Data sources
The NCBI-GEO database is a free and public database containing gene profiles. Two microarray datasets (GSE8006 about intestinal tract and GSE40334 about liver) were obtained from the GEO database (https://www.ncbi.nlm.nih.gov/gds/). The gene expression profiles of the GSE8006 dataset included 4 control liquid diet group samples and 4 alcohol liquid diet group samples. The gene expression profiles of the GSE40334 dataset included 3 colon germ free group samples and 3 colon specific pathogen-free group samples.
Identification of DEGs
GEO2R is an interactive web tool that can compare and analyze two different groups of samples under the same experimental conditions. In this study, DEGs in the GSE8006 dataset and GSE40334 dataset were analyzed using GEO2R. Subsequently, DEGs met the cutoff criteria of the adjusted P-value < 0.05 and |log fold change (FC)| > 1.0.
Functional enrichment analysis of DEGs
GO functional analysis and KEGG pathway analysis were perform to predict the potential functions and the main signaling pathways involved of the DEGs using the Database for Annotation, Visualization and Integrated Discovery (DAVID; https://david.ncifcrf.gov/; version 6.8) database. DEGs of GSE8006 and GSE40334 datasets were submitted to the DAVID online program. The top 10 items of the cellular component (CC), biological process (BP), and molecular function (MF) categories and KEGG pathways were then sorted and presented in the form of bubble maps. These bubble plots were drawn using the ggplot2 R package based on P-value (< 0.05 was considered statistically significant) through the statistical software R (version 3.6.1).
PPI network construction and hub gene identification
The Search Tool for the Retrieval of Interacting Genes database (Version 10.0, http://string-db.org) database was used to predict potential interactions between gene candidates at the protein level. A combined score of > 0.4 (medium confidence score) was considered significant. Additionally, cytoscape software (Version 7.2.0, http://www.cytoscape.org/) was utilized for constructing a PPI network. CytoHubba, a Cytoscape plugin of identifying hub objects and sub-networks from complex interactome, was utilized to explore PPI network hub genes; it provides a user-friendly interface to explore important nodes in biological networks and computes using eleven methods, of which MCC has a better performance in the PPI network.
Weighted gene co-expression network analysis (WGCNA)
The WGCNA R package was used to construct a weighted correlation network between the prognostic genes. In the network, we used the pairwise Pearson coefficient to evaluate the weighted co-expression relationship between all genes in the adjacency matrix. The soft threshold was used to ensure a scale-free network. In the unsigned co-expression network, genes with high absolute correlations were clustered into the same module. The modules were also identified by hierarchical clustering of the weighted coefficient adjacency matrix to calculate the topological overlap matrix (TOM). The absolute value of the threshold of correlation degree > 0.8. In addition, the topological overlap of the intramodules and adjacency modules was used to select the functional modules.
16S rDNA sequencing
Cecal contents were collected from some animals and frozen at -80°C. DNA was extracted using Stool DNA Extraction Kit (Qiagen) according to the manufacturer’s instructions. To check DNA quality and 16S content prior to sequencing, universal primers were used for SYBR Green quantitative polymerase chain reaction (qPCR) with the following extended cycling protocol: 95°C 10 min; 95°C 15 sec, 60°C 30 sec, 72°C 30 sec for 40 cycles. Sequencing was completed at the Cincinnati University Children’s Hospital Medical Center’s DNA Sequencing and Genotyping Facility Core (Cincinnati, OH) as described [14]. All antibiotic-treated samples failed to yield 16S rDNA sequence data; one sample each from the ethanol- and pair-fed groups was excluded based on insufficient sequence data.
UPARSE and UTAX (http://www.drive5.com/usearch/manual/cmd_utax.html) were used to generate OTU tables from 16S rDNA read data and to make taxonomic assignments. QIIME package scripts were used for calculations of α- (PD_whole_tree, chao1, observed_otus and shannon) and β- (Bray-Curtis, Un-Weighted UniFrac, and Weighted UniFrac) diversity.
