Long non-coding RNA H19 regulates matrisome signature and impacts cell behavior on MSC-engineered extracellular matrices

Abstract

Background: The vast and promising class of long non-coding RNAs (lncRNAs) has been under investigation for distinct therapeutic applications. Nevertheless, their role as molecular drivers of bone regeneration remains poorly studied. The lncRNA H19 mediates osteogenic differentiation of Mesenchymal Stem/Stromal Cells (MSCs) through the control of intracellular pathways. However, the effect of H19 on the extracellular matrix (ECM) components is still largely unknown. This research study was designed to decode the H19-mediated ECM regulatory network, and to reveal how the decellularized siH19-engineered matrices influence MSC proliferation and fate. This is particularly relevant for diseases in which the ECM regulation and remodeling processes are disrupted, such as osteoporosis.

Methods: Mass spectrometry-based quantitative proteomics analysis was used to identify ECM components, after oligonucleotides delivery to osteoporosis-derived hMSCs. Moreover, qRT-PCR, immunofluorescence and proliferation, differentiation and apoptosis assays were performed. Engineered matrices were decellularized, characterized by atomic force microscopy and repopulated with hMSC and pre-adipocytes. Clinical bone samples were characterized by histomorphometry analysis.

Results: Our study provides an in-depth proteome-wide and matrisome-specific analysis of the ECM proteins controlled by the lncRNA H19. Using bone marrow-isolated MSC from patients with osteoporosis, we identified fibrillin-1 (FBN1), vitronectin (VTN) and collagen triple helix repeat containing 1 (CTHRC1), among others, as having different pattern levels following H19 silencing. Decellularized siH19-engineered matrices are less dense and have a decreased collagen content compared with control matrices. Repopulation with naïve MSCs promotes a shift towards the adipogenic lineage in detriment of the osteogenic lineage and inhibits proliferation. In pre-adipocytes, these siH19-matrices enhance lipid droplets formation. Mechanistically, H19 is targeted by miR-29c, whose expression is decreased in osteoporotic bone clinical samples. Accordingly, miR-29c impacts MSC proliferation and collagen production, but does not influence ALP staining or mineralization, revealing that H19 silencing and miR-29c mimics have complementary but not overlapping functions.

Conclusion: Our data suggest H19 as a therapeutic target to engineer the bone ECM and to control cell behavior.

Keywords: Bone extracellular matrix; Decellularization; Fragility fractures; Gene therapy; Non-coding RNAs.

Fig. 1

Proteomic analysis of H19 downstream targets and ECM-related processes. a H19 expression after transfection of MSCs, isolated from osteoporotic patients, with two alternative small interference RNAs against H19 (siH19-1 and siH19-2) or the respective control (siCTR), cultured for 3 days in osteogenic-inducing conditions (N = 6). b H19 subcellular location in U2OS osteosarcoma cells. Levels were normalized against the whole-cell content. XIST and MALAT1, two nuclear lncRNAs, were used as quality controls. c Volcano plot of the proteins identified after transfection with siH19 versus control (FC ≥ |1.25|) and p-value < 0.05. The y-axis corresponds to the − Log₁₀(p-value) of a certain protein and the x-axis the correspondent Log₁₀(FC). d Gene ontology (GO) enrichment analysis, according to PANTHER classification system. The x-axis shows the −Log₁₀(p-value) associated with determined GO term (y-axis). e Venn diagram showing the extracellular proteins differentially expressed by H19 knockdown. f Proteins associated with the GO terms “non-structural extracellular” and “extracellular matrix” with p < 0.01 and a minimum of four unique peptides (left) and with the GO term “extracellular structural activity” (right) and respective heatmap.

Fig. 2

Effect of H19 knockdown on osteogenic differentiation, ECM composition, proliferation and apoptosis in MSCs. a mRNA expression levels of FBN1, VNT and CTHRC1 after H19 knockdown compared with control (N = 6). b Representative image and quantification of FBN1, VTN and CTHRC1 protein levels 10 days after culture in osteogenic-inducing conditions (N = 6; scale: 50 μm). c COL1 staining (left) and quantification (right) at day 7 of differentiation (COL1: red; N = 6; scale: 100 μm). d ECM components (COL1A1, COL1A2 and COL3A1) and e Osteogenic markers (ALP and RUNX2) expression levels at day 3 after transfection and incubation in osteogenic-inducing conditions (N = 6). f ALP (N = 6) and g Alizarin (N = 6) staining after transfection and culture for 7 and 14 days, respectively, in osteogenic-inducing conditions. h Representative fluorescence profile (resorufin) measured every 24 h for 7 days in transfected MSCs; quantification of the percentage of cells in proliferation (ki-67⁺/DAPI⁺) 3 days after siH19 or siCTR transfection; and representative images of MSC stained for the proliferation marker Ki-67 protein (red) and nuclear DNA labeled with DAPI (blue) (N = 6; scale: 50 μm) (RFU: relative fluorescence units). i Flow cytometry analysis of Annexin V/PI staining and quantification of viable (Annexin V⁻ PI⁻), early apoptotic (Annexin V⁺ PI⁻), or late apoptosis and dead cells (Annexin V⁻ PI⁺ or Annexin V⁺ PI⁺) (N = 6). siCTR: MSC transfected with scrambled control sequence; siH19: MSC transfected with small interference RNA against H19.

