Effect of supplementation of inulin in dietary on lactation performance, rumen fermentation, ruminal microbial prole and metabolites in dairy cows

Inulin is a kind of fructo-oligosaccharide (FOS) derived mainly from Jerusalem artichoke and chicory tubers, which is also a soluble dietary ber. Inulin has become a scientically proven prebiotic product with satisfactory effects in improving the structure of intestinal ora, regulating blood lipid and glycemia, etc., in humans and monogastric. However, unlike monogastric animals, ruminal microbes are the largest microbiota in ruminants. The microora prole and metabolism activity in the rumen are closely related to the health of dairy cows. This study investigated the effects of inulin on rumen fermentation parameters, ruminal microbiome and metabolites, as well as lactation performance and serum indexes in dairy cows. A total of 16 Holstein dairy cows with similar body condition were randomly divided into two groups (n = 8 per group), with inulin addition at 0 and 200 g/d per cow, respectively. Experiment was lasted for 6 weeks including 1 week of adaptation period and 5 weeks of treatment period. At the end of the experimental period, the milk, serum and rumen uid were sampled and analyzed. The rumen microbiota and metabolites were analyzed via 16S rRNA sequencing and untargeted metabolomics, respectively. ROC, area

The results indicated the supplementation of inulin in dietary can increase the relative abundance of commensal microbiota and SCFAs-producing bacteria, meanwhile, upregulate amino acids (AAs) metabolism and downregulate lipid metabolism in rumen of dairy cows, which might further improve lactation performance and the level of serum lipids.

Background
Inulin is generally served as a reserve polysaccharide in plants. Statistically, it was contained in approximately 36,000 plants, of which Jerusalem artichoke (Helianthus tuberosus L.) tubers were the most abundant [1]. As a kind of fructo-oligosaccharide (FOS), inulin is mainly composed of β-Dfrutofuranose and glucopyranose residues connected by β-(1, 2) glucosidic bond [2]. There is a mild smell in inulin and with the characteristics of high sweetness and low calori c value. Although the inulin has high stability in water and tends to agglomerate when dispersed in water, [3] Ronkart et al. [4] reported that it could be slowly degraded into fructose and glucose in the solution with pH < 4 and the proper temperature for a period of time. Due to the existence of β-(1, 2) glycosidic bond structure, inulin cannot be digested and decomposed in the mouth, stomach and small intestine, and can only be partially fermented by Bi dobacteria in the colon. [2] Therefore, inulin is a scienti cally proved and edible nondigestible polysaccharide that meets the de nition of prebiotics. [2] As a soluble dietary ber, inulin has been widely studied in human and monogastric animals. Inulin can promote the growth of probiotics, such as Bi dobacteria and Lactobacilli, etc., and inhibit the proliferation of pathogenic and putrid bacteria, such as Clostridium pneumoniae, Enterococcus and Mould etc. [2], and meanwhile, the changes of intestinal ora caused by inulin fermentation can regulate the immune function of intestinal lymphoid tissue. [5] Chen et al. [6] found that inulin could delay the occurrence of immune diseases such as acute pancreatitis by regulating the homeostasis of the intestinal ora. On the other hand, inulin can induce the metabolic changes. It could promote the production of endogenous active substance cathelicidin related antimicrobial peptide (CRAMP). The effect depended on the production of short-chain fatty acids (SCFAs). In addition, a large number of SCFAs and lactic acid (LA) produced by inulin fermentation in the colon can not only stimulate intestinal peristalsis and prevent constipation, 2 but also lower the pH value of the intestine and facilitate mineral absorption. [7] Additionally, inulin has been shown to have the ability to inhibit the degradation of ingested fat by lipolytic enzymes, thereby lowering blood lipids. [8] Also, the lipid-lowering effect may also be related to the regulation of cholesterol and triglycerides (TG) in blood by SCFAs produced by inulin fermentation. [9] In a word, inulin can improve the intestinal environment and body health in human and monogastric animals by changing the microbial population and metabolic processes.
However, studies on inulin in ruminants, especially in dairy cows, were quite limited. Previous studies supported that inulin could not exert prebiotic role in ruminants because the existence of the rumen, in which inulin would be strongly degraded. [10,11] The fermentation process of inulin in the colon of monogastric animals is basically analogous to the process that occurs in the rumen of ruminants. [10] An in vitro rumen fermentation test in sheep showed that adding 1.5 g DM inulin to an incubation bottles containing 1 g basic diet could signi cantly reduce the concentration of ammoniacal nitrogen (NH 3 -N), which showed the potential of inulin to improve rumen nitrogen utilization. [12] Zhao et al. [11] investigated the effects of inulin on rumen fermentation and microbial growth in goats by rumen simulation technology. The results showed that inulin treatment decreased the concentration of acetate, the ratio of acetate to propionate (A/P) and methane production, but increased the concentration of butyrate. Furthermore, the abundance of the Fibrobacter succinogenes and Ruminococcus avefaciens were reduced, which indicated that inulin may inhibit the growth of partial rumen cellulose decomposition bacteria. Nevertheless, Umucalilar et al. [13] reported that inulin could only play a limited modi er role in regulating rumen fermentation parameters in vitro experiment. Another study on nishing beef showed that the supplementation of inulin in the diet diverted the rumen fermentation pattern from acetate to propionate and butyrate. Moreover, the α-diversity index of rumen bacteria was increased, in which the abundance of Bacteroides and Firmicutes were signi cantly increased, as well as weight and feed utilization [14]. Most of the previous studies on the effects of inulin on ruminants mainly focused on in vitro tests, which may restrict the knowledge and in-depth study of inulin's effects on the actual physiological regulation and production performance in ruminants.
Intestinal microbes are the most important microbial group in monogastric animals, and ruminant microbes have the same status in ruminants. The successful application of inulin in monogastric animals to improve the intestinal environment has aroused great interest in its potential roles on rumen function, and physiology and production performance on ruminants. Up to now, however, the effects of inulin on lactation performance, immune capacity, ruminal microorganism composition and metabolites activity in dairy cows have received few research attention. In this study, 16S rRNA sequencing and untargeted metabolomics technology were applied to investigate the effects of inulin on rumen microbial pro le and the metabolites compositions, which was helpful to evaluate the feasibility of using inulin as feed additive in dairy cows.

