Transcriptional profiling of murine osteoblast differentiation based on RNA-seq expression analyses

Osteoblast MR deficiency protects against adverse ventricular remodeling after myocardial infarction

RATIONALE

Mineralocorticoid receptor (MR) antagonists have been clinically used to treat heart failure. However, the underlying cellular and molecular mechanisms remain incompletely understood.

METHODS AND RESULTS

Using osteoblast MR knockout (MR^obko) mouse in combination with myocardial infarction (MI) model, we demonstrated that MR deficiency in osteoblasts significantly improved cardiac function, promoted myocardial healing, as well as attenuated cardiac hypertrophy, fibrosis and inflammatory response after MI. Gene expression profiling using RNA sequencing revealed suppressed expression of osteocalcin (OCN) in calvaria from MR^obko mice compared to littermate control (MR^fl/fl) mice with or without MI. Plasma levels of undercarboxylated OCN (ucOCN) were also markedly decreased in MR^obko mice compared to MR^fl/fl mice. Administration of ucOCN abolished the protective effects of osteoblast MR deficiency on infarcted hearts. Mechanistically, ucOCN treatment promoted proliferation and inflammatory cytokine secretion in macrophages. Spironolactone, an MR antagonist, significantly inhibited the expression and secretion of OCN in post-MI mice. More importantly, spironolactone decreased plasma levels of ucOCN and inflammatory cytokines in heart failure patients.

CONCLUSIONS

MR deficiency in osteoblasts alleviates pathological ventricular remodeling after MI, likely through its regulation on OCN. Spironolactone may work through osteoblast MR/OCN axis to exert its therapeutic effects on pathological ventricular remodeling and heart failure in mice and human patients.

Titanium with nanotopography attenuates the osteoclast-induced disruption of osteoblast differentiation by regulating histone methylation

The bone remodeling process is crucial for titanium (Ti) osseointegration and involves the crosstalk between osteoclasts and osteoblasts. Considering the high osteogenic potential of Ti with nanotopography (Ti Nano) and that osteoclasts inhibit osteoblast differentiation, we hypothesized that nanotopography attenuate the osteoclast-induced disruption of osteoblast differentiation. Osteoblasts were co-cultured with osteoclasts on Ti Nano and Ti Control and non-co-cultured osteoblasts were used as control. Gene expression analysis using RNAseq showed that osteoclasts downregulated the expression of osteoblast marker genes and upregulated genes related to histone modification and chromatin organization in osteoblasts grown on both Ti surfaces. Osteoclasts also inhibited the mRNA and protein expression of osteoblast markers, and such effect was attenuated by Ti Nano. Also, osteoclasts increased the protein expression of H3K9me2, H3K27me3 and EZH2 in osteoblasts grown on both Ti surfaces. ChIP assay revealed that osteoclasts increased accumulation of H3K27me3 that represses the promoter regions of Runx2 and Alpl in osteoblasts grown on Ti Control, which was reduced by Ti Nano. In conclusion, these data show that despite osteoclast inhibition of osteoblasts grown on both Ti Control and Ti Nano, the nanotopography attenuates the osteoclast-induced disruption of osteoblast differentiation by preventing the increase of H3K27me3 accumulation that represses the promoter regions of some key osteoblast marker genes. These findings highlight the epigenetic mechanisms triggered by nanotopography to protect osteoblasts from the deleterious effects of osteoclasts, which modulate the process of bone remodeling and may benefit the osseointegration of Ti implants.

Morphology-Based Deep Learning Approach for Predicting Osteogenic Differentiation

Early, high-throughput, and accurate recognition of osteogenic differentiation of stem cells is urgently required in stem cell therapy, tissue engineering, and regenerative medicine. In this study, we established an automatic deep learning algorithm, i.e., osteogenic convolutional neural network (OCNN), to quantitatively measure the osteogenic differentiation of rat bone marrow mesenchymal stem cells (rBMSCs). rBMSCs stained with F-actin and DAPI during early differentiation (day 0, 1, 4, and 7) were captured using laser confocal scanning microscopy to train OCNN. As a result, OCNN successfully distinguished differentiated cells at a very early stage (24 h) with a high area under the curve (AUC) (0.94 ± 0.04) and correlated with conventional biochemical markers. Meanwhile, OCNN exhibited better prediction performance compared with the single morphological parameters and support vector machine. Furthermore, OCNN successfully predicted the dose-dependent effects of small-molecule osteogenic drugs and a cytokine. OCNN-based online learning models can further recognize the osteogenic differentiation of rBMSCs cultured on several material surfaces. Hence, this study initially demonstrated the foreground of OCNN in osteogenic drug and biomaterial screening for next-generation tissue engineering and stem cell research.

