A single-cell and spatial transcriptomic analysis of pre- and post- marsupialization conventional ameloblastoma

A single-cell and spatial transcriptomic analysis of pre- and post- marsupialization conventional ameloblastoma | 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 A single-cell and spatial transcriptomic analysis of pre- and post- marsupialization conventional ameloblastoma Yueqi Shi, Hao Lu, Shaoyi Wang, Weichen Song This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6903382/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Ameloblastoma (AM) is the most frequent odontogenic tumor with local invasiveness, and conventional ameloblastoma is the most common type of it. This study aimed to gain an unbiased overview of the cellular and spatial component of conventional ameloblastoma. We applied single-cell RNA sequencing and spatial transcriptomics on pre- and post-marsupialization conventional ameloblastoma slices, followed by cluster, trajectory, network analysis and immunofluorescence validation. We found a continuous cell trajectory from pre-marsupialization epithelium-like cell types to post- marsupialization mesenchyme-like cell types. A comparative analysis with our recently established human embryonic dental cell atlas revealed that epithelium-like cells resembled the most primitive outer enamel progenitors, whereas mesenchyme-like cells did not resemble any embryonic mesenchymal cells, representing a differentiated and abnormal state. Motif enrichment analysis and pseudotime analysis found that transcription factors DLX5 and TWIST1, together with their target genes, showed profound alterations paralleled the epithelium-to-mesenchyme transition (EMT). We verified the existence of EMT by double immunofluorescence of two EMT markers FN1 and VIM, and verified the specific expression of KRT15 in pre-marsupialization tissues. Lastly, signaling network analysis found that mesenchyme-derived proteins MMP13 and THBS2 exert feedback regulatory effects on EMT. Our study constructed the cellular and spatial landscape of conventional ameloblastoma, and highlighted hub genes like DLX5 and MMP13 that could serve as potential therapeutic targets. Biological sciences/Cancer/Oral cancer Biological sciences/Cell biology Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introductions Ameloblastoma is the most frequent odontogenic tumor, with no significant gender predilection 1 . In the fifth Edition of the World Health Organization (WHO) Classification of Head and Neck Tumors, ameloblastoma were classified as conventional, extraosseous/peripheral, unicystic, adenoid and metastasizing types 2 . Among them, conventional ameloblastoma is the most common type. Conventional ameloblastoma is defined as a benign epithelial odontogenic neoplasm characterized by a locally infiltrative growth pattern and a variably solid-cystic morphological architecture[3], and it has been a topic of significant interest among oral and maxillofacial researchers. It has a predilection for the mandible and exhibits a wide age range, albeit with a slight peak in the third and fourth decades. In terms of therapy, complete resection with a safe margin remains the gold standard, as conservative approaches often lead to high recurrence rates. However, this treatment can lead to several complications, including dentition and bone defects, lower lip numbness, and masticatory function disorders 3 . Recently, marsupialization plus second-stage curettage was reported as an effective strategy in treating ameloblastoma[4]. Marsupialization, a technique for the treatment of cystic lesions of the jaws, was initially delineated by Partsch in the late 19th century 4 . The approach entailed the externalization of the cystic lesion via the establishment of a surgical window in both the oral mucosa and the cystic wall. In addition, the advent of molecular-targeted therapies and immunotherapy has shown promise in recent studies, offering potential alternatives to the traditional surgical management of this challenging neoplasm 5 . The molecular underpinnings and origins of ameloblastoma have been the subject of extensive investigation, shedding light on the intricate interplay of genetic and epigenetic factors that contribute to its pathogenesis 6 . Mounting evidence highlights the role of aberrant signaling pathways, including the mitogen-activated protein kinase (MAPK) pathway, which has been implicated in the uncontrolled proliferation and survival of ameloblastoma cells 7 . Activating mutations in BRAF and RAS genes have been identified as key drivers of this dysregulated signaling cascade 6 – 10 . Furthermore, recent advancements in genomic technologies have unveiled recurrent mutations in genes such as SMO and PTCH1, pinpointing the involvement of the Hedgehog signaling pathway in the tumorigenesis of this neoplasm 6 . The concepts of ameloblastoma etiology are controversial. Earlier researches found it might be relevant to trauma, inflammation, and dental issues. Modern research connects its development to enamel organ remnants and odontogenic epithelium, with similarities found in tooth germ development, while other theories focus on the differentiation of pre-ameloblasts during tooth development and the role of the stratum intermedium in this process. However, despite these insights, our understanding of the precise cellular and molecular events that trigger the initial formation of ameloblastoma remains limited. Unraveling the early events in tumorigenesis, as well as the exact roles of various signaling pathways and their interactions in the progression of this neoplasm, are critical gaps in our current knowledge. Ongoing research endeavors are focused on elucidating these aspects of ameloblastoma, with the ultimate goal of identifying novel therapeutic targets and improving patient outcome. Single-cell RNA sequencing (scRNA) and spatial transcriptomics (ST) have emerged as transformative tools in the field of dental research, offering unique opportunities to address the current limitations in our understanding of the origins and progression of ameloblastoma 11 – 15 . By enabling the profiling of individual cells within a tumor, scRNA-seq allows for the identification and characterization of distinct cell populations and states, thus providing a high-resolution view of the tumor heterogeneity and the underlying molecular mechanisms driving tumorigenesis. This approach has the potential to reveal novel cellular subtypes and molecular players involved in the initiation and development of ameloblastoma, facilitating the discovery of new therapeutic targets. Furthermore, spatial transcriptomics complements scRNA-seq by preserving the spatial information of gene expression within a tissue, which is crucial for understanding the complex interplay between various cell types, as well as the tumor microenvironment. This powerful technique can elucidate the spatial organization of ameloblastoma and its various histopathological subtypes, shedding light on the role of stromal and immune cells in tumor progression and therapy response. Together, single-cell RNA sequencing and spatial transcriptomics hold great promise in advancing our knowledge of ameloblastoma, offering new avenues for the development of targeted and personalized therapeutic strategies. In light with these advantages, we carried out scRNA and ST on pre- and post-marsupialization conventional ameloblastoma tissues followed by trajectory and network analysis. We mapped the conventional ameloblastoma trajectory to our recently established human embryonic tooth cell atlas 16 and compared between pre- and post-marsupialization tissues. These analyses could provide new insights to 1) the cellular and molecular compositions of conventional ameloblastoma as well as its spatial distribution; 2) the cellular origins of conventional ameloblastoma; and 3) the mechanism and hub genes of the transition from epithelial tumor cell to non-malignment mesenchymal cell, which might be the potential targets for further therapy. Materials and Method Sample collection and preprocessing This study was approved and supervised by Ethical committee of Shanghai Tongren Hospital (“human embryonic tooth development”, 2022-10-01, approval number: K2023013). Written informed consent was provided by all participants. This study was performed in accordance with the Declaration of Helsinki. The 3 patients were all male, aged between 41 and 52 years old, and consulted our department in 2022. Pre-operative CT scans suggested they had conventional ameloblastoma of the mandible. One of the patients had undergone marsupialization in our department in 2020. The marsupialization was performed under local anaesthesia. The opening site was made at the central part of the cystic lesions, and it was enlarged to prevent the premature closure induced by new bone formation. Impacted/offending tooth was extracted. The tooth socket was then enlarged and trimmed. All the interatrial septums were perforated to ensure effective drainage. Histological examination performed on the biopsy specimen obtained from the opening sites also suggested conventional ameloblastoma. An obturator was inserted after the surgery to prevent the premature closure of the opening site. Postoperative clinical and panoramic radiograph examinations showed the range of the lesion was gradually reduced. We then performed a second-stage ameloblastoma curettage for him 2 years after the marsupialization. The remaining 2 patients had not undergone any treatment, and we performed mandibular osteotomy on them. During the surgery, tumor tissues were collected and respectively placed in tissue preservation solution (Miltenyi, German) and optimum cutting temperature compound (Sakura, U.S.A). Post-operative pathological diagnosis revealed that all 3 patients had conventional ameloblastoma. scRNA and ST The procedure of library construction, sequencing and quality control was similar to our previous work 16 . We used NovaSeq 6000 platform to sequence libraries of scRNA and ST by paired-end 150 bp sequencing (PE150). We applied Seurat 17 NormalizeData function to obtain the library size-normalized count. Specifically, the global scale normalization method "LogNormalize" normalizes gene expression measurements per cell by total expression, multiplies by a scaling factor (10,000 by default), and transforms via the Seurat: SCTransform function. Cell phases are defined by Seurat: CellCycleScoring function based on list of cell cycle genes. We manually removed all ribosomal genes and mitochondrial genes. We retained non-doublet cells with number of detected genes > 500 and < 2500, and mitochondria gene percentage < 5%. Seurat preprocessed the spatial transcriptome data in a similar fashion. Karyotype analysis We applied CopyKAT 18 on the filtered scRNA count data to define the karyotype of each cell. In brief, at least 5 genes per chromosome were required for each cell to calculate DNA copy numbers. Only genes that are expressed in LOW.DR = 0.05 to UP.DR = 0.2 fractions of cells were kept. For each segment, at least 25 genes were used to calculate copy number. Segmentation parameter was set at KS.cut = 0.1. Cluster analysis For the scRNA data, we applied cluster analysis based on Louvain community detection implemented in Seurat: FindClusters function, principal component fraction = 20, resolution = 0.5. We applied the Seurat: FindMarkers Wilcoxon test to find the top 15 markers for each cluster and define them based on existing knowledge of epithelium and mesenchyme cell types. To integrate scRNA data with previous human embryonic tooth germ scRNA data, we applied Seurat anchored-based integration analysis with the default settings, including method = canonical correlation analysis, original dimension = PCA. Same settings were also used for ST data. Pseudotime analysis We extracted the continuous EMT cell clusters and applied monocle3 19 :learn_graph and order_cells function 19 to learn the pseudotime trajectory. We then applied monocle3: graph_test function on top 5,000 variable genes detected by Seurat: FindVariableGenes function to find all genes that significantly altered along the trajectory. Genes with FDR-adjusted p < 0.01 were further used for module detection by monocle3: find_gene_modules function. We separately applied enrichment analysis and pathway analysis for each module. Transcription factor regulation network analysis We applied SCENIC 20 to construct a transcription factor (TF)-target gene regulatory network targeting epithelial and mesenchymal cells, respectively. TF-motif annotations were downloaded from the SCENIC website, corresponding to hg38 refseq gene annotations, each gene has a 10kb window. We retained the genes of interest for the extended network of each TF and applied Fisher's test to see if the modular genes detected by the pseudotime analysis were enriched in any of the TF regulatory networks. TFs with FDR-adjusted p 2 were retained for each block. We further extracted the