The developing mouse coronal suture at single-cell resolution

The developing mouse coronal suture at single-cell resolution | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article The developing mouse coronal suture at single-cell resolution D'Juan Farmer, Hana Mlcochova, Yan Zhou, Nils Koelling, Guanlin Wang, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-135455/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 10 Aug, 2021 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Abstract Sutures separate the flat bones of the skull and enable coordinated growth of the brain and overlying cranium. To uncover the cellular diversity within sutures, we generated single-cell transcriptomes and performed extensive expression validation of the embryonic murine coronal suture. We identify Erg and Pthlh as markers of osteogenic progenitors in sutures, and distinct pre-osteoblast signatures between the bone fronts and periosteum. In the ectocranial layers above the suture, we observe a ligament-like population spanning the frontal and parietal bones. In the dura mater underlying the suture, we detect a chondrocyte-like signature potentially linked to cartilage formation under pathological conditions. Genes mutated in coronal synostosis are preferentially expressed in proliferative osteogenic cells, as well as meningeal layers, suggesting discrete cell types that may be altered in different syndromes. This single-cell atlas provides a resource for understanding development of the coronal suture, the suture most commonly fused in monogenic craniosynostosis. Developmental Biology Medical Genetics bone growth coronal suture Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Cranial sutures are fibrous joints between calvarial bones that act as zones of bone growth and absorbers of physical forces 1 . They comprise the leading edges of abutting calvarial bones separated by mesenchymal tissue. New bone forms by intramembranous ossification in response to expansion of the underlying brain 1 , 2 , 3 . Growth of the bony skull requires the proliferation and osteogenic differentiation of progenitor cells, as well as the maintenance of sufficient undifferentiated cells in the suture to ensure continued bone growth during fetal and postnatal stages. Environmental and/or genetic insults that disrupt the delicate balance of proliferation and differentiation result in premature fusion of cranial sutures, a condition known as craniosynostosis 4 , 5 . The coronal suture, which separates the frontal and parietal bones, is the suture most commonly affected in monogenic craniosynostosis 6 . The mouse has proven an effective model for the study of coronal synostosis 7 , 8 , 9 , 10 . During cranial development, the coronal suture and closely associated tissues are derived from three distinct populations: the supraorbital mesenchyme which will give rise to the calvarial bones and suture mesenchyme, the meningeal mesenchyme, and non-osteogenic early migrating mesenchyme 11 . The meninges form between the brain and calvaria and are essential for development of both 12 . Coronal suture mesenchyme, derived largely from the mesoderm, forms a boundary between the neural crest-derived frontal and mesoderm-derived parietal bones 13 , 14 . Embryonic suture mesenchyme originates from Gli1 -expressing cells that migrate away from the paraxial cephalic mesoderm at embryonic day (E) 7.5 15 and expand apically to sit between the lateral dermal mesenchyme and medial meningeal layers from E12.5 onwards 11 . Whereas these macroscopic developmental steps are well established, the cellular composition of the developing coronal suture remains poorly understood. Markers that label skeletal stem cells within postnatal sutures have been identified 16 , 17 , 18 , 19 , yet none of these markers identify a distinct skeletal progenitor cell population at embryonic stages. Given recent reports that embryonic progenitor dysfunction precedes craniosynostosis 20 , identifying cell type diversity in early forming sutures will be critical for understanding the etiology of this birth defect. A better understanding of embryonic osteogenic and non-osteogenic populations will also inform how the meninges and ectocranial layers contribute to suture patency 7 , 21 , 22 . To build a cell atlas of the embryonic coronal suture, we combined single-cell transcriptomics with highly resolved in situ analysis to catalogue the cell types present in murine coronal sutures at E15.5 and E17.5. In the ectocranial compartment, we uncovered multiple layers of distinct cell types, including a ligament-like population connecting the lateral aspects of the frontal and parietal bones. Within the multiple layers of the meninges, we revealed an outer dura mater population with a chondrogenic signature, suggesting a latent capacity for chondrocyte differentiation. In the osteogenic population, pseudotime analysis revealed a putative Erg + / Pthlh + progenitor that we found to be located in the suture and along the leading edges of the bones. These progenitors fed into two distinct pre-osteoblast trajectories, one concentrated at the growing bone tips and the other localized along the periosteum more distant from the suture. Comparative expression analysis between genes associated with coronal or midline synostosis highlighted the selective expression of many coronal synostosis genes, including Twist1 and Tcf12 , within proliferative osteogenic cells, but also within ectocranial and meningeal layers, suggesting heterogenous etiologies for coronal synostosis. We also detected potential ligand-receptor interactions of neighboring ectocranial and meningeal layers with osteogenic cells within the suture, in particular the proliferative osteogenic population. This is in agreement with previous studies showing roles for the ectocranial mesenchyme and meninges in regulating suture patency 11 , 12 . This single-cell atlas reveals diversity within the developing coronal suture, some of the earliest potential markers for suture-resident osteogenic progenitors, and potential interactions between osteogenic and non-osteogenic populations likely important for proper skull expansion. Results Diverse mesenchymal heterogeneity captured by single-cell sequencing. To understand the cellular composition of the embryonic coronal suture, we performed single-cell RNA sequencing at E15.5 and E17.5 on dissected coronal sutures, including small amounts of frontal and parietal bone, after removing the skin and brain (Fig. 1 a). We filtered using Seurat 3 R-Package 23 and obtained 8279 cells at E15.5 (median of 2460 genes per cell) and 8682 cells at E17.5 (median of 3200 genes per cell) (Fig. 1 b). We identified 14 cell clusters through unsupervised graph clustering of the two datasets combined (Supplementary Table 1). Osteogenic and mesenchymal cell types were identified based on the expression of broad mesenchyme/fibroblast ( Col1a1 ) and osteoblast ( Sp7 ) markers (Fig. 1 b, dotted line; Supplementary Fig. 1a-b). The identities of clusters outside the osteogenic/mesenchymal subset were resolved using previously reported markers, and included chondrocytes, myeloid cells, mast cells, lymphocytes, pericytes, osteoclasts, endothelial cells, neurons, and glia (Fig. 1 b, c). All the major cell types in our analysis were present at E15.5 and E17.5 (Supplementary Fig. 2a). Chondrocytes were especially abundant at E15.5 (Supplementary Fig. 2b), highlighting the close proximity of the E15.5 coronal suture to the chondrocranium. Myeloid cells were more abundant at E17.5, consistent with reports of increased myeloid differentiation during late embryonic stages 24 (Supplementary Fig. 2b). To analyse the osteogenic and mesenchymal cell types that might comprise and support the coronal suture, we re-clustered the osteogenic/mesenchymal population and obtained 14 clusters present at both E15.5 and E17.5 (Fig. 1 d-f). Analysis of enriched genes for each cluster allowed us to assign probable identities to each cell type (Fig. 1 g, Supplementary Table 2), which we validated by in situ hybridization as described below. We observed one ectocranial cluster strongly over-represented at E15.5 and a meningeal cluster over-represented at E17.5. Osteogenic cells were also more abundant in the E17.5 dataset, although it is unclear whether this reflects true biological differences versus differing cell capture between the dissections (Fig. 1 a, f). Diversity of meningeal layers below the coronal suture. The meninges are involved in the development of the calvaria and underlying brain that they separate 12 . They comprise dura mater, arachnoid mater, and pia mater. To determine the identity of meningeal tissues included in our dissections, we utilized a recent transcriptomic study of murine E14 meninges as a guide 25 . Markers associated with the pia mater ( Ngfr , Lama1 , Rdh10 ) were not co-enriched in any of our clusters, consistent with the pia mater being removed with the brain during dissections (Fig. 2 a). In contrast, markers enriched within the arachnoid mater ( Aldh1a2 , Cldn11 , and Tbx18 ) were abundant in MG4, and dura mater markers ( Gja1 , Fxyd5 , and Crabp2 ) in MG3 and MG4, and to a lesser extent MG1 and MG2 (Fig. 2 a). To resolve the identity of MG4, we performed in situ experiments for a highly specific MG4 marker, Gjb6 , in combination with the arachnoid/dura mater marker, Crabp2 (Fig. 2 b). Crabp2 and Gjb6 overlapped below the bone with Crabp2 single-positive cells (MG3) found above the Crabp2 + / Gjb6 + domain (Fig. 2 c). Immunofluorescence for Crabp2 and Gja1 at E17.5 confirmed that these arachnoid/dura mater markers are excluded from the pia mater (Supplementary Fig. 3a). Rgs5 + pericytes were interspersed with Gjb6 + cells in the MG4 arachnoid layer, consistent with the prominent vasculature extending from the border of the dura mater and through the arachnoid mater to the pia mater 26 (Supplementary Fig. 3c). In the Crapb2 + / Gjb6 − layer (MG3), we also observed co-expression of Crabp2 with Nppc , with a zone of Nppc + / Crabp2 − cells (MG2) above (Fig. 2 d). In addition, Nppc was co-expressed with the known dura mater marker Fxyd5 25 (Supplementary Fig. 3b, d). MG3 may represent the dural border cell layer that has been described using EM in adult meninges 27 , 28 . The Nppc + / Crabp1 − population above Nppc + / Crabp2 + cells is consistent with MG2 identity in our UMAP analysis (Fig. 2 b, d). This domain was also positive for Matn4 and Ctgf , markers that only overlap with Nppc in MG2 (Fig. 2 f, Supplementary Fig. 3b, e). Matn4 expression does not overlap with the MG3/MG4 marker Crabp2 (Fig. 2 e), and chondrocytes in other sections express high levels of Matn4 but not Ctgf (Supplementary Fig. 3f). We also detected some Matn4 cells that were low or absent for Nppc , and these were situated closer to the suture and calvarial bones than the Nppc -high cells (Fig. 2 f). Our UMAP analysis indicates that these Matn4 + / Nppc − cells represent MG1 (Fig. 2 b). MG1 expressed higher levels of the chondrocyte markers Col2a1 and Acan than MG2, but not at the level seen in bona fide chondrocytes (Supplementary Fig. 3g). These findings are consistent with MG1/MG2 representing a periosteal dura mater population 29 primed to form cartilage, with the MG1 population having a stronger chondrogenic-like program. These data reveal a diversity of mesenchyme cell types within the meninges, with those closest to the suture sharing features with chondrogenic cells and likely functioning as a specialized connective tissue bridging the calvarial bones to the meninges (Fig. 2 g). Diversity of ectocranial layers above the coronal suture. Ectocranial mesenchyme guides the migration of early osteogenic cells and helps establish suture boundaries 7 , 21 , 22 . Consistent with analysis of the mouse frontal suture 30 , we identified clusters enriched for markers of the dermis ( Ly6a , Dpt ; EC1-3) (Fig. 1 g, Fig. 3 a). Ly6a is expressed in hypodermis of neonatal mouse skin 31 and we observe co-expression of Ly6a with additional EC1-3 markers, Pi16 and Dpt , within a single layer of mesenchyme above the skull bones (Fig. 3 b, c; Supplementary Fig. 4a, b). Our previous work had identified expression of Jag1 within both an ectocranial layer and suture mesenchyme cells 22 , and scRNAseq analysis shows high expression of Jag1 in EC1, and to a lesser extent EC2 and EC3 (Fig. 1 f-g, Supplementary Fig. 4g). Co-expression of Jag1 with Pi16 confirms Jag1 expression within the hypodermal layer (Supplementary Fig. 4h). In addition to hypodermal EC1-3 layers, we also noted a prominent EC4 population defined by C1qtnf3 and Tnmd expression in our UMAP analysis (Fig. 3 a). In situ experiments revealed two C1qtnf3 + ectocranial layers on either side of the Pi16 + hypodermis (Fig. 3 d, Supplementary Fig. 4c). However, only the C1qtnf3 + layer between the calvarial bones and hypodermis expresses Tnmd , suggesting that this corresponds to EC4, which we term “suprasutural mesenchyme” (Fig. 3 a, e; Supplementary Fig. 4d). It is likely that the outer C1qtnf3 + layer was removed along with the skin during dissection and thus not included in our single-cell analysis. Consistently, Epha4 , which has also been shown to display ectocranial expression 7 , was most strongly expressed in this outer C1qtnf3 + layer despite not being particularly enriched in clusters EC1-4 in UMAP analysis (Supplementary Fig. 4i). Thus, Jag1 and Epha4 appear to label distinct ectocranial layers. Interestingly, just underneath the C1qtnf3 + / Tnmd + layer (EC4), and sitting on top of the suture, we observed a C1qtnf3 − / Tnmd + population (LIG) that was enriched for a number of genes associated with tendon/ligament development (e.g. Scx , Tnmd , Mkx, Thbs2, Bgn ), as well as markers of smooth muscle (e.g. Acta2 , Tagln , Myl9 ) (Supplementary Fig. 4j). Two of the most specific markers for LIG were Tac1 , a gene that encodes four neuropeptides including Substance P that is implicated in tendon mechanosensation 32 , and Chodl , a membrane protein with expression in tendons of humans 33 and mouse limbs 34 (Fig. 3 a). In situ hybridization for Tac1 and Chodl revealed highly restricted expression in mesenchyme overlying the coronal suture and connecting the edges of the frontal and parietal bones (Fig. 3 f, h, Supplementary Fig. 4e). Although broader, in situ experiments showed that Scx and Tnmd were also expressed within this layer (Fig. 3 g; Supplementary Fig. 4f, k), and we observed only minimal overlap between C1qtnf3 and Tac1 expression (Supplementary Fig. 4l). Thus, the LIG cluster represents a transcriptionally distinct and spatially confined subset of the ectocranium overlying the coronal suture and connecting the adjacent bones. Distinct osteoblast trajectories in the suture versus periosteum. The suture is a source of new osteoblasts for skull bone expansion. To better understand the trajectories of osteogenesis within and around the sutures, we identified six osteogenic clusters (OG1, OG2, OG3, OG4, PO1, PO2) based on expression of the osteoblast transcription factors Runx2 and Sp7 35,36,37 , as well as Dlx5 and Dlx6 38,39,40 , (Supplementary Fig. 5). Pseudotime analysis of these clusters using Monocle3 revealed OG1 to be the earliest lineage cells, followed by two proliferative branches (PO1, PO2) and then distinct OG2 and OG3 trajectories that converged on the osteoblast cluster OG4 (Fig. 4 a, b). These pseudotime trajectories were reflected by increased expression of Runx2 and Sp7 from OG1 to OG4 and accumulation of the mature osteoblast markers Ifitm5 and Dmp1 in OG4 (Fig. 4 c). Prominent cluster markers included Erg and Pthlh (OG1), Lef1 and Inhba (OG2), Mmp13 and Podnl1 (OG3), and Ifitm5 , Dmp1 , and Sost (OG4) (Fig. 4 e). Pseudotime visualization shows Erg (OG1) appearing earliest and shutting down as Lef1 (OG2) and Mmp13 (OG3) become expressed, followed by their extinguishment and progressive appearance of Ifitm5, Dmp1 , and Sost (OG4) (Fig. 4 d). In situ validation at E15.5 showed co-expression of OG1 markers Erg and Pthlh in suture mesenchyme and extending along the edges of the frontal and parietal bone tips (Fig. 4 f, Supplementary Fig. 6a). Interestingly, Erg but not Pthlh was asymmetrically distributed along the bones, with stronger expression above the frontal and below the parietal bone. Gli1 and Prrx1 showed similar asymmetric expression above the frontal bone, and Gli1 and Six2 below the parietal bone (Supplementary Fig. 7a-c). The different asymmetric mesenchyme patterns were not attributable to disproportionate populations of mesenchyme above and below the bones, as ectocranial and meningeal layers were generally evenly distributed along the bones (Figs. 2 , 3 ). These findings highlight asymmetric distribution of the earliest osteogenic cells around the bone fronts, which may play a role in ensuring the later reproducible overlap of the parietal over the frontal bone. We next examined expression of markers for OG2 and OG3, as pseudotime analysis suggested that these represent alternative pathways to osteoblasts (Fig. 4 a). Markers for both OG2 ( Lef1 , Inhba ) and OG3 ( Podnl1 , Mmp13 ) displayed expression along the outer domains of the Sp7 + bone surface, consistent with a pre-osteoblast identity. Whereas Lef1 + / Inhba + cells were most abundant at the edges of the growing bones near the suture, Podnl1 + / Mmp13 + cells were enriched on bone surfaces further away from the suture (Fig. 4 g, h; Supplementary Fig. 6b, c). OG2 may therefore represent more specialized suture-resident pre-osteoblasts, and OG3 periosteal pre-osteoblasts more generally found on bone surfaces. Clusters PO1 and PO2 are enriched for markers of proliferation ( Mki67 , Cenpf , Top2a; Supplementary Fig. 6e), as well as expression of genes from suture-resident clusters, for example Erg (OG1) and Lef1 and Inhba (OG2) (Fig. 1 a, Fig. 4 e). In situ validation revealed Cenpf