MiR-378 exaggerates angiogenesis and bone erosion in collagen-induced arthritis mice by regulating endoplasmic reticulum stress

Abstract

Rheumatoid arthritis (RA) is a chronic autoimmune disorder marked by pain, inflammation, and discomfort in the synovial joints. It is critical to understand the pathological mechanisms of RA progression. MicroRNA-378 (miR-378) is highly expressed in the synovium of RA patients and positively correlated with disease severity, but its function and underlying mechanisms remain poorly understood. In this study, miR-378 transgenic (miR-378^high) mice were used to construct the collagen-induced arthritis (CIA) model for exploring the role of miR-378 in RA development. miR-378^high CIA mice showed accelerated RA development, as evidenced by exaggerated joint swelling and bone structural deformities. More severe endoplasmic reticulum (ER) stress and the consequent angiogenesis and osteoclastogenesis were also activated in the synovial tissue and calcaneus, respectively, in the miR-378^high group, suggesting that ER plays a significant role in miR-378-mediated RA pathogenesis. Upon in vitro RA induction, fibroblast-like synoviocytes (FLSs) isolated from miR-378^high mice showed a higher expression level of ER stress markers. The conditioned medium (CM) from RA-FLSs of miR-378^high mice stimulated more intensive angiogenesis and osteoclastogenesis. The ER stress-related protein Crebrf was identified as a downstream target of miR-378. Crebrf knockdown diminished the promoting effect of miR-378 on ER stress, as well as its downstream angiogenesis and osteoclastogenesis activities. Tail vein injection of anti-miR-378 lentivirus in an established RA mouse model was shown to ameliorate RA progression. In conclusion, miR-378 amplified RA development by promoting ER stress and downstream angiogenesis and osteoclastogenesis, thus indicating that miR-378 may be a potential therapeutic target for RA treatment.

Fig. 1. MiR-378^high CIA mice displayed exaggerated RA phenotype of synovium in collagen-induced arthritis (CIA) model.

A Representative paw photographs of WT and miR-378^high mice after PBS or Collagen II induction. B, C The side view (B) and front view (C) of micro-CT reconstruction of mice hind paw. D, E Semiquantitative analysis of bone volume (BV) (D) and bone volume/total volume (BV/TV) (E) of mice hind paw bone from different groups (n = 6; *p < 0.05, **p < 0.01, ***p < 0.001). F, G Mean clinical scores (F) and mean paw thickness (G) from different groups were assessed every day after collagen stimulation (n = 6; *p < 0.05, **p < 0.01, CIA versus sham; ^#p < 0.05, ^##p < 0.01, miR-378^high group versus WT group). H, I Ankle histopathology analysis by Safranin O&fast green (H) and H&E staining (I). Scale bars: 100 μm. J, K Immunohistochemistry analysis of RA ankle using anti-IL-1β (J) and anti-Mmp13 (K) antibodies. Scale bars: 100 μm.

Fig. 2. miR-378 overexpression severely induced ER stress in FLSs as well as the consequent angiogenesis and osteoclastogenesis.

A Representative immunohistochemical staining of ER stress-related markers in the synovium sections obtained from miR-378^high CIA mice. Scale bars: 100 μm. B, C Relative mRNA (B) and protein (C) expression level of ER stress-related markers in synovium sections from miR-378^high CIA mice (n = 6; *p < 0.05, **p < 0.01, ***p < 0.01). D Representative immunofluorescence staining of angiogenesis-related marker VEGF and CD31 in the synovium sections obtained from miR-378^high CIA mice. Scale bars: 100 μm. E Relative mRNA expression level of angiogenesis markers in synovium tissues from miR-378^high CIA mice (n = 6; *p < 0.05, **p < 0.01). F Trap staining of calcaneus sections obtained from miR-378^high CIA mice. Scale bars: 100 μm. G Relative mRNA expression level of osteoclastogenesis markers in synovium tissues from miR-378^high CIA mice (n = 6; *p < 0.05). H Protein expression pattern of VEGF and RANKL in synovium tissues from miR-378^high CIA mice.

Fig. 3. Induction of ER stress-related makers by proinflammatory cytokines was aggregated upon miR-378 overexpression.

A, B The mRNA (A) and protein (B) expression level of ER stress-related makers in RA-FLSs from WT and miR-378^high mice upon TNFα activation (n = 6; *p < 0.05, **p < 0.01, ***p < 0.001). C Both WT and miR-378^high FLSs showed ER stress after TG induction, as manifested by enlarged endoplasmic reticulum cisternae. D Semi-quantitative analysis of ER sheet width. E, F The mRNA (E) and protein (F) expression level of ER stress-related makers in RA-FLSs from WT and miR-378^high mice upon thapsigargin stimulation (n = 6; *p < 0.05, **p < 0.01). G Immunofluorescence staining of ER stress-related markers in RA-FLSs from WT and miR-378^high mice upon thapsigargin stimulation. Scale bar: 100 μm.

