TNF-α-mediated m6A modification of ELMO1 triggers
directional migration of mesenchymal stem cell in ankylosing

Study approval

This study conforms to the Declaration of Helsinki and was approved by the Ethics Committee of the Eighth Affiliated Hospital, Sun Yat-Sen University, Guangzhou, China. The experiments on mice were approved by the Institutional Animal Care and Use Committee of Sun Yat-Sen University, Guangzhou, China. All experimental procedures on mice were carried out in strict adherence to the rules and guidelines for the ethical use of animals in research.

Cell isolation and culture

A total of 15 patients with AS and 15 healthy controls were recruited for this study. All the AS patients were diagnosed according to the 1984 New York modified criteria56. After being informed of the possible risks and complications of bone marrow puncture, all the AS patients and healthy controls signed informed consent forms. The characteristics of the study subjects are presented in Supplementary Table 1. MSC were immediately isolated and purified from bone marrow samples using density gradient centrifugation. MSC were resuspended in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum and then seeded into 25-cm2 flasks and cultured at 37 °C in 5% CO2. The cells in suspension were removed, and the medium was replaced every 3 days thereafter. When the culture reached 90% confluence, MSC were digested using 0.25% trypsin containing 0.53 mM EDTA and reseeded in new flasks; these cells were expanded and used for experiments.

HEK293T cells were cultured in high-glucose DMEM supplements with 10% FBS at 37 °C under 5% CO2. At 80–90% confluence, HEK293T cells were digested with 0.25% trypsin containing 0.53 mM EDTA and reseeded in new flasks.

MSC migration assays in a Transwell system

MSC migration assays were performed using Polycarbonate Membrane Transwell® Inserts (8.0-μm pores, 24-well plate; Corning). Briefly, a total of 2 × 104 MSC in 100 μl DMEM were seeded in the upper chambers, and 600 μl DMEM without or with TNF-α at a concentration ranging from 10 to 100 ng/ml was added to the lower chambers. In the assays of Fig. 1a and Supplementary Figs. 1, 2 × 105 macrophages differentiated from CD14+ monocytes isolated from peripheral blood samples were suspended in 600 μl DMEM and then seeded in the lower chambers without or with 0.5 μg/ml anti-TNF-α, 0.05 μg/ml anti-IL-17 or 0.5 μg/ml anti-IL-23 neutralizing antibody. After coculturing for 24 h, the upper chambers were washed three times with phosphate-buffered saline (PBS) and fixed with 4% paraformaldehyde. Then, the upper chambers were stained with 0.1% crystal violet for 15 min. The migratory MSC on the lower side of the chamber were photographed and counted as the mean number of cells per 10 random fields for each chamber in two separate experimenters.

Wound-healing assays

Wound-healing assays were performed using IBIDI® culture-Inserts 2 Well for self-insertion. Each insert was placed in a well of a 12-well plate. MSC (5 × 104) in 200 μl DMEM were seeded in a well of the insert. The inserts were then removed, and DMEM without or with TNF-α at a concentration ranging from 10 to 100 ng/ml was added. After 12 h of migration, the migratory area was calculated using ImagePro Plus 6.0. The percentage of the migratory area was defined as the ratio of the MSC migratory area at 12 h to the primary wound area created by the insert at 0 h.

μ-slide chemotaxis assay

Directional migration assays were performed using IBIDI® μ-slide chemotaxis chambers. An MSC suspension (6 μl, 3 × 106 cells) was seeded into the center channel of μ-slide chambers. After 12 h of incubation, DMEM containing TNF-α was added to the left medium reservoir, and DMEM without TNF-α was added to the right one reservoir. The cell migration track was imaged every 10 min over 24 h with the BioTek Lionheart™ FX Automated Live Cell Imager with Augmented Microscopy™. Cell tracking analysis was performed using the Manual Tracking plugin for ImageJ 1.51 and the Chemotaxis and Migration Tool according to the IBIDI protocol.