global expression of each TF and all its targets (so-called regulons) per cell to verify that this expression value had a similar expression pattern to the module genes. nichenetr analysis We applied signaling network analysis using the nichenetr R package as previously described 21 . Ligands were prioritized by area under the curve (AUC) and Pearson correlation coefficient calculated by nichenetr. Top 20 ligands ranked by AUC were chosen as highlighted ligands for further analysis. To focus on EMT procedure, we only analyzed the second wave genes. Spatial transcriptome analysis After defining cell clusters on scRNA data, we projected it to spatial transcriptome data by Seurat: FindTransferAnchors and TransferData functions to estimate the cell type probability of each spot. We assign cell types with the highest probability to each spot. We used misty R package 22 to estimate the spatial relationship among highlighted cell types, TFs and expression levels of ligands and target genes. We chose the “juxta” mode in misty analysis and trained a linear model to predict each of the target gene expression value by adjacent predicting values. Spatial adjacency was quantified by importance score in the prediction model. To focus on EMT procedure, we only analyzed the second wave genes. Result Cellular and spatial components of conventional ameloblastoma We quantified the single-cell expression of 3,209 cells passing quality control (Method), which were collected from pre- and post-marsupialization conventional ameloblastoma tissues (Fig. 1 A and Figure S1 ). By applying CopyKAT 18 (Fig. 1 B and Figure S2), we confirmed that pre- marsupialization cells were mainly tumor-like aneuploid cell, which gradually transformed into diploid post-marsupialization cells via an intermediate state. This pattern was not biased by cell phase (Figure S1 B). We further applied cluster analysis (Fig. 1 A&C, Table S1 ) to dissect the heterogeneity of each karyotype. A total of 11 cell clusters were found. Aneuploid cells were mainly grouped into a cell cluster with characteristics of premature epithelium (epi-like), with high expression of KRT15 and ETV4 (Fig. 1 A & C). We further validated that the protein expression of KRT15 was restricted to pre- marsupialization tissues by immunofluorescence (Fig. 1 F). The transition cells between aneuploid and diploid cells were grouped into one cluster characterized by MALAT1 and JUN expression, which we defined as stem-like cell cluster. Diploid cells were grouped into clusters with mesenchyme characteristics, in line with the important role of epithelium to mesenchyme transition (EMT) in conventional ameloblastoma 23 . This fibroblast that expressed FN1, and COL1A2 (classified as DCN + and DCN − fibroblast), immature (FCN3 − ) and mature (FCN3 + , PLVAP + ) endothelium that expressed AQP1 and PLVAP. A small subset of SPARC + COL1A2 + SPP1 + cells was defined as SPP1 + osteoblast. Notably, a subset of diploid post- marsupialization cells expressed epithelium stem cell marker CD24 and KRT17 (ESC-like). These results confirmed previous histological observation that post-marsupialization tumor tissues underwent fibrogenesis, epitheliogenesis and osteogenesis procedure 24 , 25 . Lastly, we defined M1 and M2 macrophage, T cells and neutrophils by their corresponding markers CD14, RGS1, CD3 and FCGR3B (Fig. 1 D and Table S1 ) We then dissected slices of pre- and post-marsupialization conventional ameloblastoma (Fig. 1 D) for spatial transcriptome (ST) analysis. Applying Seurat anchor-based integration analysis 17 (Fig. 1 E), we found that pre-marsupialization conventional ameloblastoma slices (lower slices) were predominantly consisted of epi-like cluster, whereas the intermediate stem-like cluster were only found sparsely at the edge of pre-marsupialization conventional ameloblastoma. In post-marsupialization slices (Fig. 1 E and Figure S3), fibroblast was the major component, and endothelium was located in complement to fibroblast and osteoblast, which suggested the existence of osteogenesis in the post-marsupialization tissue. We also analyzed the spatial expression of cell markers (Figure S4), and also supported these patterns. Taken together, these results depicted a general landscape of cellular and spatial component of pre- and post-marsupialization conventional ameloblastoma. Furthermore, our cluster analysis and karyotype analysis suggested that aneuploid epithelial cells underwent EMT into diploid mesenchyme cells after marsupialization. Consistently, EMT markers (FN1, VIM, ACTA2, S100A4, COL1A1, TWIST1) were all highly expressed in the post-marsupialization cells (Figure S1 ). To further verify this result, we applied double immunofluorescence on EMT marker FN1 and VIM proteins, and confirmed that their expression was profoundly elevated in post-marsupialization tissue slice (Fig. 1 G). Conventional ameloblastoma resembled early outer enamel epithelium One of the key questions of conventional ameloblastoma pathology is its cellular origin. By integrating conventional ameloblastoma single-cell data with our recently established human embryonic tooth germ cell atlas15 (Fig. 2 A) with Seurat anchor integration algorithm, we analyzed the cellular origin of conventional ameloblastoma. As shown in Fig. 2 B and C, pre-marsupialization conventional ameloblastoma cells (epi-like cluster) were highly similar to the most primitive epithelial cell type (outer enamel epithelium, OEE) in embryonic tooth germ, whereas the transition state conventional ameloblastoma cells (stem-like cluster) were similar to the second primitive inner enamel epithelium (IEE). This suggested that conventional ameloblastoma might originate from premature OEE. Compared with epi-like cluster, stem-like cluster had more characteristics of differentiated dental epithelial cells of human embryo (Fig. 2 D). On the other hand, post-marsupialization ESC-like cluster did not resemble any of the tooth germ cell types (Fig. 2 D). As for other post-marsupialization cell clusters (Fig. 2 E), they resembled mature mesenchyme cells like endothelium and follicle cell, and did not show characteristics of mesenchyme stem cells or progenitor cells. This result supported that post-marsupialization mesenchyme cells were directly induced from tumor epithelial cells, not migrated from nearby connective tissues. We further integrated embryo tooth germ cell atlas with conventional ameloblastoma spatial transcriptome (Fig. 2 F and Figure S5). We found that pre-marsupialization slices were predominantly consisted of IEE-like cells, whereas cells similar to OEE and ALCAM + SC were distributed at the outer surface. We also found that at the opposite end of OEE there were regions enriched of ameloblast (Figure S5). The stratum intermedium (SI) lineage and odontoblast (Fig. 2 A) was sparse in conventional ameloblastoma slice (Figure S5). Vital role of DLX5 and TWIST1 in epithelium-to-mesenchyme transition We have found that epi-like tumor cell in pre-marsupialization conventional ameloblastoma underwent EMT after marsupialization and gained characteristics of mature mesenchyme cells. We further ask what is the key molecular events during this procedure, especially in the transition state stem-like cell cluster. We apply SCENIC to find active transcription factors (TF) as well as their target genes. As shown in Fig. 3 A and B, by combining expression levels of TF and their targets (“regulon”), we found high cell type specificity. Epi-like cluster showed high expression of regulons like DLX5, ETV5, TAF7 and ATF1, whereas post-marsupialization endothelium and fibroblast-like clusters were characterized by regulons of TWIST1, KLT2, MYC and ERG (Fig. 3 A and B). The transition state stem-like cluster showed weaken expression of both of these regulons, but also exhibited unique expression of YY1, REST and KDM5A (Fig. 3 A and B). We proposed that TF activity transition in stem-like cells might be the key regulator of EMT. To further explore this mechanism, we compared regulon activity between epi-like cell, stem-like aneuploid cell and stem-like diploid cell by t test (Fig. 3 C and Table S2). We found that most key TF showed monotonous alteration during this transition. For example, DLX5 and GTF2B had the highest expression in epi-like cluster, lower expression in stem-like aneuploid cell and the lowest expression in stem-like diploid cell, and the difference between groups were all significant (p < 10 − 8 ). On the other hand, expression of KDM5A and TWIST1 regulon were the lowest in epi-like cell and were highest in stem-like diploid cell. These results suggested that the activation of KDM5A and TWIST1 and the deactivation of DLX5 and GTF2B might play an important role in EMT and the loss of tumor characteristics of conventional ameloblastoma. Other TFs showing significant alteration during process included ETV5, ELK4, TAF7, GATA2, etc. (Table S2). We further explored the biological significance of these key TFs. Applying AUCell 20 on spatial transcriptome data (Fig. 3 D), we found that DLX5 regulon were highly expressed in the entire slice of pre-marsupialization conventional ameloblastoma, without significant region specificity. TWIST1 regulon expression in post-marsupialization slice was largely overlapped with fibroblast distribution (Fig. 3 D). These key TF in EMT were predicted to regulate several genes implicated in ameloblast pathology, including epithelial stem cell marker KRT15, COL1A1 and KRAS, etc. Pseudotime analysis of epithelium-to-mesenchyme transition We next applied pseudotime analysis 26 to reconstruct the trajectory from pre-marsupialization epithelial cells to post-marsupialization mesenchyme cells. As shown in Fig. 4 A, pseudotime analysis verified that there was a continuous monotonous trajectory on EMT, supported the theory that post-marsupialization mesenchyme cells were transferred from pre-marsupialization epithelial cells. Applying gene module analysis from monocle3 19 , we identified four clusters of genes that emerged at different time points, hereafter referred to as four waves of genes (Fig. 4 B). The first wave (Fig. 4 C) was highly expressed in epithelial cells and gradually decreased along pseudotime, including DUSP1, ID3, MSX2 and DLX5. DLX5 and MSX2, two transcription factors, were also predicted to regulate larger-than-expected proportion of first wave genes (Fisher p < 10 − 5 , Table S3). First wave genes were also involved in biological processes related to extracellular matrix organization, epithelial cell proliferation, and transform growth factor beta, MAPK, ERK signaling pathways (gene ontology analysis [GO] FDR < 0.05). In spatial transcriptome (Fig. 4 D), first wave genes were widely expressed in pre-marsupialization slices, covering all structures of conventional ameloblastoma. The second wave genes (Fig. 4 C) peaked at transition state, including NFIX, FOXP1, MALAT1 and HMGB1, etc. They were enriched in predicted targets of KDM5B and NFIA (Table S3), and were enriched in biological processes of RNA splicing, chromatin organization and Wnt signaling pathway, suggesting that these pathways might be the key process for conventional ameloblastoma to lose tumor characteristics. The spatial distribution of second wave genes is similar to the first wave, albeit with a lower expression value (Fig. 4 D). The third wave genes involved a small number of genes including KRT5 and KRT 16. They showed little TF specificity and were only enriched in a few pathways related to T cell functions. The spatial expression (Fig. 4 D) is also sparse, with limited expression at the border of pre-marsupialization slices. Lastly, the fourth wave of genes involved EGR1, VCAM1, CLDN5 other genes related to mesenchyme cell functions (Fig. 4 C). They enriched in targets of TWIST1 and RORB (Table S3), and enriched in zinc ion homeostasis, cytoskeleton organization and intermediate filament (Fig. 4 C). The spatial expression of fourth wave genes were similar to TWIST1 regulon (Fig. 4 D), suggesting that TWIST1 might be one of the key drivers of this wave of gene expression. Positive feedback regulation of EMT via MMP13 and THBS2 Lastly, we explored the upstream signaling pathways that regulated EMT in conventional ameloblastoma. We first applied nichenetr on the transition-related genes (second wave genes in monocle analysis). As shown in Fig. 5 A, transition-related genes were densely regulated by ligands from ESC-like cell clusters, including MMP13, THBS2 and COL4A1 (Table S4). Among them, MMP13 showed the strongest regulation activity with the largest number (19) of predicted targets, including high-confidence target gene like NFIB, SLC38A2, JUN, ZFP36L1 and HSP90AB1 (nichenetr regulation weight > 0.75), and were exclusively expressed in ESC-like cell clusters. In fact, all prioritized ligands were found to expressed in ESC-like cell clusters (Fig. 5 A). Since ESC-like cell cluster is a post-EMT cluster (Fig. 4 ), we proposed that cells completed EMT had positive feedback regulation