expression around the tips of the growing bones and extending into the sutures, though in comparison to Erg it appeared largely absent from the central part of the suture (Supplementary Fig. 6f). These data suggest that proliferative pre-osteoblasts are concentrated at the bone tips, consistent with previous direct evidence for proliferative osteogenic cells at the leading edges of the calvarial bones 20 . OG4 represents mature osteoblasts and osteocytes, as revealed by markers such as Ifitm5 , Dmp1 and Sost and its terminal position in the pseudotime analysis 41 (Fig. 4 a, c). We observe co-expression of Sp7 with Ifitm5 throughout most of the bone, except for the leading tips and the surfaces of bones which are Sp7 -positive only, consistent with bone growth from both the leading edges and the periosteal surfaces (Fig. 4 i, Supplementary Fig. 6d). Similarly, we observe protein expression of the osteoblast marker Cd200 in bone and the Wnt-responsive transcription factor Tcf7, a pre-osteoblast marker, along the bone tips and periosteal surfaces 43 , 44 , 45 , consistent with Cd200 expression in cluster OG4 and Tcf7 expression in both OG4 and the suture-specific pre-osteoblast cluster OG2 (Fig. 1 g, Supplementary Fig. 8a-c). The mature osteocyte marker Sost , which labels the oldest cells in pseudotime, was expressed in only a few osteocytes distant from the suture (Fig. 4 i). The spatial organization between osteoblasts, pre-osteoblasts, and progenitors was confirmed with double in situs between Podnl1 and Pthlh or Ifitm5 (Supplementary Fig. 6g-h). These findings are consistent with bone elongation occurring through osteogenesis at the suture and bone tips, with a distinct osteogenic pathway in the periosteum contributing to bone thickening (Fig. 4 j). Signaling interactions between mesenchymal layers captured from single-cell analysis. To gain insights into potential signaling between osteogenic cells and adjacent ectocranial and meningeal mesenchyme, we surveyed our datasets for expression of ligands and receptors (Fig. 5 a-b). In OG4 osteoblasts, we observed enrichment of Tgfb1 , Bmp3 , Bmp4 , Ihh , and Pdgfa , with Ihh 46 and Pdgfa 47 , 48 known to be secreted by osteoblasts. In addition to osteoprogenitors (OG1), Pthlh ligand was expressed in several meningeal clusters. In the ectocranial clusters EC4 and LIG, which are situated closest to the suture, we detected enrichment of Tgfb3 , Fgf9 , Fgf18 , Wnt5a , Wnt9a , Wnt11 , and Igf1 ligands. In the meningeal clusters closest to the suture (MG1, MG2), we also detected Tgfb3 and Igf1 ligand expression, as well as Tgfb2 that has previously been shown to be a critical meningeal-derived factor for suture morphogenesis 49 . Reciprocally, we detected expression of the Tgfβ receptor Tgfbr3 , the Fgf receptors Fgfr1 and Fgfr2 , the Wnt receptors Fzd1 , Fzd2 , and Fzd6 , the Ihh receptor Ptch1 , the Pthlh receptor Pth1r , and the Pdgfa receptor Pdgfra in various osteogenic clusters (Fig. 5 b). Interestingly, Fgfr1 was expressed more strongly in suture-resident pre-osteoblasts (OG2) and Fgfr2 in periosteal pre-osteoblasts (OG3). To capture signalling interactions in an unbiased approach, we interrogated our data using the CellPhoneDB package 50 , scoring the ligand-receptor pairings between all clusters (Fig. 6 c). Interactions between osteogenic clusters were notably weak, highlighting potential roles for osteogenic - non-osteogenic interactions in coronal suture regulation. Consistently, we noted that the mesenchymal populations closest to the suture (EC4 above, MG1/MG2 below) had the strongest interactions with osteogenic cells. The proliferative PO1 and pre-osteoblast OG3 clusters appeared to have the strongest connections with the surrounding mesenchyme, consistent with these populations residing at the surfaces of bones. Of the individual ligand-receptor interactions that were identified between EC4/MG1/M2 and osteogenic clusters, several pathways are relevant to coronal suture development, including Fgf, Tgfβ, and Wnt signalling 51 (Supplementary Fig. 9). Synostosis related genes display diverse expression patterns. To determine if genes underlying craniosynostosis in humans 51 , 52 coalesce within a specific cell type, we mapped coronal and midline synostosis genes onto our dataset. Genes associated with midline suture synostosis were less abundant, suggesting that divergent gene expression patterns may contribute to suture-specific fusions (Fig. 6 a). We noted that 8/17 coronal synostosis genes were selectively enriched in PO1 or PO2, suggesting misregulation of proliferating osteoblasts as a common mechanism of synostosis. By scoring every cell for the average expression of each coronal or midline synostosis gene, we noted that coronal synostosis scores were more restricted to the meningeal and osteogenic clusters, while midline synostosis genes were broadly distributed across clusters (Fig. 6 b). Coronal synostosis genes with particular enrichment in PO1 and PO2 include Fgfr1 , Fgfr2 , Tcf12 , Twist1 , and Zeb2 . Consistently, in situ validation revealed expression of Twist1 in suture mesenchyme and cells lining the bone fronts, with Twist1 expression also overlapping with the proliferative pre-osteoblast marker Cenpf (Fig. 6 c, Supplementary Fig. 10). Selective enrichment of Twist1 and Tcf12 in proliferative pre-osteoblasts is consistent with functional studies showing roles for these transcription factors in regulating the balance between bone differentiation and proliferation 9 , 20 . Discussion Proper formation of the coronal suture is a complex process requiring the coordinated development of multiple tissue types, which is reflected in the diverse genetic heterogeneity underlying coronal synostosis. Understanding the mechanisms that cause pathogenic suture fusion in distinct syndromic forms of synostosis has been limited by our incomplete knowledge of the cellular diversity of the developing coronal suture. In an effort to bridge this gap, we have generated a comprehensive spatial and transcriptomic map of the cell types that comprise and support the embryonic coronal suture (Fig. 6 d). The relationship of the embryonic progenitors that build the coronal suture with the adult sutural stem cells that grow the calvarium remains debated. Recent work on postnatal sutures has demonstrated that the suture mesenchyme houses resident skeletal stem cells that produce osteoblasts to grow skull bones 16 , 17 , 18 , 54 . However, these markers ( Gli1 , Prrx1 , Axin2 ) broadly label mesenchyme throughout the embryonic skull, making them unsuitable for identifying osteogenic cells at earlier stages. Here we identify Erg and Pthlh as two of the earliest markers for the putative osteoblast progenitors concentrated in the embryonic coronal suture. As Pthlh is a marker for chondrocyte stem cells in the femoral growth plate 55 , Pthlh signaling may be a more general regulator of skeletal progenitors. As opposed to the adult skull where stem cells are tightly restricted to suture mesenchyme, embryonic Erg + /Pthlh + progenitors extend away from the suture along the surfaces of the frontal and parietal bone tips. Interestingly, these progenitors are asymmetrically distributed, with more cells along the lateral surface of the frontal bone and medial surface of the parietal bone. This asymmetric organization may ensure the reproducible architecture of the coronal suture, with the parietal bone consistently overlapping above the frontal bone. In a companion study, we have found that the asymmetric distribution of these progenitors, as marked by Six2 and Grem1 protein, is lost upon disruption of Tcf12 and/or Twist1 function, consistent with bones meeting end-on-end and fusing in this craniosynostosis model (Ting et al. submitted). Within the developing bone, we resolved a spatial hierarchy of osteoblasts, with progenitors located in the suture mesenchyme and giving rise to osteoblasts that are found progressively further away from the suture as they mature. Our data also revealed distinct routes of osteoblast differentiation through either suture-resident pre-osteoblasts or periosteal cells more broadly distributed along the surfaces of bones. Suture-resident pre-osteoblasts are enriched for Wnt-related genes, perhaps reflecting the known involvement of genes from this pathway in coronal suture development 56 , 17 , as well as proliferative markers such as Cenpf . These findings support a model in which proliferative bone growth from the coronal suture extends the lengths of bones, whereas more restrained osteoblast differentiation along the bony surfaces increases the thickness of the calvarial bones, particularly at postnatal stages 57 . The underlying meninges are important in sutural and calvarial development 12 , 58 , and we uncovered multiple distinct layers associated with the coronal suture. In particular, we resolved the dura mater into a dural border cell layer and two layers consistent with periosteal dura 27 , 28 . Interestingly, the periosteal dura layer, which sits closest to the suture and calvarial bones, was distinguished by expression of several chondrogenic markers, including Matn4 , Ctgf , and lower levels of Col2a1 and Acan . Cartilage formation beneath sutures has been linked to normal and pathological suture fusion 59 , 60 , yet the precise source of such cartilage has remained unclear. It will be interesting to assess whether the periosteal dura layers contribute to the natural and ectopic cartilage formation associated with suture closure. In addition, the periosteal dura layers express multiple signalling factors implicated in suture regulation, including Tgfb2 , Fgf2 , Gdf10 , and Ctgf 61 , and CellPhoneDB analysis points to signalling interactions between the periosteal dura layers and pre-osteoblasts. Our single-cell atlas therefore provides testable hypotheses for how specific meningeal cell types may regulate calvarial bone formation in a paracrine fashion. The ectocranial mesenchyme above the coronal suture is also known to regulate suture patency 7 , 22 , and here we identify at least four distinct layers. The two outermost layers likely represent the lower limits of the skin, including dermal reticular fibroblasts ( C1qtnf3 + ) and non-osteogenic Ly6a + ( Sca1 + ) positive hypodermis 31 , 62 , 63 . Whereas the hypodermis expresses Jag1 , Epha4 is enriched in the dermal reticular fibroblast layer. As both Jag1 and Epha4 have been implicated in coronal suture formation 7 , 22 , our data suggest that multiple ectocranial layers play a role in suture regulation. Below the hypodermis, we identified a C1qtnf3 + / Tnmd + layer that we termed suprasutural, as well as a related and tightly associated layer enriched for expression of genes involved in ligament formation, cellular contraction, and mechanosensation. A dense network of collagen fibres is present within adult mouse coronal sutures 64 and may correspond to the ligament-like population in our embryonic dataset. One possibility is that this ligament-like population, which connects the lateral tips of the frontal and parietal bones, may contribute to the flexibility of the sutures, for example to accommodate compression of the calvarium during birth. More speculatively, this population could also function to interpret mechanical forces and transmit these to the osteogenic cells within the suture, thus coupling expansion of the brain to calvarial growth. As with the periosteal dura layers of the meninges, the suprasutural and ligament-like layers have strong predicted signalling interactions with osteogenic cells within the suture, including expression of members of the Tgfβ, Fgf, Wnt, and Igf1 signalling families. It will be interesting to determine the extent to which meningeal and ectocranial layers regulate osteogenic differentiation through similar or distinct means, and whether these layers help sense mechanical forces driving skull expansion. The broad distribution of craniosynostosis-related genes within our dataset suggests diverse etiologies of synostosis. However, we did note that nearly half of synostosis genes had highest expression within the proliferative osteogenic clusters, including Twist1 and Tcf12 that have been implicated in negative regulation of the rate of proliferative bone growth in the calvarium 20 . Thus, misregulation of osteogenic cell proliferation may be a common driver of coronal synostosis, although other coronal synostosis genes had preferential enrichment in meningeal ( Ctsk , Efnb1 ) or ectocranial ( Jag1 ) layers. In addition, midline synostosis-related genes had much lower expression within our dataset, which may reflect the distinct genetic sensitivities of particular cranial sutures. Comparison to a recently published dataset for the frontal suture 30 highlights both conserved and divergent features of the coronal suture (Supplementary Fig. 11). Similar hypodermis (EC1-3 HD), dura mater (MG1-4 DM), and osteoblast (OG4 OB) populations were identified within both. A subpopulation of Ctgf -expressing FS3 cells was also found below the frontal suture, which may correspond to MG1 in our coronal suture dataset. Similar populations to EC4 (FS1) and LIG (FS2) were also captured within the metopic suture datasets. However, there are prominent differences in the spatial distribution of these clusters between sutures. Whereas the EC4 suprasutural population is distributed above the bones in the coronal suture, the comparable FS1 population appears to occupy the bulk of the suture mesenchyme in the frontal suture. Similarly, the LIG population in the coronal suture connects adjacent bones, as true ligaments do, yet the comparable FS2 population is embedded within the frontal suture mesenchyme distant from bone. The biggest differences between the frontal and coronal datasets can be seen in the osteogenic mesenchyme. Whereas both datasets capture osteoblasts and proliferating pre-osteoblasts (PO1/PO2 FS4), the earliest Erg + progenitors and distinct routes through suture-resident pre-osteoblasts (OG2) and periosteal pre-osteoblasts (OG3) at the coronal suture were not apparent in the frontal suture dataset. The frontal suture cluster FS3 appears closest to these signatures, and it is possible that periosteal tissues were not included to the same extent in dissection of the frontal suture. However, there is no evidence of the asymmetric distribution of osteoprogenitors seen at the coronal suture for the frontal suture, which likely reflects that bones meeting at the frontal suture do not overlap in the same way as at the coronal suture. Our analysis therefore reveals differences in the topological arrangements of cell populations between sutures, and possibly in the identities and functions of the populations themselves. This atlas of the coronal suture therefore will be an important resource for understanding why specific sutures are reproducibly affected in distinct human craniosynostosis syndromes. Methods Coronal Suture Dissociations. E15.5 and E17.5 embryos were isolated and the bony skull was dissected away from the skin and brain. For E15.5 embryos the coronal suture was dissected away from the skull cap using microdissection scissors, and 10 pooled coronal sutures were rinsed in PBS and enzymatically dissociated with a final concentration of 3 mg/mL of Collagenase II (Worthington) and 4 units/mL of Dispase (Corning) in DMEM/F12 (Corning) for 45 min. For E17.5 embryos coronal sutures were dissected out in ice cold PBS using a scalpel blade, isolating a strip containing the overlapping frontal and parietal bone fronts (which appears opaque compared to adjacent regions) and avoiding the most apical and basal aspects of the suture. Isolated sutural strips from embryos from two litters in 3 batches (batch 1, 10 sutures from litter 1; batch 2 and 3, 3 sutures each from litter 2) were cut into small fragments in HBBS and digested using Collagenase IV (Worthington, USA; final concentration in HBBS of 2 mg/mL) for 30 min. Dissociation was terminated with 2% Fetal Bovine Serum and cells were passed through a 0.35 µM filter (E15.5) or Pluri-strainer Mini 70 µm (E17.5; pluriSelect Life Science, Germany). For E15.5 sample preparation, dead cells were removed using the Dead Cell Removal kit (Miltenyi Biotec 130-090-101) and cells counts were determined with a hemocytometer. The three batches of E17.5 dissociated cells were separately sorted by FACs to remove debris, cell doublets and likely dead cells (BD FACSAria Fusion; 100 µM nozzle) prior to library preparation. scRNA-seq library preparation and sequencing. Transcriptome libraries for single cells were captured using 10X genomics Chromium Single Cell 3’ Library and Gel Bead Kit v2 following manufacturer’s guidelines. Sequencing for E15.5 coronal sutures was performed with Illumina’s HiSeq 3000/4000 PE Cluster Kit at the Children’s Hospital Los Angeles’ Molecular Genomics Core, and for E17.5 cells, PE sequencing was run on an Illumina HiSeq 4000 at the Oxford Genomics Centre Wellcome Centre for Human Genetics, Oxford achieving an average of ~ 150,000 mean reads per cell for E15.5 and ~ 92,000 mean reads per cell for E17.5. Bioinformatics analysis. Quality control of raw reads was performed with FastQC version 0.11.7, fastq_screen version 0.7.0 and multiqc 65 version 0.9. Samples were counted individually and aggregated with cellranger (10X Genomics) version 2.1.1 using the mm10 mouse transcriptome. All other parameters were set to their default values. Data analysis was performed with Seurat 23 version 3.2.0. The aggregate gene/barcode matrix (`raw_gene_bc_matrices_mex`) was loaded using `CreateSeuratObject` and `Read10X` with `min.cells = 10, min.genes = 200`. Cells were filtered to exclude cells with fewer than 1000 UMIs, fewer than 1000 genes, more than 7.5% mitochondrial content (calculated as the fraction of reads assigned to a gene on the chromosome `MT`), and more than 3% Hbb/Hba centent (to