Cell immunofluorescence assay

For the cell immunofluorescence assay, 5 × 104 MSC in 1 ml DMEM were seeded in each well of a 12-well plate and then treated with or without 100 ng/ml TNF-α for 24 h. Then, the cells were washed with PBS, fixed with 4% paraformaldehyde and incubated with 1% Triton X-100. MSC were blocked in goat serum and then incubated with an anti-ELMO1 antibody (1:500) overnight at 4 °C. The cells were washed and incubated with a fluorescein-labeled secondary antibody (Alexa Fluor® 647; 1:3000) for 1 h, followed by staining with DAPI (Thermo Fisher) for 10 min. All images were obtained using an LSM 5 Exciter confocal imaging system (Carl Zeiss).

In vivo migration assay

MSC were transfected with a lentivirus encoding luciferase (OBiO Technology) as described below (MSC/Fluc). Mice were housed at the Laboratory Animal Center of Sun Yat-Sen University under specific pathogen-free conditions, with a 12-h light/dark cycle in a temperature ((22 ± 2 °C) and humidity-controlled room (60%) with free access to water and food. Eight-week-old BALB/c-nu/nu female mice (Laboratory Animal Center of Sun Yat-Sen University) were anesthetized via intraperitoneal injection of ketamine and xylazine. A total of 5 × 105 MSC/Fluc in 20 μl DMEM were injected subcutaneously into the dorsal sides of the BALB/c-nu/nu mice. Twenty microliters of TNF-α at a concentration of 500 ng/ml was administered at four sites around the MSC injection point at a distance of 0.5 cm. The injected MSC/Fluc were observed using a Xenogen IVIS Spectrum system (Caliper Life Sciences, Inc.) on days 0, 1, 3, and 5 after intraperitoneal injection of 3 mg D-Luciferin potassium. The bioluminescence area was analyzed using Living Image 2.12 and ImagePro Plus 6.0. On day 7 after injection, the mice were sacrificed, and the tissues at the TNF-α injection sites were obtained for immunohistochemical analysis as described below.

RNA extraction, reverse transcription, and quantitative real-time PCR

Total RNA was isolated from MSC using TRIzol and transcribed into cDNA using a PrimeScriptTM RT reagent kit according to the protocols. Quantitative real-time PCR was performed on the LightCycler®480 PCR System (Roche) using TB Green® Premix Ex TaqTM. The relative expression levels of each gene were analyzed using the 2Ct method. The forward and reverse primers for each gene are shown in Supplementary Table 2.

RNA sequencing and data analysis

MSC were plated and treated with or without 100 ng/ml TNF-α for 24 h. RNA was extracted as described above. cDNA library construction and sequencing were performed by the Beijing Genomics Institute using the BGISEQ-500 platform. The sequencing data were filtered with SOAPnuke 1.5.2, and the clean reads were mapped to a reference genome using HISAT2 2.0.4. After alignment using Bowtie2 2.2.5, the expression level of each gene was calculated by RSEM 1.2.12, and differential expression analysis was performed using DESeq2 1.4.5 with the parameters fold change ≥2 and adjusted P value ≤0.001. The sequencing data analysis, including heatmap clustering, principal component analysis (PCA), Venn diagram creation, gene ontology (GO) analysis, and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, were performed using BGI Dr. Tom 2.0.

Protein extraction and western blot

MSC were lysed in RIPA lysis buffer containing protease inhibitors and phosphatase inhibitors. The lysates were centrifuged at 1100 × g at 4 °C for 30 min, and the lysate protein concentrations were determined using a Pierce BCA protein assay kit. The immunoprecipitants was obtained in Co-IP assay as described below. After boiling with sample loading buffer, equal amounts of protein extracts or the immunoprecipitants were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis and subsequently transferred to polyvinylidene fluoride (PVDF) membranes (Millipore). The PVDF membranes were blocked and incubated overnight at 4 °C with primary antibodies against GAPDH, ELMO1, ELMO2, ELMO3, METTL3, METTL14, FTO, ALKBH5, WTAP, DOCK1, DOCK2, DOCK4, DOCK5, DOCK8, or Rac1 (1:1000). After washing three times, the PVDF membranes were incubated with a horseradish peroxidase (HRP)-conjugated secondary antibody (1:3000). Specific antibody-antigen complexes were detected using Immobilon Western Chemiluminescent HRP Substrate. The mean intensity ratio was determined and analyzed using a UVP ChemStudio PLUS system and ImagePro Plus 6.0.