Animal Experimentation
Male C57BL/6J mice (22 ± 2 g, six weeks old) and Fmo5 gene knockout (Fmo5-/-) mice (20 ± 3 g, six weeks old) were purchased from the Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China) and housed individually in cages with a 12-h light/dark cycle at 23 ± 2°C with full access to chow diet and water. All experimental protocols were approved by the Institutional Animal Care and Use Committee of the Zhengzhou University.
The AFLD mouse model was established using a modified Lieber–DeCarli liquid diet (21). The 40 mice were separated into four groups, each consisting of 10 mice. The first group (control group) and the third group (Fmo5-/- group) were fed with a normal standard growth diet while the other two groups (AFLD group and Fmo5-/-_AFLD group) were fed high-fat liquid diets (35% fat, 18% protein, and 47% carbohydrates, provided by TROPHIC Animal Feed High-Tech Co., Ltd., Nantong, China) for an acclimation period of two weeks acclimation. The amount of ethanol included increased over two weeks to reach a final concentration of 20% (v/v). Body weight gain and food intake were assessed once a week. The pair-fed control group (Con) was included in this model.
After 12 weeks, the fasted mice were euthanized, and blood samples were collected. The feces in the colon were also harvested into 2-ml sterile tubes to assess the gut microbiota. The tissues were immediately removed and weighed, and the liver coefficient (liver weight/body weight) was calculated. The left lobe of the liver was instantly fixed in 10% buffered formalin for histological analysis with the rest of the tissues frozen in liquid nitrogen and stored at −80°C until further use.
Hepatic Triglyceride Staining
The liver tissue was separated and rapidly fixed in 4% neutral buffered formalin solution for 24 h and then processed for paraffin embedding. Five-micrometer-thick paraffin sections were stained with hematoxylin and eosin (H&E) and oil-red solution. The liver steatosis status was examined under a light microscope (Olympus, Tokyo, Japan), and photographed at 100 × and 200 × magnification.
Measurement of liver function and serum pro-inflammatory cytokines
Blood samples were obtained from the common carotid artery to detect serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), tumor necrosis factor-α (TNF-α) and interleukin-6/-4 (IL-6/-4) when rats were sacrificed at 24 h after reperfusion. The levels of serum ALT and AST were measured using corresponding kits (Jiancheng, Nanjing, China) to assess the liver function. Serum TNF-α, IL-6, and IL-4 levels were measured by radioimmunoassay kits (Albert Poole Biotechnology, Beijing, China).
Apoptosis and immunohistochemical examination of liver
Harvested livers were placed into 10% formalin immediately after excision, and immersed for 24 h. Liver specimens were then embedded in paraffin, and sections were cut at 5 μm. Sections were stained for: (1) Apoptosis was detected by the TdT-mediated dUTP-biotin nick end labeling (TUNEL) method using an in situ apoptosis detection kit (Boster Biological Engineering, Wuhan, China) as previously described. The number of cells with TUNEL-positive nuclei was counted from 20 randomly selected fields at × 100 or × 200 magnification per liver sample. Results were expressed as the mean number of TUNEL-positive apoptotic hepatocytes per microscopic field; and (2) Immunobiological analysis of Fmo5, caspase-3, NF-κB p65 expression in paraffin wax-embedded liver sections were performed using a standard peroxidase-antiperoxidase technique as described previously, using a Fmo5 rabbit anti-mouse polyclonal antibody (Abcam Inc, United Kingdom) or mouse monoclonal antibodies against caspase-3 and NF-κB p65 (Santa Cruz, CA, United States) at a 1:150 dilution, with a biotinylated goat anti-rabbit or goat anti-mouse antibody (Santa Cruz, CA, United States) as the secondary antibody. Brown color in the cytoplasm of the hepatocytes was evaluated as positive staining.
