A developmental stage-specific network approach for studying dynamic co-regulation of transcription factors and microRNAs during craniofacial development

DeepFace: Deep-learning-based framework to contextualize orofacial-cleft-related variants during human embryonic craniofacial development

Orofacial clefts (OFCs) are among the most common human congenital birth defects. Previous multiethnic studies have identified dozens of associated loci for both cleft lip with or without cleft palate (CL/P) and cleft palate alone (CP). Although several nearby genes have been highlighted, the "casual" variants are largely unknown. Here, we developed DeepFace, a convolutional neural network model, to assess the functional impact of variants by SNP activity difference (SAD) scores. The DeepFace model is trained with 204 epigenomic assays from crucial human embryonic craniofacial developmental stages of post-conception week (pcw) 4 to pcw 10. The Pearson correlation coefficient between the predicted and actual values for 12 epigenetic features achieved a median range of 0.50-0.83. Specifically, our model revealed that SNPs significantly associated with OFCs tended to exhibit higher SAD scores across various variant categories compared to less related groups, indicating a context-specific impact of OFC-related SNPs. Notably, we identified six SNPs with a significant linear relationship to SAD scores throughout developmental progression, suggesting that these SNPs could play a temporal regulatory role. Furthermore, our cell-type specificity analysis pinpointed the trophoblast cell as having the highest enrichment of risk signals associated with OFCs. Overall, DeepFace can harness distal regulatory signals from extensive epigenomic assays, offering new perspectives for prioritizing OFC variants using contextualized functional genomic features. We expect DeepFace to be instrumental in accessing and predicting the regulatory roles of variants associated with OFCs, and the model can be extended to study other complex diseases or traits.

Protective effect of Sasa veitchii extract against all-trans-retinoic acid-induced inhibition of proliferation of cultured human palate cells

Cleft palate is the most common facial birth defect worldwide. It is caused by environmental factors or genetic mutations. Environmental factors such as pharmaceutical exposure in women are known to induce cleft palate. The aim of the present study was to investigate the protective effect of Sasa veitchii extract against medicine-induced inhibition of proliferation of human embryonic palatal mesenchymal cells. We demonstrated that all-trans-retinoic acid inhibited human embryonic palatal mesenchymal cell proliferation in a dose-dependent manner, whereas dexamethasone treatment had no effect on cell proliferation. Cotreatment with Sasa veitchii extract repressed all-trans-retinoic acid-induced toxicity in human embryonic palatal mesenchymal cells. We found that cotreatment with Sasa veitchii extract protected all-trans-retinoic acid-induced cyclin D1 downregulation in human embryonic palatal mesenchymal cells. Furthermore, Sasa veitchii extract suppressed all-trans-retinoic acid-induced miR-4680-3p expression. Additionally, the expression levels of the genes that function downstream of the target genes ( ERBB2 and JADE1 ) of miR-4680-3p in signaling pathways were enhanced by cotreatment with Sasa veitchii extract and all-trans-retinoic acid compared to all-trans-retinoic acid treatment. These results suggest that Sasa veitchii extract suppresses all-trans-retinoic acid-induced inhibition of cell proliferation via modulation of miR-4680-3p expression.

Single-cell multiomics decodes regulatory programs for mouse secondary palate development

Perturbations in gene regulation during palatogenesis can lead to cleft palate, which is among the most common congenital birth defects. Here, we perform single-cell multiome sequencing and profile chromatin accessibility and gene expression simultaneously within the same cells (n = 36,154) isolated from mouse secondary palate across embryonic days (E) 12.5, E13.5, E14.0, and E14.5. We construct five trajectories representing continuous differentiation of cranial neural crest-derived multipotent cells into distinct lineages. By linking open chromatin signals to gene expression changes, we characterize the underlying lineage-determining transcription factors. In silico perturbation analysis identifies transcription factors SHOX2 and MEOX2 as important regulators of the development of the anterior and posterior palate, respectively. In conclusion, our study charts epigenetic and transcriptional dynamics in palatogenesis, serving as a valuable resource for further cleft palate research.

MicroRNAs in Small Extracellular Vesicles from Amniotic Fluid and Maternal Plasma Associated with Fetal Palate Development in Mice

Cleft palate (CP) is a common congenital birth defect. Cellular and morphological processes change dynamically during palatogenesis, and any disturbance in this process could result in CP. However, the molecular mechanisms steering this fundamental phase remain unclear. One study suggesting a role for miRNAs in palate development via maternal small extracellular vesicles (SEVs) drew our attention to their potential involvement in palatogenesis. In this study, we used an in vitro model to determine how SEVs derived from amniotic fluid (ASVs) and maternal plasma (MSVs) influence the biological behaviors of mouse embryonic palatal mesenchyme (MEPM) cells and medial edge epithelial (MEE) cells; we also compared time-dependent differential expression (DE) miRNAs in ASVs and MSVs with the DE mRNAs in palate tissue from E13.5 to E15.5 to study the dynamic co-regulation of miRNAs and mRNAs during palatogenesis in vivo. Our results demonstrate that some pivotal biological activities, such as MEPM proliferation, migration, osteogenesis, and MEE apoptosis, might be directed, in part, by stage-specific MSVs and ASVs. We further identified interconnected networks and key miRNAs such as miR-744-5p, miR-323-5p, and miR-3102-5p, offering a roadmap for mechanistic investigations and the identification of early CP biomarkers.

