Skip to main content
Download PDF

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

Stem Cell Research & Therapy volume 14, Article number: 37 (2023) Cite this article

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.

Introduction

Osteoporosis is a disease that affects more than 200 million people worldwide [1, 2]. It is a skeletal disease defined by a low bone mineral density (BMD) that results in the deterioration of the bone tissue microarchitecture/quality [1, 2]. It increases bone fragility and vulnerability to fractures, which is the most important cause of morbidity and loss of life quality in patients suffering from this disease [3,4,5,6]. Although BMD measurements are widely used as a diagnostic tool for osteoporosis, it is unrealistic to predict the risk of fracture merely based on changes in the BMD [7, 8]. Therefore, it is crucial to emphasize the importance of the use of other indicators of bone health [7, 8]. Among other properties related to bone quality, bone composition, the biophysical properties and structure of these components should be taken into consideration. In fact, the impairment of the composition of the organic portion of the bone was reported in animal models (e.g., DMP-1) [9, 10] and patients (e.g., collagens and osteocalcin (OCN)) [11,12,13] with osteoporosis, as well as a structural/orientational alteration of the fibrils [10, 14, 15]. These changes impact the matrix assembly and, consequently, the structural organization of the bone, influencing the mechanical properties, widely described to be impaired [9, 16, 17].

At the cellular level, osteoporosis is caused by a disruption in the homeostasis between bone formation and resorption [5, 18, 19], leading to the consequent degradation of the bone matrix. Recent studies have shown that the disruption of the commitment into the osteogenic lineage in benefit of adipogenesis may also be associated with disease progression and/or ageing, leading to a disturbance in bone regeneration/repair [20,21,22,23].

Thus, it is crucial to determine the role of key molecular regulators of Mesenchymal Stem/Stromal Cells (MSCs) differentiation and bone extracellular matrix (ECM) components for the development of new targeted therapies in osteoporosis and fragility fractures.

Long non-coding RNAs (lncRNAs) comprise a large class of non-protein-coding genes that regulate cellular mechanisms, including bone formation, remodeling and homeostasis [24,25,26]. Their ability to regulate bone-related mechanisms, as well as their abnormal expression in bone disorders, like osteoporosis, illustrates their potential as therapeutic tools and/or biomarkers [6, 27]. H19 is an imprinted human lncRNA lacking an evolutionarily conserved open reading frame, but with a conserved secondary RNA structure, that has been described as a regulator of osteogenic differentiation [6, 28, 29]. Moreover, H19 overexpression promotes heterotopic bone formation in vivo [28, 30, 31]. It works as an endogenous competing RNA by targeting mi(cro)RNAs (eg. miR-141 and miR-22 [28], miR-138 [32], miR-188 [33]), being able to decrease HDAC4/5 and TGF-β1 [30], activate the Wnt/β-catenin pathway [28], increase SDF-1 [29] and regulate FAK expression [32] in MSCs. Furthermore, H19 is expressed at low levels in disuse osteoporosis [34]. Although several studies have shown that H19 is able to modulate different intracellular mechanisms in MSCs, the impact on the ECM production and secretion, and on the ECM-cell interactions, is still poorly explored. On the other hand, the use of genetically engineered cells, enabling the production of pro-regenerative matrices, has emerged as an alternative approach in the field of regenerative medicine. Thus, understanding the regulators of ECM components, as well as bone cells behavior on engineered-cell-derived matrices, is crucial.

In this study, we aim to investigate the role of H19 in ECM composition using osteoporosis-derived MSCs and to determine the impact of H19-engineered matrices on naïve MSCs and pre-adipocytes behavior.

Materials and methods

Clinical samples: bone tissue and human primary bone marrow-derived MSCs

Bone samples (N = 17) were collected at Centro Hospitalar Universitário São João (CHUSJ, Porto, Portugal) from discarded bone of patients that underwent hip surgery following a fragility fracture (FF). Human primary MSCs were isolated from the bone marrow derived from the femoral head of patients with fragility fractures (N = 6), following a previously described protocol [35]. MSCs from healthy donors (N = 4) were also collected from the tibia following surgery to anterior cruciate ligament injury. Patients with other medical conditions, such as tumors or inflammatory diseases, were excluded from the study. Clinicopathological data are provided in Additional file 1: Table SI (Additional file 1). Culture of MSCs is detailed in the Additional file 1: Methods section. The protocol was approved by the CHUSJ Ethics Committee for Health, according to the declaration of Helsinki, and signed informed consents were obtained from the study participants. RNA samples extracted from the trabecular bone at the distal extremity of the femoral epiphysis of patients with fragility fractures (N = 11) or osteoarthritis (N = 9) were obtained from the biobank at Instituto de Medicina Molecular (iMM, Lisbon Academic Medical Center, Lisbon, Portugal).

Osteogenic and adipogenic differentiation of bone marrow-derived MSCs

To induce osteogenic differentiation, MSCs were incubated in the presence of osteogenic supplements: 10–7 M dexamethasone, 10–2 M β-glycerophosphate and 5 × 10–5 M ascorbic acid, all from Sigma-Aldrich. To induce adipogenic differentiation, bone marrow-derived MSCs were cultured in the presence of adipogenic supplements: 10–4 M dexamethasone, 10–1 mM Indomethacin, 5 × 10–4 M 3-Isobutyl-1-methylxanthine (IBMX) and 10 μg/mL insulin, all from Sigma-Aldrich. Cells were incubated under basal or osteogenic/adipogenic conditions, and the media was changed twice a week. To evaluate osteogenic differentiation, MSCs were fixed with 40 g/L paraformaldehyde (PFA, Sigma-Aldrich) for 30 min, at room temperature (RT), before the ALP and Alizarin stainings were performed (Additional file 1: Methods). To confirm adipogenic differentiation, cells were fixed with 40 g/L PFA and stained with Oil Red O solution (Additional file 1: Methods).

RNA isolation, reverse transcription and real-time quantitative polymerase chain reaction (RT-qPCR)

RNA was isolated using TRIzol Reagent (Invitrogen), according to the manufacturer’s instructions. RNA concentration and purity were determined by measuring absorbance using a NanoDrop Spectrophotometer ND-1000 (Thermo Fisher Scientific). RNA integrity was evaluated by agarose gel electrophoresis or Agilent RNA 6 000 Pico kit (Agilent).

To test the expression of coding and long non-coding transcripts, total RNA was first treated with TURBO DNA-free kit (Life Technologies), following the manufacturer’s protocol, to remove potential DNA contaminants. RNA samples were then used to synthesize cDNA, using random hexamers (Life Technologies) and SuperScript® III Reverse Transcriptase kit (Life Technologies). qPCR reactions were performed using cDNA, iQ SYBR Green Supermix (Bio-Rad) and respective primers, according to the following conditions: 3 min at 95 °C and forty cycles of 30 s at 95 °C, 30 s at 58 °C and 30 s at 72 °C. β-actin was used as a reference gene. Primer sequences are described in Additional file 1: Table SII.

To test the expression of miRNAs, TaqMan miRNA assays (Applied Biosystems) were used, according to the manufacturer’s protocol. Briefly, complementary DNA (cDNA) was synthesized using total RNA, TaqMan MicroRNA Reverse Transcription Kit and gene specific stem-loop Reverse Transcription primers (hsa-miR-29a-3p. hsa-miR-29b-3p and hsa-miR-29c-3p, Applied Biosystems; Additional file 1: Table SIII). qPCR was carried out in a thermal cycler (MyCycler Thermal Cycler, Bio-Rad) using cDNA, SsoAdvanced™ Universal Probes Supermix (Bio-Rad) and TaqMan probes (Applied Biosystems), according to the following conditions: 10 min at 95 °C, 40 cycles of 15 s at 95 °C and 1 min at 60 °C. Small nuclear RNA U6 was used as a reference gene.

