Spatiotemporal Decoding of Tumor Contactome Unveils a Microbial-Sensitive Ctrb1-Dependent Cell Competition Architecture Driving Epithelial-Mesenchymal Defense Against Cancer

Spatiotemporal Decoding of Tumor Contactome Unveils a Microbial-Sensitive Ctrb1-Dependent Cell Competition Architecture Driving Epithelial-Mesenchymal Defense Against Cancer | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Spatiotemporal Decoding of Tumor Contactome Unveils a Microbial-Sensitive Ctrb1-Dependent Cell Competition Architecture Driving Epithelial-Mesenchymal Defense Against Cancer Zhaocai Zhou, Zhangting Zhao, Fan Chen, Yang Meng, Hui Zhang, and 10 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6524135/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract While non-immune epithelial surveillance constitutes a crucial frontline defense against early tumorigenesis, its operational mechanisms - particularly the cellular components and structural basis of cell competition - remain poorly defined. In this study, we uncover a novel cell competition architecture which we termed as proliferative containment layer (PCL), comprising coordinated epithelial-mesenchymal cell assemblies. Employing our newly developed Surface-anchored TurboID-mediated Contact-dependent Labeling (STCL) system, we achieved in situ biotinylation of membrane proteins on interacting cells, enabling the first dynamic mapping of direct tumor-stromal interactomes in a tumor-suppressive competition model. Mechanistically, direct tumor cell contact triggers a unique epithelial-mesenchymal defense against cancer (EMDAC) program, where PCL components acquire superior proliferative capacity through chymotrypsinogen B1 (CTRB1)-mediated coordination of Myc/YAP signaling axes. Notably, we demonstrate that CTRB1 serves as a molecular rheostat integrating multiple proliferation pathways to establish competitive dominance, while microbial infection unexpectedly suppresses CTRB1 expression and compromises tumor clearance. This work fundamentally advances our understanding of non-immune surveillance by: 1) Identifying PCL as a spatially organized defense unit; 2) Deciphering EMDAC as a dual-lineage competition mechanism; 3) Establishing CTRB1 as a microbial-sensitive master regulator of cellular fitness during competition. Biological sciences/Cancer/Tumour-suppressor proteins Biological sciences/Cell biology/Cell growth/HIPPO signalling Biological sciences/Cancer/Cancer microenvironment Biological sciences/Biotechnology/Sequencing/RNA sequencing Biological sciences/Microbiology/Microbial communities/Microbiome Non-immune surveillance Cell competition PCL EMDAC CTRB1 Microbial infection Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction The fate of tumor initiation—whether progressing to malignancy or being eradicated by host defenses—hinges on the dynamic equilibrium between oncogenic transformation and cellular surveillance mechanisms. While immune-mediated tumor surveillance has been extensively characterized, the functional contributions of non-immune cellular components, particularly epithelial and mesenchymal cells, remain enigmatic. Seminal studies have revealed that normal epithelia possess intrinsic tumor-suppressive capacity through neighbor elimination programs, exemplified by epithelial defense against cancer (EDAC)(Ayukawa et al., 2021 ; Tanimura and Fujita, 2020 ). However, whether mesenchymal lineages (e.g., fibroblasts, smooth muscle cells) participate in analogous surveillance programs, and how these cellular populations coordinate to establish tumor-suppressive microenvironments, constitutes a critical knowledge gap. Crucially, the spatiotemporal dynamics governing tumor-stromal interactions during early tumorigenesis, particularly the molecular logic determining competitive outcomes between malignant and host cells, await systematic elucidation. Cell competition—an evolutionarily conserved quality control mechanism—represents a fundamental mechanism for host non-immune surveillance against transformed cell(van Neerven and Vermeulen, 2023 ). This process enables supercompetitor cells to eliminate less-fit neighbors through direct interactions, playing context-dependent roles in both tumor suppression ( e.g. , epithelial homeostasis) and promotion ( e.g. , clonal selection)(Parker et al., 2021 ; van Neerven and Vermeulen, 2023 ). While oncogenic pathway rewiring (Wnt, Hippo, Notch) has been implicated in competitive interactions in a framework of EDAC(Moya et al., 2019 ; Nakai et al., 2023 ; van Neerven and Vermeulen, 2023 ), the structural basis of tumor-host competition remains obscure. Two fundamental questions persist: (1) What spatial architectures mediate competitive interactions between incipient tumor cells and their microenvironment? (2) How do direct cell contacts coordinate multi-lineage (epithelial-mesenchymal) defense programs? Recent methodological advances in proximity labeling (e.g., LIPSTIC(Nakandakari-Higa et al., 2024 ; Pasqual et al., 2018 ), EXCELL(Ge et al., 2019 ; Liu et al., 2022 ), FucoID(Liu et al., 2020 ; Qiu et al., 2022 )) have revolutionized cell-cell interaction mapping. Nevertheless, current tools face limitations in capturing transient physical interactions within complex microenvironments, particularly for non-immune cell types(Roux et al., 2012 ). The recent development of uLIPSTIC(Nakandakari-Higa et al., 2024 ), while enabling immune interaction profiling, lacks generalizability for studying epithelial-mesenchymal crosstalk. These technical constraints have hindered comprehensive analysis of direct interactomes governing tumor-stromal competition in vivo . Here, through an orthotopic gastric cancer model, we identify a proliferating containment layer (PCL)—a spatially organized epithelial-mesenchymal unit—that orchestrates tumor-suppressive competition. To decipher PCL dynamics, we developed surface-anchored TurboID-mediated contact-dependent labeling (STCL), which integrated with single cell RNA sequence (scRNA-Seq), enables in vivo mapping of direct cellular interactomes in tumor microenvironment (TME). This approach revealed a new type of non-immune surveillance i.e. , epithelial-mesenchymal defense against cancer (EMDAC), a dual-lineage mechanism in which PCL constituents outcompete tumor cells through chymotrypsinogen B1 (Ctrb1)-mediated coordination of Myc/YAP signaling. Intriguingly, microbial infection subverts this surveillance machinery by suppressing Ctrb1 expression. Our findings (1) establish PCL as a structural framework for tumor-host competition; (2) define EMDAC as a non-immune surveillance paradigm; and (3) identify Ctrb1 as a microbial-sensitive rheostat of competitive fitness. Results I. Revealing Spatiotemporal Architecture of A Proliferative Containment Layer for EMDAC To delineate non-immune surveillance mechanisms against tumorigenesis, we established an orthotopic gastric cancer model in which GFP-labeled MFC tumors underwent host-mediated clearance by D15 (day 15, same below) (Fig. 1 a, Extended Data Fig. 1 a). Spatial profiling revealed a dynamically expanding GFP − Ki67 + p roliferative c ontainment l ayer (PCL) surrounding regressing tumors (Fig. 1 b). Quantification demonstrated PCL's proliferative dominance: 59.3% (95% CI, 55.7 ~ 63.5%) Ki67 + cells in PCL zone vs. 26.9% (95% CI, 22.5 ~ 32.7%) in tumor zone at D10 (p = 0.0000, Extended Data Fig. 1 b), with progressive width expansion from D5 148.9 µm (95% CI, 140.6 ~ 169.8 µm) to D10 472.8 µm (95% CI, 391.3 ~ 585.6 µm, p = 0.0000, Fig. 1 c). These results suggest that PCL may represent a spatial barrier constraining tumor progression. Subsequently, our lineage-specific profiling revealed coordinated proliferation of epithelial and mesenchymal components in PCL. On one hand, the epithelial proliferation dynamics was shown by that CK8 + Ki67 + cells dominated PCL regions 82.1 ± 1.5% vs. 15.6 ± 1.0% in adjacent epithelia at D10, (p = 0.0000), with 2.5-fold area expansion from D5 to D10 (Fig. 1 d-e). On the other, the mesenchymal recruitment was reflected by that αSMA + Ki67 + fibroblasts occupied 79.0 ± 2.4% of PCL vs. 14.2 ± 1.7% in muscle layer at D10 (p = 0.0000), showing 5.5-fold spatial expansion by D10 (Fig. 1 f-g). These results suggest dual-lineage contribution to PCL expansion. To further characterize the cellular origin of the PCL, we utilized an inducible reporter system to genetically trace epithelial cells and fibroblasts in mice bearing MFC-derived tumors. We generated several Cre-strain Rosa26-LSL-RFP reporter models: Claudin18.2 Cre/ERT2 ; R26 RFP for gastric epithelial cells, αSMA Cre ; R26 RFP for smooth muscle-derived fibroblasts, and Col1a2 Cre/ERT2 ; R26 RFP for total fibroblast populations. Tamoxifen-induced RFP expression allowed precise tracking of these cell types during tumor formation. The results showed that Claudin18.2 Cre/ERT2 -traced cells increased from 17.2 ± 0.9% (D5) to 25.1 ± 2.9% (D10, p = 0.0034), among which the Ki67 + RFP + cells augmented from 0.88 ± 0.16% (D5) to 11.6 ± 0.5% (D10, p = 0.0000, Fig. 1 h and Extended Data Fig. 1 c), confirming the epithelial contribution to PCL structure. Meanwhile, αSMA Cre -traced proliferating fibroblasts (Ki67 + RFP + cells) dominated PCL (59.0 ± 4.3% at D10, Fig. 1 i and Extended Data Fig. 1 d), confirming the mesenchymal recruitment. Col1a2 Cre/ERT2 ; R26 RFP mice further confirmed this pattern, showing 60.1 ± 3.7% of PCL cells with an origin of fibroblasts ( Extended Data Fig. 1 e). These findings confirm that both epithelial and mesenchymal cell types contribute significantly to PCL formation. Together, these data indicate that spatiotemporal coordination between epithelial proliferation and stromal activation underpins PCL-mediated tumor containment, a process we termed as e pithelial- m esenchymal d efense a gainst c ancer (EMDAC) (Fig. 1 j). II. Myc and YAP Signaling Cooperatively Drives EMDAC To investigate the cell competition mechanisms associated with EMDAC, we performed single-cell RNA sequencing (scRNA-seq) and spatial transcriptomics on samples collected at D5, D10, and D15 during tumor clearance. These analyses revealed distinct cell types and their spatial distribution (Fig. 2 a, left panel). We observed a layered architecture comprising epithelial (EP), tumor zone, PCL, and mesenchymal (ME) layers early after MFC injection at D5 (Fig. 2 a, right panel). The EP layer primarily consisted of normal epithelium (Epcam + ) and metaplastic epithelium (Clu + ), while the ME layer predominantly included smooth muscle cell-derived fibroblasts ( Extended Data Fig. 2 a). PCL exhibited a high density of epithelial cells ( e.g. , pit cells with Cldn18 + ) and fibroblasts at D5, likely due to their reprogramming in response to tumor cells (Fig. 2 a, b). By D10 and D15, the PCL expanded toward the tumor core (Fig. 2 a). Quantification of proliferating cells showed a stronger Ki67 + index on PCL at D10 compared to D5 or D15 (Fig. 2 c), suggesting its significant contribution to EMDAC. The ME layer exhibited consistent proliferation capacity throughout the process (Fig. 2 c), aligning with its predominant role in the PCL structure. To validate that host cell proliferation is essential for EMDAC, we pretreated mice with 5-FU two days before tumor injection. We observed marked outgrowth of MFC tumors in 5-FU-pretreated mice compared to untreated controls (p = 0.0070, Extended Data Fig. 2 b). At D15, T cells (CD3e + ) and myeloid cells (Kit + ) predominantly located at the tumor periphery ( Extended Data Fig. 2 c), which was corroborated by immunostaining for F4/80 (macrophage) and CD4 (T cell) in tumor tissues from D10 ( Extended Data Fig. 2 d). These findings demonstrate that EMDAC occurs without reliance on the immune system as a primary anti-cancer defense mechanism. Next, we systematically analyzed the activation status of key signaling pathways required for cell proliferation (Myc, TP53, Kras, Hippo-Yap, Notch, and Wnt) across different cell layers during EMDAC. Our results revealed that Myc signaling in the epithelial layer (referred to as Myc EP ) and Yap signaling in the mesenchymal layer (Yap ME ) were specifically and gradually activated during this process (Fig. 2 d and Extended Data Fig. 2 e). Conversely, Notch signaling pathway was predominantly enriched in the tumor zone at D10 and D15, likely due to the clearance of tumor cells from these regions (Fig. 2 d). These findings indicate Myc EP and Yap ME signaling as molecular mechanisms driving EMDAC. Subsequently, we performed surgical separation of the epithelial layer and mesenchymal tissues adjacent to the tumor zone, as well as the tissues near the non-tumor region as controls (Fig. 2 e, upper panel ). We conducted qPCR analysis of downstream targets of Myc ( e.g. , Myc and Cdk4 ) or Yap ( e.g. , Ctgf and Cyr61 ) in these regions. Our findings confirmed that Myc and Cdk4 were upregulated in the epithelial cells adjacent to the tumor region, while Ctgf and Cyr61 were upregulated in the mesenchymal cells, without such activation in their respective control groups (Fig. 2 e, lower panel ). These results demonstrate the spatial specificity of Myc and Yap signaling in driving cell proliferation in the EP and ME layers during EMDAC. To validate whether cell type-specific signaling pathways are required for EMDAC, we pretreated mice with specific inhibitors targeting either Myc (EN4 MYC inhibitor(Han et al., 2019a ), S89228, 50 mg/kg) or Yap (Verteporfin(Liu-Chittenden et al., 2012 ), S80258, 10 mg/kg) for two weeks prior to orthotopic transplantation of MFC cells (Fig. 2 f). As expected, pretreatment with these inhibitors resulted in tumor persistence compared with the control group (Fig. 2 f-h), indicating that Myc and Yap signaling are essential during EMDAC. Furthermore, we generated αSma Cre/ERT2 ; Yap F/+ Taz F/+ genetic mice to specifically reduce Yap and Taz levels in smooth muscle cells prior to MFC orthotopic transplantation (Fig. 2 i). Consistent with the verteporfin treatment results, conditional knockdown of Yap/Taz in smooth muscle cells also led to significant tumor outgrowth (p = 0.0000, Fig. 2 i, j and Extended Data Fig. 2 f) and PCL collapse ( Extended Data Fig. 2 g), confirming that mesenchymal-specific Yap signaling is critical for EMDAC. Together, these findings identify Myc-dependent signaling in epithelial cells and Yap-dependent signaling in mesenchymal cells as essential pathways for EMDAC (Fig. 2 k). III. STCL Technology Deciphers Tumor Cell Direct Interactome during EMDAC Development of STCL as a novel contact-dependent intercellular labeling strategy To delineate direct intercellular communication between neoplastic cells and their epithelial/mesenchymal counterparts during EMDAC, we engineered a contact-dependent Surface-TurboID-mediated interCellular Labeling (STCL) system based on proximity-dependent biotinylation (Fig. 3 a). Building upon TurboID - an engineered biotin ligase optimized through directed evolution(Roux et al., 2012 ) with demonstrated efficacy in mapping organelle contact zones ( e.g. , ER-mitochondria interfaces(Cho et al., 2020 ), astrocyte-neuron synapses(Takano et al., 2020 )) - we hypothesized that cell membrane-tethered outbound TurboID could achieve dual labeling modalities: (1) cis -labeling of autologous membrane proteins, and (2) trans -labeling of interacting partner cells through direct membrane contact. For STCL implementation, we designed a chimeric construct by fusing TurboID to the extracellular end of the transmembrane (TM) domain of human PDGFRβ ( Extended Data Fig. 3 a, upper ). We then stably expressed this PDGFR-TurboID cassette in MFC cells (hereafter referred as MFC STCL , which showed proper membrane localization, as evidenced by 95.8 ± 8.0% colocalization (Pearson's coefficient) between TurboID and WGA (a dye for membrane staining) ( Extended Data Fig. 3 a, lower ). Biotin administration (12h) induced robust membrane-restricted biotinylation (94.3 ± 9.0% WGA overlap), with signal specificity confirmed by three orthogonal detection methods: immunofluorescence (IF, Extended Data Fig. 3 a, lower ), streptavidin western blot (WB, Extended Data Fig. 3 b), and flow cytometry (FC, Extended Data Fig. 3 c). Critically, the biotinylated signals were strictly dependent on TurboID expression and biotin administration, with little background or noise in parental MFC control cells. To validate the capacity of the STCL system in capturing direct intercellular contact events, we established a quantitative co-culture assay with stringent specificity controls. When MFC STCL cells were co-cultured with eFluor670-prelabeled DC2.4 dendritic cells at 1:1 ratio (Fig. 3 b), biotin treatment (12h) induced robust cis -labeling in 84.5 ± 0.2% of MFC STCL cells (n = 3; Extended Data Fig. 3 c). Crucially, flow cytometric quantification revealed significant biotinylation in 54.2 ± 1.0% of DC2.4 cells (vs. 5.0 ± 0.6% in biotin-free controls; p = 0.0000, Student's t-test), demonstrating contact-dependent trans -labeling (Fig. 3 c,d). High-resolution confocal imaging further resolved micron-scale biotinylation foci exclusively at MFC STCL -DC2.4 contact interfaces ( Extended Data Fig. 3 e), with contacting DC2.4 cells showing localized biotin signals. The generalizability of STCL was then confirmed using L929 fibroblasts, where trans -labeling efficiency reached 20.4 ± 0.9% (Fig. 3 d and Extended Data Fig. 3 d), with similar biotinylation foci at MFC STCL -L929 contact interfaces ( Extended Data Fig. 3 f). Importantly, post-co-culture magnetic sorting (purity > 99%) of DC2.4 and L-929 cells eliminated potential contamination from MFC STCL -derived biotin particles. Immunofluorescence of sorted cells confirmed membrane-localized biotinylation patterns (Fig. 3 e). These orthogonal validations exclude artifact from bystander transfer or extracellular vesicle uptake, confirming that STCL specifically labels physically contacted cells. To establish the physiological relevance of STCL, we implemented a multi-model validation strategy spanning subcutaneous and orthotopic tumor systems. In C57/BL6 mice bearing MC38 STCL -derived subcutaneous tumors (n = 3/group), biotin administration (5 mg/kg, i.p., q.d.