All animal studies were performed according to protocols approved by the Animal Ethics Committee of the Fudan University School of Basic Medical Sciences. Age- and sex-matched, 8–12-week-old C57BL/6J mice used in this study were purchased from the Experimental Animal Center of Shanghai, SLAC. CRISPR/Cas9 technology at GemPharmatech Co., Ltd. (Nanjing, China) was used to modify the Sult2b1 gene. Briefly, sgRNA targeting exon 3 of the Sult2b1 gene was transcribed in vitro, and a donor vector was constructed. Cas9, sgRNA, and donor were microinjected into fertilized eggs of C57BL/6JGpt mice. Fertilized eggs were transplanted to obtain F0-positive mice, which were confirmed using PCR and sequencing. A stable F1 generation mouse model was obtained by mating positive F0 generation mice with C57BL/6JGpt mice. The floxed mice were knocked out after mating with mice expressing Villin-Cre recombinase, resulting in the loss of function of SULT2B1 in intestinal epithelial cells. Sult2b1f/f Villin-Cre (Sult2b1∆IEC), and littermate control Sult2b1f/f mice were generated by crossing Cre-positive and Cre-negative mice. All mice were fed in laboratory animal facilities, given free access to food and water, under a controlled room temperature (22 ± 1 °C) and humidity (65 ± 5%) with a standard 12 h light-dark cycle. To establish DSS-induced colitis, mice received 2.5% (2.5 g/100 mL) DSS (MP Biomedicals, USA; MW: 36,000–50,000 Da, Cat. #160110) for six consecutive days in drinking water. The body weights of the mice were recorded every day. The FOB was tested on days 3 and 5 using a commercial kit (BASO, Zhuhai, China, Cat. #BA2020B), and colon length was determined at the end of the experiment. The mice were fed a diet supplemented with 0.004% CS or 0.005% cholesterol (Supplementary tables 3 and 4) for three days before DSS treatment and for the following six-day DSS challenge. The mice were administered the SREBP2 inhibitor betulin (100 mg/kg) (MedChemExpress LLC, MONMOUTH Junction, NJ, USA, Cat. # 473-98-3) via oral gavage for ten consecutive days, in addition to CS treatment. The control mice were administered vehicle. Male C57BL/6 interleukin-10 deficient (IL-10−/−) mice were obtained from GemPharmatech Co., Ltd. (Nanjing, China). IL-10−/− mice were born and raised in SPF facility until the day they were transferred to our clean facility (age 8–10 week).
Human biopsy collection
Following informed consent, biopsies were collected from volunteers attending endoscopy for routine colonoscopic screening (healthy) or as part of ongoing clinical care (patients with UC). We used tissues derived from patients with a proven histological diagnosis. In the immunochemical staining experiment, tissues were sampled from clinically inflamed colon and clinically non-involved regions of the same patients with UC. All patients gave informed consents for collection of tissue collection. All procedures were performed in accordance with institutional guidelines and were approved by Shanxi Provincial People’s Hospital Research Ethics Committee with the reference number 2019-70 for this study. The study was conducted in accordance with the criteria set by the Declaration of Helsinki.
Quantification of cholesterol sulfate by LC–MS
Cholesterol sulfate-d7 (CS-d7, Cat. #IR20993, ISOREAG; Shanghai ZZBio Co., Ltd.), as the internal standard (IS), was prepared at a concentration of 60 ng/mL in methanol. Stock standard of 20 mg of cholesterol sulfate (Cat. #700016P; Sigma-Aldrich) was dissolved in methanol to a volume of 100 mL as a storage solution (200 μg/mL). The CS storage solution was diluted in methanol to 2000, 1000, 400, 200, 100, 40, 20, 10, 4, 2, 1, and 0.4 ng/mL, and then mixed with the IS (the volume ratio was 1:1) to 1000, 500, 200, 100, 50, 20, 10, 5, 2, 1, 0.5, and 0.2 ng/mL, for generating a standard curve.
