Human subjects A total of 34 PBMCs from patients of SLE and 16 healthy controls were from Ren Ji Hospital Biobank of Shanghai Jiao Tong University School of Medicine. The study was approved by the Research Ethics Board of Ren Ji Hospital. Signed Informed Consent Forms were collected from all subjects. All enrolled SLE patients were diagnosed recording to SLICC criteria for SLE in 201254,55, or 2019 EULAR/ACR criteria56. The demographic clinical information of the subjects was listed in Supplementary Table 1–3.
Mice. C57BL/6N mice were purchased from Vital River Laboratories. Female NZB/BINJ mice and male NZW/LacJ mice were purchased from The Jackson Laboratory. NZB/NZW F1 mice were produced by hybridizing crosses between female NZB/BINJ mice and male NZW/LacJ mice. All animals were maintained in a specific pathogen-free facility at Shanghai Rat&Mouse Biotech Company in a 12-h light/12-h dark cycle and free acquirement to sterile water and a standard mouse diet. Animal studies were approved by the Institutional Animal Care and Use Committee at Ren Ji Hospital, Shanghai Jiao Tong University School of Medicine.
For ex vivo assessment of the effect of Diosmetin on I IFN signaling, 8 weeks old female C57BL/6N mice were intraperitoneally administrated with Diosmetin (50mg/kg) or vehicles for 48 hours, then splenocytes from these mice were stimulated with 1000 U/ml of universal type I IFN (R&D System).
To evaluate the therapeutic effects of diosmetin, 6 weeks old female NZB/NZW F1 mice were intravenously injected with 109 pfu of IFNa5-encoding adenovirus particles (ViGene Biosciences). After 6 weeks, mice were randomized and grouped regarding their urinary protein levels. These mice were intraperitoneally administrated with diosmetin (50mg/kg, Selleck Chemicals) or vehicles every other day for continuous treatment cycles of 6 weeks. TMPD induced lupus mice model was created by intraperitoneally injecting C57BL/6N mice with 0.5 ml TMPD once (0.783g/ml, Sigma). Five months later, diseased mice were randomly divided into two groups accepting diosmetin or vehicle treatment. Peripheral blood from above mice was collected in tubes containing ethylenediaminetetraacetic acid to isolate plasma. Anti-dsDNA autoantibody was detected with Mouse anti-dsDNA IgG-specific ELISA Kit (Alpha Diagnostic International), anti-ANA autoantibody was detected with Mouse anti-ANA ELISA Kit (Alpha Diagnostic International). To evaluate the hepatotoxicity of diosmetin, plasma ALT and AST level were measured separately with Mouse ALT or AST ELISA Kit (Alan Transaminase, FineTest). At a specific time point, mice were kept in separate metabolic cages for 24 hours, and the urine was collected to evaluated the kidney damage of lupus mice. Urinary protein was detected by BCA Protein Assay Kit (TIANGEN).
Cell culture THP-1 cells (Cell Bank, Shanghai Institutes for Biological Science, Chinese Academy of Sciences) were cultured in RPMI 1640 medium supplied with 10% (v/v) fetal bovine serum (Gibco), 2 mM L-Glutamine, 1mM sodium pyruvate and 10 mM HEPES at 37℃ in a cell incubator containing 5% CO2. PBMCs were collected by using Ficoll-Paque PLUS (Citiva) according to manufacturer's instructions. CD14 + monocytes were separated from PBMCs via magnetic beads sorting (Miltenyi Biotech), and the purity (> 98%) was verified by anti-CD14 antibody (BD biosciences) through flow cytometry (BD FORTESSA). THP-1 cells or primary monocytes were pretreated with 50 uM Diosmetin or vehicle for 30 minutes, then type I IFN 1000 U/ml was added. PBMCs from patients of SLE were incubated with Diosmetin 50 uM or vehicle for 3 hours. Cells were collected to extract RNA or protein for downstream analysis.
For transient transfection, THP-1 cells (2*105/well) were cultured in a 24-well plate for 24 hours, and siRNAs (200uM) targeting CYP1B1 were transfected into THP-1 cells with LipofectamineTM RNAiMAX (Invitrogen). 24 hours later, cells were stimulated with type IFN-I (1000 U/ml) and harvested at indicated timepoint for further analysis. The sequences of all siRNAs (RIBOBIO Biotech) are in Supplementary Table 4.
For stable transfection, hsa-CYP1B1 (transcripts NM_000104.4) was constructed into pLV6ltr-ZsGreen-Puro-CMV (TSINGKE), the modified plasmid was transfected into 293T cells together with package plasmids (TSINGKE). The virus supernatant was collected for concentration and titrated to 5E + 8TU/mL. Lentiviruses that carried CYP1B1 or negative control were respectively added to THP-1 cells at MOI 50. After 72h, THP-1 cells stably expressing CYP1B1 were sorted by flow cytometry (BD AIRA, BD biosciences).