×5) induced robust cis -labeling evidenced by: (1) strong streptavidin WB and IF signal versus none in MC38 Ctrl tumors (Fig. 3 f,g and Extended Data Fig. 3 g), and (2) High-dimensional cytometry revealed trans -labeling efficiency of CAF, immune cells, and endothelial cells with 8.4 ± 0.3% (vs. 0.8 ± 0.06% in MFC Ctrl TME; p = 0.0000), 42.3 ± 3.4% (vs. 0.8 ± 0.0% in MFC Ctrl TME; p = 0.0000), and 86.1 ± 0.3% (vs. 0.9 ± 0.5% in MFC Ctrl TME; p = 0.0000), respectively (Fig. 3 h), with spatial resolution showing biotin foci precisely aligned with tumor-stromal/immune interfaces (Fig. 3 i and Extended Data Fig. 3 h). Moreover, further cytometry analysis revealed variable trans -labeling efficiency of each type of immune cells, among which B cells showed highest labeling efficiency (~ 80%) ( Extended Data Fig. 3 i-j). Furthermore, we tested STCL in an orthotopic gastric cancer model (n = 6, Fig. 3 j). HE-guided microdissection showed 33.1 ± 3.5% of TME cells exhibited biotin + TurboID − signatures, confirming tumor-specific trans-labeling (Fig. 3 k). Multispectral imaging quantified contact-dependent biotin transfer to αSMA + fibroblasts (13.7 ± 0.8%), CD11b + myeloid cells (35.7 ± 1.8%), and CD3 + T cells (24.9 ± 2.3%) (Fig. 3 i and Extended Data Fig. 3 k). Integrating STCL with scRNA to decipher direct cellular interactome governing EMDAC By integrating the STCL technology with single-cell transcriptomics, we attempted to establish a spatiotemporal map of direct cellular interactomes ( contactome ) governing EMDAC. The study workflow included harvesting whole stomach tissue at D5, D10, and D15, followed by single-cell isolation using magnetic-activated cell sorting to distinguish biotin + (interacting) from biotin − (bystander) cells, which were subsequently analyzed for gene expression profiles (scRNA-seq) (Fig. 4 a). From the total of 35,172 single-cell samples obtained, we identified 10 distinct cell clusters ( Extended Data Fig. 4 a), each annotated using top principal component scores and lineage-specific markers. Quantitative analysis revealed that tumor cells interacted preferentially with certain host cell types during EMDAC progression (Fig. 4 b and Extended Data Fig. 4 b). Specifically, STCL labeling efficiency for epithelial, endothelial, and fibroblast populations showed a consistent increase from D5 to D10, followed by gradual reduction at D15 (Fig. 4 c). Meanwhile, tumor cell labeling efficiency decreased from 85–90% at D5 and D10 to approximately 40% at D15 (Fig. 4 c and Extended Data Fig. 4 c), indicating a decline in cis -labeling as tumor cells increasingly interacted with host epithelial and mesenchymal compartments during EMDAC. Notably, epithelial cells represent a predominant (> 80%) among all trans -labeled cells across the entire timeline of the EMDAC process (Fig. 4 c). Subsequently, we examined the spatial organization of different cell types in the TME at D10 when STCL-based labeling was strong and efficient (Fig. 4 c). A striking observation was the significant increase in CK8 immunofluorescence signals in the PCL (Fig. 4 d). However, the number of biotin + CK8 + cells gradually decreased as the distance from the tumor boundary increased. Importantly, the spatial distribution of these cells mirrored the STCL labeling pattern (Fig. 4 b), with fibroblasts (αSMA + ) and endothelial cells (CD31 + ) predominantly located within the PCL, while myeloid cells (CD11b + ) and T cells (CD4 + ) were more abundant at the periphery (Fig. 4 d). Furthermore, quantitative analysis revealed that tumor cells interacted with various types of host cells through direct physical contact, with frequencies comparable to those determined by scRNA-seq (Fig. 4 d). Notably, epithelial and fibroblast cells emerged as the two most frequently contacted cell types in this context (Fig. 4 e). Together, these findings provide a dynamic snapshot of tumor contactome, i.e. , tumor cell direct interactions with other types of cells within the TME during EMDAC. Tumor contacts activate EMDAC by conferring epithelial cells and fibroblasts proliferative capacity As epithelial cells (EP) and fibroblasts constituted the top-ranked direct interactors with tumor cells, we systematically mapped their interplay. UMAP-based stratification identified 14 EP clusters through cell-type-specific signatures (Fig. 5 a,b), encompassing parietal/pre-parietal, pit, neck/pre-neck, isthmus stem, chief, neuroendocrine, and intestinal metaplastic (IM) lineages. Two emergent populations dominated functional analyses: a hyper-proliferative cluster (Ki67 + Pcna + ; EP_1) and a metabolically active population (Acat1 + Apoa1 + Eno1 + ; EP_2) (Fig. 5 a-c). Systematic annotation consolidated these into eight definitive subpopulations using canonical and computational markers: Tumor, PAC (parietal lineages), Neck/Stem, IM, PIC (pit cells), EP_1, EP_2, and neuroendocrine cells (Fig. 5 a, right panel). Notably, EP_1 and EP_2 emerged at D5 and expanded progressively during EMDAC ( Extended Data 5a ). EP_1 displayed escalating trans -labeling efficiency (Bio + /total) despite declining tumor burden (Fig. 5 d). Pseudotemporal mapping revealed EP_1's origin from pre-neck, neck, Cldn18 + IM, and isthmus stem precursors (Fig. 5 e, Extended Data 5b ). Crucially, these precursors exhibited higher D5 trans -labeling efficiency than EP_1 itself (Fig. 5 f, Extended Data 5c ), indicating that tumor interactions precede hyper-proliferative conversion. We next deconvoluted tumor-proximal (biotin + ) versus bystander (biotin − ) fibroblasts across cell clusters. Integrative marker analysis resolved five CAF states: myCAF (myofibroblast), iCAF (inflammatory fibroblast), apCAF (antigen-presenting fibroblast), and eCAF (epithelial-like CAF expressing Cd24a , Cdh1 , Krt19 , Fxyd3 and Epcam ) ( Extended Data Fig. 5 d,e). Strikingly, a subset of eCAF showed the strongest tumor-derived biotinylation (Bio high eCAF; Fig. 5 g,h), marking their privileged tumor interface in the contactome. Dynamic tracking revealed eCAF trans -labeling peaked during early tumor engagement (D5-D10) but declined post-clearance, while Bio high eCAF sustained persistent tumor interactions (Fig. 5 i). GSVA (gene set variation analysis) of Bio high eCAF demonstrated enriched proliferation pathways (Ki67/Pcna; Extended Data Fig. 5 f, Fig. 5 j), implicating tumor contact as a driver of CAF hyperproliferation. This mechanism was functionally validated by EDU pulse-chase assays ( Extended Data Fig. 5 g): proliferative CAFs directly contacting tumors peaked at D5 but diminished upon tumor eradication at D15 (D15, p = 0.0755), mirroring the Bio high eCAF trajectory. Taken together, these data indicated that physical contacts by tumor cells activate EMDAC by conferring epithelial cells and fibroblasts proliferative capacity. IV. Ctrb1 as a Microbial-Sensitive Surveillance Hub Ctrb1 emerges as a molecular driver for PCL-mediated EMDAC Having established PCL's structural and functional basis for EMDAC, we profiled the gene expression of proliferating cells in the PCL and the identified chymotrypsinogen B1 (Ctrb1) as a key driver of PCL formation (Fig. 5 b and Fig. 6 a). While physiologically restricted to pancreatic tissue ( Extended Data Fig. 6 a,b), Ctrb1 was ectopically expressed in tumor-interacted EP_1 cells and fibroblasts, showing a 2.1-fold enrichment in the biotin + population compared to bystander EP_1 cells and over a 10-fold enrichment compared to bystander fibroblasts by D10 (Fig. 6 a,b). This ectopic expression followed a tumor clearance-associated trajectory: Ctrb1 levels rose sharply from D5, peaked at D10 during maximal PCL expansion (Fig. 6 b,c), and correlated with proliferative activity (30% co-expression with Ki67 in PCL cells; Fig. 6 c and Extended Data Fig. 6 c). Lineage tracing in αSma Cre/ERT2 ; R26 RFP mice confirmed this dynamic – 44.0% of smooth muscle-derived (RFP + ) PCL cells co-expressed Ctrb1 and Ki67 by D10 (Fig. 6 d, Extended Data Fig. 6 e). Importantly, tumor presence induced 4.2- and 3.1-fold upregulation of Ctrb1 transcription in peri-PCL EP and ME compartments compared to tumor-free controls (Fig. 6 e), positioning it as a tumor-responsive effector. The spatiotemporal alignment of Ctrb1 expression with 1) PCL biogenesis (D5-D10), 2) proliferative conversion of stromal cells, and 3) tumor competition dynamics (Fig. 1 b) suggests its dual role as both architectural organizer and mitotic accelerator in EMDAC. To establish causality between tumor contact and Ctrb1 induction, we deployed direct co-culture systems. Gastric epithelial cells (GEC) contacting MFC tumors exhibited 9.1-fold Ctrb1 upregulation versus controls (qPCR/IFA; p = 0.0037; Fig. 6 f-h). Crucially, this response required physical contact, as Transwell-separated co-cultures showed no induction (p = 0.1372, Extended Data Fig. 6 f). Fibroblasts displayed even greater contact-dependent Ctrb1 activation (57.3-fold vs control; 6.3-fold vs GEC) (Fig. 6 , f-h and Extended Data Fig. 6 f), revealing a conserved tumor-sensing mechanism across lineages. Genetic validation in Ctrb1 F/F mice demonstrated its non-redundant role: AAV-mediated stomach-specific Ctrb1 knockout (qPCR/WB; Extended Data Fig. 6 g,h) caused 2.2-fold tumor enlargement (Fig. 6 i,j and Extended Data Fig. 6 i) and PCL collapse ( Extended Data Fig. 6 j). Strikingly, Ctrb1 loss abolished host-mediated tumor clearance, confirming its gatekeeper function in EMDAC (Fig. 6 j). These results demonstrate that cell contact-dependent Ctrb1 activation governs antitumor competence during EMDAC. Building on our observations that epithelial Myc and mesenchymal Yap signaling are indispensable for EMDAC, and that tumor cell contact specifically upregulates Ctrb1 in these stromal compartments, we hypothesized that Ctrb1 might function as a convergence point for these oncogenic pathways. Systematic analysis of the Ctrb1 promoter uncovered conserved Myc-binding E-box elements and Yap/TEAD-binding M-CAT elements (Fig. 6 k). To functionally validate these predictions, we performed chromatin immunoprecipitation (ChIP)-qPCR in contact-activated epithelia and stromal cells. This revealed 10.7-fold (p = 0.0069) and 2.8-fold (p = 0.0157) enrichment of Myc and Yap/Tead4 binding, respectively, at the Ctrb1 promoter compared to non-contact controls (Fig. 6 k). More importantly, conditional Yap/Taz ablation in mesenchymal cells reduced Ctrb1 + Ki67 + proliferating cells by 79.8 ± 6.1% (p = 0.0015) in orthotopic MFC transplantation models (Fig. 6 l, Extended Data Fig. 2 g). Notably, epithelial-specific Myc inhibition produced comparable suppression of Ctrb1 + Ki67 + populations (8.2-fold lower than DMSO group, Extended Data Fig. 6 k). Collectively, these results establish Ctrb1 as a cell contact-responsive effector of tumor-stroma crosstalk, with its transcriptional activation being spatially regulated: epithelial cells employ Myc signaling while mesenchymal counterparts utilize Yap-Tead4 complex to drive EMDAC progression (Fig. 6 m). Ctrb1 boosts PCL cell fitness to outcompete tumor cells during EMDAC Having established the tumor cell contact-dependent induction of Ctrb1 in EMDAC, we next sought to delineate the cell-autonomous mechanisms underlying Ctrb1-mediated fitness enhancement. A 293FT isogenic competition model was established by generating cells stably expressing CTRB1-P2A-GFP (293FT CTRB1 − GFP ; Extended Data Fig. 7 a-b) and co-culturing them 1:1 with parental controls (Fig. 7 a). Strikingly, 293FT CTRB1 − GFP cells progressively dominated the co-culture, achieving 74.35 ± 2.9% dominance by D7 (vs 49.9 ± 5.1% in empty vector controls, p = 0.0000; Fig. 7 b), suggesting a selective growth advantage. Moreover, this fitness phenotype translated directly to in vivo tumorigenic potential. Orthotopically transplanted MFC STCL−Ctrb1 cells ( Extended Data Fig. 7 c) exhibited 4.2-fold larger tumor volumes than wild-type counterparts by D12 (p = 0.0100; Fig. 7 c), accompanied by complete escape from host-mediated tumor elimination (0% regression vs 74.4% in controls, p = 0.0000) (Fig. 7 c and Extended Data Fig. 7 d). To delineate the molecular basis of Ctrb1-mediated cellular fitness, we conducted comparative transcriptomic profiling between isogenic 293FT and CTRB1-overexpressing clones (Fig. 7 d, Extended Data Fig. 7 e). KEGG pathway enrichment revealed CTRB1's global modulation of oncogenic signaling hubs - including Hippo-YAP, WNT/β-catenin, MAPK, mTOR, NF-κB and Notch pathways (Fig. 7 e). Strikingly, hierarchical analysis identified pathway-specific transcriptional amplifiers: Hippo-YAP effectors AJUBA (mechanical stress sensor)(Rauskolb et al., 2014 ) and pro-apoptotic BBC3/PUMA(Subramaniam et al., 2021 ), along with WNT regulators NOTUM (extracellular β-catenin inhibitor) and SERPINF1/PEDF (angiogenesis modulator)(Liu et al., 187), emerged as top-tier CTRB1-responsive nodes (Fig. 7 f, qPCR-validated in Extended Data Fig. 7 f). Notably, our discovery of NOTUM upregulation establishes an evolutionarily conserved axis - as this tumor-derived WNT antagonist has been shown to orchestrate pre-neoplastic niche remodeling through paracrine suppression of competing epithelial clones in intestinal carcinogenesis models(Flanagan et al., 2021a ; Flanagan et al., 2021b ; Pentinmikko et al., 2019 ; van Neerven et al., 2021 ; Yum et al., 2021 ). This mechanistic convergence between CTRB1 modality and established oncogenic programs suggests a novel feedforward signaling mechanism for cellular selection during tumor initiation. Subsequently, we performed affinity purification mass spectrometry (AP-MS) in 293FT cells expressing Flag-CTRB1 ( Extended Data Fig. 7 g-j). Strikingly, the CTRB1 protein complex coalesced core signalosomes across five major pathways - STAT3 (Hippo cross-talk), GSK3β (WNT node), mTOR (metabolic master regulator), NF-κB1 (inflammatory switch), and MAPKAPK3 (AMPK interface) ( Extended Data Fig. 7 j). Systematic deconvolution of interactome topology demonstrated three functional strata: EGFR inhibitor resistance (p = 0.0368), Protein export (p = 0.0336), Pancreatic cancer (p = 0.0319), mTOR signaling pathway (p = 0.0294), Vesicular transport (p = 0.0066), Nucleocytoplasmic transport (p = 0.0000), Protein processing in ER (p = 0.0000) ( Extended Data Fig. 7 k). Compartmental mapping further revealed ER-nucleated signaling hubs, with 74.6% of interactors residing in membrane-bound organelles ( Extended Data Fig. 7 l). Given the ER's emerging role as a topological conductor for molecular transport and signal integration(Ron and Walter, 2007 ; Schwarz and Blower, 2016 ), our findings posit CTRB1 as a multi-pathway licensor that spatially coordinates proliferative signaling through: (i) physical scaffolding of key kinases/transcription factors at ER membranes, and (ii) coupling signal activation with secretory trafficking during clonal selection. Collectively, these data demonstrate that Ctrb1 functions as a cell-intrinsic fitness amplifier through enhanced proliferative capacity and multiple oncogenic signaling. Microbial infection impairs Ctrb1-driven EMDAC to promote tumorigenesis Given the emerging roles of pathobionts like Streptococcus anginosus ( s.a. )(Fu et al., 2024 ; Yuan et al., 2024 ; Zhou et al., 2022 ) and Candida albicans ( c.a. )(Dohlman et al., 2022 ) in gastric oncogenesis and context-dependent expression of Ctrb1, we hypothesized that microbial dysbiosis might subvert Ctrb1-mediated EMDAC. To verify this host-pathogen crosstalk paradigm, we established a tripartite experimental axis: 1) 4-week oral pathogen challenge with s.a. or c.a. , 2) orthotopic implantation of MFC STCL cells, and 3) multi-parametric analysis of EMDAC dynamics (Fig. 7 g). Pathogen-engrafted mice developed microbial persistence with complete failure of tumor clearance (vs. sterile controls), establishing a pathogen-dependent oncogenic permissiveness (Fig. 7 h,i). Mechanistically, microbial colonization eroded the EMDAC architecture - evidenced by 34.1% ( s.a. ) and 27.7% ( c.a. ) reduction in PCL territories and 87.5% ( s.a. ) and 89.6% ( c.a. ) depletion of Ki67 + Ctrb1 + sentinel cells (Fig. 7 j, Extended Data Fig. 7 m-o). Further immunostaining of Ki67 revealed pathogen-driven competitive reprogramming: tumor foci exhibited 2.7-fold ( s.a. ) and 2.0-fold ( c.a. ) elevated proliferation compared to adjacent PCL ( Extended Data Fig. 7 o), suggesting microbiota-induced fitness switching. These findings coalesce into a pathogenic triad model where microbial challenge: 1) silences Ctrb1-mediated surveillance, 2) disables EMDAC-mediated tumor elimination, and 3) creates permissive niche for malignant clonal expansion (Fig. 7 k). Discussion Reconceptualizing Host Defense: The EMDAC Paradigm for Non-Immune Surveillance Our study unveils Epithelial-Mesenchymal Defense Against Cancer (EMDAC) as a spatiotemporally coordinated host surveillance program that transcends canonical immune-mediated mechanisms. While prior work established Epithelial Defense Against Cancer (EDAC) as a frontline mechanism against preneoplastic clones(Ayukawa et al., 2021 ; Lima and Rodriguez, 2021 ; Moya et al., 2019 ; van Neerven and Vermeulen, 2023 ), we demonstrate that mesenchymal co-option transforms this into a multi-lineage containment strategy. Our discovery of the PCL cell competition architecture - with its dual-lineage dynamics (epithelial expansion + fibroblast mobilization) and Myc EP -Yap ME signaling axes - redefines the conceptual framework of non-immune tumor surveillance ( Extended Data Fig. 8 ). Beyond epithelial surveillance, our study supports a multi-layered defense paradigm in which coordinated action of epithelial-mesenchymal compartments assemble into proliferative barrier structures (PCL) to physically constrain tumors. This multicellular architecture distinguishes EMDAC from classical epithelial surveillance, i.e. , EDAC, by integrating stromal reprogramming. Our lineage tracing and spatial transcriptomics data demonstrate mesenchymal fibroblasts - particularly epithelial-like eCAFs with enhanced tumor-proximal activity and proliferative superiority over canonical CAF subtypes - as indispensable partners in forming the PCL. This architectural innovation expands the "guardian cell" repertoire beyond epithelial sentinels, suggesting mesenchymal lineages actively participate in tumor suppression through structural remodeling. During EMDAC, coordinated activation of Myc EP -Yap ME signaling bifurcation establishes lineage-restricted proliferation programs required for PCL formation. In this regard, the PCL's compartmentalized signaling architecture, with Myc EP /YAP ME activation mirroring embryonic morphogenetic fields, provides a biochemical blueprint for host-tumor boundary formation. Ctrb1: A Master Regulator for Cellular Fitness Myc and Yap signaling not only drive cell proliferation but also converge on Ctrb1, a key molecular effector of EMDAC. Ctrb1, normally restricted to the pancreas(Eizirik et al., 2012 ), is ectopically expressed in tumor-adjacent epithelial and mesenchymal cells, enhancing their fitness to outcompete tumor cells. This discovery positions Ctrb1 as a biomarker for supercompetitor cells in host defense against cancer. The serendipitous discovery of Ctrb1 as an ectopically expressed surveillance effector in gastric EMDAC challenges conventional views of its pancreatic-restricted function. Our findings position Ctrb1 as: 1) a contact-dependent danger signal triggered by tumor adjacency; 2) a signaling nexus integrating multiple oncogenic pathways with broader proliferative networks; 3) a microbial vulnerability node linking infection to carcinogenesis. Notably, Ctrb1's ability to simultaneously activate Hippo/YAP, Wnt, and Myc pathways suggests it functions as a proteolytic signaling amplifier - a hypothesis supported by its interaction with ER-associated signalosomes. This "moonlighting" protease activity may explain its paradoxical roles in both digestive physiology and tumor suppression, reminiscent of matrix metalloproteinases' dual functions in tissue remodeling and cancer. The ectopic induction of Ctrb1 in gastric stroma represents a paradigm shift in understanding tissue-specific tumor defense. While pancreatic-restricted under homeostasis, Ctrb1 becomes a tumor-contact sensor across lineages - epithelial cells show 9.1-fold induction upon direct tumor interaction, while fibroblasts exhibit 57.8-fold upregulation. This spatial specificity aligns with its role as a molecular rheostat: Ctrb1 coordinates Myc/Yap-driven hyperproliferation in the PCL while suppressing tumor-promoting Wnt signaling through NOTUM induction. The protease's dual capacity to 1) amplify host cell fitness via Hippo/Myc/p53 axis and 2) antagonize tumor-supportive niches through paracrine factors positions it as a pleiotropic regulator of EMDAC. STCL: Rewiring Proximity Labeling for Spatiotemporal Mapping of Tumor Contactome Cell-cell interactions are crucial for both tissue homeostasis and disease development(Bechtel et al., 2021 ; Boareto, 2020 ; Su et al., 2024 ). Our development of the STCL system represents a significant advancement in mapping direct cell-cell interactions, i.e. , physical contactome. Unlike previous methods(Chudnovskiy et al., 2024 ; Liu et al., 2022 ; Liu et al., 2020 ; Nakandakari-Higa et al., 2024 ; Pasqual et al., 2018 ; Takano et al., 2020 ), STCL allows unbiased, ligand-receptor-independent labeling of physically contacted cells, providing a comprehensive contactome in disease models. Compared to existing tools, STCL's membrane-anchored TurboID design minimizes intracellular labeling artifacts while maintaining compatibility with complex in vivo models. Therefore, the STCL platform overcomes critical limitations in existing cell interaction tools by enabling: 1) Ligand-receptor agnostic profiling - capturing transient tumor-stromal contacts likely missed by LIPSTIC/FucoID systems; 2) Temporal resolution - mapping interaction dynamics during tumor clearance phases (Days 5–15); 3) In vivo scalability - generating organ-specific contactome without requiring dual genetic manipulation. By integrating STCL with scRNA-seq and spatial transcriptomics, we deciphered the dynamic tumor contactome during EMDAC, revealing preferential interactions between tumor cells and epithelial/fibroblast populations. The identification of eCAFs as privileged tumor interactors exemplifies STCL utility in discovering non-canonical cellular crosstalk. Notably, our application of STCL in gastric models revealed an unexpected dominance of non-immune interactions (87.3% epithelial/stromal vs 9.6% immune) at D10, challenging the immunotherapy-centric view of tumor microenvironment crosstalk. Translational Implications Our findings establish EMDAC as a blueprint for engineering multicellular defense programs - a frontier that may complement current immune-centric approaches in precision oncology. In this regard, our study illuminates three translational axes: 1) Ctrb1 as a prognostic biomarker - Its expression dynamics correlate with PCL integrity and survival outcomes; 2) Microbial modulation - Pathogen-mediated Ctrb1 suppression provides mechanistic basis for infection-associated gastric cancer; 3) Therapeutic targeting - Pharmacological stabilization of PCL architecture could enhance endogenous tumor surveillance. While EMDAC demonstrates tumor suppression in an orthotopic gastric cancer model, three key questions emerge: 1) Tissue plasticity - Can Ctrb1-mediated defense be reactivated in metastasized tumors? Preliminary data show circulating Ctrb1 levels correlate with gastric cancer prognosis; 2) Microbiome crosstalk - How do H. pylori -induced Ctrb1 fluctuations impact EMDAC efficacy? 