All RT-qPCR reactions were performed in duplicate and data was analyzed using Bio-Rad CFX Manager software (Bio-Rad). Relative expression levels were calculated using the quantification cycle (Ct) method, according to MIQE guidelines [36].

In silico target predictions

miRNA FASTA format sequence annotations were obtained from the miRBase database (http://www.mirbase.org/). In silico predictions of the interactions between miRNAs with lncRNAs were performed using Encyclopedia of RNA Interactomes (ENCORI).

MSC transfection

For functional assays, MSCs were transfected with a small interference RNA against H19 (20 nM; silencer select pre-designed siRNA for H19; siH19-1, also named siH19, Assay ID: n548065; siH19-2, Assay ID: n272453) or the respective control (20 nM; silencer select negative control; siCTR); or with miR-29c-3p mimics (50 nM; mimics), miR-29c-3p inhibitor (50 nM; inhibitor) or the respective controls (50 nM; miRNA mimics negative control and miRNA inhibitors negative control), all from Ambion. Oligonucleotide transfection was performed using Lipofectamine 2 000 reagent (Invitrogen), according to manufacturer’s protocol. MSCs were plated the day before the transfection at a cell density of 0.02 × 106 cells/cm2 and incubated for 12 h with the oligonucleotides-liposomal complexes in media without penicillin/streptomycin (P/S, Invitrogen). Cells were cultured with basal growth or osteogenic media, depending on the assay.

Protein isolation and identification by nanoscale liquid chromatography coupled to tandem mass spectrometry (nano LC–MS/MS)

Proteins were isolated from siH19-MSCs or respective control using an extraction buffer for matrix proteins enrichment, following 7 days of culture in osteogenic-inducing conditions. Briefly, cells were washed four times with cold PBS 1 × and a 2 M urea solution [4% (v/v) SDS (Sigma-Aldrich), 60 mM Tris–HCl (Sigma-Aldrich), 1 mM EDTA (Sigma-Aldrich); pH = 6.8] was used to solubilize the ECM and lysate the cells in the presence of protease (1:100, Thermo Fisher Scientific) and phosphatase inhibitors (1:100, Thermo Fisher Scientific). The solution was added directly to the cell culture plate to avoid losing ECM and culture plates were incubated for 30 min under constant rotation. The protein was collected to low binding microcentrifuge tubes (Eppendorf) and the lysates/solubilized ECM were spun down at 16 000 g for 5 min. Total protein concentration was quantified using the DC protein assay kit (Bio-Rad).

Protein identification and quantitation was performed by nano LC–MS/MS, using an equipment composed by an Ultimate 3000 liquid chromatography system coupled to a Q-Exactive Hybrid Quadrupole-Orbitrap mass spectrometer (Thermo Fisher Scientific), as previously described [37]. Further details are provided in Additional file 1: Methods. Bioinformatic and data analysis were performed for proteins expressed in three independent experiments with fold-change (FC) differences ≥ 1.25 or ≤  − 1.25 and p < 0.05. A minimum of two unique peptides was required for further evaluation. The differentially expressed proteins were classified according to gene ontology (GO) annotations. The enriched biological process analysis was performed using Protein ANalysis THrough Evolutionary Relationships (PANTHER) tool. Pathway analysis was performed using the software Ingenuity Pathway Analysis (IPA). For target validation, proteins were selected according to the following criteria: more than four unique peptides that simultaneously associate with cellular terms “non-structural extracellular” and “extracellular matrix” (p-value < 0.01) and with the molecular function “extracellular structural activity”.

Immunocytochemistry

MSCs were washed and fixed with 40 g/L PFA for 30 min at RT. Cells were permeabilized with 0.1% (v/v) Triton X‐100 (Sigma-Aldrich) for 10 min, at RT, rinsed and blocked with 5 g/L BSA/ 0.1% (v/v) Triton X‐100 for 1 h at RT. Then, cells were incubated with primary antibody against collagen type I (COL1, 1:250, Rockland), osteopontin (OPN; 1:250, Santa Cruz Biotechnology), fibrillin-1 (FBN1; 1:100, Thermo Fisher Scientific), collagen Triple Helix Repeat Containing 1 (CTHRC1; 1:50, Santa Cruz Biotechnology) and vitronectin (VTN; 1:25, Santa Cruz Biotechnology) overnight at 4 °C. Cells were washed three times with PBS 1x, 5 min each, incubated with the respective secondary antibody (1:1 000; Thermo Fisher Scientific) for 1 h at RT and washed again. Cell nuclei were stained with 1 μg/mL DAPI (Invitrogen) for 5 min. The antibodies manufacturers and respective dilutions are detailed in Additional file 1: Table SIV. The stainings were visualized under the Leica DMi6000 FFW (Leica Microsystems). Cells incubated with the antibody diluent alone (without primary antibody), followed by incubation with the secondary antibody, were used as negative controls. Semi-quantification of the protein was performed using the ImageJ Fiji software, considering a minimum of 6 distinct fields per condition.

Resazurin reduction assay

MSCs were seeded with a cell density of 2 × 103 cells/well in 96-well plates. The resazurin redox dye (0.01 mg/mL, Sigma-Aldrich) was added to each well at 10% (v/v) and incubated for 2 h 30 min at 37 °C and 5% (v/v) CO2, protected from the light. The supernatant was then collected, transferred to a dark 96-well plate (Falcon) and the reduction of resazurin (non-fluorescent blue dye) into resorufin (pink fluorescent) was measured at an excitation wavelength of 530 nm and an emission wavelength of 590 nm, using the spectrophotometer microplate reader Synergy MX (Biotek Synergy). Seven biological replicates were performed for each condition and time-point.

Cell proliferation assay through Ki-67 immunostaining

MSCs were seeded at 15 000 cells/cm2 in 96-well plates, cultured for 3 days in MSC growth media, fixed and immunostained for Ki-67 (1:100; SP6, Thermo Fisher Scientific) followed by Alexa Fluor 647-conjugated secondary antibody (1:1 000; Thermo Fisher Scientific). Nuclei were stained with 1 μg/mL DAPI, for 5 min. The respective manufacturers and dilutions are detailed in Additional file 1: Table SIV. Cells incubated with the secondary antibody alone were used as a negative control. Five biological replicates were performed per condition. Images were acquired in a IN Cell Analyzer 2 000 (GE Healthcare) using a Nikon 20x/0.45 NA Plan Fluor objective and then analyzed using the IN Cell Developer software (GE Healthcare). The total number of nuclei per image was counted, and the Ki-67 positive cells were quantified. Results are presented as the percentage of Ki-67-positive cells per total number of cells (DAPI-positive).

Cell death/apoptosis assay by flow cytometry and the colorimetric lactate dehydrogenase (LDH) assay

Transfected MSCs were cultured for 2 days before labeling with the FITC-Annexin V Apoptosis Detection Kit I (BD Biosciences), according to manufacturer's instructions. Briefly, approximately 1 × 105 cells were resuspended in binding buffer 1 × and incubated with FITC-Annexin V and Propidium Iodide (PI) for 15 min, protected from the light at RT. Unstained and single cells stained with FITC-Annexin V or PI were used as controls. Flow cytometry was performed within 1 h on the FACS Canto II ™ system (BD Biosciences) and the results were analyzed using the FlowJo software (BD Biosciences). The LDH was measured using the CytoTox 96® Non-Radioactive Cytotoxicity Assay (Promega) and the details are provided in the Additional file 1: Methods section.

