Increasing the Hindgut Carbohydrate/Protein Ratio by Cecal Infusion of Corn Starch or Casein Hydrolysate Drives Gut Microbiota-Related Bile Acid Metabolism To Stimulate Colonic Barrier Function

Ethical approval.The experimental protocol and procedures for the care and treatment of the pigs were approved by the Animal Care and Use Committee of Nanjing Agricultural University (approval number SYXK 2018-0071).

Animals and experimental procedures.A total of 24 male 42-day-old cross-bred Duroc × (Landrace × Large White) weaned piglets, weighing 12.08 ± 0.28 kg, were individually housed in stainless steel metabolism cages (height, 0.85 m; length, 0.70 m; width, 0.70 m) with a nipple drinker and feeder and temperature controlled at 25°C. After a 7-day adaptation, pigs were fasted for 12 h prior to fitting a simple T-cannula in the cecum, approximately 5 cm cranial to the ileocecal sphincter. A comprehensive description of the surgical procedure and postoperative care has been given previously (20). Before surgery and during a 14-day recuperation period, the pigs were fed a 23% CP commercial starter diet. All pigs were fed individually, and water was provided ad libitum. After the 14-day recuperation, the pigs were allotted to experimental treatments according to a completely randomized design. The total duration of the experiment was 19 days, during which the pigs were infused 3 times per day (at 8:00, 12:00, and 20:00, respectively) with 10 ml sterile saline (control), 10 ml corn starch (Xinmao Group, Zhucheng, China) (50 g/day) suspended in sterile saline (starch group), or 10 ml casein hydrolysate (Qingdao Hope Bio-Technology Co. Ltd., Qingdao, China) (50 g/day) suspended in sterile saline (casein group). The ingredients and chemical composition of the basal diets are presented in Table S1 in the supplemental material. The diets were formulated to meet or exceed the nutrient demand according to the National Research Council (NRC) (50) standards for pigs.

Sample collection.The timeline of the experiment can be seen in Fig. S4. At day 19 of the experiment, the pigs were fasted for 12 h and then weighed, and blood samples (10 ml) were collected from the jugular vein prior to slaughter. Blood samples were centrifuged at 3,000 × g for 15 min at 4°C to separate the serum for chemical analysis. Within 5 min after slaughter, the colon was carefully excised and fastened in length and then weighed both with and without contents. Then, segments of the distal colon tissue were collected and washed three times in ice-cold phosphate-buffered saline. Mucosal samples were then scraped from the underlying tissue using a sterile glass slide, immediately transferred into liquid N, and then stored at −80°C. Representative samples of colonic digesta were collected, and the first part of each sample was diluted with twice the volume of distilled water, mixed, and immediately centrifuged at 2,000 × g for 10 min. The supernatants were stored at −80°C for analysis of microbial metabolites. The second part of the digesta samples were placed in a 65°C oven for 48 h and then shattered and stored at −20°C for nitrogen substrate detection. In addition, pH was measured by placing the pH probe within the residual proximal colonic digesta using a portable pH meter.

Cell culture.Human colorectal adenocarcinoma epithelial cells (Caco-2) were cultivated in Dulbecco modified Eagle medium (DMEM)–F-12 supplemented with 10% (vol/vol) fetal bovine serum (FBS) (Gibco, Grand Island, NY, USA), penicillin (100 U/ml), and streptomycin (100 mg/ml) at 37°C and in a 95% air-5% CO₂ atmosphere. The effects of bile acids (CA, DCA, and LCA) on gene expression for epithelial barrier function were tested with dosages (25, 100, and 250 μM) based on a previous study (41). Caco-2 cells were seeded onto collagen-coated 9-mm petri dishes at a density of 5.95 × 10⁴ cells per cm² and maintained at 37°C, 5% CO₂, for 10 days (i.e., for the same incubation time as cells used for paracellular permeability). Culture medium was changed every 48 h. For inhibitor studies, cells were pretreated with inhibitors for 2 h, and the concentrations of inhibitors were as follows according to the previous reports: 250 nM EGFR inhibitor AG-1478 (43), 10 μM Src inhibitor PP-2 (MedChem Express, USA) (23). Cells were plated at a density of 1.0 × 10⁴ cells/well in 96-well plates and treated with inhibitor for 24 h. Dimethyl sulfoxide (DMSO)-methanol at equimolar concentrations to AG-1478 and PP-2 concentrations served as vehicle control.

Cell viability.Cell viability was determined using the CCK-8 assay (Nanjing Jiancheng Institute of Bioengineering, Nanjing, China) according to the manufacturer’s protocol. Briefly, Caco-2 cells were plated in 96-well culture plates at a density of 1.0 × 10⁴ cells/well 24 h before treatment. Then, the cells were treated with 0, 25, 100, and 250 μM BAs (CA, DCA, and LCA) in serum-free DMEM for about 2 h. Each cell viability value was normalized to the control and then calculated as previously described (44).