3) Therapeutic hijacking - Can synthetic Ctrb1 analogs boost endogenous tumor suppression in immunotherapy-resistant cancers? Limitations and Future Directions While our study provides novel insights into EMDAC, several questions remain. First, the universality of EMDAC across other epithelial tissues ( e.g. , gut, liver, breast) warrants investigation. Second, the mechanisms underlying epithelial and mesenchymal cell recruitment to the PCL are unclear. Third, the potential role of Ctrb1 in regulating nuclear transport of transcriptional factors requires further exploration. Finally, the application of STCL to other cell types and disease models will expand our understanding of intercellular communication in health and disease. Complementary application of STCL with Cre/Lox systems will reveal bidirectional communication networks. In conclusion, our study unveils EMDAC as a novel non-immune surveillance mechanism, identifies Ctrb1 as a key molecular driver, and introduces STCL as a powerful tool for mapping cell-cell interactions. These findings provide a foundation for developing new therapeutic strategies to enhance host defense against cancer. Methods Cells HEK293FT cells were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). GEC cells were sourced from Biofeng (Hunan, China), and MFC cells were acquired from the National Infrastructure of Cell Line Resource (Beijing, China). L929 and 3T3-L1 cell lines were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA), while the DC2.4 cell line was obtained from Sigma-Aldrich (St. Louis, MO, USA). MFC, GEC, and DC2.4 cells were cultured in RPMI 1640 medium (Invitrogen, Carlsbad, CA, USA), supplemented with 10% fetal bovine serum (FBS; Biological Industries, Kibbutz Beit Haemek, Israel) and 1% penicillin/streptomycin (Thermo Fisher Scientific, Waltham, MA, USA). All other cell lines were maintained in Dulbecco’s Modified Eagle Medium (DMEM; Invitrogen) under the same supplementation conditions. Cells were incubated at 37°C in a humidified atmosphere with 5% CO2 using a Thermo Fisher Scientific incubator (Waltham, MA, USA). Mice and Genotyping Mouse Lines and Sources Ctrb1 F/F mice were generated by the Shanghai Model Organisms Center (SMOC, Shanghai, China). Claudin18.2 Cre/ERT2 mice (Stock No. T056749) were obtained from GemPharmatech (Jiangsu, China). Col1a2 Cre/ERT2 mice (Stock, 029567) were obtained from Jackson Lab. R26 RFP mice (Rosa26-LoxP-STOP-LoxP-RFP) (Han et al., 2019b ) were kindly provided by Dr. Bin Zhou (CAS Center for Excellence in Molecular Cell Science, Shanghai). αSMA Cre mice were a generous gift from Dr. Gang Wang (Fudan University, Shanghai). Yap1 F/F and Taz F/F mice were previously described(Tang et al., 2020 ). Tamoxifen Administration Cre/ERT2 recombinase activity was induced by intraperitoneal injection of tamoxifen (Sigma-Aldrich, St. Louis, MO, USA) at specified time points to activate lineage tracing. Genotyping Genotyping was performed using PCR with primers listed in Supplementary Table S1 . Animal Housing and Ethical Approval All mice used in this study were maintained on a C57BL/6 genetic background, except for the 615 strains. Both male and female mice aged 4–8 weeks were included in the analyses. Mice were housed in individually ventilated cages under group housing conditions whenever possible. They were maintained in a controlled environment with a 12-h light/dark cycle and provided ad libitum access to water and standard rodent chow. Randomization of mice into experimental groups was performed in strict accordance with the guidelines of the Institutional Animal Care and Use Committee (IACUC). All animal procedures were approved by the IACUC of Fudan University (approval ID: IDM2022037) and Tongji University (approval ID: SHDSYY-2023-P0011), ensuring compliance with ethical standards for animal research. Statistical Analysis Sample size estimation for planned comparisons of two independent means was performed using a two-tailed test in GraphPad Prism 9 (GraphPad Software, San Diego, CA, USA). Continuous variables are presented as mean ± standard deviation (s.d.), while categorical variables are reported as frequencies and proportions. Statistical comparisons of continuous data were conducted using Student’s t-tests (for two groups) or one-way ANOVA (for multiple groups). A p-value < 0.05 was considered statistically significant. All experiments were performed with at least two biological replicates to ensure reproducibility. Declarations Conflict of interest The authors declare no conflicts of interests. Author contribution Z.Zhao and F.C. performed most of the molecular, cellular and animal experiments. Y.M. performed most of Bioinformatics Analysis. H.Z., M.Z. and X.Z. did animal sample preparation. Y.H. and W.W. performed plasmid construction. Z.C. performed ChIP-Seq and RNA-Seq experiments. M.D and Z.L did bacterial-related experiments. L.A., S.J. and Z.Zhou designed the experiments, analyzed the data, and wrote the manuscript. S.J., and Z.Z. supervised the project. Acknowledgements This work was supported by the National Key R&D Program of China (2020YFA0803200), the National Natural Science Foundation of China Grants (82222052, 32070710, 31930026, 81972876, 81725014, 92168116 and 82150112), Natural Science Foundation of Shanghai (23ZR1480400), Shanghai Rising-Star Program (22QA1407300), Shanghai Sailing Program (No. 23YF1432900). Data availability RNA sequencing data reported in this paper have been deposited in the GSA-human database ( https://ngdc.cncb.ac.cn/gsa-human/ ). The accession number is subHRA014955. scRNA-seq and spatial transcriptomics for direct tumoral interactome have been deposited in National Genomics Data Center with accession number OMIX008890 ( https://ngdc.cncb.ac.cn/omix/preview/z4vQ2aXQ ). The CTRB1-interactome MS data was deposited in the iProX database under accession number IPX0010977000 ( https://www.iprox.cn/ ). References Ayukawa, S., Kamoshita, N., Nakayama, J., Teramoto, R., Pishesha, N., Ohba, K., Sato, N., Kozawa, K., Abe, H., Semba, K., et al. (2021). Epithelial cells remove precancerous cells by cell competition via MHC class I–LILRB3 interaction. Nature Immunology 22 , 1391–1402. Bechtel, T.J., Reyes-Robles, T., Fadeyi, O.O., and Oslund, R.C. (2021). Strategies for monitoring cell–cell interactions. Nature Chemical Biology 17 , 641–652. Boareto, M. (2020). Patterning via local cell-cell interactions in developing systems. Developmental Biology 460 , 77–85. Cho, K.F., Branon, T.C., Rajeev, S., Svinkina, T., Udeshi, N.D., Thoudam, T.A.-O., Kwak, C., Rhee, H.A.-O., Lee, I.K., Carr, S.A., et al. (2020). Split-TurboID enables contact-dependent proximity labeling in cells. PNAS 117 . Chudnovskiy, A.A.-O., Castro, T.B.R., Nakandakari-Higa, S., Cui, A.A.-O., Lin, C.H., Sade-Feldman, M.A.-O., Phillips, B.A.-O., Pae, J., Mesin, L.A.-O., Bortolatto, J.A.-O., et al. (2024). Proximity-dependent labeling identifies dendritic cells that drive the tumor-specific CD4(+) T cell response. Science immunology 9(100) . Dohlman, A.B., Klug, J., Mesko, M., Gao, I.H., Lipkin, S.M., Shen, X., and Iliev, I.D. (2022). A pan-cancer mycobiome analysis reveals fungal involvement in gastrointestinal and lung tumors. Cell 185 , 3807–3822. Eizirik, D., Sammeth, M., Bouckenooghe, T., Bottu, G., Sisino, G., Igoillo-Esteve, M., Ortis, F., Santin, I., Colli, M., Barthson, J., et al. (2012). The Human Pancreatic Islet Transcriptome: Expression of Candidate Genes for Type 1 Diabetes and the Impact of Pro-Inflammatory Cytokines. PLoS genetics 8 , e1002552. Flanagan, D.J., Pentinmikko, N., Luopajärvi, K., Willis, N.J., Gilroy, K., Raven, A.P., McGarry, L., Englund, J.I., Webb, A.T., Scharaw, S., et al. (2021a). NOTUM from Apc-mutant cells biases clonal competition to initiate cancer. Nature 594 , 430–435. Flanagan, D.J., Pentinmikko, N.A.-O., Luopajärvi, K., Willis, N.A.-O., Gilroy, K., Raven, A.P., McGarry, L.A.-O., Englund, J.A.-O., Webb, A.T., Scharaw, S.A.-O., et al. (2021b). NOTUM from Apc-mutant cells biases clonal competition to initiate cancer. Nature 594(7863) . Fu, K., Cheung, A.H.K., Wong, C.C., Liu, W., Zhou, Y., Wang, F., Huang, P., Yuan, K., Coker, O.O., Pan, Y., et al. (2024). Streptococcus anginosus promotes gastric inflammation, atrophy, and tumorigenesis in mice. Cell 187 , 882–896. Ge, Y., Chen, L., Liu, S., Zhao, J., Zhang, H., and Chen, P.R. (2019). Enzyme-Mediated Intercellular Proximity Labeling for Detecting Cell–Cell Interactions. Journal of the American Chemical Society 141 , 1833–1837. Han, H., Jain, A.D., Truica, M.I., Izquierdo-Ferrer, J., Anker, J.F., Lysy, B., Sagar, V., Luan, Y., Chalmers, Z.R., Unno, K., et al. (2019a). Small-Molecule MYC Inhibitors Suppress Tumor Growth and Enhance Immunotherapy. Cancer cell 36 , 483–497.e415. Han, X., Wang, Y., Pu, W., Huang, X., Qiu, L., Li, Y., Yu, W., Zhao, H., Liu, X., He, L., et al. (2019b). Lineage Tracing Reveals the Bipotency of SOX9(+) Hepatocytes during Liver Regeneration. Stem Cell Reports 12 , 624–638. Lima, A., and Rodriguez, T.A. (2021). MHC-I presents: tumor surveillance in the epithelia by cell competition. Nature Immunology 22 , 1358–1360. Liu-Chittenden, Y., Huang, B., Shim, J.S., Chen, Q., Lee, S.J., Anders, R.A., Liu, J.O., and Pan, D. (2012). Genetic and pharmacological disruption of the TEAD-YAP complex suppresses the oncogenic activity of YAP. Genes & development 26 , 1300–1305. Liu, H., Luo, H., Xue, Q., Qin, S., Qiu, S., Liu, S., Lin, J., Li, J.P., and Chen, P.R. (2022). Antigen-Specific T Cell Detection via Photocatalytic Proximity Cell Labeling (PhoXCELL). Journal of the American Chemical Society 144 , 5517–5526. Liu, Q., Tang, Y.A.-O., Huang, Y.A.-O., Wang, J., Yang, K., Zhang, Y., Pu, W., Liu, J., Shi, X., Ma, Y.A.-O., et al. (187). Insights into male androgenetic alopecia using comparative transcriptome profiling: hypoxia-inducible factor-1 and Wnt/β-catenin signalling pathways. British Journal of Dermatology 2022. Liu, Z., Li, J.P., Chen, M., Wu, M., Shi, Y., Li, W., Teijaro, J.R., and Wu, P. (2020). Detecting Tumor Antigen-Specific T Cells via Interaction-Dependent Fucosyl-Biotinylation. Cell 183, 1117–1133. Moya, I.A.-O., Castaldo, S.A.-O., Van den Mooter, L.A.-O., Soheily, S.A.-O., Sansores-Garcia, L., Jacobs, J.A.-O., Mannaerts, I.A.-O., Xie, J.A.-O., Verboven, E.A.-O., Hillen, H., et al. (2019). Peritumoral activation of the Hippo pathway effectors YAP and TAZ suppresses liver cancer in mice. Science 366 . Nakai, K., Lin, H., Yamano, S., Tanaka, S., Kitamoto, S., Saitoh, H., Sakuma, K., Kurauchi, J., Akter, E., Konno, M., et al. (2023). Wnt activation disturbs cell competition and causes diffuse invasion of transformed cells through NF-κB-MMP21 pathway. Nature communications 14 , 7048. Nakandakari-Higa, S., Walker, S., Canesso, M.C.C., van der Heide, V., Chudnovskiy, A., Kim, D.-Y., Jacobsen, J.T., Parsa, R., Bilanovic, J., Parigi, S.M., et al. (2024). Universal recording of immune cell interactions in vivo. Nature 627 , 399–406. Parker, T.M., Gupta, K., Palma, A.M., Yekelchyk, M.A.-O., Fisher, P.B., Grossman, S.R., Won, K.A.-O., Madan, E., Moreno, E.A.-O.X., and Gogna, R.A.-O. (2021). Cell competition in intratumoral and tumor microenvironment interactions. The EMBO Journal 40(17) . Pasqual, G., Chudnovskiy, A., Tas, J.M.J., Agudelo, M., Schweitzer, L.D., Cui, A., Hacohen, N., and Victora, G.D. (2018). Monitoring T cell–dendritic cell interactions in vivo by intercellular enzymatic labelling. Nature 553 , 496–500. Pentinmikko, N., Iqbal, S., Mana, M., Andersson, S., Cognetta, A.B., 3rd, Suciu, R.M., Roper, J., Luopajärvi, K., Markelin, E., Gopalakrishnan, S., et al. (2019). Notum produced by Paneth cells attenuates regeneration of aged intestinal epithelium. Nature 571 , 398–402. Qiu, S., Zhao, Z., Wu, M., Xue, Q., Yang, Y., Ouyang, S., Li, W., Zhong, L., Wang, W., Yang, R., et al. (2022). Use of intercellular proximity labeling to quantify and decipher cell-cell interactions directed by diversified molecular pairs. Science advances 8 , eadd2337. Rauskolb, C., Sun, S., Sun, G., Pan, Y., and Irvine, K.D. (2014). Cytoskeletal tension inhibits Hippo signaling through an Ajuba-Warts complex. Cell 158 . Ron, D., and Walter, P. (2007). Signal integration in the endoplasmic reticulum unfolded protein response. Nature reviews Molecular cell biology 8 , 519–529. Roux, K.J., Kim Di Fau - Raida, M., Raida M Fau - Burke, B., and Burke, B. (2012). A promiscuous biotin ligase fusion protein identifies proximal and interacting proteins in mammalian cells. Journal of Cell Biology 6 , 801–810. Schwarz, D.S., and Blower, M.D. (2016). The endoplasmic reticulum: structure, function and response to cellular signaling. Cellular and molecular life sciences: CMLS 73, 79–94. Su, J., Song, Y., Zhu, Z., Huang, X., Fan, J., Qiao, J., and Mao, F. (2024). Cell–cell communication: new insights and clinical implications. Signal Transduction and Targeted Therapy 9 , 196. Subramaniam, D.A.-O.X., Ponnurangam, S., Ramalingam, S., Kwatra, D., Dandawate, P., Weir, S.A.-O.X., Umar, S., Jensen, R.A.-O., and Anant, S.A.-O. (2021). Honokiol Affects Stem Cell Viability by Suppressing Oncogenic YAP1 Function to Inhibit Colon Tumorigenesis. LID – 10.3390/cells10071607 [doi] LID – 1607. Cells 10 . Takano, T., Wallace, J.T., Baldwin, K.T., Purkey, A.M., Uezu, A., Courtland, J.L., Soderblom, E.J., Shimogori, T., Maness, P.F., Eroglu, C., et al. (2020). Chemico-genetic discovery of astrocytic control of inhibition in vivo. Nature 588 , 296–302. Tang, Y., Fang, G., Guo, F., Zhang, H., Chen, X., An, L., Chen, M., Zhou, L., Wang, W., Ye, T., et al. (2020). Selective Inhibition of STRN3-Containing PP2A Phosphatase Restores Hippo Tumor-Suppressor Activity in Gastric Cancer. Cancer Cell 38 , 115–128.e119. Tanimura, N., and Fujita, Y. (2020). Epithelial defense against cancer (EDAC). Seminars in cancer biology 63 , 44–48. van Neerven, S.M., de Groot, N.E., Nijman, L.E., Scicluna, B.P., van Driel, M.S., Lecca, M.C., Warmerdam, D.O., Kakkar, V., Moreno, L.F., Vieira Braga, F.A., et al. (2021). Apc-mutant cells act as supercompetitors in intestinal tumour initiation. Nature 594 , 436–441. van Neerven, S.M., and Vermeulen, L. (2023). Cell competition in development, homeostasis and cancer. Nature Reviews Molecular Cell Biology 24 , 221–236. Yuan, L., Pan, L., Wang, Y., Zhao, J., Fang, L., Zhou, Y., Xia, R., Ma, Y., Jiang, Z., Xu, Z., et al. (2024). Characterization of the landscape of the intratumoral microbiota reveals that Streptococcus anginosus increases the risk of gastric cancer initiation and progression. Cell Discovery 10 , 117. Yum, M.K., Han, S., Fink, J., Wu, S.-H.S., Dabrowska, C., Trendafilova, T., Mustata, R., Chatzeli, L., Azzarelli, R., Pshenichnaya, I., et al. (2021). Tracing oncogene-driven remodelling of the intestinal stem cell niche. Nature 594 , 442–447. Zhou, C.B., Pan, S.Y., Jin, P., Deng, J.W., Xue, J.H., Ma, X.Y., Xie, Y.H., Cao, H., Liu, Q., Xie, W.F., et al. (2022). Fecal Signatures of Streptococcus anginosus and Streptococcus constellatus for Noninvasive Screening and Early Warning of Gastric Cancer. Gastroenterology 162 . Additional Declarations There is NO Competing Interest. Supplementary Files STCLSupp23.docx Supplementary information Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6524135","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":452276664,"identity":"13025bf0-9fd1-4560-9d19-e36b41b4e37f","order_by":0,"name":"Zhaocai Zhou","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAvklEQVRIiWNgGAWjYHACNiC2YWCTADOI15JGupbDDAxEa5FvP/7swce28/Z80s3PHjDU2DHwz27Ar4WxJyHdcGbb7cQ2mWPmBgzHkhkk7hzAr4VZguGYNG/b7QQ2iRyQdw4wGEgkEPCIBGMbUMs5e4iWf0Ro4ZFgZgNqOcDYBtLC2EaEFgmeNDbJGeeSIX5J7EvmkbhBQAsoxCQ+lNnZy88GhtiHb3Zy/DMIaEEFQMU8pKgfBaNgFIyCUYADAAB8lzXPiTvPPQAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-5441-3922","institution":"Fudan University","correspondingAuthor":true,"prefix":"","firstName":"Zhaocai","middleName":"","lastName":"Zhou","suffix":""},{"id":452276665,"identity":"f3156138-c8b2-4375-bd6a-9f1b701a3952","order_by":1,"name":"Zhangting Zhao","email":"","orcid":"","institution":"Tongji University","correspondingAuthor":false,"prefix":"","firstName":"Zhangting","middleName":"","lastName":"Zhao","suffix":""},{"id":452276666,"identity":"359f822d-2e17-4b3c-a75d-78ca9ecd9ece","order_by":2,"name":"Fan Chen","email":"","orcid":"","institution":"SIBCB","correspondingAuthor":false,"prefix":"","firstName":"Fan","middleName":"","lastName":"Chen","suffix":""},{"id":452276667,"identity":"7083f12f-e926-43b8-9bfa-cea520f5ee59","order_by":3,"name":"Yang Meng","email":"","orcid":"","institution":"Shanghai Eastern Hepatobiliary Surgery Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Meng","suffix":""},{"id":452276668,"identity":"49e90da6-df44-471b-8ec6-0bfd3569d4a4","order_by":4,"name":"Hui Zhang","email":"","orcid":"","institution":"School of Life Sciences, Fudan University","correspondingAuthor":false,"prefix":"","firstName":"Hui","middleName":"","lastName":"Zhang","suffix":""},{"id":452276669,"identity":"86ed3f3e-0c4d-4ca2-a077-115088db1013","order_by":5,"name":"Mengwen Zhu","email":"","orcid":"","institution":"Fudan University","correspondingAuthor":false,"prefix":"","firstName":"Mengwen","middleName":"","lastName":"Zhu","suffix":""},{"id":452276670,"identity":"2179f139-85d3-4188-8298-4a262c8aacd4","order_by":6,"name":"Yi Han","email":"","orcid":"","institution":"Shanghai Tenth People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Han","suffix":""},{"id":452276671,"identity":"50cd917b-814d-4633-82e4-9a6695e09de8","order_by":7,"name":"Xiaowei Wang","email":"","orcid":"","institution":"tongji university","correspondingAuthor":false,"prefix":"","firstName":"Xiaowei","middleName":"","lastName":"Wang","suffix":""},{"id":452276672,"identity":"69327b3f-70fd-433a-ba0c-0dfb6e9a96a6","order_by":8,"name":"Zhifa Cao","email":"","orcid":"","institution":"Chinese Academy of Sciences, University of Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Zhifa","middleName":"","lastName":"Cao","suffix":""},{"id":452276673,"identity":"c970f343-8ab7-4f25-a758-6789b0d86053","order_by":9,"name":"xinru Zhang","email":"","orcid":"","institution":"tongji university","correspondingAuthor":false,"prefix":"","firstName":"xinru","middleName":"","lastName":"Zhang","suffix":""},{"id":452276674,"identity":"18a757ca-e4dd-4e51-b8b2-21388d849755","order_by":10,"name":"meihang du","email":"","orcid":"","institution":"tongji university","correspondingAuthor":false,"prefix":"","firstName":"meihang","middleName":"","lastName":"du","suffix":""},{"id":452276675,"identity":"46c76fba-27c7-4b54-9a83-8516592b2f0c","order_by":11,"name":"Ziyuan Liu","email":"","orcid":"","institution":"tongji university","correspondingAuthor":false,"prefix":"","firstName":"Ziyuan","middleName":"","lastName":"Liu","suffix":""},{"id":452276676,"identity":"e614c77c-c3c1-4af3-a39c-00a8c7ef45a0","order_by":12,"name":"Wenjia Wang","email":"","orcid":"","institution":"Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Wenjia","middleName":"","lastName":"Wang","suffix":""},{"id":452276677,"identity":"46fc0685-0a14-4ddd-a790-7dec069a0932","order_by":13,"name":"Shi Jiao","email":"","orcid":"https://orcid.org/0000-0003-3591-8973","institution":"Fudan University","correspondingAuthor":false,"prefix":"","firstName":"Shi","middleName":"","lastName":"Jiao","suffix":""},{"id":452276678,"identity":"cbd6c2ca-70de-463f-ba8c-7534a7c871e1","order_by":14,"name":"Liwei An","email":"","orcid":"","institution":"Tongji University","correspondingAuthor":false,"prefix":"","firstName":"Liwei","middleName":"","lastName":"An","suffix":""}],"badges":[],"createdAt":"2025-04-25 01:05:42","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6524135/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6524135/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82132191,"identity":"b693b51e-3da8-4635-91ee-686453afce63","added_by":"auto","created_at":"2025-05-07 05:42:06","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":7319552,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEpithelial and Mesenchymal Cells Form a Proliferative Containment Layer (PCL) to Constrain Tumor Progression\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea.\u003c/strong\u003e Schematic of an orthotopic gastric cancer model (left panel). MFC cells were injected into the gastric submucosa-muscularis layer. Representative images show the proliferative containment layer (PCL) surrounding tumor tissue at D5 and D10 (right panel). Scale bar: 200 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb.\u003c/strong\u003e Quantification of tumor and PCL zones at D5 and D10 (n = 4 mice per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec.\u003c/strong\u003e Percentage of Ki67\u003csup\u003e+\u003c/sup\u003e cells at increasing distances from the submucosa (upper panel) and muscularis (lower panel) to the tumor core (n = 4 mice per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed, f. \u003c/strong\u003eCo-immunostaining of Ki67 with CK8 (d) or αSMA (f) in stomach tissues from the orthotopic model. Scale bars: 100 µm and 10 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee, g.