Clinical and Genetic Characteristics of Calvarial Doughnut Lesions with Bone Fragility in Three Families with a Reccurent SGMS2 Gene Variant

Calvarial doughnut lesions (CDL) with bone fragility with or without spondylometaphyseal dysplasia (MIM: #126550) is a rare autosomal dominant skeletal disorder characterized by low bone mineral density, spinal and peripheral fractures, and specific sclerotic lesions of the cranial bones. In the current classification of skeletal disorders, the disease is included in the group of bone fragility disorders along with osteogenesis imperfecta. The disease is caused by pathogenic variants in the SGMS2 gene, the protein product of which is sphingomyelin synthase 2, which primarily contributes to sphingomyelin (SM) synthesis-the main lipid component of the plasma membrane essential for bone mineralization. To date, 15 patients from eight families with CDL with bone fragility have been described in the literature, and a recurrent variant c.148C>T (p.Arg50Ter) in the SGMS2 gene has been identified, which was found in patients from six families. We diagnosed the disease in 11 more patients from three unrelated families, caused by the same heterozygous nonsense variant c.148C>T (p.Arg50Ter) in the SGMS2 gene. Our results show wide interfamilial and intrafamilial phenotypic variability in patients with a detected recurrent variant in the SGMS2 gene, the presence of which must be taken into consideration in the diagnosis of the disease. The primary analysis of this variant will contribute to optimal molecular genetic diagnostics, which can reduce diagnostic costs and time.

Early-Onset Osteoporosis: Rare Monogenic Forms Elucidate the Complexity of Disease Pathogenesis Beyond Type I Collagen

Early-onset osteoporosis (EOOP), characterized by low bone mineral density (BMD) and fractures, affects children, premenopausal women and men aged <50 years. EOOP may be secondary to a chronic illness, long-term medication, nutritional deficiencies, etc. If no such cause is identified, EOOP is regarded primary and may then be related to rare variants in genes playing a pivotal role in bone homeostasis. If the cause remains unknown, EOOP is considered idiopathic. The scope of this review is to guide through clinical and genetic diagnostics of EOOP, summarize the present knowledge on rare monogenic forms of EOOP, and describe how analysis of bone biopsy samples can lead to a better understanding of the disease pathogenesis. The diagnostic pathway of EOOP is often complicated and extensive assessments may be needed to reliably exclude secondary causes. Due to the genetic heterogeneity and overlapping features in the various genetic forms of EOOP and other bone fragility disorders, the genetic diagnosis usually requires the use of next-generation sequencing to investigate several genes simultaneously. Recent discoveries have elucidated the complexity of disease pathogenesis both regarding genetic architecture and bone tissue-level pathology. Two rare monogenic forms of EOOP are due to defects in genes partaking in the canonical WNT pathway: LRP5 and WNT1. Variants in the genes encoding plastin-3 (PLS3) and sphingomyelin synthase 2 (SGMS2) have also been found in children and young adults with skeletal fragility. The molecular mechanisms leading from gene defects to clinical manifestations are often not fully understood. Detailed analysis of patient-derived transiliac bone biopsies gives valuable information to understand disease pathogenesis, distinguishes EOOP from other bone fragility disorders, and guides in patient management, but is not widely available in clinical settings. Despite the great advances in this field, EOOP remains an insufficiently explored entity and further research is needed to optimize diagnostic and therapeutic approaches. © 2022 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Suppression of microRNA 124-3p and microRNA 340-5p ameliorates retinoic acid-induced cleft palate in mice

The etiology of cleft lip with or without cleft palate (CL/P), a common congenital birth defect, is complex, with genetic and epigenetic, as well as environmental, contributing factors. Recent studies suggest that fetal development is affected by maternal conditions through microRNAs (miRNAs), a group of short noncoding RNAs. Here, we show that miR-129-5p and miR-340-5p suppress cell proliferation in both primary mouse embryonic palatal mesenchymal cells and O9-1 cells, a neural crest cell line, through the regulation of Sox5 and Trp53 by miR-129-5p, and the regulation of Chd7, Fign and Tgfbr1 by miR-340-5p. Notably, miR-340-5p, but not miR-129-5p, was upregulated following all-trans retinoic acid (atRA; tretinoin) administration, and a miR-340-5p inhibitor rescued the cleft palate (CP) phenotype in 47% of atRA-induced CP mice. We have previously reported that a miR-124-3p inhibitor can also partially rescue the CP phenotype in atRA-induced CP mouse model. In this study, we found that a cocktail of miR-124-3p and miR-340-5p inhibitors rescued atRA-induced CP with almost complete penetrance. Taken together, our results suggest that normalization of pathological miRNA expression can be a preventive intervention for CP.