Rac1 activation assay

Rac1 activation in MSC was detected with the Active Rac1 Detection Kit through a GST pulldown assay according to the kit protocol. Briefly, protein was extracted from MSC treated with or without TNF-α as described above. Glutathione resin containing agarose beads was added to the spin cup with a collection tube and washed three times. GST-Human PAK1-PBD was added to the spin cup, followed by protein lysis and incubation for 1 h at 4 °C with gentle rocking. After washing, the spin cup containing agarose beads was incubated with reducing sample buffer at room temperature for 2 min, and the eluted sample was collected by centrifugation. GTP-Rac1 in the eluted protein fraction was detected using Western blot analysis as described above. MSC-derived protein without GST pulldown was used to detect the total Rac1 and GAPDH levels.

Coimmunoprecipitation (Co-IP) and liquid chromatograph (LC)-mass spectrometry (MS)/MS

The Co-IP assay was performed using the Dynabeads™ Protein A Immunoprecipitation Kit according to the kit protocol. Briefly, an MSC protein extract was incubated with an anti-ELMO1 antibody (1:100) or IgG control at 4 °C overnight, followed by the addition of protein G agarose beads and incubation for another 4 h. Then, the immunoprecipitants were collected for LC-MS/MS and western blotting. LC-MS/MS was performed by Shanghai Applied Protein Technology Co., Ltd. Data were analyzed using Proteome Discoverer 1.4 against the UniProt database (

Lentivirus construction and transfection

Three ELMO1, METTL14, DOCK1, DOCK4, DOCK5, YTHDF2, and YTHDF3-specific siRNAs were designed and synthesized by OBiO Technology. The sequences are shown in Supplementary Table 3. The siRNA was used to transfect MSC using a Lipofectamine RNAi MAX according to the protocol, and the siRNA with best knockdown efficiency was chosen to construct a lentivirus encoding a short hairpin RNA (shRNA) specific for ELMO1 (Lv-ELMO1) or METTL14 (Lv-M14) by OBiO Technology. ELMO1- and METTL14-overexpression lentiviruses (OE-ELMO1 and OE-M14, respectively) and their vector controls were also constructed by OBiO Technology. Lentiviruses (109 TU/mL) and 5 μg/mL polybrene were added to DMEM and incubated with MSC for 24 h at a multiplicity of infection of 50. Experiments were performed on day 4 after transfection. Adenoviruses encoding an shRNA specific for ELMO1 (Av-ELMO1) or control adenoviruses (Av-NC) for mouse experiments were also constructed and purchased from OBiO Technology.

Cell proliferation assay

MSC transfected with Lv-ELMO1 (or OE-ELMO1) and its NC control were seeded in 96-well plates. Cell proliferation ability was detected using Cell Counting Kit-8 according to the protocol. Medium without cells were used as negative controls.

m6A RNA immunoprecipitation (RIP)

The m6A RIP assay was performed using the Magna MeRIP™ m6A Kit according to the manufacturer’s instructions. Briefly, RNA was extracted and then chemically fragmented into fragments of 200 nucleotides or less. The RNA fragments were incubated with anti-m6A antibody- or IgG-conjugated Protein A/G magnetic beads at 4 °C overnight. Then, the magnetic beads were collected, and the bound m6A-modified RNA was eluted for qRT-PCR analysis as described above. Equal amounts of nonimmunoprecipitated RNA fragments were used as the input control. The fold enrichment of each target gene was calculated with the formulas below.