\u003c/strong\u003e Quantification of CK8\u003csup\u003e+\u003c/sup\u003eKi67\u003csup\u003e+\u003c/sup\u003e cells (e) and αSMA\u003csup\u003e+\u003c/sup\u003eKi67\u003csup\u003e+ \u003c/sup\u003ecells (g) at increasing distances from the submucosa and muscularis to the tumor (n = 3 mice per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eh. \u003c/strong\u003eCo-immunostaining of Ki67 and RFP in stomach tissues of \u003cem\u003eClaudin18.2\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre/ERT2\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e;R26\u003c/em\u003e\u003csup\u003e\u003cem\u003eRFP\u003c/em\u003e\u003c/sup\u003e mice at D10 post orthotopic implantation of MFC cells (left panel). Proportion of RFP\u003csup\u003e+\u003c/sup\u003e cells (middle panel) and RFP\u003csup\u003e+\u003c/sup\u003eKi67\u003csup\u003e+\u003c/sup\u003e cells (right panel) in the PCL (n = 3 mice per group). Scale bars: 100 µm and 10 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ei. \u003c/strong\u003eCo-immunostaining of Ki67 and RFP in stomach tissues of \u003cem\u003eαSma\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e;R26\u003c/em\u003e\u003csup\u003e\u003cem\u003eRFP\u003c/em\u003e\u003c/sup\u003e mice at D10 post orthotopic implantation of MFC cells (left panel). Proportion of RFP\u003csup\u003e+\u003c/sup\u003e cells (middle panel) and RFP\u003csup\u003e+\u003c/sup\u003eKi67\u003csup\u003e+\u003c/sup\u003e cells (right panel) in the PCL (n = 4 and n = 5 mice per group). Scale bar: 100 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ej.\u003c/strong\u003e Model of epithelial-mesenchymal defense against cancer (EMDAC). Epithelial and mesenchymal cells are recruited to the tumor border, where interactions with tumor cells enhance their proliferative capacity, forming a PCL that competes with and eliminates the tumor.\u003c/p\u003e\n\u003cp\u003eStatistical Analysis: Data are presented as mean ± s.d. Significant differences were determined using two-tailed, unpaired t-tests (b, h, i) or unpaired t-tests based on the areas under two curves (AUC) (c, e, g).\u003c/p\u003e\n\u003cp\u003eSee also\u003cstrong\u003e Extended Data Fig. 1\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-6524135/v1/ff5ccb630abfc14f0f9acc23.png"},{"id":82132194,"identity":"482cfa39-932d-41f4-91f6-8025bd6eb6c6","added_by":"auto","created_at":"2025-05-07 05:42:06","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":4684918,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSpatial Transcriptomics Reveal Coordinated Myc\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003eEP\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e and Yap\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003eME\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e Signaling in Driving EMDAC\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea.\u003c/strong\u003e Patho-DBiT profiling of stomach tissue from the orthotopic gastric cancer model (\u003cem\u003eo.c.\u003c/em\u003e) at three time points (left: H\u0026amp;E staining; middle: unsupervised clustering; right: schematic of unsupervised clustering). The spatial transcriptomics analysis identified four functional layers: Mesenchymal (ME) layer, Epithelial (EP) layer, Proliferative Containment Layer (PCL), and Tumor zone.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb. \u003c/strong\u003eDynamics of unsupervised clustering across the four functional layers at D5, D10, and D15 post orthotopic tumor implantation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec, d.\u003c/strong\u003e Heatmaps illustrating the dynamics of proliferation scores (c) and signaling pathway scores (d) across the four functional layers.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee. \u003c/strong\u003eSurgical isolation of epithelial and mesenchymal tissues from the tumor region and contralateral control region (upper panel). qPCR analysis of target genes downstream of Myc and Hippo-Yap pathways in the tumor region and control region (right panel) (n = 6 mice per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ef. \u003c/strong\u003eSchematic of the \u003cem\u003eo.c.\u003c/em\u003emodel. MFC cells were injected into mice after 14 days of intraperitoneal administration of YAP and Myc inhibitors. Tumors were allowed to grow for 12 days.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eg, h.\u003c/strong\u003e Representative images of gastric cancer tumors from the model in panel f (g) and quantification of tumor volumes (h) (n = 3 mice per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ei, j. \u003c/strong\u003eRepresentative images of gastric cancer tumors (i) and quantification of tumor volumes (j) (n = 8 mice per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ek. \u003c/strong\u003eModel of the epithelial-mesenchymal defense against cancer (EMDAC) ecosystem. Myc\u003csup\u003eEP\u003c/sup\u003e and Yap\u003csup\u003eME\u003c/sup\u003e signaling are essential pathways driving the EMDAC process.\u003c/p\u003e\n\u003cp\u003eStatistical Analysis: Data are presented as mean ± s.d. Significant differences were determined using two-tailed, one-way ANOVA with Dunnett’s post hoc analysis (e, h) or unpaired t-tests (j).\u003c/p\u003e\n\u003cp\u003eSee also\u003cstrong\u003e Extended Data Fig. 2\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-6524135/v1/375f16f98153faed9e103c72.png"},{"id":82136646,"identity":"67bbcb2b-d4f8-4203-957d-71df92260519","added_by":"auto","created_at":"2025-05-07 06:14:06","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":4846623,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe STCL System Enables Contact-Dependent Cellular Labeling \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eIn Vitro\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e and\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e In Vivo\u003c/strong\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea.\u003c/strong\u003e Schematic of the \u003cu\u003eS\u003c/u\u003eurface-anchored \u003cu\u003eT\u003c/u\u003eurboID-mediated \u003cu\u003eC\u003c/u\u003eontact-dependent \u003cu\u003eL\u003c/u\u003eabeling (STCL) strategy, enabling unbiased biotinylation of physically interacting cells upon biotin substrate introduction.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb.\u003c/strong\u003e Co-culture setup of MFC\u003csup\u003eSTCL\u003c/sup\u003e cells with DC2.4 and L929 cells (pre-stained with eFluor670) at a 1:1 ratio, with or without biotin treatment for 12 h, followed by flow cytometry or immunofluorescence analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec. \u003c/strong\u003eFlow cytometry analysis of labeling efficiency in DC2.4 cells contacted by MFC\u003csup\u003eSTCL \u003c/sup\u003e(left panel). Proportion of DC2.4 (\u003cem\u003etrans\u003c/em\u003e) labeled through contact with MFC\u003csup\u003eSTCL\u003c/sup\u003e (n = 3 co-culture wells per group, right panel).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed. \u003c/strong\u003eProportion of L929 cells (\u003cem\u003etrans\u003c/em\u003e) labeled through contact with MFC\u003csup\u003eSTCL\u003c/sup\u003e (n = 3 co-culture wells per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee.\u003c/strong\u003e Immunofluorescence imaging of biotin signals on the membranes of sorted DC2.4 and L929 cells. Scale bar: 20 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ef. \u003c/strong\u003eSchematic of the subcutaneous tumor model. Parental MC38\u003csup\u003eCtrl\u003c/sup\u003e and MC38\u003csup\u003eSTCL\u003c/sup\u003e cells were injected subcutaneously and allowed to grow for 7 days, followed by intraperitoneal biotin administration daily for 5 days. Tumor tissues were harvested for analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eg. \u003c/strong\u003eWestern blot analysis of cis-labeling in MC38\u003csup\u003eSTCL\u003c/sup\u003e cells, confirming TurboID- and biotin-dependent biotinylation \u003cem\u003ein vivo\u003c/em\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eh.\u003c/strong\u003e Proportions of CAF, immune and endothelial cell (\u003cem\u003etrans\u003c/em\u003e)-labeled by tumor cells in the subcutaneous model (n = 3 mice per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ei. \u003c/strong\u003eRepresentative images of biotinylated signals on MFC\u003csup\u003eSTCL\u003c/sup\u003e, TurboID, αSMA, CD11b, and CD3 cells in the subcutaneous tumor model. Scale bar: 10 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ej.\u003c/strong\u003e Schematic of the orthotopic gastric cancer (\u003cem\u003eo.c.\u003c/em\u003e) model. Tumors were allowed to grow for 5 days, followed by 5 days of intraperitoneal biotin administration for \u003cem\u003ein vivo\u003c/em\u003e contact-dependent labeling. Tumor regions were identified by H\u0026amp;E staining (left panel), with TurboID highlighting tumor cells (right panel). Scale bar: 200 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ek. \u003c/strong\u003eTurboID and biotin staining (double-positive) highlighting tumor cells and single biotin-positive (SP) cells within the tumor microenvironment (TME), representing tumor-interacted cells. Scale bars: 200 µm, 20 µm, and 10 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003el. \u003c/strong\u003eRepresentative images of biotinylated signals on αSMA, CD11b, and CD3 cells in the orthotopic model. Scale bar: 10 µm.\u003c/p\u003e\n\u003cp\u003eStatistical Analysis: Data are presented as mean ± s.d. Significant differences were determined using two-tailed, one-way ANOVA with Dunnett’s post hoc analysis (c, d, h).\u003c/p\u003e\n\u003cp\u003eSee also\u003cstrong\u003e Extended Data Fig. 3\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-6524135/v1/130a74e385a262c707280d38.png"},{"id":82134739,"identity":"71c4786b-f336-400b-bf80-473c41a81cbd","added_by":"auto","created_at":"2025-05-07 06:06:06","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":6758973,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDynamic Mapping of Tumor Contactome\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea.\u003c/strong\u003e Schematic of the STCL-scRNA workflow to identify tumor-interacting cell populations and gene signatures in the orthotopic gastric cancer (\u003cem\u003eo.c.\u003c/em\u003e) model. Whole stomach tissues, including tumors, were harvested at D5, D10, and D15. Single-cell suspensions were analyzed by flow cytometry to isolate biotin\u003csup\u003e+\u003c/sup\u003e (interacting) and biotin\u003csup\u003e-\u003c/sup\u003e (bystander) cells, followed by scRNA sequencing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb.\u003c/strong\u003e Interaction-based analysis of trans-labeling efficiencies across different cell types interacting with tumor cells, independent of time intervals.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec. \u003c/strong\u003eCellular composition of biotin\u003csup\u003e+\u003c/sup\u003e cells at each time point (upper panel). Interaction-based analysis of trans-labeling efficiencies between tumor cells and interacting cells at different time intervals (lower panel).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed. \u003c/strong\u003eSpatial distribution of tumor-interacting cell types in the tumor microenvironment (TME) atD10. Scale bar: 200 µm (left panel).\u003cstrong\u003e \u003c/strong\u003eHeatmap showing the proportion of labeled cells based on their distance from the tumor in the \u003cem\u003eo.c. \u003c/em\u003emodel (right panel).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee. \u003c/strong\u003eRepresentative images of biotinylated signals in indicated cell types (left panel). Scale bars: 5 µm and 1 µm.\u003cstrong\u003e \u003c/strong\u003eColumn plot summarizing the proportion of tumor-interacted versus bystander cells in the \u003cem\u003eo.c.\u003c/em\u003e model (n = 3 tumor samples per group, right panel).\u003c/p\u003e\n\u003cp\u003eStatistical Analysis: Data are presented as mean ± s.d. (g).\u003c/p\u003e\n\u003cp\u003eSee also\u003cstrong\u003e Extended Data Fig. 4\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-6524135/v1/a6950b7beb534885e51ac935.png"},{"id":82132192,"identity":"a7c339f4-e998-4ce2-8803-383c52730147","added_by":"auto","created_at":"2025-05-07 05:42:06","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":4026250,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTumor Contactome Analysis Reveals Dual Lineage Origins of PCL Cells\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea. \u003c/strong\u003eUniform Manifold Approximation and Projection (UMAP) plots of epithelial cells, color-coded by cell type, identified through integrated analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb. \u003c/strong\u003eHeatmap of characteristic genes for each epithelial cluster, highlighting transcriptional signatures.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec.\u003c/strong\u003e UMAP plots showing the expression patterns of \u003cem\u003eMki67\u003c/em\u003e and \u003cem\u003ePcna\u003c/em\u003e genes across epithelial clusters, indicative of proliferative activity.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed. \u003c/strong\u003eCellular composition of biotin\u003csup\u003e+ \u003c/sup\u003eepithelial cells at each time point, illustrating dynamic changes in tumor-interacting epithelial populations.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee.\u003c/strong\u003e Trajectory analysis of the EP_1 cluster, depicting its developmental path and proliferative transformation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ef.\u003c/strong\u003e \u0026nbsp;Trajectory analysis of the tumor contact status (biotin\u003csup\u003e+\u003c/sup\u003e vs biotin\u003csup\u003e-\u003c/sup\u003e) within different types of cells during EMDAC process.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eg.\u003c/strong\u003e UMAP plots of cancer-associated fibroblasts (CAFs), color-coded by cell type, identified through integrated analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eh.\u003c/strong\u003e Heatmap of characteristic genes for each CAF cluster, revealing distinct transcriptional profiles.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ei. \u003c/strong\u003eCellular composition of biotin\u003csup\u003e+\u003c/sup\u003e CAF cells at each time point, showing temporal changes in tumor-interacting CAF populations.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ej.\u003c/strong\u003e UMAP plots showing the expression patterns of \u003cem\u003eMki67\u003c/em\u003e and \u003cem\u003ePcna\u003c/em\u003e genes across CAF clusters, reflecting their proliferative capacity.\u003c/p\u003e\n\u003cp\u003eSee also\u003cstrong\u003e Extended Data Fig. 5\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-6524135/v1/b37ce36e785b31488421831a.png"},{"id":82132196,"identity":"84e8d52b-2d58-4cc7-87f1-3c5901f21456","added_by":"auto","created_at":"2025-05-07 05:42:06","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":6279231,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCtrb1 Is Essential for PCL Formation and EMDAC-Mediated Tumor Clearance\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea.\u003c/strong\u003e UMAP plots showing the transcriptional pattern of Ctrb1 across epithelial and fibroblast clusters.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb. \u003c/strong\u003eHeatmap illustrating \u003cem\u003eCtrb1\u003c/em\u003e expression levels in the EP_1 cluster and tumor-interacted versus bystander fibroblasts within the tumor microenvironment (TME).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec.\u003c/strong\u003e Co-immunostaining of Ki67 and Ctrb1 in stomach tissues of tumor-bearing mice (left panel). Quantification of Ctrb1\u003csup\u003e+\u003c/sup\u003eKi67\u003csup\u003e+\u003c/sup\u003e cells in the PCL (right panel) (n = 3 mice per group). Scale bars: 100 µm and 10 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed.\u003c/strong\u003e Immunostaining of Ki67, Ctrb1, and RFP in whole-stomach tissues of \u003cem\u003eαSma\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e;R26\u003c/em\u003e\u003csup\u003e\u003cem\u003eRFP\u003c/em\u003e\u003c/sup\u003e mice at day 10 post MFC injection (left panel). Quantification of CTRB1\u003csup\u003e+\u003c/sup\u003eKi67\u003csup\u003e+\u003c/sup\u003e cells within the RFP\u003csup\u003e+\u003c/sup\u003e PCL region (n = 4 mice per group). Scale bars: 100 µm and 10 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee.\u003c/strong\u003e qPCR analysis of target genes downstream of Myc and Hippo-Yap pathways in epithelial and mesenchymal regions (n = 6 mice per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ef.\u003c/strong\u003e Schematic of the co-culture system with MFC cells and 3T3-L1 or GEC cells (pre-stained with eFluor670) at a 1:1 ratio, followed by flow cytometry sorting for qPCR, immunofluorescence, and Cut \u0026amp; Run analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eg.\u003c/strong\u003e qPCR analysis of Ctrb1 transcription in 3T3-L1 and GEC cells from panel f (n = 7 co-culture wells per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eh.\u003c/strong\u003e Representative images of Ctrb1 expression in 3T3-L1 and GEC cells from panel f. Scale bar: 10 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ei.\u003c/strong\u003e Schematic of the orthotopic gastric cancer (\u003cem\u003eo.c.\u003c/em\u003e) model. MFC cells were injected 21 days after AAV-CMV\u003csup\u003eCre\u003c/sup\u003e-ZsGreen virus injection into the stomach. Tumor volumes were quantified (n = 6 mice per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ej.\u003c/strong\u003e Tumor regions identified by H\u0026amp;E staining in panel i. Scale bar: 100 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ek.\u003c/strong\u003e Schematic of TEAD4 and Myc binding motifs on the Ctrb1 promoter (upper panel). ChIP-qPCR analysis of 3T3-L1 and GEC cells from panel f (n = 3 wells per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003el. \u003c/strong\u003eImmunostaining of Ki67 and Ctrb1 in the PCL of \u003cem\u003eαSma\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e;Yap\u003c/em\u003e\u003csup\u003e\u003cem\u003eF/+\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e;Taz\u003c/em\u003e\u003csup\u003e\u003cem\u003eF/+\u003c/em\u003e\u003c/sup\u003e mice (left panel). Quantification of Ctrb1\u003csup\u003e+\u003c/sup\u003eKi67\u003csup\u003e+\u003c/sup\u003e cells in the PCL (right panel) (n = 3 mice per group). Scale bars: 100 µm and 10 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003em.\u003c/strong\u003e Model illustrating signaling activation in tumor-interacting epithelial and mesenchymal cells, leading to Ctrb1 transcription and PCL formation.\u003c/p\u003e\n\u003cp\u003eStatistical Analysis: Data are presented as mean ± s.d. Significant differences were determined using two-tailed, one-way ANOVA with Dunnett’s post hoc analysis (c-e, g, k) or unpaired t-tests (i, l).\u003c/p\u003e\n\u003cp\u003eSee also\u003cstrong\u003e Extended Data Fig. 6\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-6524135/v1/bb9fd960ffea0cc43ea19aac.png"},{"id":82134740,"identity":"da410d93-02ff-4eaf-9064-22c2cbdb315f","added_by":"auto","created_at":"2025-05-07 06:06:06","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":4714725,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCtrb1 as a Microbial-Sensitive Master Regulator of Cellular Fitness\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea.\u003c/strong\u003e Schematic of the co-culture system with 293FT cells and 293FT\u003csup\u003eGFP\u003c/sup\u003e or 293FT\u003csup\u003eCTRB1-GFP\u003c/sup\u003e stable cell lines at a 1:1 ratio for 7 days.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb. \u003c/strong\u003eRepresentative images of co-culture results on day 7 (left panel). Quantification of the proportions of 293FT\u003csup\u003eGFP\u003c/sup\u003e and 293FT\u003csup\u003eCTRB1-GFP\u003c/sup\u003e cells (right panel) (n = 8 wells per group). Scale bar: 20 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec. \u003c/strong\u003eRepresentative images of gastric cancer (GC) tumors derived from MFC\u003csup\u003eSTCL\u003c/sup\u003e and MFC\u003csup\u003eSTCL-Ctrb1\u003c/sup\u003e cells (left panel). Quantification of tumor volumes (right panel) (n = 5 mice per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed.\u003c/strong\u003e Volcano plot showing differentially expressed genes between 293FT\u003csup\u003eGFP\u003c/sup\u003e and 293FT\u003csup\u003eCTRB1-GFP\u003c/sup\u003e stable cell lines (n = 3 biological replicates per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee.\u003c/strong\u003e KEGG pathway analysis revealing cell proliferation-related pathways enriched in upregulated genes from panel d.