Abstract

   Cancer metastasis is the primary cause of cancer-related death, yet the
   forces that drive cancer cells through various steps and different
   routes to distinct target organs/tissues remain elusive. In this study,
   we applied a barcoding system based single-cell lineage tracing
   approach to study the metastasis rate and route of breast cancer cells
   and their interactions with the tumor microenvironment (TME) during
   metastasis. The results indicate that only a small fraction of cells,
   accounting for fewer than 3% of total barcodes, can intravasate from
   the primary site into the blood circulation, whereas more cells
   disseminate through the lymphatic system to different organs. Tumor
   cells derived from the same progenitor cell exhibit different gene
   expression patterns in different soils, and the cancer cell-TME
   communication paradigm varies significantly between primary and
   metastatic tumors. Furthermore, metastable cells require a prewired
   particular cytokine expression ability which may be specific for lymph
   metastasis route although the underlying mechanism requires further
   investigation. In summary, leveraging a single-cell lineage tracing
   system, we demonstrate that the crosstalk between tumor cells and the
   TME is the driving force controlling the preferential metastatic fate
   of cancer cells through the lymphatic system.

Graphical Abstract

   [62]graphic file with name 12943_2025_2279_Figa_HTML.jpg

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12943-025-02279-w.

   Keywords: Barcode system, Lymphatic metastasis, Lineage tracing, Tumor
   microenvironment, Soil-seed interaction

Introduction

   Cancer metastasis is a process in which tumor cells from the primary
   tumor invade through the basement membrane into blood or lymphatic
   vessels (intravasation), travel through the blood or lymphatic system,
   and colonize distant organs/tissues for further development. Metastasis
   is responsible for up to 70–90% of cancer-related mortality [[63]1,
   [64]2]. Shedding of tumor cells into the vasculature and lymphatic
   vessels occurs simultaneously in the vast majority of cancers [[65]1,
   [66]3, [67]4]. Clonal origin analysis has demonstrated distinct lineage
   relationships between the lymphatic system and blood circulation
   [[68]5, [69]6] indicating the complexity of tumor cell dissemination
   [[70]7].

   The metastasis of different tumor types demonstrates a biased
   distribution among distant organs. This phenomenon is called
   “organotropism” or “organ-specific metastasis” and is believed to be
   caused primarily by coordination between metastatic cancer cells
   (“seeds”) and their organ/tissue microenvironment (“soil”), i.e., the
   “seed and soil” hypothesis [[71]8, [72]9]. Tissue-specific gene
   signatures and signalling pathways in metastatic breast cancers have
   been identified by bulk RNA sequencing (RNA-seq) by comparing the
   primary tumor and distant lesions in different organs [[73]10].
   However, whether these distinct signatures represent different cell
   populations within the primary tumor or the same type of cell educated
   to exhibit different gene expression profiles remains unknown.

   In this study, we used a barcoding system to track individual tumor
   cells to show cell fate during metastasis, including their migration
   ability and route. Combining this lineage tracing analysis with
   single-cell RNA sequencing (scRNA-seq), we demonstrated that the
   lymphatic system is the main route of breast cancer dissemination and
   metastasis with the aid of prewired particular cytokine expression,
   which serves as an intrinsic driver for tumor cell crosstalk with the
   TME. Our data also provide evidence that cancer cells derived from the
   same progenitor cell show different gene expression patterns and a
   different tumor cell-TME communication paradigm in different soils.

Results

Intravasation is the bottleneck for tumor metastasis through the blood
circulation

   We applied the CellTag barcoding system [[74]11] to the mammary
   carcinoma cell lines 4T1 and EMT6 to trace tumor cell metastasis in
   mouse cancer models. Because both cell lines can spontaneously produce
   highly metastatic tumors in immunocompetent mice, they are ideal models
   to show the tumor cell dissemination route and fate for studying the
   intrinsic drivers and their communication patterns with stromal cells
   in the tumor immune microenvironment during metastasis.