RNA fractionation

Approximately 10 × 106 U2OS cells were collected through trypsinization and washed twice in cold PBS 1x. The cell pellet was resuspended in 1 mL of lysis buffer [10 mM Tris–HCl (pH 8–8.4) 0.14 M NaCl (Merck), 1.5 mM MgCl2 (Sigma-Aldrich) and 0.5% (v/v) NP-40 (Sigma-Aldrich)], before collection of 100 μL of the resulting lysates to be labeled as the whole-cell RNA fraction. The nuclei were spun down at 1 000 g for 4 min. The supernatant, representing the cytoplasmic fraction, was collected and centrifuged at 11 000 g for 1 min to remove the remaining nuclei. Nuclei pellets were resuspended in 1 mL of lysis buffer and 1/10 volume (100 μL) of a detergent solution [3.3% (w/v) sodium deoxycholate (Sigma-Aldrich) and 6.6% (v/v) Tween 40 (Merck)] under slow vortexing. The stripped nuclei were then centrifuged at 1 000 g, for 4 min at 4 °C. The nuclei pellet was then washed once more in lysis buffer and spun down at 1 000 g, for 4 min at 4 °C. The pellet enriched in nucleic content was resuspended in 1 mL of TRIzol using a 21-gauge syringe. TRIzol was also added to the whole-cell RNA and cytoplasmatic fraction and RNA was extracted according to the manufacturer’s instructions.

Decellularization of MSC-derived ECM following osteogenic-inducing stimuli

siH19-MSC or respective control-MSC were cultured for 10 days under osteogenic-inducing conditions to synthesize ECM. The decellularization process was performed in a flow chamber and 1% (v/v) P/S and 0.5% (v/v) fungizone (Capricorn) were added to each solution. The cell culture media was gently removed, cells were washed with PBS 1 × and treated with 0.2% (v/v) Triton X-100 for 10 min at RT. The solution was carefully removed and a 2 M urea (Sigma-Aldrich)/0.2% (v/v) Triton X-100 solution was slowly added, to avoid disturbance of the matrix, and left to incubate for 10 min with gently agitation. The decellularized matrices were treated with 120 Ku/mL of DNase-I (PanReac AppliChem) for 30 min, at 37 °C, and the insoluble ECM layer was washed three times with de-ionized H2O to ensure complete removal of the solubilized material.

Decellularization was confirmed by DAPI (1 μg/mL, for 5 min), hematoxylin and eosin staining and by DNA quantification using Pico-Green dsDNA assay (Quant-iT ™ Pico-Green™ dsDNA assay kit, Thermo Fisher Scientific). Non-decellularized matrices were used as control. Further details are provided in the Additional file 1: Methods section.

Quantification of glycosaminoglycans (GAGS) and collagen content

Content of GAGs in the decellularized matrices was measured using Blyscan™ GAG assay (Biocolor), according to the manufacturer’s instructions (Additional file 1: Methods). The amount of collagenous proteins in the decellularized matrices was measured using the SIRCOL kit (Biocolor), according to the manufacturer’s instructions (Additional file 1: Methods).

Atomic force microscopy (AFM)

AFM topography images of the ECM were obtained with a PicoPlus scanning probe microscope interfaced with a Picoscan 5500 controller (Keysight Technologies) using the PicoView 1.20 software (Keysight Technologies), coupled to an Inverted Optical Microscope (Observer Z1, Zeiss). Each sample was imaged with a 100 × 100 µm [2] piezoelectric scanner. All measurements were performed in contact mode, in air, at RT, using bar-shaped cantilever silicon tips (MLCT-Bio-DC, Bruker) with a spring constant of 0.03 N/m. The scan speed was set at 2.0 Hz.

Recellularization of the siH19-engineered matrices

Decellularized matrices were incubated with cell culture media supplemented with 1% (v/v) P/S and 0.5% (v/v) fungizone at 37 °C in sterile conditions, for 4 h. Naïve MSCs or pre-adipocytes (differentiated from MSCs cultured for 10 days in adipogenesis-inducing conditions) were detached with trypsin–EDTA, centrifuged, resuspended in fresh media with 1% (v/v) P/S and 0.5% (v/v) fungizone, and plated on top of the decellularized sheets at a cell density of 10 000 cells/cm2.

Histology and histomorphometry analysis in bone samples

Bone samples were divided for histologic and histomorphometric evaluation. Samples were cut and fixed in 10% (w/v) formalin (Bio-Optica) for 3 days. For histomorphometric analysis, undecalcified samples were processed for plastic embedding, using a previously described protocol. Briefly, samples were dehydrated in a graded series of ethanol [50, 70 and 100% (v/v)], followed by immersion in xylol for 24 h. The undecalcified bone samples were infiltrated with a plastic embedding mixture using a three-step protocol. In each step, the samples were infiltrated for three consecutive days with daily freshly made solutions, containing 75% (v/v) of methylmethacrylate (MMA, Sigma-Aldrich) and 25% (v/v) of dibutyl phthalate (Prolabo) with increase in concentrations (0 g/mL; 0.01 g/mL and 0.025 g/mL) of benzoyl peroxide (Sigma-Aldrich). Polymerization was carried out at 37 °C for a week. The plastic blocks, containing the processed bone samples, were cut into 7 µm sections using a tungsten knife (Leica). After deplasticization, the sections were stained with toluidine blue staining. Trabecular separation (Tb.Sp) was determined using the Osteomeasure bone histomorphometry software (OsteoMetrics, OsteoMetrics, Inc.). The percentage of adipose tissue was calculated using a 15 to 11 points grid [38, 39].

Statistical analysis

Statistical data analysis was performed using GraphPad Prism 7 (GraphPad Software, Inc.). Gaussian distribution was tested by the Shapiro–Wilk and Kolmogorov–Smirnov tests. For data with a normal distribution, parametric tests were used to evaluate differences between samples, namely a Student t test (between two groups) or one-way ANOVA (more than two groups). For non-normalized data, Mann–Whitney U test (between two groups) or Kruskal–Wallis, followed by Dunn’s multiple comparisons test (more than two groups) were used. Categorical data was tested using the Fisher's exact test. Pearson or Spearman correlation coefficients were calculated when the data followed a parametric and nonparametric distribution, respectively. Statistical significance was considered for p < 0.05 (* p < 0.05; ** p < 0.01 and *** p < 0.001).

Results

H19 controls matrisome of MSC-derived pre-osteoblasts

To screen for matrisome-specific proteins regulated by H19, we performed a mass spectrometry-based quantitative proteomic analysis in MSCs isolated from osteoporotic patients, following induction into the osteogenic lineage. In these cells, the levels of H19 were successfully reduced by transfection with siH19, with an average efficiency of  > 75% and 70% decrease for siH19-1 and siH19-2, respectively (Fig. 1a). Thus, siH19-1 was the sequence selected for the following experiments. Cell fractionation shows that, contrary to other typically nuclear lncRNAs, such as XIST and MALAT1, H19 transcripts are mainly located in the cytoplasm (Fig. 1b). A matrix enrichment buffer was used to facilitate the solubilization of ECM proteins derived from siH19-MSCs and siCTR-MSCs. Proteomic profiling identified 4 020 proteins, considering a minimum of two unique peptides. Results show that 66 proteins were significantly upregulated [Fold-change (FC) > 1.25; p < 0.05], whereas 102 were significantly downregulated (FC < -1.25; p < 0.05), in MSCs with reduced levels of H19 compared with the control (Fig. 1c).

Fig. 1
figure 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.