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ef. \u003c/strong\u003eTarget genes of each pathway upregulated in 293FT\u003csup\u003eCTRB1-GFP\u003c/sup\u003e cells compared to 293FT\u003csup\u003eGFP\u003c/sup\u003e cells.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eg. \u003c/strong\u003eSchematic of the orthotopic gastric cancer (\u003cem\u003eo.c.\u003c/em\u003e) model. MFC\u003csup\u003eSTCL \u003c/sup\u003ecells were injected into the stomach of 615 mice after a three-week bacterial gavage protocol.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eh, i.\u003c/strong\u003e Quantification of tumor volumes after \u003cem\u003eS. anginosus\u003c/em\u003e (h) and\u003cem\u003e C. albicans\u003c/em\u003e (i) gavage (n = 4 mice per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ej.\u003c/strong\u003e Quantification of Ki67\u003csup\u003e+\u003c/sup\u003eCtrb1\u003csup\u003e+\u003c/sup\u003e cells in tumor tissues after \u003cem\u003eS. anginosus \u003c/em\u003eand \u003cem\u003eC. albicans\u003c/em\u003e gavage (n = 3 tumor samples per group).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ek.\u003c/strong\u003e Model illustrating the role of Ctrb1 in conferring a competitive fitness advantage during EMDAC. Knockout of Ctrb1 in host cells causes PCL collapse and failure of EMDAC, while elevated Ctrb1 expression in tumor cells evades EMDAC. Chronic microbial infection impairs the stomach’s EMDAC capacity, facilitating tumor initiation and progression.\u003c/p\u003e\n\u003cp\u003eStatistical Analysis: Data are presented as mean ± s.d. Significant differences were determined using two-tailed, unpaired t-tests (b, c, h-j).\u003c/p\u003e\n\u003cp\u003eSee also\u003cstrong\u003e Extended Data Fig. 7\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-6524135/v1/2707229ee8c005bb1ec10e49.png"},{"id":83931467,"identity":"89098a24-a2a3-431c-9c6d-3806730fd683","added_by":"auto","created_at":"2025-06-04 15:34:44","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":37340752,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6524135/v1/64209e6f-b502-4530-84f9-66b829bf4b3d.pdf"},{"id":82132211,"identity":"300cd7c3-3895-46ea-86a4-6bc08b72a9c6","added_by":"auto","created_at":"2025-05-07 05:42:06","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":12277338,"visible":true,"origin":"","legend":"Supplementary information","description":"","filename":"STCLSupp23.docx","url":"https://assets-eu.researchsquare.com/files/rs-6524135/v1/d5e2daf66c09374c5c87ad75.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Spatiotemporal Decoding of Tumor Contactome Unveils a Microbial-Sensitive Ctrb1-Dependent Cell Competition Architecture Driving Epithelial-Mesenchymal Defense Against Cancer","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe fate of tumor initiation\u0026mdash;whether progressing to malignancy or being eradicated by host defenses\u0026mdash;hinges on the dynamic equilibrium between oncogenic transformation and cellular surveillance mechanisms. While immune-mediated tumor surveillance has been extensively characterized, the functional contributions of non-immune cellular components, particularly epithelial and mesenchymal cells, remain enigmatic. Seminal studies have revealed that normal epithelia possess intrinsic tumor-suppressive capacity through neighbor elimination programs, exemplified by epithelial defense against cancer (EDAC)(Ayukawa et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Tanimura and Fujita, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). However, whether mesenchymal lineages (e.g., fibroblasts, smooth muscle cells) participate in analogous surveillance programs, and how these cellular populations coordinate to establish tumor-suppressive microenvironments, constitutes a critical knowledge gap. Crucially, the spatiotemporal dynamics governing tumor-stromal interactions during early tumorigenesis, particularly the molecular logic determining competitive outcomes between malignant and host cells, await systematic elucidation.\u003c/p\u003e \u003cp\u003eCell competition\u0026mdash;an evolutionarily conserved quality control mechanism\u0026mdash;represents a fundamental mechanism for host non-immune surveillance against transformed cell(van Neerven and Vermeulen, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). This process enables supercompetitor cells to eliminate less-fit neighbors through direct interactions, playing context-dependent roles in both tumor suppression (\u003cem\u003ee.g.\u003c/em\u003e, epithelial homeostasis) and promotion (\u003cem\u003ee.g.\u003c/em\u003e, clonal selection)(Parker et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; van Neerven and Vermeulen, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). While oncogenic pathway rewiring (Wnt, Hippo, Notch) has been implicated in competitive interactions in a framework of EDAC(Moya et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Nakai et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; van Neerven and Vermeulen, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), the structural basis of tumor-host competition remains obscure. Two fundamental questions persist: (1) What spatial architectures mediate competitive interactions between incipient tumor cells and their microenvironment? (2) How do direct cell contacts coordinate multi-lineage (epithelial-mesenchymal) defense programs?\u003c/p\u003e \u003cp\u003eRecent methodological advances in proximity labeling (e.g., LIPSTIC(Nakandakari-Higa et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Pasqual et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), EXCELL(Ge et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Liu et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), FucoID(Liu et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Qiu et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)) have revolutionized cell-cell interaction mapping. Nevertheless, current tools face limitations in capturing transient physical interactions within complex microenvironments, particularly for non-immune cell types(Roux et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). The recent development of uLIPSTIC(Nakandakari-Higa et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), while enabling immune interaction profiling, lacks generalizability for studying epithelial-mesenchymal crosstalk. These technical constraints have hindered comprehensive analysis of direct interactomes governing tumor-stromal competition \u003cem\u003ein vivo\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eHere, through an orthotopic gastric cancer model, we identify a proliferating containment layer (PCL)\u0026mdash;a spatially organized epithelial-mesenchymal unit\u0026mdash;that orchestrates tumor-suppressive competition. To decipher PCL dynamics, we developed surface-anchored TurboID-mediated contact-dependent labeling (STCL), which integrated with single cell RNA sequence (scRNA-Seq), enables \u003cem\u003ein vivo\u003c/em\u003e mapping of direct cellular interactomes in tumor microenvironment (TME). This approach revealed a new type of non-immune surveillance \u003cem\u003ei.e.\u003c/em\u003e, epithelial-mesenchymal defense against cancer (EMDAC), a dual-lineage mechanism in which PCL constituents outcompete tumor cells through chymotrypsinogen B1 (Ctrb1)-mediated coordination of Myc/YAP signaling. Intriguingly, microbial infection subverts this surveillance machinery by suppressing Ctrb1 expression. Our findings (1) establish PCL as a structural framework for tumor-host competition; (2) define EMDAC as a non-immune surveillance paradigm; and (3) identify Ctrb1 as a microbial-sensitive rheostat of competitive fitness.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eI. Revealing Spatiotemporal Architecture of A Proliferative Containment Layer for EMDAC\u003c/h2\u003e \u003cp\u003eTo delineate non-immune surveillance mechanisms against tumorigenesis, we established an orthotopic gastric cancer model in which GFP-labeled MFC tumors underwent host-mediated clearance by D15 (day 15, same below) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea, \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea). Spatial profiling revealed a dynamically expanding GFP\u003csup\u003e\u0026minus;\u003c/sup\u003eKi67\u003csup\u003e+\u003c/sup\u003e \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003ep\u003c/span\u003eroliferative \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003ec\u003c/span\u003eontainment \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003el\u003c/span\u003eayer (PCL) surrounding regressing tumors (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb). Quantification demonstrated PCL's proliferative dominance: 59.3% (95% CI, 55.7\u0026thinsp;~\u0026thinsp;63.5%) Ki67\u003csup\u003e+\u003c/sup\u003e cells in PCL zone vs. 26.9% (95% CI, 22.5\u0026thinsp;~\u0026thinsp;32.7%) in tumor zone at D10 (p\u0026thinsp;=\u0026thinsp;0.0000, \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb), with progressive width expansion from D5 148.9 \u0026micro;m (95% CI, 140.6\u0026thinsp;~\u0026thinsp;169.8 \u0026micro;m) to D10 472.8 \u0026micro;m (95% CI, 391.3\u0026thinsp;~\u0026thinsp;585.6 \u0026micro;m, p\u0026thinsp;=\u0026thinsp;0.0000, Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec). These results suggest that PCL may represent a spatial barrier constraining tumor progression.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eSubsequently, our lineage-specific profiling revealed coordinated proliferation of epithelial and mesenchymal components in PCL. On one hand, the epithelial proliferation dynamics was shown by that CK8\u003csup\u003e+\u003c/sup\u003eKi67\u003csup\u003e+\u003c/sup\u003e cells dominated PCL regions 82.1\u0026thinsp;\u0026plusmn;\u0026thinsp;1.5% vs. 15.6\u0026thinsp;\u0026plusmn;\u0026thinsp;1.0% in adjacent epithelia at D10, (p\u0026thinsp;=\u0026thinsp;0.0000), with 2.5-fold area expansion from D5 to D10 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ed-e). On the other, the mesenchymal recruitment was reflected by that αSMA\u003csup\u003e+\u003c/sup\u003eKi67\u003csup\u003e+\u003c/sup\u003e fibroblasts occupied 79.0\u0026thinsp;\u0026plusmn;\u0026thinsp;2.4% of PCL vs. 14.2\u0026thinsp;\u0026plusmn;\u0026thinsp;1.7% in muscle layer at D10 (p\u0026thinsp;=\u0026thinsp;0.0000), showing 5.5-fold spatial expansion by D10 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ef-g). These results suggest dual-lineage contribution to PCL expansion.\u003c/p\u003e \u003cp\u003eTo further characterize the cellular origin of the PCL, we utilized an inducible reporter system to genetically trace epithelial cells and fibroblasts in mice bearing MFC-derived tumors. We generated several Cre-strain Rosa26-LSL-RFP reporter models: \u003cem\u003eClaudin18.2\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre/ERT2\u003c/em\u003e\u003c/sup\u003e;\u003cem\u003eR26\u003c/em\u003e\u003csup\u003e\u003cem\u003eRFP\u003c/em\u003e\u003c/sup\u003e for gastric epithelial cells, \u003cem\u003eαSMA\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre\u003c/em\u003e\u003c/sup\u003e;\u003cem\u003eR26\u003c/em\u003e\u003csup\u003e\u003cem\u003eRFP\u003c/em\u003e\u003c/sup\u003e for smooth muscle-derived fibroblasts, and \u003cem\u003eCol1a2\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre/ERT2\u003c/em\u003e\u003c/sup\u003e;\u003cem\u003eR26\u003c/em\u003e\u003csup\u003e\u003cem\u003eRFP\u003c/em\u003e\u003c/sup\u003e for total fibroblast populations. Tamoxifen-induced RFP expression allowed precise tracking of these cell types during tumor formation. The results showed that \u003cem\u003eClaudin18.2\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre/ERT2\u003c/em\u003e\u003c/sup\u003e-traced cells increased from 17.2\u0026thinsp;\u0026plusmn;\u0026thinsp;0.9% (D5) to 25.1\u0026thinsp;\u0026plusmn;\u0026thinsp;2.9% (D10, p\u0026thinsp;=\u0026thinsp;0.0034), among which the Ki67\u003csup\u003e+\u003c/sup\u003eRFP\u003csup\u003e+\u003c/sup\u003e cells augmented from 0.88\u0026thinsp;\u0026plusmn;\u0026thinsp;0.16% (D5) to 11.6\u0026thinsp;\u0026plusmn;\u0026thinsp;0.5% (D10, p\u0026thinsp;=\u0026thinsp;0.0000, Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eh \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec), confirming the epithelial contribution to PCL structure. Meanwhile, \u003cem\u003eαSMA\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre\u003c/em\u003e\u003c/sup\u003e-traced proliferating fibroblasts (Ki67\u003csup\u003e+\u003c/sup\u003eRFP\u003csup\u003e+\u003c/sup\u003e cells) dominated PCL (59.0\u0026thinsp;\u0026plusmn;\u0026thinsp;4.3% at D10, Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ei \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ed), confirming the mesenchymal recruitment. \u003cem\u003eCol1a2\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre/ERT2\u003c/em\u003e\u003c/sup\u003e;\u003cem\u003eR26\u003c/em\u003e\u003csup\u003e\u003cem\u003eRFP\u003c/em\u003e\u003c/sup\u003e mice further confirmed this pattern, showing 60.1\u0026thinsp;\u0026plusmn;\u0026thinsp;3.7% of PCL cells with an origin of fibroblasts (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ee). These findings confirm that both epithelial and mesenchymal cell types contribute significantly to PCL formation.\u003c/p\u003e \u003cp\u003eTogether, these data indicate that spatiotemporal coordination between epithelial proliferation and stromal activation underpins PCL-mediated tumor containment, a process we termed as \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003ee\u003c/span\u003epithelial-\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003em\u003c/span\u003eesenchymal \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003ed\u003c/span\u003eefense \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003ea\u003c/span\u003egainst \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003ec\u003c/span\u003eancer (EMDAC) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ej).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eII. Myc and YAP Signaling Cooperatively Drives EMDAC\u003c/h3\u003e\n\u003cp\u003eTo investigate the cell competition mechanisms associated with EMDAC, we performed single-cell RNA sequencing (scRNA-seq) and spatial transcriptomics on samples collected at D5, D10, and D15 during tumor clearance. These analyses revealed distinct cell types and their spatial distribution (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea, left panel). We observed a layered architecture comprising epithelial (EP), tumor zone, PCL, and mesenchymal (ME) layers early after MFC injection at D5 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea, right panel). The EP layer primarily consisted of normal epithelium (Epcam\u003csup\u003e+\u003c/sup\u003e) and metaplastic epithelium (Clu\u003csup\u003e+\u003c/sup\u003e), while the ME layer predominantly included smooth muscle cell-derived fibroblasts (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). PCL exhibited a high density of epithelial cells (\u003cem\u003ee.g.\u003c/em\u003e, pit cells with Cldn18\u003csup\u003e+\u003c/sup\u003e) and fibroblasts at D5, likely due to their reprogramming in response to tumor cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea, b). By D10 and D15, the PCL expanded toward the tumor core (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). Quantification of proliferating cells showed a stronger Ki67\u003csup\u003e+\u003c/sup\u003e index on PCL at D10 compared to D5 or D15 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec), suggesting its significant contribution to EMDAC. The ME layer exhibited consistent proliferation capacity throughout the process (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec), aligning with its predominant role in the PCL structure.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo validate that host cell proliferation is essential for EMDAC, we pretreated mice with 5-FU two days before tumor injection. We observed marked outgrowth of MFC tumors in 5-FU-pretreated mice compared to untreated controls (p\u0026thinsp;=\u0026thinsp;0.0070, \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). At D15, T cells (CD3e\u003csup\u003e+\u003c/sup\u003e) and myeloid cells (Kit\u003csup\u003e+\u003c/sup\u003e) predominantly located at the tumor periphery (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec), which was corroborated by immunostaining for F4/80 (macrophage) and CD4 (T cell) in tumor tissues from D10 (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed). These findings demonstrate that EMDAC occurs without reliance on the immune system as a primary anti-cancer defense mechanism.\u003c/p\u003e \u003cp\u003eNext, we systematically analyzed the activation status of key signaling pathways required for cell proliferation (Myc, TP53, Kras, Hippo-Yap, Notch, and Wnt) across different cell layers during EMDAC. Our results revealed that Myc signaling in the epithelial layer (referred to as Myc\u003csup\u003eEP\u003c/sup\u003e) and Yap signaling in the mesenchymal layer (Yap\u003csup\u003eME\u003c/sup\u003e) were specifically and gradually activated during this process (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee). Conversely, Notch signaling pathway was predominantly enriched in the tumor zone at D10 and D15, likely due to the clearance of tumor cells from these regions (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed). These findings indicate Myc\u003csup\u003eEP\u003c/sup\u003e and Yap\u003csup\u003eME\u003c/sup\u003e signaling as molecular mechanisms driving EMDAC.\u003c/p\u003e \u003cp\u003eSubsequently, we performed surgical separation of the epithelial layer and mesenchymal tissues adjacent to the tumor zone, as well as the tissues near the non-tumor region as controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee, \u003cb\u003eupper panel\u003c/b\u003e). We conducted qPCR analysis of downstream targets of Myc (\u003cem\u003ee.g.\u003c/em\u003e, \u003cem\u003eMyc\u003c/em\u003e and \u003cem\u003eCdk4\u003c/em\u003e) or Yap (\u003cem\u003ee.g.\u003c/em\u003e, \u003cem\u003eCtgf\u003c/em\u003e and \u003cem\u003eCyr61\u003c/em\u003e) in these regions. Our findings confirmed that \u003cem\u003eMyc\u003c/em\u003e and \u003cem\u003eCdk4\u003c/em\u003e were upregulated in the epithelial cells adjacent to the tumor region, while \u003cem\u003eCtgf\u003c/em\u003e and \u003cem\u003eCyr61\u003c/em\u003e were upregulated in the mesenchymal cells, without such activation in their respective control groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee, \u003cb\u003elower panel\u003c/b\u003e). These results demonstrate the spatial specificity of Myc and Yap signaling in driving cell proliferation in the EP and ME layers during EMDAC.\u003c/p\u003e \u003cp\u003eTo validate whether cell type-specific signaling pathways are required for EMDAC, we pretreated mice with specific inhibitors targeting either Myc (EN4 MYC inhibitor(Han et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2019a\u003c/span\u003e), S89228, 50 mg/kg) or Yap (Verteporfin(Liu-Chittenden et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), S80258, 10 mg/kg) for two weeks prior to orthotopic transplantation of MFC cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef). As expected, pretreatment with these inhibitors resulted in tumor persistence compared with the control group (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef-h), indicating that Myc and Yap signaling are essential during EMDAC. Furthermore, we generated \u003cem\u003eαSma\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre/ERT2\u003c/em\u003e\u003c/sup\u003e; \u003cem\u003eYap\u003c/em\u003e\u003csup\u003e\u003cem\u003eF/+\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eTaz\u003c/em\u003e \u003csup\u003e\u003cem\u003eF/+\u003c/em\u003e\u003c/sup\u003e genetic mice to specifically reduce Yap and Taz levels in smooth muscle cells prior to MFC orthotopic transplantation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ei). Consistent with the verteporfin treatment results, conditional knockdown of Yap/Taz in smooth muscle cells also led to significant tumor outgrowth (p\u0026thinsp;=\u0026thinsp;0.0000, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ei, j \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef) and PCL collapse (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eg), confirming that mesenchymal-specific Yap signaling is critical for EMDAC. Together, these findings identify Myc-dependent signaling in epithelial cells and Yap-dependent signaling in mesenchymal cells as essential pathways for EMDAC (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ek).