   First, we transfected cells with a combinatory cell indexing barcode
   tag fused to a GFP marker and then sorted the cells by
   fluorescence-activated cell sorting (FACS) to select labelled tumor
   cells. To ensure that each cell contained a unique barcode, we used a
   low multiplicity of infection (MOI) of 0.01 for infection and sorted
   10,000 GFP-positive cells, a much lower number than the number of
   different barcodes (19,973). Then, the GFP-positive cells were expanded
   in culture for 7 days prior to DNA sequencing (DNA-seq), which detected
   7,665 and 7,809 unique barcodes in 4T1 and EMT6 cells, respectively. We
   then inoculated cells into the 4th mammary fat pad of BALB/c mice to
   model tumor metastasis (Fig. [75]1A; n = 10 mice per cell line). Mice
   were monitored twice a week, and blood samples were collected at 10-day
   intervals. A complete autopsy was performed to assess metastatic tumors
   when the mice were moribund (30 days for 4T1-derived tumors and 47 days
   for EMT6-derived tumors) (Figure S1A).

Fig. 1.

   [76]Fig. 1
   [77]Open in a new tab

   Lineage tracing of tumor cell metastatic rate and route. A Schematic
   illustration of the strategy for tracing tumor cell migration fate
   using single-cell barcoding integrated with scRNA-seq (n = 10). B
   CellTag compositions analysis between the primary tumor, blood CTCs
   collected from different days and liver/lung metastasis tumors (Liver
   MT, Lung MT), which were collected at the end timepoint. C CellTag
   barcode types identified in lung and liver metastatic tumors were
   analysed in the tail vein injection model and compared with those in
   the mammary fat pad inoculation model. D Summary of CellTag barcode
   types were compared between different organs/tissues and blood
   disseminated tumor cells in the 4T1 metastasis model (n = 10 mice). E
   Workflow of the sample collection strategy to demonstrate the tumor
   cell lymphatic and hematogenous migration routes. F The CellTag barcode
   types were compared in different organs/tissues in the 4T1 metastasis
   model on day 30 (n = 7 mice). G Heatmap presentation of the CellTag
   composition in the blood, lymph nodes, and distant organs in the 4T1
   cell metastasis model. CellTags were defined based on their presence in
   at least 5% of tumors or more than 3 samples

   All recipient mice exhibited severe metastasis in distant organs,
   especially the lung and liver (Figure S1B). DNA-seq detected 4,741
   unique barcodes for 4T1 cells and 5,227 unique barcodes for EMT6 cells,
   on average, from each primary tumor (Table S1, Table S2). Notably, the
   diversity of the barcodes identified in the blood and distant organs
   decreased dramatically to approximately 1.04–2.46% of that in the
   primary tumour (Fig. [78]1B). A similar pattern was observed in EMT6
   tumor models (Figure S1C). Based on the barcode system, we were able to
   trace metastasis fate of tumor cells in each of all animals. Using a
   tumor derived from 4T1 cells as an example (mouse #540), which contains
   4,456 different barcodes (i.e. different types of individual cells), we
   found very small fraction of cells (68, 38 and 39 barcodes in the blood
   collected at D10, D20 and D30) could escape from the primary tumor into
   the blood (Figure S1D). Further analysis detected a total of 86
   distinct barcodes with 26 common and varying number of barcodes
   specific for each time point in this tumor (Figure S1E). Next, we
   traced the fate of these circulating tumor cells (CTCs) by studying
   their presence in the metastasis tumors, and found that 33 out of 68
   (48.5%) CTCs at day 10; 24 out of 38 (63%) CTCs at day 20; and 25 out
   of 38 (66%) CTCs at day 30 could eventually colonize in the lung and/or
   liver (Figure S1F). Analysis of all 10 tumors derived from both 4T1 and
   EMT6 models revealed a similar pattern, i.e. only small fractions, on
   average < 3%, of all barcodes were detected in the blood at these time
   points (Fig. [79]1B, Figure S1G, H). These data suggest that the step
   of cancer cell intravasation from the primary tumor is a major
   bottleneck for tumor metastasis. To verify this hypothesis, we directly
   injected the same amount of tumor cells into a blood vessel through the
   tail vein. By DNA-seq, we found that the average number of barcodes in
   lung metastatic tumors was 1,839, approximately 15 times higher than
   that (121 CellTags) detected in the lungs of mice implanted with the
   same number of cancer cells in the mammary fat pad (Fig. [80]1C, Figure
   S1I, J, Table S3).