Full size image

The GO analysis on the biological processes revealed that extracellular matrix organization (GO:0030198; fold-enrichment (FE) = 5.55, p = 3.52 × 10–7; false discovery rate (FDR) = 3.09 × 10–4) and extracellular structure organization (GO:0043062; FE = 5.54, p = 3.63 × 10–7; FDR = 3.01 × 10–4) are among the top 20 terms with higher significance and low FDR (Fig. 1d). Further analysis revealed that the extracellular matrix structural constituent (GO:0005201) is the fourth term with the highest FE score, and is within the top ten with the lowest p-value (p = 2.98 × 10–6; FDR = 2.03 × 10–3) (Fig. 1d). The following proteins are associated with this term: collagen alpha-1(I) chain (COL1A1; P02452); EGF-containing fibulin-like extracellular matrix protein 1 (EFEMP1; Q12805); collagen triple helix repeat containing 1 (CTHRC1; Q96CG8); fibrinogen beta chain (FGB; P02675); fibrillin-1 (FBN1; P35555); fibulin-1 (FBLN1; P23142); elastin (ELN; P15502); vitronectin (VTN; P04004) and lactadherin (MFGE8; Q08431). Among the 20 most enriched cellular components terms are collagen-containing extracellular matrix (GO:0062023; FE = 7.45, p = 1.67 × 10–12; FDR = 4.75 × 10–10) and extracellular matrix (GO:0031012; FE = 5.56, p = 3.09 × 10–10; FDR = 6.16 × 10–8) (Fig. 1d). Also, IPA shows that the growth hormone signaling pathway and the Wnt/β-catenin signaling pathway are predicted to be changed by H19 knockdown.

A Venn diagram was used to compare the differently expressed proteins from the proteomic analysis and the GO cellular terms associated with ECM proteins. Results show that 38 of the differently expressed proteins in the siH19-engineered MSCs versus control MSCs are associated with the cellular functions “extracellular matrix” and “non-structural extracellular” (Fig. 1e). Out of those, 14 proteins are upregulated, while 24 are downregulated following H19 knockdown, compared with the control. Specifically, the differently expressed proteins FBN1 (FC = 1.53, p = 8.06 × 10–6), VTN (FC = − 1.72, p = 5.09 × 10–5) and COL1A1 (FC = − 1.31, p = 0.03) are associated with ECM-related GO terms (Fig. 1f), while CTHRC1 (FC = − 1.374, p = 0.0085) is associated with collagen biogenesis.

ECM components are regulated by H19

To validate novel ECM proteins regulated by H19, the levels of FBN1, VTN and CTHRC1 were evaluated at the transcriptional level. RT-qPCR results show a 42.5% upregulation for FBN1 and a 24.1% and 42.0% downregulation for VTN and CTHRC1, respectively, in siH19-MSCs compared with the control, following 7 days of osteogenic differentiation (Fig. 2a). Furthermore, immunocytochemistry was used to detect these proteins in the MSC-derived matrices. In agreement with the RT-qPCR, results show an increase for FBN1, while VTN is decreased in siH19-MSCs versus control MSCs (Fig. 2b). Although not statistically significant, these also show a tendency for a reduction in CTHRC1 protein levels (Fig. 2b).

Fig. 2
figure 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.

Full size image

Being collagen type I (COL1) a major component of bone ECM, its levels were also quantified by immunocytochemistry. COL1 is reduced in siH19-MSCs versus control MSCs (97.4%), following 7 days of osteogenic induction (Fig. 2c). In agreement, mRNA expression of COL1A1, COL1A2 and COL3A1 is reduced by siH19 (Fig. 2d). The same decrease in the levels of these transcripts is also present when using a distinct sequence for H19 silencing (siH19-2; Additional file 1: Fig. S1).

H19 impairs osteogenic differentiation and mineralization

To evaluate the biological impact of H19 on MSCs, loss-of-function studies were performed. The decrease in H19 levels caused a reduction in expression levels of ALP, OPN, OSX and RUNX2, four early osteogenic markers (Fig. 2e), but did not change the expression of the late osteogenic marker OCN. This was further confirmed through ALP staining, which shows a significant decrease (p < 0.05) in siH19-MSCs compared with the control, after culture for 7 days in osteogenic-inducing conditions (Fig. 2f). Furthermore, H19 knockdown revealed a 79.7% decrease (p < 0.05) in the deposition of calcium nodules after 10 days of differentiation (Fig. 2g), which demonstrates that H19 has an impact on mineralization.

Moreover, results also show a significant reduction in MSCs proliferation, assessed through measurement of the metabolic activity, after H19 depletion compared with control MSCs (Fig. 2h). To further confirm this, cells were stained for the proliferation marker Ki-67. Results show that H19 disturbs the MSCs proliferative capacity, considering the reduction in the percentage of Ki-67+/DAPI+ (p < 0.05) in the siH19-MSCs (Fig. 2h). Nevertheless, flow cytometry shows that the inhibition of H19 does not affect early apoptosis or late apoptosis/cell death (Fig. 2i). Taken together, these results indicate that the inhibition of H19 expression exerted an osteogenic- and proliferative-suppressive effect in MSCs in vitro.

ECM derived from siH19-engineered cells affects the behavior of naïve MSCs

Considering the impact of H19 on the ECM components and osteogenesis, we next evaluated the osteogenic behavior of naïve MSCs on the matrices derived from siH19 cells (Fig. 3a). Firstly, MSCs were transfected with either siH19 or the control, and were allowed to differentiate for 10 days in osteogenic-inducing conditions to promote the deposition of an ECM network. Next, matrices were decellularized and characterized. No DAPI-positive cells or Gill’s hematoxylin stained cells were found in the decellularized matrices, compared with non-decellularized MSCs, which is in agreement with the quantification of low amounts of DNA content (Additional file 1: Fig. S2). There are no differences in GAGs content, when comparing decellularized matrices derived from siH19-MSCs with control MSCs (Fig. 3b). As expected, the collagenous content in decellularized matrices derived from siH19-MSCs, following osteogenic differentiation, is significantly reduced (63.8%, p < 0.05) compared with the controls (Fig. 3c). Furthermore, AFM microscopy images of the decellularized matrices show the presence of a continuous and thicker layer of ECM on the control matrix, in opposition to a thinner and porous network of extracellular matrix fibers on the siH19-derived matrices (Fig. 3d), suggesting an impairment in the ECM production by H19 silencing.

Fig. 3
figure 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).

Full size image

To evaluate if the changes in the matrices secreted by siH19-engineered MSCs influence MSC lineage commitment, naïve cells were used to repopulate the decellularized matrices. Following 7 days of culture, results show that OPN, OSX and RUNX2, three osteogenic key markers, are significantly downregulated, while CEBPB, an adipogenic marker is significantly upregulated, when MSCs are cultured on top of siH19-derived matrices, compared with control-derived matrices (Fig. 3e). No significant differences in gene expression were found for ALP, OCN and CEBPA, nor for the chondrogenic markers, DLX5 and SOX9 (Fig. 3e). Furthermore, immunocytochemistry results show that OPN protein is decreased when naïve MSCs are reseeded on siH19-derived matrices (Fig. 3f), suggesting that the secreted matrices are detrimental for osteogenesis when H19 levels are reduced.

To evaluate if the MSC proliferation is influenced by siH19-derived matrices, naïve MSCs were cultured on top of the decellularized matrices. Results show a decrease in the metabolic activity (Fig. 3g) and in the proliferative capacity (Fig. 3h), evaluated by the percentage of Ki-67+/DAPI+ cells, in naïve MSC cultured on top of siH19-decellularized versus control-decellularized matrices. Also, levels of LDH on the conditioned-media were measured to assess cell viability/cytotoxicity. No differences were found between MSCs cultured on top of the different matrices (Fig. 3i), with LDH levels being similar to those detected when cells were cultured in tissue culture treated plastics (TCPS) flasks, which excludes a cytotoxic effect. Overall, these results indicate that changes in the matrix induced by siH19 affect the lineage commitment and the proliferative capacity of naïve MSCs derived from osteoporotic patients.