Fig. 4. ER stressor-induced miR-378^high FLSs promoted angiogenesis and osteoclastogenesis.

A Diagram to show the collection of conditioned medium from RA-FLSs and the co-culture system for angiogenesis and osteoclastogenesis induction. B The protein expression level of angiogenesis and osteoclastogenesis makers in RA-FLSs from WT and miR-378^high mice upon thapsigargin activation. C Images of tube formation at 3 and 6 h post-induction with the treatment of conditioned medium from miR-378^high FLSs upon thapsigargin stimulation. D, E Semiquantitative analysis of total branching length of each group (D) and number of junctions (E) in each group (n = 6; *p < 0.05, **p < 0.01). F TRAP staining assay at day 5 with the treatment of conditioned medium from miR-378^high FLSs upon thapsigargin stimulation. Scale bar: 100 μm. G Quantitative data of osteoclast fold change (n = 6; *p < 0.05, **p < 0.01). H The mRNA expression level of osteoclastogenesis-related markers measured by qRT-PCR (n = 6; *p < 0.05, **p < 0.01).

Fig. 5. Crebrf was characterized to be the key hub gene in miR-378 regulated ER stress in FLSs and mediated the promoting effect of miR-378^high FLSs on angiogenesis and osteoclastogenesis.

A Schematic diagram to show the miRNA target gene screening procedure. The TargetScan and ENCORI software were applied for in silico analysis of conserved 3’-UTR binding sites of miR-378 target genes, respectively. The overlapped genes were listed. B Conservation of the miR-378-3p binding site on Crebrf 3′-UTR (shaded region) among different species. The wild type (wt) and mutation (mut) form of the Crebrf 3′-UTR luciferase reporter vector were shown. C Effects of miR-378-3p on the luciferase activity of pmiRGLO vectors incorporated with Crebrf-wt or Crebrf-mut sequence were measured (n = 6; ***p < 0.001). D The mRNA and protein expression levels of Crebrf in WT and miR-378 synoviocytes were measured respectively (n = 6; *p < 0.05). E The mRNA expression levels of ER stress markers of WT and miR-378^high synoviocytes upon Crebrf knockdown under CIA condition were assessed by real-time PCR (n = 6; *p < 0.05, **p < 0.01, ***p < 0.001). F Representative immunofluorescence staining of ER stress markers Grp78 and Atf6 upon Crebrf knockdown in FLSs obtained from miR-378^high CIA mice. Scale bar: 100 μm. G Western blot analysis of VEGF and RANCL protein expression in miR-378^high FLSs upon Crebrf knockdown. H Images of tube formation at 3 and 6 h post-induction with the treatment of conditioned medium from miR-378^high FLSs upon Crebrf knockdown. I, J Semiquantitative analysis of total branching length of each group (I) and number of junctions (J) in each group. K TRAP staining assay at day 5 with the treatment of conditioned medium from miR-378^high FLSs upon Crebrf knockdown. Scale bar: 100 μm. L Quantitative data of osteoclast fold change (n = 6; *p < 0.05, **p < 0.01). M The mRNA expression level of osteoclastogenesis-related markers measured by qRT-PCR (n = 6; *p < 0.05, **p < 0.01, ***p < 0.001).

Fig. 6. Anti-miR-378 injection ameliorated RA phenotype in synovium in miR-378^high CIA mice.

A Representative paw photographs and the side view and front view of micro-CT reconstruction of hind paw of miR-378^high mice after anti-miR-NC or anti-miR-378 mediated lentivirus treatment. B Histopathology analysis of RA joints by Safranin O&fast green and H&E staining, as well as immunohistochemistry staining using anti-IL-1β and anti-Mmp13 antibodies. Scale bars: 100 μm. C, D Semiquantitative analysis of bone volume (BV) (C) and bone volume/total volume (BV/TV) (D) of mice hind paw bone from different groups (n = 6; *p < 0.05, **p < 0.01). E, F Mean clinical scores (E) and mean paw thickness (F) from different groups were assessed every day after collagen stimulation (n = 6; **p < 0.01). G Representative immunohistochemical staining of ER stress-related markers in the synovium sections after anti-miR-378 lentivirus treatment. Scale bars: 100 μm. H Relative mRNA expression level of ER stress-related markers in synovium sections (n = 6; *p < 0.05, **p < 0.01). I Representative immunofluorescence staining of angiogenesis marker VEGF and CD31 in the synovium sections. Scale bars: 100 μm. J Relative mRNA expression level of angiogenesis markers in synovium tissues (n = 6; *p < 0.05, **p < 0.01). K Trap staining of synovium sections. Scale bars: 100 μm. L Relative mRNA expression level of angiogenesis markers in synovium tissues (n = 6; *p < 0.05).