The lymphatic system is the main route of tumor dissemination

   Next, we compared the barcode types in the lung and liver with those in
   the blood, and the data revealed there are a total of 1290 different
   barcodes in the lung and liver, however, as much as 83% (1077/1290) of
   them were not detected in the blood (Fig. [81]1D, Figure S2A). The
   analysis of tumors derived from EMT6 also revealed a similar pattern,
   i.e., about 71% (155/221) of barcodes detected in the lung and liver
   metastatic tumors were not found in the blood (Figure S2B, C). The
   detailed analysis and comparison of barcodes regarding their appearance
   and abundance in the blood at all three time points from all the mice,
   in the lung and liver metastatic tumors revealed similar patterns
   (Figure S2D, Table S4). Altogether, these data suggest that while CTCs
   were definitely appeared in the metastatic tumors, large fraction of
   metastatic cells might disseminate through other routes.

   To identify the tumor cell migration route, we specifically checked the
   CellTag information of the lymphatic-disseminated tumor cells in the
   inguinal, axillary, and brachial lymph nodes (Fig. [82]1E, Table S5).
   To further show the cell migration route, we first defined lung and
   liver metastatic cells based on the barcode information in the lung and
   liver on day 30. Then, the barcode composition was compared with that
   of lymph node-disseminated tumor cells collected at different time
   points. As the Venn diagrams show, all the tumor cells in the blood
   circulation were found in the lymphatic system, and more metastatic
   tumor cells were detected in lymph nodes than in the blood
   (Fig. [83]1F). This result may indicate that tumor cells have a
   stronger ability to intravasate into the lymphatic system than into the
   blood circulation [[84]7, [85]12–[86]14], while during lymphatic
   circulation, the majority of tumor cells gradually die of ferroptosis
   or other mechanisms during the journey before colonizing remote
   tissues/organs to establish metastatic tumors [[87]15, [88]16]. Further
   global cell fate comparison defined based on 7 replicates also
   indicated that lymphatic-disseminated tumor cells exhibited the highest
   diversity, covering almost all blood circulating tumor cells (CTCs),
   and formed metastatic tumors in distinct organs (Fig. [89]1G). These
   data indicate that the lymphatic system is the route of tumor
   metastasis, which has been widely reported in clinical [[90]17].

   Thus, by using the CellTag system, we successfully discovered the tumor
   cell metastatic route and consequently defined different types of
   metastatic tumor cells. Therefore, based on this lineage information,
   we can determine the fates of cells in primary tumors, i.e., which
   cells could migrate into the blood and/or lymphatic system, which cells
   could colonize the lung or liver via the blood circulation and which
   cells could disseminate into the lung or liver through the lymphatic
   system.

   These results indicate that tumor cells may already be wired for a
   particular migration route before disseminating from the primary site
   [[91]15, [92]17, [93]18], a finding that demonstrated heritability and
   reproducibility among the mouse models.

Communication interactions between tumor cells and stromal cells correlate
with their metastasis route

   To identify the intrinsic features that govern the metastatic fate of
   tumor cells, we conducted scRNA-seq on the primary tumors, and 6,066
   cells passed quality control. Dimensionality reduction via uniform
   manifold approximation and projection (UMAP) clustering yielded 10
   distinct clusters, including tumor cells, fibroblasts, macrophages, T
   cells, and other types of stromal cells, based on distinct marker gene
   expression patterns (Figure S3, S4A). Then, we focused specifically on
   tumor cells for dimensionality reduction to reveal the differences
   between different types of metastatic cells and nonmetastatic cells
   (Fig. [94]2A).