to further promote EMT. We further applied misty on spatial transcriptome data to analyze the spatial adjacency of these ligands and targets, using misty juxta mode. As shown in Fig. 5 B in Table S5, ligands with the most significant adjacency with transition-related genes included THBS2, SERPING1, COL4A1 and LAMB2, indicating that their spatial expression were close to transition-related genes. For example, THBS2 were found to be enriched around MALAT1, FOXP1 and HSP90AB1, three genes that were also predicted to be targets of THBS2 by nichenetr. These concordant results supported the regulatory role of THBS2 on these transition-related genes. The spatial expression of many of these highlighted ligands, including YHBS2, SERPING1 and LAMB2, were enriched in post-marsupialization tissues and were enriched in mesenchyme tissues (Fig. 5 C), further supporting our conclusion that EMT was promoted by signals from post-EMT cells, which formed a positive feedback loop. Discussion In this study, we applied scRNA and ST on conventional ameloblastoma tissues to construct a spatial cell atlas, providing new clues on its underlying mechanism. We managed to find the cellular origin of conventional ameloblastoma by mapping it to the human embryonic dental scRNA reference, and deciphered the transition procedures from tumor epithelial tissues to mesenchymal tissues after marsupialization by trajectory and network analysis. The origin of ameloblastoma remains a subject of debate. It is hypothesized to arise from several sources: (1) cells of the enamel organ rest, (2) cells from Hertwig's epithelial root sheath or the epithelial cell rest of Malassez, (3) the epithelial lining of an odontogenic cyst, especially a dentigerous cyst, and (4) basal cells of the oral mucosa. Nevertheless, concrete evidence supporting each of these hypotheses is currently absent 27 . Through the integration of conventional ameloblastoma and fetal tooth germ cell atlas data, our findings reveal that conventional ameloblastoma tumor cells predominantly exhibit molecular signatures akin to primitive OEE cells. This observation suggests that conventional ameloblastoma likely arises from an early stage of dental epithelial development. The cellular origin of conventional ameloblastoma can thus be traced to these primitive OEE cells, which possess the ability to self-renew and differentiate into various cell types contributing to the diverse histopathological subtypes observed in the tumor 28 . This insight into the developmental trajectory of conventional ameloblastoma not only furthers our understanding of its pathogenesis but also opens new avenues for targeted therapeutic interventions that may exploit the vulnerabilities inherent in these stem cell-like populations 6 . Additionally, the identification of key regulators and signaling pathways specific to the primitive dental epithelial stem cells could provide valuable information for the development of novel strategies to prevent or halt the initiation and progression of ameloblastoma 29 . The epithelial–mesenchymal transition (EMT) is a crucial event during cell development. In this process, epithelial cells undergo a transformation, adopting mesenchymal fibroblast-like features characterized by decreased intercellular adhesion and enhanced motility. This transition can be categorized into three distinct types, based on the specific biological environment: Types 1, 2, and 3. Type 1 EMT is related with processes such as implantation, embryogenesis, and organogenesis. These events necessitate rapid cellular differentiation into various phenotypes and cell migration within a constrained temporal window. Type 2 EMT plays a role in wound healing and tissue regeneration, leading to the formation of repair-associated mesenchymal cells or fibroblasts. Type 3 EMT plays a pivotal role in tumor progression and metastasis by facilitating the generation, growth, and dissemination of cancerous cells 30 , 31 . We found that marsupialization of conventional ameloblastoma induced the occurrence of type 2 EMT, in which aneuploid tumor epithelial cells transformed into euploid mesenchymal cells, including fibroblasts, vascular endothelial cells, and osteoblasts. These cells are involved in the tissue repair of the jawbone. Our result showed the high expression of several type 2 EMT markers (FN1, VIM, ACTA2, S100A4, COL1A1, TWIST1) in post-marsupialization cells. Among them, TWIST1 has been shown to have an important role in cell plasticity in lung cancers 32 , whereas the type-2 EMT is one form of cell plasticity. Furthermore, the feed-back regulation from post-EMT mesenchyme were also found by previous studies of other tumors 33 – 35 . Double immunofluorescence showed an increase in expression of mesenchymal markers fibronectin and vimentin after marsupialization of conventional ameloblastoma, which also confirmed this finding. Through an in-depth network analysis of the epithelial-to-mesenchymal transition (EMT) process, we identified DLX5 and MSX2 as key transcription factors of the first Wave of EMT genes. Both DLX5 and MSX2 are homeobox genes. Previous study has shown that they have an important role in tooth morphogenesis and development, and that mutations in homeobox genes cause developmental disorders such as odontogenic lesions 36 . In our analysis, they both showed high expression in pre-marsupialization conventional ameloblastoma cells and were decreased during EMT. Furthermore, our signaling network analysis revealed that MMP13 could regulate the downstream expression of various EMT-related genes, making it a potential hub in the EMT procedure of conventional ameloblastoma. Matrix metalloproteinases (MMPs) are crucial for various tissue functions, including cellular regeneration, programmed death, angiogenesis, and other vital processes. These functions are integral to both normal development and pathological events, such as the EMT 37 . MMP13 is not only an essential factor for bone development and bone healing but also has a pivotal role in promoting osteogenic differentiation of mesenchymal stem cells (MSCs) to induce reparative jawbone tissue formation. It was reported that the osteogenic marker genes, ALP and RUNX2, were upregulated in MMP13-overexpressing MSCs through p38 phosphorylation 38 . These results revealed novel potential therapeutic targets in conventional ameloblastoma. Marsupialization has traditionally been the primary treatment for odontogenic cystic lesions, especially large ones in the mandible. Its advantages are new bone formation in the cystic cavity, preservation of oral tissues, pulp vitality maintenance, fewer dental extractions, and protection of crucial anatomical structures. Given this, some clinicians have explored marsupialization for ameloblastoma 3 . It was reported that marsupialization resulted with 18% recurrence rate in unicystic ameloblastoma 39 . However, the rate exceeded 50% in multicystic ameloblastoma 40 , which potentially due to periosteum damage during curettage or incomplete marsupialization because of the exist of bony septae in multilocular lesions. Our research offers a potential solution for cases where marsupialization may be unsuitable or ineffective. Our study presents certain limitations. Non-conventional subtypes of ameloblastoma are relatively rare, and our investigation solely encompassed the conventional type; therefore, our conclusions may not be universally applicable to all patients. Additionally, the sample size is relatively small, which may be subject to individual variation. Future large-scale studies covering a broader range of subtypes and timepoints are warranted to elucidate the comprehensive landscape of ameloblastoma. Conclusion In summary, we have deciphered the cellular and spatial composition of conventional ameloblastoma, revealing its primary origin from early-stage epithelial stem cells. We have also underscored the role of epithelial-to-mesenchymal transition (EMT) in conventional ameloblastoma and identified genes such as DLX5 and MMP13 as key players in this process, rendering them promising therapeutic targets. Continued exploration of the molecular mechanisms governing ameloblastoma and the development of targeted therapies have the potential to significantly improve patient outcomes and advance the management of this challenging neoplasm. Declarations Acknowledgement The authors declare that they have not use AI-generated work in this manuscript Funding This work was supported by National Science Foundation of China (82401759), Research Fund of Shanghai Tongren Hospital, Shanghai Jiaotong University School of Medicine (NO: TRYJ2021JC14, TRKYRC -xx202207). Availability of data and material scRNA and ST data could be obtained from https://data.mendeley.com/datasets/4mdwc4h8rb/1. Ethics approval and consent to participate This study was approved and supervised by Ethical committee of Shanghai Tongren Hospital (“human embryonic tooth development”, 2022-10-01, approval number: K2023013). Written informed consent was provided by all participants. Competing interests All the authors declared no conflict of interest. Consent for publication Not applicable. Author’s contribution Y.S designed and conduct the study, H.L collected and preprocessed the sample and data, S.W interpreted the data, W.S and W.W supervised the data, Y.S drafted the manuscript, all authors read, revised and approved the manuscript. References Johnson, N. R., Gannon, O. M., Savage, N. W. & Batstone, M. D. Frequency of odontogenic cysts and tumors: a systematic review. J. Investig. Clin. Dent. 5 , 9–14 (2014). Vered, M. & Wright, J. M. Update from the 5th Edition of the World Health Organization Classification of Head and Neck Tumors: Odontogenic and Maxillofacial Bone Tumours. Head Neck Pathol. 16 , 63–75 (2022). Yang, Z. et al. Marsupialization of mandibular cystic ameloblastoma: Retrospective study of 7 years. Head Neck 40 , 2172–2180 (2018). Kolari, V., Dhabaria, H., Sengupta, S., Sait, A. I. & Shah, A. Use of Marsupialisation for a Conservative Approach to Huge Cystic Lesions of the Jaws - A Report of three Cases. Ann. Maxillofac. Surg. 12 , 244–247 (2022). Effiom, O. A., Ogundana, O. M., Akinshipo, A. O. & Akintoye, S. O. Ameloblastoma: current etiopathological concepts and management. Oral Dis. 24 , 307–316 (2018). Sweeney, R. T. et al. Identification of recurrent SMO and BRAF mutations in ameloblastomas. Nat. Genet. 46 , 722–5 (2014). Brown, N. A. et al. Activating FGFR2-RAS-BRAF mutations in ameloblastoma. Clin. Cancer Res. 20 , 5517–5526 (2014). Kurppa, K. J. et al. High frequency of BRAFV600E mutations in ameloblastoma. J. Pathol. 232 , 492–498 (2014). Zhang, H. L. et al. Comparative analysis of cellular expression pattern of schizophrenia risk genes in human versus mouse cortex. Cell Biosci. 9 , (2019). Shi, Y., Li, M., Yu, Y., Zhou, Y. & Wang, S. Whole exome sequencing and system biology analysis support the ‘two-hit’ mechanism in the onset of Ameloblastoma. Med. Oral Patol. Oral Cir. Bucal 26 , e510 (2021). Lee, S. et al. Single-Cell RNA Sequencing Analysis of Human Dental Pulp Stem Cell and Human Periodontal Ligament Stem Cell. J. Endod. 48 , 240–248 (2022). Ren, H., Wen, Q., Zhao, Q., Wang, N. & Zhao, Y. Atlas of human dental pulp cells at multiple spatial and temporal levels based on single-cell sequencing analysis. Front. Physiol. 13 , (2022). Hermans, F. et al. Establishment of inclusive single-cell transcriptome atlases from mouse and human tooth as powerful resource for dental research. Front. Cell Dev. Biol. 10 , 1990 (2022). Pagella, P., de Vargas Roditi, L., Stadlinger, B., Moor, A. E. & Mitsiadis, T. A. A single-cell atlas of human teeth. iScience 24 , 102405 (2021). Belkina, A. C. et al. Single-Cell Analysis of the Periodontal Immune Niche in Type 2 Diabetes. J. Dent. Res. 99 , 855–862 (2020). Shi, Y. et al. Spatiotemporal cell landscape of human embryonic tooth development. Cell Prolif. 2023.03.01.530693 (2024) doi:10.1111/cpr.13653. Stuart, T. et al. Comprehensive Integration of Single-Cell Data. Cell 177 , 1888-1902.e21 (2019). Gao, R. et al. Delineating copy number and clonal substructure in human tumors from single-cell transcriptomes. Nat. Biotechnol. 39 , 599–608 (2021). Cao, J. et al. The single-cell transcriptional landscape of mammalian organogenesis. Nature 566 , 496–502 (2019). Aibar, S. et al. SCENIC: single-cell regulatory network inference and clustering. Nat. Methods 14 , 1083–1086 (2017). Shi, Y. et al. A single-cell interactome of human tooth germ from growing third molar elucidates signaling networks regulating dental development. Cell Biosci. 