remove red blood cells, and cells with high contamination for red blood cell specific genes). Filtered cells were normalized with SCTransform, and cell cycle scoring and regression was performed for each dataset with the default list of human cell-cycle genes (converted to mouse gene symbols). Datasets were integrated based on the tutorial “Integration and Label Transfer” from Seurat. In brief, an object list including both datasets was created, 3000 features were selected using the SelectIntegrationFeatures following by PrepSCTIntegration, anchors were identified using FindIntegrationAnchors, and the data was integrated using IntegrateData. Dimensionality reduction was performed using the first 30 principal components (PCs) with UMAP 66 (`min_dist = 0.5, n_neighbors = 50`). Clustering was performed with `FindClusters` and `resolution = 1`. Cluster marker genes were identified with `FindAllMarkers` using parameters `min.pct = 0.2, logfc.threshold = 0.5, max.cells.per.ident = 1000, min.cells.gene = 5`. Secondary dimensionality reduction and clustering of the osteogenic and mesenchymal subset was performed as above, after subsetting for cluster numbers 0, 1, 2, 4, 5, 7, and 9. Trajectory analysis was carried out on the osteogenic subset (clusters OG1-4, PO1-2) using Monocle 3 67 . The Seurat integrated object was converted into a Monocle cell dataset and the cluster information and UMAP coordinates carried over from the Seurat object prior to following the Monocle3 recommended protocols. Cell–cell communication analysis was performed using CellphoneDB v2.1.4 50 , after transforming mouse genes to human homologs. We prioritised potential interactions (p-value < 0.01) and manually selected those that were of biological relevance. Synostosis scores was determined using AddModuleScore in Seurat. RNAScope and immunohistochemistry. RNAscope in situ hybridization was performed the RNAscope Multiplex Fluorescent Kit v2 (Advanced Cell Diagnostics, Newark, CA) according to the manufacturer’s protocol for fixed-frozen sections, with one modification. To retain optimal sectioning quality, the heat antigen retrieval was omitted. TSA® Plus (Fluorescein, Cy3 and. Cy5) reagents were used at 1:1000. For immunohistochemistry, wild-type C57BL/6 embryos were collected at E17.5 and rinsed in ice-cold PBS for 30 minutes followed by head dissection, skin removal and overnight incubation in 4% PFA at room temperature. Heads were then washed multiple times in PBS, decalcified for 2 hours in Calci-ClearTM Rapid (HS-105, National Diagnostics), dehydrated, and paraffin embedded. Embedded tissue was sectioned (5 µm) and then rehydrated, stained and visualized using ImmPRESS® HRP Anti-Rabbit IgG (Peroxidase) Polymer Detection Kit (MP-7451, Vector Laboratories). Primary antibodies were diluted in cold TBS and incubated overnight at 4 °C. In order to detect primary antibodies other than those raised in rabbit, donkey anti-sheep/goat/rat IgG-HRP was used (all 1:200, A16041, sc-2020, A18739, respectively). The tissue sections were subsequently imaged using an Olympus BX60 Microscope (Olympus) and/or NanoZoomer 2.0 HT (Hamamatsu). Secondary antibodies include donkey anti-sheep IgG-HRP (A16041), donkey anti-goat IgG-HRP (sc-2020), and donkey anti-rat IgG-HRP (A18739). To study Tcf7/Cd200 and Tcf7/Dmp1 localization, double immunofluorescence staining was performed on 5 µm E17.5 sections. Following deparaffinisation, rehydration, and heat-mediated antigen retrieval in 10 mM sodium citrate buffer solution (pH 6), samples were blocked in 4% Donkey Serum (D9663, Sigma-Aldrich) for 30 min. Individual sections were then incubated overnight at 4 °C with a mixture of Tcf7 (1:200; C63D9, Cell Signaling Technology) and Cd200 (1:200; AF2724, R&D systems) or Tcf7 (1:200; C63D9, Cell Signaling Technology) and Dmp1 (1:400; AF4386, R&D systems) primary antibodies. Antigen detection was performed using appropriate combination of the Alexa Fluor 488, 555 and 647 secondary antibodies (all 1:500; A21206, A21432 or A11015, A31573; Thermo Fisher Scientific) for 2 h at room temperature in the dark. All primary/secondary antibodies were diluted in SignalBoost™ Immunoreaction Enhancer Kit (407207–1KIT, Calbiochem). After three washes in PBS, sections were incubated with DAPI (1 µg/mL) (Roche, cat # 10 236 276 001). Following multiple washes in PBS, slides were mounted using Vectashield® Antifade Mounting Medium (H-1000Vector Laboratories, Inc.). Imaging was performed using a Zeiss LSM 780 Upright Multi-Photon Confocal Microscope with LD LCI PA 25× /0.8 DIC WD = 0.57 mm Imm Corr (UV)VIS-IR (Oil-Immersion) and Plan-Apochromat 63x/1.4 Oil objectives. Images were obtained using ZEISS ZEN Microscope software. When using two rabbit antibodies, for example the co-localisation of Gja1 (1:100; 3512, Cell Signalling Technology) and Crabp2 (1:750; 10225-1-AP, Proteintech), we used the TSA Cyanine 3 Plus Evaluation Kit (NEL744E001KT, Perkin Elmer) following the manufacturers’ instructions. Primary antibodies were visualised using ImmPRESS® HRP Anti-Rabbit IgG (MP-7451, Vector Laboratories) followed by incubation with Fluorescein (FP1168) or Cyanine 3 (FP1170) Amplification Reagents (both 1:200) and imaged by a Zeiss LSM 780 Upright Multi-Photon Confocal Microscope (same parameters as above). Declarations Data availability The RNAseq data are available in the GEO repository, accession: GSE163693. Acknowledgments We thank Kevin Clarke and Craig Waugh from the WIMM Flow Cytometry Facility, and Claire Arata, Maxwell Serowoky, and Francesca Mariani from the University of Southern California for help with single-cell dissociation and isolation. We thank Julie Siegenthaler from the University of Colorado Anschutz Medical Campus for helpful advice and feedback. We thank the Oxford Genomics Centre at the Wellcome Centre for Human Genetics and the Children’s Hospital Los Angeles’ Molecular Pathology Genomics Core for next-generation sequencing. Work was supported by Wellcome (102731 to AOMW), Action Medical Research (GN2483 to SRFT), VTCT Foundation Fellowship (SRFT, AOMW), the MRC through the WIMM Strategic Alliance (G0902418 and MC_UU_12025), Burroughs Wellcome Trust (DTF), HHMI Hanna H. Gray Fellows Porgram (DTF), National Institutes of Health (R01DE026339 to JGC and REM). The views expressed in this publication are those of the authors and not necessarily those of funding sources. Contributions D.T.F., A.O.M.W., R.E.M., S.R.F.T. and J.G.C. conceived and designed the study. D.T.F., Y.Z. N.K., N.A. and S.R.F.T. carried out the single-cell and bioinformatic analysis. D.T.F. performed the RNAScope validation experiments, and H.M. the immunolocalizations. S.R.F.T., A.O.M.W. and J.G.C. supervised the research. D.T.F., S.R.F.T. and J.G.C wrote the paper with contributions from all authors. Conflict of Interest Statement The authors report no conflicts of interest. Corresponding authors Correspondence to Stephen RF Twigg and J Gage Crump. References Rice, D. P. Developmental anatomy of craniofacial sutures. Front. Oral Biol. 12, 1-21 (2008). Al-Rekabi, Z., Cunningham, M. L. & Sniadecki, N. J. Cell Mechanics of Craniosynostosis. ACS Biomater Sci. Eng. 3, 2733-2743 (2017). Lee, C., Richtsmeier, J. T. & R. H. Kraft, A coupled reaction-diffusion-strain model predicts cranial vault formation in development and disease. Biomech. Model Mechanobiol. 18, 1197-1211 (2019). Lajeunie, E., Le Merrer, M., Bonaïti-Pellie, C., Marchac, D., & Renier, D. Genetic study of nonsyndromic coronal craniosynostosis. Am. J. Med. Genet. 55, 500-504 (1995). Cornelissen, M. et al. Increase of prevalence of craniosynostosis. J. Craniomaxillofac. Surg. 44, 1273–1279 (2016). Wilkie, A. O. M., Johnson, D. & Wall, S. A. Clinical genetics of craniosynostosis. Curr. Opin. Pediatr. 29, 622–628 (2017). Merrill, A. E. et. al. Cell mixing at a neural crest-mesoderm boundary and deficient ephrin-Eph signaling in the pathogenesis of craniosynostosis. Hum. Mol. Genet. 15, 1319-1328 (2006). Holmes, G. et al. Early onset of craniosynostosis in an Apert mouse model reveals critical features of this pathology. Dev. Biol. 328, 273-284 (2009). Sharma, V.P. et al. Mutations in TCF12 , encoding a basic helix-loop-helix partner of TWIST1, are a frequent cause of coronal craniosynostosis. Nat. Genet. 45, 304–307 (2013). Lee, K. L. L., Stanier, P. & Pauws, E. Mouse models of syndromic craniosynostosis. Mol. Syndromol. 10, 58–73 (2019). Ferguson, J. W. & Atit, R. P. A tale of two cities: The genetic mechanisms governing calvarial bone development. Genesis 57, e23248 (2018). Dasgupta, K., & Jeong, J. Developmental biology of the meninges. Genesis 57 , e23288 (2019). Jiang, X., Iseki, S., Maxson, R. E., Sucov, H. M. & Morriss-Kay, G. M. Tissue origins and interactions in the mammalian skull vault. Dev. Biol. 241, 106–116 (2002). Yoshida, T. et al. Cell lineage in mammalian craniofacial mesenchyme. Mech. Dev. 125 , 797-808 (2008). Deckelbaum, R. A. et al. Regulation of cranial morphogenesis and cell fate at the neural crest-mesoderm boundary by engrailed 1. Development 139, 1346-58 (2012). Zhao, H., et al. The suture provides a niche for mesenchymal stem cells of craniofacial bones. Nat. Cell Biol. 17, 386–396 (2015). Maruyama, T., Jeong, J., Sheu, T. -J., & Hsu, W. Stem cells of the suture mesenchyme in craniofacial bone development, repair and regeneration. Nat. Commun. 7, 10526 (2016). Wilk, K., et al. Postnatal calvarial skeletal stem cells expressing PRX1 reside exclusively in the calvarial sutures and are required for bone regeneration. Stem Cell Rep. 8, 933–946 (2017). Doro, D. H. , Grigoriadis, A. E. & Liu, K. J. Calvarial suture-derived stem cells and their contribution to cranial bone repair. Front. Physiol. 8, 956 (2017). Teng, C.S., et al., Altered bone growth dynamics prefigure craniosynostosis in a zebrafish model of Saethre-Chotzen syndrome. Elife 7, e37024 (2018). Ting, M. C. et al. EphA4 as an effector of Twist1 in the guidance of osteogenic precursor cells during calvarial bone growth and in craniosynostosis. Development 136, 855-864 (2009). Yen, H. Y., Ting, M. C., & Maxson, R.E. Jagged1 functions downstream of Twist1 in the specification of the coronal suture and the formation of a boundary between osteogenic and non-osteogenic cells. Dev. Biol. 347, 258-270 (2010). Stuart, T. et al. Comprehensive integration of single-cell data. Cell 177, 1888–1902 (2019). Ayturk, U. M. et al. Single-cel RNA sequencing of calvarial and long bone endocortical cells. J. Bone Miner. Res. 35, 1981-1991 (2020). Desisto, J. et al. Single-cell transcriptomic analyses of the developing meninges reveal meningeal fibroblast diversity and function Dev. Cell 54, 43-59 (2020). Lopes, M. B. S. Meninges: Embryology. in Meningiomas (ed. Lee JH) 25-29 (Springer London, 2009). Nabeshima, S., Reese, T. S., Landis, D. M., & Brightman, M. W. Junctions in the meninges and marginal glia. J. Comp. Neurol. 164, 127–169 (1975). Vandenabeele, F., Creemers, J. & Lambrichts, I. Ultrastructure of the human spinal arachnoid mater and dura mater. J. Anat. 189, 417-430 (1996). Adeeb, N., Mortazavi, M. M., Tubbs, R. S. & Cohen-Gadol, A. A. The cranial dura mater: a review of its history, embryology, and anatomy. Childs Nerv. Syst. 28 , 827–837(2012). Holmes, G. et al. Integrated transcriptome and network analysis reveals spatiotemporal dynamics of calvarial suturogenesis. Cell Reports 32, 107871 (2020). Wolnicka-Glubisz, A., King, W. & Noonan, F. P. Sca-1 + cells with an adipocyte phenotype in neonatal mouse skin . J. Invest. Dermato.l 125, 383–385 (2005). Backman, L. J., Andresson, G., Wennstig, G., Forsgren, S. & Danilson, P. Endogenous substance P production in the Achilles tendon increases with loading in an in vivo model of tendinopathy-peptidergic elevation preceding tendinosis-like tissue changes. J Musculoskelet. Neuronal interact. 11, 133-140 (2011). Jelinsky, S A. Archambault, J., Li, L. & Seeherman H. Tendon‐selective genes identified from rat and human musculoskeletal tissues. J. Orthop. Res. 28, 289-297 (2010). Havis, E. et al. Transcriptomic analysis of mouse limb tendon cells during development. Development 141, 3683-3696. Ducy, P., Zhang, R., Geoffroy, V., Ridall A. L. & Karsenty, G. Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89, 747-754 (1997). Otto, F. et al. Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development. Cell 89, 765-771 (1997). Nakashima, K. et al. The novel zinc finger-containing transcription factor osterix is required for osteoblast differentiation and bone formation. Cell 108, 17-29 (2002). Robledo, R. F., Rajan, L., Li, X. & Lufkin, T. The Dlx5 and Dlx6 homeobox genes are essential for craniofacial, axial, and appendicular skeletal development. Genes Dev. 16, 1089-1101 (2002). Lee, M-H. et al. Dlx5 specifically regulates Runx2 type II expression by binding to homeodomain-response elements in the Runx2 distal promoter. J. Biol. Chem. 280, 35579-35587 (2005). Holleville, N., Matéos, S., Bontoux, M., Bollerot, K. & Monsoro-Burq, A-H. Dlx5 drives Runx2 expression and osteogenic differentiation in developing cranial suture mesenchyme. Dev. Biol. 304, 860-874 (2007). Paic, F. et al. Identification of differentially expressed genes between osteoblasts and osteocytes. Bone 45, 682–692 (2009). Hanagata, N. et al. Characterization of the osteoblast-specific transmembrane protein IFITM5 and analysis of IFITM5-deficient mice. J. Bone Miner. Metab. 29, 279–290 (2011). Pontikoglou, C. et al. CD200 expression in human cultured bone marrow mesenchymal stem cells is induced by pro-osteogenic and pro-inflammatory cues. J. Cell. Mol. Med. 20, 655-665 (2016). Guo, Y. et al. BMP-IHH-mediated interplay between mesenchymal stem cells and osteoclasts supports calvarial bone homeostasis and repair. Bone Res. 6, 30 (2018). Glass, D. A. et al. Canonical Wnt Signaling in Differentiated Osteoblasts Controls Osteoclast Differentiation. Dev. Cell 8, 751-764 (2005). Veistinen, L. K., et al. Regulation of calvarial osteogenesis by concomitant de-repression of GLI3 and activation of IHH targets. Front. Physiol. 8, 1036 (2017). Zhang, L., Leeman, E., Carnes, D. C. & Graves, D. T. Human osteoblasts synthesize and respond to platelet-derived growth factor. Am. J. Physiol. 261, C348-354 (1991). Andrew, J. G., Hoyland, J. A., Freemont, A. J. & Marsh, D. A. Platelet-derived growth factor expression in normally healing human fractures. Bone 16 , 455-460 (1995). Opperman, L. A., Adab, K. & Gakunga, P. T. Transforming growth factor-beta 2 and TGF-beta 3 regulate fetal rat cranial suture morphogenesis by regulating rates of cell proliferation and apoptosis. Dev. Dyn. 219, 237–247 (2000). Efremova, M. et al. CellPhoneDB: inferring cell–cell communication from combined expression of multi-subunit ligand–receptor complexes. Nat. Protoc. 15, 1484–1506 (2020). Ishii, M., Sun, J., Ting, M. -C. & Maxson, R. E. The Development of the Calvarial Bones and Sutures and the Pathophysiology of Craniosynostosis. Curr. Top. Dev. Biol. 115, 131–156 (2015). Twigg, S. R. F. & Wilkie, A. O. M. A Genetic-Pathophysiological Framework for Craniosynostosis. Am. J. Hum. Genet. 97, 359–377 (2015). Goos, J. A. C. & Mathijssen, I. M. J. Genetic causes of craniosynostosis: an update. Mol. Syndromol. 10, 6-23 (2019). Debnath, S., et al. Discovery of a periosteal stem cell mediating intramembranous bone formation. Nature 562, 133-139 (2018). Mizuhashi, K. et al. Resting zone of the growth plate houses a unique class of skeletal stem cells. Nature 563, 254-258 (2018). Yu, H. M. I. et al. The role of Axin2 in calvarial morphogenesis and craniosynostosis. Development 132, 1995-2005 (2005). Morriss-Kay, G. M. & Wilkie, A. O. M. Growth of the normal skull vault and its alteration in craniosynostosis: insights from human genetics and experimental studies. J. Anat. 207 , 637–53 (2005). Opperman, L. A. Cranial sutures as intramembranous bone growth sites. Dev. Dyn. 219, 472–485 (2000). Sahar, D. E., Longaker, M. T. & Quarto, N. et al. Sox9 neural crest determinant gene controls patterning and closure of the posterior frontal cranial suture. Dev. Biol. 280 , 344-361 (2005). He, F. & Soriano, P. Dysregulated PDGFRα signaling alters coronal suture morphogenesis and leads to craniosynostosis through endochondral ossification. Development 144, 4026-4036 (2017). Tang, X. et al. Connective tissue growth factor contributes to joint homeostasis and osteoarthritis severity by controlling the matrix sequestration and activation of latent TGFβ. Ann. Rheum. Dis. 77, 1372-1380 (2018). Steenhuis, P., Carr, K. M., Pettway, G. J. & Ignelzi, M. A. Osteogenic and adipogenic cell fractions isolated from postnatal mouse calvaria. Cell Tissues Organs 190, 150-157 (2009). Haydont, V. et al. Fibroblasts from the human skin dermo-hypodermal junction are distinct from dermal papillary and reticular fibroblasts and from mesenchymal stem cells and exhibit a specific molecular profile related to extracellular matrix organization and modelling. Cells 9, 368 (2020). Khonsari, R. H, Di Rocco, F., Arnaud, E., Sanchez, S. & Tafforeau, P. High-resolution imaging of craniofacial sutures: new tools for understanding the origins of craniosynostoses. Childs Nerv. Syst. 28, 1465-1469 (2012). Ewels, P., Magnusson, M., Lundin, S. & Kaller, M. MultiQC: summarize analysis results for multiple tools and samples in a single report. Bioinformatics 32, 3047-3048 (2016). McInnes, L., Healy, J., Saul, N. & Großberger, L. UMAP: uniform manifold approximation and projection. J. Open Source Softw. 