For NADP + and NADPH quantification, THP-1 cells, THP1-NC and THP1-CYP1B1 cells were individually pretreated with 50 uM Diosmetin or vehicle for 30 minutes, then type I IFN 1000 U/ml was added for the designated time. Then each sample was harvested and resuspended by 200 µL NADP+/ NADPH extraction solution (Beyotime, China), the extracts were centrifuged at 12000 g for 10 min at 4°C to collect the supernatant. The supernatant was added to the 96 well plates with G6PDH working solution and incubated at 37°C for 10 min. Then the absorbance was read at 450 nm absorbance to calculate the total quantity of NADP + and NADPH. Next, NADP + was water bath inactivated at 60°C for 30 min. And the supernatant was measured as above to calculate the quantification of NADPH.
For the measurement of intracellular pH (ipH), THP-1 cells were pretreated with 50 uM Diosmetin or vehicle for 30 minutes, then type I IFN 1000 U/ml was added. And cells were harvested and washed twice with PBS (Ca2+ -free, Mg2+-free, PH = 7.4, BasalMedia, Shanghai), then cells were incubated in PBS containing 2 µM BCECF-AM, a fluorescent probe that binds specifically to hydrogen ions, for 30 min at 37°C in 5% humidified CO2. Next, cells were then washed with PBS three times. The ipH level was measured by the mean fluorescence intensity through flow cytometry (BD FORTESSA) with an excitation wavelength 488nm and an emission wavelength 535nm.
Hematoxylin and eosin (H&E) staining, IHC assay and immunofluorescence staining of kidney tissue. Kidneys from lupus mice model were fixed in 4% formaldehyde solution immediately after the mice were sacrificed. Then the fixed tissue were embedded in paraffin and 5 um thickness sections were made and stained with hematoxylin and eosin (H&E). Blinded assessment of renal pathological scores was performed by a senior renal pathologist, as previously described 57.
For IHC assay, sections above proceeded to deparaffinization and hydration. Antigen was retrieved in citrate buffer (0.01 M, pH 6.0) at 95°C for 10 min, and then the sections were incubated with blocking buffer (5% BSA in PBS) for 60 min at room temperature, stained with anti-CYP1B1 (ABclonal Technology) at 4°C overnight. On day 2, sections were incubated with secondary antibody. Then the slides were incubated with secondary antibody at the second day before incubating with DAB and DAB buffer (1:19 mixture) for 10 minutes (Dako REAL™ EnVision™ Detection System). Slides were counterstained for 5 minutes with Schmidt hematoxylin, followed by several rinses with wash buffer and distilled water. Subsequently, slides were dehydrated with increasing concentrations of ethanol and cleared with xylene. At last, the image was recorded by Nikon Eclipse Ci-L microscope. The proportion of areas staining positive for CYP1B1 was quantified by IHC profiler plugin in ImageJ (NIH image program, version 2.9.0). Immune complexes precipitation in the glomerulus is stained by immunofluorescence with anti-mouse IgG2a (Santa Cruz Biotechnology) and anti-mouse C3 (Abcam). The immunofluorescence was assessed by TissueFAXS Viewer software.
Cellular immunofluorescence staining. Cells were collected and washed with PBS to remove stimuli, then cells were stained with DiOC18(3) (3,3'-dioctadecyloxacarbocyanine perchlorate) (Cell Plasma Membrane Staining Kit ,Beyotime). Then, cells were fixed and permeabilized by BD Cytofix/Cytoperm™ kit (BD Bioscience). Next, the cells were blocked by 10% FBS for 30 minutes, then they were individually incubated by primary antibody IFNAR1/IFNAR2/CYP1B1 (ABclonal Technology) and duramycin-LC-biotin (Molecular Targeting Technologies) at 4 ℃ overnight. On day 2, the secondary antibody Cy3 Goat Anti-Rabbit IgG (H + L) (ABclonal Technology) and APC-conjugated streptavidin (BD Bioscience) were added. Finally, for nuclear staining, DAPI (1:5000) (BD Bioscience) was incubated for 10 minutes. Images of cells were acquired by the confocal microscope (Leica, SP8). ImageJ software with plugin JACoP47 and RGB Profiler were applied to analyze the fluorescence intensity in marked regions, calculate the Manders’ coefficient and quantify the RGB intensity58.