Transepithelial electrical resistance (TEER) measurements.To assess paracellular permeability, Caco-2 cells were seeded onto the apical surface of rat-tail collagen (Corning, NY, USA)-coated polyethylene terephthalate cell culture inserts (pore size 0.4 μm, surface area 1.12 cm²), which were kept in 12-well companion plates (Becton Dickinson, Falcon labware; Oxford, United Kingdom). Caco-2 cells were seeded at 2.5 × 10⁵ cells in each insert (i.e., 5.95 × 10⁴ cells per cm²), and standard growth medium (1.5 ml) was added to each well. Two inserts were cell free with medium only to account for resistance due to the insert membrane. Medium was changed in all inserts every 48 h. Transepithelial electrical resistance (TEER) value was monitored by a Millicell ERS meter (Millipore, Bedford, MA, USA).

Colonic mucosal bacterial composition analysis.Total genomic DNA from mucosal scrapings was extracted from 0.3 g of sample using a bead-beating and phenol-chloroform extraction method as described previously (45). The V3-V4 regions of the bacterial 16S rRNA gene were amplified using a universal forward primer (5′-ACTCCTRCGGGAGGCAGCAG-3′) and a reverse primer (5′-GGACTACCVGGGTATCTAAT-3′). Amplicons were purified using an AxyPrep DNA gel extraction kit according to the manufacturer’s instructions (Axygen Biosciences, Union City, CA, USA) and performed on the Illumina MiSeq high-throughput sequencing platform. Total bacterial and functional microbial genes including baiJ and bsh involved in conversion of primary to secondary BAs were quantified by real-time quantitative PCR (qPCR) using group-specific primers (Table S2) and SYBR green premix (TaKaRa Biotechnology, Dalian, China) in the Step One Plus qPCR system (Life Technologies Inc., CA, USA) as previously described (46).

Raw sequence data generated from MiSeq sequencing were processed in the Mothur version 1.36.1 software (47). Sequences with lengths of <200 bp, having one or more ambiguous bases, or containing a homopolymer length of ≥8 bp were deleted. Then, sequences were normalized to contain an equal number of sequences in all groups before the sequences were presumptively identified and aligned to the high-quality 16S rRNA bacterial sequence database derived from the bacterial Silva database (Silva version 108). These sequences were used for phylotype analysis at the 97% operational taxonomic unit (OTU) level. Representative sequences from each OTU were taxonomically classified with a confidence level of 90% using the Ribosomal Database Project (RDP) classifier. Bacterial diversity including rarefaction analysis, Shannon and inverse Simpson diversity indices, and ACE and Chao richness estimators was assessed using Mothur (47). Principal-coordinate analysis (PCoA) was conducted based on an unweighted UniFrac distance. An unweighted distance-based analysis of molecular variance (AMOVA) was conducted to assess significant differences among samples using the Mothur program (47).

Colonic bile acids and nutrient substrate measurements.The concentrations of CA, CDCA, DCA, and LCA were determined in three replicates for each digesta sample using enzyme-linked immunosorbent assay kits (12) (R&D Systems, Shanghai Enzyme-linked Biotechnology CO., Ltd.) according to the manufacturer’s instructions. Samples of colonic contents were analyzed for dry matter (DM), crude protein (CP) (N × 6.25), Ash, and ether extract (EE) according to AOAC (51) procedures. The content of carbohydrate was calculated as 100 − [(100 − DM) (%) + CP (%) + Ash (%) + EE (%)]. The content of nitrogenous compound was calculated as CP (%)/6.25.

Serum d-lactate and DAO measurements.The concentrations of d-lactate and diamine oxidase (DAO) in serum were measured by corresponding assay kits (Nanjing Jiancheng Institute of Bioengineering, Nanjing, China) according to the manufacturer’s instructions.

RT-qPCR.Total RNA was extracted from colonic mucosa and Caco-2 cells using TRIzol reagent (TaKaRa Bio, Otsu, Japan), and 1 μg RNA was reverse transcribed using a PrimeScript RT reagent kit with cDNA Eraser (TaKaRa Bio) according to the manufacturer’s instructions. Quantitative reverse transcription-PCR (RT-qPCR) of the target genes was performed using the ABI 7300 real-time PCR system (Applied Biosystems, Foster City, CA, USA) with fluorescence detection of SYBR green dye. All the primers were synthesized by Invitrogen Life Technologies (Invitrogen, Shanghai, China) with their sequences shown in Table S2. The reaction system and PCR conditions used in the present study were referring to a previous study (48). The relative quantifications of gene expressions were analyzed by the cycle threshold (C_T) method described previously (48).