One hundred microliters of serum sample was mixed with 100 μL of IS (1.5 μg/mL). Protein precipitation occurred after adding 300 μL of methanol and vortexing for 1 min. Fifty milligrams of cecal content sample or colon tissue sample was homogenized in 300 μL of methanol and 200 μL of IS (1.5 μg/mL). After centrifugation at 14,000 g for 10 min, the upper phase was transferred to a new tube and diluted with methanol. LC-MS was used to quantitate the extracted cholesterol sulfate in this study by PANOMIX Biomedical Tech Co., LTD (Suzhou, China). Chromatography was performed on an Agilent Poroshell 120EC-C8, 2.1 mm × 100 mm, 1.9 μm HPLC column at 40 °C and subjected to gradient elution. The flow rate was 0.3 mL/min and the injection volume was 5 μL. The mobile phase consisted of solvent A (water with 5 mM formic acid) and solvent B (acetonitrile). The percentage of mobile phase B was 5% in the first 1 min, changed linearly from 5% to 95% over 3 min, maintained for 2 min, then linearly returned to 5% over 10 s, and maintained for 2.9 min. After separation, eluting compounds were ionized using ESI in the negative ion mode. The ionization source parameters were as follows: ion source temperature, 500°C; ion source voltage, −4500 V; collision gas, 6 psi; air curtain gas, 30 psi; atomized gas, 50 psi; auxiliary gas, 50 psi. Multiple reaction monitoring (MRM) scanning was performed. The ion pair used for CS quantitative analysis was 465/97; DP: −40, EP: −10, CE: −49, and CXP: −1. The ion pair used for CS-d7 quantitative analysis was 472/97; DP: −40, EP: −10, CE: −47, and CXP: −3. The linear regression equation was y = 0.028 * x + 0.0402; r = 0.9950. “x” denotes the CS concentration. “y” denotes the peak area ratio of CS versus CS-d7. The linear range was 0.5–1000 ng/mL and the limit of quantitation was 0.5 ng/mL.
A commercial desorption electrospray ionization (DESI) source (Prosolia, Indianapolis, U.S.) coupled to an LTQ Orbitrap Elite mass spectrometer (Thermo Fisher Scientific, San Jose, CA) was used for mass spectrometry imaging (MSI). The main parameters of DESI-MSI were as follows: acetonitrile-water (7:3, v/v) was constructed as the spray solvent system with the flow rate set at 2.0 μL/min. Nitrogen (1.0 MPa) was used for the nebulizing gas. A high voltage of −4 kV was applied onto the sprayer head to form the charged microdroplets. Given these conditions above, the sprayed droplets spot on the glass slide as well as corresponding lateral resolution can reach to approximate 200 μm. The impact angle and distance between the sprayer and substrate were 55° and 4.0 mm, respectively. The distance between the sprayer tip and the transport tube (coupled to the MS inlet) was 4.5 mm. The MS capillary temperature was set at 275 °C, and the S lens voltage was set at 55 V. The automatic gain control was set at 3E6 and maximum injection time set at 400 ms. By preliminary investigation, the DESI scan rate can be kept constant under these orbitrap parameter settings. DESI-MSI was performed in the negative ion mode within the range of m/z 400-500 that covered the deprotonated cholesterol sulfate ion at m/z 465.3044 [M-H]− (molecular formula: C27H46O4S). To guarantee the correction of the peak assignment above, the CID-MS/MS experiment was also conducted with the collision energy set at 15 V. Mouse colonic tissues were opened longitudinally, rolled into Swiss rolls and embedded in OCT. After thawing mounted on glass slides, tissue cryosections (10 μm) were fully dehydrated at room temperature before analysis. To compare the CS contents, a diluted series of CS standard-spiked simulative tissue was prepared to construct the quantitation curve according to previous reports41,42. After DESI-MSI data acquisition, Xcalibur (Thermo Scientific) was first employed for converting a batch of raw data files into cdf files. Then, the cdf files were imported into a massimager (Chemmind Technologies Co., Ltd, China) for further ion image reconstruction and CS quantitation. Briefly, the CS ion signal variation was first normalized by the total ion current (TIC, range from m/z 400–500) pixel-by-pixel. After the serial simulated intestine section were scanned under completely same DESI condition, a quantitation curve was constructed by fitting the normalized CS ion intensity with its content in each section. Then, the CS content in the real colonic tissue can be estimated by putting each pixel’s CS ion intensity into the quantitation curve.
