Preparation of BL powder
BL powder used in this study was crude powder without any extraction process and was supplied by Hebei Biotechnology Co., Ltd. (Jiaxing, China). Briefly, fresh leaves of Hordeum vulgare L. (cultivated in Hangzhou, China) were washed with water, cut into pieces and dried in a freeze dryer OE-950 (Labor, MIM, Budapest, Hungary) at -60°C for 24 h. The dried BL was pulverized with a blender (KA-2610, Jworld Tech, Korea) for 1 min, screened through a 300 mesh sieve and stored at -20°C until use. The proximate analysis of the BL powder was performed by Pony Testing International Group (Beijing, China) (Additional file 2: Table S1).
Eight-week-old female C57Bl/6J mice were obtained from Vital River Laboratory Animal Technology (Beijing, China) and maintained under specific-pathogen-free (SPF) facility with a 12-h light and dark cycle. All mice were adapted to the laboratory conditions with ad libitum access to food and water for at least one week. Animal experiments were conducted in accordance with the Guidelines for Animal Experimentation of Peking University Health Science Center (Beijing, China), and the protocols were approved by the Animal Ethics Committee of this institution. Mice were randomly distributed and fed with either a standard chow diet (CD) or an isocaloric diet where BL was supplemented at a ratio of 2.5%, a dosage translated from a previous human intervention study . The composition of two diets was listed in Additional file 2: Table S2 and prepared by Hfk Biotech Co, Ltd (Beijing, China).
For the treatment with PPARγ antagonist GW9662 or A2AR antagonist SCH58261, mice were treated with 3 mg/kg/day of GW9662 intragastrically  or treated with 2 mg/kg/day of SCH58261 intraperitoneally .
For antibiotic experiment, a combination of neomycin (100 mg/l), streptomycin (50 mg/l), penicillin (100 mg/l), vancomycin (50 mg/l) and metronidazole (100 mg/l) were administered in the drinking water.
For the treatment with inosine, mice were treated intragastrically with 800 mg/kg/day of inosine (Sigma-Aldrich) dissolved in PBS at the concentration of 40 mg/ml .
Dextran sodium sulfate (DSS)-induced colitis
Colitis was induced by administering 2.5% DSS (molecular weight 36,000–50,000 kDa; MP Biomedicals) dissolved in drinking water for 7 days. The mice were weighed daily and monitored for signs of stool consistency and rectal bleeding. Evaluation of disease activity index was performed by combining the parameters of weight loss, stool consistency and rectal bleeding as described previously . The disease activity index was the mean of the total score of the three parameters. Mice were sacrificed on day 7 and colon length was measured.
Assessment of intestinal permeability
Briefly, fluorescein isothiocyanate (FITC)-dextran (4 kDa; Sigma) was dissolved in PBS at a concentration of 100 mg/ml. After mice were fasted for 4 h and orally administrated with FITC-dextran (60 mg/100 gm body weight). Blood was collected following another 4 h and was centrifuged at 1,000 rpm for 20 min. Serum was collected and fluorescence was quantified at an excitation wavelength of 485 nm and 535 nm emission wavelength.
16S rRNA gene sequencing and analysis
Total DNA was extracted from colonic contents or fecal culture samples using a QIAamp-DNA Stool Mini Kit (Qiagen, Hilden, Germany). The integrity of the extracted DNA was examined by electrophoresis in 1% (wt/vol) agarose gels. Based on the quantity and the quality of the DNA extracted, samples were selected to perform the consequent sequencing.
