Figure legends
Figure 1 Effects of AZM on RA FLSs. (A) Following exposure to TNF-α and IL-1β for 6 h, RA FLSs (n=3) were treated by AZM or Vehicle for another 24 h. IL-6 (A) , IL-8 (B) , MMP-1(C) and MMP-3 (D) levels in the supernatant of RA FLSs were analysed by ELISA. The migration (E) and invasion(F) abilities were assessed in AZM or Vehicle treated RA FLSs from six different patients (3 males and 3 females). Data represent independent experiments performed in triplicate, and six different fields were selected for cell counting (graph below). RA FLSs passing through the polycarbonate membrane with ECM coating shows that cell invasion requires the ECM proteolysis step in addition to migration. RA FLSs (n=3) were treated by AZM or Vehicle for 24 h, followed by TNF-α or IL-1β stimulation for 6 h. CXCL9 (G) , CXCL10 (H)levels in the culture supernatant were measured by ELISA. (I)After treatment as indicated in (A-D), cell-free RA FLSs supernatant was collected and used as a chemo-tactic source for healthy donor peripheral blood leucocytes (n=3) in a transwell migration system for 6 h. The number of migrating leucocytes in the lower compartment was counted. *P <0.05, **P <0.01 and ***P <0.001 compared with Vehicle. (J) The production of VEGF decreased when RA FLSs were exposed to AZM.(K) Tube formation assay was applied to determine the angiogenic ability of HUVECs after treatment with conditioned medium of AZM or its Vehicle treated RA FLSs. Data in A-D, G and H are expressed as the mean of three samples ± SD and represent three independent experiments. *P <0.05, **P <0.01 and ***P <0.001 compared with Control. IL-1β, interleukin-1β; TNF-α, tumour necrosis factor-α; AZM, azithromycin; Ctrl, control.
Figure 2 Anti-arthritis effects of AZM on CIA. Mice immunized with CII were randomly divided into 4 groups (n=6 mice for each group and time point) and administered AZM or Vehicle at the indicated doses twice a day after the initial immunization. The data are representative of four independent experiments with similar results. (A)Arthritis scores were monitored once per five days. (B) Hind paw thickness was calibrated from the 21st day following the first immunization. AZM (low): #P <0.05,##P <0.01,###P <0.001 compared with Vehicle. AZM (medium): *P <0.05, **P <0.01, ***P <0.001 compared with Vehicle. AZM (high):&P <0.05,&&P <0.01,&&&P <0.001 compared with Vehicle.(C) Fore paw (upper) and hind paw (lower) photographs obtained on day 42 from mice with CIA from the day of first immunization.(D) Representative histology images by H&E staining about interphalangeal joint (upper) and ankle joint (lower) obtained on day 67 from mice with CIA with the indicated AZM treatment. Pathological changes, including synovial proliferation (red arrowhead) and joint destruction (yellow arrowhead), are shown. (E) Inflammation, hyperplasia, cartilage degradation and bone destruction were measured through a scoring system (n= 12 mice per group). (F)Representative micro-CT images of hind paws and interphalangeal joints (red square). (G) BV/TV, Tb. BMD, Tb. Th, Tb. N and Tb. Sp in the distal tibia were assayed by micro-CT and 3D reconstruction. E and G: ns: not significant, *P <0.05, **P <0.01 and ***P <0.001 compared with Vehicle group. CIA, collagen-induced arthritis; BV/TV, Bone volume fraction; Tb. BMD, trabecular bone mineral density; Tb. Th, trabecular thickness; Tb. N, trabecular number; Tb. Sp, trabecular separation.