Ethics statement
The experimental animals, designs and operations in the present study were approved by the Animal Ethics Committee of the Chinese Academy of Agricultural Sciences (Beijing, China) (approval number: IAS-2019-9) and were in accordance with the recommendations of the academy's guidelines for animal research.

Inulin
The inulin (> 85% purity, other sugars including fructose, sucrose and glucose < 15%) added in this study was in powder form and extracted from Jerusalem artichoke (Helianthus tuberosus L.) tubers, which provided by Langfang Academy of Agriculture and Forestry (Hebei, China).

Experimental Animals, Diets and Design
The experiment was carried out on a well-managed and large-scale dairy farm in the suburbs of Beijing, China, which lasted 6 weeks composed by 1 week of adaptation and 5 week of treatments from mid-September to late October. A total of 16 Holstein dairy cows with similar initial parity (1.83 ± 0.91), body weight (BW) (553 ± 17.3 kg), days in milk (DIM) (166 ± 48.8 d), milk yield (33.7 ± 2.64 kg/d) and milk somatic cell counts (SCC) (338 ± 64.5 × 10 3 cells/mL) were randomly assigned to two groups, including control and inulin group (n=8). The cows in the two groups were fed 3 times a day at 0700, 1300, and 1900h, respectively, with same total mixed rations (TMR; concentrate-to-forage ratio of 40:60). Cows were fed TMR and drink freely. The TMR ingredients and nutrient components are listed in Additional le 1: Table S1. Based on the basal diet, the inulin addition dose were 0 and 200 g/d per cow in control and inulin group, respectively. The addition level of inulin in the current study was determined according to an independent dose screening test, which showed that 200 g/d per cow had a signi cant effect on rumen fermentation and lactation performance in dairy cows (the data to be published). Therefore, this addition dose was select in the present study. The inulin was offered through oral dosing after the morning feeding.

Milk Sample Collection and Body Weight Measurement
The cows were milking 3 times a day at 0600, 1200 and 1800h, respectively, through automatic milking systems (A milk, Israel). At the last 3 consecutive days in the end phase of experiment, the milk samples were collected, the milk yield and BW were recorded. The BW was measured once a day on a oor scale before morning feeding. The average milk yield and BW in these 3 days were used as the representatives for the experiment period. The milk samples from the 3 times in a day were mixed at the ratio of 4:3:3 [15]. A part of the mixed milk samples were added with 0.6 mg/mL potassium dichromate as a preservative and stored at 4 ℃ for the analysis of milk composition. The other part of mixed milk samples were stored at -20℃ for the analysis of milk fatty acids (FAs).

Milk Composition and Fatty Acids Analysis
Milk protein, fat, lactose, urea nitrogen (MUN) and SCC were measured at the Beijing Dairy Cattle Center (Beijing, China) by Milk composition analyzer (Lactoscan SP, Funke Gerber, Berlin, Germany) subsequently. The milk FAs was detected by acetyl chloride-methanolmethyl esteri cation method. The milk sample pretreatment referenced González-Córdova and Vallejo-Cordoba [16]. In addition, 4.5 mL of toluene was added to 0.5 mL FA triglycerides and mixed thoroughly as a standard measurement solution.
The milk FAs composition were analyzed with gas chromatograph (Agilent 8860 GC, California, USA). The details were as follows: Column, dicyanopropyl polysiloxane column (100 m × 0.25 mm , 0.20 µm); The temperature of the column oven was 140℃ for 5 min, then was increased to 240℃ at 4℃/min for 5 min; The injector and detector temperature were 260℃ and 280℃, respectively [16].