3, 861 (2018). Cao, J. et al. The single-cell transcriptional landscape of mammalian organogenesis. Nature 566, 496–502 (2019). Additional Declarations There is NO Competing Interest. Supplementary Files FarmeretalSupplementaryTable1.xls Supplementary Table 1 FarmeretalSupplementaryTable2.xls Supplementary Table 2 FarmeretelSupplementaryFigures.pdf Supplementary Figures nrreportingsummaryFarmeretal.pdf Reporting Summary Cite Share Download PDF Status: Published Journal Publication published 10 Aug, 2021 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-135455","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":7589173,"identity":"86503610-9f4c-40e4-b7b6-9785c9022fbd","order_by":0,"name":"D'Juan Farmer","email":"","orcid":"","institution":"University of Southern California","correspondingAuthor":false,"prefix":"","firstName":"D'Juan","middleName":"","lastName":"Farmer","suffix":""},{"id":7589174,"identity":"27f52ded-fcdd-45a4-8339-7ba6de633fce","order_by":1,"name":"Hana Mlcochova","email":"","orcid":"","institution":"University of Oxford","correspondingAuthor":false,"prefix":"","firstName":"Hana","middleName":"","lastName":"Mlcochova","suffix":""},{"id":7589175,"identity":"88e77af7-1176-434a-80b2-9276c0b903ae","order_by":2,"name":"Yan Zhou","email":"","orcid":"","institution":"Cancer Research UK Beatson Institute","correspondingAuthor":false,"prefix":"","firstName":"Yan","middleName":"","lastName":"Zhou","suffix":""},{"id":7589176,"identity":"a8b12a46-cdfd-488a-8115-4b9db0cc9cd9","order_by":3,"name":"Nils Koelling","email":"","orcid":"","institution":"University of Oxford","correspondingAuthor":false,"prefix":"","firstName":"Nils","middleName":"","lastName":"Koelling","suffix":""},{"id":7589177,"identity":"65e16bca-fe92-4dcd-b032-0601259d502f","order_by":4,"name":"Guanlin Wang","email":"","orcid":"","institution":"University of Oxford","correspondingAuthor":false,"prefix":"","firstName":"Guanlin","middleName":"","lastName":"Wang","suffix":""},{"id":7589178,"identity":"12c2fdea-3afb-4d55-880c-2afa7bb9fc89","order_by":5,"name":"Neil Ashley","email":"","orcid":"","institution":"University of Oxford","correspondingAuthor":false,"prefix":"","firstName":"Neil","middleName":"","lastName":"Ashley","suffix":""},{"id":7589179,"identity":"c2800286-f50c-4e3d-9212-6e481f9cb74f","order_by":6,"name":"Robert Maxson","email":"","orcid":"","institution":"University of Southern California/Norris Comprehensive Cancer Cente","correspondingAuthor":false,"prefix":"","firstName":"Robert","middleName":"","lastName":"Maxson","suffix":""},{"id":7589180,"identity":"0d1b7017-9a66-407e-9606-e8166b7f25ca","order_by":7,"name":"Andrew Wilkie","email":"","orcid":"https://orcid.org/0000-0002-2972-5481","institution":"University of Oxford, John Radcliffe Hospital","correspondingAuthor":false,"prefix":"","firstName":"Andrew","middleName":"","lastName":"Wilkie","suffix":""},{"id":7589181,"identity":"6b3a9c41-0254-4984-bb14-821130ee8efa","order_by":8,"name":"J Crump","email":"","orcid":"https://orcid.org/0000-0002-3209-0026","institution":"University of Southern California","correspondingAuthor":false,"prefix":"","firstName":"J","middleName":"","lastName":"Crump","suffix":""},{"id":7589182,"identity":"e445222f-55cc-4895-8fcc-bc41bdad3921","order_by":9,"name":"Steve Twigg","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABE0lEQVRIiWNgGAWjYJACZgjF2MDAUCGBJM6DQz0bVAsPWMsZ0rSALGpDlsKhRX5+87HHBRV3GOwlkptffp1nEa3bwHxMgqHGjsHgzAGsWgyOsaUbzzjzjIFHIrHNWnabRO62A2zJBgzHkhkMzjZg18LGYybN23aYgUc6sc1YEqyFx/ABA9sBBoPzOBzWxv9NmvcfTMscsBaDAwz/cGthOMbDJs3bANbS/PBjA9QWxrYDuB12LM3cmOfYYR6e+w/bmBmOAbUcBvolsS+ZRxKH9+WbDz97zFNzWI695/jjjz9q6nK3HW8+JvHhm50c35kEHC4DxQwkDtikwTEBiqYE3BEJ1wJW+/EHHmWjYBSMglEwcgEAuYxaSaccm0QAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-5024-049X","institution":"University of Oxford","correspondingAuthor":true,"prefix":"","firstName":"Steve","middleName":"","lastName":"Twigg","suffix":""}],"badges":[],"createdAt":"2020-12-24 09:16:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-135455/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-135455/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41467-021-24917-9","type":"published","date":"2021-08-10T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":7852582,"identity":"dd6b1f72-afc1-41de-977b-2d64143463e4","added_by":"auto","created_at":"2021-04-09 21:28:22","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1171257,"visible":true,"origin":"","legend":"Single-cell RNA-sequencing analysis of the E15.5 and E17.5 coronal suture. a Schematic of dissection strategy for E15.5 and E17.5 coronal sutures. Yellow dashed lines outline dissected regions. b Uniform Manifold Approximation and Projection (UMAP) plot of integrated E15.5 (8279 cells) and E17.5 (8682 cells) datasets. Osteogenic and mesenchymal cell types outlined by dashed lines. c Dot plot depicting selected markers (determined by adjusted p-value) enriched for each ancillary cell type outside of the osteogenic and mesenchymal population. d UMAP analysis of re-clustered osteogenic and mesenchymal subset outlined in (b) resolves 14 clusters. e The osteogenic and mesenchymal subset separated by developmental stage. f Graphical depiction of the cluster proportions from the osteogenic/mesenchymal subset, plotted as ratio between E15.5 and E17.5 cells within each cluster. g Dot plot showing markers enriched for each cluster within the osteogenic/mesenchymal subset. ","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-135455/v1/a37a8377eddd7119833a9d0d.jpg"},{"id":7852140,"identity":"080c12c1-14af-42bb-8bc6-19d4b8eb64fe","added_by":"auto","created_at":"2021-04-09 21:22:22","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1514398,"visible":true,"origin":"","legend":"Diverse meningeal cell types are resolved by scRNA-seq. a Dot plot depicting selected markers previously associated with the pia mater, the arachnoid, and the dura mater. b Feature plots of genes validated by in situ experiments. c-f Combinatorial in situ analysis of coronal sutures for indicated markers at E15.5. Sp7 marks the frontal (F) and parietal (P) bones, except for insets (boxed regions) below. c Crabp2 and Gjb6. Arrow, Crabp2+/Gjb6+; arrowhead, Crabp2+. d Crabp2 and Nppc. Arrow, Crabp2+/Nppc+; arrowhead, Nppc+. e Matn4 and Crabp2. Arrow, Matn4+; arrowhead, Crabp2+. f Matn4 and Nppc. Arrow, Matn4+/Nppc+; arrowhead, Nppc+; double arrows, Matn4+. g Model summarizing gene expression patterns of meningeal layers captured from single cell analysis. B, Brain. Scale bar = 50 µm. ","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-135455/v1/1c6599fd0514cd11acb07582.jpg"},{"id":7852325,"identity":"67586fec-9a74-43cc-9a15-374fa3eac01b","added_by":"auto","created_at":"2021-04-09 21:25:22","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1210010,"visible":true,"origin":"","legend":"Multiple ectocranial layers overlay the coronal suture. a Feature plots of genes validated by in situ experiments. b-h Combinatorial in situ analysis of coronal sutures for indicated markers at E15.5. Sp7 marks the frontal (F) and parietal (P) bones. b Ly6a and Pi16. c Ly6a and Dpt. Arrow, Dpt+; arrowhead, Ly6a+/Dpt+. d Pi16 and C1qtnf3. Arrows, C1qtnf3+ layers; Arrowhead, Pi16+ layer. e Tnmd and C1qtnf3. Arrow, C1qtnf3+; arrowhead, Tnmd+/C1qtnf3+. Asterisk, Tnmd expression in MG1. f Tac1 and Chodl. Arrowhead, Tac1+/Chodl+ ligament-like population. g Tac1 and Scx. Arrowhead, Tac1+/Scx+ ligament-like population; asterisk, Tac1+/Scx+ suture mesenchyme. h Tac1. i Model summarizing gene expression patterns of ectocranial layers captured from single cell analysis. Scale bars = 50 µm. ","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-135455/v1/097290c84e8b315a9310384b.jpg"},{"id":7852326,"identity":"f256f0b1-b56c-4267-b512-ebbba155a9d1","added_by":"auto","created_at":"2021-04-09 21:25:22","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1629544,"visible":true,"origin":"","legend":"scRNA-seq captures various subtypes of osteoblasts at the coronal suture. a Lineage analysis and b Pseudotime analysis of osteoblast subset using Monocle 3. c Feature plot of selected markers of osteogenesis within the osteoblast subset. d Expression plots of selected genes across pseudotime. e Feature plots of genes validated by in situ experiments. f-i Combinatorial in situ analysis of coronal sutures for indicated markers at E15.5. Sp7 marks the frontal (F) and parietal (P) bones in the left image of each pair. f Erg and Pthlh. Arrowhead marks suture mesenchyme. g Lef1 and Inhba. Arrowheads mark Lef1+/Inhba+ expression near bone tips. h Podnl1 and Mmp13. Arrowheads mark Podnl1+/Mmp13+ expression in periosteum distant from suture. i Ifitm5 and Sost. Arrows, Sp7+ expression in presumptive newly formed osteoblasts; arrowhead, Sost+. j Model summarizing gene expression patterns within the developing bones and coronal suture mesenchyme. Scale bar = 50 µm. ","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-135455/v1/da7c5c7c8b1770e96ff274b9.jpg"},{"id":7852583,"identity":"36a537ee-4bb0-432a-b608-f5961927275a","added_by":"auto","created_at":"2021-04-09 21:28:22","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":693615,"visible":true,"origin":"","legend":"Ligand and receptor expression and predicted interactions at the coronal suture. a Dot plot of a selected group of secreted factors expressed within the coronal suture. b Dot plot of a selected group of receptors expressed within the coronal suture. c Heatmap of interaction scores between clusters from CellPhoneDB analysis. Ectocranial and meningeal clusters are arranged based on their validated distance from the coronal suture and bones. Heatmap scale shows counts of interactions.","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-135455/v1/e5ce34eab2afd9b87c406789.jpg"},{"id":7852330,"identity":"f1048c86-f5f1-49f9-85d5-ab3b3f0fa4e7","added_by":"auto","created_at":"2021-04-09 21:25:22","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1139990,"visible":true,"origin":"","legend":"Genes associated with synostosis are expressed across multiple cell types in the coronal suture. a Dot plot of genes associated with coronal and midline suture synostosis. b Module score for coronal or midline synostosis genes plotted on to the osteogenic/mesenchymal subset UMAP. c Double in situ analysis of Twist1 and Cenpf expression in the coronal suture relative to the frontal (F) and parietal (P) bones marked by Sp7 expression (dashed box shows magnified region of suture mesenchyme below). d Model summarizing coronal suture cell types. Scale bars = 50 µm. ","description":"","filename":"6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-135455/v1/95cb152eb2b96e56c0870fad.jpg"},{"id":15780828,"identity":"0b83b3ea-b232-445c-a8e6-6d7c9326c8ca","added_by":"auto","created_at":"2021-11-22 15:42:07","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1533686,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-135455/v1/26173438-0fb8-425d-8fe3-798004af41b8.pdf"},{"id":7852136,"identity":"0920a1ef-808a-4f3f-97f5-3853bbb3487f","added_by":"auto","created_at":"2021-04-09 21:22:21","extension":"xls","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2459136,"visible":true,"origin":"","legend":"Supplementary Table 1","description":"","filename":"FarmeretalSupplementaryTable1.xls","url":"https://assets-eu.researchsquare.com/files/rs-135455/v1/e798ae83dc84396e3ff57cdd.xls"},{"id":7852137,"identity":"375bc12f-c686-4f94-89b8-bd5ebe340848","added_by":"auto","created_at":"2021-04-09 21:22:22","extension":"xls","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":2313216,"visible":true,"origin":"","legend":"Supplementary Table 2","description":"","filename":"FarmeretalSupplementaryTable2.xls","url":"https://assets-eu.researchsquare.com/files/rs-135455/v1/c43516035fb374852dd8f7dc.xls"},{"id":7852328,"identity":"020606ec-b161-435c-a831-fb576e69f94c","added_by":"auto","created_at":"2021-04-09 21:25:22","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":7441288,"visible":true,"origin":"","legend":"Supplementary Figures","description":"","filename":"FarmeretelSupplementaryFigures.pdf","url":"https://assets-eu.researchsquare.com/files/rs-135455/v1/172f48da6fda8da321276336.pdf"},{"id":7852145,"identity":"ba975ae6-ef33-4c54-94ce-f595233ddec8","added_by":"auto","created_at":"2021-04-09 21:22:22","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":323688,"visible":true,"origin":"","legend":"Reporting Summary","description":"","filename":"nrreportingsummaryFarmeretal.pdf","url":"https://assets-eu.researchsquare.com/files/rs-135455/v1/7f8ca90e7b4c2edb12ac0fab.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"The developing mouse coronal suture at single-cell resolution","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCranial sutures are fibrous joints between calvarial bones that act as zones of bone growth and absorbers of physical forces\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. They comprise the leading edges of abutting calvarial bones separated by mesenchymal tissue. New bone forms by intramembranous ossification in response to expansion of the underlying brain\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. Growth of the bony skull requires the proliferation and osteogenic differentiation of progenitor cells, as well as the maintenance of sufficient undifferentiated cells in the suture to ensure continued bone growth during fetal and postnatal stages. Environmental and/or genetic insults that disrupt the delicate balance of proliferation and differentiation result in premature fusion of cranial sutures, a condition known as craniosynostosis\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe coronal suture, which separates the frontal and parietal bones, is the suture most commonly affected in monogenic craniosynostosis\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. The mouse has proven an effective model for the study of coronal synostosis\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. During cranial development, the coronal suture and closely associated tissues are derived from three distinct populations: the supraorbital mesenchyme which will give rise to the calvarial bones and suture mesenchyme, the meningeal mesenchyme, and non-osteogenic early migrating mesenchyme\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. The meninges form between the brain and calvaria and are essential for development of both\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Coronal suture mesenchyme, derived largely from the mesoderm, forms a boundary between the neural crest-derived frontal and mesoderm-derived parietal bones\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e13\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Embryonic suture mesenchyme originates from \u003cem\u003eGli1\u003c/em\u003e-expressing cells that migrate away from the paraxial cephalic mesoderm at embryonic day (E) 7.5\u003csup\u003e15\u003c/sup\u003e and expand apically to sit between the lateral dermal mesenchyme and medial meningeal layers from E12.5 onwards\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. Whereas these macroscopic developmental steps are well established, the cellular composition of the developing coronal suture remains poorly understood. Markers that label skeletal stem cells within postnatal sutures have been identified\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e, yet none of these markers identify a distinct skeletal progenitor cell population at embryonic stages. Given recent reports that embryonic progenitor dysfunction precedes craniosynostosis\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e, identifying cell type diversity in early forming sutures will be critical for understanding the etiology of this birth defect. A better understanding of embryonic osteogenic and non-osteogenic populations will also inform how the meninges and ectocranial layers contribute to suture patency\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eTo build a cell atlas of the embryonic coronal suture, we combined single-cell transcriptomics with highly resolved in situ analysis to catalogue the cell types present in murine coronal sutures at E15.5 and E17.5. In the ectocranial compartment, we uncovered multiple layers of distinct cell types, including a ligament-like population connecting the lateral aspects of the frontal and parietal bones. Within the multiple layers of the meninges, we revealed an outer dura mater population with a chondrogenic signature, suggesting a latent capacity for chondrocyte differentiation. In the osteogenic population, pseudotime analysis revealed a putative \u003cem\u003eErg\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e/\u003cem\u003ePthlh\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e progenitor that we found to be located in the suture and along the leading edges of the bones. These progenitors fed into two distinct pre-osteoblast trajectories, one concentrated at the growing bone tips and the other localized along the periosteum more distant from the suture. Comparative expression analysis between genes associated with coronal or midline synostosis highlighted the selective expression of many coronal synostosis genes, including \u003cem\u003eTwist1\u003c/em\u003e and \u003cem\u003eTcf12\u003c/em\u003e, within proliferative osteogenic cells, but also within ectocranial and meningeal layers, suggesting heterogenous etiologies for coronal synostosis. We also detected potential ligand-receptor interactions of neighboring ectocranial and meningeal layers with osteogenic cells within the suture, in particular the proliferative osteogenic population. This is in agreement with previous studies showing roles for the ectocranial mesenchyme and meninges in regulating suture patency\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. This single-cell atlas reveals diversity within the developing coronal suture, some of the earliest potential markers for suture-resident osteogenic progenitors, and potential interactions between osteogenic and non-osteogenic populations likely important for proper skull expansion.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eDiverse mesenchymal heterogeneity captured by single-cell sequencing.