Colon colitis scores
For the colon histological analysis, the middle colon segments were embedded in paraffin, sectioned at 5 μm, and stained with haematoxylin and eosin. The colitis score was determined using a standard histologic colitis score (Supplementary Table 5).
Qualification of cholesterol
Extraction of cholesterol from colonic tissues of patients or mice was performed as described in the above extraction method used for CS measurement. The cholesterol qualification of these extracts as well as of mouse serum was carried out using an Amplex® Red Cholesterol Assay Kit (Invitrogen, Carlsbad, CA, USA Cat. # A12216).
Preparation of the cholesterol solution used in the in vitro experiments
β-cyclodextrin (0.4 g) was dissolved in 1 mL of physiological saline solution and shaken for 30 min at 37 °C. 11.58 mg cholesterol dissolved in 200 μL CHCL3 was dried by nitrogen. The dried cholesterol was dissolved in 1 mL of the above 40% β-cyclodextrin solution to form cholesterol stock solution (30 mM) and shaken overnight at 37 °C. The cholesterol stock solution was shaken for at least 2 h at 37 °C before use.
HEK293T cell line, human colon carcinoma cell lines HT-29, LOVO, SW480, HCT116, and SW1116 were from National Collection of Authenticated Cell Cultures (Shanghai, China, Cat. #SCSP-502, #SCSP-5032, #SCSP-514, #SCSP-5033, #TCHu99, #TCHu174, respectively).
Construction of the SULT2B1
−/− HT-29 cell line
To construct the SULT2B1 knockout HT-29 cell line, an online sgRNA design tool based on the clustered, regularly interspaced, short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (CAS9) (CRISPR/CAS9) system (http://crispr.mit.edu/) was utilized. Two sgRNA oligos targeting exon 2 of the SULT2B1 gene were selected as follows: 5′-CGAGTACAGGCCGACGGGGAAGG-3′ (sgRNA-A) and 5′-CGGAGAACACCCAAGATGTGCGG-3′ (sgRNA-B). sgRNA-A and sgRNA-B were cloned into the pSpCas9n(BB)−2A-Puro vector (pX462 vector, Addgene plasmid # 48141). Then, the constructed plasmids containing sgRNA-A and sgRNA-B were co-transfected into HT-29 cells with Lipofectamine 3000 (Invitrogen, CA, USA). SULT2B1 knockout was confirmed by western blotting and gene sequencing.
RNA sequencing (RNA-seq)
RNA purified from tissues and cells as indicated was converted into cDNA libraries using the TrueLib mRNA library prep kit for Illumina (ExCell. Bio, China). The library quality was assessed using an Agilent 2100 Bioanalyzer. Sequencing was performed on an Illumina HiSeq X Ten sequencer at Annoroad Gene Technology Co., Ltd. (Beijing, China).
Mice were injected with EdU (5 mg/kg) intraperitoneally 6 h before sacrifice. The colon tissues were embedded in paraffin for the subsequent staining. Cell proliferation was assessed by the Cell-Light EdU Apollo 567 In Vitro Kit (RiboBio, Guangzhou, Guangdong, China. Cat. #C00003), according to the manufacturer’s instructions. In brief, the tissue was permeabilized with 0.5% Triton X-100 and reacted with 1 × Apollo reaction cocktail for 30 min. Subsequently, the DNA contents of the cells were stained with DAPI for 30 min and visualized under a fluorescence microscope (Leica Microsystems GmbH, DMI6000B).
Cells were seeded on coverslips placed in 24-well plates. After additional cholesterol or cholesterol sulfate treatment, cells were washed with PBS 3 times and fixed in 4% paraformaldehyde for 20 min at room temperature. Cells were then incubated with 50 μg/mL filipin III (Thermo Fisher Scientific) for 1 h at room temperature. The staining was examined with ultraviolet excitation using a confocal microscope (ZEISS, Germany, LSM 7 10).