Fig. 2.

   [95]Fig. 2
   [96]Open in a new tab

   scRNA-seq identifies the intrinsic features correlated with distinct
   metastatic fates. A Schematic illustration of the strategy for
   identifying the distinct features between metastatic cells and
   nonmetastatic cells using single-cell barcoding integrated with
   scRNA-seq. B Ligand–receptor pairs in tumor-–stromal interactions in
   primary tumors; only the differential pairs among a variety of tumor
   cells were used in the plot. The colour and size of the circles
   indicate the significance and the enrichment score, respectively. C
   Schematic showing the traditional bulk strategy for gene expression
   comparison analysis among primary tumor and metastasis tumors. All
   cells in the different organs were used for the following analysis. D
   Schematic showing the strategy for comparative analysis of gene
   expression. Cells with the same tags in both primary tumors and
   metastatic tumors were used for the following analysis. E Venn diagram
   showing the CellTags detected by the scRNA-seq of primary, lung
   metastatic, and liver metastatic tumors. F Heatmap showing that the
   same cells labelled with identical tags were educated for differential
   regulation of their gene expression pattern in distinct organs during
   metastasis. G Differentially expressed genes identified by the
   traditional bulk RNA analysis approach and the identical tag approach
   were compared. H GSEA of the differentially expressed genes in
   identical tagged cells during metastasis from the primary tumor to the
   lung and from the primary tumor to the liver, as well as distinct
   metastasis organs

   To investigate the driving force responsible for the metastatic fate of
   different types of tumor cells, we conducted cell–cell interaction
   analysis by using the CellCall program [[97]19]. The results indicated
   that tumor cells communicated closely with a variety of stromal cells,
   especially macrophages, fibroblasts, endothelial cells, and dendritic
   cells (Figure S4B). Different types of tumor cells interacted
   differently with immune cells (Figure S4C, D). Analysis of signal
   transduction from tumor cells to stromal cells showed that different
   types of cells secreted different cytokines or chemokines into the
   microenvironment, which might help to direct their migration route.
   Specifically, most of the tumor cells, except the Blood-trap tumor
   cells, demonstrated a high expression level of Tgfβ2 to interact with
   endothelial cells and macrophages. In contrast, the Blood-trap tumor
   cells communicated strongly with macrophages and monocytes. The
   LyN-trap and Other-met tumor cells showed very similar interaction
   patterns, suggesting that Other-met tumor cells also disseminated
   through the lymphatic system but could not be detected, possibly
   because they were moderately difficult to detect (Fig. [98]2B).

   Through comparative analysis of the cell–cell interactions between the
   abovementioned groups as well as the nonmetastasized cells, we
   identified several specific features that correlated with lymphatic and
   haematogenous dissemination and with lung or liver metastasis. Notably,
   the observation that the Harmony cells lacked Il2-Il2rg, Il15-Il2rg,
   Ccl9-Ccr1, Flt3l-Flt3–mediated communication with stromal cells
   indicating the “prewired” cytokine signalling played a key role in
   tumor metastasis (Fig. [99]2B, S4E).

Tumor cells demonstrate distinct properties in congenial soil

   It is a well know concept that metastases develop only when the seed
   and soil are compatible. Samples from the same organ/tissue
   demonstrated comparable CellTag diversity even in different mice
   (Figure S5A, B), indicating that tumor cells exhibited greater
   similarity in more similar sample types and that highly conserved
   features existed in the tumor cells to modulate the migration fate,
   which could be identified by the dissimilar CellTags.

   To further demonstrate the features of cancer cells during metastasis,
   we also conducted scRNA-seq on primary, lung and liver metastatic
   tumors to provide greater insights into the features of organ-specific
   tumor metastasis. A total of 23,843 cells in the 4T1 model passed
   quality control, of which 6,066, 8,582, and 9,195 cells were derived
   from the primary tumors, lung metastatic tumors, and liver metastatic
   tumors, respectively. Dimensionality reduction via UMAP clustering
   analysis yielded 10 distinct clusters, including tumor cells,
   fibroblasts, macrophages, T cells, and other types of stromal cells,
   based on distinct specific marker gene expression patterns (Figure S5C,
   D).