\u003c/strong\u003e To understand the cellular composition of the embryonic coronal suture, we performed single-cell RNA sequencing at E15.5 and E17.5 on dissected coronal sutures, including small amounts of frontal and parietal bone, after removing the skin and brain (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003ea). We filtered using Seurat 3 R-Package\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e and obtained 8279 cells at E15.5 (median of 2460 genes per cell) and 8682 cells at E17.5 (median of 3200 genes per cell) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eb). We identified 14 cell clusters through unsupervised graph clustering of the two datasets combined (Supplementary Table\u0026nbsp;1). Osteogenic and mesenchymal cell types were identified based on the expression of broad mesenchyme/fibroblast (\u003cem\u003eCol1a1\u003c/em\u003e) and osteoblast (\u003cem\u003eSp7\u003c/em\u003e) markers (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eb, dotted line; Supplementary Fig.\u0026nbsp;1a-b). The identities of clusters outside the osteogenic/mesenchymal subset were resolved using previously reported markers, and included chondrocytes, myeloid cells, mast cells, lymphocytes, pericytes, osteoclasts, endothelial cells, neurons, and glia (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eb, c). All the major cell types in our analysis were present at E15.5 and E17.5 (Supplementary Fig.\u0026nbsp;2a). Chondrocytes were especially abundant at E15.5 (Supplementary Fig.\u0026nbsp;2b), highlighting the close proximity of the E15.5 coronal suture to the chondrocranium. Myeloid cells were more abundant at E17.5, consistent with reports of increased myeloid differentiation during late embryonic stages\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e (Supplementary Fig.\u0026nbsp;2b).\u003c/p\u003e\n\u003cp\u003eTo analyse the osteogenic and mesenchymal cell types that might comprise and support the coronal suture, we re-clustered the osteogenic/mesenchymal population and obtained 14 clusters present at both E15.5 and E17.5 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003ed-f). Analysis of enriched genes for each cluster allowed us to assign probable identities to each cell type (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eg, Supplementary Table\u0026nbsp;2), which we validated by in situ hybridization as described below. We observed one ectocranial cluster strongly over-represented at E15.5 and a meningeal cluster over-represented at E17.5. Osteogenic cells were also more abundant in the E17.5 dataset, although it is unclear whether this reflects true biological differences versus differing cell capture between the dissections (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003ea, f).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDiversity of meningeal layers below the coronal suture.\u003c/strong\u003e The meninges are involved in the development of the calvaria and underlying brain that they separate\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. They comprise dura mater, arachnoid mater, and pia mater. To determine the identity of meningeal tissues included in our dissections, we utilized a recent transcriptomic study of murine E14 meninges as a guide\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. Markers associated with the pia mater (\u003cem\u003eNgfr\u003c/em\u003e, \u003cem\u003eLama1\u003c/em\u003e, \u003cem\u003eRdh10\u003c/em\u003e) were not co-enriched in any of our clusters, consistent with the pia mater being removed with the brain during dissections (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ea). In contrast, markers enriched within the arachnoid mater (\u003cem\u003eAldh1a2\u003c/em\u003e, \u003cem\u003eCldn11\u003c/em\u003e, and \u003cem\u003eTbx18\u003c/em\u003e) were abundant in MG4, and dura mater markers (\u003cem\u003eGja1\u003c/em\u003e, \u003cem\u003eFxyd5\u003c/em\u003e, and \u003cem\u003eCrabp2\u003c/em\u003e) in MG3 and MG4, and to a lesser extent MG1 and MG2 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ea).\u003c/p\u003e\n\u003cp\u003eTo resolve the identity of MG4, we performed \u003cem\u003ein situ\u003c/em\u003e experiments for a highly specific MG4 marker, \u003cem\u003eGjb6\u003c/em\u003e, in combination with the arachnoid/dura mater marker, \u003cem\u003eCrabp2\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eb). \u003cem\u003eCrabp2\u003c/em\u003e and \u003cem\u003eGjb6\u003c/em\u003e overlapped below the bone with \u003cem\u003eCrabp2\u003c/em\u003e single-positive cells (MG3) found above the \u003cem\u003eCrabp2\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e/\u003cem\u003eGjb6\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e domain (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ec). Immunofluorescence for Crabp2 and Gja1 at E17.5 confirmed that these arachnoid/dura mater markers are excluded from the pia mater (Supplementary Fig.\u0026nbsp;3a). \u003cem\u003eRgs5\u003c/em\u003e\u0026thinsp;+\u0026thinsp;pericytes were interspersed with \u003cem\u003eGjb6\u0026thinsp;+\u003c/em\u003e\u0026thinsp;cells in the MG4 arachnoid layer, consistent with the prominent vasculature extending from the border of the dura mater and through the arachnoid mater to the pia mater\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e (Supplementary Fig.\u0026nbsp;3c). In the \u003cem\u003eCrapb2\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e/\u003cem\u003eGjb6\u003c/em\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e layer (MG3), we also observed co-expression of \u003cem\u003eCrabp2\u003c/em\u003e with \u003cem\u003eNppc\u003c/em\u003e, with a zone of \u003cem\u003eNppc\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e/\u003cem\u003eCrabp2\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u0026minus;\u003c/em\u003e\u003c/sup\u003e cells (MG2) above (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ed). In addition, \u003cem\u003eNppc\u003c/em\u003e was co-expressed with the known dura mater marker \u003cem\u003eFxyd5\u003c/em\u003e\u003csup\u003e\u003cem\u003e25\u003c/em\u003e\u003c/sup\u003e (Supplementary Fig.\u0026nbsp;3b, d). MG3 may represent the dural border cell layer that has been described using EM in adult meninges\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe \u003cem\u003eNppc\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e/\u003cem\u003eCrabp1\u003c/em\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e population above \u003cem\u003eNppc\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e/\u003cem\u003eCrabp2\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e cells is consistent with MG2 identity in our UMAP analysis (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eb, d). This domain was also positive for \u003cem\u003eMatn4\u003c/em\u003e and \u003cem\u003eCtgf\u003c/em\u003e, markers that only overlap with \u003cem\u003eNppc\u003c/em\u003e in MG2 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ef, Supplementary Fig.\u0026nbsp;3b, e). \u003cem\u003eMatn4\u003c/em\u003e expression does not overlap with the MG3/MG4 marker \u003cem\u003eCrabp2\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ee), and chondrocytes in other sections express high levels of \u003cem\u003eMatn4\u003c/em\u003e but not \u003cem\u003eCtgf\u003c/em\u003e (Supplementary Fig.\u0026nbsp;3f). We also detected some \u003cem\u003eMatn4\u003c/em\u003e cells that were low or absent for \u003cem\u003eNppc\u003c/em\u003e, and these were situated closer to the suture and calvarial bones than the \u003cem\u003eNppc\u003c/em\u003e-high cells (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ef). Our UMAP analysis indicates that these \u003cem\u003eMatn4\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e/\u003cem\u003eNppc\u003c/em\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e cells represent MG1 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eb). MG1 expressed higher levels of the chondrocyte markers \u003cem\u003eCol2a1\u003c/em\u003e and \u003cem\u003eAcan\u003c/em\u003e than MG2, but not at the level seen in bona fide chondrocytes (Supplementary Fig.\u0026nbsp;3g). These findings are consistent with MG1/MG2 representing a periosteal dura mater population\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e primed to form cartilage, with the MG1 population having a stronger chondrogenic-like program. These data reveal a diversity of mesenchyme cell types within the meninges, with those closest to the suture sharing features with chondrogenic cells and likely functioning as a specialized connective tissue bridging the calvarial bones to the meninges (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eg).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDiversity of ectocranial layers above the coronal suture.\u003c/strong\u003e Ectocranial mesenchyme guides the migration of early osteogenic cells and helps establish suture boundaries\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Consistent with analysis of the mouse frontal suture\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e, we identified clusters enriched for markers of the dermis (\u003cem\u003eLy6a\u003c/em\u003e, \u003cem\u003eDpt\u003c/em\u003e; EC1-3) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eg, Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003ea). \u003cem\u003eLy6a\u003c/em\u003e is expressed in hypodermis of neonatal mouse skin\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e and we observe co-expression of \u003cem\u003eLy6a\u003c/em\u003e with additional EC1-3 markers, \u003cem\u003ePi16\u003c/em\u003e and \u003cem\u003eDpt\u003c/em\u003e, within a single layer of mesenchyme above the skull bones (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eb, c; Supplementary Fig.\u0026nbsp;4a, b). Our previous work had identified expression of \u003cem\u003eJag1\u003c/em\u003e within both an ectocranial layer and suture mesenchyme cells\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e, and scRNAseq analysis shows high expression of \u003cem\u003eJag1\u003c/em\u003e in EC1, and to a lesser extent EC2 and EC3 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003ef-g, Supplementary Fig.\u0026nbsp;4g). Co-expression of \u003cem\u003eJag1\u003c/em\u003e with \u003cem\u003ePi16\u003c/em\u003e confirms \u003cem\u003eJag1\u003c/em\u003e expression within the hypodermal layer (Supplementary Fig.\u0026nbsp;4h).\u003c/p\u003e\n\u003cp\u003eIn addition to hypodermal EC1-3 layers, we also noted a prominent EC4 population defined by \u003cem\u003eC1qtnf3\u003c/em\u003e and \u003cem\u003eTnmd\u003c/em\u003e expression in our UMAP analysis (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003ea). In situ experiments revealed two \u003cem\u003eC1qtnf3\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e ectocranial layers on either side of the \u003cem\u003ePi16\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e hypodermis (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003ed, Supplementary Fig.\u0026nbsp;4c). However, only the \u003cem\u003eC1qtnf3\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e layer between the calvarial bones and hypodermis expresses \u003cem\u003eTnmd\u003c/em\u003e, suggesting that this corresponds to EC4, which we term \u0026ldquo;suprasutural mesenchyme\u0026rdquo; (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003ea, e; Supplementary Fig.\u0026nbsp;4d). It is likely that the outer \u003cem\u003eC1qtnf3\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e layer was removed along with the skin during dissection and thus not included in our single-cell analysis. Consistently, \u003cem\u003eEpha4\u003c/em\u003e, which has also been shown to display ectocranial expression\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e, was most strongly expressed in this outer \u003cem\u003eC1qtnf3\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e layer despite not being particularly enriched in clusters EC1-4 in UMAP analysis (Supplementary Fig.\u0026nbsp;4i). Thus, \u003cem\u003eJag1\u003c/em\u003e and \u003cem\u003eEpha4\u003c/em\u003e appear to label distinct ectocranial layers.\u003c/p\u003e\n\u003cp\u003eInterestingly, just underneath the \u003cem\u003eC1qtnf3\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e/\u003cem\u003eTnmd\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e layer (EC4), and sitting on top of the suture, we observed a \u003cem\u003eC1qtnf3\u003c/em\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e/\u003cem\u003eTnmd\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e population (LIG) that was enriched for a number of genes associated with tendon/ligament development (e.g. \u003cem\u003eScx\u003c/em\u003e, \u003cem\u003eTnmd\u003c/em\u003e, \u003cem\u003eMkx, Thbs2, Bgn\u003c/em\u003e), as well as markers of smooth muscle (e.g. \u003cem\u003eActa2\u003c/em\u003e, \u003cem\u003eTagln\u003c/em\u003e, \u003cem\u003eMyl9\u003c/em\u003e) (Supplementary Fig.\u0026nbsp;4j). Two of the most specific markers for LIG were \u003cem\u003eTac1\u003c/em\u003e, a gene that encodes four neuropeptides including Substance P that is implicated in tendon mechanosensation\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e, and \u003cem\u003eChodl\u003c/em\u003e, a membrane protein with expression in tendons of humans\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e and mouse limbs\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003ea). In situ hybridization for \u003cem\u003eTac1\u003c/em\u003e and \u003cem\u003eChodl\u003c/em\u003e revealed highly restricted expression in mesenchyme overlying the coronal suture and connecting the edges of the frontal and parietal bones (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003ef, h, Supplementary Fig.\u0026nbsp;4e). Although broader, in situ experiments showed that \u003cem\u003eScx\u003c/em\u003e and \u003cem\u003eTnmd\u003c/em\u003e were also expressed within this layer (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eg; Supplementary Fig.\u0026nbsp;4f, k), and we observed only minimal overlap between \u003cem\u003eC1qtnf3\u003c/em\u003e and \u003cem\u003eTac1\u003c/em\u003e expression (Supplementary Fig.\u0026nbsp;4l). Thus, the LIG cluster represents a transcriptionally distinct and spatially confined subset of the ectocranium overlying the coronal suture and connecting the adjacent bones.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDistinct osteoblast trajectories in the suture versus periosteum.\u003c/strong\u003e The suture is a source of new osteoblasts for skull bone expansion. To better understand the trajectories of osteogenesis within and around the sutures, we identified six osteogenic clusters (OG1, OG2, OG3, OG4, PO1, PO2) based on expression of the osteoblast transcription factors \u003cem\u003eRunx2\u003c/em\u003e and \u003cem\u003eSp7\u003c/em\u003e\u003csup\u003e35,36,37\u003c/sup\u003e, as well as \u003cem\u003eDlx5 and Dlx6\u003c/em\u003e\u003csup\u003e38,39,40\u003c/sup\u003e, (Supplementary Fig.\u0026nbsp;5). Pseudotime analysis of these clusters using Monocle3 revealed OG1 to be the earliest lineage cells, followed by two proliferative branches (PO1, PO2) and then distinct OG2 and OG3 trajectories that converged on the osteoblast cluster OG4 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ea, b). These pseudotime trajectories were reflected by increased expression of \u003cem\u003eRunx2\u003c/em\u003e and \u003cem\u003eSp7\u003c/em\u003e from OG1 to OG4 and accumulation of the mature osteoblast markers \u003cem\u003eIfitm5\u003c/em\u003e and \u003cem\u003eDmp1\u003c/em\u003e in OG4 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ec). Prominent cluster markers included \u003cem\u003eErg\u003c/em\u003e and \u003cem\u003ePthlh\u003c/em\u003e (OG1), \u003cem\u003eLef1\u003c/em\u003e and \u003cem\u003eInhba\u003c/em\u003e (OG2), \u003cem\u003eMmp13\u003c/em\u003e and \u003cem\u003ePodnl1\u003c/em\u003e (OG3), and \u003cem\u003eIfitm5\u003c/em\u003e, \u003cem\u003eDmp1\u003c/em\u003e, and \u003cem\u003eSost\u003c/em\u003e (OG4) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ee). Pseudotime visualization shows \u003cem\u003eErg\u003c/em\u003e (OG1) appearing earliest and shutting down as \u003cem\u003eLef1\u003c/em\u003e (OG2) and \u003cem\u003eMmp13\u003c/em\u003e (OG3) become expressed, followed by their extinguishment and progressive appearance of \u003cem\u003eIfitm5, Dmp1\u003c/em\u003e, and \u003cem\u003eSost\u003c/em\u003e (OG4) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ed).\u003c/p\u003e\n\u003cp\u003eIn situ validation at E15.5 showed co-expression of OG1 markers \u003cem\u003eErg\u003c/em\u003e and \u003cem\u003ePthlh\u003c/em\u003e in suture mesenchyme and extending along the edges of the frontal and parietal bone tips (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ef, Supplementary Fig.\u0026nbsp;6a). Interestingly, \u003cem\u003eErg\u003c/em\u003e but not \u003cem\u003ePthlh\u003c/em\u003e was asymmetrically distributed along the bones, with stronger expression above the frontal and below the parietal bone. \u003cem\u003eGli1\u003c/em\u003e and \u003cem\u003ePrrx1\u003c/em\u003e showed similar asymmetric expression above the frontal bone, and \u003cem\u003eGli1\u003c/em\u003e and \u003cem\u003eSix2\u003c/em\u003e below the parietal bone (Supplementary Fig.\u0026nbsp;7a-c). The different asymmetric mesenchyme patterns were not attributable to disproportionate populations of mesenchyme above and below the bones, as ectocranial and meningeal layers were generally evenly distributed along the bones (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e, \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). These findings highlight asymmetric distribution of the earliest osteogenic cells around the bone fronts, which may play a role in ensuring the later reproducible overlap of the parietal over the frontal bone.