acid deposition in muscle tissue, potentially contributing to enhanced flavor development during cooking. However, it is important to interpret these findings with caution due to the limited sample size for fatty acid analysis (n = 3 per group). Additionally, while the numerically higher levels of unsaturated fatty acids (UFAs) observed in the INU group may benefit flavor, they were accompanied by numerically lower PUFA/SFA and n-3/n-6 ratios compared to the CON group. These trends may have implications for the nutritional quality and oxidative stability of pork, as lower n-3/n-6 ratios have been associated with reduced health benefits [31, 32], and higher UFA content may potentially shorten meat shelf life [33]. These descriptive trends indicate that inulin supplementation may modulate the fatty acid composition of pork, but further studies with larger sample sizes are needed to confirm these observations and to explore the balance between flavor enhancement and nutritional quality in pork products.

The observed increase in umami-related amino acids, such as glutamic acid, in the INU group suggests a potential improvement in pork flavor, although these findings should be interpreted with caution due to the limited sample size for amino acid analysis (n = 3 per group). Descriptive analysis revealed numerically higher levels of umami-related amino acids, particularly glutamic acid, in the INU group compared to the CON group. Glutamic acid is widely recognized as a key contributor to the umami taste of meat [34, 35]. Serine and alanine primarily influence the meat’s sweetness [36], while arginine generates pyrazine compounds during ripening [37]. Valine produces caramel via the Maillard reaction when combined with glucose [38], and phenylalanine degradation leads to benzaldehyde, imparting a nutty flavor [39]. These descriptive trends indicate that inulin supplementation may modulate the amino acid composition of pork, potentially contributing to improved flavor profiles. However, further studies with larger sample sizes are warranted to confirm these observations and to elucidate the underlying mechanisms linking dietary inulin, gut microbiota modulation, and amino acid metabolism in muscle tissue. Additionally, inulin’s ability to increase the abundance of beneficial microbes, such as Lactobacillus and Lachnospiraceae_NK4A136_group, likely supported the production of flavor precursors. These microbes are known for their roles in gut health, nutrient metabolism, and reducing inflammation [40, 41].

Analogous to our finding, the addition of dietary fiber to the feed of livestock and poultry leads to a decrease in α-diversity [8, 9]. This phenomenon may be characterized by a specific increase in the abundance of key/beneficial intestinal bacteria and a reduction in the abundance of less dominant miscellaneous bacteria. The INU group exhibited a reduction in alpha-diversity and an increase in the abundance of some beneficial genera, such as Turicibacter and Romboutsia. These microbial changes were associated with improved systemic health, as well as increased deposition of fatty acids and amino acids, which enhance pork flavor [42]. The observed increase in the F/B ratio aligns with findings from a recent study [43], further supporting the role of inulin supplementation in modulating gut microbiota (i.e., Lactobacillus and Bifidobacterium) to enhance nutrient absorption and energy metabolism. The genus Escherichia-Shigella is almost entirely composed of Escherichia coli (the species level, refer to Supplementary Materials) which is a group of bacteria that naturally constitutes a part of the intestinal flora of warm-blooded animals, including humans. The majority of Escherichia coli strains are non-pathogenic and play a crucial role in the normal functioning of a healthy intestine [13], the inulin enhanced its abundance, establishing Escherichia coli as a keystone species in the intestinal tract of a healthy host. Additionally, Lu et al. reported increased intramuscular fat content and enhanced redness (a*) in pork, signifying improved meat color and quality. In contrast, harmful genera such as Treponema and Rikenellaceae_RC9_gut_group were more prevalent in the CON group, correlating with unfavorable phenotypes and metabolic outcomes [44–47]. These results support the hypothesis that inulin supplementation reduces the prevalence of potentially harmful microbes, thereby conferring protection against pathogen-associated issues. Furthermore, the increase abundance of UCG-005, a member of the Oscillospiraceae family is essential for host fatness [42], in the INU group suggests the inulin may have the potential to promote production performance, consistent with the rise in the F/B ratio.