   To assess the dynamic changes in tumor cells during metastasis, we
   conducted gene set enrichment analysis (GSEA) specifically on cancer
   cells to show the differences among primary tumors and metastatic
   tumors. First, we included all tumor cells, similar to the traditional
   bulk approach, for comparative analysis (Fig. [100]2C). After tumor
   cells migrated into the lung, the cytokine response, including the TGFβ
   signalling pathway, was upregulated, whereas the Toll-like
   receptor-related pathway and metabolism-regulated growth-related
   pathway were downregulated (Figure S5E). Tumor cells demonstrated
   similar gene expression patterns, except for the upregulation of Myc
   related pathways, after they colonized the liver (Figure S5F).

   As demonstrated in our previous results, only a small portion of tumor
   cells could eventually colonize distant organs. To demonstrate the
   faithful properties of seeds in different soils, we focused
   specifically on cells that carried the same tag in primary tumors, lung
   metastatic tumors, and liver metastatic tumors (Fig. [101]2D). Based on
   the scRNA-seq data, eight types of cells fit the criteria and were
   therefore used for further analysis (Fig. [102]2E). Next, we conducted
   transcriptomic analysis by comparing the expression patterns of cells
   with the same barcodes in the primary tumor, the lung and liver
   metastatic tumors. Of 8 different cell tags analysed, the data revealed
   a site-specific expression pattern, i.e. the same types of cells
   elicited distinct expression patterns when they were at different
   sites. The expression in the lung and liver metastatic tumors was quite
   distinct (Fig. [103]2F). On the other hand, the difference was less
   obvious when these 8 different types of cells were compared in the same
   sites (lung metastasis, liver metastasis and primary tumors). These
   data suggest that soil play a significant role in regulating expression
   of cancer cells.

   Compared with the traditional bulk approach, the CellTag-based analysis
   identified different marker genes during tumor cell migration
   (Fig. [104]2G, Figure S6A, B). As predicted, we identified a group of
   genes that were specifically differentially regulated after cell
   colonization of the lung and/or liver (Figure S6C). Specifically, Tufm
   and Atp1b1 were specifically upregulated after cells were colonized in
   the live, while some genes were down-regulated to adapt liver
   environment, such as Acot1, Orc5, Evc2 et al. There was a group of
   genes, including Sftpc, Bak1, Cnpy4, Gins1 et al., were activated to
   promote the cell growth in lung (Fig. [105]2F). GSEA and KEGG pathway
   enrichment analysis indicated that in addition to the terms identified
   by the bulk sequencing approach, the CellTag-based analysis approach
   also allowed us to identify more specific functional terms related to
   tumor metastasis, i.e., reactive oxygen species detoxification,
   hypoxia, DNA methylation, and mitochondria terms (Fig. [106]2H, Figure
   S6D). These data indicate that identification of the cell migration
   fate provides detailed information on tumor metastasis.

The seed-soil communication paradigm differs among primary tumors and
metastatic tumors

   To further investigate cell–cell interactions and communication in the
   TME, we used CellCall to estimate the putative intercellular networks
   based on not only ligand and receptor expression but also receptor
   downstream transcription factors (TFs). These analyses revealed
   comprehensive crosstalk among tumor cells and stromal cells (Figure
   S7A-B). Specifically, tumor cells at the primary site demonstrated much
   stronger interactions with endothelial cells and fibroblasts, for
   example, Vegfc-Kdr/Flt4/Flt1 and Pdgfc-Pdgfrb/Pdgfra/Kdr/Flt4/Flt1
   interactions (Figure S7C). Functional analysis indicated that some of
   these genes are mainly growth factors that consistently exhibit high
   expression levels in primary tumors. In addition, groups of
   ligand–receptor-transcription factor interaction pairs were commonly
   upregulated in all tumor metastasis steps, including Tslp-Il17r/Crlf2,
   Il11-Il16st/Il16ra, Tgfβ1/3-Tfgbe1/2, Csf3-Csf3r, and Csf1-Csf1r. These
   interactions were functionally enriched in cytokine-, chemokine- and
   interferon-gamma response-related genes, as well as the Tgfβ signalling
   pathway.