ECM derived from siH19-engineered cells affect the behavior of pre-adipocytes

The commitment of MSC into the osteogenic lineage instead of the adipogenic lineage is crucial for a healthy bone. To understand the impact of the matrices on the behavior of pre-adipocytes, MSCs were differentiated for 10 days in adipogenic-inducing conditions (pre-adipocytes), and then used to reseed the decellularized matrix produced by osteogenic-induced MSCs with H19 reduced levels or control. Results show that after 7 days in culture, the expression of CEBPB was significantly increased (p < 0.01) on pre-adipocytes cultured on top of the siH19-derived matrix, compared with cells cultured on top of the control matrix, with no differences regarding osteogenic or chondrogenic markers (ALP, OCN, OPN, OSX and RUNX2; DLX5 and SOX9) (Fig. 4a). Interestingly, changes in the number of cells with lipid droplets (stained with Oil Red O) were observed with only 3 days of culture on top of the decellularized matrices. Specifically, the percentage of Oil Red O+ cells was significantly increased in the siH19-derived matrices (p < 0.001) (Fig. 4b), compared with the control matrices, following normalization to the total number of cells (number of nuclei stained with hematoxylin). Similarly to the results from naïve MSC, we did not detect a cytotoxic effect of the decellularized matrices on the pre-adipocytes (Fig. 4c). Globally, the results suggest that H19 knockdown promotes adipogenesis and increases the number of cells with lipid droplets. Interestingly, in bone clinical samples from osteoporosis patients, we found a significant positive correlation between the percentage of area covered by adipose tissue and the trabecular separation (Tb.Sp), which also positively correlates with age (Fig. 4d).

Fig. 4
figure 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).

Full size image

miR-29c-3p controls H19 expression

To identify potential regulators of H19, an in silico analysis on miRNAs was performed using the ENCORI platform (http://starbase.sysu.edu.cn/panCancer.php), and potential binding sites for the miR-29 family, including miR-29a, miR-29b and miR-29c members, are shown in Additional file 1: Fig. S3. To understand the relevance of those miRNAs in osteoporosis, we evaluated their expression levels in bone samples from osteoporotic fragility fractures (FF) and osteoarthritis (OA) patients. RT-qPCR results show that miR-29b-3p is poorly expressed in the bone samples or even undetected (Cq > 35). However, miR-29a-3p and miR-29c-3p were expressed and significantly decreased in FF versus OA bone samples (Fig. 5a). A multivariate analysis was performed to exclude possible associations between miRNA expression and clinically relevant variables, including age. When using miRNA expression and age as independent variables, and disease as a dependent variable, the results from the multivariate linear regression analysis with the stepwise forward method show that only miR-29c-3p, but not miR-29a-3p, is statistically significant (Fig. 5b), being a strong predictive factor of the disease. The levels of miR-29c-3p were also decreased in MSC isolated from osteoporosis patients compared with healthy donors. In the same samples, H19 expression shows an antagonistic profile, being upregulated in MSC derived from osteoporosis patients versus healthy donors (Fig. 5c). Interestingly, during the early stages of osteogenic differentiation H19 expression is increased in MSC from healthy donors (p < 0.05, day 7). However, this increase is not present during osteogenic differentiation of osteoporosis-derived MSC, even in the presence of osteogenic-inducing stimuli (Additional file 1: Fig. S4). This reinforces the need for studies to be conducted in samples derived from osteoporosis patients.

Fig. 5
figure 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)

Full size image

Next, to evaluate if miR-29c-3p regulates H19, the expression levels of H19 were tested after modulation of miR-29c-3p in MSCs cultured for 3 days in osteogenic-inducing conditions (Additional file 1: Fig. S5). Results show that miR-29c-3p overexpression significantly decreases H19 (FC = 0.58), while miR-29c-3p inhibition increases H19 levels (FC = 2.44). (Fig. 5d).

miR-29c-3p reduces collagen expression and MSC proliferation but do not impact ALP and mineralization

To understand if miR-29c-3p impacts H19-mediated processes, collagen secretion, ALP level, mineralization, proliferation and apoptosis in MSCs were investigated. Immunostaining against COL1, after 7 days of culture in osteogenic-inducing conditions, shows that COL1 proteins are reduced by 94.4% when MSCs are transfected with miR-29c-3p mimics, while COL1 protein levels are highly increased following inhibition of miR-29c-3p, compared with the respective controls (Fig. 6a). At the transcription level, a reduction was found for COL1A1, COL1A2 and COL3A1 in miR-29c-3p-overexpressing-MSCs, whereas the opposite was observed in miR-29c-3p-depleted-MSCs (Fig. 6b). Interestingly, in osteoporotic bone samples, there is a negative correlation between miR-29c-3p expression levels and COL1A1 (Fig. 6c). To test if miR-29c-3p regulates osteogenesis, osteogenic differentiation markers, ALP staining and mineralization were analyzed. However, no differences were found, suggesting that the modulation of miR-29c-3p does not impair osteogenic differentiation, ALP or calcium deposition (Fig. 6d, 6e and 6f). On the other hand, differences were found regarding proliferation. The metabolic activity is significantly increased when miR-29c-3p levels are reduced and, although not statistically significant, there is a tendency for a reduction in the metabolic activity following miR-29c-3p overexpression (Fig. 6g). These results were confirmed when analyzing the proliferation marker Ki-67. The percentage of Ki-67+/DAPI+ is significantly decreased by miR-29c-3p and increased following miR-29c-3p inhibition (Fig. 6g). Furthermore, analysis of Annexin V/PI staining by flow cytometry showed that the miR-29c-3p overexpression significantly decreases the percentage of viable cells, while increasing the percentage of cells in late apoptosis/cell death (Fig. 6h). Taken together, these results suggest that miR-29c-3p regulates levels of collagen proteins and MSCs proliferation, but do not impair ALP or mineralization.

Fig. 6
figure 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).

Full size image

Discussion

In osteoporosis, the composition and structure of the organic and inorganic bone ECM are affected by the disruption in the interplay between osteoblasts and osteoclasts [11, 19]. In order to restore bone ECM, it is essential to identify the key parameters that regulate ECM components. On one hand, ECM proteins are regulated by RNA transcripts; on the other hand, the ECM is able to modulate the transcription of certain RNAs. LncRNAs have been gaining interest as therapeutic targets for bone diseases, due to their involvement in bone metabolism, remodeling and metastasis [6]. However, up-to-date, the role of lncRNAs has only been revealed for a small portion of transcripts.

LncRNA H19 is a pro-osteogenic molecule that has been found to be deregulated in osteoporosis, both in human patients [40, 41] and animal models [34, 41, 42]. In this study, we performed, for the first time, an ECM-enriched proteomic analysis in H19 depleted MSCs derived from osteoporotic patients, to screen for novel matrix proteins regulated by H19. Besides collagen synthesis, our study reveals that H19 modulates multiple ECM proteins, including FBN1, VTN and CTHRC1, among others. Specifically, FBN1 is a structural glycoprotein, that together with fibrillin 2 (FBN2), composes 1–3% of the skeleton [43]. Both proteins regulate the bioavailability of locally produced transforming growth factor beta (TGF-β) and bone morphogenetic protein (BMP) complexes, affecting osteoblast maturation [44,45,46]. Mutations in FBN1 result in skeletal abnormalities [47]. In contrast, the ECM glycoprotein VTN is a key positive regulator of the mammalian tissue repair and remodeling activity [48,49,50]. Recently, a vitronectin-derived peptide was shown to reverse ovariectomy-induced bone loss due to its dual antagonist role on bone formation and resorption [51]. Moreover, it mediates cell attachment, spreading and migration to the ECM [51, 52], and was identified as a survival factor for MSCs submitted to serum deprivation-induced apoptosis [53], highlighting its importance as a strategy to increase MSCs survival post-transplantation. Furthermore, the adhesion of MSC to VTN promotes osteogenic differentiation by increasing OCN, COL1 and ALP levels [54]. It also induces mineralization in the presence or absence of osteogenic-inducing supplements. Considering this, the decrease in VTN caused by H19 depletion may partly explain the decrease in the osteogenic capacity of MSCs, when reseeded on top of the siH19-decellularized matrices. Futhermore, transcript levels of the glycoprotein CTHRC1 were reduced by H19 silencing. This protein has been widely described to promote cell migration [55] and abundant expression levels of CTHRC1 are detected in the bone matrix during embryogenesis and in adult mice [56]. Ablation of CTHRC1 results in the impairment of bone formation [57], while increased levels of CTHRC1 stimulate differentiation and mineralization of osteoprogenitor cells [57, 58].

