Spatial transcriptomics reveals coordinated epithelial invasion and stromal remodeling in adenomyosis

Spatial transcriptomics reveals coordinated epithelial invasion and stromal remodeling in adenomyosis | 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 Help Center Sign In Submit a Preprint Cite Share Download PDF Article Spatial transcriptomics reveals coordinated epithelial invasion and stromal remodeling in adenomyosis Wenli Gu, Dan Yang, Lulu Shi, Hui Dong, Caihong Yang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9819300/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Background: Adenomyosis is characterized by ectopic endometrial tissue invading the myometrium, yet the spatiotemporal dynamics and molecular programs governing epithelial invasion and subsequent tissue remodeling remain poorly understood. Methods: High-resolution spatial transcriptomic profiling (Stereo-seq) was performed on three pathologically confirmed human adenomyosis specimens. Spatial data were integrated with publicly available single-cell RNA sequencing datasets. Cell type composition was inferred using Cell2location, while differential gene expression and cell–cell communication networks were analyzed to characterize transcriptional programs and signaling pathways at the invasive front. Results: Spatial transcriptomic mapping showed region-associated changes in epithelial adhesion programs during adenomyotic invasion, with ciliated and non-ciliated epithelial cells displaying enrichment of focal adhesion and cell-junction pathways in invasive regions. Cell-cell communication analysis further identified CXCL12-CXCR4 chemokine signaling and PDGF/IGF-related growth factor signaling as spatially associated interaction networks involving epithelial, immune, and stromal cell populations. These findings suggest coordinated epithelial-stromal crosstalk at the invasive front, while functional validation is required to establish causal roles for these pathways. Conclusions: This study identifies the CXCL12–CXCR4 and PDGF/IGF signaling axes as central spatial regulators of epithelial invasion and stromal remodeling in adenomyosis. By providing a spatiotemporal framework for epithelial–stromal crosstalk, these findings advance mechanistic understanding of adenomyosis pathogenesis and offer molecular insights that may inform emerging disease models and targeted therapeutic strategies, including recently identified prolactin receptor–based interventions. Biological sciences/Cell biology Biological sciences/Computational biology and bioinformatics Health sciences/Diseases Biological sciences/Molecular biology Adenomyosis spatial transcriptomics epithelial invasion CXCL12–CXCR4 axis PDGF signaling prolactin receptor cell–cell communication Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1 Introduction Adenomyosis is a common gynecological disorder characterized by the presence of endometrial glands and stroma within the myometrium, frequently accompanied by smooth muscle hyperplasia, fibrosis, and chronic inflammation( 1 – 4 ). Clinically, adenomyosis is strongly associated with dysmenorrhea, abnormal uterine bleeding, infertility, and a substantial reduction in quality of life for affected women( 1 , 2 , 4 – 6 ). Despite its high prevalence and clinical burden, the cellular and molecular mechanisms that drive lesion initiation, invasion, and progressive myometrial remodeling remain incompletely understood( 3 – 7 ). Several pathogenic hypotheses have been proposed to explain adenomyosis development, including invagination of the endometrial–myometrial junction following tissue microinjury, hormonal dysregulation, aberrant wound-healing responses, and stem or progenitor cell involvement( 7 , 8 ). While these models provide partial explanations, they largely fail to capture adenomyosis as a dynamic, multistep process( 3 , 5 , 7 ). In particular, key mechanistic questions remain unresolved: how endometrial epithelial cells acquire invasive capacity, how they migrate directionally into the myometrium, and how their presence subsequently induces the characteristic hypertrophy and fibrosis of the surrounding smooth muscle tissue( 7 , 9 ). Recent advances in single-cell RNA sequencing (scRNA-seq) have substantially expanded our understanding of cellular heterogeneity in adenomyosis, revealing distinct epithelial, stromal, immune, and vascular subpopulations with altered transcriptional programs( 9 – 13 ). However, by necessity, scRNA-seq disrupts native tissue architecture, eliminating spatial information that is essential for studying processes such as invasion, guided migration, and localized stromal activation( 14 – 17 ). As a result, although transcriptionally invasive cell states have been identified, their spatial organization, interaction partners, and functional roles within intact tissue remain largely inferred rather than directly observed( 14 – 17 ). Spatial transcriptomics provides a powerful framework to overcome these limitations by enabling transcriptome-wide profiling while preserving tissue context( 14 – 18 ). Emerging spatial studies have begun to map adenomyotic lesions across anatomical compartments, offering valuable descriptive atlases of gene expression( 12 , 13 ). Nevertheless, a mechanistic understanding of how spatially organized cell–cell communication programs coordinate epithelial invasion and myometrial remodeling is still lacking( 14 – 17 ). In particular, the temporal sequence linking epithelial adhesion remodeling, immune-guided migration, and stromal activation has not been systematically resolved( 3 , 7 , 18 , 19 ). In this study, we integrated high-resolution Stereo-seq spatial transcriptomics with publicly available scRNA-seq datasets to construct a spatiotemporal cellular atlas of human adenomyosis( 20 ). By delineating pre-invasive and invasive regions within intact tissue sections, we systematically compared transcriptional programs across disease stages and cell types( 20 ). Our analysis focused on epithelial cell state transitions during invasion, the chemokine-mediated cues guiding their directional migration, and the growth factor signaling networks through which invaded epithelial cells reshape the myometrial microenvironment( 21 – 24 ). Accordingly, the present study was designed to define spatially resolved epithelial, immune, and stromal cell states across invasive and adjacent non-invasive adenomyotic regions and to identify candidate signaling programs associated with epithelial invasion and myometrial remodeling. Together, our findings define adenomyosis progression as a spatially coordinated, multistage process encompassing epithelial adhesion plasticity, immune-guided migration, and epithelial-driven stromal remodeling( 9 , 19 ). This integrative framework not only refines the mechanistic understanding of adenomyosis pathogenesis but also highlights the CXCL12–CXCR4 and PDGF/IGF signaling axes as potential therapeutic targets for disease intervention( 31 – 35 ). 2 Materials and Methods 2.1 Clinical samples and ethical approval Human adenomyosis tissues were obtained from patients undergoing total hysterectomy at Ningxia Medical University General Hospital. None of the patients had received hormonal therapy prior to surgery. The study was conducted following the guidelines of the Declaration of Helsinki and the International Ethical Guidelines for Health-related Research Involving Humans. The study protocol was approved by the Institutional Review Board of Ningxia Medical University General Hospital (Approval No. KYLL-2023-0272), and written informed consent was obtained from all participants. A total of 12 surgical specimens were initially collected. Adenomyosis was diagnosed based on histopathological criteria, defined as the presence of endometrial glands and stroma invading the myometrium at a depth greater than 2 mm. Following pathological evaluation and RNA quality assessment, three specimens that met strict diagnostic and technical criteria were selected for downstream spatial transcriptomic analysis. 2.2 Tissue processing and spatial transcriptomics (Stereo-seq) Fresh surgical tissues were embedded in optimal cutting temperature (OCT) compound and stored at − 80°C. Cryosections of 10 µm thickness were prepared using a cryostat and mounted onto DNA nanoball (DNB)–patterned spatial transcriptomic chips. Sections with satisfactory RNA integrity (RNA integrity number > 7) were processed according to the Stereo-seq protocol and sequenced on the BGI STOmics platform. Raw sequencing data were processed using the Spatial Analysis Workflow (SAW, v7.1.0) pipeline. Quality control was performed to remove low-quality spatial bins, defined as bins with fewer than 500 total UMI counts or with mitochondrial gene content exceeding 15%. To enhance signal-to-noise ratio, adjacent 50 × 50 DNBs were merged into one spatial bin (bin50), corresponding to an effective spatial resolution of 25 µm. Spatial visualization and downstream analyses were conducted using the Stereopy package (v1.6.0). 2.3 Single-cell RNA-seq data processing and integration Publicly available human adenomyosis single-cell RNA-sequencing datasets (SRA accessions SRR12791871, SRR12791872, and SRR12791873) were downloaded from the NCBI Sequence Read Archive and used as reference profiles for spatial deconvolution. These datasets were selected because they were derived from adenomyosis tissue and contained epithelial, stromal, vascular, and immune cell populations relevant to the spatial transcriptomic specimens analyzed in this study. Raw reads were processed using Cell Ranger (v7.0.0) to generate gene expression matrices. Data quality control, normalization, and integration were performed using Seurat (v4.3.0). Cells with fewer than 500 detected genes, fewer than 1,000 UMIs, or mitochondrial gene content greater than 15% were excluded. Highly variable genes were identified, followed by principal component analysis. Batch effects across samples were corrected using the Harmony algorithm. Cell type annotation was performed using the ScType tool based on curated marker gene sets. 2.4 Spatial cell type deconvolution Spatial cell type composition was inferred using Cell2location, a Bayesian framework that integrates spatial transcriptomic data with reference single-cell RNA-seq profiles. Annotated single-cell expression signatures were used as reference input. Model hyperparameters were set as follows: detection_alpha = 20 and N_cells_per_location = 2. For each spatial bin, the cell type with the highest inferred abundance was assigned as the dominant cell identity for visualization and comparative analyses. 2.5 Region annotation and differential expression analysis Based on histopathological features and spatial cell type distributions, invasive lesions and adjacent non-invasive regions were manually annotated on spatial transcriptomic sections using StereoMap software. Gene expression matrices corresponding to selected regions were extracted using the SAW pipeline. Differential gene expression analysis between invasive and non-invasive regions was performed in Stereopy using a two-sided t -test with Bonferroni correction for multiple testing. Genes with an absolute log₂ fold change greater than 0.5 and an adjusted P value < 0.05 were considered differentially expressed. 2.6 Functional enrichment analysis Gene Ontology (GO) biological process terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were conducted using the clusterProfiler package (v4.14.0). Enrichment significance was assessed using the Benjamini–Hochberg method to control the false discovery rate. 2.7 Cell–cell communication analysis Cell–cell communication networks were inferred using CellChat (v2.1.2) based on spatially resolved cell type annotations. Ligand–receptor interactions were identified using the built-in curated database, and communication probabilities were computed using default parameters. Signaling pathways related to chemokines (CXCL family) and growth factors (PDGF and IGF families) were analyzed, and spatial interaction networks were visualized to assess pathway activity across tissue regions. 