\u003c/p\u003e\n\u003cp\u003eWe next examined expression of markers for OG2 and OG3, as pseudotime analysis suggested that these represent alternative pathways to osteoblasts (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ea). Markers for both OG2 (\u003cem\u003eLef1\u003c/em\u003e, \u003cem\u003eInhba\u003c/em\u003e) and OG3 (\u003cem\u003ePodnl1\u003c/em\u003e, \u003cem\u003eMmp13\u003c/em\u003e) displayed expression along the outer domains of the \u003cem\u003eSp7\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e bone surface, consistent with a pre-osteoblast identity. Whereas \u003cem\u003eLef1\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e/\u003cem\u003eInhba\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e cells were most abundant at the edges of the growing bones near the suture, \u003cem\u003ePodnl1\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e/\u003cem\u003eMmp13\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e cells were enriched on bone surfaces further away from the suture (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eg, h; Supplementary Fig.\u0026nbsp;6b, c). OG2 may therefore represent more specialized suture-resident pre-osteoblasts, and OG3 periosteal pre-osteoblasts more generally found on bone surfaces.\u003c/p\u003e\n\u003cp\u003eClusters PO1 and PO2 are enriched for markers of proliferation (\u003cem\u003eMki67\u003c/em\u003e, \u003cem\u003eCenpf\u003c/em\u003e, \u003cem\u003eTop2a;\u003c/em\u003e Supplementary Fig.\u0026nbsp;6e), as well as expression of genes from suture-resident clusters, for example \u003cem\u003eErg\u003c/em\u003e (OG1) and \u003cem\u003eLef1\u003c/em\u003e and \u003cem\u003eInhba\u003c/em\u003e (OG2) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003ea, Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ee). In situ validation revealed \u003cem\u003eCenpf\u003c/em\u003e expression around the tips of the growing bones and extending into the sutures, though in comparison to \u003cem\u003eErg\u003c/em\u003e it appeared largely absent from the central part of the suture (Supplementary Fig.\u0026nbsp;6f). These data suggest that proliferative pre-osteoblasts are concentrated at the bone tips, consistent with previous direct evidence for proliferative osteogenic cells at the leading edges of the calvarial bones\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eOG4 represents mature osteoblasts and osteocytes, as revealed by markers such as \u003cem\u003eIfitm5\u003c/em\u003e, \u003cem\u003eDmp1\u003c/em\u003e and \u003cem\u003eSost\u003c/em\u003e and its terminal position in the pseudotime analysis\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ea, c). We observe co-expression of \u003cem\u003eSp7\u003c/em\u003e with \u003cem\u003eIfitm5\u003c/em\u003e throughout most of the bone, except for the leading tips and the surfaces of bones which are \u003cem\u003eSp7\u003c/em\u003e-positive only, consistent with bone growth from both the leading edges and the periosteal surfaces (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ei, Supplementary Fig.\u0026nbsp;6d). Similarly, we observe protein expression of the osteoblast marker Cd200 in bone and the Wnt-responsive transcription factor Tcf7, a pre-osteoblast marker, along the bone tips and periosteal surfaces\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e43\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e44\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e, consistent with \u003cem\u003eCd200\u003c/em\u003e expression in cluster OG4 and \u003cem\u003eTcf7\u003c/em\u003e expression in both OG4 and the suture-specific pre-osteoblast cluster OG2 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eg, Supplementary Fig.\u0026nbsp;8a-c). The mature osteocyte marker \u003cem\u003eSost\u003c/em\u003e, which labels the oldest cells in pseudotime, was expressed in only a few osteocytes distant from the suture (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ei). The spatial organization between osteoblasts, pre-osteoblasts, and progenitors was confirmed with double in situs between \u003cem\u003ePodnl1\u003c/em\u003e and \u003cem\u003ePthlh\u003c/em\u003e or \u003cem\u003eIfitm5\u003c/em\u003e (Supplementary Fig.\u0026nbsp;6g-h). These findings are consistent with bone elongation occurring through osteogenesis at the suture and bone tips, with a distinct osteogenic pathway in the periosteum contributing to bone thickening (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ej).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSignaling interactions between mesenchymal layers captured from single-cell analysis.\u003c/strong\u003e To gain insights into potential signaling between osteogenic cells and adjacent ectocranial and meningeal mesenchyme, we surveyed our datasets for expression of ligands and receptors (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003ea-b). In OG4 osteoblasts, we observed enrichment of \u003cem\u003eTgfb1\u003c/em\u003e, \u003cem\u003eBmp3\u003c/em\u003e, \u003cem\u003eBmp4\u003c/em\u003e, \u003cem\u003eIhh\u003c/em\u003e, and \u003cem\u003ePdgfa\u003c/em\u003e, with Ihh\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e and Pdgfa\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e47\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e known to be secreted by osteoblasts. In addition to osteoprogenitors (OG1), \u003cem\u003ePthlh\u003c/em\u003e ligand was expressed in several meningeal clusters. In the ectocranial clusters EC4 and LIG, which are situated closest to the suture, we detected enrichment of \u003cem\u003eTgfb3\u003c/em\u003e, \u003cem\u003eFgf9\u003c/em\u003e, \u003cem\u003eFgf18\u003c/em\u003e, \u003cem\u003eWnt5a\u003c/em\u003e, \u003cem\u003eWnt9a\u003c/em\u003e, \u003cem\u003eWnt11\u003c/em\u003e, and \u003cem\u003eIgf1\u003c/em\u003e ligands. In the meningeal clusters closest to the suture (MG1, MG2), we also detected \u003cem\u003eTgfb3\u003c/em\u003e and \u003cem\u003eIgf1\u003c/em\u003e ligand expression, as well as \u003cem\u003eTgfb2\u003c/em\u003e that has previously been shown to be a critical meningeal-derived factor for suture morphogenesis\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. Reciprocally, we detected expression of the Tgf\u0026beta; receptor \u003cem\u003eTgfbr3\u003c/em\u003e, the Fgf receptors \u003cem\u003eFgfr1\u003c/em\u003e and \u003cem\u003eFgfr2\u003c/em\u003e, the Wnt receptors \u003cem\u003eFzd1\u003c/em\u003e, \u003cem\u003eFzd2\u003c/em\u003e, and \u003cem\u003eFzd6\u003c/em\u003e, the Ihh \u003cem\u003ereceptor Ptch1\u003c/em\u003e, the Pthlh receptor \u003cem\u003ePth1r\u003c/em\u003e, and the Pdgfa receptor \u003cem\u003ePdgfra\u003c/em\u003e in various osteogenic clusters (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eb). Interestingly, \u003cem\u003eFgfr1\u003c/em\u003e was expressed more strongly in suture-resident pre-osteoblasts (OG2) and \u003cem\u003eFgfr2\u003c/em\u003e in periosteal pre-osteoblasts (OG3). To capture signalling interactions in an unbiased approach, we interrogated our data using the CellPhoneDB package\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e, scoring the ligand-receptor pairings between all clusters (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003ec). Interactions between osteogenic clusters were notably weak, highlighting potential roles for osteogenic - non-osteogenic interactions in coronal suture regulation. Consistently, we noted that the mesenchymal populations closest to the suture (EC4 above, MG1/MG2 below) had the strongest interactions with osteogenic cells. The proliferative PO1 and pre-osteoblast OG3 clusters appeared to have the strongest connections with the surrounding mesenchyme, consistent with these populations residing at the surfaces of bones. Of the individual ligand-receptor interactions that were identified between EC4/MG1/M2 and osteogenic clusters, several pathways are relevant to coronal suture development, including Fgf, Tgf\u0026beta;, and Wnt signalling\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e (Supplementary Fig.\u0026nbsp;9).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSynostosis related genes display diverse expression patterns.\u003c/strong\u003e To determine if genes underlying craniosynostosis in humans\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e51\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e coalesce within a specific cell type, we mapped coronal and midline synostosis genes onto our dataset. Genes associated with midline suture synostosis were less abundant, suggesting that divergent gene expression patterns may contribute to suture-specific fusions (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003ea). We noted that 8/17 coronal synostosis genes were selectively enriched in PO1 or PO2, suggesting misregulation of proliferating osteoblasts as a common mechanism of synostosis. By scoring every cell for the average expression of each coronal or midline synostosis gene, we noted that coronal synostosis scores were more restricted to the meningeal and osteogenic clusters, while midline synostosis genes were broadly distributed across clusters (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eb). Coronal synostosis genes with particular enrichment in PO1 and PO2 include \u003cem\u003eFgfr1\u003c/em\u003e, \u003cem\u003eFgfr2\u003c/em\u003e, \u003cem\u003eTcf12\u003c/em\u003e, \u003cem\u003eTwist1\u003c/em\u003e, and \u003cem\u003eZeb2\u003c/em\u003e. Consistently, in situ validation revealed expression of \u003cem\u003eTwist1\u003c/em\u003e in suture mesenchyme and cells lining the bone fronts, with \u003cem\u003eTwist1\u003c/em\u003e expression also overlapping with the proliferative pre-osteoblast marker \u003cem\u003eCenpf\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003ec, Supplementary Fig.\u0026nbsp;10). Selective enrichment of \u003cem\u003eTwist1\u003c/em\u003e and \u003cem\u003eTcf12\u003c/em\u003e in proliferative pre-osteoblasts is consistent with functional studies showing roles for these transcription factors in regulating the balance between bone differentiation and proliferation\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eProper formation of the coronal suture is a complex process requiring the coordinated development of multiple tissue types, which is reflected in the diverse genetic heterogeneity underlying coronal synostosis. Understanding the mechanisms that cause pathogenic suture fusion in distinct syndromic forms of synostosis has been limited by our incomplete knowledge of the cellular diversity of the developing coronal suture. In an effort to bridge this gap, we have generated a comprehensive spatial and transcriptomic map of the cell types that comprise and support the embryonic coronal suture (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003ed).\u003c/p\u003e\n\u003cp\u003eThe relationship of the embryonic progenitors that build the coronal suture with the adult sutural stem cells that grow the calvarium remains debated. Recent work on postnatal sutures has demonstrated that the suture mesenchyme houses resident skeletal stem cells that produce osteoblasts to grow skull bones\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. However, these markers (\u003cem\u003eGli1\u003c/em\u003e, \u003cem\u003ePrrx1\u003c/em\u003e, \u003cem\u003eAxin2\u003c/em\u003e) broadly label mesenchyme throughout the embryonic skull, making them unsuitable for identifying osteogenic cells at earlier stages. Here we identify \u003cem\u003eErg\u003c/em\u003e and \u003cem\u003ePthlh\u003c/em\u003e as two of the earliest markers for the putative osteoblast progenitors concentrated in the embryonic coronal suture. As \u003cem\u003ePthlh\u003c/em\u003e is a marker for chondrocyte stem cells in the femoral growth plate\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e, Pthlh signaling may be a more general regulator of skeletal progenitors. As opposed to the adult skull where stem cells are tightly restricted to suture mesenchyme, embryonic \u003cem\u003eErg\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e/Pthlh\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e progenitors extend away from the suture along the surfaces of the frontal and parietal bone tips. Interestingly, these progenitors are asymmetrically distributed, with more cells along the lateral surface of the frontal bone and medial surface of the parietal bone. This asymmetric organization may ensure the reproducible architecture of the coronal suture, with the parietal bone consistently overlapping above the frontal bone. In a companion study, we have found that the asymmetric distribution of these progenitors, as marked by \u003cem\u003eSix2\u003c/em\u003e and Grem1 protein, is lost upon disruption of \u003cem\u003eTcf12\u003c/em\u003e and/or \u003cem\u003eTwist1\u003c/em\u003e function, consistent with bones meeting end-on-end and fusing in this craniosynostosis model (Ting et al. submitted).\u003c/p\u003e\n\u003cp\u003eWithin the developing bone, we resolved a spatial hierarchy of osteoblasts, with progenitors located in the suture mesenchyme and giving rise to osteoblasts that are found progressively further away from the suture as they mature. Our data also revealed distinct routes of osteoblast differentiation through either suture-resident pre-osteoblasts or periosteal cells more broadly distributed along the surfaces of bones. Suture-resident pre-osteoblasts are enriched for Wnt-related genes, perhaps reflecting the known involvement of genes from this pathway in coronal suture development\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e56\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e, as well as proliferative markers such as \u003cem\u003eCenpf\u003c/em\u003e. These findings support a model in which proliferative bone growth from the coronal suture extends the lengths of bones, whereas more restrained osteoblast differentiation along the bony surfaces increases the thickness of the calvarial bones, particularly at postnatal stages\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe underlying meninges are important in sutural and calvarial development\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e, and we uncovered multiple distinct layers associated with the coronal suture. In particular, we resolved the dura mater into a dural border cell layer and two layers consistent with periosteal dura\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. Interestingly, the periosteal dura layer, which sits closest to the suture and calvarial bones, was distinguished by expression of several chondrogenic markers, including \u003cem\u003eMatn4\u003c/em\u003e, \u003cem\u003eCtgf\u003c/em\u003e, and lower levels of \u003cem\u003eCol2a1\u003c/em\u003e and \u003cem\u003eAcan\u003c/em\u003e. Cartilage formation beneath sutures has been linked to normal and pathological suture fusion\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e59\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e, yet the precise source of such cartilage has remained unclear. It will be interesting to assess whether the periosteal dura layers contribute to the natural and ectopic cartilage formation associated with suture closure. In addition, the periosteal dura layers express multiple signalling factors implicated in suture regulation, including \u003cem\u003eTgfb2\u003c/em\u003e, \u003cem\u003eFgf2\u003c/em\u003e, \u003cem\u003eGdf10\u003c/em\u003e, and \u003cem\u003eCtgf\u003c/em\u003e\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e, and CellPhoneDB analysis points to signalling interactions between the periosteal dura layers and pre-osteoblasts. Our single-cell atlas therefore provides testable hypotheses for how specific meningeal cell types may regulate calvarial bone formation in a paracrine fashion.\u003c/p\u003e\n\u003cp\u003eThe ectocranial mesenchyme above the coronal suture is also known to regulate suture patency\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e, and here we identify at least four distinct layers. The two outermost layers likely represent the lower limits of the skin, including dermal reticular fibroblasts (\u003cem\u003eC1qtnf3\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e) and non-osteogenic \u003cem\u003eLy6a\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e (\u003cem\u003eSca1\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e) positive hypodermis\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e62\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e. Whereas the hypodermis expresses \u003cem\u003eJag1\u003c/em\u003e, \u003cem\u003eEpha4\u003c/em\u003e is enriched in the dermal reticular fibroblast layer. As both \u003cem\u003eJag1\u003c/em\u003e and \u003cem\u003eEpha4\u003c/em\u003e have been implicated in coronal suture formation\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e, our data suggest that multiple ectocranial layers play a role in suture regulation. Below the hypodermis, we identified a \u003cem\u003eC1qtnf3\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e/\u003cem\u003eTnmd\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e layer that we termed suprasutural, as well as a related and tightly associated layer enriched for expression of genes involved in ligament formation, cellular contraction, and mechanosensation. A dense network of collagen fibres is present within adult mouse coronal sutures\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e64\u003c/span\u003e\u003c/sup\u003e and may correspond to the ligament-like population in our embryonic dataset. One possibility is that this ligament-like population, which connects the lateral tips of the frontal and parietal bones, may contribute to the flexibility of the sutures, for example to accommodate compression of the calvarium during birth. More speculatively, this population could also function to interpret mechanical forces and transmit these to the osteogenic cells within the suture, thus coupling expansion of the brain to calvarial growth. As with the periosteal dura layers of the meninges, the suprasutural and ligament-like layers have strong predicted signalling interactions with osteogenic cells within the suture, including expression of members of the Tgf\u0026beta;, Fgf, Wnt, and Igf1 signalling families. It will be interesting to determine the extent to which meningeal and ectocranial layers regulate osteogenic differentiation through similar or distinct means, and whether these layers help sense mechanical forces driving skull expansion.