As a non-protein amino acid, taurine reduces oxidative stress and supports lipid metabolism by forming taurocholate with bile acids, promoting fat digestion and lowering serum triglycerides and cholesterol levels [48–50]. Riboflavin (vitamin B₂) also plays a key role in protein and lipid metabolism [51]. In our study, inulin supplementation reduced protein and lipid degradation by modulating related metabolic pathways.

The alanine, aspartate and glutamate metabolism pathway, central to amino acid metabolism and pork flavor [52], is linked to arginine synthesis initiated by glutamate [53]. Additionally, linoleic acid, precursor for conjugated linoleic acid, enhanced pork flavor by increasing fatty acid content [54, 55]. Collectively, our findings demonstrate that inulin can effectively inhibit the breakdown of flavor-enhancing amino acids and preserve umami compounds, as evidenced by the high glutamate content and increased fatty acid levels in pork. These results confirm the down-regulation of fatty acid catabolism pathways observed in the omics analysis.

Conclusions

In conclusion, dietary supplementation with 0.5% inulin in finishing pigs modulated the gut microbiome (decreased alpha-diversity, increased beneficial genera) and downregulated metabolic pathways related to amino acid and fatty acid degradation. These changes were associated with improved serum biochemical indicators, enhanced meat quality traits (reduced drip loss and shear force, increased redness), and numerically higher levels of umami-related amino acids and fatty acids in pork. Notably, these improvements were achieved without compromising growth performance.

Collectively, these findings suggest that inulin supplementation is a promising strategy for enhancing pork quality and systemic health in finishing pigs. However, due to the limited sample size for amino acid and fatty acid analyses (n = 3 per group), further studies with larger sample sizes are warranted to validate these observations and elucidate underlying mechanisms.

Methods

Experimental design, animals, and feeding management

The middle-and-late fattening period is a crucial stage for the deposition of flavor substance in pork. Considering the target weight of lean-meat-type commercial pigs and the experimental period, thirty-six healthy Duroc × Landrace × Yorkshire (DLY) pigs with an average body weight of 75.0 ± 1.5 kg were randomly allocated into six pens (4 m × 8 m), with six pigs per pen. The pens were equally divided between two dietary treatment groups: a control group (CON) fed a standard diet and an inulin-supplemented group (INU) fed a diet containing 5 g/kg inulin. Each treatment was replicated three times. Dietary formulation (Table 8) was based on the China national standard outlined in GB/T 39235 − 2020 [56] to satisfy the nutrient requirements of fatting pigs, and the crude protein (the method of 954.01), crude fat (920.39), ash (942.05), Ca (927.02) and TP (965.17) content in diet were all analysed according to the AOAC [57]. The dietary digestible energy (DE) and metabolizable energy (ME) was calculated according to the following formula: DE (ME) = corn × DE1 (ME₁) + wheat middlings × DE2 (ME₂₎ + wheat bran × DE3 (ME₃₎ + … + minerals × DE14 (ME₁₄₎. All values from DE₁ (ME₁) to DE₁₄ (ME₁₄) were based on nutritional value listed on the GB/T 39235 − 2020 [56]. The inulin was provided by Gansu Lircon Biological Co., Ltd. The molar mass of inulin is 1,800 g/mol, and mean degree of polymerization (DP) is 12. Inulin was first combined with a premix that was subsequently mixed with other ingredients and then stored in covered containers. All diets were prepared in a single batch and stored in a cool warehouse. The pigs were provided ad libitum access to fresh water and fed three times daily (6:00 a.m., 12:00 p.m., and 6:00 p.m.). Feed intake was calculated based on the quantity provided and leftovers. After a 7-day acclimation period, the pigs were fed their respective diets for 60 days.

Table 8.