Blood Sampling and Analysis
Blood samples were collected through the caudal vein into procoagulant inert separation tubes (Saiao An Weiye Technology Development Co., Ltd., Beijing, China) at approximately 1 h before morning feeding on the last day of the experiment. . The blood samples were placed in a foam box with ice packs and centrifuged at 3000 × g for 15 min after brought back to the laboratory. The separated serum from each cow were collected and stored at -20 °C for the further analysis of biochemical indicators. The serum TG and total cholesterol (TC) concentration were analyzed by serum triglyceride determination kit (TR0100, Sigma-Aldrich, Missouri, USA) and cholesterol quantitation kit (MAK043, Sigma-Aldrich, Missouri, USA), respectively. The concentration of total protein (TP) was de ned through biuret speci c absorbance colorimetric method using total protein detection kit (Bai Olaibo technology Co., Ltd., Beijing, China). The serum albumin (ALB) concentration was detected by bromocresol green (BCG) method using BCG albumin assay kit (MAK124, Sigma-Aldrich, Missouri, USA). The concentration of serum globulin (GLO) was calculated by the difference between the concentration of TP and ALB. The blood urea nitrogen (BUN) concentration was determined by urea nitrogen kit (MS1812, Leigen Biotechnology Co., Ltd., Beijing, China).

Rumen Fluid Sampling
Rumen uid samples were collected on the last day of this experiment. At 1 h before morning feeding, a gastric tube type rumen uid sampler (MDW15, Colebo Equipment Co., Ltd., Wuhan, China) and a 200 mL syringe were used to collect rumen uid sample from each cow. The rst two tubes of rumen uid were discarded to avoid saliva contamination, and 100 mL rumen uid sample was collected from each cow [17]. The pH value of rumen uid samples were immediately measured after collection by a portable pH meter (Bell Analytical Instruments Co., Ltd., Liaoning, China). Each rumen uid sample was ltered through 4 layers of gauze and divided into six parts for the analysis of volatile fatty acids (VFAs), NH 3 -N, rumen urea nitrogen (RUN), LA, ruminal bacteria, and metabolites, respectively.
The enzyme-labeled instrument (Multiskan TM FC, Thermo Fisher, Beijing, China) was used to measure NH 3 -N concentration [18]. The concentration of RUN and LA were detected by urea nitrogen kit (MS1812, Leigen Biotechnology Co., Ltd., Beijing, China) and LA detection kit (TC0733, Leigen Biotechnology Co., Ltd., Beijing, China), respectively. The ruminal micro ora and metabolites were analyzed by 16S rRNA sequencing and untargeted metabolomics techniques.

Analysis of Sequencing Data
The Trimmomatic software (version v 0.36) was used for quality control (QC) of raw sequences, and FLASH software (version 1.2.11, https://ccb.jhu.edu/software/F LASH/index.shtml) was used for splicing: i) The bases with tail mass value below 20 and the reads less than 50 bp length after QC were ltered, respectively, and the reads including N bases were removed; ii) According to overlap relation between PE reads, pairs of reads were merged into a sequence, with minimum overlap length of 10 bp; iii) The overlap area of mosaics sequence were allowed the maximum error ratio of 0.2, the unmatched sequences were ltered out; iv) Samples were distinguished and the sequence direction was adjusted according to the barcode and primer at both ends of the sequences. The allowed mismatches of barcode and primer mismatch were 0 and 2, respectively. The UPARSE software (version 7.0.1090, http://drive5.com/uparse/) was used to perform the operational taxonomic units (OTU) sequence clustering, and select the sequences with more than 97% similar to the representative sequences of OTU, meanwhile, the chimeras were eliminated by UCHIME software (version 7.0, http://www. drive5.com/usearch/) and then the OTU table was generated. The ribosomal database project (RDP) classi er (http://rdp.cme.msu.edu/) was used to classify and annotate each sequence through comparing to Silva (Release132, http://www.arb-silva.de) with the comparison threshold of 70% [19].
The alpha diversity analysis at OTU level was conducted by Mothur software (version 1.30.2, https://www.mothur.org /wiki/Download_mothur). The difference test of the alpha diversity index (Sobs, ACE, Chao, Shannon and Simpson) in the two groups were conducted by Wilcoxon rank-sum test. The abundance at various taxonomic levels and beta diversity distance were calculated through Qiime software (version 1.9.1, http://qiime.org/install/index.html). In the beta diversity analysis, the principal coordinates analysis (PCoA) and non-metric multidimensional scaling (NMDS) were analyzed at OTU level with the distance algorithm of weighted normalized Unifrac. The differential bacteria were analyzed through linear discriminant analysis effect size (LEfSe) software (http://huttenhower.sph.harvard.edu/galaxy/root?Tool_id=lefse_Upload). Speci cally, non-parametric factorial Kruskal-Wallis (KW) sum-rank test was applied to detect the taxa with signi cantly differential abundance. The P-value from KW sum-rank test were adjusted by false discovery rate (FDR), with statistical signi cances declared at FDR-adjusted P < 0.05. And then, linear discriminant analysis (LDA) was used to estimate the in uence of abundance of each species on difference effect [20].