DNA samples were used as the template for PCR amplification of the V3-V4 region of bacterial 16S rRNA genes. PCR amplification was performed on ABI GeneAmp®9700 PCR System (AppliedBiosystems, Foster City, CA, USA) and the PCR amplification products were quantified with a QuantiFluorTM-ST Handheld Fluorometer with UV/Blue Channels (Promega Corporation, Madison, WI, USA)
Sequencing of the PCR amplification products was performed on an Illumina Miseq platform at Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China). Briefly, the 16S rRNA gene sequencing data was filtered and trimmed, and further classified into operational taxonomic units (OTUs) within a 0.03 difference (equivalent to 97% similarity). A representative set of sequences was then generated and was assigned taxonomy using the SILVA database (Release115 http://www.arb-silva.de). Analyses for rarefaction curves, and calculation of richness estimators and diversity indices were performed using the MOTHUR program. Taxonomic community structure and phylogeny were analyzed through visualization of the data sets of the microbial diversities and abundances of different samples.
Inflammatory cytokine analysis
Cytokine levels in serum and colonic tissues were detected by enzyme-linked immunosorbent assay (ELISA) kit (Nanjing Jiancheng Bioengineering Institute Co., Ltd. Nanjing, China) according to the manufacturer's instructions.
Collected colon samples were fixed in 4% paraformaldehyde and embedded in paraffin. Paraffin-embedded colonic tissues were sectioned (4µm in thicknesses) and were subjected to hematoxylin and eosin (HE) and alcian blue (AB) staining by using commercial kits (Beijing Solarbio Technology Co., Ltd. Beijing, China) according to the manufacturer's instructions. The crypt height and muscular layer width in the large intestine were measured using Image J (Media Cybernetics, Inc., Rockville, MD, USA). Mucusproducing goblet cells were observed and counted under light microscope (Leica DM500). Six microscopic fields of every section of the testes were randomly selected.
Histological pathology was scored according to a previously established scoring system, as follows: crypt damage (0–4 scale), severity of inflammation (0–3 scale) and depth of injury (0–3 scale) .
Intestinal motility assessment
The intestinal motility was determined through evaluating the intestinal transit time and defecation frequency. After mice were fasted for 4 h, individual mouse was gavaged with the 10 µL/g of 6% carmine solution (in 0.5% methylcellulose). The time from gavage to initial appearance of carmine in the feces was recorded as the gut transit time for a given mouse.
Defecation frequency was recorded by separating mice into a sterile cage and the cumulative fecal pellets within 60 min were recorded as the corresponding defecation number for a given mouse.
Scanning electron microscopy
Colonic tissues were fixed with 2.5% PBS-diluted glutaraldehyde. After being rinsed three times in PBS, samples were post-fixed in 1% osmium tetroxide for 1 h, dehydrated in alcohol, and then critical-point-dried using CO2. The samples were coated with gold and observed in a Hitachi S-3400 scanning electron microscope (Hitachi, Japan).
Transmission electron microscopy
Colonic tissues were fixed with 2.5% glutaraldehyde. PBS was used to remove excess fixative. Samples were then fixed with 1% osmium tetroxide at 4°C for 2 h, dehydrated in acetone. Samples were finally infiltrated with 1:1 propylene oxide and EPON resin for 1 h followed by overnight infiltration in 100% EPON's resin. The tissues were embedded in flat molds in 100% EPON for 36 h at 60°C. Ultra-thin sections of 70 nm were stained with uranyl acetate and lead citrate (10 min each) and viewed under H-7650 transmission electron microscope (Hitachi, Japan).
Immunohistochemistry and immunofluorescence
Following deparaffinization and rehydration, colonic sections were blocked with 5% bovine serum albumin for 30 min at room temperature and then washed with PBS. Tissue sections were incubated with primary antibody Muc-2 (Santa Cruz, sc-515032), Ly6G (Abcam, ab25377) and F4/80 (Abcam, ab6640) overnight at 4oC. Slides were washed three times in PBS before applying peroxidase-conjugated secondary antibody (Invitrogen) for 2 h at room temperature.
For immunofluorescence staining, paraffin-embedded colonic sections or cells were incubated with primary antibody PPARγ (Abcam, ab59256) at 4oC overnight. Next day, Alexa Fluor488 donkey antirabbit secondary antibody (Invitrogen) was used.