2.8 Statistical analysis and visualization All statistical analyses were performed in R (v4.2.0) and Python (v3.9). Spatial visualizations were generated using Stereopy and Scanpy. No statistical methods were used to pre-determine sample size. 3 Results 3.1 High-resolution spatial cellular mapping reveals the cellular geography of the invasive front in adenomyosis To characterize the spatial transcriptional landscape of adenomyosis, we performed Stereo-seq–based spatial transcriptomic profiling on tissue sections from three pathologically confirmed adenomyosis specimens (A1, A7, and A12). Spatial visualization of total unique molecular identifier (UMI) counts revealed pronounced transcriptional heterogeneity across tissue sections (Fig. 1 A), with localized regions exhibiting markedly higher gene detection levels. To resolve the cellular composition underlying this spatial heterogeneity, we first processed publicly available adenomyosis single-cell RNA-sequencing data as a reference dataset, which supported cluster annotation and marker-based identification of 12 major cell populations (Figure S1 A-B). We then integrated these reference profiles with the spatial transcriptomic data and applied Cell2location for high-resolution spatial deconvolution. This approach identified smooth muscle cells, fibroblasts, ciliated epithelial cells, non-ciliated epithelial cells, endothelial cells, pericytes, and immune cell populations across the spatial sections (Fig. 1 B-C). Analysis of cell type abundance demonstrated that smooth muscle cells, fibroblasts, and epithelial cells constituted the dominant populations across all samples (Fig. 1 D; Figure S1 C). Spatial distribution maps further revealed distinct localization patterns among these cell types. Smooth muscle cells were predominantly distributed within the myometrial compartment, whereas fibroblasts, ciliated epithelial cells, and non-ciliated epithelial cells displayed elongated or clustered distributions extending from the endometrium toward the myometrium (Fig. 1 E-F; Figure S2 ). These spatial patterns suggested that epithelial and fibroblast populations were closely associated with regions of myometrial invasion. Functional enrichment analysis of marker genes for these cell populations showed that fibroblasts and epithelial cells were significantly enriched for pathways related to cell adhesion, tight junctions, focal adhesion, and extracellular matrix organization (Figure S1 D–E). Consistent with these functional signatures, spatial distribution analysis of smooth muscle cells, fibroblasts, ciliated epithelial cells, non-ciliated epithelial cells, and pericytes (Figure S1 F) revealed that fibroblasts as well as both ciliated and non-ciliated epithelial cells had breached their normal anatomical boundaries and infiltrated the myometrial region. Together, these functional and spatial features indicate that fibroblasts and epithelial cells are likely key contributors to tissue invasion during adenomyosis progression. 3.2 Distinct transcriptional programs characterize invasive and non-invasive tissue regions To investigate molecular changes associated with tissue invasion, invasive lesions and adjacent non-invasive regions were manually delineated on spatial transcriptomic sections based on histopathological features and spatial cell type composition (Fig. 2 A). Comparative analysis of cell type abundance showed that invasive regions were dominated by smooth muscle cells and contained variable enrichment of fibroblasts and epithelial cell populations relative to adjacent non-invasive regions across the three samples (Fig. 2 B). Differential gene expression analysis between invasive and non-invasive regions was performed for each major cell type. Gene Ontology (GO) enrichment analysis of upregulated genes in invasive regions revealed cell type-specific transcriptional responses (Fig. 2 C), whereas GO analysis of downregulated genes is shown in Fig. 2 D. In fibroblasts and smooth muscle cells, differentially expressed genes were enriched for cytoskeletal organization, extracellular matrix-related processes, and smooth muscle-associated programs, indicating remodeling of contractile and structural features rather than a simple increase in contractile activity. In immune cell populations, enriched processes included immune signaling and cytoplasmic translation. Notably, in both ciliated and non-ciliated epithelial cells, upregulated genes were enriched in adhesion-related biological processes (Fig. 2 C). KEGG pathway analysis further showed that upregulated genes in epithelial cells were enriched in focal adhesion, adherens junction, and related cell-junction pathways (Fig. 2 E). Downregulated genes were enriched in disease- and inflammation-associated pathways, including MAPK and TNF signaling, as well as TGF-beta-related processes (Fig. 2 F). These results indicate that epithelial cells in invasive regions exhibit a transcriptional profile characterized by adhesion-related pathway enrichment accompanied by shifts in inflammatory and intracellular signaling programs. 3.3 Epithelial cell subtypes exhibit coordinated adhesion remodeling during invasion To further dissect epithelial cell–specific transcriptional changes, differential expression analyses were performed separately for ciliated and non-ciliated epithelial cells between invasive and non-invasive regions (Table S1 ). Gene Ontology enrichment analysis revealed that upregulated genes in both epithelial subtypes were associated with epithelial cell migration, proliferation, and regulation of apoptosis (Figure S3 A). In non-ciliated epithelial cells, additional enrichment was observed for pathways related to positive regulation of epithelial–mesenchymal transition. Analysis of downregulated genes showed enrichment in processes related to epithelial–mesenchymal transition and apoptosis regulation, particularly in ciliated epithelial cells (Figure S3 B), indicating heterogeneity in epithelial cell states following invasion. KEGG pathway analysis demonstrated that, in both epithelial subtypes, upregulated genes were enriched in pathways associated with focal adhesion, adherens junctions, tight junctions, and gap junctions (Fig. 3 A). In contrast, downregulated genes in both subtypes were consistently enriched in the MAPK signaling pathway (Fig. 3 B). Consistent with these enrichment results, Table S1 showed that focal adhesion, adherens junction, tight junction, and gap junction pathways were significantly enriched among upregulated genes in both epithelial subtypes. Because the current figure set summarizes pathway enrichment rather than individual spatial expression of adhesion genes, these findings support an association between invasive-region epithelial cells and adhesion-related transcriptional remodeling, but do not by themselves establish the spatial expression pattern of each adhesion molecule (Fig. 3 A-B; Table S1 ). 3.4 Chemokine signaling mediates directed epithelial migration toward the myometrium To assess whether epithelial invasion is associated with coordinated cell–cell signaling, we performed cell–cell communication analysis using CellChat. Global interaction analysis revealed multiple chemokine ligand–receptor pairs involving epithelial and immune cells (Figure S4 A–C; Table S2 ). Among these, the CXCL12–CXCR4 signaling axis emerged as the most consistently active chemokine pathway across all three samples. CellChat analysis of the CXCL signaling network identified CXCL12-CXCR4 and related chemokine ligand-receptor pairs among epithelial and immune cell populations across samples (Fig. 4 A-C; Table S2 ). The current network plots support predicted communication links and relative pathway contributions, but they do not provide a direct quantitative measurement of cell co-localization or ligand-receptor co-expression at single-cell resolution. Therefore, these results are interpreted as evidence that CXCL signaling is spatially associated with epithelial-immune communication during adenomyotic invasion. In sample A1, both ciliated and non-ciliated epithelial cells in non-invasive regions were predicted to receive CXCL12 signals from immune cells, while epithelial cells in invasive regions acted as CXCL12 signal sources (Fig. 4 A). Similar interaction patterns were observed in samples A7 and A12, although the relative contributions of specific immune cell types varied between samples (Fig. 4 B–C). In addition to CXCL12–CXCR4, CX3CL1–CX3CR1 signaling was detected in sample A1 (Figure S4 D; Table S2 ), indicating sample-specific chemokine interactions. Collectively, these results suggest that chemokine signaling, particularly through the CXCL12–CXCR4 axis, is spatially associated with epithelial cell migration during adenomyosis. 3.5 Invaded epithelial cells activate stromal cells through PDGF and IGF signaling To examine how epithelial cells influence the surrounding myometrial microenvironment following invasion, epithelial cells within invasive regions were designated as signal-sending populations for further cell–cell communication analysis. Multiple growth factor–related signaling pathways were identified, including fibroblast growth factor (FGF), PDGF, IGF, and transforming growth factor-β (TGF-β) pathways (Table S2 ). Among these, PDGF and IGF signaling pathways were consistently active across all three samples. Spatial interaction analysis revealed that epithelial cells in invasive regions expressed PDGF ligands that interacted with receptor-expressing smooth muscle cells, pericytes, and fibroblasts (Fig. 5 A–C). In sample A1, both ciliated and non-ciliated epithelial cells contributed to PDGF signaling toward smooth muscle cells and pericytes, whereas in samples A7 and A12, PDGF signaling was predominantly driven by ciliated epithelial cells targeting fibroblasts. Similarly, IGF signaling analysis showed that epithelial cells in invasive regions acted as major ligand sources, with predicted interactions involving smooth muscle cells, fibroblasts, pericytes, and immune cells (Fig. 5 D–F). While IGF signaling in sample A1 targeted multiple stromal and immune cell types, signaling in samples A7 and A12 primarily involved interactions with smooth muscle cells and fibroblasts. In addition, FGF and TGF-β signaling pathways were detected in subsets of samples (Figure S5 A–F), suggesting supplementary growth factor signaling interactions in specific tissue contexts. Overall, these results indicate that epithelial cells that have invaded the myometrium exhibit active growth factor signaling toward surrounding stromal cells. 3.6 Spatial heterogeneity reveals potential disease-associated epithelial subpopulations Exploratory spatial analysis suggested epithelial heterogeneity at the invasive front, including small epithelial subsets with features that may warrant further investigation in relation to progenitor-like states. Because LGR5 is commonly interpreted as a stem/progenitor-associated marker and the present figure set does not include dedicated validation of LGR5-positive epithelial cells, this observation should be regarded as hypothesis-generating and requires confirmation by targeted spatial visualization or immunostaining. 