11 , 1–17 (2021). Tanevski, J., Flores, R. O. R., Gabor, A., Schapiro, D. & Saez-Rodriguez, J. Explainable multiview framework for dissecting spatial relationships from highly multiplexed data. Genome Biol. 23 , 1–31 (2022). Siar, C. H. & Ng, K. H. Epithelial-to-mesenchymal transition in ameloblastoma: focus on morphologically evident mesenchymal phenotypic transition. Pathology 51 , 494–501 (2019). Nakamura, N., Higuchi, Y., Tashiro, H. & Ohishi, M. Marsupialization of cystic ameloblastoma: a clinical and histopathologic study of the growth characteristics before and after marsupialization. J. Oral Maxillofac. Surg. 53 , 748–754 (1995). Mohamed, A. A. S. et al. Volumetric change of bony cavity and shrinkage speed after marsupialization for odontogenic keratocyst and unicystic ameloblastoma. Int. J. Oral Maxillofac. Surg. 52 , 670–678 (2023). Qiu, X. et al. Reversed graph embedding resolves complex single-cell trajectories. Nat. Methods 14 , 979–982 (2017). McClary, A. C. et al. Ameloblastoma: a clinical review and trends in management. Eur. Arch. Otorhinolaryngol. 273 , 1649–1661 (2016). R, N. B. et al. Origin of Ameloblastoma From Basal Cells of the Oral Epithelium- Establishing the Relation Using Neuroectodermal Markers. J. Clin. Diagn. Res. 8 , ZC44 (2014). Adeel, M., Rajput, M. S. A., Arain, A. A., Baloch, M. & Khan, M. Ameloblastoma: Management and Outcome. Cureus (2018) doi:10.7759/cureus.3437. Zeisberg, M. & Neilson, E. G. Biomarkers for epithelial-mesenchymal transitions. J. Clin. Invest. 119 , 1429–1437 (2009). Vered, M. et al. Cancer-associated fibroblasts and epithelial-mesenchymal transition in metastatic oral tongue squamous cell carcinoma. Int. J. cancer 127 , 1356–1362 (2010). Groves, S. M. et al. Involvement of Epithelial–Mesenchymal Transition Genes in Small Cell Lung Cancer Phenotypic Plasticity. Cancers (Basel). 15 , 1477 (2023). Preca, B. T. et al. A novel ZEB1/HAS2 positive feedback loop promotes EMT in breast cancer. Oncotarget 8 , 11530–11543 (2017). Hong, T. et al. An Ovol2-Zeb1 Mutual Inhibitory Circuit Governs Bidirectional and Multi-step Transition between Epithelial and Mesenchymal States. PLoS Comput. Biol. 11 , (2015). Gregory, P. A. et al. An autocrine TGF-beta/ZEB/miR-200 signaling network regulates establishment and maintenance of epithelial-mesenchymal transition. Mol. Biol. Cell 22 , 1686–1698 (2011). Ruhin-Poncet, B. et al. Msx and dlx homeogene expression in epithelial odontogenic tumors. J. Histochem. Cytochem. 57 , 69–78 (2009). Scheau, C. et al. The Role of Matrix Metalloproteinases in the Epithelial-Mesenchymal Transition of Hepatocellular Carcinoma. Anal. Cell. Pathol. (Amst). 2019 , (2019). Arai, Y. & Lee, S. H. MMP13-Overexpressing Mesenchymal Stem Cells Enhance Bone Tissue Formation in the Presence of Collagen Hydrogel. Tissue Eng. Regen. Med. 20 , 461–471 (2023). Lau, S. L. & Samman, N. Recurrence related to treatment modalities of unicystic ameloblastoma: a systematic review. Int. J. Oral Maxillofac. Surg. 35 , 681–690 (2006). Nakamura, N., Higuchi, Y., Mitsuyasu, T., Sandra, F. & Ohishi, M. Comparison of long-term results between different approaches to ameloblastoma. Oral Surg. Oral Med. Oral Pathol. Oral Radiol. Endod. 93 , 13–20 (2002). Additional Declarations No competing interests reported. Supplementary Files supplementary.zip Table Legend Table S1: marker genes of each cluster. p_val and avg_log2FC were p values and fold change calculated by Seurat fnd_cluster function. Table S2: Differential expression of transcription factor regulons between cell subclusters. Table S3: Enrichment of transcription factor regulons in each wave of genes along ameloblastoma differentiation trajectory. P value and odds ratio are calculated by Fisher test. Table S4: Ligand-target association weights estimated by nichenetr. Table S5: adjacency of each transcription factors to transition state cell in spatial transcriptomics estimated by misty. Cite Share Download PDF Status: Posted 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-6903382","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":478643966,"identity":"8cc5aad6-45d5-46d6-afa5-68caefcc9f2a","order_by":0,"name":"Yueqi Shi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCElEQVRIiWNgGAWjYJACgw8/bOQYJBgYmEE8NgYGxgOEtBTO7EkzZpBghmoB6iGo5TMH2+HEBpgWBkJa5Gckb9zMwHM4ff7s/mOfCyrq8vjkmx8cYKixicbpjxtpxcYFFum5G+4cZp4948zhYjY2NoMDDMfSchtwaZHIMTOewWOdu0EimZmZt+1AYhsbg8EBxobDOLXIz8gx/83DxpwOdCFISx1QC/sHvFoYbuQYGPOwOScw3ABrYQZq4cFvi8GZZwWGwEA23HAj2ZiZB+yXnIIDCXj8It+evAEUlfLyMxIfM/MAQ0y++fjGBx9qbHA7TCDBAIWfgETiAPwHsGkZBaNgFIyCUYAEAJwdWixr0wGVAAAAAElFTkSuQmCC","orcid":"","institution":"Shanghai Jiao Tong University School of Medicine","correspondingAuthor":true,"prefix":"","firstName":"Yueqi","middleName":"","lastName":"Shi","suffix":""},{"id":478643967,"identity":"04873868-a81d-4055-acd8-91ebbbe189ee","order_by":1,"name":"Hao Lu","email":"","orcid":"","institution":"Shanghai Jiao Tong University School of Medicine, Shanghai Jiao Tong University, Shanghai Research Institute of Stomatology","correspondingAuthor":false,"prefix":"","firstName":"Hao","middleName":"","lastName":"Lu","suffix":""},{"id":478643968,"identity":"36b587d0-8139-4f1e-8396-0aeaa43588d2","order_by":2,"name":"Shaoyi Wang","email":"","orcid":"","institution":"Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai Jiao Tong University, Shanghai Research Institute of Stomatology","correspondingAuthor":false,"prefix":"","firstName":"Shaoyi","middleName":"","lastName":"Wang","suffix":""},{"id":478643969,"identity":"092f6651-023f-4395-bfd7-a2afbbf4034a","order_by":3,"name":"Weichen Song","email":"","orcid":"","institution":"Shanghai Jiao Tong University","correspondingAuthor":false,"prefix":"","firstName":"Weichen","middleName":"","lastName":"Song","suffix":""}],"badges":[],"createdAt":"2025-06-16 08:38:20","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6903382/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6903382/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":85749014,"identity":"685d02c8-7528-4938-a17b-13300107c8e2","added_by":"auto","created_at":"2025-07-01 09:52:26","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":387650,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eoverview of the single-cell and spatial transcriptome. \u003c/strong\u003eA: UMAP projection and clustering of single cell data. B: Karyotype of each cell estimated by CopyKAT. C: marker expression of each cell cluster. D: H-E staining of ameloblastoma slices used for spatial transcriptome. E: Cell type similarity spatial distribution. Cell type definition was the same as A. \u0026nbsp;F \u0026amp; G: Immunofluorescence staining of KRT15, VIM and FN1.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6903382/v1/b127edcfb9c4141273843e92.png"},{"id":85749025,"identity":"2f5f4099-b87a-4813-8c76-47563b26dd04","added_by":"auto","created_at":"2025-07-01 09:52:26","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":259163,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIntegration of embryonic tooth germ and ameloblastoma cell atlas. \u003c/strong\u003eA: epithelial and mesenchymal cell lineage used in the current study. See ref. for detail. B: Similarity between ameloblastoma cells and embryonic outer enamel epithelium (OEE) cell. Colors indicated similarity between each cell and OEE, as calculated by Seurat anchor analysis. C: Similarity between ameloblastoma cells and embryonic inner enamel epithelium (IEE) cell. Colors indicated similarity between each cell and IEE, as calculated by Seurat anchor analysis.D: Box plot summarizing similarity between each cell of stem-like, epi-like and ESC-like cluster and each of embryonic epithelial clusters. Dashed line represented median value. E: Box plot summarizing similarity between each cell of DCN- fibroblast, DCN+ fibroblast and FCN3- endothelium cluster and each of embryonic mesenchymal clusters. F: Similarity between each spatial spot and embryonic cell clusters.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6903382/v1/2c2974d4c8c0eb3db5f6dcf6.png"},{"id":85749008,"identity":"30b32551-06b0-42b5-8fe9-865117f9be8f","added_by":"auto","created_at":"2025-07-01 09:52:26","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":342359,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTranscription factor activity of ameloblastoma. \u003c/strong\u003eA: Cell type-specific transcription factor regulon expression level. Regulon: overall expression level of a transcription factor as well as all its predicted targets, as calculated by SCENIC. B: regulon expression of DLX5 and KDM5A regulon of each cell on UMAP projection. C: regulon expression comparison among cell clusters. Epi-like_aneu: aneuploid cells from epi-like cluster. Stem-like_aneu: aneuploid cells from stem-like cluster. Stem-like_dip: diploid cells from stem-like cluster. D: regulon expression level in spatial transcriptome data estimated by AUCell. E: TF-target network of highlighted regulon. Line color denoted regulation weight.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6903382/v1/ffaa37566ef3c8e8314aaec5.png"},{"id":85750105,"identity":"bf002895-8197-4120-854c-a15458d950a3","added_by":"auto","created_at":"2025-07-01 10:00:26","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":453656,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePseudotime analysis of epithelium-to-mesenchyme transition. \u003c/strong\u003eA: UMAP plot and pseudotime analysis of epithelial and mesenchymal cells. Arrows marked development direction. B: Expression of four waves of gene along epithelium-to-mesenchyme transition. C: Heatmap of module gene expression along pseudotime. Yellow bar denoted that the gene was predicted to be target of the corresponding transcription factor. Word cloud showed GO term enriched by each wave. Font size denoted enrichment odds ratio, color denoted enrichment p value. D: Expression of four waves of genes on spatial transcriptome estimated by AUCell.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6903382/v1/905ba8a5df592bdc5ef0ffd9.png"},{"id":85749029,"identity":"9f61a517-3fb1-46be-9c23-f337edaee227","added_by":"auto","created_at":"2025-07-01 09:52:26","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":279365,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSignaling networks of epithelium-to-mesenchyme transition. \u003c/strong\u003eA: left heatmap showed the average expression level of each ligand in each cell cluster. Right heatmap showed the regulation weight of each ligand (row) on each second wave gene (column), predicted by nichenetr. B: Spatial adjacency between each ligand (row) and each second wave gene (column), predicted by misty. C: Spatial expression level of highlighted ligands.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6903382/v1/e72fef0dd106e143e4dc3bb8.png"},{"id":104479226,"identity":"67c459c8-0d4f-41b2-94c5-5e8e0a3069f8","added_by":"auto","created_at":"2026-03-12 08:57:46","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2524890,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6903382/v1/7b6cc8ab-5c67-47a8-8126-18f9b0773c2b.pdf"},{"id":85750113,"identity":"6ea66871-e312-4c4d-83c3-b6b659b8f266","added_by":"auto","created_at":"2025-07-01 10:00:26","extension":"zip","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":471293,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTable Legend\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTable S1: marker genes of each cluster. p_val and avg_log2FC were p values and fold change calculated by Seurat fnd_cluster function.\u003c/p\u003e\n\u003cp\u003eTable S2: Differential expression of transcription factor regulons between cell subclusters.\u003c/p\u003e\n\u003cp\u003eTable S3: Enrichment of transcription factor regulons in each wave of genes along ameloblastoma differentiation trajectory. P value and odds ratio are calculated by Fisher test.\u003c/p\u003e\n\u003cp\u003eTable S4: Ligand-target association weights estimated by nichenetr.\u003c/p\u003e\n\u003cp\u003eTable S5: adjacency of each transcription factors to transition state cell in spatial transcriptomics estimated by misty.