LC-MS Metabolomics Analysis
One hundred microliter thawed rumen uid sample from each cow was transferred into a 1.5 mL centrifuge tube, and 400 μL extraction solution (acetonitrile: methanol = 1: 1) was added and mixed thoroughly for 30s. After treated with low-temperature ultrasound for 30 min (5℃, 40 KHz), the solution was placed at 4 ℃ for 30 min and centrifuged at 13,000 × g for 15 min. The supernatant was removed and the precipitate was dried with nitrogen, and 120 µL reconstituted solution (acetonitrile: water = 1:1) was added to reconstitute the remains. The solution was treated with low temperature ultrasonic for 5min (5℃, 40 KHz) and centrifuged at 13,000 × g for 5 min again. The supernatant was pipetted into a sample injection vial with an inner cannula for LC-MS analysis [17]. The metabolites from all samples with equal volume were mixed to prepare a QC sample. Inserting a QC sample for every 10 samples to check the repeatability of the entire analysis process.
The LC-MS analysis was performed through the UPLC-Triple TOF system (Triple TOF5600, AB SCIEX, Massachusetts, USA). Ten microliter sample from each cow was detected through mass spectrometry after separated by BEH C18 chromatographic column (100 mm × 2.1 mm i.d., 1.8 µm) (Waters, Massachusetts, USA). The conditions of LC-MS were as follows: mobile phase A: water (containing 0.1% formic acid); mobile phase B: acetonitrile: isopropanol (1: 1) (containing 0.1% formic acid); Separation gradient of mobile phase (A: B): 80%: 20% from 0 to 3 min, 5%: 95% from 3 to 9 min and maintained for 2 min, 95%: 5% from 13.0 to 13.1 min and maintained for 3 min; The ow rate was 0.40 mL/min, and the column temperature was 40°C. The signal acquisition of mass spectrum was adopted the positive and negative ion scanning mode with the scanning range of 50-1000 m/z. The ion spray voltage in positive and negative ion mode were 5,000 V and 4,000 V, respectively. The ion source heating temperature was 500℃ with voltage of 20-60 V [21].

Metabolomics Data Processing and Analysis
The raw data was processed by Progenesis QI (Waters, Milford, USA) software for baseline ltering, peak identi cation, integration, retention time (RT) correction, peak alignment, and nally a data matrix including RT, mass-to-charge ratio (M/Z), and peak intensity was obtained. In order to reduce the error caused by sample preparation and instrument instability, the response intensity of the sample mass spectrum peak is normalized by the sum normalization method. Meanwhile, the variables in QC samples with the relative standard deviation (RSD) > 30% were deleted to obtain the nal data matrix for subsequent analysis. The MS spectrometry information was matched with the human metabolome database (HMDB) (http://www.hmdb.ca/) and Metlin database (https://metlin.scripps.edu/) to obtain metabolites information [21].
The preprocessed data was uploaded to the Majorbio Cloud Platform (https://cloud.majorbio.com) for further analysis. The ropls package of R program (Version1.6.2) was used to perform principal component analysis (PCA) and orthogonal least partial squares discriminant analysis (OPLS-DA), which accompanied by 7 cycles of interactive veri cation to evaluate the stability of the model. In addition, Student's t-test (unpaired) was used for the determination of signi cant differences. The fold change (FC) was used to evaluate the change trend (up-regulation or down-regulation) of differential metabolites. The selection of signi cantly differential metabolites was based on the variable important in projection (VIP) obtained from the OPLS-DA model and the P-value from Student's t-test, which adjusted by FDR. The metabolites with VIP > 1 and FDR-adjusted P-value < 0.05 were identi ed as the signi cantly different metabolites between the two groups. The diagram of Venn was printed by Venn Diagram in R packages (version 1.6.20). The hierarchical clustering analysis (HCA) for signi cantly different metabolites was performed by Scipy package in Python software (version 1.0.0). The metabolite distance algorithm was bray Curtis. The hierarchical clustering method of metabolites was average. The metabolic pathways in which the differential metabolites clustered were annotated through the KEGG database (https://www.kegg.jp/kegg/pathway.html). The Scipy package in Python software (version 1.0.0) was conducted for pathway enrichment analysis [21]. The receiver operator characteristic (ROC) analysis was used to examine the metabolites that were critical to the intergroup differentiation by ROC ANALYSIS of SPSS Statistics (version 22, IBM, New York, USA). The closer the area under curve (AUC) value to 1, the higher the accuracy of the prediction was.

Correlation Analysis
The correlation analysis were conducted through the OmicShare tools (http://www.omicshare.com/tools). The correlations between differential bacteria and metabolites, rumen fermentation parameters and milk compositions, as well as signi cantly differential metabolites and milk compositions, were calculated by Spearman's correlation coe cient, respectively. The range of the correlation coe cient (r) was from -1 to 1. The r > 0 and < 0 were represented positive correlation and negative correlation, respectively. The |r| value denoted the degree of correlation between variables. In particular, r = -1, 0 and 1 were meant a completely negative correlation, uncorrelated and a completely positive correlation, respectively. Correlation signi cance P-value below 0.05 and 0.01 were served as signi cant and extremely signi cant correlation, respectively.