\u003c/p\u003e\n\u003ch3\u003eIII. STCL Technology Deciphers Tumor Cell Direct Interactome during EMDAC\u003c/h3\u003e\n\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eDevelopment of STCL as a novel contact-dependent intercellular labeling strategy\u003c/h2\u003e \u003cp\u003eTo delineate direct intercellular communication between neoplastic cells and their epithelial/mesenchymal counterparts during EMDAC, we engineered a contact-dependent Surface-TurboID-mediated interCellular Labeling (STCL) system based on proximity-dependent biotinylation (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea). Building upon TurboID - an engineered biotin ligase optimized through directed evolution(Roux et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) with demonstrated efficacy in mapping organelle contact zones (\u003cem\u003ee.g.\u003c/em\u003e, ER-mitochondria interfaces(Cho et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), astrocyte-neuron synapses(Takano et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)) - we hypothesized that cell membrane-tethered outbound TurboID could achieve dual labeling modalities: (1) \u003cem\u003ecis\u003c/em\u003e-labeling of autologous membrane proteins, and (2) \u003cem\u003etrans\u003c/em\u003e-labeling of interacting partner cells through direct membrane contact.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFor STCL implementation, we designed a chimeric construct by fusing TurboID to the extracellular end of the transmembrane (TM) domain of human PDGFRβ (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea, \u003cb\u003eupper\u003c/b\u003e). We then stably expressed this PDGFR-TurboID cassette in MFC cells (hereafter referred as MFC\u003csup\u003eSTCL\u003c/sup\u003e, which showed proper membrane localization, as evidenced by 95.8\u0026thinsp;\u0026plusmn;\u0026thinsp;8.0% colocalization (Pearson's coefficient) between TurboID and WGA (a dye for membrane staining) (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea, \u003cb\u003elower\u003c/b\u003e). Biotin administration (12h) induced robust membrane-restricted biotinylation (94.3\u0026thinsp;\u0026plusmn;\u0026thinsp;9.0% WGA overlap), with signal specificity confirmed by three orthogonal detection methods: immunofluorescence (IF, \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea, \u003cb\u003elower\u003c/b\u003e), streptavidin western blot (WB, \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb), and flow cytometry (FC, \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec). Critically, the biotinylated signals were strictly dependent on TurboID expression and biotin administration, with little background or noise in parental MFC control cells.\u003c/p\u003e \u003cp\u003eTo validate the capacity of the STCL system in capturing direct intercellular contact events, we established a quantitative co-culture assay with stringent specificity controls. When MFC\u003csup\u003eSTCL\u003c/sup\u003e cells were co-cultured with eFluor670-prelabeled DC2.4 dendritic cells at 1:1 ratio (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb), biotin treatment (12h) induced robust \u003cem\u003ecis\u003c/em\u003e-labeling in 84.5\u0026thinsp;\u0026plusmn;\u0026thinsp;0.2% of MFC\u003csup\u003eSTCL\u003c/sup\u003e cells (n\u0026thinsp;=\u0026thinsp;3; \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec). Crucially, flow cytometric quantification revealed significant biotinylation in 54.2\u0026thinsp;\u0026plusmn;\u0026thinsp;1.0% of DC2.4 cells (vs. 5.0\u0026thinsp;\u0026plusmn;\u0026thinsp;0.6% in biotin-free controls; p\u0026thinsp;=\u0026thinsp;0.0000, Student's t-test), demonstrating contact-dependent \u003cem\u003etrans\u003c/em\u003e-labeling (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec,d). High-resolution confocal imaging further resolved micron-scale biotinylation foci exclusively at MFC\u003csup\u003eSTCL\u003c/sup\u003e-DC2.4 contact interfaces (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee), with contacting DC2.4 cells showing localized biotin signals.\u003c/p\u003e \u003cp\u003eThe generalizability of STCL was then confirmed using L929 fibroblasts, where \u003cem\u003etrans\u003c/em\u003e-labeling efficiency reached 20.4\u0026thinsp;\u0026plusmn;\u0026thinsp;0.9% (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed), with similar biotinylation foci at MFC\u003csup\u003eSTCL\u003c/sup\u003e-L929 contact interfaces (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ef). Importantly, post-co-culture magnetic sorting (purity\u0026thinsp;\u0026gt;\u0026thinsp;99%) of DC2.4 and L-929 cells eliminated potential contamination from MFC\u003csup\u003eSTCL\u003c/sup\u003e-derived biotin particles. Immunofluorescence of sorted cells confirmed membrane-localized biotinylation patterns (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee). These orthogonal validations exclude artifact from bystander transfer or extracellular vesicle uptake, confirming that STCL specifically labels physically contacted cells.\u003c/p\u003e \u003cp\u003eTo establish the physiological relevance of STCL, we implemented a multi-model validation strategy spanning subcutaneous and orthotopic tumor systems. In C57/BL6 mice bearing MC38\u003csup\u003eSTCL\u003c/sup\u003e-derived subcutaneous tumors (n\u0026thinsp;=\u0026thinsp;3/group), biotin administration (5 mg/kg, i.p., q.d.\u0026times;5) induced robust \u003cem\u003ecis\u003c/em\u003e-labeling evidenced by: (1) strong streptavidin WB and IF signal versus none in MC38\u003csup\u003eCtrl\u003c/sup\u003e tumors (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ef,g \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eg), and (2) High-dimensional cytometry revealed \u003cem\u003etrans\u003c/em\u003e-labeling efficiency of CAF, immune cells, and endothelial cells with 8.4\u0026thinsp;\u0026plusmn;\u0026thinsp;0.3% (vs. 0.8\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06% in MFC\u003csup\u003eCtrl\u003c/sup\u003e TME; p\u0026thinsp;=\u0026thinsp;0.0000), 42.3\u0026thinsp;\u0026plusmn;\u0026thinsp;3.4% (vs. 0.8\u0026thinsp;\u0026plusmn;\u0026thinsp;0.0% in MFC\u003csup\u003eCtrl\u003c/sup\u003e TME; p\u0026thinsp;=\u0026thinsp;0.0000), and 86.1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.3% (vs. 0.9\u0026thinsp;\u0026plusmn;\u0026thinsp;0.5% in MFC\u003csup\u003eCtrl\u003c/sup\u003e TME; p\u0026thinsp;=\u0026thinsp;0.0000), respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eh), with spatial resolution showing biotin foci precisely aligned with tumor-stromal/immune interfaces (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ei \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eh). Moreover, further cytometry analysis revealed variable \u003cem\u003etrans\u003c/em\u003e-labeling efficiency of each type of immune cells, among which B cells showed highest labeling efficiency (~\u0026thinsp;80%) (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ei-j).\u003c/p\u003e \u003cp\u003eFurthermore, we tested STCL in an orthotopic gastric cancer model (n\u0026thinsp;=\u0026thinsp;6, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ej). HE-guided microdissection showed 33.1\u0026thinsp;\u0026plusmn;\u0026thinsp;3.5% of TME cells exhibited biotin\u003csup\u003e+\u003c/sup\u003eTurboID\u003csup\u003e\u0026minus;\u003c/sup\u003e signatures, confirming tumor-specific trans-labeling (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ek). Multispectral imaging quantified contact-dependent biotin transfer to αSMA\u003csup\u003e+\u003c/sup\u003e fibroblasts (13.7\u0026thinsp;\u0026plusmn;\u0026thinsp;0.8%), CD11b\u003csup\u003e+\u003c/sup\u003e myeloid cells (35.7\u0026thinsp;\u0026plusmn;\u0026thinsp;1.8%), and CD3\u003csup\u003e+\u003c/sup\u003e T cells (24.9\u0026thinsp;\u0026plusmn;\u0026thinsp;2.3%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ei \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ek).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eIntegrating STCL with scRNA to decipher direct cellular interactome governing EMDAC\u003c/h3\u003e\n\u003cp\u003eBy integrating the STCL technology with single-cell transcriptomics, we attempted to establish a spatiotemporal map of direct cellular interactomes (\u003cspan type=\"ItalicUnderline\" class=\"ItalicUnderline\" name=\"Emphasis\"\u003econtactome\u003c/span\u003e) governing EMDAC. The study workflow included harvesting whole stomach tissue at D5, D10, and D15, followed by single-cell isolation using magnetic-activated cell sorting to distinguish biotin\u003csup\u003e+\u003c/sup\u003e (interacting) from biotin\u003csup\u003e\u0026minus;\u003c/sup\u003e (bystander) cells, which were subsequently analyzed for gene expression profiles (scRNA-seq) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea). From the total of 35,172 single-cell samples obtained, we identified 10 distinct cell clusters (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea), each annotated using top principal component scores and lineage-specific markers.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eQuantitative analysis revealed that tumor cells interacted preferentially with certain host cell types during EMDAC progression (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb). Specifically, STCL labeling efficiency for epithelial, endothelial, and fibroblast populations showed a consistent increase from D5 to D10, followed by gradual reduction at D15 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec). Meanwhile, tumor cell labeling efficiency decreased from 85\u0026ndash;90% at D5 and D10 to approximately 40% at D15 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec), indicating a decline in \u003cem\u003ecis\u003c/em\u003e-labeling as tumor cells increasingly interacted with host epithelial and mesenchymal compartments during EMDAC. Notably, epithelial cells represent a predominant (\u0026gt;\u0026thinsp;80%) among all \u003cem\u003etrans\u003c/em\u003e-labeled cells across the entire timeline of the EMDAC process (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec).\u003c/p\u003e \u003cp\u003eSubsequently, we examined the spatial organization of different cell types in the TME at D10 when STCL-based labeling was strong and efficient (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec). A striking observation was the significant increase in CK8 immunofluorescence signals in the PCL (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed). However, the number of biotin\u003csup\u003e+\u003c/sup\u003eCK8\u003csup\u003e+\u003c/sup\u003e cells gradually decreased as the distance from the tumor boundary increased. Importantly, the spatial distribution of these cells mirrored the STCL labeling pattern (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb), with fibroblasts (αSMA\u003csup\u003e+\u003c/sup\u003e) and endothelial cells (CD31\u003csup\u003e+\u003c/sup\u003e) predominantly located within the PCL, while myeloid cells (CD11b\u003csup\u003e+\u003c/sup\u003e) and T cells (CD4\u003csup\u003e+\u003c/sup\u003e) were more abundant at the periphery (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed). Furthermore, quantitative analysis revealed that tumor cells interacted with various types of host cells through direct physical contact, with frequencies comparable to those determined by scRNA-seq (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed). Notably, epithelial and fibroblast cells emerged as the two most frequently contacted cell types in this context (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ee). Together, these findings provide a dynamic snapshot of tumor contactome, \u003cem\u003ei.e.\u003c/em\u003e, tumor cell direct interactions with other types of cells within the TME during EMDAC.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eTumor contacts activate EMDAC by conferring epithelial cells and fibroblasts proliferative capacity\u003c/h2\u003e \u003cp\u003eAs epithelial cells (EP) and fibroblasts constituted the top-ranked direct interactors with tumor cells, we systematically mapped their interplay. UMAP-based stratification identified 14 EP clusters through cell-type-specific signatures (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea,b), encompassing parietal/pre-parietal, pit, neck/pre-neck, isthmus stem, chief, neuroendocrine, and intestinal metaplastic (IM) lineages. Two emergent populations dominated functional analyses: a hyper-proliferative cluster (Ki67\u003csup\u003e+\u003c/sup\u003ePcna\u003csup\u003e+\u003c/sup\u003e; EP_1) and a metabolically active population (Acat1\u003csup\u003e+\u003c/sup\u003eApoa1\u003csup\u003e+\u003c/sup\u003eEno1\u003csup\u003e+\u003c/sup\u003e; EP_2) (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea-c). Systematic annotation consolidated these into eight definitive subpopulations using canonical and computational markers: Tumor, PAC (parietal lineages), Neck/Stem, IM, PIC (pit cells), EP_1, EP_2, and neuroendocrine cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea, right panel).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eNotably, EP_1 and EP_2 emerged at D5 and expanded progressively during EMDAC (\u003cb\u003eExtended Data 5a\u003c/b\u003e). EP_1 displayed escalating \u003cem\u003etrans\u003c/em\u003e-labeling efficiency (Bio\u003csup\u003e+\u003c/sup\u003e/total) despite declining tumor burden (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ed). Pseudotemporal mapping revealed EP_1's origin from pre-neck, neck, Cldn18\u003csup\u003e+\u003c/sup\u003eIM, and isthmus stem precursors (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ee, \u003cb\u003eExtended Data 5b\u003c/b\u003e). Crucially, these precursors exhibited higher D5 \u003cem\u003etrans\u003c/em\u003e-labeling efficiency than EP_1 itself (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ef, \u003cb\u003eExtended Data 5c\u003c/b\u003e), indicating that tumor interactions precede hyper-proliferative conversion.\u003c/p\u003e \u003cp\u003eWe next deconvoluted tumor-proximal (biotin\u003csup\u003e+\u003c/sup\u003e) versus bystander (biotin\u003csup\u003e\u0026minus;\u003c/sup\u003e) fibroblasts across cell clusters. Integrative marker analysis resolved five CAF states: myCAF (myofibroblast), iCAF (inflammatory fibroblast), apCAF (antigen-presenting fibroblast), and eCAF (epithelial-like CAF expressing \u003cem\u003eCd24a\u003c/em\u003e, \u003cem\u003eCdh1\u003c/em\u003e, \u003cem\u003eKrt19\u003c/em\u003e, \u003cem\u003eFxyd3\u003c/em\u003e and \u003cem\u003eEpcam\u003c/em\u003e) (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ed,e). Strikingly, a subset of eCAF showed the strongest tumor-derived biotinylation (Bio\u003csup\u003ehigh\u003c/sup\u003e eCAF; Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eg,h), marking their privileged tumor interface in the contactome.\u003c/p\u003e \u003cp\u003eDynamic tracking revealed eCAF \u003cem\u003etrans\u003c/em\u003e-labeling peaked during early tumor engagement (D5-D10) but declined post-clearance, while Bio\u003csup\u003ehigh\u003c/sup\u003e eCAF sustained persistent tumor interactions (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ei). GSVA (gene set variation analysis) of Bio\u003csup\u003ehigh\u003c/sup\u003e eCAF demonstrated enriched proliferation pathways (Ki67/Pcna; \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ef, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ej), implicating tumor contact as a driver of CAF hyperproliferation. This mechanism was functionally validated by EDU pulse-chase assays (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eg): proliferative CAFs directly contacting tumors peaked at D5 but diminished upon tumor eradication at D15 (D15, p\u0026thinsp;=\u0026thinsp;0.0755), mirroring the Bio\u003csup\u003ehigh\u003c/sup\u003e eCAF trajectory.\u003c/p\u003e \u003cp\u003eTaken together, these data indicated that physical contacts by tumor cells activate EMDAC by conferring epithelial cells and fibroblasts proliferative capacity.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eIV. Ctrb1 as a Microbial-Sensitive Surveillance Hub\u003c/h3\u003e\n\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eCtrb1 emerges as a molecular driver for PCL-mediated EMDAC\u003c/h2\u003e \u003cp\u003eHaving established PCL's structural and functional basis for EMDAC, we profiled the gene expression of proliferating cells in the PCL and the identified chymotrypsinogen B1 (Ctrb1) as a key driver of PCL formation (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb \u003cb\u003eand\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea). While physiologically restricted to pancreatic tissue (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea,b), Ctrb1 was ectopically expressed in tumor-interacted EP_1 cells and fibroblasts, showing a 2.1-fold enrichment in the biotin\u003csup\u003e+\u003c/sup\u003e population compared to bystander EP_1 cells and over a 10-fold enrichment compared to bystander fibroblasts by D10 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea,b). This ectopic expression followed a tumor clearance-associated trajectory: Ctrb1 levels rose sharply from D5, peaked at D10 during maximal PCL expansion (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb,c), and correlated with proliferative activity (30% co-expression with Ki67 in PCL cells; Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec). Lineage tracing in \u003cem\u003eαSma\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre/ERT2\u003c/em\u003e\u003c/sup\u003e;\u003cem\u003eR26\u003c/em\u003e\u003csup\u003e\u003cem\u003eRFP\u003c/em\u003e\u003c/sup\u003e mice confirmed this dynamic \u0026ndash; 44.0% of smooth muscle-derived (RFP\u003csup\u003e+\u003c/sup\u003e) PCL cells co-expressed Ctrb1 and Ki67 by D10 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ed, \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ee).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eImportantly, tumor presence induced 4.2- and 3.1-fold upregulation of \u003cem\u003eCtrb1\u003c/em\u003e transcription in peri-PCL EP and ME compartments compared to tumor-free controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ee), positioning it as a tumor-responsive effector. The spatiotemporal alignment of Ctrb1 expression with 1) PCL biogenesis (D5-D10), 2) proliferative conversion of stromal cells, and 3) tumor competition dynamics (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb) suggests its dual role as both architectural organizer and mitotic accelerator in EMDAC.\u003c/p\u003e \u003cp\u003eTo establish causality between tumor contact and Ctrb1 induction, we deployed direct co-culture systems. Gastric epithelial cells (GEC) contacting MFC tumors exhibited 9.1-fold \u003cem\u003eCtrb1\u003c/em\u003e upregulation versus controls (qPCR/IFA; p\u0026thinsp;=\u0026thinsp;0.0037; Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ef-h). Crucially, this response required physical contact, as Transwell-separated co-cultures showed no induction (p\u0026thinsp;=\u0026thinsp;0.1372, \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ef). Fibroblasts displayed even greater contact-dependent Ctrb1 activation (57.3-fold vs control; 6.3-fold vs GEC) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, f-h \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ef), revealing a conserved tumor-sensing mechanism across lineages. Genetic validation in \u003cem\u003eCtrb1\u003c/em\u003e\u003csup\u003e\u003cem\u003eF/F\u003c/em\u003e\u003c/sup\u003e mice demonstrated its non-redundant role: AAV-mediated stomach-specific Ctrb1 knockout (qPCR/WB; \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eg,h) caused 2.2-fold tumor enlargement (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ei,j \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ei) and PCL collapse (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ej). Strikingly, Ctrb1 loss abolished host-mediated tumor clearance, confirming its gatekeeper function in EMDAC (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ej). These results demonstrate that cell contact-dependent Ctrb1 activation governs antitumor competence during EMDAC.\u003c/p\u003e \u003cp\u003eBuilding on our observations that epithelial Myc and mesenchymal Yap signaling are indispensable for EMDAC, and that tumor cell contact specifically upregulates Ctrb1 in these stromal compartments, we hypothesized that Ctrb1 might function as a convergence point for these oncogenic pathways. Systematic analysis of the Ctrb1 promoter uncovered conserved Myc-binding E-box elements and Yap/TEAD-binding M-CAT elements (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ek). To functionally validate these predictions, we performed chromatin immunoprecipitation (ChIP)-qPCR in contact-activated epithelia and stromal cells. This revealed 10.7-fold (p\u0026thinsp;=\u0026thinsp;0.0069) and 2.8-fold (p\u0026thinsp;=\u0026thinsp;0.0157) enrichment of Myc and Yap/Tead4 binding, respectively, at the Ctrb1 promoter compared to non-contact controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ek). More importantly, conditional Yap/Taz ablation in mesenchymal cells reduced Ctrb1\u003csup\u003e+\u003c/sup\u003eKi67\u003csup\u003e+\u003c/sup\u003e proliferating cells by 79.8\u0026thinsp;\u0026plusmn;\u0026thinsp;6.1% (p\u0026thinsp;=\u0026thinsp;0.0015) in orthotopic MFC transplantation models (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003el, \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eg). Notably, epithelial-specific Myc inhibition produced comparable suppression of Ctrb1\u003csup\u003e+\u003c/sup\u003eKi67\u003csup\u003e+\u003c/sup\u003e populations (8.2-fold lower than DMSO group, \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ek).