$$\Delta {{{{{{\rm{CT}}}}}}}_{{{{{{\rm{target}}}}}}\,{{{{{\rm{gene}}}}}}} = {{{{{{\rm{CT}}}}}}}_{{{{{{\rm{target}}}}}}\,{{{{{\rm{gene}}}}}}}-{{{{{{\rm{CT}}}}}}}_{{{{{{\rm{Input}}}}}}}\\ \Delta {{{{{{\rm{CT}}}}}}}_{{{{{{\rm{IgG}}}}}}} = {{{{{{\rm{CT}}}}}}}_{{{{{{\rm{IgG}}}}}}}-{{{{{\rm{C}}}}}}{{{{{{\rm{T}}}}}}}_{{{{{{\rm{Input}}}}}}}\\ \Delta \Delta {{{{{\rm{CT}}}}}} = \Delta {{{{{{\rm{CT}}}}}}}_{{{{{{\rm{target}}}}}}\,{{{{{\rm{gene}}}}}}}-\Delta {{{{{{\rm{CT}}}}}}}_{{{{{{\rm{IgG}}}}}}}\\ {{{{{\rm{Fold}}}}}}\,{{{{{\rm{enrichment}}}}}} = {2}^{-\Delta \Delta {{{{{\rm{CT}}}}}}}$$

The relative fold enrichment was calculated by normalizing to the fold enrichment of HC-MSC without TNF-α stimulation.

RNA stability assays

MSC were seeded in a 12-well plate and treated with actinomycin D at a concentration of 20 μg/ml for 0, 1, 2, and 3 h. After treatment, MSC RNA was immediately extracted, and qRT-PCR assays were performed as described above. The turnover rate and half-life of each target gene were calculated as described in a previous study57.

Cross-linking and RNA immunoprecipitation (CLIP) assay

MSC were treated with 150 mJ/cm2 UVC irradiation at 254 nm for 40 s. The CLIP assay was then performed using the EZ-Magna RIP™ RNA-Binding Protein Immunoprecipitation Kit according to the manufacturer’s instructions. Briefly, MSC were lysed using RIP lysis buffer, and the lysate was incubated with antibodies against METTL14, YTHDC1, YTHDC2, YTHDF1, YTHDF2, YTHDF3 (1:50) or IgG control with protein A/G magnetic beads at 4 °C overnight. The magnetic beads were immobilized, and then the immunoprecipitant containing specific protein with its bound RNA was eluted and treated with proteinase K. The remaining RNA was extracted for qRT-PCR analysis as described above. Equal amounts of nonimmunoprecipitated RNA fragments were used as the input control. The %Input was calculated with the formulas below.

$$\Delta {{{{{{\rm{CT}}}}}}}_{{{{{{\rm{target}}}}}}\,{{{{{\rm{gene}}}}}}\,{{{{{\rm{or}}}}}}\,{{{{{\rm{IgG}}}}}}} = {{{{{{\rm{CT}}}}}}}_{{{{{{\rm{target}}}}}}\,{{{{{\rm{gene}}}}}}\,{{{{{\rm{or}}}}}}\,{{{{{\rm{IgG}}}}}}}-{{{{{{\rm{CT}}}}}}}_{{{{{{\rm{Input}}}}}}}\\ \% {{{{{\rm{Input}}}}}} ={2}^{-\Delta {{{{{\rm{C}}}}}}{{{{{\rm{T}}}}}}}$$

Dual-luciferase reporter assay

A METTL14 expression vector and its mutant vector (METTL14-R298P), as well as luciferase reporter vectors containing the C-terminal DNA fragment of ELMO1 or its mutant with mutations in m6A modification sites (A replaced by T), were synthesized by OBiO Technology. The possible m6A modification sites were predicted by the m6AVar database58. HEK293T cells were seeded in 12-well plates, and then 1.5 μg luciferase reporter vector, 1.5 μg METTL14 or its mutant vector, and 1.5 μg Renilla luciferase reporter vector were cotransfected into HEK293T cells using the Lipofectamine 3000 Transfection Kit according to the manufacturer’s instructions. After 48 h of transfection, the luciferase activity was detected using the Dual-Luciferase® Reporter Assay System Kit (Promega) according to the protocol. Relative Fluc/Rluc activity was calculated by normalizing the activity of firefly luciferase to that of Renilla luciferase.