RNA extraction and RT-qPCR
Total RNA was extracted from cells by TRIzol reagent (Invitrogen, Carlsbad, CA, USA Cat. #12183555) following the manufacturer’s instructions and used to generate cDNA using the ReverTra Ace qPCR RT Kit (TOYOBO, Cat. #QPK-201). SYBR green (Yeasen, Shanghai, China, Cat. #11203ES03)-based qPCR was performed with the primers listed in Supplementary Table 6.
Different samples were lysed in RIPA buffer containing a protease and phosphatase inhibitor cocktail (Bimake, Cat. # 510026, Cat. # 410043). The protein concentrations were detected using a BCA protein quantification kit (Biocolor, Cat. # K3000). Then, the same amounts of protein were separated by SDS-PAGE and transferred onto PVDF membranes. The membranes were blocked with 3% BSA in TBST and probed with anti-human SREBP2 (ABclonal, Cat. #A13049, 1:1000), anti-human SULT2B1 (R&D Systems, Cat. #AF6174, 1:1000), anti-mouse SULT2B1 (Santa Cruz Biotechnology, Cat. #sc-166423, 1:1000), anti-human NPC2 (Proteintech, Cat. #19888-1-AP, 1:1000), Monoclonal ANTI-FLAG® M2 antibody (Sigma-Aldrich, Cat. #F1804, 1:2000), HA-tag rabbit mAb (Cell Signaling Technology, #3724), and anti-β-actin (Proteintech, Cat. #66009-1, 1:2000) or anti-HSP90 (Proteintech, Cat. #60318-1-Ig, 1:2000) for the loading control. Then, the membranes were washed with PBST and incubated with peroxidase-conjugated goat anti-rabbit secondary antibodies (Yeasen, Cat. #33101ES60, 1:2500), peroxidase-conjugated goat anti-mouse secondary antibodies (Yeasen, Cat. #33201ES60, 1:2500) or peroxidase-conjugated rabbit anti-sheep secondary antibodies (Abcam, Cat. #ab6747, 1:2500). Immunoreactivity was observed with an ECL Western Blotting Substrate kit (Tanon, Shanghai, China, Cat. #180-501) using a gel imaging system (Tanon-4200, Shanghai, China).
The cells were seeded on coverslips and treated as indicated. The slides were incubated with anti-human SREBP2 (ABclonal, Cat. #A13049, 1:200) or anti-human SLC10A6(Invitrogen, Cat. PA5-53468, 1:200) at 4 °C overnight and then incubated with a fluorescent secondary antibody at room temperature for 90 min. The nuclei were visualized using DAPI. Representative images were acquired with a fluorescence microscope (Leica Microsystems GmbH, DMI6000B).
Human biopsies were embedded in paraffin. Sections were incubated with primary antibodies, namely, anti-human SULT2B1 (R&D Systems, Cat. #AF6174, 1:100) at 4 °C overnight, followed by incubation with a peroxidase-conjugated rabbit anti-Sheep secondary antibody (Abcam, Cat. #ab6747, 1:500). at room temperature for 90 min, visualization with DAB, and counterstaining with haematoxylin. The H-SCORE calculated by multiplying the quantity and intensity scores was used to semi-quantify SULT2B1 staining for each tissue sample. Scoring criteria for quantity (the estimated proportion of staining in intestinal epithelial cells that was positively stained) were as follows: score = 0, none; score = 1, 1–25%; score = 2, 26–50%; score = 3, 51–75%; and score = 4, 76–100%. An intensity score represented the average intensity of the positive cells, as follows: 0 (none); 1 (weak); 2 (intermediate); and 3 (strong). The proportion and intensity scores were then multiplied to obtain a total score, which could range from 0 to 12. Each sample was evaluated in a blinded manner by two senior pathologists, and conflicting cases were reanalysed by a third pathologist.