Western blot analysis.The frozen colonic samples were powdered under liquid nitrogen and lysed in RIPA buffer (150 mM NaCl, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris-HCl [pH 7.4]) plus a protease inhibitor (Thermo Fisher Scientific, Rockford, IL). After centrifugation at 13,000 × g for 10 min at 4°C, the protein concentration in the supernatant fluid was determined using a bicinchoninic acid (BCA) protein assay kit (Beyotime Biotechnology, Beijing, China). Equal amounts of proteins (60 mg) and prestained protein markers were electrophoresed on SDS-polyacrylamide gels. Proteins were transferred to polyvinylidene difluoride (PVDF) membranes (Millipore, Billerica, MA, USA) and blocked in 5% fat-free milk at room temperature in Tris-Tween-buffered saline (TTBS; 20 mM Tris-150 mM NaCl, pH 7.5, and 0.1% Tween 20) for 1 h. Then, samples were incubated with an antibody for ZO-1, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (1:1,000 dilution; CST, USA), MUC1, OCLD (1:1,000 dilution; Abcam, Cambridge, United Kingdom), Src, EGFR (1:1,000 dilution; Affinity Biosciences, OH, USA), and β-actin (1:1,500 dilution, Abcam, Cambridge, United Kingdom) at 4°C overnight with gentle rocking. The manufacturer has validated all the primary antibodies for use in swine. After being washed three times with TTBS, the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (goat anti-rabbit IgG; Thermo Fisher Scientific, Rockford, IL) at 1:5,000 dilution for 1 h at room temperature. Finally, the membranes were washed five times with TTBS and then developed (SuperSignal West Dura extended-duration substrate; Pierce, Rockford, IL). Band densities were detected on chemiluminescence (Applygen Technologies Inc., Beijing, China) and quantified using Alpha Imager 2200 (Alpha Innotech Corporation, CA, USA) software.

Immunohistochemistry analysis.Colonic tissue samples were removed, fixed in 4% paraformaldehyde, and embedded in paraffin, and 5-μm-thick sections were cut. Briefly, sections were deparaffinized and washed in phosphate-buffered saline (PBS), soaked in 3% H₂O₂ for 10 min, washed in PBS, and incubated with goat serum albumin for 20 min. Sections were then incubated with an antibody for ZO-1, OCLD, and MUC1 (1:1,000 dilution; Abcam, Cambridge, United Kingdom) at 37°C for 2.5 h. The embedded tissues were washed with PBS. After the sections were incubated with biotinylated anti-rabbit IgG and then processed by the horseradish peroxidase-conjugated streptavidin, color was developed in the diaminobenzidine (DAB) substrate solution. The sections were then counterstained with hematoxylin, dehydrated, cleared, and permanently mounted. Each sample was used to prepare four slides, and each slide had four sections. The integral optical density (IOD) of each specimen was analyzed by Image Pro Plus 6.0 software.

Histological measurements.Paraformaldehyde-fixed colonic tissues were embedded in paraffin, sectioned (6 μm), and stained with hematoxylin and eosin (H&E) or with periodic acid-Schiff stain (PAS). Measurements of crypt length were made using the 40× lens objective. Three slides per pig were prepared, and two images were captured per slide for a total of 36 replicates per measurement per group. Colonic crypt length and number of PAS-positive cells per crypt length were determined by analysis of 30 to 40 well-oriented crypts per pig (n = 8/group). Image Pro Plus 6.0 software was used to measure predefined criteria as previously described (49).

Statistical analysis.Statistical analysis of all data was performed using SPSS 20.0 (Chicago, IL, USA). For microbiota analysis, prior to investigating the probability of intergroup differences, normality and homogeneity of variances were studied with a Kolmogorov-Smirnov and Levene test, respectively. If the data were normally distributed and variances were equal, a one-way analysis of variance (ANOVA) with (post hoc) Bonferroni test was performed. Otherwise, a Kruskal-Wallis test was applied, and P values were corrected with the false-discovery rate (FDR) using the Benjamini-Hochberg method (q value < 0.05). The FDR q value < 0.05 was considered statistically significant. For data on nutrient substrates, microbial metabolites, gene expression, protein expression, and cell numbers, a one-way ANOVA with (post hoc) Bonferroni test was also performed, and a P value of <0.05 was considered statistically significant. Correlation analysis was assessed by Pearson’s correlation test using GraphPad Prism version 5.00 (GraphPad Software, San Diego, CA, USA). Correlation was considered significant when correlation coefficient was >0.4 and significance was P < 0.05.

Data availability.The raw reads were deposited into the NCBI Sequence Read Archive (SRA) database (accession number SRP174399).