Perspective of the GEMSTONE Consortium on Current and Future Approaches to Functional Validation for Skeletal Genetic Disease Using Cellular, Molecular and Animal-Modeling Techniques

The availability of large human datasets for genome-wide association studies (GWAS) and the advancement of sequencing technologies have boosted the identification of genetic variants in complex and rare diseases in the skeletal field. Yet, interpreting results from human association studies remains a challenge. To bridge the gap between genetic association and causality, a systematic functional investigation is necessary. Multiple unknowns exist for putative causal genes, including cellular localization of the molecular function. Intermediate traits ("endophenotypes"), e.g. molecular quantitative trait loci (molQTLs), are needed to identify mechanisms of underlying associations. Furthermore, index variants often reside in non-coding regions of the genome, therefore challenging for interpretation. Knowledge of non-coding variance (e.g. ncRNAs), repetitive sequences, and regulatory interactions between enhancers and their target genes is central for understanding causal genes in skeletal conditions. Animal models with deep skeletal phenotyping and cell culture models have already facilitated fine mapping of some association signals, elucidated gene mechanisms, and revealed disease-relevant biology. However, to accelerate research towards bridging the current gap between association and causality in skeletal diseases, alternative in vivo platforms need to be used and developed in parallel with the current -omics and traditional in vivo resources. Therefore, we argue that as a field we need to establish resource-sharing standards to collectively address complex research questions. These standards will promote data integration from various -omics technologies and functional dissection of human complex traits. In this mission statement, we review the current available resources and as a group propose a consensus to facilitate resource sharing using existing and future resources. Such coordination efforts will maximize the acquisition of knowledge from different approaches and thus reduce redundancy and duplication of resources. These measures will help to understand the pathogenesis of osteoporosis and other skeletal diseases towards defining new and more efficient therapeutic targets.

Cdkn1a deletion or suppression by cyclic stretch enhance the osteogenic potential of bone marrow mesenchymal stem cell-derived cultures

CDKN1A/P21 is a potent inhibitor of cell cycle progression and its overexpression is thought to be associated with inhibition of normal bone regenerative osteogenesis during spaceflight. To test whether CDKN1A/P21 regulates osteogenesis in response to mechanical loading we studied cyclic stretch versus static culture of Cdkn1a^-/- (null) or wildtype primary mouse bone marrow osteoprogenitors during 21-day ex-vivo mineralization assays. Cyclically stretched Cdkn1a^-/- cells are 3.95-fold more proliferative than wildtype, while static Cdkn1a^-/- cells show a 2.50-fold increase. Furthermore, stage-specific single cell RNAseq analyses show expression of Cdkn1a is strongly suppressed by cyclic stretch in early and late osteoblasts, and minimally in the progenitor population. Lastly, both stretch and/or Cdkn1a deletion cause population shift from osteoprogenitors to osteoblasts, also indicating increased differentiation. Collectively, our results support the hypothesis that Cdkn1a constitutively plays a mechano-reversible anti-proliferative role during osteogenesis and suggests a new molecular target to counter regenerative deficits caused by disuse.

Transcriptome sequencing profiling identifies miRNA-331-3p as an osteoblast-specific miRNA in infected bone nonunion

Bone nonunion caused by bacterial infection accounts for bone fractures, bone trauma and bone transplantation surgeries. Severe consequences include delayed unions and amputation and result in functional limitations, work disability, and poor quality of life. However, the mechanism of bone nonunion remains unknown. In this study, we aimed to screen the miRNA biomarkers of bacterial bone infection and investigated whether miRNAs regulate the osteoblasts and thus contribute to bone nonunion. We established a miRNA-mRNA network based on high-throughput RNA sequencing to compare the model rabbits infected with Staphylococcus aureus with the control rabbits. After validation experiments, miRNA-331-3p and fibroblast growth factor 23 (FGF23) were found to be inversely correlated with the pathways of osteoblast mineralization and pathology of infected bone nonunion. In in vitro experiments, miRNA-331-3p was downregulated and FGF23 was upregulated in lipopolysaccharide (LPS)-induced mouse calvarial osteoblasts. Further studies of the mechanism showed that mutated of putative miRNA-331-3p can bind to FGF23 3'-untranslated region sites. MiRNA-331-3p acted as an osteoblast mineralization promoter by directly targeting FGF23. Downregulation of miRNA-331-3p led to inhibition of osteoblast mineralization by regulating the DKK1/β-catenin mediated signaling. Thus, we established an improved animal model and identified new bone-related biomarkers in the infected bone nonunion. The miRNA-331-3p biomarker was demonstrated to regulate osteoblast mineralization by targeting FGF23. The novel mechanism can be used as potential diagnostic biomarkers and therapeutic targets in the infected bone nonunion and other inflammatory bone disorders.

Transcriptomics: a Solution for Renal Osteodystrophy?

PURPOSE OF REVIEW

The molecular mechanisms of the bone disease associated with chronic kidney disease (CKD), called renal osteodystrophy (ROD), are poorly understood. New transcriptomics technologies may provide clinically relevant insights into the pathogenesis of ROD. This review summarizes current progress and limitations in the study and treatment of ROD, and in transcriptomics analyses of skeletal tissues.