Figure 3 Functional enrichment analysis of differentially expressed genes. (A) Hierarchical clustering of the dysregulated mRNAs in RA FLSs. The expression values were represented by a color scale. The intensity increased from blue (relatively lower expression) to red (relatively higher expression). Each column represented one tissue sample, and each row represented a single mRNA. (B) GO analyses of the host genes of differentially expressed mRNAs and GO annotations of the host genes of differentially expressed mRNAs. The bar plot presented the enrichment scores (−loge[p value]) of the top 10 significantly enriched GO terms in biological processes, cellular components and molecular functions. (C) KEGG analyses of the host genes of differentially expressed mRNAs and KEGG annotations of the host genes of differentially expressed mRNAs. The bubble diagram presented the enrichment scores (−loge[p value]) of the top 10 significantly enriched KEGG terms in molecular function and signal pathway. (D) After stimulation with AZM for 24 h, total RNA was extracted from RA FLSs and subjected to qRT-PCR for the assessment of CHOP mRNA levels. qRT-PCR data are expressed as the mean ± SD of six samples (3 males and 3 females) from two independent experiments. NS: not significant, *P <0.05 and ***P <0.001 compared with vehicle. (E) Using similar treatments as (D) , total and phosphorylated levels of PERK and elF2α, and the increased expression of IRE1α, ATF4 were analysed by western blot. Western blot data are representative of two independent experiments from six different patients (3 male and 3 female) with similar results. (F) Under the stimulation of TNF-α and IL-1β, similar treatment as (D) was used, total and phosphorylated levels of p38 and Akt, and the increased expression of CREB3L2 were analysed by western blot. Western blot data are representative of two independent experiments from six different patients (3 male and 3 female) with similar results. (G) RA FLSs (n = 3) were treated with AZM or its vehicle for 24 h, Annexin X staining were applied to detect the cell apoptosis of RA FLSs (n = 3). The statistical significance of differences between AZM and vehicle-treated groups was determined. Data represent the mean ± SD of three independent replicates of three samples. **P <0.01 and ***P <0.001 compared with Control. DMSO, Dimethyl Sulfoxide; Tm, tunicamycin; Tg, thapsigargin.
Figure 4 GRP78 is a novel target of AZM. (A) Coomassie blue staining of DARTS assay. The band with molecular weight around 80 kDa protected by AZM was indicated by arrow. (B) DARTS and Western blot to confirm AZM’s binding targets. (C and D) CETSA melt response. (E) DARTS assay with serial deletion constructs encoding Flag-tagged mutants of GRP78. GRP78 (aa 1–750), GRP78 (aa 126–750), GRP78 (aa 406–750) and GRP78 (aa 1–479). 293T cells were transfected with Flag tagged mutants of GRP78 plasmids, as indicated. DARTS assay samples were detected by Flag antibody. (F)Potential targeting sites of AZM towards GRP78. (G) DARTS assay for Asp-178 point mutant of cPLA2. The Asp-178 of cPLA2 was substituted with Ala505. 293T cells were transfected with the point mutant plasmid and DARTS was performed. Point mutated cPLA2 was detected by Flag antibody. (H) The double reciprocal diagram of ATP concentration and luminescence intensity under different concentrations of AZM showed that the treatment of AZM might be related to ATP.(I) The enzyme activity of GRP78 was measured in the presence of AZM or HA15. The initial concentration of Pronase was 10 mg/kg. ns: not significant, **P <0.01 and ***P <0.001 compared with Vehicle. Temp, Temperature.
Figure 5 AZM inhibits GRP78 activity through binding to its catalytic domain. (A) RA FLSs were stimulated with AZM for 24 h, immunoprecipitation and immunoblotting were performed on total protein with anti-GRP78 or anti- PERK, IRE1α or ATF6α antibodies. (B)Equal aliquots of nuclear and membrane protein from AZM or its vehicle treated RA FLSs with GRP78 overexpression or not were pooled (total, 30 mg) and subjected to SDS-PAGE and immunoblot analysis for the indicated protein. A and B: The results are representative of four samples (2 male and 2 female) from independent experiment.