4 Discussion In this study, we integrated high-resolution spatial transcriptomics with single-cell transcriptomic references to delineate the cellular architecture and signaling networks associated with adenomyosis progression( 9 – 11 , 14 , 15 ). Recent single-cell studies have revealed epithelial, stromal, immune, and vascular heterogeneity in adenomyosis, including paracrine epithelial-stromal interactions such as WNT/SFRP signaling( 9 ), while emerging spatial transcriptomic studies have described ectopic endometrial penetration and lesion-associated spatial organization( 10 , 11 ). Extending these observations, our analysis directly compared invasive and adjacent non-invasive regions within intact tissue sections and identified coordinated epithelial adhesion programs, chemokine-associated epithelial-immune communication, and growth factor-related epithelial-stromal interactions as spatial features of the invasive front. A central finding of our study is the dynamic regulation of epithelial adhesion programs during invasion( 25 – 27 ). Both ciliated and non-ciliated epithelial cells exhibited a transition from a low-adhesion state in non-invasive regions to a re-anchored state within the myometrium, characterized by upregulation of focal adhesion and cell–cell junction pathways( 25 – 27 ). This “detachment–reattachment” behavior reflects a form of epithelial plasticity that enables migration while preserving epithelial identity( 26 , 27 ). Importantly, unlike malignant invasion, epithelial cells in adenomyosis do not undergo irreversible dedifferentiation or sustained loss of junctional structures( 26 , 28 ). Instead, adhesion remodeling appears to be tightly regulated and reversible, consistent with invasion occurring within a benign but mechanically and hormonally dynamic tissue environment(26,28). Our spatial analysis further implicates the CXCL12-CXCR4 axis in epithelial-immune communication at the invasive front( 21 – 23 , 30 ). Previous work has linked CXCL12-CXCR4 signaling to immune-cell recruitment, fibrosis, and tissue remodeling in multiple disease contexts( 21 – 24 ), but its spatial relationship to adenomyotic epithelial invasion has been less clearly defined. In our data, predicted CXCL signaling networks involved epithelial and immune cell populations across all three samples, suggesting that chemokine-mediated communication may contribute to the microenvironmental context in which epithelial cells invade the myometrium. However, because these findings are based on transcriptomic ligand-receptor inference, they should be interpreted as candidate interaction networks rather than direct proof of chemokine gradients or directed migration. Beyond migration-associated signaling, our findings suggest that invaded epithelial cells participate in stromal remodeling through growth factor-related communication( 31 – 35 ). Spatial cell-cell communication analysis predicted PDGF and IGF signaling from epithelial cells in invasive regions toward smooth muscle cells, fibroblasts, and pericytes( 31 – 35 ). These results are consistent with prior evidence that growth factor pathways regulate proliferation, extracellular matrix remodeling, and fibrosis( 31 – 36 ), and they provide a spatially resolved candidate mechanism linking ectopic epithelial compartments to myometrial hyperplasia and fibrosis. Nevertheless, this epithelial-stromal model should be tested experimentally, because transcriptomic inference cannot determine whether PDGF or IGF signaling is sufficient to induce stromal activation in adenomyosis. The spatial integration of adhesion remodeling, chemokine-guided migration, and growth factor–mediated stromal activation provides a unified framework for understanding adenomyosis as a coordinated tissue-level process( 9 – 11 ). Importantly, these signaling axes appear to operate at distinct but interconnected stages of disease development( 9 – 11 ). Chemokine signaling through CXCL12–CXCR4 is closely associated with the initiation and directionality of epithelial migration, whereas PDGF and IGF signaling likely contribute to lesion stabilization and amplification through sustained stromal activation( 31 – 36 ). Within this framework, recently identified targets such as the prolactin receptor may intersect with or modulate these spatial programs, underscoring the multifactorial nature of adenomyosis pathogenesis and the potential need for combinatorial therapeutic strategies( 37 – 39 ). Several limitations of this study should be acknowledged. First, the spatial transcriptomic cohort included three specimens selected after pathological review and RNA quality control. This sample size is appropriate for exploratory high-resolution spatial profiling but limits statistical power, the ability to model inter-patient heterogeneity, and the generalizability of region-level comparisons. Second, ligand-receptor interactions inferred from CellChat represent probabilistic predictions based on transcript abundance rather than direct biochemical measurements of protein binding, secretion, or receptor activation( 16 , 17 ). Third, the cross-sectional nature of human tissue sampling limits the ability to reconstruct temporal sequences with certainty, and inferred invasion stages are based on spatial context rather than longitudinal observation( 14 – 17 ). Finally, the study did not include independent experimental validation such as multiplex immunostaining, RNAscope, organoid/assembloid perturbation, or in vivo functional assays. Future studies with larger clinically annotated cohorts and targeted validation experiments will be essential to determine the causal roles and therapeutic relevance of the candidate CXCL12-CXCR4 and PDGF/IGF signaling axes( 37 – 39 ). In summary, our study defines adenomyosis progression as a spatially organized process associated with epithelial plasticity, epithelial-immune communication, and epithelial-stromal remodeling( 9 – 11 ). By identifying CXCL12-CXCR4 and PDGF/IGF signaling as candidate components of this invasion-remodeling axis, we provide a spatially informed framework that links cellular behavior with tissue-scale pathology and offers a basis for future mechanistic and therapeutic studies( 37 – 39 ). Declarations Conflict of Interest The authors declare no competing interests. Author Contributions Caihong Yang: Conceptualization, Methodology, Supervision,Formal analysis. Hui Dong: Data curation, Visualization,software. Lulu Shi:Funding acquisition. Wenli Gu,Dan Yang:Writing – original draft , Writing – review & editing. Funding This study was supported by the Natural Science Foundation of Ningxia (Grant No. 2024AAC03628). Acknowledgements We thank all patients who participated in this study. We acknowledge BGI-STOmics (Shenzhen, China) for providing Stereo-seq technical support and Wuhan Ru Biotechnology Co., Ltd. for assistance with data analysis. Data Availability Statement The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive (Genomics, Proteomics & Bioinformatics 2025) in the National Genomics Data Center (Nucleic Acids Res 2025), China National Center for Bioinformation / Beijing Institute of Genomics, Chinese Academy of Sciences, under BioProject accession number PRJCA057653 (sample/experiment accession: HRA016686 ). The data are publicly accessible at https://ngdc.cncb.ac.cn/gsa-human as of February 10, 2026. References Chapron, C. et al. Diagnosing adenomyosis: an integrated clinical and imaging approach. Hum. Reprod. Update 15 avr . 26 (3), 392–411 (2020). Bourdon, M. et al. Adenomyosis: An update regarding its diagnosis and clinical features. J. Gynecol. Obstet. Hum. Reprod. déc . 50 (10), 102228 (2021). Khan, K. N., Fujishita, A. & Mori, T. Pathogenesis of Human Adenomyosis: Current Understanding and Its Association with Infertility. JCM 13 juill . 11 (14), 4057 (2022). Vannuccini, S. & Petraglia, F. Recent advances in understanding and managing adenomyosis. F1000Res 13 mars . 8 , 283 (2019). Santulli, P., Vannuccini, S., Bourdon, M., Chapron, C. & Petraglia, F. Adenomyosis: the missed disease. Reproductive Biomed. Online avr . 50 (4), 104837 (2025). Giudice, L. C., Liu, B. & Irwin, J. C. Endometriosis and adenomyosis unveiled through single-cell glasses. Am. J. Obstet. Gynecol. avr . 232 (4), S105–S123 (2025). Bulun, S. E., Yildiz, S., Adli, M. & Wei, J. J. Adenomyosis pathogenesis: insights from next-generation sequencing. Hum. Reprod. Update . 18 (6), 1086–1097 (oct 2021). Zhai, J., Vannuccini, S., Petraglia, F. & Giudice, L. C. Adenomyosis: Mechanisms and Pathogenesis. Semin Reprod. Med. mai . 38 (02/03), 129–143 (2020). Yildiz, S. et al. Adenomyosis: single-cell transcriptomic analysis reveals a paracrine mesenchymal–epithelial interaction involving the WNT/SFRP pathway. Fertility Steril. mai . 119 (5), 869–882 (2023). Li, B. et al. From Invaginating Site to Deep Lesion: Spatial Transcriptomics Unravels Ectopic Endometrial Penetration Features in Adenomyosis. Adv. Sci. mai . 12 (20), 2411752 (2025). Yang, X. et al. Spatial and single-cell transcriptomics landscape of adenomyosis. J. Adv. Res. déc 2025;S209012322501032X. Wang, R. et al. Single-cell RNA sequencing identifies the prolactin receptor as a therapeutic target in adenomyosis. Sig Transduct. Target. Ther. 13 août . 10 (1), 258 (2025). Li, Y. et al. Signaling pathways and advances in targeted therapy for adenomyosis. Front. Cell. Dev. Biol. 6 nov . 13 , 1685525 (2025). Tian, L., Chen, F. & Macosko, E. Z. The expanding vistas of spatial transcriptomics. Nat. Biotechnol. juin . 41 (6), 773–782 (2023). Chen, T. Y., You, L., Hardillo, J. A. U. & Chien, M. P. Spat. Transcriptomic Technol. Cells 10 août ; 12 (16):2042. (2023). Kleshchevnikov, V. et al. Cell2location maps fine-grained cell types in spatial transcriptomics. Nat. Biotechnol. mai . 40 (5), 661–671 (2022). Jin, S. et al. Inference and analysis of cell-cell communication using CellChat. Nat. Commun. 17 févr . 12 (1), 1088 (2021). Chen, A. et al. Spatiotemporal transcriptomic atlas of mouse organogenesis using DNA nanoball-patterned arrays. Cell. mai . 185 (10), 1777–1792e21 (2022). Wahafu, W. et al. Spatial relationships and interactions of immune cell niches are linked to the pathologic response of muscle-invasive bladder cancer to neoadjuvant therapy. J Transl Med. 27 mars. ;23(1):375. (2025). Fang, S. et al. Stereopy: modeling comparative and spatiotemporal cellular heterogeneity via multi-sample spatial transcriptomics. Nat. Commun. 21 avr . 16 (1), 3741 (2025). Wu, X. et al. CXCL12/CXCR4: An amazing challenge and opportunity in the fight against fibrosis. Ageing Res. Reviews janv . 83 , 101809 (2023). Daniel, S. K., Seo, Y. D. & Pillarisetty, V. G. The CXCL12-CXCR4/CXCR7 axis as a mechanism of immune resistance in gastrointestinal malignancies. Seminars Cancer Biology oct. 65 , 176–188 (2020). Zhou, C., Gao, Y., Ding, P., Wu, T. & Ji, G. The role of CXCL family members in different diseases. Cell. Death Discov 1 juill . 9 (1), 212 (2023). Kohli, K., Pillarisetty, V. G. & Kim, T. S. Key chemokines direct migration of immune cells in solid tumors. Cancer Gene Ther. janv . 29 (1), 10–21 (2022). Campàs, O., Noordstra, I. & Yap, A. S. Adherens junctions as molecular regulators of emergent tissue mechanics. Nat. Rev. Mol. Cell. Biol. avr . 25 (4), 252–269 (2024). Hirano, S., Aoki, K. & Ueno, N. Dynamic behavior of cell-cell adhesion factors in collective cell migration. Cells Dev. déc . 184 , 203995 (2025). Pourjafar, M. & Tiwari, V. K. Plasticity in cell migration modes across development, physiology, and disease. Front. Cell. Dev. Biol. 23 avr . 12 , 1363361 (2024). Sun, X., Joost, S. & Kasper, M. Plasticity of Epithelial Cells during Skin Wound Healing. Cold Spring Harb Perspect. Biol. mai . 15 (5), a041232 (2023). Bijonowski, B. M. et al. Intercellular adhesion boots collective cell migration through elevated membrane tension. Nat. Commun. 12 févr . 16 (1), 1588 (2025). Garg, P., Jallepalli, V. R. & Verma, S. Unravelling the CXCL12/CXCR4 Axis in breast cancer: Insights into metastasis, microenvironment interactions, and therapeutic opportunities. Hum. Gene mai . 40 , 201272 (2024). Zhao, M. et al. Targeting fibrosis: mechanisms and clinical trials. Sig Transduct Target Ther. 30 juin. ;7(1):206. (2022). Ku, J. C., Raiten, J. & Li, Y. Understanding fibrosis: Mechanisms, clinical implications, current therapies, and prospects for future interventions. Biomedical Eng. Adv. juin . 7 , 100118 (2024). Irma, J. et al. From Growth Factors to Structure: PDGF and TGF-β in Granulation Tissue Formation. A Literature Review. J. Cell. Mol. Medi juin . 29 (11), e70374 (2025). Khan, M. Z., Zugaza, J. L. & Torres Aleman, I. The signaling landscape of insulin-like growth factor 1. J. Biol. Chem. janv . 301 (1), 108047 (2025). Sechrist, Z. R. et al. Pathologic Signaling and Disease Implications of Insulin-like Growth Factor Binding Proteins in Cancer, Cardiovascular Disease, and Fibrosis. IJMS 22 (21), 10248 (oct 2025). Di, X., Li, Y., Wei, J., Li, T. & Liao, B. Targeting Fibrosis: From Molecular Mechanisms to Advanced Therapies. Adv. Sci. janv . 12 (3), 2410416 (2025). Rawlings, T. M., Makwana, K., Tryfonos, M. & Lucas, E. S. Organoids to model the endometrium: implantation and beyond. Reprod. Fertility 8 sept . 2 (3), R85–101 (2021). Stratopoulou, C. A. et al. Generation of epithelial-stromal assembloids as an advanced in vitro model of impaired adenomyosis-related endometrial receptivity. Fertility Steril. févr . 123 (2), 350–360 (2025). Xu, Y. et al. Decoding adenomyosis pathogenesis using an assembloid model. Sci. China Life Sci. janv . 