Statistical Analysis
Data statistics was performed by SPSS Statistics (version 22, IBM, New York, USA). The data of body condition information in dairy cows (parity, BW, DIM, and SCC), milk yield and compositions, milk fatty acids compositions, rumen fermentation parameters, and alpha diversity indices were analyzed through one-way ANOVA and Student's t-test, with statistical significances declared at P < 0.05, a tendency was declared at 0.05 < P < 0.10.

Effect of Inulin Supplementation on Lactation Performance
The milk yield and compositions are showed in Table 1. With the supplementation of inulin, the dry matter intake (DMI) (P = 0.003), milk yield (P = 0.001), energy corrected milk (ECM), fat corrected milk (FCM) (P < 0.001), milk protein (P = 0.042) and lactose (P = 0.004) were signi cantly increased, while milk fat was showed a tendency of increase (P = 0.075). Whereas, the MUN (P = 0.023), SCC (P = 0.036) and the fat to protein ratio (F/P) was signi cantly decreased (P = 0.027), but the F/P was still in the normal rage (1.12-1.36).

Effect of Inulin Supplementation on Serum Indexes
As shown in Table 3, compared with control group, the concentration of TC (P = 0.008) and TG (P = 0.01) were signi cantly declined in the inulin group, while the level of TP, ALB, GLO and BUN had no signi cant difference (P > 0.05).

Effects of Inulin Supplementation on Rumen Fermentation Characteristics
The rumen fermentation parameters were listed in Table 4. The pH value in rumen was signi cantly declined (P = 0.040) with the addition of inulin, which accompanied by the signi cant increase of the concentration of acetate (P < 0.001), propionate (P = 0.003), butyrate (P < 0.001), isobutyrate (P = 0.002), valetate (P = 0.001), isovaletate (P < 0.001) and LA (P = 0.043). Meanwhile, the signi cant decrease of the concentration of NH 3 -N (P = 0.024) was also observed in inulin group.

Effect of Dietary Supplementation of Inulin on the Richness, Diversity and Composition of Ruminal Bacteria
A total of 1,349,120 effective 16S rRNA sequences were detected in 16 rumen uid samples and 2,337 OTUs were obtained by performing OTU clustering on non-repetitive sequences according to 97% similarity. Rarefaction curves (Additional le 1: Figure S1) showed that the current sequencing depth and sample size were su cient to assess the microbial diversity, total species richness and core species number of rumen uid samples. The α-diversity analysis revealed that the ACE (P = 0.031), Chao (P = 0.017) and Shannon (P = 0.026) indexes in inulin group were signi cantly increased, the Simpson index also showed a tendency of rise (P = 0.071), which illustrated that the addition of inulin increased the ruminal microbial community richness and diversity ( Table 5).

Effect of Dietary Supplementation of Inulin on Ruminal Metabolites
The rumen metabolites were analyzed through untargeted metabolomics techniques. The total ion chromatograms (TIC) plot of QC samples in positive and negative ion mode are shown in Additional le 1: Figure S4 A and B. The overlap of QC samples revealed the well repeatability and high-accuracy of the data. The unsupervised multivariate statistical analysis, PCA, generally re ected that a distinct difference existed in the ruminal metabolites between the two groups and a less degree of variation among samples within a group in positive and negative ion modes (Additional le 1: Figure S5 A and B). Further OPLS-DA provided a supervised discriminant analysis method, which could further distinguish the differences of ruminal metabolites between control and inulin groups and improve the effectiveness and analytical capabilities of the model (Figure 4). In OPLS-DA plots, R 2 Y and Q 2 were used to evaluate the modeling and prediction ability of OPLS-DA model respectively. The cumulative value of R 2 Y and Q 2 in positive (0.995 and 0.806) and negative (0.907 and 0.839) ion model were all above 0.80, which illustrated the stability and reliability of the model (Figure 4 A and C). Response permutation testing (RPT) was a randomized sequencing method to evaluate the accuracy of OPLS-DA models. As shown in Figure 4 B and D, the value of R 2 (0.953 and 0.931) and Q 2 (-0.0067 and -0.082 < 0) in positive and negative ion models revealed a well accuracy of OPLS-DA models.

Signi cantly Different Ruminal Metabolites between the Control and Inulin Groups
A total of 99 differential metabolites in rumen (64 in positive and 35 in negative ion models) between the control and inulin groups were detected with VIP > 1 and FDR-adjusted P < 0.05 (Additional le 1: Table  S4). Among them, lipids and lipid-like molecules, organic acids and derivatives, organic oxygen compounds and organoheterocyclic compounds were accounted for 37.0 ± 0.48%, 22.2 ± 0.32%, 16.7 ± 0.11% and 14.8 ± 0.24%, respectively (Additional le 1: Figure S6). The top 70 ruminal metabolites were selected for HCA analysis (Additional le 1: Figure S7). The differential metabolites between the two groups were divided into two cluster.  Figure S8 and Table S4).
Metabolic pathway enrichment analysis of differentially abundant metabolites KEGG pathway enrichment analysis showed that the supplementation of inulin mainly affected the lipid and amino acid metabolism, vitamin metabolism, biosynthesis of plant secondary metabolites and protein metabolism in the rumen of dairy cows (Table 6).
Correlation analysis among differential ruminal bacteria, metabolites, lactation and rumen fermentation performance.
Spearman correlation coe cient was used to calculate the correlation among several indicators ( Figure   6). Correlation analysis between signi cantly differential bacteria and milk components showed that the milk yield was positively associated with Muribaculaceae (r = 0.521, FDR-adjusted P = 0.042), Butyrivibrio  Figure 6B).
The correlation between signi cantly differential metabolites and microbacteria is showed in Figure 6C. As shown in Figure 6D, the relevance between signi cantly differential metabolites and milk compositions showed that, the milk SCC was positively associated with LysoPC (