Basic diet composition and nutritional level

Items	Content
Ingredients
Corn, %	60.50
Wheat middlings, %	10.00
Wheat bran, %	10.00
Soybean meal, %	9.00
Defatted rice bran, %	3.00
Peanut meal, %	4.00
CaHPO4·2 H2O, %	0.34
NaCl, %	0.45
Soybean oil, %	1.00
Sulfate lysine (70%), %	0.30
DL-methionine (99%), %	0.01
Limestone, %	1.00
Vitamins1, %	0.20
Minerals2, %	0.20
Total, %	100.00
Nutrients content 3, 4
DE, MJ/kg	14.08
ME, MJ/kg	13.46
CP, %	14.43
EE, %	3.71
Lys, %	0.84
Lys/ME, g/MJ	0.62
Ca, %	0.57
TP, %	0.47
Met, %	0.25
Thr, %	0.51
Cys, %	0.29

¹The Vitamins provided the following per kg DM of diets: Vitamin A 8000 IU, Vitamin D₃ 3500 IU, Vitamin E 30 IU, Vitamin K₃ 1.0 mg, Vitamin B₁ 1.5 mg, Vitamin B₂ 5.5 mg, Vitamin B₆ 2.0 mg, Vitamin B₁₂ 0.02 mg, Pantothenic acid 7.50 mg, Niacin 26.5 mg, Biotin 0.07 mg, Folic acid 0.6 mg;

² The Minerals provided the following per kg DM of diets: Mn (as manganese sulfate) 40 mg, Fe (as ferrous sulfate) 70 mg, Zn (as zinc sulfate) 80 mg, Cu (as copper sulfate) 25 mg, I (as potassium iodide) 0.2 mg, Se (as sodium selenite) 0.4 mg;

³ The content of CP, EE, Ca, and TP were measured values (n = 6), methionine (Met), lysine (Lys), Threonine (Thr), and Cystine (Cys) were the actual addition. The value of dietary DE and ME was calculated based on the DE and ME values of each ingredient in the formula

⁴ DE, digestible energy; ME, Metabolizable energy; CP, Crude protein; EE, Ether extract; Ca, Calcium; TP, Total phosphorus

Sample collection

At the end of trial, 12 pigs (two randomly selected per pen, six per treatment group) were chosen for sample collection. Blood samples were collected from fasted pigs on the day before slaughter after 12 h fasting, serum samples were collected via centrifuging the samples, and determined using an automatic biochemical analyzer (7072, Hitachi, Japan). Immediately following blood collection, the pigs were provided with feed. Subsequently, the pigs were fasted again for 12 h prior to slaughter, which were euthanized using electrical stunning following guidelines from the Animal Ethics Committee of Shandong Agricultural University. Longissimus dorsi muscle samples were collected to access meat quality and nutrient composition. Gut content was sampled from the junction of the cecum and colon using sterile surgical procedures. Samples were immediately flash-frozen in liquid nitrogen and stored for further microbiome and metabolome analysis.

Meat quality assessment

Muscle samples were evaluated for physical traits, including the nutrient content, drip loss, cooking loss and shear force. The nutritional values of the pork also determined by the methods mentioned in the AOAC [57], which the moisture according to the method of 950.46, while the 928.08 for analysing the crude protein, 960.39 for the ether extract, and the ash refer to the 920.153. Drip loss was determined as a percentage of weight loss for a sample (30 ± 3 g) after hanging at 4 °C for 24 h. Similarly, cooking loss was calculated as the weight difference of muscle (50 ± 5 g) before and after being water-bathed in a sealed bag at 80 °C for 60 min. The cooked samples were then trimmed into rectangular chunks (30 mm × 10 mm × 10 mm) parallel to the muscle fibers. The chunks were subsequently sliced perpendicular to the myofiber orientation using a tenderness meter (C-LM38, NEAU, China) to determine the pork’s shear force.

Briefly, the sarcolemma and attached fat were meticulously removed from the longissimus dorsi muscle, and a freeze - drying process was conducted. Specifically, the ether extract of 1 gram of freeze-dried pork was determined using a Soxhlet extractor with petroleum ether for 5 h. Concurrently, the crude protein content of the samples (0.4 g) was analyzed using a protein analyzer (K9860, Hanon, China), and the ash content of 2 g sample was measured by combusting it at temperatures ranging from 550 to 600 °C in a Muffle furnace. pH values were measured at 45 min post-mortem (PHBJ-260, INESA, China), and color parameters (L* (lightness), a* (redness), and b* (yellowness)) were accessed using a colorimeter (CR 300, Minolta, Japan) [58].