The present study demonstrates that H19 is crucial for ECM protein production (including collagen, FBN1, VNT, CTHRC1, among others), MSC differentiation into osteoblasts, mineralization and proliferative capacity. Importantly, lncRNAs not only mediate intracellular molecular interactions but may have an indirect effect on the interactions between the ECM and the cells. As such, we demonstrated that the decellularized matrices produced by H19-depleted MSCs, cultured under osteogenic-inducing conditions, are sufficient to shift the MSCs commitment into the adipogenic, in detriment of the osteogenic lineage, and to negatively impact MSCs proliferation. Considering levels of multiple ECM proteins and ECM-associated proteins are modulated by H19, this effect is probably the consequence of the simultaneous changes in multiple proteins, rather than individual molecules.

Furthermore, we showed that the H19 expression is regulated by miR-29c-3p, whose levels are decreased in bone samples from osteoporotic patients, and that it negatively changes COL1 expression, which is in agreement with previous studies [59, 60]. In a recent study published by our group, miR-29c-3p was identified as a marker of secondary osteoporosis in multiple myeloma plasma samples [61]. Modulation of miR-29c-3p affects MSC proliferative capacity, without impacting cell death or osteogenic differentiation and mineralization in the early stages of differentiation. Thus, although miR-29c-3p regulates H19 expression, the functions of these two transcripts do not fully overlap, since differences were found for H19 regarding osteogenesis, mineralization and several bone ECM proteins. Importantly, the regulation of H19 by miRNAs is extended far beyond the miR-29 family, as several other miRNAs have been shown to interact with H19 [28, 32, 33]. On the other hand, the function of miR-29c-3p is not only determined by the targeting of a single molecule but by the balance in levels of all the regulated transcripts, which might explain why the functions of the lncRNA H19 and the miR-29c-3p are only partially complementary.

Collectively, the findings presented in this study support H19 as a potential therapeutic target for osteoporosis and repair of fragility fractures through the modulation of the ECM components and the ECM-cell interactions.

Conclusions

Matrisome signature is altered following H19 silencing and among the proteins impacted were FBN1, VTN and CTHRC1. siH19-engineered matrices impair MSC proliferation and fate, while enhancing lipid droplets formation in pre-adipocytes. H19 and miR-29c-3p are antagonistic partners with partially independent functional profiles.

In summary, this study advances lncRNA-based therapeutics for the control of ECM homeostasis.

Availability of data and materials

The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE [62] partner repository with the dataset identifier PXD039001 and 10 .6019/PXD039001.

Abbreviations

ALP:

Alkaline phosphatase

AFM:

Atomic force microscopy

CEBPA:

CCAAT Enhancer Binding Protein Alpha

CEBPB:

CCAAT Enhancer Binding Protein Beta

CTHRC1:

Collagen triple helix repeat containing 1

COL1A1:

Collagen Type I Alpha 1 Chain

COL1A2:

Collagen Type I Alpha 2 Chain

COL 1:

Collagen type I protein

COL3A1:

Collagen Type III Alpha 1 Chain

DLX5:

Distal-Less Homeobox 5

ECM:

Extracellular matrix

FDR:

False discovery rate

FBN1:

Fibrillin-1

FE:

Fold enrichment

FF:

Fragility fractures

GO:

Gene ontology

GAGs:

Glycosaminoglycans

IBMX:

3-Isobutyl-1-methylxanthine

LDH:

Lactate dehydrogenase

lncRNAs:

Long non-coding RNAs

MSCs:

Mesenchymal Stem/Stromal Cells

OA:

Osteoarthritis

RFU:

Relative fluorescence units

RUNX2:

RUNX Family Transcription Factor 2

OPN:

Secreted Phosphoprotein 1/Osteopontin

siH19:

Small interfering RNA against H19

SOX9:

SRY-Box Transcription Factor 9

TCPS:

Tissue culture treated plastics

Tb.Sp:

Trabecular separation

VTN:

Vitronectin

References

  1. Bartl R, Frisch B. Osteoporosis. Berlin: Springer; 2004. https://doi.org/10.1007/978-3-662-09163-0.

    Book  Google Scholar 

  2. Kanis JA, et al. SCOPE 2021: a new scorecard for osteoporosis in Europe. Arch Osteoporos. 2021;16:82.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Harvey N, Dennison E, Cooper C. Osteoporosis: impact on health and economics. Nat Rev Rheumatol. 2010;6:99–105.

    Article  PubMed  Google Scholar 

  4. Hernlund E, et al. Osteoporosis in the European Union: medical management, epidemiology and economic burden. A report prepared in collaboration with the International Osteoporosis Foundation (IOF) and the European Federation of Pharmaceutical Industry Associations (EFPIA). Arch Osteoporos. 2013;8:136.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Kanis JA, McCloskey EV, Johansson H, Oden A. Approaches to the targeting of treatment for osteoporosis. Nat Rev Rheumatol. 2009;5:425–31.

    Article  CAS  PubMed  Google Scholar 

  6. Silva AM, et al. Long noncoding RNAs: a missing link in osteoporosis. Bone Res. 2019;7:10.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Cranney A, Jamal SA, Tsang JF, Josse RG, Leslie WD. Low bone mineral density and fracture burden in postmenopausal women. CMAJ. 2007;177:575–80.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Bouxsein ML. Bone quality: where do we go from here? Osteoporos Int. 2003;14:118–27.

    Article  Google Scholar 

  9. Zhang D, et al. Osteoporosis-decreased extracellular matrix stiffness impairs connexin 43-mediated gap junction intercellular communication in osteocytes. Acta Biochim Biophys Sin. 2020;52:517–26.

    Article  PubMed  Google Scholar 

  10. Ozasa R, et al. Osteoporosis changes collagen/apatite orientation and young’s modulus in vertebral cortical bone of rat. Calcif Tissue Int. 2019;104:449–60.

    Article  CAS  PubMed  Google Scholar 

  11. Rodrigues AM, et al. Low osteocalcin/collagen type I bone gene expression ratio is associated with hip fragility fractures. Bone. 2012;51:981–9.

    Article  CAS  PubMed  Google Scholar 

  12. De-Ugarte L, et al. MiRNA profiling of whole trabecular bone: identification of osteoporosis-related changes in MiRNAs in human hip bones. BMC Med Genomics. 2015;8:75.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Hopwood B, Tsykin A, Findlay DM, Fazzalari NL. Gene expression profile of the bone microenvironment in human fragility fracture bone. Bone. 2009;44:87–101.

    Article  CAS  PubMed  Google Scholar 

  14. Kafantari H, Kounadi E, Fatouros M, Milonakis M, Tzaphlidou M. Structural alterations in rat skin and bone collagen fibrils induced by ovariectomy. Bone. 2000;26:349–53.

    Article  CAS  PubMed  Google Scholar 

  15. Müller R, et al. Analysis of microscopic bone properties in an osteoporotic sheep model: a combined biomechanics, FE and ToF-SIMS study. J R Soc Interface. 2019;16:20180793.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Nyman JS, Makowski AJ. The contribution of the extracellular matrix to the fracture resistance of bone. Curr Osteoporos Rep. 2012;10:169–77.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Schlesinger PH, et al. Cellular and extracellular matrix of bone, with principles of synthesis and dependency of mineral deposition on cell membrane transport. Am J Physiol Cell Physiol. 2020;318:C111–24.