   In addition, tumor cells exhibited distinct responses to stromal
   signals in different soils through differentially expressed receptors,
   which were also reflected by the levels of their downstream
   transcription factors (Figure S7D, E). For example, tumor cells highly
   expressed Bmpr2 after migrating into the lungs, with subsequent
   activation of its downstream targets, including Id1, Id2, Id4, and
   Smad4.

   These results are consistent with those of GSEA, indicating that tumor
   cells have different gene expression patterns in different soils and
   that the cancer cell-TME communication paradigm varies significantly
   between primary tumors and metastatic tumors.

Discussion

   Tumor cells metastasize mainly through the blood circulation and the
   lymphatic system. Numerous studies have been carried out to distinguish
   the role of both systems in the various stages of cancer metastasis
   [[107]1, [108]5, [109]10, [110]20], yet which one is the “preferred”
   route for distant metastasis remains unknown. Phylogenetic analysis of
   colorectal and breast cancers has shown that in the majority of cases,
   lymphatic and distant metastases arose from independent subclones in
   the primary tumor; moreover, the two sites are subject to different
   levels of selection, indicating that the origin cells of lymphatic and
   distant metastases are fundamentally different [[111]5, [112]6].

   Cancer cells enter the bloodstream through intravasation to become CTCs
   [[113]1], which have been proven to be highly correlated with the
   survival outcomes of breast cancer patients [[114]21]. The lymphatic
   system is reported to be critical for metastasis. Lymphatic
   system-experienced melanoma cells were more resistant to ferroptosis
   and formed more metastases [[115]16]. In primary tumor resection
   models, cancer cells can further disseminate from lymph nodes into
   blood vessels and form lung metastatic tumors [[116]22]. Although
   metastatic spread is feasible via both routes, the relative importance
   of each process for a given cancer type and target organ is unknown
   owing to the lack of appropriate quantitative tools. Leveraging the
   CellTag system, we demonstrated that the majority of lung and liver
   metastatic cells can be detected in the lymphatic system instead of as
   blood-derived CTCs, indicating that lymphatic vessels could be the
   preferred route for tumor cell metastasis, at least in breast cancer.

   Tumor cells demonstrate high intra- and intertumor heterogeneity
   [[117]23] and exhibit multiclonal evolution during tumorigenesis, drug
   treatment, and metastasis [[118]24–[119]26]. Via the cell labelling
   system, we were able to perform lineage tracing to define individual
   cell fates. In our metastasis models, different tagged cells, even in
   established cell lines, consistently demonstrated migration route bias
   and organotropic metastasis, indicating that intrinsic mechanisms were
   prewired in each tumor cell before it disseminated from the primary
   site, controlling its heritable and reproducible migration fate in
   individual mice. Owing to its tremendous complexity and diversity, the
   immune system performs multifaceted functions in the TME, with
   different immune cells playing distinct or sometimes even opposing
   roles. Emerging evidence indicates that intrinsic tumor genetic
   aberrations shape the tumor milieu. Intratumor heterogeneity
   predisposes tumors to an elusive tumor-TME crosstalk pattern, making
   exploitation of the intrinsic properties of cancer cells impossible
   without high-throughput approaches for cell fate classification.
   Further scRNA seq and cell–cell interaction analyses allowed us to
   identify the unique features of metastatic tumor cells in primary
   tumors, for example, Il2-Il2rg and Flt3l-Flt3, which mediate
   interactions between dendritic cells and endothelial cells. Consistent
   with this, in our subsequent preliminary tests, we found silencing Il2
   in 4T1 cells could dramatic decrease lymphatic-disseminated tumor
   cells. We will continually work on this to further under the mechanism
   of lymphatic dissemination and explore the therapeutic disruptions in
   our future efforts.