Cell culture
Human normal hepatocyte cell line L02 was purchased from the Chinese Academy of Science (Shanghai, China). Cells were cultivated in RPMI-1640 medium (HyClone, Logan, UT, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 100 U/ml penicillin, and 100 µg/ml streptomycin according to the manufacturer's instructions. Cells were maintained at 37°C in a humidified atmosphere with 5% CO2. Half of the growth medium was changed each day. L02 cells at 60-70% confluency after the fourth passage were plated for the experiments.
Construction of AFLD cell model
To deplete L02 cells of iron, the chelator, desferrioxamine mesylate (DFO) (100 μM, D9533; Sigma-Aldrich), was added to the cell culture medium [16]. Iron loading of L02 cells was achieved by incubating cells in the presence of iron-dextran (0–100 μg/ml; Sigma-Aldrich). Cells were collected 24 h after treatment and at 48 h after treatment for extraction of mRNA and protein, respectively. For alcohol treatment, L02 cells were treated for 48 h with 200 nM alcohol (Sigma-Aldrich); then, cells were treated for 24 h with DFO (100 μM) in the presence of 200 mM alcohol.
Cell viability
The cells were seeded in 96-well plates at a density of 5 × 103 cells/well and cultured overnight. Subsequently, the cells were pretreated with various concentrations of ethyl alcohol (0-200 nM) for 72 h at 37°C. Cell viability was estimated using the cell counting kit-8 colorimetric assay (CCK-8 assay, Dojindo Molecular Technologies, Inc., Kumamoto, Japan). In addition, the viability was estimated using the cell counting kit-8 colorimetric assay.
Oil Red O staining
The Oil Red O stock solution was composed of 5 g Oil Red O in 100 ml isopropyl alcohol. Fresh Oil Red O working solution was prepared by diluting the Oil Red O stock with distilled water in a 3:2 ratio. Cells were washed twice in PBS after PA treatment, fixed with 4% formaldehyde for 30 min, and stained with Oil Red O working solution for 40 min at room temperature. Then, L02 cells were washed with 60% isopropyl alcohol once followed by washing with water twice. The nuclei of the cells were stained with Hematoxylin. The number of Oil Red O-positive cells was counted in replicates in 15 randomly selected fields at × 100 and × 200 magnification under a bright-field microscope.
Double immunofluorescent staining
The L02 cells were simultaneously incubated with primary antibodies against Fmo5 (1:50) and PPARα (1:50). The effect concentration of tetraethyl rhodamine isothiocyanate (TRITC) and fluorescein isothiocyanate (FITC) was 1:50. 4′,6-diamidino-2-phenylindole (DAPI) counterstained nuclei. At least three independent experiments were done.
Co-immunoprecipitation (Co-IP)
For Co-IP, anti-Fmo5 was coupled with protein A-Sepharose beads (Sigma) in RIPA buffer overnight at 4°C. The immune complex was then added to cell lysate and incubated at 4°C for 2 h. Endogenous Fmo5 was immunoprecipitated using an Fmo5 antibody (Santa Cruz Biotechnology, Inc., Dallas, TX, USA). The IP samples were analyzed by western blot using the following antibodies: Fmo5 (Abcam, Cambridge, MA, USA) and PPARα (Abcam).
Triglyceride (TG) assay
Cells from the different groups were harvested and washed twice with PBS. The intracellular triglycerides were measured using a triglyceride assay kit according to the manufacturer’s instruction (Applygen Technologies Inc., Beijing, China). The TG concentrations were normalized to the total cell protein concentration.
Intracellular ROS measurement
The intracellular ROS level is measured using the probe 2’,7’-dichlorofluorescein diacetate (DCFH-DA; Solarbio; Beijing; China). Briefly, cells were seeded in a 96-well plate at a density of 2 × 104 cells/well and cultured with their respective treatments for 24 h. After incubation with 10 μM DCFH-DA for 30 min at 37 °C, cells were washed twice with PBS. The fluorescence intensity is monitored using an excitation wavelength of 488 nm and an emission wavelength of 530 nm using a fluorescence microplate reader (Molecular Devices, Sunnyvale, CA).
