Alterations of serum levels of BDNF-related miRNAs in patients with depression

Abstract

Depression is a serious and potentially life-threatening mental disorder with unknown etiology. Emerging evidence shows that brain-derived neurotrophic factor (BDNF) and microRNAs (miRNAs) play critical roles in the etiology of depression. Here this study was aimed to identify and characterize the roles of BDNF and its putative regulatory miRNAs in depression. First, we identified that miR-182 may be a putative miRNA that regulates BDNF levels by bioinformatic studies, and characterized the effects of miR-182 on the BDNF levels using cell-based studies, side by side with miR-132 (a known miRNA that regulates BDNF expression). We showed that treatment of miR-132 and miR-182 respectively decreased the BDNF protein levels in a human neuronal cell model, supporting the regulatory roles of miR-132 and miR-182 on the BDNF expression. Furthermore, we explored the roles of miR-132 and miR-182 on the BDNF levels in depression using human subjects by assessing their serum levels. Compared with the healthy controls, patients with depression showed lower serum BDNF levels (via the enzyme-linked immunosorbent assays) and higher serum miR-132 and miR-182 levels (via the real-time PCR). Finally, the Pearson's (or Spearman's) correlation coefficient was calculated to study whether there was a relationship among the Self-Rating Depression Scale score, the serum BDNF levels, and serum BDNF-related miRNA levels. Our results revealed that there was a significant negative correlation between the SDS scores and the serum BDNF levels, and a positive correlation between the SDS scores and miR-132 levels. In addition, we found a reverse relationship between the serum BDNF levels and the miR-132/miR-182 levels in depression. Collectively, we provided evidence supporting that miR-182 is a putative BDNF-regulatory miRNA, and suggested that the serum BDNF and its related miRNAs may be utilized as important biomarkers in the diagnosis or as therapeutic targets of depression.

Figure 1. BDNF serum levels and its related miRNAs.

(A) BDNF concentrations in patients and controls. The serum levels of BDNF in the patients with depression (n = 40; 20.05±5.98 ng/ml) were significantly decreased compared with those of the healthy controls (n = 40; 25.22±5.17 ng/ml). (B) The targeting site on BDNF-mRNA-3′UTR by miR-182, which was aligned with the mRNA-3′UTR of human bdnf gene with the nucleotide position. Vertical lines indicate identity.

Figure 2. miR-132/182 regulating BDNF expression.

(A, B) Detection of BDNF by western blotting. SH-SY5Y cells were treated with miR-182 or miR-132, the results showed that BDNF expression in the miR-182- or miR-132-treated cultures was much lower than that of negative control miRNA cultures. *P<0.05, miR-182- or miR-132-treated cultures vs. control cultures. Relative values for BDNF vs. Actin are indicated in Fig.2B.

Figure 3. Serum miR-132 and miR-182 levels detected by real-time PCR.

(A, B) Serum miR-132 or miR-182 levels in patients with depression and their controls, respectively. Real-time PCR showed that serum miR-132 (or miR-182) levels in depressed patients (n = 40) were much higher than those in healthy controls (n = 40, P<0.01).

Figure 4. Relationship between levels of serum BDNF and its related miRNAs.

(A) A significant negative correlation (Spearman r_s = −0.307, P = 0.006) between the serum BDNF and miR-132 levels in depressed patients (n = 40) and controls (n = 40). (B) No significant correlation (Spearman r_s = 0.098, P = 0.385) was found between the serum BDNF and miR-182 levels in depressed patients (n = 40) and controls (n = 40).

Figure 5. Relationship between serum levels of BDNF-related miRNAs and SDS score.

(A) A significant negative correlation (Pearson r = −0.427, P = 7.75E-05) was found between serum BDNF levels and SDS score in depressed patients (n = 40) and controls (n = 40). (B) An obvious positive correlation (Spearman r_s =0.347, P = 0.002) was revealed between the serum miR-132 levels and SDS score in patients with depression (n = 40) and controls (N = 40). (C) There was an obvious positive correlation (Spearman r_s =0.242, P = 0.030) between the serum miR-182 levels and SDS score in depressed patients (n = 40) and healthy controls (N = 40).