Silencing miR-146a influences B cells and ameliorates experimental autoimmune myasthenia gravis

Abstract

MicroRNAs have been shown to be important regulators of immune homeostasis as patients with aberrant microRNA expression appeared to be more susceptible to autoimmune diseases. We recently found that miR-146a was up-regulated in activated B cells in response to rat acetylcholine receptor (AChR) α-subunit 97-116 peptide, and this up-regulation was significantly attenuated by AntagomiR-146a. Our data also demonstrated that silencing miR-146a with its inhibitor AntagomiR-146a effectively ameliorated clinical myasthenic symptoms in mice with ongoing experimental autoimmune myasthenia gravis. Furthermore, multiple defects were observed after miR-146a was knocked down in B cells, including decreased anti-R97-116 antibody production and class switching, reduced numbers of plasma cells, memory B cells and B-1 cells, and weakened activation of B cells. Previously, miR-146a has been identified as a nuclear factor-κB-dependent gene and predicted to base pair with the tumour necrosis factor receptor-associated factor 6 (TRAF6) and interleukin-1 receptor-associated kinase 1 (IRAK1) genes to regulate the immune response. However, our study proved that miR-146a inhibition had no effect on the expression of TRAF6 and IRAK1 in B cells. This result suggests that the function of miR-146a in B cells does not involve these two target molecules. We conclude that silencing miR-146a exerts its therapeutic effects by influencing the B-cell functions that contribute to the autoimmune pathogenesis of myasthenia gravis.

Keywords: RNA interference; experimental autoimmune myasthenia gravis; interleukin-1 receptor-associated kinase-1; miR-146a; tumour necrosis factor receptor-associated factor-6.

Figure 1

miR-146a was up-regulated in B cells following activation. The miR-146a mRNA was determined by quantitative PCR analysis in sensitized B cells. These B cells were cultured with a second immunization by R97-116 peptide in the absence or presence of AntagomiR-146a and/or AntagomiR Negative Control (NC). Non-activated B cells from the complete Freund's adjuvant (CFA) group were used as negative controls. The data were from three independent experiments and are shown as means ± SEM, with n = 3. The results showed that miR-146a expression was decreased significantly in R97-116-stimulated plus AntagomiR-146a-inhibited B cells (R97-116+ AntagomiR-146a subgroup) when compared with R97-116-stimulated but plus NC-inhibited B cells (R97-116+ NC subgroup) and normal R97-116-stimulated B cells (Positive control) (**P < 0·01).

Figure 2

B cells with miR-146a knockdown secreted less total IgG in vitro. Samples were the same cultured B cells as the previous experiments. Supernatants from these cultures were harvested 72 hr after co-culture with T cells and assayed by ELISA for total IgG. The data were from three independent experiments and shown as means ± SEM, with n = 3. The results showed that the secretion levels of total IgG in R97-116+ AntagomiR-146a subgroup was significantly lower than in the R97-116+ NC subgroup and Positive control. (*P < 0·05).

Figure 3

Treatment with AntagomiR-146a ameliorated clinical myasthenic symptoms in mice with ongoing experimental autoimmune myasthenia gravis (EAMG). Each symbol represents the mean clinical score (MCS) of mice in the AntagomiR-146a group (n = 10), the NC group (n = 10) and the PBS group (n = 10) at various times after treatment with respective treatment drugs via the caudal vein for 3 days continuously. Differences of the MCS were statistically significant between three groups since the sixth days after enrolment began. The MCS of AntagomiR-146a group was significantly lower than NC and PBS groups. At the end of the experiment, the MCS of the AntagomiR-146a group was 0·63 ± 0·33, the NC group was 2·01 ± 0·41, and the PBS group was 2·14 ± 0·55 (*P < 0·05; **P < 0·01).

Figure 4

miR-146a knockdown inhibited the production of anti-R97-116 antibodies in vivo. The secretion levels of total IgG, IgG1 and IgG2b were significantly lower in the AntagomiR-146a group compared with the NC group and PBS group. There were no significant differences of IgG2a, IgG2c and IgM in all groups. The complete Freund's adjuvant (CFA) group, which served as the control and exhibited no secretion of antigen-specific antibody, demonstrated the success of experimental autoimmune myasthenia gravis induction in the three experimental groups (*P < 0·05; **P < 0·01).

Figure 5

AntagomiR-146a could effectively silence miR-146a of B cells in vivo. At day 10 after the treatment, the mice were killed to collect spleen cells and B cells were isolated. (a) The high purity of sorted B cells, the left was negative control. (b) The expression of miR-146a in B cells was significantly reduced in the AntagomiR-146a group (0·6747 ± 0·0645) compared with the NC group (1·7971 ± 0·1031) and PBS group (1·8140 ± 0·1021). The complete Freund's adjuvant (CFA) group demonstrated low expression of miR-146a (0·1487 ± 0·0836) (***P < 0·001).

Figure 6

AntagomiR-146a influenced the amount of plasma cells, memory B cells and B-1 cells in vivo. (a) Flow cytometry analysis of CD45R⁺ CD27⁺ CD138⁺ plasma cells in spleen cells of three group mice(n = 10/group). In the AntagomiR-146a group it was 1·24 ± 0·57%, the NC group was 3·87 ± 0·78% and the PBS group was 4·15 ± 0·83%. (b) Flow cytometry analysis of CD45R⁺ CD27⁺ CD44⁺ memory cells. In the AntagomiR-146a group it was 1·57 ± 0·61%, the NC group was 7·62 ± 0·68% and the PBS group was 7·41 ± 1·07%. (c) Flow cytometry analysis of CD19⁺ CD5⁺ B-1 cells. In the AntagomiR-146a group it was 4·01 ± 0·63%, the NC group was 6·16 ± 0·76% and the PBS group was 5·97 ± 0·77%. In the dot plots at right, each symbol represents an individual mouse, and the horizontal bars represent the median values. The blank control was used to set the cross mark into four quadrants by the mean of making its probability of the upper right quadrant approximately 0% (*P < 0·05; **P < 0·01).

Figure 7

AntagomiR-146a decreased the expression of CD40, CD80 and CD86 on B cells. (a) Flow cytometry analysis of CD40 on B-cell surface. In the AntagomiR-146a group it was 25·72 ± 3·16%, the NC group was 78·57 ± 2·07% and the PBS group was 79·62 ± 1·43%. (b) Flow cytometry analysis of CD80. In the AntagomiR-146a group it was 10·37 ± 2·33%, the NC group was 30·68 ± 1·98% and the PBS group 32·75 ± 1·65%. (c) Flow cytometry analysis of CD86. In the AntagomiR-146a group it was 22·92 ± 1·38%, the NC group was 67·62 ± 3·72% and the PBS group was 70·15 ± 2·41%. The blank control was used to set the gate by the mean of making the region (R4/R3/R2) approximately 1% (*P < 0·05).

Figure 8

Tumour necrosis factor receptor-associated factor 6 (TRAF6) and interleukin-1 receptor-associated kinase 1 (IRAK1) were not the targets for miR-146a-induced immunomodulation in B cells of experimental autoimmune myasthenia gravis (EAMG). Quantitative PCR measurements of TRAF-6 and IRAK-1 mRNA were taken to analyse the possible targets of miR-146a in B cells. (a) Amplification plot and melt curve for IRAK1 and β-actin. (b) Amplification plot and melt curve for TRAF6. (c) Data analysis showed that the expression differences of TRAF6 (P = 0·296) and IRAK1 (P = 0·463) in mouse B cells from three groups were not statistically significant.