\u003c/p\u003e","description":"","filename":"supplementary.zip","url":"https://assets-eu.researchsquare.com/files/rs-6903382/v1/4de27f7a761a98e73c225596.zip"}],"financialInterests":"No competing interests reported.","formattedTitle":"A single-cell and spatial transcriptomic analysis of pre- and post- marsupialization conventional ameloblastoma","fulltext":[{"header":"Introductions","content":"\u003cp\u003eAmeloblastoma is the most frequent odontogenic tumor, with no significant gender predilection \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. In the fifth Edition of the World Health Organization (WHO) Classification of Head and Neck Tumors, ameloblastoma were classified as conventional, extraosseous/peripheral, unicystic, adenoid and metastasizing types\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Among them, conventional ameloblastoma is the most common type. Conventional ameloblastoma is defined as a benign epithelial odontogenic neoplasm characterized by a locally infiltrative growth pattern and a variably solid-cystic morphological architecture[3], and it has been a topic of significant interest among oral and maxillofacial researchers. It has a predilection for the mandible and exhibits a wide age range, albeit with a slight peak in the third and fourth decades. In terms of therapy, complete resection with a safe margin remains the gold standard, as conservative approaches often lead to high recurrence rates. However, this treatment can lead to several complications, including dentition and bone defects, lower lip numbness, and masticatory function disorders\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. Recently, marsupialization plus second-stage curettage was reported as an effective strategy in treating ameloblastoma[4]. Marsupialization, a technique for the treatment of cystic lesions of the jaws, was initially delineated by Partsch in the late 19th century\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. The approach entailed the externalization of the cystic lesion via the establishment of a surgical window in both the oral mucosa and the cystic wall. In addition, the advent of molecular-targeted therapies and immunotherapy has shown promise in recent studies, offering potential alternatives to the traditional surgical management of this challenging neoplasm\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe molecular underpinnings and origins of ameloblastoma have been the subject of extensive investigation, shedding light on the intricate interplay of genetic and epigenetic factors that contribute to its pathogenesis\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Mounting evidence highlights the role of aberrant signaling pathways, including the mitogen-activated protein kinase (MAPK) pathway, which has been implicated in the uncontrolled proliferation and survival of ameloblastoma cells\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Activating mutations in BRAF and RAS genes have been identified as key drivers of this dysregulated signaling cascade\u003csup\u003e\u003cspan additionalcitationids=\"CR7 CR8 CR9\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. Furthermore, recent advancements in genomic technologies have unveiled recurrent mutations in genes such as SMO and PTCH1, pinpointing the involvement of the Hedgehog signaling pathway in the tumorigenesis of this neoplasm\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. The concepts of ameloblastoma etiology are controversial. Earlier researches found it might be relevant to trauma, inflammation, and dental issues. Modern research connects its development to enamel organ remnants and odontogenic epithelium, with similarities found in tooth germ development, while other theories focus on the differentiation of pre-ameloblasts during tooth development and the role of the stratum intermedium in this process. However, despite these insights, our understanding of the precise cellular and molecular events that trigger the initial formation of ameloblastoma remains limited. Unraveling the early events in tumorigenesis, as well as the exact roles of various signaling pathways and their interactions in the progression of this neoplasm, are critical gaps in our current knowledge. Ongoing research endeavors are focused on elucidating these aspects of ameloblastoma, with the ultimate goal of identifying novel therapeutic targets and improving patient outcome.\u003c/p\u003e \u003cp\u003eSingle-cell RNA sequencing (scRNA) and spatial transcriptomics (ST) have emerged as transformative tools in the field of dental research, offering unique opportunities to address the current limitations in our understanding of the origins and progression of ameloblastoma\u003csup\u003e\u003cspan additionalcitationids=\"CR12 CR13 CR14\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. By enabling the profiling of individual cells within a tumor, scRNA-seq allows for the identification and characterization of distinct cell populations and states, thus providing a high-resolution view of the tumor heterogeneity and the underlying molecular mechanisms driving tumorigenesis. This approach has the potential to reveal novel cellular subtypes and molecular players involved in the initiation and development of ameloblastoma, facilitating the discovery of new therapeutic targets. Furthermore, spatial transcriptomics complements scRNA-seq by preserving the spatial information of gene expression within a tissue, which is crucial for understanding the complex interplay between various cell types, as well as the tumor microenvironment. This powerful technique can elucidate the spatial organization of ameloblastoma and its various histopathological subtypes, shedding light on the role of stromal and immune cells in tumor progression and therapy response. Together, single-cell RNA sequencing and spatial transcriptomics hold great promise in advancing our knowledge of ameloblastoma, offering new avenues for the development of targeted and personalized therapeutic strategies.\u003c/p\u003e \u003cp\u003eIn light with these advantages, we carried out scRNA and ST on pre- and post-marsupialization conventional ameloblastoma tissues followed by trajectory and network analysis. We mapped the conventional ameloblastoma trajectory to our recently established human embryonic tooth cell atlas\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e and compared between pre- and post-marsupialization tissues. These analyses could provide new insights to 1) the cellular and molecular compositions of conventional ameloblastoma as well as its spatial distribution; 2) the cellular origins of conventional ameloblastoma; and 3) the mechanism and hub genes of the transition from epithelial tumor cell to non-malignment mesenchymal cell, which might be the potential targets for further therapy.\u003c/p\u003e"},{"header":"Materials and Method","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n \u003ch2\u003eSample collection and preprocessing\u003c/h2\u003e\n \u003cp\u003eThis study was approved and supervised by Ethical committee of Shanghai Tongren Hospital (\u0026ldquo;human embryonic tooth development\u0026rdquo;, 2022-10-01, approval number: K2023013). Written informed consent was provided by all participants. This study was performed in accordance with the Declaration of Helsinki. The 3 patients were all male, aged between 41 and 52 years old, and consulted our department in 2022. Pre-operative CT scans suggested they had conventional ameloblastoma of the mandible. One of the patients had undergone marsupialization in our department in 2020. The marsupialization was performed under local anaesthesia. The opening site was made at the central part of the cystic lesions, and it was enlarged to prevent the premature closure induced by new bone formation. Impacted/offending tooth was extracted. The tooth socket was then enlarged and trimmed. All the interatrial septums were perforated to ensure effective drainage. Histological examination performed on the biopsy specimen obtained from the opening sites also suggested conventional ameloblastoma. An obturator was inserted after the surgery to prevent the premature closure of the opening site. Postoperative clinical and panoramic radiograph examinations showed the range of the lesion was gradually reduced. We then performed a second-stage ameloblastoma curettage for him 2 years after the marsupialization. The remaining 2 patients had not undergone any treatment, and we performed mandibular osteotomy on them. During the surgery, tumor tissues were collected and respectively placed in tissue preservation solution (Miltenyi, German) and optimum cutting temperature compound (Sakura, U.S.A). Post-operative pathological diagnosis revealed that all 3 patients had conventional ameloblastoma.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003escRNA and ST\u003c/h3\u003e\n\u003cp\u003eThe procedure of library construction, sequencing and quality control was similar to our previous work\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. We used NovaSeq 6000 platform to sequence libraries of scRNA and ST by paired-end 150 bp sequencing (PE150). We applied Seurat\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e NormalizeData function to obtain the library size-normalized count. Specifically, the global scale normalization method \u0026quot;LogNormalize\u0026quot; normalizes gene expression measurements per cell by total expression, multiplies by a scaling factor (10,000 by default), and transforms via the Seurat: SCTransform function. Cell phases are defined by Seurat: CellCycleScoring function based on list of cell cycle genes. We manually removed all ribosomal genes and mitochondrial genes. We retained non-doublet cells with number of detected genes\u0026thinsp;\u0026gt;\u0026thinsp;500 and \u0026lt;\u0026thinsp;2500, and mitochondria gene percentage\u0026thinsp;\u0026lt;\u0026thinsp;5%. Seurat preprocessed the spatial transcriptome data in a similar fashion.\u003c/p\u003e\n\u003ch3\u003eKaryotype analysis\u003c/h3\u003e\n\u003cp\u003eWe applied CopyKAT\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e on the filtered scRNA count data to define the karyotype of each cell. In brief, at least 5 genes per chromosome were required for each cell to calculate DNA copy numbers. Only genes that are expressed in LOW.DR\u0026thinsp;=\u0026thinsp;0.05 to UP.DR\u0026thinsp;=\u0026thinsp;0.2 fractions of cells were kept. For each segment, at least 25 genes were used to calculate copy number. Segmentation parameter was set at KS.cut\u0026thinsp;=\u0026thinsp;0.1.\u003c/p\u003e\n\u003ch3\u003eCluster analysis\u003c/h3\u003e\n\u003cp\u003eFor the scRNA data, we applied cluster analysis based on Louvain community detection implemented in Seurat: FindClusters function, principal component fraction\u0026thinsp;=\u0026thinsp;20, resolution\u0026thinsp;=\u0026thinsp;0.5. We applied the Seurat: FindMarkers Wilcoxon test to find the top 15 markers for each cluster and define them based on existing knowledge of epithelium and mesenchyme cell types. To integrate scRNA data with previous human embryonic tooth germ scRNA data, we applied Seurat anchored-based integration analysis with the default settings, including method\u0026thinsp;=\u0026thinsp;canonical correlation analysis, original dimension\u0026thinsp;=\u0026thinsp;PCA. Same settings were also used for ST data.\u003c/p\u003e\n\u003ch3\u003ePseudotime analysis\u003c/h3\u003e\n\u003cp\u003eWe extracted the continuous EMT cell clusters and applied monocle3\u003csup\u003e19\u003c/sup\u003e:learn_graph and order_cells function\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e to learn the pseudotime trajectory. We then applied monocle3: graph_test function on top 5,000 variable genes detected by Seurat: FindVariableGenes function to find all genes that significantly altered along the trajectory. Genes with FDR-adjusted p\u0026thinsp;\u0026lt;\u0026thinsp;0.01 were further used for module detection by monocle3: find_gene_modules function. We separately applied enrichment analysis and pathway analysis for each module.\u003c/p\u003e\n\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\n \u003ch2\u003eTranscription factor regulation network analysis\u003c/h2\u003e\n \u003cp\u003eWe applied SCENIC\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e to construct a transcription factor (TF)-target gene regulatory network targeting epithelial and mesenchymal cells, respectively. TF-motif annotations were downloaded from the SCENIC website, corresponding to hg38 refseq gene annotations, each gene has a 10kb window. We retained the genes of interest for the extended network of each TF and applied Fisher\u0026apos;s test to see if the modular genes detected by the pseudotime analysis were enriched in any of the TF regulatory networks. TFs with FDR-adjusted p\u0026thinsp;\u0026lt;\u0026thinsp;0.01 and odds ratios\u0026thinsp;\u0026gt;\u0026thinsp;2 were retained for each block. We further extracted the global expression of each TF and all its targets (so-called regulons) per cell to verify that this expression value had a similar expression pattern to the module genes.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003enichenetr analysis\u003c/h3\u003e\n\u003cp\u003eWe applied signaling network analysis using the nichenetr R package as previously described\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. Ligands were prioritized by area under the curve (AUC) and Pearson correlation coefficient calculated by nichenetr. Top 20 ligands ranked by AUC were chosen as highlighted ligands for further analysis. To focus on EMT procedure, we only analyzed the second wave genes.\u003c/p\u003e\n\u003ch3\u003eSpatial transcriptome analysis\u003c/h3\u003e\n\u003cp\u003eAfter defining cell clusters on scRNA data, we projected it to spatial transcriptome data by Seurat: FindTransferAnchors and TransferData functions to estimate the cell type probability of each spot. We assign cell types with the highest probability to each spot. We used misty R package\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e to estimate the spatial relationship among highlighted cell types, TFs and expression levels of ligands and target genes. We chose the \u0026ldquo;juxta\u0026rdquo; mode in misty analysis and trained a linear model to predict each of the target gene expression value by adjacent predicting values. Spatial adjacency was quantified by importance score in the prediction model. To focus on EMT procedure, we only analyzed the second wave genes.