RECENT FINDINGS

ROD is characterized by poor bone quality and strength leading to increased risk of fracture. Recent studies indicate permanent alterations in bone cell populations during ROD. Single-cell transcriptomics analyses, successful at identifying specialized cell subpopulations in bone, have not yet been performed in ROD. ROD is a widespread poorly understood bone disease with limited treatment options. Transcriptomics analyses of bone are needed to identify the bone cell subtypes and their role in the pathogenesis of ROD, and to develop adequate diagnosis and treatment strategies.

Bone morphogenetic protein-7 upregulates genes associated with osteoblast differentiation, including collagen I, Sp7 and IBSP in gingiva-derived stem cells

The present study was performed to evaluate the effects of short-term application of bone morphogenetic protein-7 (BMP-7) on human gingiva-derived mesenchymal stem cells with next-generation sequencing. Human gingiva-derived stem cells were treated with a final concentration of 100 ng/ml BMP-7 and the same concentration of a vehicle control. mRNA sequencing and data analysis were performed along using gene ontology and pathway analysis. RT-qPCR of mRNA of collagen I, Sp7, IBSP and western blot analysis of collagen I, osterix and bone sialoprotein was also performed. A total of 25,737 mRNAs were identified to be differentially expressed. Regarding osteoblast differentiation, 14 mRNAs were upregulated and 10 were downregulated when the results of the BMP-7 at 3 h were compared with the control at 3 h. The expression of collagen I was increased following the application of BMP-7 at 3 h, and this increase was also observed following western blot analysis. The effects of BMP-7 on stem cells were evaluated with mRNA sequencing, and the expression was validated with RT-qPCR and western blot analysis. The short-term application of BMP-7 produced an increased expression of collagen I, which was associated with target genes selected for osteoblast differentiation. This study may provide novel insights into the role of BMP-7 using mRNA sequencing.

RNA-seq in Skeletal Biology

PURPOSE OF REVIEW

The goal of this paper is to review state-of-the-art transcriptome profiling methods and their recent applications in the field of skeletal biology.

RECENT FINDINGS

Next-generation sequencing of mRNA (RNA-seq) methods have been established and routinely used in skeletal biology research. RNA-seq has led to the identification of novel genes and transcription factors involved in skeletal development and disease, through its application in small and large animal models, as well as human tissue and cells. With the availability of advanced techniques such as single-cell RNA-seq, novel cell types in skeletal tissues are being identified. As the sequencing technologies are rapidly evolving, the exciting discoveries supported by transcriptomics will continue to emerge and improve our understanding of the biology of the skeleton.

High Fidelity of Mouse Models Mimicking Human Genetic Skeletal Disorders

The 2019 International Skeletal Dysplasia Society nosology update lists 441 genes for which mutations result in rare human skeletal disorders. These genes code for enzymes (33%), scaffolding proteins (18%), signal transduction proteins (16%), transcription factors (14%), cilia proteins (8%), extracellular matrix proteins (5%), and membrane transporters (4%). Skeletal disorders include aggrecanopathies, channelopathies, ciliopathies, cohesinopathies, laminopathies, linkeropathies, lysosomal storage diseases, protein-folding and RNA splicing defects, and ribosomopathies. With the goal of evaluating the ability of mouse models to mimic these human genetic skeletal disorders, a PubMed literature search identified 260 genes for which mutant mice were examined for skeletal phenotypes. These mouse models included spontaneous and ENU-induced mutants, global and conditional gene knockouts, and transgenic mice with gene over-expression or specific base-pair substitutions. The human X-linked gene ARSE and small nuclear RNA U4ATAC, a component of the minor spliceosome, do not have mouse homologs. Mouse skeletal phenotypes mimicking human skeletal disorders were observed in 249 of the 260 genes (96%) for which comparisons are possible. A supplemental table in spreadsheet format provides PubMed weblinks to representative publications of mutant mouse skeletal phenotypes. Mutations in 11 mouse genes (Ccn6, Cyp2r1, Flna, Galns, Gna13, Lemd3, Manba, Mnx1, Nsd1, Plod1, Smarcal1) do not result in similar skeletal phenotypes observed with mutations of the homologous human genes. These discrepancies can result from failure of mouse models to mimic the exact human gene mutations. There are no obvious commonalities among these 11 genes. Body BMD and/or radiologic dysmorphology phenotypes were successfully identified for 28 genes by the International Mouse Phenotyping Consortium (IMPC). Forward genetics using ENU mouse mutagenesis successfully identified 37 nosology gene phenotypes. Since many human genetic disorders involve hypomorphic, gain-of-function, dominant-negative and intronic mutations, future studies will undoubtedly utilize CRISPR/Cas9 technology to examine transgenic mice having genes modified to exactly mimic variant human sequences. Mutant mice will increasingly be employed for drug development studies designed to treat human genetic skeletal disorders.

SIGNIFICANCE

Great progress is being made identifying mutant genes responsible for human rare genetic skeletal disorders and mouse models for genes affecting bone mass, architecture, mineralization and strength. This review organizes data for 441 human genetic bone disorders with regard to heredity, gene function, molecular pathways, and fidelity of relevant mouse models to mimic the human skeletal disorders. PubMed weblinks to citations of 249 successful mouse models are provided.