   Besides, communication interactions between tumor cells and stromal
   cells correlate with their metastasis route. For instance, we observed
   that fibroblasts exhibited extensive communication with tumor cells,
   particularly through the secretion of cytokines and growth factors such
   as TGFβ and PDGF. These interactions are consistent with previous
   studies showing that cancer-associated fibroblasts (CAFs) promote tumor
   cell invasion, migration, and survival by remodeling the extracellular
   matrix (ECM) and creating a pro-metastatic niche [[120]27, [121]28].
   Specifically, our single-cell RNA sequencing (scRNA-seq) data reveal
   that fibroblasts in the primary tumor microenvironment expressed high
   levels of TGFβ2, which is known to enhance tumor cell invasiveness and
   epithelial-mesenchymal transition (EMT) [[122]29, [123]30].
   Additionally, fibroblasts were found to interact with tumor cells
   through ligand-receptor pairs such as PDGFC-PDGFRB, which are
   implicated in promoting tumor cell proliferation and survival
   [[124]31]. These findings suggest that fibroblasts not only support
   tumor cell dissemination but also play a role in preparing the "soil"
   for metastatic colonization. Beyond fibroblasts, our data also show
   that tumor cells interact closely with endothelial cells, particularly
   through VEGF and TGFβ signaling pathways. VEGF, a well-known angiogenic
   factor, is highly expressed by tumor cells and interacts with
   endothelial cell receptors such as KDR and FLT1. This interaction is
   crucial for tumor cell intravasation, as VEGF signaling promotes
   vascular permeability, allowing tumor cells to enter the bloodstream or
   lymphatic system [[125]32]. Moreover, our study highlights the role of
   TGFβ signaling in endothelial cell-tumor cell communication. TGFβ2,
   expressed by tumor cells, is found to interact with endothelial cells,
   potentially promoting endothelial cell activation and facilitating
   tumor cell extravasation at distant sites. This is consistent with
   previous studies showing that TGFβ signaling enhances endothelial cell
   permeability and supports the formation of metastatic niches [[126]33].

   Tumor cells colonizing distinct organs/tissues to form metastatic
   tumors usually exhibit tremendously different gene expression patterns
   [[127]10]. Because only a small portion of shed tumor cells can form
   metastatic tumors, RNA-seq (at either the bulk or single-cell level)
   cannot determine whether organotropism is generated because different
   seeds preferentially migrate to different organs or because the same
   seeds are educated for differential gene expression in different soils.
   Therefore, this lineage tracing approach coupled with scRNA-seq
   analysis provided a powerful means to distinguish between these
   alternatives. In this regard, our data precisely elucidated that tumor
   cells derived from the same progenitor cell exhibit different gene
   expression patterns in different soils and that the cancer cell-TME
   communication paradigm varies significantly between primary tumors and
   metastatic tumors.

   In summary, leveraging the CellTag system, we demonstrated the route of
   breast cancer metastasis and identified the potent intrinsic drivers of
   lymphatic dissemination (Figure S8). To strengthen the clinical utility
   of tumor metastasis prediction and therapy, prospective studies should
   exploit effective cellular markers for cellular and molecular
   properties and identify targets for pharmacological treatment to block
   or eliminate cancer cell dissemination.

Supplementary Information

   [128]12943_2025_2279_MOESM1_ESM.docx^ (56.8MB, docx)

   Additional file 1. Table S1-Celltag information in 4T1 tumor-bearing
   mice. Table S2-Celltag information in EMT6 tumor-bearing mice. Table
   S3-Celltag information in tail vein injection mice model. Table
   S4-Candidate celltag for tumor cell dissemination. Table S5-Lymph node
   Celltag information in 4T1 tumor-bearing mice. Table S6 primers for
   Bulk DNA barcode sequencing.

Acknowledgements