\u003c/p\u003e"},{"header":"Result","content":"\u003ch2\u003eCellular and spatial components of conventional ameloblastoma\u003c/h2\u003e\u003cp\u003eWe quantified the single-cell expression of 3,209 cells passing quality control (Method), which were collected from pre- and post-marsupialization conventional ameloblastoma tissues (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA and Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). By applying CopyKAT\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB and Figure S2), we confirmed that pre- marsupialization cells were mainly tumor-like aneuploid cell, which gradually transformed into diploid post-marsupialization cells via an intermediate state. This pattern was not biased by cell phase (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eB). We further applied cluster analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA\u0026amp;C, Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e) to dissect the heterogeneity of each karyotype. A total of 11 cell clusters were found. Aneuploid cells were mainly grouped into a cell cluster with characteristics of premature epithelium (epi-like), with high expression of KRT15 and ETV4 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA \u0026amp; C). We further validated that the protein expression of KRT15 was restricted to pre- marsupialization tissues by immunofluorescence (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF). The transition cells between aneuploid and diploid cells were grouped into one cluster characterized by MALAT1 and JUN expression, which we defined as stem-like cell cluster. Diploid cells were grouped into clusters with mesenchyme characteristics, in line with the important role of epithelium to mesenchyme transition (EMT) in conventional ameloblastoma\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. This fibroblast that expressed FN1, and COL1A2 (classified as DCN\u003csup\u003e+\u003c/sup\u003e and DCN\u003csup\u003e−\u003c/sup\u003e fibroblast), immature (FCN3\u003csup\u003e−\u003c/sup\u003e) and mature (FCN3\u003csup\u003e+\u003c/sup\u003e, PLVAP\u003csup\u003e+\u003c/sup\u003e) endothelium that expressed AQP1 and PLVAP. A small subset of SPARC + COL1A2 + SPP1 + cells was defined as SPP1 + osteoblast. Notably, a subset of diploid post- marsupialization cells expressed epithelium stem cell marker CD24 and KRT17 (ESC-like). These results confirmed previous histological observation that post-marsupialization tumor tissues underwent fibrogenesis, epitheliogenesis and osteogenesis procedure\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. Lastly, we defined M1 and M2 macrophage, T cells and neutrophils by their corresponding markers CD14, RGS1, CD3 and FCGR3B (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD and Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e)\u003c/p\u003e\u003cp\u003eWe then dissected slices of pre- and post-marsupialization conventional ameloblastoma (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD) for spatial transcriptome (ST) analysis. Applying Seurat anchor-based integration analysis\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE), we found that pre-marsupialization conventional ameloblastoma slices (lower slices) were predominantly consisted of epi-like cluster, whereas the intermediate stem-like cluster were only found sparsely at the edge of pre-marsupialization conventional ameloblastoma. In post-marsupialization slices (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE and Figure S3), fibroblast was the major component, and endothelium was located in complement to fibroblast and osteoblast, which suggested the existence of osteogenesis in the post-marsupialization tissue. We also analyzed the spatial expression of cell markers (Figure S4), and also supported these patterns. Taken together, these results depicted a general landscape of cellular and spatial component of pre- and post-marsupialization conventional ameloblastoma.\u003c/p\u003e\u003cp\u003eFurthermore, our cluster analysis and karyotype analysis suggested that aneuploid epithelial cells underwent EMT into diploid mesenchyme cells after marsupialization. Consistently, EMT markers (FN1, VIM, ACTA2, S100A4, COL1A1, TWIST1) were all highly expressed in the post-marsupialization cells (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). To further verify this result, we applied double immunofluorescence on EMT marker FN1 and VIM proteins, and confirmed that their expression was profoundly elevated in post-marsupialization tissue slice (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eG).\u003c/p\u003e\u003ch2\u003eConventional ameloblastoma resembled early outer enamel epithelium\u003c/h2\u003e\u003cp\u003eOne of the key questions of conventional ameloblastoma pathology is its cellular origin. By integrating conventional ameloblastoma single-cell data with our recently established human embryonic tooth germ cell atlas15 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA) with Seurat anchor integration algorithm, we analyzed the cellular origin of conventional ameloblastoma. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB and C, pre-marsupialization conventional ameloblastoma cells (epi-like cluster) were highly similar to the most primitive epithelial cell type (outer enamel epithelium, OEE) in embryonic tooth germ, whereas the transition state conventional ameloblastoma cells (stem-like cluster) were similar to the second primitive inner enamel epithelium (IEE). This suggested that conventional ameloblastoma might originate from premature OEE. Compared with epi-like cluster, stem-like cluster had more characteristics of differentiated dental epithelial cells of human embryo (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). On the other hand, post-marsupialization ESC-like cluster did not resemble any of the tooth germ cell types (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). As for other post-marsupialization cell clusters (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE), they resembled mature mesenchyme cells like endothelium and follicle cell, and did not show characteristics of mesenchyme stem cells or progenitor cells. This result supported that post-marsupialization mesenchyme cells were directly induced from tumor epithelial cells, not migrated from nearby connective tissues.\u003c/p\u003e\u003cp\u003eWe further integrated embryo tooth germ cell atlas with conventional ameloblastoma spatial transcriptome (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF and Figure S5). We found that pre-marsupialization slices were predominantly consisted of IEE-like cells, whereas cells similar to OEE and ALCAM + SC were distributed at the outer surface. We also found that at the opposite end of OEE there were regions enriched of ameloblast (Figure S5). The stratum intermedium (SI) lineage and odontoblast (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA) was sparse in conventional ameloblastoma slice (Figure S5).\u003c/p\u003e\u003ch2\u003eVital role of DLX5 and TWIST1 in epithelium-to-mesenchyme transition\u003c/h2\u003e\u003cp\u003eWe have found that epi-like tumor cell in pre-marsupialization conventional ameloblastoma underwent EMT after marsupialization and gained characteristics of mature mesenchyme cells. We further ask what is the key molecular events during this procedure, especially in the transition state stem-like cell cluster. We apply SCENIC to find active transcription factors (TF) as well as their target genes. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA and B, by combining expression levels of TF and their targets (“regulon”), we found high cell type specificity. Epi-like cluster showed high expression of regulons like DLX5, ETV5, TAF7 and ATF1, whereas post-marsupialization endothelium and fibroblast-like clusters were characterized by regulons of TWIST1, KLT2, MYC and ERG (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA and B). The transition state stem-like cluster showed weaken expression of both of these regulons, but also exhibited unique expression of YY1, REST and KDM5A (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA and B). We proposed that TF activity transition in stem-like cells might be the key regulator of EMT.\u003c/p\u003e\u003cp\u003eTo further explore this mechanism, we compared regulon activity between epi-like cell, stem-like aneuploid cell and stem-like diploid cell by t test (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC and Table S2). We found that most key TF showed monotonous alteration during this transition. For example, DLX5 and GTF2B had the highest expression in epi-like cluster, lower expression in stem-like aneuploid cell and the lowest expression in stem-like diploid cell, and the difference between groups were all significant (p \u0026lt; 10\u003csup\u003e− 8\u003c/sup\u003e). On the other hand, expression of KDM5A and TWIST1 regulon were the lowest in epi-like cell and were highest in stem-like diploid cell. These results suggested that the activation of KDM5A and TWIST1 and the deactivation of DLX5 and GTF2B might play an important role in EMT and the loss of tumor characteristics of conventional ameloblastoma. Other TFs showing significant alteration during process included ETV5, ELK4, TAF7, GATA2, etc. (Table S2).\u003c/p\u003e\u003cp\u003eWe further explored the biological significance of these key TFs. Applying AUCell\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e on spatial transcriptome data (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD), we found that DLX5 regulon were highly expressed in the entire slice of pre-marsupialization conventional ameloblastoma, without significant region specificity. TWIST1 regulon expression in post-marsupialization slice was largely overlapped with fibroblast distribution (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). These key TF in EMT were predicted to regulate several genes implicated in ameloblast pathology, including epithelial stem cell marker KRT15, COL1A1 and KRAS, etc.\u003c/p\u003e\u003ch2\u003ePseudotime analysis of epithelium-to-mesenchyme transition\u003c/h2\u003e\u003cp\u003eWe next applied pseudotime analysis\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e to reconstruct the trajectory from pre-marsupialization epithelial cells to post-marsupialization mesenchyme cells. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA, pseudotime analysis verified that there was a continuous monotonous trajectory on EMT, supported the theory that post-marsupialization mesenchyme cells were transferred from pre-marsupialization epithelial cells. Applying gene module analysis from monocle3\u003csup\u003e19\u003c/sup\u003e, we identified four clusters of genes that emerged at different time points, hereafter referred to as four waves of genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). The first wave (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC) was highly expressed in epithelial cells and gradually decreased along pseudotime, including DUSP1, ID3, MSX2 and DLX5. DLX5 and MSX2, two transcription factors, were also predicted to regulate larger-than-expected proportion of first wave genes (Fisher p \u0026lt; 10\u003csup\u003e− 5\u003c/sup\u003e, Table S3). First wave genes were also involved in biological processes related to extracellular matrix organization, epithelial cell proliferation, and transform growth factor beta, MAPK, ERK signaling pathways (gene ontology analysis [GO] FDR \u0026lt; 0.05). In spatial transcriptome (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD), first wave genes were widely expressed in pre-marsupialization slices, covering all structures of conventional ameloblastoma.\u003c/p\u003e\u003cp\u003eThe second wave genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC) peaked at transition state, including NFIX, FOXP1, MALAT1 and HMGB1, etc. They were enriched in predicted targets of KDM5B and NFIA (Table S3), and were enriched in biological processes of RNA splicing, chromatin organization and Wnt signaling pathway, suggesting that these pathways might be the key process for conventional ameloblastoma to lose tumor characteristics. The spatial distribution of second wave genes is similar to the first wave, albeit with a lower expression value (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD).\u003c/p\u003e\u003cp\u003eThe third wave genes involved a small number of genes including KRT5 and KRT 16. They showed little TF specificity and were only enriched in a few pathways related to T cell functions. The spatial expression (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD) is also sparse, with limited expression at the border of pre-marsupialization slices.