Cell transfection
SiRNA was used to specifically target Fmo5 or PPARα. L02 cells were transiently transfected with Fmo5 siRNA or PPARα siRNA (Genesis Biotechnology, Wuhan, China) or negative control siRNA (Santa Cruz, CA, USA) according to the standard protocols. Briefly, cells were seeded at 2 × 105 in 12-well plates without antibiotics. Next, cells with 30–50% confluence were transfected with 50 nM siRNA or negative control using Lipofectamine 2000 (Invitrogen, USA) according to the manufacturer’s protocol.
Fmo5 expressing plasmids or PPARα expressing plasmids were obtained from Genesis Biotechnology (Wuhan, China). Briefly, cells were seeded at 2 × 105 in 12-well plates without antibiotics. Next, cells with 30–50% confluence were transfected with 30 nM expressing plasmids or negative control using Lipofectamine 2000 (Invitrogen, USA) according to the manufacturer’s protocol.
Cell treatment
The cells were seeded in 96-well plates at a density of 5 × 103 cells/well and cultured overnight. Subsequently, the cells were pretreated with different concentrations lipopolysaccharide (LPS; Sigma-Aldrich; 0, 50, and 100 ng/mL) for 12 h at 37°C.
Determination of apoptosis by Annexin V-FITC/PI staining
Chondrocyte apoptosis was determined by Annexin V-FITC/PI double labeling according to the manufacturer’s protocol. Briefly, after indicated cultures, chondrocytes were digested and suspended in binding buffer. Then 5 μl Annexin V and 5 μl PI solutions were added into cells and incubated for 15 min. Apoptotic rate was detected by BD FACSVerse™ flow cytometer (Becton Dickinson, Heidelberg, Germany) and analyzed by Cell Quest software (BD Biosciences). Experiments were conducted in triplicate.
Intracellular reactive oxygen species (ROS) levels detection by flow cytometry
Intracellular reactive oxygen species (ROS) levels were measured by ROS Assay Kit (Beyotime, Shanghai, China) with an oxidation-sensitive fluorescent probe dye, DCFH-DA (2, 7- dichlorodihydro fluorescein diacetate). In brief, cells were seeded in 6-well plates (5 × 103 cells per well) and incubated at 37°C, 5% CO2 atmosphere for 24 h. Next, the cells were collected and washed with PBS, and then incubated at 37°C with DMEM containing 10 μM DCFH-DA for 30 min. After that, cells were washed three times using serum-free DMEM. The samples were finally analyzed using a FACScan flow cytometer and Flowjo 7.6.1 software.
Immunofluorescence staining
The cells were seeded onto 96 black well plate (5 × 104 cells/well) and cultured until 80% confluence. After 30 min of incubation, cells were fixed with 4% paraformaldehyde and permeabilized with 0.05% Triton X. Then, cells were incubated overnight with primary antibodies against NF-ĸB p65 (1:400) at 4°C. Samples were then washed with PBS, incubated for an hour with Alexa Fluor 488-conjugated goat anti-rabbit IgG (1:400), followed by DNA staining for 15 min with hoechst 33342 staining (1: 1000). The fluorescently labeled NF-ĸB p65 was detected by Operetta High-Content Imaging System.
Western blot analyses of tissue and cells
Dissected livers were snap-frozen in liquid nitrogen and protein lysates for the detection of HIF1α and HIF2α proteins were extracted with the NE-PER kit (Pierce). Primary hepatocytes and Hep3B cells were lysed in buffer of 50 mM Тris–HCl, pH 8, 150 mM NaCl, 0.5% NP-40, 1 mM PMSF and a complete protease inhibitor cocktail tablet (Roche), kept on ice for 20 min and centrifuged at 12 000 g for 20 min at 4°C. 100 μg of proteins from liver or cells were resolved by SDS-PAGE, blotted, and probed with the following primary antibodies: Fmo5, PPARα, caspase-3, NF-kB p65, IL-6, IL-4, TNF-a, and GAPDH. The secondary antibody, either anti-mouse or anti-rabbit, was conjugated to horseradish peroxidase (1:5000; Bio-Rad Laboratories). The Pierce ECL system (ThermoScientific) was used for detection.