\u003c/p\u003e\n\u003cp\u003eThe broad distribution of craniosynostosis-related genes within our dataset suggests diverse etiologies of synostosis. However, we did note that nearly half of synostosis genes had highest expression within the proliferative osteogenic clusters, including \u003cem\u003eTwist1\u003c/em\u003e and \u003cem\u003eTcf12\u003c/em\u003e that have been implicated in negative regulation of the rate of proliferative bone growth in the calvarium\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. Thus, misregulation of osteogenic cell proliferation may be a common driver of coronal synostosis, although other coronal synostosis genes had preferential enrichment in meningeal (\u003cem\u003eCtsk\u003c/em\u003e, \u003cem\u003eEfnb1\u003c/em\u003e) or ectocranial (\u003cem\u003eJag1\u003c/em\u003e) layers. In addition, midline synostosis-related genes had much lower expression within our dataset, which may reflect the distinct genetic sensitivities of particular cranial sutures.\u003c/p\u003e\n\u003cp\u003eComparison to a recently published dataset for the frontal suture\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e highlights both conserved and divergent features of the coronal suture (Supplementary Fig.\u0026nbsp;11). Similar hypodermis (EC1-3 \u0026lt;-\u0026gt; HD), dura mater (MG1-4 \u0026lt;-\u0026gt; DM), and osteoblast (OG4 \u0026lt;-\u0026gt; OB) populations were identified within both. A subpopulation of \u003cem\u003eCtgf\u003c/em\u003e-expressing FS3 cells was also found below the frontal suture, which may correspond to MG1 in our coronal suture dataset. Similar populations to EC4 (FS1) and LIG (FS2) were also captured within the metopic suture datasets. However, there are prominent differences in the spatial distribution of these clusters between sutures. Whereas the EC4 suprasutural population is distributed above the bones in the coronal suture, the comparable FS1 population appears to occupy the bulk of the suture mesenchyme in the frontal suture. Similarly, the LIG population in the coronal suture connects adjacent bones, as true ligaments do, yet the comparable FS2 population is embedded within the frontal suture mesenchyme distant from bone. The biggest differences between the frontal and coronal datasets can be seen in the osteogenic mesenchyme. Whereas both datasets capture osteoblasts and proliferating pre-osteoblasts (PO1/PO2 \u0026lt;-\u0026gt; FS4), the earliest \u003cem\u003eErg\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e progenitors and distinct routes through suture-resident pre-osteoblasts (OG2) and periosteal pre-osteoblasts (OG3) at the coronal suture were not apparent in the frontal suture dataset. The frontal suture cluster FS3 appears closest to these signatures, and it is possible that periosteal tissues were not included to the same extent in dissection of the frontal suture. However, there is no evidence of the asymmetric distribution of osteoprogenitors seen at the coronal suture for the frontal suture, which likely reflects that bones meeting at the frontal suture do not overlap in the same way as at the coronal suture. Our analysis therefore reveals differences in the topological arrangements of cell populations between sutures, and possibly in the identities and functions of the populations themselves. This atlas of the coronal suture therefore will be an important resource for understanding why specific sutures are reproducibly affected in distinct human craniosynostosis syndromes.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cspan class=\"ItalicUnderline\"\u003eCoronal Suture Dissociations.\u003c/span\u003e \u003cspan class=\"Underline\"\u003eE15.5 and E17.5 embryos were isolated and the bony skull was dissected away from the skin and brain. For E15.5 embryos the coronal suture was dissected away from the skull cap using microdissection scissors, and 10 pooled coronal sutures were rinsed in PBS and enzymatically dissociated with a final concentration of 3\u0026nbsp;mg/mL of Collagenase II (Worthington) and 4 units/mL of Dispase (Corning) in DMEM/F12 (Corning) for 45\u0026nbsp;min. For E17.5 embryos coronal sutures were dissected out in ice cold PBS using a scalpel blade, isolating a strip containing the overlapping frontal and parietal bone fronts (which appears opaque compared to adjacent regions) and avoiding the most apical and basal aspects of the suture. Isolated sutural strips from embryos from two litters in 3 batches (batch 1, 10 sutures from litter 1; batch 2 and 3, 3 sutures each from litter 2) were cut into small fragments in HBBS and digested using Collagenase IV (Worthington, USA; final concentration in HBBS of 2\u0026nbsp;mg/mL) for 30\u0026nbsp;min. Dissociation was terminated with 2% Fetal Bovine Serum and cells were passed through a 0.35\u0026nbsp;\u0026micro;M filter (E15.5) or Pluri-strainer Mini 70\u0026nbsp;\u0026micro;m (E17.5; pluriSelect Life Science, Germany). For E15.5 sample preparation, dead cells were removed using the Dead Cell Removal kit (Miltenyi Biotec 130-090-101) and cells counts were determined with a hemocytometer. The three batches of E17.5 dissociated cells were separately sorted by FACs to remove debris, cell doublets and likely dead cells (BD FACSAria Fusion; 100\u0026nbsp;\u0026micro;M nozzle) prior to library preparation.\u003c/span\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003escRNA-seq library preparation and sequencing.\u003c/em\u003e Transcriptome libraries for single cells were captured using 10X genomics Chromium Single Cell 3\u0026rsquo; Library and Gel Bead Kit v2 following manufacturer\u0026rsquo;s guidelines. Sequencing for E15.5 coronal sutures was performed with Illumina\u0026rsquo;s HiSeq 3000/4000 PE Cluster Kit at the Children\u0026rsquo;s Hospital Los Angeles\u0026rsquo; Molecular Genomics Core, and for E17.5 cells, PE sequencing was run on an Illumina HiSeq 4000\u0026nbsp;at the Oxford Genomics Centre Wellcome Centre for Human Genetics, Oxford achieving an average of ~\u0026thinsp;150,000 mean reads per cell for E15.5 and ~\u0026thinsp;92,000 mean reads per cell for E17.5.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eBioinformatics analysis.\u003c/em\u003e Quality control of raw reads was performed with FastQC version 0.11.7, fastq_screen version 0.7.0 and multiqc\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e version 0.9. Samples were counted individually and aggregated with cellranger (10X Genomics) version 2.1.1 using the mm10 mouse transcriptome. All other parameters were set to their default values. Data analysis was performed with Seurat\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e version 3.2.0. The aggregate gene/barcode matrix (`raw_gene_bc_matrices_mex`) was loaded using `CreateSeuratObject` and `Read10X` with `min.cells\u0026thinsp;=\u0026thinsp;10, min.genes\u0026thinsp;=\u0026thinsp;200`. Cells were filtered to exclude cells with fewer than 1000 UMIs, fewer than 1000 genes, more than 7.5% mitochondrial content (calculated as the fraction of reads assigned to a gene on the chromosome `MT`), and more than 3% Hbb/Hba centent (to remove red blood cells, and cells with high contamination for red blood cell specific genes). Filtered cells were normalized with SCTransform, and cell cycle scoring and regression was performed for each dataset with the default list of human cell-cycle genes (converted to mouse gene symbols). Datasets were integrated based on the tutorial \u0026ldquo;Integration and Label Transfer\u0026rdquo; from Seurat. In brief, an object list including both datasets was created, 3000 features were selected using the SelectIntegrationFeatures following by PrepSCTIntegration, anchors were identified using FindIntegrationAnchors, and the data was integrated using IntegrateData. Dimensionality reduction was performed using the first 30 principal components (PCs) with UMAP\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e (`min_dist\u0026thinsp;=\u0026thinsp;0.5, n_neighbors\u0026thinsp;=\u0026thinsp;50`). Clustering was performed with `FindClusters` and `resolution\u0026thinsp;=\u0026thinsp;1`. Cluster marker genes were identified with `FindAllMarkers` using parameters `min.pct\u0026thinsp;=\u0026thinsp;0.2, logfc.threshold\u0026thinsp;=\u0026thinsp;0.5, max.cells.per.ident\u0026thinsp;=\u0026thinsp;1000, min.cells.gene\u0026thinsp;=\u0026thinsp;5`. Secondary dimensionality reduction and clustering of the osteogenic and mesenchymal subset was performed as above, after subsetting for cluster numbers 0, 1, 2, 4, 5, 7, and 9. Trajectory analysis was carried out on the osteogenic subset (clusters OG1-4, PO1-2) using Monocle 3\u003csup\u003e67\u003c/sup\u003e. The Seurat integrated object was converted into a Monocle cell dataset and the cluster information and UMAP coordinates carried over from the Seurat object prior to following the Monocle3 recommended protocols. Cell\u0026ndash;cell communication analysis was performed using CellphoneDB v2.1.4\u003csup\u003e50\u003c/sup\u003e, after transforming mouse genes to human homologs. We prioritised potential interactions (p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.01) and manually selected those that were of biological relevance. Synostosis scores was determined using AddModuleScore in Seurat.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eRNAScope and immunohistochemistry.\u003c/em\u003e RNAscope in situ hybridization was performed the RNAscope Multiplex Fluorescent Kit v2 (Advanced Cell Diagnostics, Newark, CA) according to the manufacturer\u0026rsquo;s protocol for fixed-frozen sections, with one modification. To retain optimal sectioning quality, the heat antigen retrieval was omitted. TSA\u0026reg; Plus (Fluorescein, Cy3 and. Cy5) reagents were used at 1:1000. For immunohistochemistry, wild-type C57BL/6 embryos were collected at E17.5 and rinsed in ice-cold PBS for 30 minutes followed by head dissection, skin removal and overnight incubation in 4% PFA at room temperature. Heads were then washed multiple times in PBS, decalcified for 2 hours in Calci-ClearTM Rapid (HS-105, National Diagnostics), dehydrated, and paraffin embedded. Embedded tissue was sectioned (5\u0026nbsp;\u0026micro;m) and then rehydrated, stained and visualized using ImmPRESS\u0026reg; HRP Anti-Rabbit IgG (Peroxidase) Polymer Detection Kit (MP-7451, Vector Laboratories). Primary antibodies were diluted in cold TBS and incubated overnight at 4\u0026nbsp;\u0026deg;C. In order to detect primary antibodies other than those raised in rabbit, donkey anti-sheep/goat/rat IgG-HRP was used (all 1:200, A16041, sc-2020, A18739, respectively). The tissue sections were subsequently imaged using an Olympus BX60 Microscope (Olympus) and/or NanoZoomer 2.0 HT (Hamamatsu). Secondary antibodies include donkey anti-sheep IgG-HRP (A16041), donkey anti-goat IgG-HRP (sc-2020), and donkey anti-rat IgG-HRP (A18739). To study Tcf7/Cd200 and Tcf7/Dmp1 localization, double immunofluorescence staining was performed on 5\u0026nbsp;\u0026micro;m E17.5 sections. Following deparaffinisation, rehydration, and heat-mediated antigen retrieval in 10\u0026nbsp;mM sodium citrate buffer solution (pH 6), samples were blocked in 4% Donkey Serum (D9663, Sigma-Aldrich) for 30\u0026nbsp;min. Individual sections were then incubated overnight at 4\u0026nbsp;\u0026deg;C with a mixture of Tcf7 (1:200; C63D9, Cell Signaling Technology) and Cd200 (1:200; AF2724, R\u0026amp;D systems) or Tcf7 (1:200; C63D9, Cell Signaling Technology) and Dmp1 (1:400; AF4386, R\u0026amp;D systems) primary antibodies. Antigen detection was performed using appropriate combination of the Alexa Fluor 488, 555 and 647 secondary antibodies (all 1:500; A21206, A21432 or A11015, A31573; Thermo Fisher Scientific) for 2\u0026nbsp;h at room temperature in the dark. All primary/secondary antibodies were diluted in SignalBoost\u0026trade; Immunoreaction Enhancer Kit (407207\u0026ndash;1KIT, Calbiochem). After three washes in PBS, sections were incubated with DAPI (1\u0026nbsp;\u0026micro;g/mL) (Roche, cat # 10 236 276 001). Following multiple washes in PBS, slides were mounted using Vectashield\u0026reg; Antifade Mounting Medium (H-1000Vector Laboratories, Inc.). Imaging was performed using a Zeiss LSM 780 Upright Multi-Photon Confocal Microscope with LD LCI PA 25\u0026times; /0.8 DIC WD\u0026thinsp;=\u0026thinsp;0.57\u0026nbsp;mm Imm Corr (UV)VIS-IR (Oil-Immersion) and Plan-Apochromat 63x/1.4 Oil objectives. Images were obtained using ZEISS ZEN Microscope software. When using two rabbit antibodies, for example the co-localisation of Gja1 (1:100; 3512, Cell Signalling Technology) and Crabp2 (1:750; 10225-1-AP, Proteintech), we used the TSA Cyanine 3 Plus Evaluation Kit (NEL744E001KT, Perkin Elmer) following the manufacturers\u0026rsquo; instructions. Primary antibodies were visualised using ImmPRESS\u0026reg; HRP Anti-Rabbit IgG (MP-7451, Vector Laboratories) followed by incubation with Fluorescein (FP1168) or Cyanine 3 (FP1170) Amplification Reagents (both 1:200) and imaged by a Zeiss LSM 780 Upright Multi-Photon Confocal Microscope (same parameters as above).\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe RNAseq data are available in the GEO repository, accession: GSE163693.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Kevin Clarke and Craig Waugh from the WIMM Flow Cytometry Facility, and Claire Arata, Maxwell Serowoky, and Francesca Mariani from the University of Southern California for help with single-cell dissociation and isolation. We thank Julie Siegenthaler from the University of Colorado Anschutz Medical Campus for helpful advice and feedback. We thank the Oxford Genomics Centre at the Wellcome Centre for Human Genetics and the Children\u0026rsquo;s Hospital Los Angeles\u0026rsquo; Molecular Pathology Genomics Core for next-generation sequencing. Work was supported by Wellcome (102731 to AOMW), Action Medical Research (GN2483 to SRFT), VTCT Foundation Fellowship (SRFT, AOMW), the MRC through the WIMM Strategic Alliance (G0902418 and MC_UU_12025), Burroughs Wellcome Trust (DTF), HHMI Hanna H. Gray Fellows Porgram (DTF), National Institutes of Health (R01DE026339 to JGC and REM). The views expressed in this publication are those of the authors and not necessarily those of funding sources.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eContributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eD.T.F., A.O.M.W., R.E.M., S.R.F.T. and J.G.C. conceived and designed the study. D.T.F., Y.Z. N.K., N.A. and S.R.F.T. carried out the single-cell and bioinformatic analysis. D.T.F. performed the RNAScope validation experiments, and H.M. the immunolocalizations. S.R.F.T., A.O.M.W. and J.G.C. supervised the research. D.T.F., S.R.F.T. and J.G.C wrote the paper with contributions from all authors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors report no conflicts of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorresponding authors\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCorrespondence to Stephen RF Twigg and J Gage Crump.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eRice, D. P. Developmental anatomy of craniofacial sutures. \u003cem\u003eFront. Oral Biol.\u003c/em\u003e \u003cstrong\u003e12,\u003c/strong\u003e 1-21 (2008).\u003c/li\u003e\n\u003cli\u003eAl-Rekabi, Z., Cunningham, M. L. \u0026amp; Sniadecki, N. J. Cell Mechanics of Craniosynostosis. \u003cem\u003eACS Biomater Sci. Eng.\u003c/em\u003e \u003cstrong\u003e3,\u003c/strong\u003e 2733-2743 (2017).\u003c/li\u003e\n\u003cli\u003eLee, C., Richtsmeier, J. T. \u0026amp; R. H. Kraft, A coupled reaction-diffusion-strain model predicts cranial vault formation in development and disease. \u003cem\u003eBiomech. Model Mechanobiol.\u003c/em\u003e \u003cstrong\u003e18, \u003c/strong\u003e1197-1211 (2019).\u003c/li\u003e\n\u003cli\u003eLajeunie, E., Le Merrer, M., Bona\u0026iuml;ti-Pellie, C., Marchac, D., \u0026amp; Renier, D. Genetic study of nonsyndromic coronal craniosynostosis. \u003cem\u003eAm. J. Med. Genet.\u003c/em\u003e \u003cstrong\u003e55,\u003c/strong\u003e 500-504 (1995).\u003c/li\u003e\n\u003cli\u003eCornelissen, M. et al. Increase of prevalence of craniosynostosis. \u003cem\u003eJ. Craniomaxillofac. Surg.\u003c/em\u003e \u003cstrong\u003e44,\u003c/strong\u003e 1273\u0026ndash;1279 (2016).\u003c/li\u003e\n\u003cli\u003eWilkie, A. O. M., Johnson, D. \u0026amp; Wall, S. A. Clinical genetics of craniosynostosis. \u003cem\u003eCurr. Opin. Pediatr.\u003c/em\u003e \u003cstrong\u003e29,\u003c/strong\u003e 622\u0026ndash;628 (2017).\u003c/li\u003e\n\u003cli\u003eMerrill, A. E. et. al. Cell mixing at a neural crest-mesoderm boundary and deficient ephrin-Eph signaling in the pathogenesis of craniosynostosis. \u003cem\u003eHum. Mol. Genet.\u003c/em\u003e \u003cstrong\u003e15,\u003c/strong\u003e 1319-1328 (2006).\u003c/li\u003e\n\u003cli\u003eHolmes, G. et al. Early onset of craniosynostosis in an Apert mouse model reveals critical features of this pathology. \u003cem\u003eDev. Biol.\u003c/em\u003e \u003cstrong\u003e328,\u003c/strong\u003e 273-284 (2009).\u003c/li\u003e\n\u003cli\u003eSharma, V.P. et al. Mutations in \u003cem\u003eTCF12\u003c/em\u003e, encoding a basic helix-loop-helix partner of TWIST1, are a frequent cause of coronal craniosynostosis. \u003cem\u003eNat. Genet.\u003c/em\u003e \u003cstrong\u003e45,\u003c/strong\u003e 304\u0026ndash;307 (2013).