\u003c/p\u003e\u003cp\u003eLastly, the fourth wave of genes involved EGR1, VCAM1, CLDN5 other genes related to mesenchyme cell functions (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). They enriched in targets of TWIST1 and RORB (Table S3), and enriched in zinc ion homeostasis, cytoskeleton organization and intermediate filament (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). The spatial expression of fourth wave genes were similar to TWIST1 regulon (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD), suggesting that TWIST1 might be one of the key drivers of this wave of gene expression.\u003c/p\u003e\u003ch2\u003ePositive feedback regulation of EMT via MMP13 and THBS2\u003c/h2\u003e\u003cp\u003eLastly, we explored the upstream signaling pathways that regulated EMT in conventional ameloblastoma. We first applied nichenetr on the transition-related genes (second wave genes in monocle analysis). As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, transition-related genes were densely regulated by ligands from ESC-like cell clusters, including MMP13, THBS2 and COL4A1 (Table S4). Among them, MMP13 showed the strongest regulation activity with the largest number (19) of predicted targets, including high-confidence target gene like NFIB, SLC38A2, JUN, ZFP36L1 and HSP90AB1 (nichenetr regulation weight \u0026gt; 0.75), and were exclusively expressed in ESC-like cell clusters. In fact, all prioritized ligands were found to expressed in ESC-like cell clusters (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). Since ESC-like cell cluster is a post-EMT cluster (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e), we proposed that cells completed EMT had positive feedback regulation to further promote EMT.\u003c/p\u003e\u003cp\u003eWe further applied misty on spatial transcriptome data to analyze the spatial adjacency of these ligands and targets, using misty juxta mode. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB in Table S5, ligands with the most significant adjacency with transition-related genes included THBS2, SERPING1, COL4A1 and LAMB2, indicating that their spatial expression were close to transition-related genes. For example, THBS2 were found to be enriched around MALAT1, FOXP1 and HSP90AB1, three genes that were also predicted to be targets of THBS2 by nichenetr. These concordant results supported the regulatory role of THBS2 on these transition-related genes. The spatial expression of many of these highlighted ligands, including YHBS2, SERPING1 and LAMB2, were enriched in post-marsupialization tissues and were enriched in mesenchyme tissues (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC), further supporting our conclusion that EMT was promoted by signals from post-EMT cells, which formed a positive feedback loop.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we applied scRNA and ST on conventional ameloblastoma tissues to construct a spatial cell atlas, providing new clues on its underlying mechanism. We managed to find the cellular origin of conventional ameloblastoma by mapping it to the human embryonic dental scRNA reference, and deciphered the transition procedures from tumor epithelial tissues to mesenchymal tissues after marsupialization by trajectory and network analysis.\u003c/p\u003e \u003cp\u003eThe origin of ameloblastoma remains a subject of debate. It is hypothesized to arise from several sources: (1) cells of the enamel organ rest, (2) cells from Hertwig's epithelial root sheath or the epithelial cell rest of Malassez, (3) the epithelial lining of an odontogenic cyst, especially a dentigerous cyst, and (4) basal cells of the oral mucosa. Nevertheless, concrete evidence supporting each of these hypotheses is currently absent\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Through the integration of conventional ameloblastoma and fetal tooth germ cell atlas data, our findings reveal that conventional ameloblastoma tumor cells predominantly exhibit molecular signatures akin to primitive OEE cells. This observation suggests that conventional ameloblastoma likely arises from an early stage of dental epithelial development. The cellular origin of conventional ameloblastoma can thus be traced to these primitive OEE cells, which possess the ability to self-renew and differentiate into various cell types contributing to the diverse histopathological subtypes observed in the tumor\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. This insight into the developmental trajectory of conventional ameloblastoma not only furthers our understanding of its pathogenesis but also opens new avenues for targeted therapeutic interventions that may exploit the vulnerabilities inherent in these stem cell-like populations\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Additionally, the identification of key regulators and signaling pathways specific to the primitive dental epithelial stem cells could provide valuable information for the development of novel strategies to prevent or halt the initiation and progression of ameloblastoma\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe epithelial\u0026ndash;mesenchymal transition (EMT) is a crucial event during cell development. In this process, epithelial cells undergo a transformation, adopting mesenchymal fibroblast-like features characterized by decreased intercellular adhesion and enhanced motility. This transition can be categorized into three distinct types, based on the specific biological environment: Types 1, 2, and 3. Type 1 EMT is related with processes such as implantation, embryogenesis, and organogenesis. These events necessitate rapid cellular differentiation into various phenotypes and cell migration within a constrained temporal window. Type 2 EMT plays a role in wound healing and tissue regeneration, leading to the formation of repair-associated mesenchymal cells or fibroblasts. Type 3 EMT plays a pivotal role in tumor progression and metastasis by facilitating the generation, growth, and dissemination of cancerous cells\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. We found that marsupialization of conventional ameloblastoma induced the occurrence of type 2 EMT, in which aneuploid tumor epithelial cells transformed into euploid mesenchymal cells, including fibroblasts, vascular endothelial cells, and osteoblasts. These cells are involved in the tissue repair of the jawbone. Our result showed the high expression of several type 2 EMT markers (FN1, VIM, ACTA2, S100A4, COL1A1, TWIST1) in post-marsupialization cells. Among them, TWIST1 has been shown to have an important role in cell plasticity in lung cancers\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e, whereas the type-2 EMT is one form of cell plasticity. Furthermore, the feed-back regulation from post-EMT mesenchyme were also found by previous studies of other tumors\u003csup\u003e\u003cspan additionalcitationids=\"CR34\" citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. Double immunofluorescence showed an increase in expression of mesenchymal markers fibronectin and vimentin after marsupialization of conventional ameloblastoma, which also confirmed this finding.\u003c/p\u003e \u003cp\u003eThrough an in-depth network analysis of the epithelial-to-mesenchymal transition (EMT) process, we identified DLX5 and MSX2 as key transcription factors of the first Wave of EMT genes. Both DLX5 and MSX2 are homeobox genes. Previous study has shown that they have an important role in tooth morphogenesis and development, and that mutations in homeobox genes cause developmental disorders such as odontogenic lesions\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. In our analysis, they both showed high expression in pre-marsupialization conventional ameloblastoma cells and were decreased during EMT. Furthermore, our signaling network analysis revealed that MMP13 could regulate the downstream expression of various EMT-related genes, making it a potential hub in the EMT procedure of conventional ameloblastoma. Matrix metalloproteinases (MMPs) are crucial for various tissue functions, including cellular regeneration, programmed death, angiogenesis, and other vital processes. These functions are integral to both normal development and pathological events, such as the EMT\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. MMP13 is not only an essential factor for bone development and bone healing but also has a pivotal role in promoting osteogenic differentiation of mesenchymal stem cells (MSCs) to induce reparative jawbone tissue formation. It was reported that the osteogenic marker genes, ALP and RUNX2, were upregulated in MMP13-overexpressing MSCs through p38 phosphorylation\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. These results revealed novel potential therapeutic targets in conventional ameloblastoma.\u003c/p\u003e \u003cp\u003eMarsupialization has traditionally been the primary treatment for odontogenic cystic lesions, especially large ones in the mandible. Its advantages are new bone formation in the cystic cavity, preservation of oral tissues, pulp vitality maintenance, fewer dental extractions, and protection of crucial anatomical structures. Given this, some clinicians have explored marsupialization for ameloblastoma\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. It was reported that marsupialization resulted with 18% recurrence rate in unicystic ameloblastoma\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. However, the rate exceeded 50% in multicystic ameloblastoma\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e, which potentially due to periosteum damage during curettage or incomplete marsupialization because of the exist of bony septae in multilocular lesions. Our research offers a potential solution for cases where marsupialization may be unsuitable or ineffective.\u003c/p\u003e \u003cp\u003eOur study presents certain limitations. Non-conventional subtypes of ameloblastoma are relatively rare, and our investigation solely encompassed the conventional type; therefore, our conclusions may not be universally applicable to all patients. Additionally, the sample size is relatively small, which may be subject to individual variation. Future large-scale studies covering a broader range of subtypes and timepoints are warranted to elucidate the comprehensive landscape of ameloblastoma.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, we have deciphered the cellular and spatial composition of conventional ameloblastoma, revealing its primary origin from early-stage epithelial stem cells. We have also underscored the role of epithelial-to-mesenchymal transition (EMT) in conventional ameloblastoma and identified genes such as DLX5 and MMP13 as key players in this process, rendering them promising therapeutic targets. Continued exploration of the molecular mechanisms governing ameloblastoma and the development of targeted therapies have the potential to significantly improve patient outcomes and advance the management of this challenging neoplasm.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have not use AI-generated work in this manuscript\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by National Science Foundation of China (82401759), Research Fund of Shanghai Tongren Hospital, Shanghai Jiaotong University School of Medicine (NO: TRYJ2021JC14, TRKYRC -xx202207).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and material\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003escRNA and ST data could be obtained from https://data.mendeley.com/datasets/4mdwc4h8rb/1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved and supervised by Ethical committee of Shanghai Tongren Hospital (“human embryonic tooth development”, 2022-10-01, approval number: K2023013). Written informed consent was provided by all participants.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll the authors declared no conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor’s contribution\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eY.S designed and conduct the study, H.L collected and preprocessed the sample and data, S.W interpreted the data, W.S and W.W supervised the data, Y.S drafted the manuscript, all authors read, revised and approved the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eJohnson, N. R., Gannon, O. M., Savage, N. W. \u0026amp; Batstone, M. D. Frequency of odontogenic cysts and tumors: a systematic review. \u003cem\u003eJ. Investig. Clin. Dent.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 9\u0026ndash;14 (2014).\u003c/li\u003e\n \u003cli\u003eVered, M. \u0026amp; Wright, J. M. Update from the 5th Edition of the World Health Organization Classification of Head and Neck Tumors: Odontogenic and Maxillofacial Bone Tumours. \u003cem\u003eHead Neck Pathol.\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 63\u0026ndash;75 (2022).\u003c/li\u003e\n \u003cli\u003eYang, Z. \u003cem\u003eet al.