    Article  CAS  PubMed  Google Scholar 

  18. Al-Bari AA, Al Mamun, A. Current advances in regulation of bone homeostasis. FASEB Bioadv 2, 668–679 (2020).

  19. Tella SH, Gallagher JC. Prevention and treatment of postmenopausal osteoporosis. J Steroid Biochem Mol Biol. 2014;142:155–70.

    Article  CAS  PubMed  Google Scholar 

  20. Bethel M, Chitteti BR, Srour EF, Kacena MA. The changing balance between osteoblastogenesis and adipogenesis in aging and its impact on hematopoiesis. Curr Osteoporos Rep. 2013;11:99–106.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Chen Q, et al. Fate decision of mesenchymal stem cells: adipocytes or osteoblasts? Cell Death Differ. 2016;23:1128–39.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Han L, et al. The shift in the balance between osteoblastogenesis and adipogenesis of mesenchymal stem cells mediated by glucocorticoid receptor. Stem Cell Res Ther. 2019;10:377.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Li J, Zuo B, Zhang L, Dai L, Zhang X. Osteoblast versus adipocyte: bone marrow microenvironment-guided epigenetic control. Case Rep Orthop Res. 2018;1:2–18.

    Article  Google Scholar 

  24. Hassan MQ, Tye CE, Stein GS, Lian JB. Non-coding RNAs: epigenetic regulators of bone development and homeostasis. Bone. 2015;81:746–56.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Patil S, Dang K, Zhao X, Gao Y, Qian A. Role of LncRNAs and CircRNAs in bone metabolism and osteoporosis. Front Genet. 2020;11: 584118.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Fu X-D. Non-coding RNA: a new frontier in regulatory biology. Natl Sci Rev. 2014;1:190–204.

    Article  CAS  PubMed  Google Scholar 

  27. Aurilia C, et al. The involvement of long non-coding RNAs in bone. Int J Mol Sci. 2021;22:3909.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Liang W-C, et al. H19 activates Wnt signaling and promotes osteoblast differentiation by functioning as a competing endogenous RNA. Sci Rep. 2016;6:20121.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Li G, et al. Long non-coding RNA-H19 stimulates osteogenic differentiation of bone marrow mesenchymal stem cells via the microRNA-149/SDF-1 axis. J Cell Mol Med. 2020;24:4944–55.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Huang Y, Zheng Y, Jia L, Li W. Long noncoding RNA H19 promotes osteoblast differentiation via TGF-β1/Smad3/HDAC signaling pathway by deriving miR-675. Stem Cells. 2015;33:3481–92.

    Article  CAS  PubMed  Google Scholar 

  31. Behera J, Kumar A, Voor MJ, Tyagi N. Exosomal lncRNA-H19 promotes osteogenesis and angiogenesis through mediating Angpt1/Tie2-NO signaling in CBS-heterozygous mice. Theranostics. 2021;11:7715–34.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Wu J, et al. Long non-coding RNA H19 mediates mechanical tension-induced osteogenesis of bone marrow mesenchymal stem cells via FAK by sponging miR-138. Bone. 2018;108:62–70.

    Article  CAS  PubMed  Google Scholar 

  33. Wang Y, et al. Long noncoding RNA H19 mediates LCoR to impact the osteogenic and adipogenic differentiation of mBMSCs in mice through sponging miR-188. J Cell Physiol. 2018;233:7435–46.

    Article  CAS  PubMed  Google Scholar 

  34. Li B, et al. Overexpression of DNMT1 leads to hypermethylation of H19 promoter and inhibition of Erk signaling pathway in disuse osteoporosis. Bone. 2018;111:82–91.

    Article  PubMed  Google Scholar 

  35. Almeida MI, et al. miR-195 in human primary mesenchymal stromal/stem cells regulates proliferation, osteogenesis and paracrine effect on angiogenesis. Oncotarget. 2016;7:7–22.

    Article  PubMed  Google Scholar 

  36. Bustin SA, et al. The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem. 2009;55:611–22.

    Article  CAS  PubMed  Google Scholar 

  37. Moura SR, et al. miR-99a in bone homeostasis: regulating osteogenic lineage commitment and osteoclast differentiation. Bone. 2020;134: 115303.

    Article  CAS  PubMed  Google Scholar 

  38. Gundersen HJG, et al. Some new, simple and efficient stereological methods and their use in pathological research and diagnosis. APMIS. 1988;96:379–94.

    Article  CAS  PubMed  Google Scholar 

  39. Gundersen HJG, Jensen EB. The efficiency of systematic sampling in stereology and its prediction*. J Microsc. 1987;147:229–63.

    Article  CAS  PubMed  Google Scholar 

  40. Xiaoling G, Shuaibin L, Kailu L. MicroRNA-19b-3p promotes cell proliferation and osteogenic differentiation of BMSCs by interacting with lncRNA H19. BMC Med Genet. 2020;21:11.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Li T, Jiang H, Li Y, Zhao X, Ding H. Estrogen promotes lncRNA H19 expression to regulate osteogenic differentiation of BMSCs and reduce osteoporosis via miR-532-3p/SIRT1 axis. Mol Cell Endocrinol. 2021;527: 111171.

    Article  CAS  PubMed  Google Scholar 

  42. Li B, et al. LncRNA-H19 modulates Wnt/β-catenin signaling by targeting Dkk4 in Hindlimb unloaded rat. Orthop Surg. 2017;9:319–27.

    Article  PubMed  PubMed Central  Google Scholar 

  43. Smaldone S, Ramirez F. Fibrillin microfibrils in bone physiology. Matrix Biol. 2016;52–54:191–7.

    Article  PubMed  Google Scholar 

  44. Nistala H, et al. Fibrillin-1 and -2 differentially modulate endogenous TGF-β and BMP bioavailability during bone formation. J Cell Biol. 2010;190:1107–21.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Smaldone S, Carta L, Ramirez F. Establishment of fibrillin-deficient osteoprogenitor cell lines identifies molecular abnormalities associated with extracellular matrix perturbation of osteogenic differentiation. Cell Tissue Res. 2011;344:511–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Cook JR, et al. Generation of Fbn1 conditional null mice implicates the extracellular microfibrils in osteoprogenitor recruitment. Genesis. 2012;50:635–41.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Keene DR, et al. Fibrillin-1 in human cartilage: developmental expression and formation of special banded fibers. J Histochem Cytochem. 1997;45:1069–82.

    Article  CAS  PubMed  Google Scholar 

  48. Leavesley DI, et al. Vitronectin–master controller or micromanager? IUBMB Life. 2013;65:807–18.

    CAS  PubMed  Google Scholar 

  49. Chow S, Di Girolamo N. Vitronectin: a migration and wound healing factor for human corneal epithelial cells. Invest Ophthalmol Vis Sci. 2014;55:6590–600.

    Article  CAS  PubMed  Google Scholar 

  50. Scaffidi AK, Moodley YP, Weichselbaum M, Thompson PJ, Knight DA. Regulation of human lung fibroblast phenotype and function by vitronectin and vitronectin integrins. J Cell Sci. 2001;114:3507–16.

    Article  CAS  PubMed  Google Scholar 

  51. Min S-K, Kang HK, Jung SY, Jang DH, Min B-M. A vitronectin-derived peptide reverses ovariectomy-induced bone loss via regulation of osteoblast and osteoclast differentiation. Cell Death Differ. 2018;25:268–81.

    Article  CAS  PubMed  Google Scholar 

  52. Schvartz I, Seger D, Shaltiel S. Vitronectin. Int J Biochem Cell Biol. 1999;31:539–44.

    Article  CAS  PubMed  Google Scholar 

  53. Goyal U, Ta M. A novel role of vitronectin in promoting survival of mesenchymal stem cells under serum deprivation stress. Stem Cell Res Ther. 2020;11:181.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  54. Salasznyk RM, Williams WA, Boskey A, Batorsky A, Plopper GE. Adhesion to vitronectin and collagen I promotes osteogenic differentiation of human mesenchymal stem cells. J Biomed Biotechnol. 2004;2004:24–34.

