miRNAs in bone tissue correlate to bone mineral density and circulating miRNAs are gender independent in osteoporotic patients

Abstract

We previously demonstrated the aberrant expression of nine specific miRNAs in serum from osteoporotic patients. In the present study, we further evaluated the expression of these miRNAs in bone tissue, osteoblasts, and osteoclasts from 28 patients. We hypothesize that miRNA expression in serum from osteoporotic patients may be gender-independent. A further hypothesis is that the miRNA expression in bone could be correlated with BMD values. Moreover, intracellular expression of these osteoporosis-related miRNAs may indicate the role of these molecules during osteoporosis. Our results indeed show that miRNA expression in serum was gender-independent except for miR125b-5p. A correlation with BMD was confirmed for miR-21-5p, miR-24-3p, miR-93-5p, miR-100-5p and miR125b-5p with linear correlation coefficients r > 0.9. Intracellular studies revealed a simultaneous up-regulation of miR-21-5p, miR-93-5p, miR-100-5p and miR125b-5p in osteoblasts and in osteoclasts. miR-148a-3p up-regulation in cells was specific for osteoporotic osteoclasts. Altogether, miR-21-5p, miR-93-5p, miR-100-5p, and miR-125b-5p showed significant upregulation in serum, tissue and bone cells of osteoporotic patients. All except miR-125b-5p showed gender independent expression and good correlation to BMD values. Our results suggest that these miRNAs may be important for an earlier diagnosis of osteoporosis.

Figure 1

Relative expression of in serum circulating miRNA in male and female subjects with and without osteoporosis (each N = 7). The expression of miRNA-21-5p (A), miRNA-23a-3p (B), miRNA-24-3p (C), miRNA-93-5p (D), miRNA-100-5p (E), miRNA-122-5p (F), miRNA-124-3p (G), miRNA-125b-5p (H) and miRNA-148a-3p (I) are shown. Significant differences are indicated by *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Normality of the data was tested by D’Agostino Pearson normality test. Statistical analysis was performed by means of a two-tailed Mann-Whitney test.

Figure 2

Relative miRNA expression and its correlation with BMD in bone tissue samples. Harvested femur heads (N = 28) were analyzed by qCT and the BMD was calculated (A). Column charts show miRNA relative expression for bone samples analyzed grouped by obtained BMD value (BMD ≥ 120 mg/cm³ normal, 120 mg/cm³ > BMD > 80 mg/cm³ osteopenic and BMD ≤ 80 mg/cm³ osteoporotic). The expression of miRNA-21-5p (B), miRNA-23a-3p (C), miRNA-24-3p (D), miRNA-93-5p (E), miRNA-100-5p (F), miRNA-122-5p (G), miRNA-124-3p (H), miRNA-125b-5p (I) and miRNA-148a-3p (J) are shown. Significant differences are indicated by *p < 0.05, **p < 0.01. Normality of the data was tested by D’Agostino Pearson test. Statistical analysis was performed by means of Kruskal-Wallis test corrected by Dunn’s test for multiple comparisons. Linear regression analysis was performed for miRNA relative expression vs. BMD and the resulting correlation coefficient (r) and coefficient of determination (r²) are shown for each graph.

Figure 3

Relative expression of analyzed miRNAs in isolated osteoblasts. Osteoblasts were isolated from bone samples of osteoporotic and non-osteoporotic patients and cultured up to 14 days in vitro. The expression levels of miRNA-21-5p (A), miRNA-23a-3p (B), miRNA-24-3p (C), miRNA-93-5p (D), miRNA-100-5p (E), miRNA-122-5p (F), miRNA-124-3p (G), miRNA-125b-5p (H) and miRNA-148a-3p (I) are illustrated. Significant differences are indicated by *p < 0.05, **p < 0.01, ***p < 0.001. Normality of the data was tested by D’Agostino Pearson test. Statistical analysis was performed by means of Kruskal-Wallis test corrected by Dunn’s test for multiple comparisons.

Figure 4

Relative expression of analyzed miRNAs in osteoclasts. Osteoclasts were generated from PBMCs isolated from blood samples of osteoporotic and non-osteoporotic patients. PBMCs were stimulated with M-CSF and RANKL for a period of 21 days or 28 days to generate the osteoclasts. The expression levels of miRNA-21-5p (A), miRNA-23a-3p (B), miRNA-24-3p (C), miRNA-93-5p (D), miRNA-100-5p (E), miRNA-122-5p (F), miRNA-124-3p (G), miRNA-125b-5p (H) and miRNA-148a-3p (I) are illustrated. Significant differences are indicated by *p < 0.05, **p < 0.01. Normality of the data was tested by D’Agostino Pearson test. Statistical analysis was performed by means of Kruskal-Wallis test corrected by Dunn’s test for multiple comparisons.

Figure 5

Schematic representation of the role of miRNA-21-5p (A), miRNA-23a-3p and miRNA-24-3p (B), miRNA-93-5p and miRNA-100-5p (C), miRNA-122-5p, miRNA-124-3p and miRNA-125b-5p (D) and miRNA-148a-3p (E) in bone remodeling during osteoporosis. For some of these miRNAs, the target(s) mRNA are known and they have been added to the schema. Solid lines in the diagram indicate direct action while dashed lines denote diminishing this action. Abbreviations used in the diagram are PDCD4: Programmed cell death protein 4, Smad7: Mothers against decapentaplegic homolog 7, Spry1: Protein sprouty homolog 1, RunX2: Runt-related transcription factor 2, SATB2: Special ATrich-sequence-binding protein 2, BMP-R2: Bone morphogenetic protein receptor type II, RANKL: Receptor activator of NF-κB ligand, TRAF6: Tumor necrosis receptor-associated factor 6, NFATc1: Nuclear factor of activated T-cells cytoplasmic 1 and MAFB: V-maf musculoaponeurotic fibrosarcoma oncogene homolog B.