\u003c/em\u003e Marsupialization of mandibular cystic ameloblastoma: Retrospective study of 7 years. \u003cem\u003eHead Neck\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 2172\u0026ndash;2180 (2018).\u003c/li\u003e\n \u003cli\u003eKolari, V., Dhabaria, H., Sengupta, S., Sait, A. I. \u0026amp; Shah, A. Use of Marsupialisation for a Conservative Approach to Huge Cystic Lesions of the Jaws - A Report of three Cases. \u003cem\u003eAnn. Maxillofac. Surg.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 244\u0026ndash;247 (2022).\u003c/li\u003e\n \u003cli\u003eEffiom, O. A., Ogundana, O. M., Akinshipo, A. O. \u0026amp; Akintoye, S. O. Ameloblastoma: current etiopathological concepts and management. \u003cem\u003eOral Dis.\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 307\u0026ndash;316 (2018).\u003c/li\u003e\n \u003cli\u003eSweeney, R. T. \u003cem\u003eet al.\u003c/em\u003e Identification of recurrent SMO and BRAF mutations in ameloblastomas. \u003cem\u003eNat. Genet.\u003c/em\u003e \u003cstrong\u003e46\u003c/strong\u003e, 722\u0026ndash;5 (2014).\u003c/li\u003e\n \u003cli\u003eBrown, N. A. \u003cem\u003eet al.\u003c/em\u003e Activating FGFR2-RAS-BRAF mutations in ameloblastoma. \u003cem\u003eClin. Cancer Res.\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 5517\u0026ndash;5526 (2014).\u003c/li\u003e\n \u003cli\u003eKurppa, K. J. \u003cem\u003eet al.\u003c/em\u003e High frequency of BRAFV600E mutations in ameloblastoma. \u003cem\u003eJ. Pathol.\u003c/em\u003e \u003cstrong\u003e232\u003c/strong\u003e, 492\u0026ndash;498 (2014).\u003c/li\u003e\n \u003cli\u003eZhang, H. L. \u003cem\u003eet al.\u003c/em\u003e Comparative analysis of cellular expression pattern of schizophrenia risk genes in human versus mouse cortex. \u003cem\u003eCell Biosci.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, (2019).\u003c/li\u003e\n \u003cli\u003eShi, Y., Li, M., Yu, Y., Zhou, Y. \u0026amp; Wang, S. Whole exome sequencing and system biology analysis support the \u0026lsquo;two-hit\u0026rsquo; mechanism in the onset of Ameloblastoma. \u003cem\u003eMed. Oral Patol. Oral Cir. Bucal\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, e510 (2021).\u003c/li\u003e\n \u003cli\u003eLee, S. \u003cem\u003eet al.\u003c/em\u003e Single-Cell RNA Sequencing Analysis of Human Dental Pulp Stem Cell and Human Periodontal Ligament Stem Cell. \u003cem\u003eJ. Endod.\u003c/em\u003e \u003cstrong\u003e48\u003c/strong\u003e, 240\u0026ndash;248 (2022).\u003c/li\u003e\n \u003cli\u003eRen, H., Wen, Q., Zhao, Q., Wang, N. \u0026amp; Zhao, Y. Atlas of human dental pulp cells at multiple spatial and temporal levels based on single-cell sequencing analysis. \u003cem\u003eFront. Physiol.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, (2022).\u003c/li\u003e\n \u003cli\u003eHermans, F. \u003cem\u003eet al.\u003c/em\u003e Establishment of inclusive single-cell transcriptome atlases from mouse and human tooth as powerful resource for dental research. \u003cem\u003eFront. Cell Dev. Biol.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 1990 (2022).\u003c/li\u003e\n \u003cli\u003ePagella, P., de Vargas Roditi, L., Stadlinger, B., Moor, A. E. \u0026amp; Mitsiadis, T. A. A single-cell atlas of human teeth. \u003cem\u003eiScience\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 102405 (2021).\u003c/li\u003e\n \u003cli\u003eBelkina, A. C. \u003cem\u003eet al.\u003c/em\u003e Single-Cell Analysis of the Periodontal Immune Niche in Type 2 Diabetes. \u003cem\u003eJ. Dent. Res.\u003c/em\u003e \u003cstrong\u003e99\u003c/strong\u003e, 855\u0026ndash;862 (2020).\u003c/li\u003e\n \u003cli\u003eShi, Y. \u003cem\u003eet al.\u003c/em\u003e Spatiotemporal cell landscape of human embryonic tooth development. \u003cem\u003eCell Prolif.\u003c/em\u003e 2023.03.01.530693 (2024) doi:10.1111/cpr.13653.\u003c/li\u003e\n \u003cli\u003eStuart, T. \u003cem\u003eet al.\u003c/em\u003e Comprehensive Integration of Single-Cell Data. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e177\u003c/strong\u003e, 1888-1902.e21 (2019).\u003c/li\u003e\n \u003cli\u003eGao, R. \u003cem\u003eet al.\u003c/em\u003e Delineating copy number and clonal substructure in human tumors from single-cell transcriptomes. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, 599\u0026ndash;608 (2021).\u003c/li\u003e\n \u003cli\u003eCao, J. \u003cem\u003eet al.\u003c/em\u003e 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 \u003cli\u003eAibar, S. \u003cem\u003eet al.\u003c/em\u003e SCENIC: single-cell regulatory network inference and clustering. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 1083\u0026ndash;1086 (2017).\u003c/li\u003e\n \u003cli\u003eShi, Y. \u003cem\u003eet al.\u003c/em\u003e A single-cell interactome of human tooth germ from growing third molar elucidates signaling networks regulating dental development. \u003cem\u003eCell Biosci.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 1\u0026ndash;17 (2021).\u003c/li\u003e\n \u003cli\u003eTanevski, J., Flores, R. O. R., Gabor, A., Schapiro, D. \u0026amp; Saez-Rodriguez, J. Explainable multiview framework for dissecting spatial relationships from highly multiplexed data. \u003cem\u003eGenome Biol.\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 1\u0026ndash;31 (2022).\u003c/li\u003e\n \u003cli\u003eSiar, C. H. \u0026amp; Ng, K. H. Epithelial-to-mesenchymal transition in ameloblastoma: focus on morphologically evident mesenchymal phenotypic transition. \u003cem\u003ePathology\u003c/em\u003e \u003cstrong\u003e51\u003c/strong\u003e, 494\u0026ndash;501 (2019).\u003c/li\u003e\n \u003cli\u003eNakamura, N., Higuchi, Y., Tashiro, H. \u0026amp; Ohishi, M. Marsupialization of cystic ameloblastoma: a clinical and histopathologic study of the growth characteristics before and after marsupialization. \u003cem\u003eJ. Oral Maxillofac. Surg.\u003c/em\u003e \u003cstrong\u003e53\u003c/strong\u003e, 748\u0026ndash;754 (1995).\u003c/li\u003e\n \u003cli\u003eMohamed, A. A. S. \u003cem\u003eet al.\u003c/em\u003e Volumetric change of bony cavity and shrinkage speed after marsupialization for odontogenic keratocyst and unicystic ameloblastoma. \u003cem\u003eInt. J. Oral Maxillofac. Surg.\u003c/em\u003e \u003cstrong\u003e52\u003c/strong\u003e, 670\u0026ndash;678 (2023).\u003c/li\u003e\n \u003cli\u003eQiu, X. \u003cem\u003eet al.\u003c/em\u003e Reversed graph embedding resolves complex single-cell trajectories. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 979\u0026ndash;982 (2017).\u003c/li\u003e\n \u003cli\u003eMcClary, A. C. \u003cem\u003eet al.\u003c/em\u003e Ameloblastoma: a clinical review and trends in management. \u003cem\u003eEur. Arch. Otorhinolaryngol.\u003c/em\u003e \u003cstrong\u003e273\u003c/strong\u003e, 1649\u0026ndash;1661 (2016).\u003c/li\u003e\n \u003cli\u003eR, N. B. \u003cem\u003eet al.\u003c/em\u003e Origin of Ameloblastoma From Basal Cells of the Oral Epithelium- Establishing the Relation Using Neuroectodermal Markers. \u003cem\u003eJ. Clin. Diagn. Res.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, ZC44 (2014).\u003c/li\u003e\n \u003cli\u003eAdeel, M., Rajput, M. S. A., Arain, A. A., Baloch, M. \u0026amp; Khan, M. Ameloblastoma: Management and Outcome. \u003cem\u003eCureus\u003c/em\u003e (2018) doi:10.7759/cureus.3437.\u003c/li\u003e\n \u003cli\u003eZeisberg, M. \u0026amp; Neilson, E. G. Biomarkers for epithelial-mesenchymal transitions. \u003cem\u003eJ. Clin. Invest.\u003c/em\u003e \u003cstrong\u003e119\u003c/strong\u003e, 1429\u0026ndash;1437 (2009).\u003c/li\u003e\n \u003cli\u003eVered, M. \u003cem\u003eet al.\u003c/em\u003e Cancer-associated fibroblasts and epithelial-mesenchymal transition in metastatic oral tongue squamous cell carcinoma. \u003cem\u003eInt. J. cancer\u003c/em\u003e \u003cstrong\u003e127\u003c/strong\u003e, 1356\u0026ndash;1362 (2010).\u003c/li\u003e\n \u003cli\u003eGroves, S. M. \u003cem\u003eet al.\u003c/em\u003e Involvement of Epithelial\u0026ndash;Mesenchymal Transition Genes in Small Cell Lung Cancer Phenotypic Plasticity. \u003cem\u003eCancers (Basel).\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1477 (2023).\u003c/li\u003e\n \u003cli\u003ePreca, B. T. \u003cem\u003eet al.\u003c/em\u003e A novel ZEB1/HAS2 positive feedback loop promotes EMT in breast cancer. \u003cem\u003eOncotarget\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 11530\u0026ndash;11543 (2017).\u003c/li\u003e\n \u003cli\u003eHong, T. \u003cem\u003eet al.\u003c/em\u003e An Ovol2-Zeb1 Mutual Inhibitory Circuit Governs Bidirectional and Multi-step Transition between Epithelial and Mesenchymal States. \u003cem\u003ePLoS Comput. Biol.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, (2015).\u003c/li\u003e\n \u003cli\u003eGregory, P. A. \u003cem\u003eet al.\u003c/em\u003e An autocrine TGF-beta/ZEB/miR-200 signaling network regulates establishment and maintenance of epithelial-mesenchymal transition. \u003cem\u003eMol. Biol. Cell\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 1686\u0026ndash;1698 (2011).\u003c/li\u003e\n \u003cli\u003eRuhin-Poncet, B. \u003cem\u003eet al.\u003c/em\u003e Msx and dlx homeogene expression in epithelial odontogenic tumors. \u003cem\u003eJ. Histochem. Cytochem.\u003c/em\u003e \u003cstrong\u003e57\u003c/strong\u003e, 69\u0026ndash;78 (2009).\u003c/li\u003e\n \u003cli\u003eScheau, C. \u003cem\u003eet al.\u003c/em\u003e The Role of Matrix Metalloproteinases in the Epithelial-Mesenchymal Transition of Hepatocellular Carcinoma. \u003cem\u003eAnal. Cell. Pathol. (Amst).\u003c/em\u003e \u003cstrong\u003e2019\u003c/strong\u003e, (2019).\u003c/li\u003e\n \u003cli\u003eArai, Y. \u0026amp; Lee, S. H. MMP13-Overexpressing Mesenchymal Stem Cells Enhance Bone Tissue Formation in the Presence of Collagen Hydrogel. \u003cem\u003eTissue Eng. Regen. Med.\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 461\u0026ndash;471 (2023).\u003c/li\u003e\n \u003cli\u003eLau, S. L. \u0026amp; Samman, N. Recurrence related to treatment modalities of unicystic ameloblastoma: a systematic review. \u003cem\u003eInt. J. Oral Maxillofac. Surg.\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, 681\u0026ndash;690 (2006).\u003c/li\u003e\n \u003cli\u003eNakamura, N., Higuchi, Y., Mitsuyasu, T., Sandra, F. \u0026amp; Ohishi, M. Comparison of long-term results between different approaches to ameloblastoma. \u003cem\u003eOral Surg. Oral Med. Oral Pathol. Oral Radiol. Endod.\u003c/em\u003e \u003cstrong\u003e93\u003c/strong\u003e, 13\u0026ndash;20 (2002).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6903382/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6903382/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAmeloblastoma (AM) is the most frequent odontogenic tumor with local invasiveness, and conventional ameloblastoma is the most common type of it. This study aimed to gain an unbiased overview of the cellular and spatial component of conventional ameloblastoma. We applied single-cell RNA sequencing and spatial transcriptomics on pre- and post-marsupialization conventional ameloblastoma slices, followed by cluster, trajectory, network analysis and immunofluorescence validation. We found a continuous cell trajectory from pre-marsupialization epithelium-like cell types to post- marsupialization mesenchyme-like cell types. A comparative analysis with our recently established human embryonic dental cell atlas revealed that epithelium-like cells resembled the most primitive outer enamel progenitors, whereas mesenchyme-like cells did not resemble any embryonic mesenchymal cells, representing a differentiated and abnormal state. Motif enrichment analysis and pseudotime analysis found that transcription factors DLX5 and TWIST1, together with their target genes, showed profound alterations paralleled the epithelium-to-mesenchyme transition (EMT). We verified the existence of EMT by double immunofluorescence of two EMT markers FN1 and VIM, and verified the specific expression of KRT15 in pre-marsupialization tissues. Lastly, signaling network analysis found that mesenchyme-derived proteins MMP13 and THBS2 exert feedback regulatory effects on EMT. Our study constructed the cellular and spatial landscape of conventional ameloblastoma, and highlighted hub genes like DLX5 and MMP13 that could serve as potential therapeutic targets.\u003c/p\u003e","manuscriptTitle":"A single-cell and spatial transcriptomic analysis of pre- and post- marsupialization conventional ameloblastoma","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-01 09:52:21","doi":"10.21203/rs.3.rs-6903382/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"23aa157b-4eb0-4306-82e9-40e7b4f72aa9","owner":[],"postedDate":"July 1st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":50821722,"name":"Biological sciences/Cancer/Oral cancer"},{"id":50821723,"name":"Biological sciences/Cell biology"}],"tags":[],"updatedAt":"2026-03-12T08:57:27+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-01 09:52:21","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6903382","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6903382","identity":"rs-6903382","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}