Quantitative real-time PCR (qPCR)
Total RNA of colonic tissues was isolated by using TRIZOL reagent (Invitrogen, USA). The quantity and purity of RNA was assessed by absorbance at 260 nm and 280 nm. Reverse transcription of 10 µl RNA was performed using PrimeScript RT Master Kit (Takara, Japan). Gene expression was measured in a MyiQ single color Real-Time PCR detection system (Bio-Rad) with SYBR Real time PCR Kit (Takara, Japan). GAPDH was used as endogenous control. Average Ct values from triplicate analyses were normalized from average Ct values of GAPDH. The primer sequences are described and synthesized by Sunbiotech Co. (Beijing, China) (Additional file 2: Table S3).
Colon tissues were collected and the total RNA was extracted by using TRIZOL reagent (Invitrogen, USA). RNA quality was evaluated by electrophoresis using an Agilent 2100 Bioanalyzer (Agilent Technologies, San Diego, CA, USA). Samples with RNA integrity numbers (RINs) > 9.4 and with 260/280 nm absorbance ratios from 1.9 to 2.1 were used for construction of RNA-Seq libraries. Libraries were constructed using the TruSeq™ RNA Sample Prep kit (Illumina, San Diego, CA, USA) according to the manufacturer’s instructions.
Sequencing of the Libraries was performed on an Illumina HiSeq2000 instrument by Shanghai Majorbio Biopharm Biotechnology (Shanghai, China), and individually assessed for quality using FastQC. Analysis of differential expression was carried out using DESeq2 . Statistical significance was assessed using a negative binomial Wald test, then corrected for multiple hypothesis testing with the Benjamini-Hochberg method. Functional enrichment cluster analysis was performed for Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis.
Gas chromatography-mass spectrometry (GC-MS)-based metabolomics was performed by ProfLeader Biotech (Shanghai, China). Briefly, serum samples were separated from the blood by centrifugation and stored at − 80°C. Colonic contents or colon tissue samples mixed with water were vortexed prior to centrifugation. The supernatant was transferred to a GC vial containing internal standards. The mixture was dried under gentle nitrogen stream and then added with methoxyamine hydrochloride in pyridine. The resultant mixture was vortexed vigorously and incubated at 37°C for 90 min. Derivatization was performed by adding BSTFA (with 1% TMCS) into the mixture. The derivatized samples were analyzed by an Agilent 7890A gas chromatography system coupled to an Agilent 5975C inert MSD system (Agilent Technologies, CA, USA). A HP-5MS fused-silica capillary column was utilized to separate the derivatives. Helium was used as a carrier gas at a constant flow rate through the column. The samples were analyzed in a random sequence.
The acquired GC/MS data were imported to SIMCA Statistical Analysis (version 13.0, Umetrics AB, Umeå, Sweden), where multivariate statistical analysis including principal component analysis (PCA) and partial least-squares discriminant analysis (PLS-DA) were performed. The differential metabolites were determined by the combination of the Variable importance in the projection (VIP) value (> 1) of PLS-DA model and the P values (< 0.05) from two-tailed Student's t test on the normalized peak intensities. Fold change was calculated as a binary logarithm of the average normalized peak area ratio between the two groups. The structural identification of differential metabolites was performed using AMDIS software, where the purified mass spectra were automatically matched with an in-house standard library including retention time and mass spectra, Agilent Fiehn GC/MS Metabolomics RTL library and Golm Metabolome Database, respectively.
Quantitative analysis of purine metabolites
Serum, colonic contents or fecal culture samples were mixed with distilled water containing internal standard (4-aminosalicylic acid). The mixture was extracted by ethyl acetate. Following centrifugation, the top layer was transferred and evaporated to dryness under nitrogen stream. The dry residue was reconstituted in acetonitrile and BSTFA (with 1% TMCS), and derivatized to perform GC-MS analysis. A mixed standard solution containing inosine and guanosine stock standard solutions was prepared at a concentration of 100 µg/mL for quantitation.