RNA Extraction and Real-Time PCR
The total RNA was extracted by using a TRIzol reagent. The first-strand cDNA was synthesized from 1 μg of total RNA using the Reverse Transcription System Bestar qPCR RT Kit according to the manufacturer instruction. Real-time PCR was carried out with an ABI 7500 Real-Time PCR System (Applied Biosystems, Lincoln Centre Drive, Foster City, CA 94404, USA). Each assay was performed in triplicate. The relative amount of Fmo5 and PPARα were calculated using with a 2−ΔΔCt method and normalized using GAPDH as an internal control. The primers used in this study were shown below: for Fmo5, 5’-CCAGTTACGTGAATGATTCG-3’ (forward), 5’- AGCGCGTGTGAATGCAGGCC-3’ (reverse); for PPARα: 5’- CTTATAACTGCGGGAGGAC-3’ (forward), 5’-TACGTGCTGAACATGAT-3’ (reverse); for IL-6, 5’-CTGGTGTGTCACTGTCAACA-3’ (forward), 5’- CTGGAAACACGTGACAAAC-3’ (reverse); for IL-4, 5’-ACCTGTGTCGCTGAAACTGGGCCC-3’ (forward), 5’- TGGCGCAACACAGTGCACCA-3’ (reverse); for TNFα, 5’-CTGAACGCTGTGCTGAAAAGCT-3’ (forward), 5’- ATTCGCATTGTCATGCCATGTT-3’ (reverse); and for GAPDH: 5’-CATCGATTAGGGCATGCGC-3’ (forward), 5’- TCGTAACTAGGGCTACCGC-3’ (reverse).
Measurement of TG content in liver
TG content in liver was determined as previously described. Briefly, 250 mg of liver sample was homogenized in 1.5 ml of methanol, and then added with 5.0 ml of MTBE and shaken at room temperature for 1 h. Subsequently, 1.25 ml of high purity water was added and mixed for 10 min, followed by centrifuging at 1,000 g for 10 min, and the upper organic layer was collected. The aqueous layer was re-extracted with MTBE/methanol/water mixture (10/3/2.5 v/v/v; 2 ml) and the combined organic layers were dried under nitrogen. The extracted dried lipids were dissolved in a mixture of triton X-114/methanol (2:1v/v, 60 μl) and analyzed for triglyceride (L-Type TGH) using commercially available kits (Wako Diagnostics, Richmond, VA, USA).
Statistical analysis
Sequence data were processed using Quantitative Insights into Microbial Ecology (QIIME, version 1.9.1). The low-quality sequences, which had lengths of <150 bp and average Phred scores of <20 and contained ambiguous bases and mononucleotide repeats of >8 bp, were filtered using the following criteria. Paired-end reads were assembled using FLASH. The remaining high-quality sequences were clustered into operational taxonomic units (OTUs) at 97% sequence identity by UCLUST (Edgar 2010). OTU taxonomic classification was conducted by BLAST and the OTUs containing more than 99.999% of the total sequences across all samples were reserved.
QIIME (version 1.9.1) was used to calculate all diversity indexes in these samples, and R software (version 3.5.2) was used for visual analysis. LEfSe software (LEfSe 1.0) was used for linear discriminant analysis. Principal coordinate analysis (PCoA) was performed using Bray-Curtis. The functions of gut microbiota were predicted for the CON and AFLD groups using Tax4Fun in the KEGG database.
All results are presented as means ± standard deviation. Data were analyzed with SPSS 19.0 using One-way analysis of variance (ANOVA), and, when appropriate, using a two-tailed Student’s t test between different groups. Differences among groups were evaluated for significance with the comparable variances, followed by Tukey’s and least significant difference (LSD) tests. P < 0.05 was considered statistically significant.