    Article  PubMed  PubMed Central  Google Scholar 

  55. Myngbay A, Manarbek L, Ludbrook S, Kunz J. The role of collagen triple helix repeat-containing 1 protein (CTHRC1) in rheumatoid arthritis. Int J Mol Sci. 2021;22:2426.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Durmus T, et al. Expression analysis of the novel gene collagen triple helix repeat containing-1 (Cthrc1). Gene Expr Patterns. 2006;6:935–40.

    Article  CAS  PubMed  Google Scholar 

  57. Kimura H, et al. Cthrc1 is a positive regulator of osteoblastic bone formation. PLoS ONE. 2008;3: e3174.

    Article  PubMed  PubMed Central  Google Scholar 

  58. Wang C, et al. CTHRC1 promotes osteogenic differentiation of periodontal ligament stem cells by regulating TAZ. J Mol Hist. 2017;48:311–9.

    Article  CAS  Google Scholar 

  59. Sengupta S, et al. MicroRNA 29c is down-regulated in nasopharyngeal carcinomas, up-regulating mRNAs encoding extracellular matrix proteins. Proc Natl Acad Sci U S A. 2008;105:5874–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. van Rooij E, et al. Dysregulation of microRNAs after myocardial infarction reveals a role of miR-29 in cardiac fibrosis. Proc Natl Acad Sci U S A. 2008;105:13027–32.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Moura SR, et al. Circulating microRNAs correlate with multiple myeloma and skeletal osteolytic lesions. Cancers. 2021;13:5258.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  62. Perez-Riverol Y, et al. The PRIDE database resources in 2022: a hub for mass spectrometry-based proteomics evidences. Nucleic Acids Res. 2022;50:D543–52.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

The authors would like to thank to Serviço de Ortopedia of Centro Hospitalar Universitário de São João (CHUSJ) for kindly donating the bone samples of osteoporotic patients. We also thank to IMM Biobank for providing RNA from bone samples. We also acknowledge the Advanced Light Microscopy, BioSciences Screening and Histology and Electron Microscopy, all members of the PPBI (PPBI-POCI-01-0145-FEDER-022122); Cell Culture and Genotyping (CCGEN); and Proteomics (ROTEIRO/0028/2013; LISBOA-01-0145-FEDER-022125), services/platforms at i3S for the technical support. We also thank to Dr. Joana Caldeira for the advice on the decellularization process; and to Dr. João Frazão and Dr. Manuel Pestana, from the Departamento de Nefrologia of CHUSJ and from the Nephrology & Infectious Diseases group at i3S, for allowing us to perform the histomorphometric analysis in their department.

Funding

The project was supported by Portuguese funds through FCT-Fundação para a Ciência e a Tecnologia (FCT)/Ministério da Ciência, Tecnologia e Ensino Superior in the framework of the project POCI-010145-FEDER-031402-R2Bone (FEDER-Fundo Europeu de Desenvolvimento Regional funds through the COMPETE 2020-Operacional Programme for Competitiveness and Internationalisation—POCI, Portugal 2020). The study was also supported by AO CMF (AOCMFS-21-23A). SRM and MIA are supported by FCT (SFRH/BD/147229/2019 and BiotechHealth Program; CEECINST/00091/2018/CP1500/CT0011, respectively).

Author information

Authors and Affiliations

  1. Instituto de Investigação E Inovação Em Saúde (i3S), Universidade Do Porto, Rua Alfredo Allen, 208, 4200-135, Porto, Portugal

    Sara Reis Moura, Jaime Freitas, Cláudia Ribeiro-Machado, Nuno Neves, Mário Adolfo Barbosa & Maria Inês Almeida

  2. Instituto de Engenharia Biomédica (INEB), Universidade Do Porto, Porto, Portugal

    Sara Reis Moura, Jaime Freitas, Cláudia Ribeiro-Machado, Nuno Neves, Mário Adolfo Barbosa & Maria Inês Almeida

  3. Instituto de Ciências Biomédicas Abel Salazar (ICBAS), Universidade Do Porto, Porto, Portugal

    Sara Reis Moura, Mário Adolfo Barbosa & Maria Inês Almeida

  4. Departamento de Ortopedia, Centro Hospitalar Universitário São João (CHUSJ), Porto, Portugal

    Jorge Lopes & Nuno Neves

  5. Hospital CUF, Porto, Portugal

    Nuno Neves

  6. Faculdade de Medicina (FMUP), Universidade Do Porto, Porto, Portugal

    Nuno Neves

  7. NOVA Medical School - Faculdade de Ciências Médicas, Universidade Nova de Lisboa, Lisbon, Portugal

    Helena Canhão & Ana Maria Rodrigues

  8. Comprehensive Health Research Center (CHRC), Universidade Nova de Lisboa, Lisbon, Portugal

    Helena Canhão & Ana Maria Rodrigues

Authors
  1. Sara Reis Moura
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jaime Freitas
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Cláudia Ribeiro-Machado
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Jorge Lopes
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Nuno Neves
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Helena Canhão
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Ana Maria Rodrigues
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Mário Adolfo Barbosa
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Maria Inês Almeida
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

S.R.M. and M.I.A. were involved in conceptualization; S.R.M., J.F., J.L., N.N., H.C., A.M.R and M.A.B helped in methodology; S.R.M. and J.F. contributed to data collection; S.R.M. and C.R.-M. were involved in statistical analysis; S.R.M., J.F. and M.I.A. helped in writing—original draft preparation; S.R.M., J.F., J.L., N.N., H.C., A.M.R., M.A.B. and M.I.A. helped in writing-review and editing; M.I.A. and M.A.B contributed to funding acquisition. All authors have read and agreed to the published version of the manuscript.

Corresponding author

Correspondence to Maria Inês Almeida.

Ethics declarations

Ethics approval and consent to participate

Ethics Committee of the Centro Hospitalar Universitário São João; Approval Reference number 70/18.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Fig. S1.

Effect of siH19-2 on the expression levels of collagens, osteogenic markers and other H19-target candidates, assessed through RT-qPCR, 3 days after culture under osteogenic-inducing conditions (N = 6). Fig. S2. Decellularization of MSCs matrices. Primary MSCs from osteoporotic patients were differentiated for 10 days under osteogenic-inducing conditions before decellularization. After treatment with the decellularization buffer (urea 2 M), samples were stained with a) DAPI (blue) to visualize nuclei (scale: 100 μm) and b) hematoxylin (blue-purple color) to visualize nuclei and matrix integrity (scale: 100 μm). c) DNA and sGAGs content were measured. Non-decellularized matrices were used as a control. Fig. S3. In silico predictions for the binding site between miR-29-a/b/c and lncRNA H19. Fig. S4. Expression profile of H19 during early osteogenic differentiation stages in MSCs from healthy donors (N = 4) and osteoporotic patients (N = 4). Fig. S5. miR-29c-3p expression after transfection of MSCs with miR-29c-3p mimics, miR-29c-3p inhibitor or respective controls (NC-mimics and NC-inhibitor), and cultured for 3 days in osteogenic-inducing conditions (N = 6). Table SI. Donors of Human Mesenchymal Stem/Stromal Cells. Table SII. Primers used for reverse transcription quantitative real-time PCR. Table SIII. Mature miRNAs sequences according to miRbase annotations (http://www.mirbase.org/). Table SIV. Antibodies used for immunocytochemistry stainings.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Moura, S.R., Freitas, J., Ribeiro-Machado, C. et al. Long non-coding RNA H19 regulates matrisome signature and impacts cell behavior on MSC-engineered extracellular matrices. Stem Cell Res Ther 14, 37 (2023). https://doi.org/10.1186/s13287-023-03250-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13287-023-03250-6

Keywords

Stem Cell Research & Therapy

ISSN: 1757-6512