Dynamic Expression and Gene Regulation of MicroRNAs During Bighead Carp (Hypophthalmichthys nobilis) Early Development

The early development of fish is regulated through dynamic and complex mechanisms involving the regulation of various genes. Many genes are subjected to post-transcriptional regulation by microRNAs (miRNAs). In the Chinese aquaculture industry, the native species bighead carp (Hypophthalmichthys nobilis) is important. However, the genetic regulation related to the early development of bighead carp is unknown. Here, we generated developmental profiles by miRNA sequencing to study the dynamic regulation of miRNAs during bighead carp early development. This study identified 1 046 miRNAs, comprising 312 known miRNAs and 734 uncharacterized miRNAs. Changes in miRNA expression were identified in the six early development stages. An obviously increased expression trend was detected during the development process, with the main burst of activity occurring after the earliest stage (early blastula, DS1). Investigations revealed that several miRNAs were dominantly expressed during the development process, especially in the later stages (e.g., miR-10b-5p, miR-21, miR-92a-3p, miR-206-3p, and miR-430a-3p), suggesting that these miRNAs exerted important functions during embryonic development. The differentially expressed miRNAs (DEMs) and time-serial analysis (profiles) of DEMs were analyzed. A total of 372 miRNAs were identified as DEMs (fold-change &gt;2, and false discovery rate &lt;0.05), and three expression profiles of the DEMs were detected to have co-expression patterns (r &gt; 0.7, and p &lt; 0.05). The broad negative regulation of target genes by miRNAs was speculated, and many development-related biological processes and pathways were enriched for the targets of the DEMs, which might be associated with maternal genome degradation and embryogenesis processes. In conclusion, we revealed the repertoire of miRNAs that are active during early development of bighead carp. These findings will increase our understanding of the regulatory mechanisms of early development of fish.

Mutations in SIX1 Associated with Branchio-oto-Renal Syndrome (BOR) Differentially Affect Otic Expression of Putative Target Genes

Several single-nucleotide mutations in SIX1 underlie branchio-otic/branchio-oto-renal (BOR) syndrome, but the clinical literature has not been able to correlate different variants with specific phenotypes. We previously assessed whether variants in either the cofactor binding domain (V17E, R110W) or the DNA binding domain (W122R, Y129C) might differentially affect early embryonic gene expression, and found that each variant had a different combination of effects on neural crest and placode gene expression. Since the otic vesicle gives rise to the inner ear, which is consistently affected in BOR, herein we focused on whether the variants differentially affected the otic expression of genes previously found to be likely Six1 targets. We found that V17E, which does not bind Eya cofactors, was as effective as wild-type Six1 in reducing most otic target genes, whereas R110W, W122R and Y129C, which bind Eya, were significantly less effective. Notably, V17E reduced the otic expression of prdm1, whereas R110W, W122R and Y129C expanded it. Since each mutant has defective transcriptional activity but differs in their ability to interact with Eya cofactors, we propose that altered cofactor interactions at the mutated sites differentially interfere with their ability to drive otic gene expression, and these differences may contribute to patient phenotype variability.

Analysis of Facial Skeletal Morphology: Nasal Bone, Maxilla, and Mandible

The growth and development of facial bones are closely related to each other. The present study investigated the differences in the nasomaxillary and mandibular morphology among different skeletal patterns. Cephalograms of 240 participants were divided into 3 groups based on the skeletal pattern (Class I, Class II, and Class III). The dimensions of nasomaxilla (nasal bone length, nasal ridge length, nasal depth, palatal length, and maxillary height) and mandible (condylar length, ramus length, body length, symphysis length, and entire mandibular length) were measured. One-way analysis of variance and Pearson's correlation test were used for statistical analysis. No significant differences were observed among the skeletal patterns in terms of nasal bone length, palatal length, maxillary height, or condylar length. Class II had a significantly shorter ramus, mandibular body, and entire mandibular length compared with those of Class I and Class III. Nasal ridge length exhibited a significant moderate correlated with nasal bone length (correlation coefficient: 0.433) and maxillary height (correlation coefficient: 0.535). The entire mandibular length exhibited a significant moderate correlated with ramus length (correlation coefficient: 0.485) and body length (correlation coefficient: 0.536). In conclusion, nasal and maxillary dimensions exhibited no significant difference among the 3 skeletal patterns. Mandibular body and entire mandibular lengths were significantly positively correlations with Class III skeletal patterns.