In vitro anaerobic culturing
Fresh mice feces (0.5 g) from 8-week-old male C57Bl/6 mice were resuspended in 10 ml sterile PBS and co-cultured with BL powder with a final concentration of 10 mg/ml under anaerobic conditions (10% H2, 10% CO2, 80% N2). After incubation for 12 and 24 h, cultured samples were collected and used for metabolites detection and bacterial composition analysis.
Cell culture and treatment
Human colon epithelial carcinoma cell lines HT29 and Caco2 were purchased from the American Type Culture Collection (ATCC). Cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin in a humidified incubator (5% CO2, 95% air, 37 oC). Cells were grown until 80 to 90% confluence and treated for the indicated time periods with inosine (Sigma-Aldrich) and guanosine (Sigma-Aldrich) from a stock in PBS to a final concentration of 0.5, 1, 2 and 4mM. Cell viability was evaluated with the Cell Counting Kit (CCK)-8 cell viability assay kit from GenMed Inc. (Shanghai, China) according to the manufacture’s instruction.
Luciferase reporter assay
PPARγ-luciferase plasmid vector was purchased from RiboBio Co., Ltd (Guangzhou, China). HT29 cells were transiently transfected using lipofectamine 3000 reagent (ThermoFisher Scientific). Briefly, 1 × 105 cells were seeded on 6-well plates and grown for 24 h. The transfection complex containing 1 µg of plasmid DNA and transfection reagent was added to each well in absence of FBS. After 6 h, medium containing 10% FBS was added and cells were treated with inosine and guanosine (0.5, 1, 2 and 4 mM) for 24 h. The luciferase activity was measured with Luciferase Assay System (Promega) using a luminometer (Perkin Elmer, Covina, CA).
Western blot analysis
Cells were lysed with lysis buffer with 1% protease inhibitor cocktail to harvest total cellular protein. The concentration of extracted protein was quantified by bicinchoninic acid (BCA) protein assay kit (Beyotime). An equal amount of protein sample was loaded on sodium dodecyl sulphate-polyacrylamide gel electrophoresis gel and then transferred to a polyvinylidene difluoride membrane (Millipore, Billerica, MA, USA). After the membrane was blocked with skimmed milk, it was incubated with primary antibodies against Muc2 (Santa Cruz, sc-515032), PPARγ (Abcam, ab59256), A2AR (Abcam, ab3461) and GAPDH (Abcam, ab8245). Next day, secondary antibodies conjugated with Horseradish peroxidase were probed for 2–3 h. Protein signals were detected with enhanced ECL chemiluminescence reagent based on the manufacturer’s instructions.
Small interfering RNA (siRNA)
The siRNA against PPARγ or A2AR was purchased from GenePharma (Shanghai, China). For knockdown experiments, 1 × 105 HT29 and Caco2 cells were plated in 6 well plates and grown for 24 h. The PPARγ, A2AR or control-siRNA was transfected into cells using lipofectamine 3000 reagent (ThermoFisher Scientific). After 24 h of transfections, cells were induced with inosine (0, 2 and 4 mM) for 24 h. And then cells were lysed using RIPA buffer and the protein expression of A2AR, PPARγ or Muc2 was detected by western blot.
Statistical analysis was performed using Prism 6.0 (GraphPad Software, CA). Data are presented as means ± SEM. Significant differences between the two groups were evaluated by two-tailed unpaired Student’s t test. Significant differences more than two groups were evaluated by one-way or two-way analysis of variance (ANOVA) followed by Tukey’s multiple comparison’s test.
When analyzing gut microbiota sequencing data, two-tailed Wilcoxon rank-sum test by R Project were performed. When analyzing the differences of abundance distributions among metabolites, Mann-Whitney U test with Benjamini-Hochberg false discovery rate correction were performed. The differences were considered to be significant at p < 0.05.