Figure 6 AZM activates UPR and promote apoptosis via GRP78. (A)GRP78 expression in Hela cells when GRP78 was knocked out using CRISPR-Cas9 technique. (B) Total and phosphorylated levels of PERK, elF2α and IRE1α, the expression of ATF6α (p50), ATF4, CHOP, as well as the expression of m-SERBP-1c in membrane or nuclear compartments, were detected in RA FLSs with GRP78 deficiency or not. β-Actin was used as a loading control. (C) Silencing efficiency of siRNA targeting GRP78 (siGRP78) was detected by western blot. The two siRNAs were combined at equal concentrations for the subsequent experiments. The negative control siRNA is referred to as siCtrl. The results are representative of two independent experiments with three different samples in each. (D) Following exposure to TNF-α and IL-1β for 6 h, RA FLSs with GRP78 knocking down or not were treated by AZM or Vehicle for another 24 h. IL-6, IL-8, MMP-1, MMP-3, CXCL9 and CXCL10 levels in the supernatant of RA FLSs were analysed by ELISA. The data are expressed as the mean ± SD of three samples and are representative of three independent experiments. ***P <0.001 compared with SiCtrl. (E) RA FLSs with GRP78 silencing or not were treated with AZM or its vehicle for 24 h, Annexin X staining were applied to detect the cell apoptosis of RA FLSs (n = 3). ***P <0.001 compared with SiCtrl. GRP78wt, GRP78 wild type. GRP78-/-, GRP78 deficiency. GRP78mt, GRP78 mutation type.
Supplementary Figure 1 Effects of Etanercept and AZM on the production of pro-inflammatory factors. Following exposure to TNF-α and IL-1β for 6 h, RA FLSs were treated by AZM, Etanercept or in combinaton for another 24 h. IL-6 (A) , IL-8 (B) , MMP-1(C) and MMP-3 (D) levels in the supernatant of RA FLSs (n=3) were analysed by ELISA. Data are expressed as the mean ± SD of three samples and are representative of three independent experiments. ns: not significant, *P <0.05, **P <0.01 and ***P <0.001 compared with Vehicle group. ETC, etanercept.
Supplementary Figure 2 Effects of AZM on the production of pro-inflammatory factors in PBMC. Following exposure to TNF-α and IL-1β for 6 h, RA FLSs were treated by AZM, Etanercept or in combination for another 24 h. IL-6 (A) , IL-8 (B) , TNF-α (C) , IL-1α (D) and IL-1β (E) levels in the supernatant of RA FLSs were analysed by ELISA. Data are expressed as the mean ± SD of six samples (3 male and 3 female) and are representative of three independent experiments. ns: not significant, *P <0.05, **P <0.01 and ***P <0.001 compared with Vehicle group.
Supplementary Figure 3 Effects of AZM on cytokine production in CIA models. The serum from AZM or Vehicle treated CIA mice (n=6 mice per group) were collected for ELISA detection of IL-1β and IL-6, COMP, IL-10, IL-13, OPG and RANKL on the 42nd day after the first immunization. This experiment data is representative of two independent experiments from 6 mice per group. **P <0.01 and ***P <0.001 compared with Vehicle in the Control or CIA group.
Supplementary Figure 4 qRT-PCR validation of fourteen differentially expressed mRNAs in 20 pairs of RA FLSs samples. The relative expression levels of 11 up-regulated and 3 down-regulated mRNAs (selected from the top 10 dysregulated mRNAs) in 20 pairs of AZM and Vehicle treated RA FLSs. ns: not significant, *P <0.05, **P <0.01 and ***P <0.001 compared with vehicle.
Supplementary Table 1 Clinical characteristics of RA patients.
Supplementary Table 2 Primer sequences of genes in real-time fluorescent quantitative PCR (qRT-PCR).
Supplementary Table 3 List of potential AZM targets in RA FLSs identified by mass spectrometry.
Supplementary Table 4 Primer sequences of sgGRP78 in CRISPR/Cas9.