Discussion
Even the encouraging results of effects of inulin on human and monogastric animal health especially the gut ora ecology have motivated researchers to explore its potentiality on ruminants like dairy cattle, beef, goat and sheep etc. [10][11][12][13][14], limited information was available for the effect of inulin in ruminants presently, and no consensus has been reached yet [22]. In the present study, the increase of milk yield after inulin addition in dietary might be attributed to the sweetness of inulin, 2 which could increase its palatability and DMI. Moreover, as a nonstructural carbohydrate, the intake of inulin signi cantly increased the concentration of VFAs in the rumen, which provided su cient energy for lactation [22]. In the current study, the increase in milk protein ratio might be due to the following two reasons: i) the inulin addition provided energy substrate, which reduced the amount of AAs used for energy supply, and thus increased microbial protein (MCP) synthesis in rumen, along with the decreased MUN and NH 3 -N concentration observed in this study, and the nitrogen utilization by rumen microorganisms was enhanced. Besides, the increase of MCP entering the small intestine promoted the synthesis of milk proteins in the mammary gland [23]; ii) the increase of propionate concentration in the rumen after the inulin supplementation could stimulate insulin secretion, which increased the absorption of AAs by the mammary glands, thereby increasing the milk protein percentage [22,23]. The increase of AAs levels were observed in the results of metabolomics in our study. The uptake of glucose by the mammary glands directly affects the milk lactose content. After inulin supplementation, the increased precursor of gluconeogenesis in rumen, such as propionate, LA and AAs, were absorbed into the blood through the rumen wall, which may promote more glucose from the blood were obtained by mammary gland [24].
Further analysis of milk FA composition showed that the proportion of SFA, especially SMCFA (C6:0, C8:0, C10:0 and C12:0), was signi cantly increased with the addition of inulin, but the ratio of C18:0 and C22:0 were decreased. The increase in SMCFA might due to the increase in the concentration of acetate and butyrate in the rumen, which are the substrates for de novo synthesis of milk FA by mammary epithelial cells [25], while about half of the C16:0 and other long-chain FA in milk fat are directly absorbed from blood lipids [25]. On the other hand, the results of metabolomics in our study revealed that the supplementation of inulin signi cantly downregulated the lipid metabolism in dairy cows. The increase in the ratio of SFA leaded to a decrease in the ratio of UFA [25]. The decrease in C18:1 trans-9 and C18:2 cis-6 might be related to the hydrogenation of rumen microorganisms, which could reduce the toxicity of UFA [26]. In addition, studies reported that milk fat percentage was negatively correlated with trans-C18:1 in milk [27]. Trans-FAs (TFAs) could restrain the expression of milk fat synthase gene, such as acetylcoenzyme A carboxylase and FA synthase [27].
In accordance with the improvement effect of inulin on intestinal ora in humans and monogastric animals, in the current study, inulin also showed a positive effect on ruminal microbiome through increasing the relative abundance and diversities of rumen micro ora. The possible reasons may due to the increase of feed intake and the optimized rumen microbiota structure induced by inulin [14,28]. In the current study, the supplementation of inulin signi cantly increased the relative abundance of Bacteroides, which could regulate nutrient absorption and metabolism of exogenous substances, such as polysaccharides [29]. In addition, inulin intake mainly increased the relative abundance of several probiotics and SCFA-producing bacteria in the rumen. Muribaculaceae (also called S24-7) belonged to Bacteroides has been found to act as a probiotic in multiple studies, or related with the innate immune system [30]. For example, homeostatic IgA responses could target Muribaculaceae resided in small intestine [30]. In addition, the relative abundance of the Muribaculaceae was improved with the increase of dietary ber levels [31]. Besides, Muribaculaceae has the function of degrading complex carbohydrates and is positively correlated with the concentration of intestinal acetate, propionate and butyrate [31]. Another interesting bacteria with signi cant change in abundance induced by inulin addition is Butyrivibrio. Butyrivibrio was the main butyrate-producing bacteria, followed by LA [32]. As a similar shortchain fatty acid producer, the main products of Prevotellaceae_NK3B31_group fermented FOS are acetate and a small amount of isobutyrate, isovalerate and LA [31]. Furthermore, Acetitomaculum was another well-known acetate and LA producing bacteria [33]. The increase of these bacteria abundance in rumen was in accordance with the elevated SCFA ratio in blood, which might further promote milk fat synthesis [34]. On the other hand, the increase of dietary ber level might promote the growth of a series of cellulolytic bacteria in the gastrointestinal tract. Eubacterium_hallii_group belonged to the Lacetospirillum family could also degrade fructose and cellobiose, which contributed to the formation of propionate and LA in rumen [35]. In addition, Eubacterium_hallii_group could convert glycerol to 3hydroxypropionaldehyde, and formed a multi-compound system called reuterin in aqueous solution, which has antibacterial properties [35]. Both Treponema and Saccharofermentans could degrade cellulose and hemicellulose to generate large amounts of propionate [36,37]. Therefore, the signi cant increase of the above bacteria may be the main reason of the increase of propionate concentration in rumen and milk lactose rate.
On the contrary, the signi cantly decreased ruminal micro ora induced by inulin were mainly several conditional pathogens, high-abundant bacteria after high-fat feeding and microorganisms involved in FA hydrogenation. Escherichia-Shigella, Anaerobiospirillum and Clostridia have been reported to cause gastrointestinal infection or in ammation [38,39,40]. Similarly, as a conditioned pathogens, Erysipelotrichaceae that resided in the mouth and intestine of animals could result in endogenous infections, and signi cantly increased in the intestinal tract of colon cancer patients. 41 The current study showed that the above bacteria were positively correlated with milk SCC and negatively correlated with milk fat, protein and lactose, which implicated the adverse effects of them on milk compositions. Moreover, Erysipelotrichaceae, Clostridia, Syntrophococcus and RF39 have been reported to increase signi cantly when feeding a high-fat diet. Studies have shown that the abundance of Erysipelotrichaceae in the intestines of rats with long-term high-fat or Western diets was 2.5 times higher than that of the control group [41]. Martinez et al. proved that there was a positive correlation between the level of Erysipelotrichaceae and host cholesterol metabolites [42]. The increase in the abundance of Clostridia in the intestines of rats with a high-fat diet might be related to low ber content [40]. Additionally, the study on growth requirements of Syntrophococcus in rumen uid showed that TG and phospholipids could promote the proliferation of Syntrophococcus [43]. RF39 belongs to the Tenericutes phylum, while a highfat diet could increase the abundance of Tenericutes, which made it easier for the host to absorb lipids [44]. On the other hand, Selenomonas and Bacteroidales_BS11_gut_group were reported to be involved in the hydrogenation of C18:3n3 and C18:2n6c, respectively. The probable reasons were that they provided energy substance, ATP, for the hydrogenation of these two FAs [45,46].