69 (1), 136–150 (2026). Additional Declarations No competing interests reported. Supplementary Files TableS1invasionvsnormalA1A7A12.xlsx TableS2Ligandreceptor.xlsx SupplementaryMaterial.docx Supplementary Materials Supplementary Figures S1–S5 illustrate the spatial transcriptomic landscape of human adenomyosis, including differential expression analyses of epithelial cells between invasive and non-invasive regions, as well as chemokine and growth factor ligand–receptor–mediated cell–cell interaction analyses. Supplementary Table S1 presents GO and KEGG enrichment analyses of differentially expressed genes in ciliated and non-ciliated epithelial cells between invasive and pre-invasive regions. Supplementary Table S2 summarizes the ligand–receptor interactions among different cell types identified from the cell–cell communication analysis. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 21 Aug, 2026 Reviewers agreed at journal 21 Aug, 2026 Reviewers invited by journal 18 Aug, 2026 Editor assigned by journal 22 Jun, 2026 Editor invited by journal 05 Jun, 2026 Submission checks completed at journal 03 Jun, 2026 First submitted to journal 03 Jun, 2026 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 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-9819300","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":703656360,"identity":"6a98de89-4bb5-49d8-b723-c625d4e006f3","order_by":0,"name":"Wenli Gu","email":"","orcid":"","institution":"Ningxia Medical University General Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Wenli","middleName":"","lastName":"Gu","suffix":""},{"id":703656361,"identity":"e388010a-925b-4d92-9c61-6f42d6653f2e","order_by":1,"name":"Dan Yang","email":"","orcid":"","institution":"Ningxia Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Dan","middleName":"","lastName":"Yang","suffix":""},{"id":703656362,"identity":"163882b6-cbe5-41cc-84d4-c774c21d6a62","order_by":2,"name":"Lulu Shi","email":"","orcid":"","institution":"Xijing Hospital, Air Force Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Lulu","middleName":"","lastName":"Shi","suffix":""},{"id":703656363,"identity":"bf86da4f-0660-4f7c-a30e-841f171d5f5a","order_by":3,"name":"Hui Dong","email":"","orcid":"","institution":"General Hospital of Ningxia Medical University,","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Hui","middleName":"","lastName":"Dong","suffix":""},{"id":703656364,"identity":"ed744f5e-6119-427e-b0ef-9b65dcb3c4d7","order_by":4,"name":"Caihong Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA40lEQVRIiWNgGAWjYBACNv7mg49/GPyX42dvP/ggoaKGsBY+iWPJxgwVzMaSPWeSDR6cOUZYixxDjpk0wxnmRIMbCWaSD1uYiXAYw7E06cI2tgSglrSKxAY2Bv727gT8WpibD1vPbOPJkzzz8NiNxB0yDBJnzm4gZEviDd42iWK+4wlpNxLPsDEYSOQS0pJjIMHbZpDYcCDBrCCxjZkoLUbSPGcSEiecSDBjIE4LMJANZ1QcAAeyRMKZYzwE/SLf33zwwQeDA+Co/PijokaOv70XvxYMwEOa8lEwCkbBKBgFWAEA9l1PXYhgWP0AAAAASUVORK5CYII=","orcid":"","institution":"Ningxia Medical University General Hospital","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Caihong","middleName":"","lastName":"Yang","suffix":""}],"badges":[],"createdAt":"2026-05-26 01:53:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9819300/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9819300/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":118410486,"identity":"e5dd2f93-8833-4c51-9833-60df07527562","added_by":"auto","created_at":"2026-08-26 14:30:37","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":6153841,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSpatial transcriptomic profiling of human adenomyosis tissues.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Spatial scatter plots showing total unique molecular identifier (UMI) counts across tissue sections from samples A1, A7, and A12.\u003c/p\u003e\n\u003cp\u003e(B) Spatial scatter plots showing cell type annotations of spatial transcriptomic data inferred using Cell2location. Colors indicate annotated cell types.\u003c/p\u003e\n\u003cp\u003e(C) Heatmap displaying the expression of representative marker genes used for cell type annotation.\u003c/p\u003e\n\u003cp\u003e(D) Bar plots showing the relative proportions of major cell types in each sample. Colors indicate individual samples.\u003c/p\u003e\n\u003cp\u003e(E) Spatial distribution maps of unciliated epithelial cells, ciliated epithelial cells, smooth muscle cells, and fibroblasts.\u003c/p\u003e\n\u003cp\u003e(F) Spatial visualization of the expression levels of EPCAM, WFDC2, ACTG2, and COL3A1 across tissue sections.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-9819300/v1/6031f93c60a216bca0656f2d.png"},{"id":118466620,"identity":"faec1b52-90da-4720-b003-191a590e8b5c","added_by":"auto","created_at":"2026-08-27 07:12:03","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":3351077,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDifferential gene expression analysis between invasive and non-invasive tissue regions.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Manual annotation of invasive and non-invasive regions using the lasso tool in StereoMap software. Invasive regions are indicated in yellow, and non-invasive regions are indicated in light blue.\u003c/p\u003e\n\u003cp\u003e(B) Bar plots showing the relative proportions of major cell types in invasive and non-invasive regions for samples A1 (left), A7 (middle), and A12 (right).\u003c/p\u003e\n\u003cp\u003e(C) GO biological process enrichment analysis of upregulated DEGs between invasive and non-invasive regions. The top three enriched terms for each cell type are shown. Dot size represents gene ratio, and color indicates adjusted \u003cem\u003eP\u003c/em\u003evalue.\u003c/p\u003e\n\u003cp\u003e(D) GO biological process enrichment analysis of downregulated DEGs between invasive and non-invasive regions, presented as in panel (C).\u003c/p\u003e\n\u003cp\u003e(E) KEGG pathway enrichment analysis of upregulated DEGs between invasive and non-invasive regions. The top three enriched pathways for each cell type are shown.\u003c/p\u003e\n\u003cp\u003e(F) KEGG pathway enrichment analysis of downregulated DEGs between invasive and non-invasive regions, presented as in panel (E).\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-9819300/v1/76847bfae59708fc05f86be7.png"},{"id":118410489,"identity":"aba305b6-b9aa-4c8a-9bff-009cbde0bc0c","added_by":"auto","created_at":"2026-08-26 14:30:37","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":445792,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDifferential gene expression analysis of epithelial cells between invasive and non-invasive regions.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Bubble plots showing KEGG pathway enrichment of upregulated DEGs in epithelial cells from invasive regions compared with non-invasive regions (left, ciliated epithelial cells; right, unciliated epithelial cells).\u003c/p\u003e\n\u003cp\u003e(B) Bubble plots showing KEGG pathway enrichment of downregulated DEGs in epithelial cells from invasive regions compared with non-invasive regions (left, ciliated epithelial cells; right, unciliated epithelial cells).\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-9819300/v1/0b940f7c453e571ab5b822f9.png"},{"id":118410497,"identity":"dd3c02ae-9eec-463b-924d-de9aac4ff95e","added_by":"auto","created_at":"2026-08-26 14:30:38","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":2595715,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eChemokine ligand–receptor interaction analysis based on spatial cell–cell communication.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Heatmap showing centrality scores of each cell type as signal senders, receivers, mediators, and influencers in the CXCL signaling network in sample A1 (left); spatial plot showing inferred CXCL signaling interactions in sample A1 (middle); bar plot showing the contribution of individual ligand–receptor pairs to the overall CXCL signaling pathway in sample A1 (right).\u003c/p\u003e\n\u003cp\u003e(B) Corresponding CXCL signaling network analyses for sample A7, presented as in panel (A).\u003c/p\u003e\n\u003cp\u003e(C) Corresponding CXCL signaling network analyses for sample A12, presented as in panel (A).\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-9819300/v1/cf73ca1217260e85e8aff592.png"},{"id":118410495,"identity":"d9c91c6b-86cd-4d40-bb03-894278f74e8b","added_by":"auto","created_at":"2026-08-26 14:30:38","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":4361556,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGrowth factor ligand–receptor interaction analysis based on spatial cell–cell communication.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Heatmap showing centrality scores of each cell type as signal senders, receivers, mediators, and influencers in the PDGF signaling network in sample A1 (left); spatial plot showing inferred PDGF signaling interactions in sample A1 (middle); bar plot showing the contribution of individual ligand–receptor pairs to the overall PDGF signaling pathway in sample A1 (right).\u003c/p\u003e\n\u003cp\u003e(B) Corresponding PDGF signaling network analyses for sample A7, presented as in panel (A).\u003c/p\u003e\n\u003cp\u003e(C) Corresponding PDGF signaling network analyses for sample A12, presented as in panel (A).\u003c/p\u003e\n\u003cp\u003e(D) Heatmap showing centrality scores of each cell type as signal senders, receivers, mediators, and influencers in the IGF signaling network in sample A1 (left); spatial plot showing inferred IGF signaling interactions in sample A1 (middle); bar plot showing the contribution of individual ligand–receptor pairs to the overall IGF signaling pathway in sample A1 (right).\u003c/p\u003e\n\u003cp\u003e(E) Corresponding IGF signaling network analyses for sample A7, presented as in panel (D).\u003c/p\u003e\n\u003cp\u003e(F) Corresponding IGF signaling network analyses for sample A12, presented as in panel (D).\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-9819300/v1/3325fc06cae9f0f6d4e2dce8.png"},{"id":118468796,"identity":"535fcb12-0189-49a3-8088-447b83cd9dc8","added_by":"auto","created_at":"2026-08-27 07:18:20","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":16629682,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9819300/v1/42d3ec5f-80ad-4ecd-b9a1-8f9041a271d1.pdf"},{"id":118410487,"identity":"380ece59-a9d2-4dfe-a24a-f148c063a732","added_by":"auto","created_at":"2026-08-26 14:30:37","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":172867,"visible":true,"origin":"","legend":"","description":"","filename":"TableS1invasionvsnormalA1A7A12.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9819300/v1/10573681e5933f096dee9052.xlsx"},{"id":118410491,"identity":"ba3f3ece-dfa6-494f-a267-78bd573553f4","added_by":"auto","created_at":"2026-08-26 14:30:37","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":499894,"visible":true,"origin":"","legend":"","description":"","filename":"TableS2Ligandreceptor.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9819300/v1/72622d53f1e75e36fa30a60c.xlsx"},{"id":118410490,"identity":"fdbb7433-04fa-4547-becb-bcbbf8a8bf74","added_by":"auto","created_at":"2026-08-26 14:30:37","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":7060668,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSupplementary Figures S1–S5 illustrate the spatial transcriptomic landscape of human adenomyosis, including differential expression analyses of epithelial cells between invasive and non-invasive regions, as well as chemokine and growth factor ligand–receptor–mediated cell–cell interaction analyses. Supplementary Table S1 presents GO and KEGG enrichment analyses of differentially expressed genes in ciliated and non-ciliated epithelial cells between invasive and pre-invasive regions. Supplementary Table S2 summarizes the ligand–receptor interactions among different cell types identified from the cell–cell communication analysis.\u003c/p\u003e","description":"","filename":"SupplementaryMaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-9819300/v1/be09da9df31f1faacc2379c6.