Fig. 3

Decellularized ECM produced by siH19-engineered MSC and its effect on naïve MSC. a Schematic representation of the experiment. b sGAGs and c collagenous content quantification after decellularization of the matrices produced by control-MSC (CTR-matrix) or siH19-MSC (siH19-matrix) cultured under osteogenic-inducing conditions (N = 6). d Representative AFM image of the topography of matrices produced by MSC depleted in H19 and control, after 10 days in osteogenic-inducing conditions, and following the decellularization process. e Expression levels of osteogenic, adipogenic and chondrogenic markers assessed by RT-qPCR, 7 days after repopulation of decellularized matrices with naïve MSCs (N = 6). f Representative image and quantification of immunostaining of Osteopontin (OPN: green; Nuclei: blue) (N = 6; scale: 100 μm). g Measurement of the metabolic activity through resazurin assay (N = 6; scale: 50 μm). h Representative image of MSC stained for the proliferation marker Ki-67 protein (red) and nuclear DNA labeled with DAPI (blue); and quantification of the percentage of MSC in proliferation (Ki-67⁺/DAPI⁺) (N = 6) after 7 days of culture on top of decellularized matrices. i Quantification of LDH levels on the conditioned-media from naïve MSC cultured on the decellularized matrices (N = 6).

Fig. 4

Effect of siH19-MSCs derived matrices on pre-adipocytes phenotype. a Osteogenic, adipogenic and chondrogenic markers expression levels assessed by RT-qPCR, 7 days after repopulation with pre-adipocytes on decellularized matrices (N = 6). b Representative image and quantification of the percentage of positive cells for the Oil Red O staining after 3 days of culture on decellularized matrices (Hematoxylin: blue-purple color; Oil Red O⁺ staining: red color) (N = 6; scale: 100 μm). c Quantification of LDH levels on the conditioned-media of pre-adipocytes cultured on decellularized matrices (N = 6). d Correlation of trabecular separation (Tb.Sp) with the percentage of adipose tissue in bone samples from patients with fragility fractures (N = 16) (left panel). Representative images (scale: 100 μm) and correlation coefficient of trabecular separation (Tb.Sp) and age (N = 17) (right panel).

Fig. 5

miR-29c-3p expression in bone samples and the impact of miR-29c-3p in H19 levels. a Box and whiskers plots show miRNA expression levels (2^−ΔCq) in bone samples from patients with fragility fractures (FF) or osteoarthritis (OA). b Adjusted p-values for miRNA expression considering age. c Expression levels of H19 and miR-29c-3p on MSC from healthy and FF patients (N = 3). d Effect of the modulation of the levels of miR-29c-3p on the lncRNA H19 expression, 3 days post-transfection in osteogenic-inducing conditions (N = 6). (Mimics-NC: MSC transfected with scrambled miRNA mimics negative control; Mimics: miR-29c-3p transfected MSC; Inhibitor-NC: MSC transfected with scrambled miRNA inhibitor negative control; Inhibitor: miR-29c-3p-inhibitor transfected MSC)

Fig. 6

Effect of miR-29c-3p in ECM composition, osteogenic differentiation, cell proliferation and apoptosis. a COL1 staining and quantification at day 7 of differentiation. (COL1: red; N = 6; scale: 100 μm) b Expression of ECM components (COL1A1, COL1A2 and COL3A1) by qPCR after 3 days in osteogenic-inducing conditions (N = 6). c Correlation between miR-29c-3p expression level and collagen expression on bone samples. d Expression levels of osteogenic markers at day 3 after transfection and incubation in osteogenic-inducing conditions (N = 6). e ALP and f Alizarin staining (N = 6) and Alizarin solubilization quantification (N = 6), after transfection and culture for 7 and 14 days, respectively, in osteogenic-inducing conditions. g Representative fluorescence profile (resorufin) measured every 24 h for 7 days in transfected MSCs and quantification of the percentage of cells in proliferation (Ki-67⁺/DAPI⁺) after 3 days in culture (N = 6). h Quantification of viable (Annexin V⁻ PI⁻), early apoptotic (Annexin V⁺ PI⁻), or late apoptosis and dead MSCs (Annexin V⁻ PI⁺ and Annexin V⁺ PI⁺) (N = 6). (Mimics-NC: MSC transfected with scrambled miRNA mimics negative control; Mimics: miR-29c-3p transfected MSC; Inhibitor-NC: MSC transfected with scrambled miRNA inhibitor negative control; Inhibitor: miR-29c-3p-inhibitor transfected MSC).