Fatty acids and amino acids analysis

The profiles of free fatty acids and amino acids in pork samples were analyzed by Shanghai Sanshu Biotechnology Co., Ltd in line with the methods we reported [59]. Briefly, the free fatty acids in meat samples were extracted and derivatized in two steps: (1) Homogenization with chloroform-methanol–NaCl, centrifugation to collect the chloroform layer; (2) Re-extraction of the aqueous phase with dichloromethane, combination and nitrogen-drying of organic layers, methylation with sulfuric acid–methanol at 80 °C for 2 h, followed by hexane–water partitioning and isolation of the upper organic phase. The resulting fatty acid methyl esters were dissolved in iso-octane for analysis. Free amino acids were extracted from meat samples by utilizing 0.5 mL of 0.5 M HCl at room temperature. The extracts were filtered via a 0.22 μm membrane. Each sample (10 µL) was combined with 70 µL borate buffer and 20 µL AccQ•Tag reagent in phosphorus vials. The mixture was heated and subsequently cooled prior to analysis. Fatty acids were identified using gas chromatography-mass spectrometry (GC-MS, Agilent Technologies Inc. CA, USA), and amino acids were determined using ultra-high-performance liquid chromatography (UPLC, (Vanquish, Thermo, USA)) coupled with high-resolution mass spectrometry (Q Exactive, Thermo, USA).

The GC system utilized Agilent 6890 (Agilent Technologies, USA), with an INU-Sil 88(100 m×0.25 mm×0.25 μm) column and an injection volume of 1 µL at a split ratio of 10:1 and nitrogen flow rate of 1.0 mL/min. The initial temperature in the column chamber was set at 100 ℃ for a duration of 5 min, followed by a heating ramp to reach 240 ℃ at a rate of 4 ℃/min. For MS detection, Agilent 5977 (Agilent Technologies, USA) equipped with electron impact ion source (EI) source and MassHunter workstations were used. The MS conditions were as follows: the injection port temperature was maintained at 260 ℃ while the Quadrupole temperature remained constant at150 ℃; full SCAN mode was selected for detection with a Mass (m/z) scan range from30-550.

The UPLC system was equipped with a Waters BEH C18 column (50 × 2.1 mm, 1.7 μm). Ultra-pure water containing 0.1% formic acid was used as mobile phase A, while acetonitrile containing 0.1% formic acid served as mobile phase B. The flow rate was 0.5 mL/min, the temperature was set at 55 °C, and the injection volume was 1 µL. The MS employed electrospray ionization (ESI) with the following settings: sheath gas, 40 arb; assisted gas, 10 arb; ion spray voltage, + 3000 V; temperature, 350 °C; capillary temperature, 320 °C. The scanning mode was set to FullScan using the positive ion method.

A total of three pigs per treatment group were randomly selected for free fatty acid and amino acid profiling. Each sample originated from an individual animal, and no sample pooling was conducted. Data were presented descriptively as range and median without statistical hypothesis testing due to the exploratory nature of the analysis and the limited number of replicates.

Microbiome and metabolome analysis

Gut microbiome composition was analyzed through 16 S rRNA sequencing, and metabolome profiles were determined using liquid chromatography-tandem mass spectrometry (LC-MS/MS), as previously described [60]. Microbial alpha- and beta-diversity were evaluated, and key metabolic pathways were identified using KEGG pathway enrichment analysis.