\u003c/p\u003e \u003cp\u003eCollectively, these results establish Ctrb1 as a cell contact-responsive effector of tumor-stroma crosstalk, with its transcriptional activation being spatially regulated: epithelial cells employ Myc signaling while mesenchymal counterparts utilize Yap-Tead4 complex to drive EMDAC progression (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003em).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eCtrb1 boosts PCL cell fitness to outcompete tumor cells during EMDAC\u003c/h2\u003e \u003cp\u003eHaving established the tumor cell contact-dependent induction of Ctrb1 in EMDAC, we next sought to delineate the cell-autonomous mechanisms underlying Ctrb1-mediated fitness enhancement. A 293FT isogenic competition model was established by generating cells stably expressing CTRB1-P2A-GFP (293FT\u003csup\u003eCTRB1\u0026thinsp;\u0026minus;\u0026thinsp;GFP\u003c/sup\u003e; \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea-b) and co-culturing them 1:1 with parental controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea). Strikingly, 293FT\u003csup\u003eCTRB1\u0026thinsp;\u0026minus;\u0026thinsp;GFP\u003c/sup\u003e cells progressively dominated the co-culture, achieving 74.35\u0026thinsp;\u0026plusmn;\u0026thinsp;2.9% dominance by D7 (vs 49.9\u0026thinsp;\u0026plusmn;\u0026thinsp;5.1% in empty vector controls, p\u0026thinsp;=\u0026thinsp;0.0000; Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb), suggesting a selective growth advantage. Moreover, this fitness phenotype translated directly to \u003cem\u003ein vivo\u003c/em\u003e tumorigenic potential. Orthotopically transplanted MFC\u003csup\u003eSTCL\u0026minus;Ctrb1\u003c/sup\u003e cells (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ec) exhibited 4.2-fold larger tumor volumes than wild-type counterparts by D12 (p\u0026thinsp;=\u0026thinsp;0.0100; Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ec), accompanied by complete escape from host-mediated tumor elimination (0% regression vs 74.4% in controls, p\u0026thinsp;=\u0026thinsp;0.0000) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ec \u003cb\u003eand Extended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ed).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo delineate the molecular basis of Ctrb1-mediated cellular fitness, we conducted comparative transcriptomic profiling between isogenic 293FT and CTRB1-overexpressing clones (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ed, \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ee). KEGG pathway enrichment revealed CTRB1's global modulation of oncogenic signaling hubs - including Hippo-YAP, WNT/β-catenin, MAPK, mTOR, NF-κB and Notch pathways (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ee). Strikingly, hierarchical analysis identified pathway-specific transcriptional amplifiers: Hippo-YAP effectors AJUBA (mechanical stress sensor)(Rauskolb et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) and pro-apoptotic BBC3/PUMA(Subramaniam et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), along with WNT regulators NOTUM (extracellular β-catenin inhibitor) and SERPINF1/PEDF (angiogenesis modulator)(Liu et al., 187), emerged as top-tier CTRB1-responsive nodes (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ef, qPCR-validated in \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ef). Notably, our discovery of NOTUM upregulation establishes an evolutionarily conserved axis - as this tumor-derived WNT antagonist has been shown to orchestrate pre-neoplastic niche remodeling through paracrine suppression of competing epithelial clones in intestinal carcinogenesis models(Flanagan et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2021a\u003c/span\u003e; Flanagan et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2021b\u003c/span\u003e; Pentinmikko et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; van Neerven et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Yum et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). This mechanistic convergence between CTRB1 modality and established oncogenic programs suggests a novel feedforward signaling mechanism for cellular selection during tumor initiation.\u003c/p\u003e \u003cp\u003eSubsequently, we performed affinity purification mass spectrometry (AP-MS) in 293FT cells expressing Flag-CTRB1 (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eg-j). Strikingly, the CTRB1 protein complex coalesced core signalosomes across five major pathways - STAT3 (Hippo cross-talk), GSK3β (WNT node), mTOR (metabolic master regulator), NF-κB1 (inflammatory switch), and MAPKAPK3 (AMPK interface) (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ej). Systematic deconvolution of interactome topology demonstrated three functional strata: EGFR inhibitor resistance (p\u0026thinsp;=\u0026thinsp;0.0368), Protein export (p\u0026thinsp;=\u0026thinsp;0.0336), Pancreatic cancer (p\u0026thinsp;=\u0026thinsp;0.0319), mTOR signaling pathway (p\u0026thinsp;=\u0026thinsp;0.0294), Vesicular transport (p\u0026thinsp;=\u0026thinsp;0.0066), Nucleocytoplasmic transport (p\u0026thinsp;=\u0026thinsp;0.0000), Protein processing in ER (p\u0026thinsp;=\u0026thinsp;0.0000) (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ek). Compartmental mapping further revealed ER-nucleated signaling hubs, with 74.6% of interactors residing in membrane-bound organelles (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003el). Given the ER's emerging role as a topological conductor for molecular transport and signal integration(Ron and Walter, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Schwarz and Blower, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), our findings posit CTRB1 as a multi-pathway licensor that spatially coordinates proliferative signaling through: (i) physical scaffolding of key kinases/transcription factors at ER membranes, and (ii) coupling signal activation with secretory trafficking during clonal selection.\u003c/p\u003e \u003cp\u003eCollectively, these data demonstrate that Ctrb1 functions as a cell-intrinsic fitness amplifier through enhanced proliferative capacity and multiple oncogenic signaling.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eMicrobial infection impairs Ctrb1-driven EMDAC to promote tumorigenesis\u003c/h2\u003e \u003cp\u003eGiven the emerging roles of pathobionts like \u003cem\u003eStreptococcus anginosus\u003c/em\u003e (\u003cem\u003es.a.\u003c/em\u003e)(Fu et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Yuan et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zhou et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) and \u003cem\u003eCandida albicans\u003c/em\u003e (\u003cem\u003ec.a.\u003c/em\u003e)(Dohlman et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) in gastric oncogenesis and context-dependent expression of Ctrb1, we hypothesized that microbial dysbiosis might subvert Ctrb1-mediated EMDAC. To verify this host-pathogen crosstalk paradigm, we established a tripartite experimental axis: 1) 4-week oral pathogen challenge with \u003cem\u003es.a.\u003c/em\u003e or \u003cem\u003ec.a.\u003c/em\u003e, 2) orthotopic implantation of MFC\u003csup\u003eSTCL\u003c/sup\u003e cells, and 3) multi-parametric analysis of EMDAC dynamics (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eg). Pathogen-engrafted mice developed microbial persistence with complete failure of tumor clearance (vs. sterile controls), establishing a pathogen-dependent oncogenic permissiveness (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eh,i). Mechanistically, microbial colonization eroded the EMDAC architecture - evidenced by 34.1% (\u003cem\u003es.a.\u003c/em\u003e) and 27.7% (\u003cem\u003ec.a.\u003c/em\u003e) reduction in PCL territories and 87.5% (\u003cem\u003es.a.\u003c/em\u003e) and 89.6% (\u003cem\u003ec.a.\u003c/em\u003e) depletion of Ki67\u003csup\u003e+\u003c/sup\u003eCtrb1\u003csup\u003e+\u003c/sup\u003e sentinel cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ej, \u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003em-o). Further immunostaining of Ki67 revealed pathogen-driven competitive reprogramming: tumor foci exhibited 2.7-fold (\u003cem\u003es.a.\u003c/em\u003e) and 2.0-fold (\u003cem\u003ec.a.\u003c/em\u003e) elevated proliferation compared to adjacent PCL (\u003cb\u003eExtended Data\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eo), suggesting microbiota-induced fitness switching. These findings coalesce into a pathogenic triad model where microbial challenge: 1) silences Ctrb1-mediated surveillance, 2) disables EMDAC-mediated tumor elimination, and 3) creates permissive niche for malignant clonal expansion (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ek).\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eReconceptualizing Host Defense: The EMDAC Paradigm for Non-Immune Surveillance\u003c/h2\u003e \u003cp\u003eOur study unveils Epithelial-Mesenchymal Defense Against Cancer (EMDAC) as a spatiotemporally coordinated host surveillance program that transcends canonical immune-mediated mechanisms. While prior work established Epithelial Defense Against Cancer (EDAC) as a frontline mechanism against preneoplastic clones(Ayukawa et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Lima and Rodriguez, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Moya et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; van Neerven and Vermeulen, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), we demonstrate that mesenchymal co-option transforms this into a multi-lineage containment strategy. Our discovery of the PCL cell competition architecture - with its dual-lineage dynamics (epithelial expansion + fibroblast mobilization) and Myc\u003csup\u003eEP\u003c/sup\u003e-Yap\u003csup\u003eME\u003c/sup\u003e signaling axes - redefines the conceptual framework of non-immune tumor surveillance (\u003cb\u003eExtended Data Fig.\u0026nbsp;8\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eBeyond epithelial surveillance, our study supports a multi-layered defense paradigm in which coordinated action of epithelial-mesenchymal compartments assemble into proliferative barrier structures (PCL) to physically constrain tumors. This multicellular architecture distinguishes EMDAC from classical epithelial surveillance, \u003cem\u003ei.e.\u003c/em\u003e, EDAC, by integrating stromal reprogramming. Our lineage tracing and spatial transcriptomics data demonstrate mesenchymal fibroblasts - particularly epithelial-like eCAFs with enhanced tumor-proximal activity and proliferative superiority over canonical CAF subtypes - as indispensable partners in forming the PCL. This architectural innovation expands the \"guardian cell\" repertoire beyond epithelial sentinels, suggesting mesenchymal lineages actively participate in tumor suppression through structural remodeling. During EMDAC, coordinated activation of Myc\u003csup\u003eEP\u003c/sup\u003e-Yap\u003csup\u003eME\u003c/sup\u003e signaling bifurcation establishes lineage-restricted proliferation programs required for PCL formation. In this regard, the PCL's compartmentalized signaling architecture, with Myc\u003csup\u003eEP\u003c/sup\u003e/YAP\u003csup\u003eME\u003c/sup\u003e activation mirroring embryonic morphogenetic fields, provides a biochemical blueprint for host-tumor boundary formation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eCtrb1: A Master Regulator for Cellular Fitness\u003c/h2\u003e \u003cp\u003eMyc and Yap signaling not only drive cell proliferation but also converge on Ctrb1, a key molecular effector of EMDAC. Ctrb1, normally restricted to the pancreas(Eizirik et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), is ectopically expressed in tumor-adjacent epithelial and mesenchymal cells, enhancing their fitness to outcompete tumor cells. This discovery positions Ctrb1 as a biomarker for supercompetitor cells in host defense against cancer. The serendipitous discovery of Ctrb1 as an ectopically expressed surveillance effector in gastric EMDAC challenges conventional views of its pancreatic-restricted function. Our findings position Ctrb1 as: 1) a contact-dependent danger signal triggered by tumor adjacency; 2) a signaling nexus integrating multiple oncogenic pathways with broader proliferative networks; 3) a microbial vulnerability node linking infection to carcinogenesis.\u003c/p\u003e \u003cp\u003eNotably, Ctrb1's ability to simultaneously activate Hippo/YAP, Wnt, and Myc pathways suggests it functions as a proteolytic signaling amplifier - a hypothesis supported by its interaction with ER-associated signalosomes. This \"moonlighting\" protease activity may explain its paradoxical roles in both digestive physiology and tumor suppression, reminiscent of matrix metalloproteinases' dual functions in tissue remodeling and cancer. The ectopic induction of Ctrb1 in gastric stroma represents a paradigm shift in understanding tissue-specific tumor defense. While pancreatic-restricted under homeostasis, Ctrb1 becomes a tumor-contact sensor across lineages - epithelial cells show 9.1-fold induction upon direct tumor interaction, while fibroblasts exhibit 57.8-fold upregulation. This spatial specificity aligns with its role as a molecular rheostat: Ctrb1 coordinates Myc/Yap-driven hyperproliferation in the PCL while suppressing tumor-promoting Wnt signaling through NOTUM induction. The protease's dual capacity to 1) amplify host cell fitness via Hippo/Myc/p53 axis and 2) antagonize tumor-supportive niches through paracrine factors positions it as a pleiotropic regulator of EMDAC.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eSTCL: Rewiring Proximity Labeling for Spatiotemporal Mapping of Tumor Contactome\u003c/h2\u003e \u003cp\u003eCell-cell interactions are crucial for both tissue homeostasis and disease development(Bechtel et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Boareto, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Su et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Our development of the STCL system represents a significant advancement in mapping direct cell-cell interactions, \u003cem\u003ei.e.\u003c/em\u003e, physical contactome. Unlike previous methods(Chudnovskiy et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Liu et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Liu et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Nakandakari-Higa et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Pasqual et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Takano et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), STCL allows unbiased, ligand-receptor-independent labeling of physically contacted cells, providing a comprehensive contactome in disease models. Compared to existing tools, STCL's membrane-anchored TurboID design minimizes intracellular labeling artifacts while maintaining compatibility with complex \u003cem\u003ein vivo\u003c/em\u003e models. Therefore, the STCL platform overcomes critical limitations in existing cell interaction tools by enabling: 1) Ligand-receptor agnostic profiling - capturing transient tumor-stromal contacts likely missed by LIPSTIC/FucoID systems; 2) Temporal resolution - mapping interaction dynamics during tumor clearance phases (Days 5–15); 3) \u003cem\u003eIn vivo\u003c/em\u003e scalability - generating organ-specific contactome without requiring dual genetic manipulation.\u003c/p\u003e \u003cp\u003eBy integrating STCL with scRNA-seq and spatial transcriptomics, we deciphered the dynamic tumor contactome during EMDAC, revealing preferential interactions between tumor cells and epithelial/fibroblast populations. The identification of eCAFs as privileged tumor interactors exemplifies STCL utility in discovering non-canonical cellular crosstalk. Notably, our application of STCL in gastric models revealed an unexpected dominance of non-immune interactions (87.3% epithelial/stromal vs 9.6% immune) at D10, challenging the immunotherapy-centric view of tumor microenvironment crosstalk.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eTranslational Implications\u003c/h2\u003e \u003cp\u003eOur findings establish EMDAC as a blueprint for engineering multicellular defense programs - a frontier that may complement current immune-centric approaches in precision oncology. In this regard, our study illuminates three translational axes: 1) Ctrb1 as a prognostic biomarker - Its expression dynamics correlate with PCL integrity and survival outcomes; 2) Microbial modulation - Pathogen-mediated Ctrb1 suppression provides mechanistic basis for infection-associated gastric cancer; 3) Therapeutic targeting - Pharmacological stabilization of PCL architecture could enhance endogenous tumor surveillance.\u003c/p\u003e \u003cp\u003eWhile EMDAC demonstrates tumor suppression in an orthotopic gastric cancer model, three key questions emerge: 1) Tissue plasticity - Can Ctrb1-mediated defense be reactivated in metastasized tumors? Preliminary data show circulating Ctrb1 levels correlate with gastric cancer prognosis; 2) Microbiome crosstalk - How do \u003cem\u003eH. pylori\u003c/em\u003e-induced Ctrb1 fluctuations impact EMDAC efficacy? 3) Therapeutic hijacking - Can synthetic Ctrb1 analogs boost endogenous tumor suppression in immunotherapy-resistant cancers?\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eLimitations and Future Directions\u003c/h2\u003e \u003cp\u003eWhile our study provides novel insights into EMDAC, several questions remain. First, the universality of EMDAC across other epithelial tissues (\u003cem\u003ee.g.\u003c/em\u003e, gut, liver, breast) warrants investigation. Second, the mechanisms underlying epithelial and mesenchymal cell recruitment to the PCL are unclear. Third, the potential role of Ctrb1 in regulating nuclear transport of transcriptional factors requires further exploration. Finally, the application of STCL to other cell types and disease models will expand our understanding of intercellular communication in health and disease. Complementary application of STCL with Cre/Lox systems will reveal bidirectional communication networks.\u003c/p\u003e \u003cp\u003eIn conclusion, our study unveils EMDAC as a novel non-immune surveillance mechanism, identifies Ctrb1 as a key molecular driver, and introduces STCL as a powerful tool for mapping cell-cell interactions. These findings provide a foundation for developing new therapeutic strategies to enhance host defense against cancer.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003cdiv id=\"Sec20\" class=\"Section3\"\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003cdiv id=\"Sec22\" class=\"Section3\"\u003e \u003c/div\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section2\"\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Methods","content":"\u003ch2\u003eCells\u003c/h2\u003e\u003cp\u003eHEK293FT cells were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). GEC cells were sourced from Biofeng (Hunan, China), and MFC cells were acquired from the National Infrastructure of Cell Line Resource (Beijing, China). L929 and 3T3-L1 cell lines were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA), while the DC2.4 cell line was obtained from Sigma-Aldrich (St. Louis, MO, USA).\u003c/p\u003e\u003cp\u003eMFC, GEC, and DC2.4 cells were cultured in RPMI 1640 medium (Invitrogen, Carlsbad, CA, USA), supplemented with 10% fetal bovine serum (FBS; Biological Industries, Kibbutz Beit Haemek, Israel) and 1% penicillin/streptomycin (Thermo Fisher Scientific, Waltham, MA, USA). All other cell lines were maintained in Dulbecco’s Modified Eagle Medium (DMEM; Invitrogen) under the same supplementation conditions. Cells were incubated at 37°C in a humidified atmosphere with 5% CO2 using a Thermo Fisher Scientific incubator (Waltham, MA, USA).