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Spatial transcriptomics reveals coordinated epithelial invasion and stromal remodeling in adenomyosis","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eAdenomyosis is a common gynecological disorder characterized by the presence of endometrial glands and stroma within the myometrium, frequently accompanied by smooth muscle hyperplasia, fibrosis, and chronic inflammation(\u003cspan additionalcitationids=\"CR2 CR3\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). Clinically, adenomyosis is strongly associated with dysmenorrhea, abnormal uterine bleeding, infertility, and a substantial reduction in quality of life for affected women(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Despite its high prevalence and clinical burden, the cellular and molecular mechanisms that drive lesion initiation, invasion, and progressive myometrial remodeling remain incompletely understood(\u003cspan additionalcitationids=\"CR4 CR5 CR6\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSeveral pathogenic hypotheses have been proposed to explain adenomyosis development, including invagination of the endometrial\u0026ndash;myometrial junction following tissue microinjury, hormonal dysregulation, aberrant wound-healing responses, and stem or progenitor cell involvement(\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). While these models provide partial explanations, they largely fail to capture adenomyosis as a dynamic, multistep process(\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). In particular, key mechanistic questions remain unresolved: how endometrial epithelial cells acquire invasive capacity, how they migrate directionally into the myometrium, and how their presence subsequently induces the characteristic hypertrophy and fibrosis of the surrounding smooth muscle tissue(\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eRecent advances in single-cell RNA sequencing (scRNA-seq) have substantially expanded our understanding of cellular heterogeneity in adenomyosis, revealing distinct epithelial, stromal, immune, and vascular subpopulations with altered transcriptional programs(\u003cspan additionalcitationids=\"CR10 CR11 CR12\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). However, by necessity, scRNA-seq disrupts native tissue architecture, eliminating spatial information that is essential for studying processes such as invasion, guided migration, and localized stromal activation(\u003cspan additionalcitationids=\"CR15 CR16\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). As a result, although transcriptionally invasive cell states have been identified, their spatial organization, interaction partners, and functional roles within intact tissue remain largely inferred rather than directly observed(\u003cspan additionalcitationids=\"CR15 CR16\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSpatial transcriptomics provides a powerful framework to overcome these limitations by enabling transcriptome-wide profiling while preserving tissue context(\u003cspan additionalcitationids=\"CR15 CR16 CR17\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). Emerging spatial studies have begun to map adenomyotic lesions across anatomical compartments, offering valuable descriptive atlases of gene expression(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). Nevertheless, a mechanistic understanding of how spatially organized cell\u0026ndash;cell communication programs coordinate epithelial invasion and myometrial remodeling is still lacking(\u003cspan additionalcitationids=\"CR15 CR16\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). In particular, the temporal sequence linking epithelial adhesion remodeling, immune-guided migration, and stromal activation has not been systematically resolved(\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn this study, we integrated high-resolution Stereo-seq spatial transcriptomics with publicly available scRNA-seq datasets to construct a spatiotemporal cellular atlas of human adenomyosis(\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). By delineating pre-invasive and invasive regions within intact tissue sections, we systematically compared transcriptional programs across disease stages and cell types(\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). Our analysis focused on epithelial cell state transitions during invasion, the chemokine-mediated cues guiding their directional migration, and the growth factor signaling networks through which invaded epithelial cells reshape the myometrial microenvironment(\u003cspan additionalcitationids=\"CR22 CR23\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAccordingly, the present study was designed to define spatially resolved epithelial, immune, and stromal cell states across invasive and adjacent non-invasive adenomyotic regions and to identify candidate signaling programs associated with epithelial invasion and myometrial remodeling.\u003c/p\u003e \u003cp\u003eTogether, our findings define adenomyosis progression as a spatially coordinated, multistage process encompassing epithelial adhesion plasticity, immune-guided migration, and epithelial-driven stromal remodeling(\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). This integrative framework not only refines the mechanistic understanding of adenomyosis pathogenesis but also highlights the CXCL12\u0026ndash;CXCR4 and PDGF/IGF signaling axes as potential therapeutic targets for disease intervention(\u003cspan additionalcitationids=\"CR32 CR33 CR34\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e).\u003c/p\u003e"},{"header":"2 Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Clinical samples and ethical approval\u003c/h2\u003e \u003cp\u003eHuman adenomyosis tissues were obtained from patients undergoing total hysterectomy at Ningxia Medical University General Hospital. None of the patients had received hormonal therapy prior to surgery. The study was conducted following the guidelines of the Declaration of Helsinki and the International Ethical Guidelines for Health-related Research Involving Humans. The study protocol was approved by the Institutional Review Board of Ningxia Medical University General Hospital (Approval No. KYLL-2023-0272), and written informed consent was obtained from all participants.\u003c/p\u003e \u003cp\u003eA total of 12 surgical specimens were initially collected. Adenomyosis was diagnosed based on histopathological criteria, defined as the presence of endometrial glands and stroma invading the myometrium at a depth greater than 2 mm. Following pathological evaluation and RNA quality assessment, three specimens that met strict diagnostic and technical criteria were selected for downstream spatial transcriptomic analysis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Tissue processing and spatial transcriptomics (Stereo-seq)\u003c/h2\u003e \u003cp\u003eFresh surgical tissues were embedded in optimal cutting temperature (OCT) compound and stored at \u0026minus;\u0026thinsp;80\u0026deg;C. Cryosections of 10 \u0026micro;m thickness were prepared using a cryostat and mounted onto DNA nanoball (DNB)\u0026ndash;patterned spatial transcriptomic chips. Sections with satisfactory RNA integrity (RNA integrity number\u0026thinsp;\u0026gt;\u0026thinsp;7) were processed according to the Stereo-seq protocol and sequenced on the BGI STOmics platform.\u003c/p\u003e \u003cp\u003eRaw sequencing data were processed using the Spatial Analysis Workflow (SAW, v7.1.0) pipeline. Quality control was performed to remove low-quality spatial bins, defined as bins with fewer than 500 total UMI counts or with mitochondrial gene content exceeding 15%. To enhance signal-to-noise ratio, adjacent 50 \u0026times; 50 DNBs were merged into one spatial bin (bin50), corresponding to an effective spatial resolution of 25 \u0026micro;m. Spatial visualization and downstream analyses were conducted using the Stereopy package (v1.6.0).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Single-cell RNA-seq data processing and integration\u003c/h2\u003e \u003cp\u003ePublicly available human adenomyosis single-cell RNA-sequencing datasets (SRA accessions SRR12791871, SRR12791872, and SRR12791873) were downloaded from the NCBI Sequence Read Archive and used as reference profiles for spatial deconvolution. These datasets were selected because they were derived from adenomyosis tissue and contained epithelial, stromal, vascular, and immune cell populations relevant to the spatial transcriptomic specimens analyzed in this study. Raw reads were processed using Cell Ranger (v7.0.0) to generate gene expression matrices.\u003c/p\u003e \u003cp\u003eData quality control, normalization, and integration were performed using Seurat (v4.3.0). Cells with fewer than 500 detected genes, fewer than 1,000 UMIs, or mitochondrial gene content greater than 15% were excluded. Highly variable genes were identified, followed by principal component analysis. Batch effects across samples were corrected using the Harmony algorithm. Cell type annotation was performed using the ScType tool based on curated marker gene sets.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Spatial cell type deconvolution\u003c/h2\u003e \u003cp\u003eSpatial cell type composition was inferred using Cell2location, a Bayesian framework that integrates spatial transcriptomic data with reference single-cell RNA-seq profiles. Annotated single-cell expression signatures were used as reference input. Model hyperparameters were set as follows: \u003cem\u003edetection_alpha\u003c/em\u003e\u0026thinsp;=\u0026thinsp;20 and \u003cem\u003eN_cells_per_location\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2. For each spatial bin, the cell type with the highest inferred abundance was assigned as the dominant cell identity for visualization and comparative analyses.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Region annotation and differential expression analysis\u003c/h2\u003e \u003cp\u003eBased on histopathological features and spatial cell type distributions, invasive lesions and adjacent non-invasive regions were manually annotated on spatial transcriptomic sections using StereoMap software. Gene expression matrices corresponding to selected regions were extracted using the SAW pipeline.\u003c/p\u003e \u003cp\u003eDifferential gene expression analysis between invasive and non-invasive regions was performed in Stereopy using a two-sided \u003cem\u003et\u003c/em\u003e-test with Bonferroni correction for multiple testing. Genes with an absolute log₂ fold change greater than 0.5 and an adjusted \u003cem\u003eP\u003c/em\u003e value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were considered differentially expressed.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Functional enrichment analysis\u003c/h2\u003e \u003cp\u003eGene Ontology (GO) biological process terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were conducted using the clusterProfiler package (v4.14.0). Enrichment significance was assessed using the Benjamini\u0026ndash;Hochberg method to control the false discovery rate.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.7 Cell\u0026ndash;cell communication analysis\u003c/h2\u003e \u003cp\u003eCell\u0026ndash;cell communication networks were inferred using CellChat (v2.1.2) based on spatially resolved cell type annotations. Ligand\u0026ndash;receptor interactions were identified using the built-in curated database, and communication probabilities were computed using default parameters. Signaling pathways related to chemokines (CXCL family) and growth factors (PDGF and IGF families) were analyzed, and spatial interaction networks were visualized to assess pathway activity across tissue regions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e2.8 Statistical analysis and visualization\u003c/h2\u003e \u003cp\u003eAll statistical analyses were performed in R (v4.2.0) and Python (v3.9). Spatial visualizations were generated using Stereopy and Scanpy. No statistical methods were used to pre-determine sample size.\u003c/p\u003e \u003c/div\u003e"},{"header":"3 Results","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.1 High-resolution spatial cellular mapping reveals the cellular geography of the invasive front in adenomyosis\u003c/h2\u003e \u003cp\u003eTo characterize the spatial transcriptional landscape of adenomyosis, we performed Stereo-seq\u0026ndash;based spatial transcriptomic profiling on tissue sections from three pathologically confirmed adenomyosis specimens (A1, A7, and A12). Spatial visualization of total unique molecular identifier (UMI) counts revealed pronounced transcriptional heterogeneity across tissue sections (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA), with localized regions exhibiting markedly higher gene detection levels.