Generation of a stable SREBP2 knockdown HT-29 cell line
The small hairpin RNA sequence against human SREBF2 was obtained from Sigma-Aldrich. The target shRNA sequence was 5′‐CCGGCCTCAGATCATCAAGACAGATCTCGAGATCTGTCTTGATGATCTGAGGTTTTT‐3′. It was cloned into the mCherry PTSB-SH-mCherry-2A-NEO vector. Lentivirus‐encoded shRNA against human SREBF2 and the control were prepared. Lentiviral particles were generated through transient transfection of retroviral vectors into the 293T packaging cell line. Supernatants were harvested two days after transfection. HT-29 cells were infected with lentiviral particles and selected against G418. Successful knockdown was confirmed by RT-qPCR and western blot analysis.
Luciferase Reporter Assay
The 6 × SREBP2 response element (SRE) consensus sequence (ATCACCCCAC- ATCACGCCAC-GTCACCCCAT-ATCACCCCAC-ATCACGCCAC-GTCACCCCAT) were cloned into the pCDH-NC-EF1-PURO-F2A-RLUC Lenti-vector. HEK293T cells were cotransfected with a lentiviral plasmid system containing SRE-Reporter-LUC or control-LUC as transfer vector, PMD2.G, and PxPSA2.0 as packaging vectors. The lentivirus-containing supernatants were collected, filtered at 48 h and 72 h after transfection. HT-29 cells were infected with the respective lentivirus. The infected cells were seeded in 24-well plates, and then treated with CS or DMSO. After 24 h, Cells were lysed in passive lysis buffer (PLB). Luciferase activity was measured by using a luciferase reporter assay (Promega, Cat. #E1960).
Transient transfection of cells
pCDH-CMV-HA-SREBF2 is an expression plasmid that encodes, in sequential order from the NH2-termius, a HA epitope tag (YPYDVPDYA) and human full-length SREBF2. pCDNA3-SCAP is an expression plasmid that encodes human full-length SCAP. The above plasmids were generated by standard cloning methods. All plasmids were verified by sequencing the entire coding region. 1.5 µg pCDH-CMV-HA-SREBF2 and 0.75 µg pCDNA3-SCAP were transiently co-transfected into 293T cells cultured in 6-well plates using polyethylenimine (PEI). After 6 h, the serum-free medium was changed to fresh complete medium. After 24 h, CS in the final concentration 25 μM or 50 μM was added to the medium. After further incubation for 6, 12, or 18 h, cells were harvested, and equal fractions of whole cell lysates were subjected to western-blot analysis of SREBP2 (anti-HA) and β-actin.
Cellular Thermal Shift Assay (CETSA)
For a CETSA in living HT-29 cells, cells were seeded in 60 mm dishes and exposed to CS or DMSO at the indicated concentrations for 1 h in the cell culture incubator. Following incubation, the cells were washed with PBS and harvested in 500 μL of PBS supplemented with complete protease inhibitor cocktail. The cell suspensions were freeze-thawed three times using lipid nitrogen. The lysates were centrifuged at 20,000 x g for 20 min at 4 °C. The supernatants were transferred to new tubes. For a CETSA, the cell lysates were divided into smaller aliquots and heated individually at different temperatures for 3 min followed by cooling for 3 min at room temperature. In order to separate the soluble fractions from precipitates, the heated lysates were centrifuged at 20,000 x g for 20 min at 4 °C. The supernatants were transferred to new microtubes and analyzed by SDS-PAGE followed by western blot analysis.
Data are presented as the mean ± standard error of the mean (SEM). GraphPad Prism 8.0 Software (San Diego, CA, USA) was used for calculations, statistical analyses, and graphic generations. Statistical analyses comparing two parameters (between treatments or genotypes) were conducted using the two-tailed Student’s t-test. Statistics for multiparameter analyses were determined by one-way analysis of variance (ANOVA) followed by the recommended post hoc tests in GraphPad Prism 8.0 Software. Two-way analysis of variance (ANOVA) was used for time-series experiments. For in vitro assays, experiments were repeated at least three times. P < 0.05 was considered significant.
Further information on research design is available in the Nature Research Reporting Summary linked to this article.