\u003c/li\u003e\n\u003cli\u003eLee, K. L. L., Stanier, P. \u0026amp; Pauws, E. Mouse models of syndromic craniosynostosis. \u003cem\u003eMol. Syndromol.\u003c/em\u003e \u003cstrong\u003e10,\u003c/strong\u003e 58\u0026ndash;73 (2019).\u003c/li\u003e\n\u003cli\u003eFerguson, J. W. \u0026amp; Atit, R. P. A tale of two cities: The genetic mechanisms governing calvarial bone development. \u003cem\u003eGenesis\u003c/em\u003e \u003cstrong\u003e57,\u003c/strong\u003e e23248 (2018).\u003c/li\u003e\n\u003cli\u003eDasgupta, K., \u0026amp; Jeong, J. Developmental biology of the meninges.\u003cem\u003e Genesis\u003c/em\u003e \u003cstrong\u003e57\u003c/strong\u003e, e23288 (2019).\u003c/li\u003e\n\u003cli\u003eJiang, X., Iseki, S., Maxson, R. E., Sucov, H. M. \u0026amp; Morriss-Kay, G. M. Tissue origins and interactions in the mammalian skull vault. \u003cem\u003eDev. Biol.\u003c/em\u003e \u003cstrong\u003e241,\u003c/strong\u003e 106\u0026ndash;116 (2002).\u003c/li\u003e\n\u003cli\u003eYoshida, T. et al. Cell lineage in mammalian craniofacial mesenchyme. \u003cem\u003eMech. Dev.\u003c/em\u003e \u003cstrong\u003e125\u003c/strong\u003e, 797-808 (2008).\u003c/li\u003e\n\u003cli\u003eDeckelbaum, R. A. et al. Regulation of cranial morphogenesis and cell fate at the neural crest-mesoderm boundary by engrailed 1. \u003cem\u003eDevelopment\u003c/em\u003e \u003cstrong\u003e139,\u003c/strong\u003e 1346-58 (2012).\u003c/li\u003e\n\u003cli\u003eZhao, H., et al. The suture provides a niche for mesenchymal stem cells of craniofacial bones. \u003cem\u003eNat. Cell Biol.\u003c/em\u003e\u003cstrong\u003e17, \u003c/strong\u003e386\u0026ndash;396 (2015).\u003c/li\u003e\n\u003cli\u003eMaruyama, T., Jeong, J., Sheu, T. -J., \u0026amp; Hsu, W. Stem cells of the suture mesenchyme in craniofacial bone development, repair and regeneration.\u003cem\u003e Nat. Commun.\u003c/em\u003e \u003cstrong\u003e7,\u003c/strong\u003e 10526 (2016).\u003c/li\u003e\n\u003cli\u003eWilk, K., et al. Postnatal calvarial skeletal stem cells expressing PRX1 reside exclusively in the calvarial sutures and are required for bone regeneration. \u003cem\u003eStem Cell Rep.\u003c/em\u003e \u003cstrong\u003e8,\u003c/strong\u003e 933\u0026ndash;946 (2017).\u003c/li\u003e\n\u003cli\u003eDoro, D. H. , Grigoriadis, A. E. \u0026amp; Liu, K. J. Calvarial suture-derived stem cells and their contribution to cranial bone repair. \u003cem\u003eFront. Physiol.\u003c/em\u003e \u003cstrong\u003e8,\u003c/strong\u003e 956 (2017).\u003c/li\u003e\n\u003cli\u003eTeng, C.S., et al., Altered bone growth dynamics prefigure craniosynostosis in a zebrafish model of Saethre-Chotzen syndrome. \u003cem\u003eElife\u003c/em\u003e \u003cstrong\u003e7, \u003c/strong\u003ee37024 (2018).\u003c/li\u003e\n\u003cli\u003eTing, M. C. et al. EphA4 as an effector of Twist1 in the guidance of osteogenic precursor cells during calvarial bone growth and in craniosynostosis. \u003cem\u003eDevelopment \u003c/em\u003e\u003cstrong\u003e136,\u003c/strong\u003e 855-864 (2009).\u003c/li\u003e\n\u003cli\u003eYen, H. Y., Ting, M. C., \u0026amp; Maxson, R.E. Jagged1 functions downstream of Twist1 in the specification of the coronal suture and the formation of a boundary between osteogenic and non-osteogenic cells. \u003cem\u003eDev. Biol.\u003c/em\u003e \u003cstrong\u003e347,\u003c/strong\u003e 258-270 (2010).\u003c/li\u003e\n\u003cli\u003eStuart, T. et al. Comprehensive integration of single-cell data. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e177,\u003c/strong\u003e 1888\u0026ndash;1902 (2019).\u003c/li\u003e\n\u003cli\u003eAyturk, U. M. et al. Single-cel RNA sequencing of calvarial and long bone endocortical cells. \u003cem\u003eJ. Bone Miner. Res.\u003c/em\u003e \u003cstrong\u003e35,\u003c/strong\u003e 1981-1991 (2020).\u003c/li\u003e\n\u003cli\u003eDesisto, J. et al. Single-cell transcriptomic analyses of the developing meninges reveal meningeal fibroblast diversity and function \u003cem\u003eDev. Cell\u003c/em\u003e \u003cstrong\u003e54,\u003c/strong\u003e 43-59 (2020).\u003c/li\u003e\n\u003cli\u003eLopes, M. B. S. Meninges: Embryology. in \u003cem\u003eMeningiomas\u003c/em\u003e (ed. Lee JH) 25-29\u0026nbsp;(Springer London, 2009).\u003c/li\u003e\n\u003cli\u003eNabeshima, S., Reese, T. S., Landis, D. M., \u0026amp; Brightman, M. W. Junctions in the meninges and marginal glia. \u003cem\u003eJ. Comp. Neurol.\u003c/em\u003e \u003cstrong\u003e164,\u003c/strong\u003e 127\u0026ndash;169 (1975).\u003c/li\u003e\n\u003cli\u003eVandenabeele, F., Creemers, J. \u0026amp; Lambrichts, I. Ultrastructure of the human spinal arachnoid mater and dura mater. \u003cem\u003eJ. Anat.\u003c/em\u003e \u003cstrong\u003e189,\u003c/strong\u003e 417-430 (1996).\u003c/li\u003e\n\u003cli\u003eAdeeb, N., Mortazavi, M. M., Tubbs, R. S. \u0026amp; Cohen-Gadol, A. A. The cranial dura mater: a review of its history, embryology, and anatomy. \u003cem\u003eChilds Nerv. Syst.\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 827\u0026ndash;837(2012).\u003c/li\u003e\n\u003cli\u003eHolmes, G. et al. Integrated transcriptome and network analysis reveals spatiotemporal dynamics of calvarial suturogenesis. \u003cem\u003eCell Reports\u003c/em\u003e \u003cstrong\u003e32,\u003c/strong\u003e 107871 (2020).\u003c/li\u003e\n\u003cli\u003eWolnicka-Glubisz, A., King, W. \u0026amp; Noonan, F. P. Sca-1\u003csup\u003e+\u003c/sup\u003e cells with an adipocyte phenotype in neonatal mouse skin\u003cem\u003e. J. Invest. Dermato.l\u003c/em\u003e \u003cstrong\u003e125,\u003c/strong\u003e383\u0026ndash;385 (2005).\u003c/li\u003e\n\u003cli\u003eBackman, L. J., Andresson, G., Wennstig, G., Forsgren, S. \u0026amp; Danilson, P. Endogenous substance P production in the Achilles tendon increases with loading in an in vivo model of tendinopathy-peptidergic elevation preceding tendinosis-like tissue changes. \u003cem\u003eJ Musculoskelet. Neuronal interact.\u003c/em\u003e \u003cstrong\u003e11,\u003c/strong\u003e 133-140 (2011).\u003c/li\u003e\n\u003cli\u003eJelinsky, S A. Archambault, J., Li, L. \u0026amp; Seeherman H. Tendon‐selective genes identified from rat and human musculoskeletal tissues. \u003cem\u003eJ. Orthop. Res.\u003c/em\u003e \u003cstrong\u003e28,\u003c/strong\u003e 289-297 (2010).\u003c/li\u003e\n\u003cli\u003eHavis, E. et al. Transcriptomic analysis of mouse limb tendon cells during development. \u003cem\u003eDevelopment\u003c/em\u003e \u003cstrong\u003e141,\u003c/strong\u003e 3683-3696.\u003c/li\u003e\n\u003cli\u003eDucy, P., Zhang, R., Geoffroy, V., Ridall A. L. \u0026amp; Karsenty, G.\u0026nbsp;Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation.\u0026nbsp;\u003cem\u003eCell\u003c/em\u003e\u003cstrong\u003e89,\u003c/strong\u003e 747-754\u0026nbsp;(1997).\u003c/li\u003e\n\u003cli\u003eOtto, F. et al. Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development.\u0026nbsp;\u003cem\u003eCell\u003c/em\u003e\u003cstrong\u003e89,\u003c/strong\u003e 765-771 (1997).\u003c/li\u003e\n\u003cli\u003eNakashima, K. et al. The novel zinc finger-containing transcription factor osterix is required for osteoblast differentiation and bone formation. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e108,\u003c/strong\u003e 17-29 (2002).\u003c/li\u003e\n\u003cli\u003eRobledo, R. F., Rajan, L., Li, X. \u0026amp; Lufkin, T. The Dlx5 and Dlx6 homeobox genes are essential for craniofacial, axial, and appendicular skeletal development. \u003cem\u003eGenes Dev.\u003c/em\u003e \u003cstrong\u003e16,\u003c/strong\u003e 1089-1101 (2002).\u003c/li\u003e\n\u003cli\u003eLee, M-H. et al. Dlx5 specifically regulates Runx2 type II expression by binding to homeodomain-response elements in the Runx2 distal promoter. \u003cem\u003eJ. Biol. Chem.\u003c/em\u003e \u003cstrong\u003e280,\u003c/strong\u003e 35579-35587 (2005).\u003c/li\u003e\n\u003cli\u003eHolleville, N., Mat\u0026eacute;os, S., Bontoux, M., Bollerot, K. \u0026amp; Monsoro-Burq, A-H. \u003cem\u003eDlx5\u003c/em\u003e drives \u003cem\u003eRunx2\u003c/em\u003e expression and osteogenic differentiation in developing cranial suture mesenchyme. \u003cem\u003eDev. Biol.\u003c/em\u003e \u003cstrong\u003e304,\u003c/strong\u003e 860-874 (2007).\u003c/li\u003e\n\u003cli\u003ePaic, F. et al. Identification of differentially expressed genes between osteoblasts and osteocytes. \u003cem\u003eBone\u003c/em\u003e \u003cstrong\u003e45,\u003c/strong\u003e 682\u0026ndash;692 (2009).\u003c/li\u003e\n\u003cli\u003eHanagata, N. et al. Characterization of the osteoblast-specific transmembrane protein IFITM5 and analysis of IFITM5-deficient mice. \u003cem\u003eJ. Bone Miner. Metab.\u003c/em\u003e \u003cstrong\u003e29,\u003c/strong\u003e 279\u0026ndash;290 (2011).\u003c/li\u003e\n\u003cli\u003ePontikoglou, C. et al. CD200 expression in human cultured bone marrow mesenchymal stem cells is induced by pro-osteogenic and pro-inflammatory cues. \u003cem\u003eJ. Cell. Mol. Med.\u003c/em\u003e \u003cstrong\u003e20,\u003c/strong\u003e 655-665 (2016).\u003c/li\u003e\n\u003cli\u003eGuo, Y. et al. BMP-IHH-mediated interplay between mesenchymal stem cells and osteoclasts supports calvarial bone homeostasis and repair. \u003cem\u003eBone Res.\u003c/em\u003e \u003cstrong\u003e6,\u003c/strong\u003e 30 (2018).\u003c/li\u003e\n\u003cli\u003eGlass, D. A. et al. Canonical Wnt Signaling in Differentiated Osteoblasts Controls Osteoclast Differentiation. \u003cem\u003eDev. Cell\u003c/em\u003e \u003cstrong\u003e8,\u003c/strong\u003e 751-764 (2005).\u003c/li\u003e\n\u003cli\u003eVeistinen, L. K., et al. Regulation of calvarial osteogenesis by concomitant de-repression of GLI3 and activation of IHH targets. \u003cem\u003eFront. Physiol. \u003c/em\u003e\u003cstrong\u003e8,\u003c/strong\u003e 1036 (2017).\u003c/li\u003e\n\u003cli\u003eZhang, L., Leeman, E., Carnes, D. C. \u0026amp; Graves, D. T. Human osteoblasts synthesize and respond to platelet-derived growth factor. Am. J. Physiol. 261, C348-354 (1991).\u003c/li\u003e\n\u003cli\u003eAndrew, J. G., Hoyland, J. A., Freemont, A. J. \u0026amp; Marsh, D. A. Platelet-derived growth factor expression in normally healing human fractures. \u003cem\u003eBone\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 455-460 (1995).\u003c/li\u003e\n\u003cli\u003eOpperman, L. A., Adab, K. \u0026amp; Gakunga, P. T. Transforming growth factor-beta 2 and TGF-beta 3 regulate fetal rat cranial suture morphogenesis by regulating rates of cell proliferation and apoptosis. \u003cem\u003eDev. Dyn.\u003c/em\u003e \u003cstrong\u003e219,\u003c/strong\u003e 237\u0026ndash;247 (2000).\u003c/li\u003e\n\u003cli\u003eEfremova, M. et al. CellPhoneDB: inferring cell\u0026ndash;cell communication from combined expression of multi-subunit ligand\u0026ndash;receptor complexes. \u003cem\u003eNat. Protoc.\u003c/em\u003e \u003cstrong\u003e15,\u003c/strong\u003e 1484\u0026ndash;1506 (2020).\u003c/li\u003e\n\u003cli\u003eIshii, M., Sun, J., Ting, M. -C. \u0026amp; Maxson, R. E. The Development of the Calvarial Bones and Sutures and the Pathophysiology of Craniosynostosis. \u003cem\u003eCurr. Top. Dev. Biol.\u003c/em\u003e \u003cstrong\u003e115,\u003c/strong\u003e 131\u0026ndash;156 (2015).\u003c/li\u003e\n\u003cli\u003eTwigg, S. R. F. \u0026amp; Wilkie, A. O. M. A Genetic-Pathophysiological Framework for Craniosynostosis. \u003cem\u003eAm. J. Hum. Genet.\u003c/em\u003e \u003cstrong\u003e97,\u003c/strong\u003e 359\u0026ndash;377 (2015).\u003c/li\u003e\n\u003cli\u003eGoos, J. A. C. \u0026amp; Mathijssen, I. M. J. Genetic causes of craniosynostosis: an update. \u003cem\u003eMol. Syndromol.\u003c/em\u003e \u003cstrong\u003e10,\u003c/strong\u003e 6-23 (2019).\u003c/li\u003e\n\u003cli\u003eDebnath, S., et al. Discovery of a periosteal stem cell mediating intramembranous bone formation. \u003cem\u003eNature \u003c/em\u003e\u003cstrong\u003e562,\u003c/strong\u003e 133-139 (2018).\u003c/li\u003e\n\u003cli\u003eMizuhashi, K. et al. Resting zone of the growth plate houses a unique class of skeletal stem cells. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e563,\u003c/strong\u003e 254-258 (2018).\u003c/li\u003e\n\u003cli\u003eYu, H. M. I. et al. The role of Axin2 in calvarial morphogenesis and craniosynostosis.\u0026nbsp;\u003cem\u003eDevelopment\u003c/em\u003e \u003cstrong\u003e132,\u003c/strong\u003e 1995-2005 (2005).\u003c/li\u003e\n\u003cli\u003eMorriss-Kay, G. M. \u0026amp; Wilkie, A. O. M. Growth of the normal skull vault and its alteration in\u0026nbsp; craniosynostosis: insights from human genetics and experimental studies. \u003cem\u003eJ. Anat.\u003c/em\u003e \u003cstrong\u003e207\u003c/strong\u003e, 637\u0026ndash;53 (2005).\u003c/li\u003e\n\u003cli\u003eOpperman, L. A. Cranial sutures as intramembranous bone growth sites. \u003cem\u003eDev. Dyn.\u003c/em\u003e \u003cstrong\u003e219,\u003c/strong\u003e 472\u0026ndash;485 (2000).\u003c/li\u003e\n\u003cli\u003eSahar, D. E., Longaker, M. T. \u0026amp; Quarto, N. et al. Sox9 neural crest determinant gene controls patterning and closure of the posterior frontal cranial suture. \u003cem\u003eDev. Biol.\u003c/em\u003e \u003cstrong\u003e280\u003c/strong\u003e, 344-361 (2005).\u003c/li\u003e\n\u003cli\u003eHe, F. \u0026amp; Soriano, P. Dysregulated PDGFR\u0026alpha; signaling alters coronal suture morphogenesis and leads to craniosynostosis through endochondral ossification. \u003cem\u003eDevelopment\u003c/em\u003e \u003cstrong\u003e144,\u003c/strong\u003e 4026-4036 (2017).\u003c/li\u003e\n\u003cli\u003eTang, X. et al. Connective tissue growth factor contributes to joint homeostasis and osteoarthritis severity by controlling the matrix sequestration and activation of latent TGF\u0026beta;. \u003cem\u003eAnn. Rheum. Dis.\u003c/em\u003e \u003cstrong\u003e77,\u003c/strong\u003e 1372-1380 (2018).\u003c/li\u003e\n\u003cli\u003eSteenhuis, P., Carr, K. M., Pettway, G. J. \u0026amp; Ignelzi, M. A. Osteogenic and adipogenic cell fractions isolated from postnatal mouse calvaria. \u003cem\u003eCell Tissues Organs \u003c/em\u003e\u003cstrong\u003e190,\u003c/strong\u003e 150-157 (2009).\u003c/li\u003e\n\u003cli\u003eHaydont, V. et al. Fibroblasts from the human skin dermo-hypodermal junction are distinct from dermal papillary and reticular fibroblasts and from mesenchymal stem cells and exhibit a specific molecular profile related to extracellular matrix organization and modelling. \u003cem\u003eCells\u003c/em\u003e \u003cstrong\u003e9,\u003c/strong\u003e 368 (2020).\u003c/li\u003e\n\u003cli\u003eKhonsari, R. H, Di Rocco, F., Arnaud, E., Sanchez, S. \u0026amp; Tafforeau, P. High-resolution imaging of craniofacial sutures: new tools for understanding the origins of craniosynostoses. \u003cem\u003eChilds Nerv. Syst. \u003c/em\u003e\u003cstrong\u003e28,\u003c/strong\u003e 1465-1469 (2012).\u003c/li\u003e\n\u003cli\u003eEwels, P., Magnusson, M., Lundin, S. \u0026amp; Kaller, M. MultiQC: summarize analysis results for multiple tools and samples in a single report. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e32,\u003c/strong\u003e 3047-3048 (2016).\u003c/li\u003e\n\u003cli\u003eMcInnes, L., Healy, J., Saul, N. \u0026amp; Gro\u0026szlig;berger, L. UMAP: uniform manifold approximation and projection. \u003cem\u003eJ. Open Source Softw.\u003c/em\u003e \u003cstrong\u003e3,\u003c/strong\u003e 861 (2018).\u003c/li\u003e\n\u003cli\u003eCao, J. et al. The single-cell transcriptional landscape of mammalian organogenesis. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e566,\u003c/strong\u003e 496\u0026ndash;502 (2019).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"bone growth, coronal suture","lastPublishedDoi":"10.21203/rs.3.rs-135455/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-135455/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSutures separate the flat bones of the skull and enable coordinated growth of the brain and overlying cranium. To uncover the cellular diversity within sutures, we generated single-cell transcriptomes and performed extensive expression validation of the embryonic murine coronal suture. We identify \u003cem\u003eErg\u003c/em\u003e and \u003cem\u003ePthlh\u003c/em\u003e as markers of osteogenic progenitors in sutures, and distinct pre-osteoblast signatures between the bone fronts and periosteum. In the ectocranial layers above the suture, we observe a ligament-like population spanning the frontal and parietal bones. In the dura mater underlying the suture, we detect a chondrocyte-like signature potentially linked to cartilage formation under pathological conditions. Genes mutated in coronal synostosis are preferentially expressed in proliferative osteogenic cells, as well as meningeal layers, suggesting discrete cell types that may be altered in different syndromes. This single-cell atlas provides a resource for understanding development of the coronal suture, the suture most commonly fused in monogenic craniosynostosis.\u003c/p\u003e","manuscriptTitle":"The developing mouse coronal suture at single-cell resolution","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2021-04-09 21:22:20","doi":"10.21203/rs.3.rs-135455/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"881f7b2b-490e-4295-a41d-ccca6362781b","owner":[],"postedDate":"April 9th, 2021","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":3536313,"name":"Developmental Biology"},{"id":3536314,"name":"Medical Genetics"}],"tags":[],"updatedAt":"2021-11-22T15:37:59+00:00","versionOfRecord":{"articleIdentity":"rs-135455","link":"https://doi.org/10.1038/s41467-021-24917-9","journal":{"identity":"nature-communications","isVorOnly":false,"title":"Nature Communications"},"publishedOn":"2021-08-10 04:00:00","publishedOnDateReadable":"August 10th, 2021"},"versionCreatedAt":"2021-04-09 21:22:20","video":"","vorDoi":"10.1038/s41467-021-24917-9","vorDoiUrl":"https://doi.org/10.1038/s41467-021-24917-9","workflowStages":[]},"version":"v1","identity":"rs-135455","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-135455","identity":"rs-135455","version":["v1"]},"buildId":"_2-kVJe1T_tPrBINL-cwx","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}