Quantification of Gene Expression. Total RNA was extracted with TRIzol reagent (Invitrogen), then RNA was synthesized to complementary DNA with PrimeScript RT Reagent Kit (Takara Bio). The expression of different genes was quantified with SYBR Premix Ex Taq Kit (Takara Bio). Then genes were amplified on a QuanStudio 7 Flex Real-Time PCR System (Applied Biosystems). The expression of genes was normalized to RPL13a by the 2-△△Ct value. All primers are listed in Supplementary Table 5.
Immunoblotting. Cells were collected and lysed in RIPA Lysis Buffer supplemented with Halt™ Protease and Phosphatase Inhibitor Cocktail (Thermo Fisher Scientific). Individual cell lysates with 10 ug/lane were separated by 10% SDS PAGE and then transferred to PVDF membranes (Millipore). Next, the membranes were blocked by BSA buffer (Sangon Biotech). Targeted proteins were detected by corresponding primary antibodies and corresponding secondary peroxidase-conjugated antibodies. The antibodies used in the study are listed in Supplementary Table 6. Finally, bands were detected by Pierce ECL Western Blotting Substrate (Thermo Fisher Scientific). Images were captured by Tanon 5500 Imaging System and measured by ImageJ software.
Gene expression profiling. THP-1 cells were collected and proceeded to RNA extraction with TRIzol reagent. RNA integrity was evaluated using the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). The samples with RNA Integrity Number (RIN) ≥ 7 were quality acceptable and subjected to the subsequent sequencing. The libraries were constructed with TruSeq Stranded Total RNA with Ribo-Zero Gold. These libraries were then sequenced on the Illumina sequencing platform (HiSeqTM 2500 or other platforms) and fragments with 150/125 bp paired-end reads were obtained. Differential screening analysis and functional analysis were proceeded by estimateSizeFactors function of the DESeq R package to normalize the counts and then by nbinomTest function analysis to calculate the P-value and fold change values for the difference comparison. Transcripts with p-values < = 0.05 and fold change > = 2 were selected. Analysis of differential mRNA GO and KEGG enrichment is by Hypergeometric Distribution Test.
LC–MS lipid profiling and data processing. Celluar samples were collected in liquid nitrogen immediately and proceeded a modified Folch lipid extraction procedure. Subsequently, above samples were analysed by LC-MS/MS lipid profiling. LC separations was performed on ExionLC™ System and MS was QTRAP® 6500+ (Sciex, USA) equipped with an IonDriveTM Turbo V source.
A total of 15 kinds of glycerophospholipids, glycerolipids, and sphingolipids isotope-labeled internal mix standards were from Avanti Polar Lipids (Birmingham, AL, USA), Palmitic acid-16,16,16-d3, Stearic acid-18,18,18-d3, and different Acyl carnitine-d3 were purchased from Sigma-Aldrich (St. Louis, MO, USA). All organic solvents and water used in sample and mobile phase preparations were HPLC grade and obtained from Fisher Scientific (Fair Lawn, NJ, USA).
The raw data generated by LC-MS/MS was first evaluated by QC sample. Qualified samples were further processed using MRMPROBS software to perform peak integration, calibration, and quantitation for each lipid. Data matrix of lipids was further analyzed on MetaboAnalyst 5.0. Partial Least Sequences- Discriminant Analysis (PLS-DA) was used to identify the important features in different groups. Importance in Projection (VIP) is a weighted sum of squares of the PLS loadings taking into account the amount of explained Y-variation in each dimension. VIP score id calculated for each component. When more than one component is used to calculate the feature importance, the average of the VIP scores is used. The other important measure of PLS-DA is the weighted sum of PLS-regression.
For star pattern recognition and statistical analysis, the concentration of three phosphatidylcholines (PC) and three phosphatidylethanolamines (PE) were detected by LC-MS/MS lipid profiling. Then, the quantitative level of each PC and PE was normalized to the corresponding mean value of DMSO group. These normalized values were then used to draw star graphs.
Molecular docking. The molecular docking studies were carried out between diosmetin and five potential ligands predicted by the STITCH database (http://http://stitch.embl.de/). The 2D structure of diosmetin and selected ligands was obtained from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/), then it was transferred to the 3D structure by Chem 3D 19.0 software. All bound substances (ligands and cofactors) and water molecules associated with the receptor were removed before in silico docking process. The prepared ligands were docked with the prepared structure of diosmetin using Discovery Studio 2019 software. The docking results were evaluated by Libscores which were generated in the software.31
Statistical analysis. Results were analyzed and figures were generated with GraphPad Prism (Version 9.0.0). The corresponding nonparametric Mann-Whitney U-test, Spearman’s correlation test, Student’s paired t-test, two-way ANOVA, and log-rank test were applied according to different data types. A significant difference was considered when the P value was < 0.05.