Mouse induction, treatment and, scoring

Male SKG mice were purchased from CLEA Japan, Inc. and housed as described above. All mice were handled in accordance with the guidelines for animal care approved by the Institutional Animal Care and Use Committee of Sun Yat-Sen University. Disease was induced at 8 weeks of age using 3 mg curdlan administered by intraperitoneal injection. The SKG mice were randomly divided into three groups: a PBS group, an Av-NC group, and an Av-ELMO1 group. The SKG mice in the Av-ELMO1 group were treated with 5 × 1010 Av-ELMO1 via intravenous tail vein injection at the time of disease induction, and the SKG mice in the Av-NC group or PBS group were separately treated with equal amounts of control adenoviruses or PBS. Clinical features in the mice were monitored weekly by two independent observers who were blinded to the treatment groups. The score criteria were defined as previously reported59: 0 = no swelling or redness, 0.1 = swelling or redness of the digits, 0.5 = mild swelling and/or redness of the wrist or ankle joints, and 1 = severe swelling of the larger joints. At week 8 after induction and treatment, the mice were sacrificed, and the tissues were obtained for micro-CT examination, hematoxylin and eosin (H&E) staining, immunohistochemical analysis, and immunofluorescence as described below.

Micro-CT scanning

Micro-CT was performed to analyze the structures of the spine and ankle. Obtained tissues were fixed with 4% polyoxymethylene and then scanned using a Siemens Inveon CT scanner with a resolution of 19 μm. The figure data were analyzed using RadiAnt DICOM Viewer 5.0.2 software.

H&E staining

Harvested spine and ankle tissue samples were fixed with 4% polyoxymethylene for 24 h, decalcified in 20% EDTA for 14 days and then embedded in paraffin for sectioning. The sections were stained with hematoxylin for 5 min. After washing with PBS for 10 min, the sections were stained with eosin for 3 min. Then, the sections were dehydrated and observed using a microscope.

Tissue immunohistochemical assay

For the SKG mouse tissue immunohistochemical assay, sections were incubated in 10 mM citrate buffer and microwaved at 750 W for 30 min for antigen retrieval. The sections were treated with 3% H2O2 for 20 min and blocked with 5% normal goat serum for 1 h. Then, the sections were separately incubated with anti-CD68, anti-TNF-α or anti-CD105 (1:100) antibodies at 4 °C overnight. Secondary antibody (1:500) incubation and color development were performed using the SP Rabbit & Mouse HRP DAB Kit according to the kit protocol. For mouse tissue from the in vivo migration assay, the immunohistochemical assay was performed as described above except that an antibody specific for human HLA Class 1 ABC was used.

Tissue immunofluorescence assay

For the human enthesis tissue immunofluorescence assay, nine AS patients and 9 non-AS patients were recruited. Sites of ossifying enthesis were confirmed via presurgical image analyses and visual observations during surgery. Ossifying tissues were obtained during lumbar spine surgeries. The characteristics of the study subjects are presented in Supplementary Table 4. The tissue samples obtained were successively fixed, decalcified, and embedded in paraffin. Sections were deparaffinized, hydrated, and incubated in 1% Triton X-100/PBS. After antigen retrieval in citrate buffer and blocking in goat serum, the sections were incubated with anti-CD105, anti-ELMO1 or anti-METTL14 antibodies (1:100) overnight at 4 °C. The sections were washed and incubated with a fluorescein-conjugated secondary antibody (1:400) for 1 h and then with DAPI for another 10 min. All images were obtained using an LSM 5 Exciter confocal imaging system (Carl Zeiss). The mean fluorescence intensity was quantified using the ImagePro Plus 6.0.

Statistical analysis

All the results were determined based on at least three separate in vitro experiments containing at least triplicate samples. The Shapiro–Wilk normality test was used to check the normality of the data, and data with a Shapiro–Wilk test P > 0.05 were considered to fit a normal distribution. The two group comparisons were performed using a 2-tailed Student’s t-test, and comparisons of three or more different groups were performed by a one-way ANOVA, followed by Bonferroni’s post hoc comparisons. Data are expressed as the means±standard deviations. Statistical analysis was performed with SPSS (SPSS Inc.). The n values indicate the numbers of individuals in each experiment. P-values less than 0.05 were considered statistically significant.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Read original article here.