In summary, after supplementation of inulin in dietary, the decrease of rumen pH and the increase of VFAs concentration might be attributed to the proliferation of SCFAs-producing bacteria. On the other hand, the acidic environment in rumen effectively inhibited the proliferation of pathogens, which might be another reason for the decrease in milk SCC. Besides, more microorganisms increase the utilization of protein in the rumen and further increase the milk protein percentage, which might also explain the negative correlation between the dominant bacteria in the inulin group and MUN and NH 3 -N.
The result of metabolomics further revealed that supplementation of inulin signi cantly affected the amino acid and lipid metabolism in dairy cows. In the current study, the decrease in the level of LysoPC(18:1(9Z)), LysoPC(16:0) and LysoPC(18:2(9Z,12Z)) in the rumen of cows fed inulin downregulated the glycerophospholipid and choline metabolism pathways. Lysophosphatidylcholine (LysoPC) is derived from the hydrolysis of lecithin or the oxidation of very low density lipoprotein (VLDL). It was reported that inulin could reduce the number of VLDL particles in plasma and inhibit the expression of enzymes involved in lipid synthesis, such as acetyl-CoA carboxylase, fatty acid synthase and malic enzyme in the liver, which could reduce the de novo synthesis of FAs and the levels of TG and cholesterol in serum, [47] which was consist with our results. Another possible reason that the decrease in LysoPCs was that inulin could promote the proliferation of SCFA-producing bacteria and reduce the pH of the rumen, which led to the reduction of solubility of bile acids and increase in the excretion of bile acids and steroids in feces [48]. This might explain the negative correlation between LysoPC (18:1 (9Z)), LysoPC (16:0), LysoPC (18:2 (9Z, 12Z)), and Muribaculaceae, Prevotellaceae_NK3B31_group and other SCFAproducing bacteria. However, in our study, the decline of the level of LysoPCs were not showed a signi cant negative correlation with milk fat. The possible reason was that the LysoPCs was not a key factor affecting milk fat synthesis. On the other hand, LysoPCs have been reported to have proin ammatory activity, which was related to cytosolic phospholipase A2 mediating phosphatidylcholine (PCs) to produce arachidonic acid [49]. This might be the reason that the positive correlation between LysoPC(18:1(9Z)), LysoPC(16:0), LysoPC(18:2(9Z,12Z)) and milk SCC as well as several pathogens abundance, including Escherichia-Shigella and Anaerobiospirillu, etc., observed in this study. Other downregulated lipids included Phenylmethylglycidic ester and 8-Methylnonenoate.
Phenylmethylglycidic ester was a sweet, fatty, and oral compound [50]. However, its metabolic pathway was needed to be further explored. 8-methyl-nonenoic acid was a long-chain fatty acid derivative from leucine/valine pathway, and also a component of fatty acyl groups [51]. In the current study, the decrease of 8-methyl-nonenoic acid level was consistent with the decrease in the proportion of long-chain fatty acids in milk. Therefore, the declined levels the above ruminal metabolites might imply the potential of inulin to downregulate the lipid metabolism in dairy cows.
N-Acetylcadaverine is the acetylated form of the polyamine cadaverine, which is formed by the decarboxylation of lysine [52]. In addition, N-Acetylcadaverine was mainly a metabolite of Escherichia coli. Also, it has been found that there were variants of the N-Acetylcadaverine in Bacillus species [52]. The accumulation of a large amount of N-Acetylcadaverine in the body was related to the production of bacteriocins and toxin activity, which also had a speci c induction effect on in ammatory factors [52].
This might illustrate the positive correlation between N-Acetylcadaverine and milk SCC, Escherichia-Shigella, Erysipelotrichaceae (Bacillus class).
The AAs in the rumen were mainly derived from the degradation of feed proteins by rumen microorganisms. In the present study, the signi cant increase in the levels of L-Lysine, L-Proline, L-Phenylalanine and L-Tyrosine in cows fed inulin might be attributed to the proliferation of the bene cial symbiotic bacteria in rumen, which accelerated the decomposition of dietary protein and synthesis of MCP, and further promoted the synthesis of milk protein [17]. Consistent with Lykos et al., the increasing of nonstructural carbohydrates in the rumen could increase the level of non-essential AAs in the blood [53]. On the other hand, inulin regarded as a prebiotics could provide substrates for intestinal ora to generate SCFAs [54]. Davila et al. proposed that AAs could synthesize SCFAs by the prebiotic-activated intestinal ora [54]. Lysine, phenylalanine and tyrosine could be used as precursors of acetate and butyrate. Butyrate was derived from glutamic acid and lysine, and propionate was synthesized from alanine and threonine [54]. Meanwhile, isoacids, including isobutyrate, isovalerate and valerate, were generated by degradation of proline [55]. In the present study, the increase of uracil upregulated the nucleic acid metabolism, which might indicate that supplementation of inulin might promote the synthesis of rumen MCP [17]. Additionally, the decomposed product of uracil was β-alanine, which could further synthesize propionate [58]. This might explain the positive correlation between uracil and lactose. Deltonin and daidzein were reported to have anti-tumor activity and inhibit in ammatory factors in serum [59,60]. The negative correlation between deltonin, daidzein and milk SCC might reveal the enhancement of immune and disease resistance in dairy cows. However, the effect of inulin on the secondary metabolites of plants in the rumen still needs to be further explored.
Throughout the study, the effects of supplementation of inulin in dietary on lactation performance and rumen fermentation characteristics in dairy cows might be mainly through the in uence on the ruminal bacteria pro le, which further affected the kinds and concentrations of rumen metabolites. The correlation among them indicated that the shifts in composition and abundance of ruminal microbiome as well as the concentration and activity of rumen metabolites caused by the increase of dietary ber intake might be a key factor affecting performance and physiological indicators in dairy cows.