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo resolve the cellular composition underlying this spatial heterogeneity, we first processed publicly available adenomyosis single-cell RNA-sequencing data as a reference dataset, which supported cluster annotation and marker-based identification of 12 major cell populations (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eA-B). We then integrated these reference profiles with the spatial transcriptomic data and applied Cell2location for high-resolution spatial deconvolution. This approach identified smooth muscle cells, fibroblasts, ciliated epithelial cells, non-ciliated epithelial cells, endothelial cells, pericytes, and immune cell populations across the spatial sections (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB-C).\u003c/p\u003e \u003cp\u003eAnalysis of cell type abundance demonstrated that smooth muscle cells, fibroblasts, and epithelial cells constituted the dominant populations across all samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD; Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eC). Spatial distribution maps further revealed distinct localization patterns among these cell types. Smooth muscle cells were predominantly distributed within the myometrial compartment, whereas fibroblasts, ciliated epithelial cells, and non-ciliated epithelial cells displayed elongated or clustered distributions extending from the endometrium toward the myometrium (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE-F; Figure \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e). These spatial patterns suggested that epithelial and fibroblast populations were closely associated with regions of myometrial invasion.\u003c/p\u003e \u003cp\u003eFunctional enrichment analysis of marker genes for these cell populations showed that fibroblasts and epithelial cells were significantly enriched for pathways related to cell adhesion, tight junctions, focal adhesion, and extracellular matrix organization (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eD\u0026ndash;E). Consistent with these functional signatures, spatial distribution analysis of smooth muscle cells, fibroblasts, ciliated epithelial cells, non-ciliated epithelial cells, and pericytes (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eF) revealed that fibroblasts as well as both ciliated and non-ciliated epithelial cells had breached their normal anatomical boundaries and infiltrated the myometrial region. Together, these functional and spatial features indicate that fibroblasts and epithelial cells are likely key contributors to tissue invasion during adenomyosis progression.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Distinct transcriptional programs characterize invasive and non-invasive tissue regions\u003c/h2\u003e \u003cp\u003eTo investigate molecular changes associated with tissue invasion, invasive lesions and adjacent non-invasive regions were manually delineated on spatial transcriptomic sections based on histopathological features and spatial cell type composition (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Comparative analysis of cell type abundance showed that invasive regions were dominated by smooth muscle cells and contained variable enrichment of fibroblasts and epithelial cell populations relative to adjacent non-invasive regions across the three samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eDifferential gene expression analysis between invasive and non-invasive regions was performed for each major cell type. Gene Ontology (GO) enrichment analysis of upregulated genes in invasive regions revealed cell type-specific transcriptional responses (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC), whereas GO analysis of downregulated genes is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD. In fibroblasts and smooth muscle cells, differentially expressed genes were enriched for cytoskeletal organization, extracellular matrix-related processes, and smooth muscle-associated programs, indicating remodeling of contractile and structural features rather than a simple increase in contractile activity. In immune cell populations, enriched processes included immune signaling and cytoplasmic translation. Notably, in both ciliated and non-ciliated epithelial cells, upregulated genes were enriched in adhesion-related biological processes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003eKEGG pathway analysis further showed that upregulated genes in epithelial cells were enriched in focal adhesion, adherens junction, and related cell-junction pathways (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE). Downregulated genes were enriched in disease- and inflammation-associated pathways, including MAPK and TNF signaling, as well as TGF-beta-related processes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF). These results indicate that epithelial cells in invasive regions exhibit a transcriptional profile characterized by adhesion-related pathway enrichment accompanied by shifts in inflammatory and intracellular signaling programs.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Epithelial cell subtypes exhibit coordinated adhesion remodeling during invasion\u003c/h2\u003e \u003cp\u003eTo further dissect epithelial cell\u0026ndash;specific transcriptional changes, differential expression analyses were performed separately for ciliated and non-ciliated epithelial cells between invasive and non-invasive regions (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Gene Ontology enrichment analysis revealed that upregulated genes in both epithelial subtypes were associated with epithelial cell migration, proliferation, and regulation of apoptosis (Figure \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003eA). In non-ciliated epithelial cells, additional enrichment was observed for pathways related to positive regulation of epithelial\u0026ndash;mesenchymal transition.\u003c/p\u003e \u003cp\u003eAnalysis of downregulated genes showed enrichment in processes related to epithelial\u0026ndash;mesenchymal transition and apoptosis regulation, particularly in ciliated epithelial cells (Figure \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003eB), indicating heterogeneity in epithelial cell states following invasion.\u003c/p\u003e \u003cp\u003eKEGG pathway analysis demonstrated that, in both epithelial subtypes, upregulated genes were enriched in pathways associated with focal adhesion, adherens junctions, tight junctions, and gap junctions (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). In contrast, downregulated genes in both subtypes were consistently enriched in the MAPK signaling pathway (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eConsistent with these enrichment results, Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e showed that focal adhesion, adherens junction, tight junction, and gap junction pathways were significantly enriched among upregulated genes in both epithelial subtypes. Because the current figure set summarizes pathway enrichment rather than individual spatial expression of adhesion genes, these findings support an association between invasive-region epithelial cells and adhesion-related transcriptional remodeling, but do not by themselves establish the spatial expression pattern of each adhesion molecule (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA-B; Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Chemokine signaling mediates directed epithelial migration toward the myometrium\u003c/h2\u003e \u003cp\u003eTo assess whether epithelial invasion is associated with coordinated cell\u0026ndash;cell signaling, we performed cell\u0026ndash;cell communication analysis using CellChat. Global interaction analysis revealed multiple chemokine ligand\u0026ndash;receptor pairs involving epithelial and immune cells (Figure \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003eA\u0026ndash;C; Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e). Among these, the CXCL12\u0026ndash;CXCR4 signaling axis emerged as the most consistently active chemokine pathway across all three samples.\u003c/p\u003e \u003cp\u003eCellChat analysis of the CXCL signaling network identified CXCL12-CXCR4 and related chemokine ligand-receptor pairs among epithelial and immune cell populations across samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA-C; Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e). The current network plots support predicted communication links and relative pathway contributions, but they do not provide a direct quantitative measurement of cell co-localization or ligand-receptor co-expression at single-cell resolution. Therefore, these results are interpreted as evidence that CXCL signaling is spatially associated with epithelial-immune communication during adenomyotic invasion.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn sample A1, both ciliated and non-ciliated epithelial cells in non-invasive regions were predicted to receive CXCL12 signals from immune cells, while epithelial cells in invasive regions acted as CXCL12 signal sources (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Similar interaction patterns were observed in samples A7 and A12, although the relative contributions of specific immune cell types varied between samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB\u0026ndash;C).\u003c/p\u003e \u003cp\u003eIn addition to CXCL12\u0026ndash;CXCR4, CX3CL1\u0026ndash;CX3CR1 signaling was detected in sample A1 (Figure \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003eD; Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e), indicating sample-specific chemokine interactions. Collectively, these results suggest that chemokine signaling, particularly through the CXCL12\u0026ndash;CXCR4 axis, is spatially associated with epithelial cell migration during adenomyosis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Invaded epithelial cells activate stromal cells through PDGF and IGF signaling\u003c/h2\u003e \u003cp\u003eTo examine how epithelial cells influence the surrounding myometrial microenvironment following invasion, epithelial cells within invasive regions were designated as signal-sending populations for further cell\u0026ndash;cell communication analysis. Multiple growth factor\u0026ndash;related signaling pathways were identified, including fibroblast growth factor (FGF), PDGF, IGF, and transforming growth factor-β (TGF-β) pathways (Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAmong these, PDGF and IGF signaling pathways were consistently active across all three samples. Spatial interaction analysis revealed that epithelial cells in invasive regions expressed PDGF ligands that interacted with receptor-expressing smooth muscle cells, pericytes, and fibroblasts (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA\u0026ndash;C). In sample A1, both ciliated and non-ciliated epithelial cells contributed to PDGF signaling toward smooth muscle cells and pericytes, whereas in samples A7 and A12, PDGF signaling was predominantly driven by ciliated epithelial cells targeting fibroblasts.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eSimilarly, IGF signaling analysis showed that epithelial cells in invasive regions acted as major ligand sources, with predicted interactions involving smooth muscle cells, fibroblasts, pericytes, and immune cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD\u0026ndash;F). While IGF signaling in sample A1 targeted multiple stromal and immune cell types, signaling in samples A7 and A12 primarily involved interactions with smooth muscle cells and fibroblasts.\u003c/p\u003e \u003cp\u003eIn addition, FGF and TGF-β signaling pathways were detected in subsets of samples (Figure \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003eA\u0026ndash;F), suggesting supplementary growth factor signaling interactions in specific tissue contexts. Overall, these results indicate that epithelial cells that have invaded the myometrium exhibit active growth factor signaling toward surrounding stromal cells.