\u003c/p\u003e\u003ch2\u003eMice and Genotyping\u003c/h2\u003e\u003ch2\u003eMouse Lines and Sources\u003c/h2\u003e\u003cp\u003e \u003cem\u003eCtrb1\u003c/em\u003e \u003csup\u003e \u003cem\u003eF/F\u003c/em\u003e \u003c/sup\u003e mice were generated by the Shanghai Model Organisms Center (SMOC, Shanghai, China). \u003cem\u003eClaudin18.2\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre/ERT2\u003c/em\u003e\u003c/sup\u003e mice (Stock No. T056749) were obtained from GemPharmatech (Jiangsu, China). \u003cem\u003eCol1a2\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre/ERT2\u003c/em\u003e\u003c/sup\u003e mice (Stock, 029567) were obtained from Jackson Lab. \u003cem\u003eR26\u003c/em\u003e\u003csup\u003e\u003cem\u003eRFP\u003c/em\u003e\u003c/sup\u003e mice (Rosa26-LoxP-STOP-LoxP-RFP) (Han et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2019b\u003c/span\u003e) were kindly provided by Dr. Bin Zhou (CAS Center for Excellence in Molecular Cell Science, Shanghai). \u003cem\u003eαSMA\u003c/em\u003e\u003csup\u003e\u003cem\u003eCre\u003c/em\u003e\u003c/sup\u003e mice were a generous gift from Dr. Gang Wang (Fudan University, Shanghai). \u003cem\u003eYap1\u003c/em\u003e\u003csup\u003e\u003cem\u003eF/F\u003c/em\u003e\u003c/sup\u003e and \u003cem\u003eTaz\u003c/em\u003e\u003csup\u003e\u003cem\u003eF/F\u003c/em\u003e\u003c/sup\u003e mice were previously described(Tang et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003ch2\u003eTamoxifen Administration\u003c/h2\u003e\u003cp\u003eCre/ERT2 recombinase activity was induced by intraperitoneal injection of tamoxifen (Sigma-Aldrich, St. Louis, MO, USA) at specified time points to activate lineage tracing.\u003c/p\u003e\u003ch2\u003eGenotyping\u003c/h2\u003e\u003cp\u003eGenotyping was performed using PCR with primers listed in \u003cb\u003eSupplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e.\u003c/p\u003e\u003ch2\u003eAnimal Housing and Ethical Approval\u003c/h2\u003e\u003cp\u003eAll mice used in this study were maintained on a C57BL/6 genetic background, except for the 615 strains. Both male and female mice aged 4–8 weeks were included in the analyses. Mice were housed in individually ventilated cages under group housing conditions whenever possible. They were maintained in a controlled environment with a 12-h light/dark cycle and provided ad libitum access to water and standard rodent chow.\u003c/p\u003e\u003cp\u003e Randomization of mice into experimental groups was performed in strict accordance with the guidelines of the Institutional Animal Care and Use Committee (IACUC). All animal procedures were approved by the IACUC of Fudan University (approval ID: IDM2022037) and Tongji University (approval ID: SHDSYY-2023-P0011), ensuring compliance with ethical standards for animal research.\u003c/p\u003e\u003ch2\u003eStatistical Analysis\u003c/h2\u003e\u003cp\u003eSample size estimation for planned comparisons of two independent means was performed using a two-tailed test in GraphPad Prism 9 (GraphPad Software, San Diego, CA, USA). Continuous variables are presented as mean ± standard deviation (s.d.), while categorical variables are reported as frequencies and proportions. Statistical comparisons of continuous data were conducted using Student’s t-tests (for two groups) or one-way ANOVA (for multiple groups). A p-value \u0026lt; 0.05 was considered statistically significant. All experiments were performed with at least two biological replicates to ensure reproducibility.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eConflict of interest\u003c/h2\u003e \u003cp\u003eThe authors declare no conflicts of interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAuthor contribution\u003c/h2\u003e \u003cp\u003eZ.Zhao and F.C. performed most of the molecular, cellular and animal experiments. Y.M. performed most of Bioinformatics Analysis. H.Z., M.Z. and X.Z. did animal sample preparation. Y.H. and W.W. performed plasmid construction. Z.C. performed ChIP-Seq and RNA-Seq experiments. M.D and Z.L did bacterial-related experiments. L.A., S.J. and Z.Zhou designed the experiments, analyzed the data, and wrote the manuscript. S.J., and Z.Z. supervised the project.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eThis work was supported by the National Key R\u0026amp;D Program of China (2020YFA0803200), the National Natural Science Foundation of China Grants (82222052, 32070710, 31930026, 81972876, 81725014, 92168116 and 82150112), Natural Science Foundation of Shanghai (23ZR1480400), Shanghai Rising-Star Program (22QA1407300), Shanghai Sailing Program (No. 23YF1432900).\u003c/p\u003e\n\u003ch3\u003eData availability\u003c/h3\u003e\n\u003cp\u003eRNA sequencing data reported in this paper have been deposited in the GSA-human database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://ngdc.cncb.ac.cn/gsa-human/\u003c/span\u003e\u003cspan address=\"https://ngdc.cncb.ac.cn/gsa-human/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The accession number is subHRA014955. scRNA-seq and spatial transcriptomics for direct tumoral interactome have been deposited in National Genomics Data Center with accession number OMIX008890 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://ngdc.cncb.ac.cn/omix/preview/z4vQ2aXQ\u003c/span\u003e\u003cspan address=\"https://ngdc.cncb.ac.cn/omix/preview/z4vQ2aXQ\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The CTRB1-interactome MS data was deposited in the iProX database under accession number IPX0010977000 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.iprox.cn/\u003c/span\u003e\u003cspan address=\"https://www.iprox.cn/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAyukawa, S., Kamoshita, N., Nakayama, J., Teramoto, R., Pishesha, N., Ohba, K., Sato, N., Kozawa, K., Abe, H., Semba, K., \u003cem\u003eet al.\u003c/em\u003e (2021). Epithelial cells remove precancerous cells by cell competition via MHC class I\u0026ndash;LILRB3 interaction. Nature Immunology \u003cem\u003e22\u003c/em\u003e, 1391\u0026ndash;1402.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBechtel, T.J., Reyes-Robles, T., Fadeyi, O.O., and Oslund, R.C. (2021). Strategies for monitoring cell\u0026ndash;cell interactions. Nature Chemical Biology \u003cem\u003e17\u003c/em\u003e, 641\u0026ndash;652.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoareto, M. (2020). Patterning via local cell-cell interactions in developing systems. Developmental Biology \u003cem\u003e460\u003c/em\u003e, 77\u0026ndash;85.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCho, K.F., Branon, T.C., Rajeev, S., Svinkina, T., Udeshi, N.D., Thoudam, T.A.-O., Kwak, C., Rhee, H.A.-O., Lee, I.K., Carr, S.A., \u003cem\u003eet al.\u003c/em\u003e (2020). Split-TurboID enables contact-dependent proximity labeling in cells. PNAS \u003cem\u003e117\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChudnovskiy, A.A.-O., Castro, T.B.R., Nakandakari-Higa, S., Cui, A.A.-O., Lin, C.H., Sade-Feldman, M.A.-O., Phillips, B.A.-O., Pae, J., Mesin, L.A.-O., Bortolatto, J.A.-O., \u003cem\u003eet al.\u003c/em\u003e (2024). Proximity-dependent labeling identifies dendritic cells that drive the tumor-specific CD4(+) T cell response. Science immunology \u003cem\u003e9(100)\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDohlman, A.B., Klug, J., Mesko, M., Gao, I.H., Lipkin, S.M., Shen, X., and Iliev, I.D. (2022). A pan-cancer mycobiome analysis reveals fungal involvement in gastrointestinal and lung tumors. Cell \u003cem\u003e185\u003c/em\u003e, 3807\u0026ndash;3822.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEizirik, D., Sammeth, M., Bouckenooghe, T., Bottu, G., Sisino, G., Igoillo-Esteve, M., Ortis, F., Santin, I., Colli, M., Barthson, J., \u003cem\u003eet al.\u003c/em\u003e (2012). The Human Pancreatic Islet Transcriptome: Expression of Candidate Genes for Type 1 Diabetes and the Impact of Pro-Inflammatory Cytokines. PLoS genetics \u003cem\u003e8\u003c/em\u003e, e1002552.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFlanagan, D.J., Pentinmikko, N., Luopaj\u0026auml;rvi, K., Willis, N.J., Gilroy, K., Raven, A.P., McGarry, L., Englund, J.I., Webb, A.T., Scharaw, S., \u003cem\u003eet al.\u003c/em\u003e (2021a). NOTUM from Apc-mutant cells biases clonal competition to initiate cancer. Nature \u003cem\u003e594\u003c/em\u003e, 430\u0026ndash;435.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFlanagan, D.J., Pentinmikko, N.A.-O., Luopaj\u0026auml;rvi, K., Willis, N.A.-O., Gilroy, K., Raven, A.P., McGarry, L.A.-O., Englund, J.A.-O., Webb, A.T., Scharaw, S.A.-O., \u003cem\u003eet al.\u003c/em\u003e (2021b). NOTUM from Apc-mutant cells biases clonal competition to initiate cancer. Nature \u003cem\u003e594(7863)\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFu, K., Cheung, A.H.K., Wong, C.C., Liu, W., Zhou, Y., Wang, F., Huang, P., Yuan, K., Coker, O.O., Pan, Y., \u003cem\u003eet al.\u003c/em\u003e (2024). Streptococcus anginosus promotes gastric inflammation, atrophy, and tumorigenesis in mice. Cell \u003cem\u003e187\u003c/em\u003e, 882\u0026ndash;896.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGe, Y., Chen, L., Liu, S., Zhao, J., Zhang, H., and Chen, P.R. (2019). Enzyme-Mediated Intercellular Proximity Labeling for Detecting Cell\u0026ndash;Cell Interactions. Journal of the American Chemical Society \u003cem\u003e141\u003c/em\u003e, 1833\u0026ndash;1837.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan, H., Jain, A.D., Truica, M.I., Izquierdo-Ferrer, J., Anker, J.F., Lysy, B., Sagar, V., Luan, Y., Chalmers, Z.R., Unno, K., \u003cem\u003eet al.\u003c/em\u003e (2019a). Small-Molecule MYC Inhibitors Suppress Tumor Growth and Enhance Immunotherapy. Cancer cell \u003cem\u003e36\u003c/em\u003e, 483\u0026ndash;497.e415.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan, X., Wang, Y., Pu, W., Huang, X., Qiu, L., Li, Y., Yu, W., Zhao, H., Liu, X., He, L., \u003cem\u003eet al.\u003c/em\u003e (2019b). Lineage Tracing Reveals the Bipotency of SOX9(+) Hepatocytes during Liver Regeneration. Stem Cell Reports \u003cem\u003e12\u003c/em\u003e, 624\u0026ndash;638.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLima, A., and Rodriguez, T.A. (2021). MHC-I presents: tumor surveillance in the epithelia by cell competition. Nature Immunology \u003cem\u003e22\u003c/em\u003e, 1358\u0026ndash;1360.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu-Chittenden, Y., Huang, B., Shim, J.S., Chen, Q., Lee, S.J., Anders, R.A., Liu, J.O., and Pan, D. (2012). Genetic and pharmacological disruption of the TEAD-YAP complex suppresses the oncogenic activity of YAP. Genes \u0026amp; development \u003cem\u003e26\u003c/em\u003e, 1300\u0026ndash;1305.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, H., Luo, H., Xue, Q., Qin, S., Qiu, S., Liu, S., Lin, J., Li, J.P., and Chen, P.R. (2022). Antigen-Specific T Cell Detection via Photocatalytic Proximity Cell Labeling (PhoXCELL). Journal of the American Chemical Society \u003cem\u003e144\u003c/em\u003e, 5517\u0026ndash;5526.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, Q., Tang, Y.A.-O., Huang, Y.A.-O., Wang, J., Yang, K., Zhang, Y., Pu, W., Liu, J., Shi, X., Ma, Y.A.-O., \u003cem\u003eet al.\u003c/em\u003e (187). Insights into male androgenetic alopecia using comparative transcriptome profiling: hypoxia-inducible factor-1 and Wnt/β-catenin signalling pathways. British Journal of Dermatology 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, Z., Li, J.P., Chen, M., Wu, M., Shi, Y., Li, W., Teijaro, J.R., and Wu, P. (2020). Detecting Tumor Antigen-Specific T Cells via Interaction-Dependent Fucosyl-Biotinylation. Cell 183, 1117\u0026ndash;1133.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoya, I.A.-O., Castaldo, S.A.-O., Van den Mooter, L.A.-O., Soheily, S.A.-O., Sansores-Garcia, L., Jacobs, J.A.-O., Mannaerts, I.A.-O., Xie, J.A.-O., Verboven, E.A.-O., Hillen, H., \u003cem\u003eet al.\u003c/em\u003e (2019). Peritumoral activation of the Hippo pathway effectors YAP and TAZ suppresses liver cancer in mice. Science \u003cem\u003e366\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNakai, K., Lin, H., Yamano, S., Tanaka, S., Kitamoto, S., Saitoh, H., Sakuma, K., Kurauchi, J., Akter, E., Konno, M., \u003cem\u003eet al.\u003c/em\u003e (2023). Wnt activation disturbs cell competition and causes diffuse invasion of transformed cells through NF-κB-MMP21 pathway. Nature communications \u003cem\u003e14\u003c/em\u003e, 7048.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNakandakari-Higa, S., Walker, S., Canesso, M.C.C., van der Heide, V., Chudnovskiy, A., Kim, D.-Y., Jacobsen, J.T., Parsa, R., Bilanovic, J., Parigi, S.M., \u003cem\u003eet al.\u003c/em\u003e (2024). Universal recording of immune cell interactions in vivo. Nature \u003cem\u003e627\u003c/em\u003e, 399\u0026ndash;406.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eParker, T.M., Gupta, K., Palma, A.M., Yekelchyk, M.A.-O., Fisher, P.B., Grossman, S.R., Won, K.A.-O., Madan, E., Moreno, E.A.-O.X., and Gogna, R.A.-O. (2021). Cell competition in intratumoral and tumor microenvironment interactions. The EMBO Journal \u003cem\u003e40(17)\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePasqual, G., Chudnovskiy, A., Tas, J.M.J., Agudelo, M., Schweitzer, L.D., Cui, A., Hacohen, N., and Victora, G.D. (2018). Monitoring T cell\u0026ndash;dendritic cell interactions in vivo by intercellular enzymatic labelling. Nature \u003cem\u003e553\u003c/em\u003e, 496\u0026ndash;500.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePentinmikko, N., Iqbal, S., Mana, M., Andersson, S., Cognetta, A.B., 3rd, Suciu, R.M., Roper, J., Luopaj\u0026auml;rvi, K., Markelin, E., Gopalakrishnan, S., \u003cem\u003eet al.\u003c/em\u003e (2019). Notum produced by Paneth cells attenuates regeneration of aged intestinal epithelium. Nature \u003cem\u003e571\u003c/em\u003e, 398\u0026ndash;402.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQiu, S., Zhao, Z., Wu, M., Xue, Q., Yang, Y., Ouyang, S., Li, W., Zhong, L., Wang, W., Yang, R., \u003cem\u003eet al.\u003c/em\u003e (2022). Use of intercellular proximity labeling to quantify and decipher cell-cell interactions directed by diversified molecular pairs. Science advances \u003cem\u003e8\u003c/em\u003e, eadd2337.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRauskolb, C., Sun, S., Sun, G., Pan, Y., and Irvine, K.D. (2014). Cytoskeletal tension inhibits Hippo signaling through an Ajuba-Warts complex. Cell \u003cem\u003e158\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRon, D., and Walter, P. (2007). Signal integration in the endoplasmic reticulum unfolded protein response. Nature reviews Molecular cell biology \u003cem\u003e8\u003c/em\u003e, 519\u0026ndash;529.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoux, K.J., Kim Di Fau - Raida, M., Raida M Fau - Burke, B., and Burke, B. (2012). A promiscuous biotin ligase fusion protein identifies proximal and interacting proteins in mammalian cells. Journal of Cell Biology \u003cem\u003e6\u003c/em\u003e, 801\u0026ndash;810.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchwarz, D.S., and Blower, M.D. (2016). The endoplasmic reticulum: structure, function and response to cellular signaling. Cellular and molecular life sciences: CMLS 73, 79\u0026ndash;94.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSu, J., Song, Y., Zhu, Z., Huang, X., Fan, J., Qiao, J., and Mao, F. (2024). Cell\u0026ndash;cell communication: new insights and clinical implications. Signal Transduction and Targeted Therapy \u003cem\u003e9\u003c/em\u003e, 196.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSubramaniam, D.A.-O.X., Ponnurangam, S., Ramalingam, S., Kwatra, D., Dandawate, P., Weir, S.A.-O.X., Umar, S., Jensen, R.A.-O., and Anant, S.A.-O. (2021). Honokiol Affects Stem Cell Viability by Suppressing Oncogenic YAP1 Function to Inhibit Colon Tumorigenesis. LID \u0026ndash;\u0026thinsp;10.3390/cells10071607 [doi] LID \u0026ndash;\u0026thinsp;1607. Cells \u003cem\u003e10\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTakano, T., Wallace, J.T., Baldwin, K.T., Purkey, A.M., Uezu, A., Courtland, J.L., Soderblom, E.J., Shimogori, T., Maness, P.F., Eroglu, C., \u003cem\u003eet al.\u003c/em\u003e (2020). Chemico-genetic discovery of astrocytic control of inhibition in vivo. Nature \u003cem\u003e588\u003c/em\u003e, 296\u0026ndash;302.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang, Y., Fang, G., Guo, F., Zhang, H., Chen, X., An, L., Chen, M., Zhou, L., Wang, W., Ye, T., \u003cem\u003eet al.\u003c/em\u003e (2020). Selective Inhibition of STRN3-Containing PP2A Phosphatase Restores Hippo Tumor-Suppressor Activity in Gastric Cancer. Cancer Cell \u003cem\u003e38\u003c/em\u003e, 115\u0026ndash;128.e119.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTanimura, N., and Fujita, Y. (2020). Epithelial defense against cancer (EDAC). Seminars in cancer biology \u003cem\u003e63\u003c/em\u003e, 44\u0026ndash;48.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Neerven, S.M., de Groot, N.E., Nijman, L.E., Scicluna, B.P., van Driel, M.S., Lecca, M.C., Warmerdam, D.O., Kakkar, V., Moreno, L.F., Vieira Braga, F.A., \u003cem\u003eet al.\u003c/em\u003e (2021). Apc-mutant cells act as supercompetitors in intestinal tumour initiation. Nature \u003cem\u003e594\u003c/em\u003e, 436\u0026ndash;441.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Neerven, S.M., and Vermeulen, L. (2023). Cell competition in development, homeostasis and cancer. Nature Reviews Molecular Cell Biology \u003cem\u003e24\u003c/em\u003e, 221\u0026ndash;236.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYuan, L., Pan, L., Wang, Y., Zhao, J., Fang, L., Zhou, Y., Xia, R., Ma, Y., Jiang, Z., Xu, Z., \u003cem\u003eet al.\u003c/em\u003e (2024). Characterization of the landscape of the intratumoral microbiota reveals that Streptococcus anginosus increases the risk of gastric cancer initiation and progression. Cell Discovery \u003cem\u003e10\u003c/em\u003e, 117.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYum, M.K., Han, S., Fink, J., Wu, S.-H.S., Dabrowska, C., Trendafilova, T., Mustata, R., Chatzeli, L., Azzarelli, R., Pshenichnaya, I., \u003cem\u003eet al.\u003c/em\u003e (2021). Tracing oncogene-driven remodelling of the intestinal stem cell niche. Nature \u003cem\u003e594\u003c/em\u003e, 442\u0026ndash;447.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, C.B., Pan, S.Y., Jin, P., Deng, J.W., Xue, J.H., Ma, X.Y., Xie, Y.H., Cao, H., Liu, Q., Xie, W.F., \u003cem\u003eet al.\u003c/em\u003e (2022). Fecal Signatures of Streptococcus anginosus and Streptococcus constellatus for Noninvasive Screening and Early Warning of Gastric Cancer. Gastroenterology \u003cem\u003e162\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Non-immune surveillance, Cell competition, PCL, EMDAC, CTRB1, Microbial infection","lastPublishedDoi":"10.21203/rs.3.rs-6524135/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6524135/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWhile non-immune epithelial surveillance constitutes a crucial frontline defense against early tumorigenesis, its operational mechanisms - particularly the cellular components and structural basis of cell competition - remain poorly defined. In this study, we uncover a novel cell competition architecture which we termed as proliferative containment layer (PCL), comprising coordinated epithelial-mesenchymal cell assemblies. Employing our newly developed Surface-anchored TurboID-mediated Contact-dependent Labeling (STCL) system, we achieved in situ biotinylation of membrane proteins on interacting cells, enabling the first dynamic mapping of direct tumor-stromal interactomes in a tumor-suppressive competition model. Mechanistically, direct tumor cell contact triggers a unique epithelial-mesenchymal defense against cancer (EMDAC) program, where PCL components acquire superior proliferative capacity through chymotrypsinogen B1 (CTRB1)-mediated coordination of Myc/YAP signaling axes. Notably, we demonstrate that CTRB1 serves as a molecular rheostat integrating multiple proliferation pathways to establish competitive dominance, while microbial infection unexpectedly suppresses CTRB1 expression and compromises tumor clearance. This work fundamentally advances our understanding of non-immune surveillance by: 1) Identifying PCL as a spatially organized defense unit; 2) Deciphering EMDAC as a dual-lineage competition mechanism; 3) Establishing CTRB1 as a microbial-sensitive master regulator of cellular fitness during competition.\u003c/p\u003e","manuscriptTitle":"Spatiotemporal Decoding of Tumor Contactome Unveils a Microbial-Sensitive Ctrb1-Dependent Cell Competition Architecture Driving Epithelial-Mesenchymal Defense Against Cancer","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-07 05:42:01","doi":"10.21203/rs.3.rs-6524135/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"401de67a-6b4f-457b-bdf1-34c083cf6644","owner":[],"postedDate":"May 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":48099666,"name":"Biological sciences/Cancer/Tumour-suppressor proteins"},{"id":48099667,"name":"Biological sciences/Cell biology/Cell growth/HIPPO signalling"},{"id":48099668,"name":"Biological sciences/Cancer/Cancer microenvironment"},{"id":48099669,"name":"Biological sciences/Biotechnology/Sequencing/RNA sequencing"},{"id":48099670,"name":"Biological sciences/Microbiology/Microbial communities/Microbiome"}],"tags":[],"updatedAt":"2025-06-04T15:26:19+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-07 05:42:01","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6524135","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6524135","identity":"rs-6524135","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}