Conclusion
The supplementation of inulin in diets of dairy cows could increase the abundance of bene cial symbiotic bacteria and SCFAs-producing bacteria in the rumen, meanwhile, it promoted AAs synthesis and metabolism, and inhibited lipid metabolism. The shift of rumen microorganisms and metabolites further resulted in the increase in concentration of VFAs in the rumen, while decline the level of TC and TG in serum. The change of these precursor of milk composition might the reason that the increase in milk yield, protein, lactose rate and SFAs proportion in milk, but decrease in the percentage of PUFAs in milk.
However, the metabolic processes and mechanism of the action of inulin in the rumen of dairy cows merit further investigation.         Linear discriminant analysis (LDA) bar showed the impact of the abundance of each species on the difference between the two groups. P-value > 0. 05 and LDA score > 2.5 were de ned as signi cant difference.

Figure 1
The  Correlation analysis between (A) signi cantly differential bacteria and milk components, (B) signi cantly differential bacteria and rumen fermentation parameters, (C) signi cantly differential bacteria and differential metabolites, as well as (D) signi cantly differential metabolites and milk compositions. LA, lactic acid; SCC, somatic cell counts; MUN, milk urea nitrogen. Red indicates a positive correlation, while the blue indicates a negative correlation.  Receiver operating characteristic (ROC) curves to evaluate the differential metabolites that have key impact on the differentiation between the two groups. ROC curve re ects the relationship between sensitivity and speci city. The x-axis is speci city (false positive rate). The closer the x-axis is to zero, the higher the accuracy will be. The y-axis is sensitivity (true positive rate). The larger the y-axis is, the better the accuracy is. The Area under curve (AUC) is used to indicate the accuracy of prediction. AUC value