Statistical analysis

Individual pigs were used for phenotypes, pork traits data collection, microbiome analysis, and metabolome analysis (n = 6). Statistical differences were assessed using IBM SPSS 26.0 (SPSS Inc., Chicago, IL, USA) to perform the independent two-sample t-tests, the statistical analysis model used as follows:

where Y_i is the outcome variable, X_i is a binary indicator (0 = Group A, 1 = Group B), β₀ represents the mean of Group A, β₁ is the mean difference between Group B and Group A are independent normally distributed errors. The null hypothesis H₀: β₁ = 0 was tested against the alternative H₁: β₁ ≠ 0. The P-value < 0.05 was considered to have statistical significance. For fatty acid and amino acid datasets (n = 3 per group), only descriptive statistics (Range and Median) were reported. No inferential statistical tests were applied to these variables due to the limited sample size, and thus no P-values are presented in the corresponding tables (Tables 4 and 5).

Gut microbiome data were analyzed and plotted using Prism 8.1 (GraphPad, LaJolla, CA, USA) and http://www.ehbio.com/Cloud_Platform/. Metabolome data were analysed using MetaboAnalyst 6.0, and heatmaps were generated using https://cloud.metware.cn/.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

DLY

Duroc × Landrace × Yorkshire pigs

CON

Control group

INU

Inulin group

DP

the degree of polymerization

BW

Body weight

EMA

Eye muscle area

ALT

Alanine aminotransferase

AST

Aspartatea aminotransferase

TP

Total protein

ALB

Albumin

GLU

Glucose

TG

Triglycerides

TCHO

Total cholesterol

HDL

High-density lipoprotein

LDL

Low-density lipoprotein

DM

Dry matter

CP

Crude protein

EE

Ether extract

SFA

Saturated fatty acid

UFA

Unsaturated fatty acid

MUFA

Monounsaturated fatty acid

PUFA

Polyunsaturated fatty acid

TAAs

Total amino acids

EAAs

Essential amino acids

UAAs

Umami amino acids

SAAs

Sweet amino acids

OTUs

Operational taxonomic units

PCoA

Principal coordinate analysis

LefSe

Linear discriminant analysis effect size

LDA

Linear discriminant analysis

LC-MS/MS

Liquid chromatography-tandem mass spectrometry

OPLS-DA

Orthogonal partial least squares discriminant analysis

VIP

Variable importance in projection

FC

Fold change

KEGG

Kyoto encyclopedia of genes and genomes

AOAC

Association of official analytical chemists

DE

Digestible energy

ME

Metabolizable energy

GC-MS

Gas chromatography-mass spectrometry

UPLC

Ultra-high-performance liquid chromatography

EI

Electron impact ion source

Author contributions

Conceptualization, Yunkyoung Lee and Guiguo Zhang; Methodology, Yunpeng Wang and Kayeon Ko; Investigation, Yunpeng Wang and Eunyoung Kim; Formal Analysis and Writing - Original Draft, Yunpeng Wang; Writing - Review & Editing, Yunkyoung Lee, Miroslava Kačániová, and Guiguo Zhang; Funding Acquisition and Supervision, Yunkyoung Lee and Guiguo Zhang; All authors read and approved the final manuscript.

Funding

This work was supported by the National Key R&D Program of China-Korea cooperative project (2019YFE0107700, NRF-2019K1A3A1A20081146), the National Research Foundation Grant of Korea (2020R1A2C2004144, RS-2024-00334577), the key project for foreign experts of Shandong Province (WRS2023075), the Forage Industrial Innovation Team Project (SDAIT-23-05), and the Key R&D Program of Shandong Province (2022TZXD0018).

Data availability

The Raw 16 S rRNA sequences datasets described in this study have been deposited in the NCBI database (https://www.ncbi.nlm.nih.gov: Accession number PRJNA971324). The metabolome data reported in this paper have been deposited in the OMIX, China National Center for Bioinformation / Beijing Institute of Genomics, Chinese Academy of Sciences at https://ngdc.cncb.ac.cn/omix: Accession number OMIX 010210. All the other original data are not publicly available due to privacy or ethical restrictions, but will be made available upon request from zhanggg@sdau.edu.cn.

Declarations

Ethical approval

The current experiment adhered to animal welfare guidelines and experimental protocols established by the Research Ethics Committee of Shandong Agricultural University. Approval was granted by the Laboratory Animal Management Committee of Shandong Agricultural University (Protocol No. S20240136).

Competing interests

The authors declare no competing interests.