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e3.6 Spatial heterogeneity reveals potential disease-associated epithelial subpopulations\u003c/h2\u003e \u003cp\u003eExploratory spatial analysis suggested epithelial heterogeneity at the invasive front, including small epithelial subsets with features that may warrant further investigation in relation to progenitor-like states. Because LGR5 is commonly interpreted as a stem/progenitor-associated marker and the present figure set does not include dedicated validation of LGR5-positive epithelial cells, this observation should be regarded as hypothesis-generating and requires confirmation by targeted spatial visualization or immunostaining.\u003c/p\u003e \u003c/div\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eIn this study, we integrated high-resolution spatial transcriptomics with single-cell transcriptomic references to delineate the cellular architecture and signaling networks associated with adenomyosis progression(\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Recent single-cell studies have revealed epithelial, stromal, immune, and vascular heterogeneity in adenomyosis, including paracrine epithelial-stromal interactions such as WNT/SFRP signaling(\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e), while emerging spatial transcriptomic studies have described ectopic endometrial penetration and lesion-associated spatial organization(\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). Extending these observations, our analysis directly compared invasive and adjacent non-invasive regions within intact tissue sections and identified coordinated epithelial adhesion programs, chemokine-associated epithelial-immune communication, and growth factor-related epithelial-stromal interactions as spatial features of the invasive front.\u003c/p\u003e \u003cp\u003eA central finding of our study is the dynamic regulation of epithelial adhesion programs during invasion(\u003cspan additionalcitationids=\"CR26\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). Both ciliated and non-ciliated epithelial cells exhibited a transition from a low-adhesion state in non-invasive regions to a re-anchored state within the myometrium, characterized by upregulation of focal adhesion and cell\u0026ndash;cell junction pathways(\u003cspan additionalcitationids=\"CR26\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). This \u0026ldquo;detachment\u0026ndash;reattachment\u0026rdquo; behavior reflects a form of epithelial plasticity that enables migration while preserving epithelial identity(\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). Importantly, unlike malignant invasion, epithelial cells in adenomyosis do not undergo irreversible dedifferentiation or sustained loss of junctional structures(\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). Instead, adhesion remodeling appears to be tightly regulated and reversible, consistent with invasion occurring within a benign but mechanically and hormonally dynamic tissue environment(26,28).\u003c/p\u003e \u003cp\u003eOur spatial analysis further implicates the CXCL12-CXCR4 axis in epithelial-immune communication at the invasive front(\u003cspan additionalcitationids=\"CR22\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). Previous work has linked CXCL12-CXCR4 signaling to immune-cell recruitment, fibrosis, and tissue remodeling in multiple disease contexts(\u003cspan additionalcitationids=\"CR22 CR23\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e), but its spatial relationship to adenomyotic epithelial invasion has been less clearly defined. In our data, predicted CXCL signaling networks involved epithelial and immune cell populations across all three samples, suggesting that chemokine-mediated communication may contribute to the microenvironmental context in which epithelial cells invade the myometrium. However, because these findings are based on transcriptomic ligand-receptor inference, they should be interpreted as candidate interaction networks rather than direct proof of chemokine gradients or directed migration.\u003c/p\u003e \u003cp\u003eBeyond migration-associated signaling, our findings suggest that invaded epithelial cells participate in stromal remodeling through growth factor-related communication(\u003cspan additionalcitationids=\"CR32 CR33 CR34\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e). Spatial cell-cell communication analysis predicted PDGF and IGF signaling from epithelial cells in invasive regions toward smooth muscle cells, fibroblasts, and pericytes(\u003cspan additionalcitationids=\"CR32 CR33 CR34\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e). These results are consistent with prior evidence that growth factor pathways regulate proliferation, extracellular matrix remodeling, and fibrosis(\u003cspan additionalcitationids=\"CR32 CR33 CR34 CR35\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e), and they provide a spatially resolved candidate mechanism linking ectopic epithelial compartments to myometrial hyperplasia and fibrosis. Nevertheless, this epithelial-stromal model should be tested experimentally, because transcriptomic inference cannot determine whether PDGF or IGF signaling is sufficient to induce stromal activation in adenomyosis.\u003c/p\u003e \u003cp\u003eThe spatial integration of adhesion remodeling, chemokine-guided migration, and growth factor\u0026ndash;mediated stromal activation provides a unified framework for understanding adenomyosis as a coordinated tissue-level process(\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). Importantly, these signaling axes appear to operate at distinct but interconnected stages of disease development(\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). Chemokine signaling through CXCL12\u0026ndash;CXCR4 is closely associated with the initiation and directionality of epithelial migration, whereas PDGF and IGF signaling likely contribute to lesion stabilization and amplification through sustained stromal activation(\u003cspan additionalcitationids=\"CR32 CR33 CR34 CR35\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e). Within this framework, recently identified targets such as the prolactin receptor may intersect with or modulate these spatial programs, underscoring the multifactorial nature of adenomyosis pathogenesis and the potential need for combinatorial therapeutic strategies(\u003cspan additionalcitationids=\"CR38\" citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSeveral limitations of this study should be acknowledged. First, the spatial transcriptomic cohort included three specimens selected after pathological review and RNA quality control. This sample size is appropriate for exploratory high-resolution spatial profiling but limits statistical power, the ability to model inter-patient heterogeneity, and the generalizability of region-level comparisons. Second, ligand-receptor interactions inferred from CellChat represent probabilistic predictions based on transcript abundance rather than direct biochemical measurements of protein binding, secretion, or receptor activation(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). Third, the cross-sectional nature of human tissue sampling limits the ability to reconstruct temporal sequences with certainty, and inferred invasion stages are based on spatial context rather than longitudinal observation(\u003cspan additionalcitationids=\"CR15 CR16\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). Finally, the study did not include independent experimental validation such as multiplex immunostaining, RNAscope, organoid/assembloid perturbation, or in vivo functional assays. Future studies with larger clinically annotated cohorts and targeted validation experiments will be essential to determine the causal roles and therapeutic relevance of the candidate CXCL12-CXCR4 and PDGF/IGF signaling axes(\u003cspan additionalcitationids=\"CR38\" citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn summary, our study defines adenomyosis progression as a spatially organized process associated with epithelial plasticity, epithelial-immune communication, and epithelial-stromal remodeling(\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). By identifying CXCL12-CXCR4 and PDGF/IGF signaling as candidate components of this invasion-remodeling axis, we provide a spatially informed framework that links cellular behavior with tissue-scale pathology and offers a basis for future mechanistic and therapeutic studies(\u003cspan additionalcitationids=\"CR38\" citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e).\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eConflict of Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCaihong Yang: Conceptualization, Methodology, Supervision,Formal analysis.\u003c/p\u003e\n\u003cp\u003eHui Dong: Data curation, Visualization,software.\u003c/p\u003e\n\u003cp\u003eLulu Shi:Funding acquisition.\u003c/p\u003e\n\u003cp\u003eWenli Gu,Dan Yang:Writing \u0026ndash; original draft , Writing \u0026ndash; review \u0026amp; editing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by the Natural Science Foundation of Ningxia (Grant No. 2024AAC03628).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank all patients who participated in this study. We acknowledge BGI-STOmics (Shenzhen, China) for providing Stereo-seq technical support and Wuhan Ru Biotechnology Co., Ltd. for assistance with data analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe raw sequence data reported in this paper have been deposited in the Genome Sequence Archive (Genomics, Proteomics \u0026amp; Bioinformatics 2025) in the National Genomics Data Center (Nucleic Acids Res 2025), China National Center for Bioinformation / Beijing Institute of Genomics, Chinese Academy of Sciences, under BioProject accession number \u003cstrong\u003ePRJCA057653\u003c/strong\u003e (sample/experiment accession: \u003cstrong\u003eHRA016686\u003c/strong\u003e). The data are publicly accessible at https://ngdc.cncb.ac.cn/gsa-human as of February 10, 2026.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eChapron, C. et al. Diagnosing adenomyosis: an integrated clinical and imaging approach. \u003cem\u003eHum. Reprod. Update 15 avr\u003c/em\u003e. \u003cb\u003e26\u003c/b\u003e (3), 392\u0026ndash;411 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBourdon, M. et al. Adenomyosis: An update regarding its diagnosis and clinical features. \u003cem\u003eJ. Gynecol. Obstet. Hum. Reprod. d\u0026eacute;c\u003c/em\u003e. \u003cb\u003e50\u003c/b\u003e (10), 102228 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhan, K. N., Fujishita, A. \u0026amp; Mori, T. Pathogenesis of Human Adenomyosis: Current Understanding and Its Association with Infertility. \u003cem\u003eJCM 13 juill\u003c/em\u003e. \u003cb\u003e11\u003c/b\u003e (14), 4057 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVannuccini, S. \u0026amp; Petraglia, F. Recent advances in understanding and managing adenomyosis. \u003cem\u003eF1000Res 13 mars\u003c/em\u003e. \u003cb\u003e8\u003c/b\u003e, 283 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSantulli, P., Vannuccini, S., Bourdon, M., Chapron, C. \u0026amp; Petraglia, F. Adenomyosis: the missed disease. \u003cem\u003eReproductive Biomed. Online avr\u003c/em\u003e. \u003cb\u003e50\u003c/b\u003e (4), 104837 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiudice, L. C., Liu, B. \u0026amp; Irwin, J. C. Endometriosis and adenomyosis unveiled through single-cell glasses. \u003cem\u003eAm. J. Obstet. Gynecol. avr\u003c/em\u003e. \u003cb\u003e232\u003c/b\u003e (4), S105\u0026ndash;S123 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBulun, S. E., Yildiz, S., Adli, M. \u0026amp; Wei, J. J. Adenomyosis pathogenesis: insights from next-generation sequencing. \u003cem\u003eHum. Reprod. Update\u003c/em\u003e. \u003cb\u003e18\u003c/b\u003e (6), 1086\u0026ndash;1097 (oct 2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhai, J., Vannuccini, S., Petraglia, F. \u0026amp; Giudice, L. C. Adenomyosis: Mechanisms and Pathogenesis. \u003cem\u003eSemin Reprod. Med. mai\u003c/em\u003e. \u003cb\u003e38\u003c/b\u003e (02/03), 129\u0026ndash;143 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYildiz, S. et al. Adenomyosis: single-cell transcriptomic analysis reveals a paracrine mesenchymal\u0026ndash;epithelial interaction involving the WNT/SFRP pathway. \u003cem\u003eFertility Steril. mai\u003c/em\u003e. \u003cb\u003e119\u003c/b\u003e (5), 869\u0026ndash;882 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, B. et al. From Invaginating Site to Deep Lesion: Spatial Transcriptomics Unravels Ectopic Endometrial Penetration Features in Adenomyosis. \u003cem\u003eAdv. Sci. mai\u003c/em\u003e. \u003cb\u003e12\u003c/b\u003e (20), 2411752 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang, X. et al. Spatial and single-cell transcriptomics landscape of adenomyosis. \u003cem\u003eJ. Adv. Res. d\u0026eacute;c\u003c/em\u003e 2025;S209012322501032X.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, R. et al. Single-cell RNA sequencing identifies the prolactin receptor as a therapeutic target in adenomyosis. \u003cem\u003eSig Transduct. Target. Ther. 13 ao\u0026ucirc;t\u003c/em\u003e. \u003cb\u003e10\u003c/b\u003e (1), 258 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, Y. et al. Signaling pathways and advances in targeted therapy for adenomyosis. \u003cem\u003eFront. Cell. Dev. Biol. 6 nov\u003c/em\u003e. \u003cb\u003e13\u003c/b\u003e, 1685525 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTian, L., Chen, F. \u0026amp; Macosko, E. Z. The expanding vistas of spatial transcriptomics. \u003cem\u003eNat. Biotechnol. juin\u003c/em\u003e. \u003cb\u003e41\u003c/b\u003e (6), 773\u0026ndash;782 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, T. Y., You, L., Hardillo, J. A. U. \u0026amp; Chien, M. P. \u003cem\u003eSpat. Transcriptomic Technol. Cells 10 ao\u0026ucirc;t\u003c/em\u003e ;\u003cb\u003e12\u003c/b\u003e(16):2042. (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKleshchevnikov, V. et al. Cell2location maps fine-grained cell types in spatial transcriptomics. \u003cem\u003eNat. Biotechnol. mai\u003c/em\u003e. \u003cb\u003e40\u003c/b\u003e (5), 661\u0026ndash;671 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJin, S. et al. Inference and analysis of cell-cell communication using CellChat. \u003cem\u003eNat. Commun. 17 f\u0026eacute;vr\u003c/em\u003e. \u003cb\u003e12\u003c/b\u003e (1), 1088 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, A. et al. Spatiotemporal transcriptomic atlas of mouse organogenesis using DNA nanoball-patterned arrays. \u003cem\u003eCell. mai\u003c/em\u003e. \u003cb\u003e185\u003c/b\u003e (10), 1777\u0026ndash;1792e21 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWahafu, W. et al. Spatial relationships and interactions of immune cell niches are linked to the pathologic response of muscle-invasive bladder cancer to neoadjuvant therapy. J Transl Med. 27 mars. ;23(1):375. (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFang, S. et al. Stereopy: modeling comparative and spatiotemporal cellular heterogeneity via multi-sample spatial transcriptomics. \u003cem\u003eNat. Commun. 21 avr\u003c/em\u003e. \u003cb\u003e16\u003c/b\u003e (1), 3741 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu, X. et al. CXCL12/CXCR4: An amazing challenge and opportunity in the fight against fibrosis. \u003cem\u003eAgeing Res. Reviews janv\u003c/em\u003e. \u003cb\u003e83\u003c/b\u003e, 101809 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDaniel, S. K., Seo, Y. D. \u0026amp; Pillarisetty, V. G. The CXCL12-CXCR4/CXCR7 axis as a mechanism of immune resistance in gastrointestinal malignancies. \u003cem\u003eSeminars Cancer Biology oct.\u003c/em\u003e \u003cb\u003e65\u003c/b\u003e, 176\u0026ndash;188 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, C., Gao, Y., Ding, P., Wu, T. \u0026amp; Ji, G. The role of CXCL family members in different diseases. \u003cem\u003eCell. Death Discov 1 juill\u003c/em\u003e. \u003cb\u003e9\u003c/b\u003e (1), 212 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKohli, K., Pillarisetty, V. G. \u0026amp; Kim, T. S. Key chemokines direct migration of immune cells in solid tumors. \u003cem\u003eCancer Gene Ther. janv\u003c/em\u003e. \u003cb\u003e29\u003c/b\u003e (1), 10\u0026ndash;21 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCamp\u0026agrave;s, O., Noordstra, I. \u0026amp; Yap, A. S. Adherens junctions as molecular regulators of emergent tissue mechanics. \u003cem\u003eNat. Rev. Mol. Cell. Biol. avr\u003c/em\u003e. \u003cb\u003e25\u003c/b\u003e (4), 252\u0026ndash;269 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHirano, S., Aoki, K. \u0026amp; Ueno, N. Dynamic behavior of cell-cell adhesion factors in collective cell migration. \u003cem\u003eCells Dev. d\u0026eacute;c\u003c/em\u003e. \u003cb\u003e184\u003c/b\u003e, 203995 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePourjafar, M. \u0026amp; Tiwari, V. K. Plasticity in cell migration modes across development, physiology, and disease. \u003cem\u003eFront. Cell. Dev. Biol. 23 avr\u003c/em\u003e. \u003cb\u003e12\u003c/b\u003e, 1363361 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun, X., Joost, S. \u0026amp; Kasper, M. Plasticity of Epithelial Cells during Skin Wound Healing. \u003cem\u003eCold Spring Harb Perspect. Biol. mai\u003c/em\u003e. \u003cb\u003e15\u003c/b\u003e (5), a041232 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBijonowski, B. M. et al. Intercellular adhesion boots collective cell migration through elevated membrane tension. \u003cem\u003eNat. Commun. 12 f\u0026eacute;vr\u003c/em\u003e. \u003cb\u003e16\u003c/b\u003e (1), 1588 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarg, P., Jallepalli, V. R. \u0026amp; Verma, S. Unravelling the CXCL12/CXCR4 Axis in breast cancer: Insights into metastasis, microenvironment interactions, and therapeutic opportunities. \u003cem\u003eHum. Gene mai\u003c/em\u003e. \u003cb\u003e40\u003c/b\u003e, 201272 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao, M. et al. Targeting fibrosis: mechanisms and clinical trials. Sig Transduct Target Ther. 30 juin. ;7(1):206. (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKu, J. C., Raiten, J. \u0026amp; Li, Y. Understanding fibrosis: Mechanisms, clinical implications, current therapies, and prospects for future interventions. \u003cem\u003eBiomedical Eng. Adv. juin\u003c/em\u003e. \u003cb\u003e7\u003c/b\u003e, 100118 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIrma, J. et al. From Growth Factors to Structure: PDGF and TGF-β in Granulation Tissue Formation. A Literature Review. \u003cem\u003eJ. Cell. Mol. Medi juin\u003c/em\u003e. \u003cb\u003e29\u003c/b\u003e (11), e70374 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhan, M. Z., Zugaza, J. L. \u0026amp; Torres Aleman, I. The signaling landscape of insulin-like growth factor 1. \u003cem\u003eJ. Biol. Chem. janv\u003c/em\u003e. \u003cb\u003e301\u003c/b\u003e (1), 108047 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSechrist, Z. R. et al. Pathologic Signaling and Disease Implications of Insulin-like Growth Factor Binding Proteins in Cancer, Cardiovascular Disease, and Fibrosis. \u003cem\u003eIJMS\u003c/em\u003e \u003cb\u003e22\u003c/b\u003e (21), 10248 (oct 2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDi, X., Li, Y., Wei, J., Li, T. \u0026amp; Liao, B. Targeting Fibrosis: From Molecular Mechanisms to Advanced Therapies. \u003cem\u003eAdv. Sci. janv\u003c/em\u003e. \u003cb\u003e12\u003c/b\u003e (3), 2410416 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRawlings, T. M., Makwana, K., Tryfonos, M. \u0026amp; Lucas, E. S. Organoids to model the endometrium: implantation and beyond. \u003cem\u003eReprod. Fertility 8 sept\u003c/em\u003e. \u003cb\u003e2\u003c/b\u003e (3), R85\u0026ndash;101 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStratopoulou, C. A. et al. Generation of epithelial-stromal assembloids as an advanced in vitro model of impaired adenomyosis-related endometrial receptivity. \u003cem\u003eFertility Steril. f\u0026eacute;vr\u003c/em\u003e. \u003cb\u003e123\u003c/b\u003e (2), 350\u0026ndash;360 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu, Y. et al. Decoding adenomyosis pathogenesis using an assembloid model. \u003cem\u003eSci. China Life Sci. janv\u003c/em\u003e. \u003cb\u003e69\u003c/b\u003e (1), 136\u0026ndash;150 (2026).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Adenomyosis, spatial transcriptomics, epithelial invasion, CXCL12–CXCR4 axis, PDGF signaling, prolactin receptor, cell–cell communication","lastPublishedDoi":"10.21203/rs.3.rs-9819300/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9819300/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground:\u003c/h2\u003e \u003cp\u003eAdenomyosis is characterized by ectopic endometrial tissue invading the myometrium, yet the spatiotemporal dynamics and molecular programs governing epithelial invasion and subsequent tissue remodeling remain poorly understood.\u003c/p\u003e\u003ch2\u003eMethods:\u003c/h2\u003e \u003cp\u003eHigh-resolution spatial transcriptomic profiling (Stereo-seq) was performed on three pathologically confirmed human adenomyosis specimens. Spatial data were integrated with publicly available single-cell RNA sequencing datasets. Cell type composition was inferred using Cell2location, while differential gene expression and cell\u0026ndash;cell communication networks were analyzed to characterize transcriptional programs and signaling pathways at the invasive front.\u003c/p\u003e\u003ch2\u003eResults:\u003c/h2\u003e \u003cp\u003eSpatial transcriptomic mapping showed region-associated changes in epithelial adhesion programs during adenomyotic invasion, with ciliated and non-ciliated epithelial cells displaying enrichment of focal adhesion and cell-junction pathways in invasive regions. Cell-cell communication analysis further identified CXCL12-CXCR4 chemokine signaling and PDGF/IGF-related growth factor signaling as spatially associated interaction networks involving epithelial, immune, and stromal cell populations. These findings suggest coordinated epithelial-stromal crosstalk at the invasive front, while functional validation is required to establish causal roles for these pathways.\u003c/p\u003e\u003ch2\u003eConclusions:\u003c/h2\u003e \u003cp\u003eThis study identifies the CXCL12\u0026ndash;CXCR4 and PDGF/IGF signaling axes as central spatial regulators of epithelial invasion and stromal remodeling in adenomyosis. By providing a spatiotemporal framework for epithelial\u0026ndash;stromal crosstalk, these findings advance mechanistic understanding of adenomyosis pathogenesis and offer molecular insights that may inform emerging disease models and targeted therapeutic strategies, including recently identified prolactin receptor\u0026ndash;based interventions.\u003c/p\u003e","manuscriptTitle":"Spatial transcriptomics reveals coordinated epithelial invasion and stromal remodeling in adenomyosis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-08-26 14:27:53","doi":"10.21203/rs.3.rs-9819300/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"24040887931247321056772173954598410496","date":"2026-08-21T05:23:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"280667353462875249536906733512698570175","date":"2026-08-21T04:38:28+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-08-18T23:44:40+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-06-22T15:50:46+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-06-05T12:26:08+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-06-03T13:00:25+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-06-03T09:04:48+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"40d68522-2a07-4857-b5aa-f2709a3ee545","owner":[],"postedDate":"August 26th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"24040887931247321056772173954598410496","date":"2026-08-21T05:23:19+00:00","index":103,"fulltext":""},{"type":"reviewerAgreed","content":"280667353462875249536906733512698570175","date":"2026-08-21T04:38:28+00:00","index":102,"fulltext":""},{"type":"reviewersInvited","content":"20","date":"2026-08-18T23:44:40+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":74689805,"name":"Biological sciences/Cell biology"},{"id":74689806,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":74689807,"name":"Health sciences/Diseases"},{"id":74689808,"name":"Biological sciences/Molecular biology"}],"tags":[],"updatedAt":"2026-08-26T14:27:53+00:00","versionOfRecord":[],"versionCreatedAt":"2026-08-26 14:27:53","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9819300","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9819300","identity":"rs-9819300","version":["v1"]},"buildId":"cTy_lsJlmDsVRNrSptgXS","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}