Abstract
Tumours can obtain nutrients and oxygen required to progress and metastasize through the blood supply1. Inducing angiogenesis involves the sprouting of established vessel beds and their maturation into an organized network2,3. Here we generate a comprehensive atlas of tumour vasculature at single-cell resolution, encompassing approximately 200,000 cells from 372 donors representing 31 cancer types. Trajectory inference suggested that tumour angiogenesis was initiated from venous endothelial cells and extended towards arterial endothelial cells. As neovascularization elongates (through angiogenic stages SI, SII and SIII), APLN+ tip cells at the SI stage (APLN+ TipSI) advanced to TipSIII cells with increased Notch signalling. Meanwhile, stalk cells, following tip cells, transitioned from high chemokine expression to elevated TEK (also known as Tie2) expression. Moreover, APLN+ TipSI cells not only were associated with disease progression and poor prognosis but also hold promise for predicting response to anti-VEGF therapy. Lymphatic endothelial cells demonstrated two distinct differentiation lineages: one responsible for lymphangiogenesis and the other involved in antigen presentation. In pericytes, endoplasmic reticulum stress was associated with the proangiogenic BASP1+ matrix-producing pericytes. Furthermore, intercellular communication analysis showed that neovascular endothelial cells could shape an immunosuppressive microenvironment conducive to angiogenesis. This study depicts the complexity of tumour vasculature and has potential clinical significance for anti-angiogenic therapy.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 51 print issues and online access
$199.00 per year
only $3.90 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
Raw sequencing data, processed scRNA and spatial transcriptome datasets reported in this paper have been deposited at the Genome Sequence Archive at the National Genomics Data Center (Beijing, China) under BioProject PRJCA018695. The data deposited and made public are compliant with the regulations of the Ministry of Science and Technology of the People’s Republic of China. Integrated gene expression data can be accessed through an online data browser (http://resource.yin-lab.com/Panvascular/). Public scRNA datasets and spacial transcriptomiocs data used in this study are shown in Supplementary Table 1 and TCGA datasets are from https://portal.gdc.cancer.gov/.
Code availability
Custom codes have been deposited at GitHub (https://github.com/bio-Pixel/panVC) and Zenodo91 (https://zenodo.org/records/11188740).
References
Chung, A. S., Lee, J. & Ferrara, N. Targeting the tumour vasculature: insights from physiological angiogenesis. Nat. Rev. Cancer 10, 505–514 (2010).
Potente, M. & Mäkinen, T. Vascular heterogeneity and specialization in development and disease. Nat. Rev. Mol. Cell Biol. 18, 477–494 (2017).
Jain, R. K. et al. Angiogenesis in brain tumours. Nat. Rev. Neurosci. 8, 610–622 (2007).
Bergers, G. & Benjamin, L. E. Tumorigenesis and the angiogenic switch. Nat. Rev. Cancer 3, 401–410 (2003).
Stapor, P. C., Sweat, R. S., Dashti, D. C., Betancourt, A. M. & Murfee, W. L. Pericyte dynamics during angiogenesis: new insights from new identities. J. Vasc. Res. 51, 163–174 (2014).
Lee, H.-W., Shin, J. H. & Simons, M. Flow goes forward and cells step backward: endothelial migration. Exp. Mol. Med. 54, 711–719 (2022).
Kalucka, J. et al. Single-cell transcriptome atlas of murine endothelial cells. Cell 180, 764–779 (2020).
Niu, G. & Chen, X. Vascular endothelial growth factor as an anti-angiogenic target for cancer therapy. Curr. Drug Targets 11, 1000–1017 (2010).
Khan, K. A. & Kerbel, R. S. Improving immunotherapy outcomes with anti-angiogenic treatments and vice versa. Nat. Rev. Clin. Oncol. 15, 310–324 (2018).
Huinen, Z. R., Huijbers, E. J. M., van Beijnum, J. R., Nowak-Sliwinska, P. & Griffioen, A. W. Anti-angiogenic agents—overcoming tumour endothelial cell anergy and improving immunotherapy outcomes. Nat. Rev. Clin. Oncol. 18, 527–540 (2021).
Zhu, R. et al. Chemokine expression profiles of ovarian endometriotic stromal cells in three-dimensional culture. J. Reprod. Immunol. 138, 103100 (2020).
Hernández-García, R., Iruela-Arispe, M. L., Reyes-Cruz, G. & Vázquez-Prado, J. Endothelial RhoGEFs: a systematic analysis of their expression profiles in VEGF-stimulated and tumor endothelial cells. Vasc. Pharmacol. 74, 60–72 (2015).
Zou, Z. et al. A single-cell transcriptomic atlas of human skin aging. Dev. Cell 56, 383–397 (2021).
Haghverdi, L., Buettner, F. & Theis, F. J. Diffusion maps for high-dimensional single-cell analysis of differentiation data. Bioinformatics 31, 2989–2998 (2015).
Gulati, G. S. et al. Single-cell transcriptional diversity is a hallmark of developmental potential. Science 367, 405–411 (2020).
Nunes, S. S. et al. Implanted microvessels progress through distinct neovascularization phenotypes. Microvasc. Res. 79, 10–20 (2010).
De Smet, F., Segura, I., De Bock, K., Hohensinner, P. J. & Carmeliet, P. Mechanisms of vessel branching: filopodia on endothelial tip cells lead the way. Arterioscler. Thromb. Vasc. Biol. 29, 639–649 (2009).
Aibar, S. et al. SCENIC: single-cell regulatory network inference and clustering. Nat. Methods 14, 1083–1086 (2017).
Siemerink, M. J., Klaassen, I., Van Noorden, C. J. F. & Schlingemann, R. O. Endothelial tip cells in ocular angiogenesis: potential target for anti-angiogenesis therapy. J. Histochem. Cytochem. 61, 101–115 (2012).
Whittall, C. et al. A chemokine self-presentation mechanism involving formation of endothelial surface microstructures. J. Immunol. 190, 1725–1736 (2013).
Savant, S. et al. The orphan receptor Tie1 controls angiogenesis and vascular remodeling by differentially regulating Tie2 in Tip and stalk cells. Cell Rep. 12, 1761–1773 (2015).
Augustin, H. G., Young Koh, G., Thurston, G. & Alitalo, K. Control of vascular morphogenesis and homeostasis through the angiopoietin–Tie system. Nat. Rev. Mol. Cell Biol. 10, 165–177 (2009).
Park, J. S. et al. Normalization of tumor vessels by Tie2 activation and Ang2 inhibition enhances drug delivery and produces a favorable tumor microenvironment. Cancer Cell 30, 953–967 (2016).
Scharpfenecker, M., Fiedler, U., Reiss, Y. & Augustin, H. G. The Tie-2 ligand angiopoietin-2 destabilizes quiescent endothelium through an internal autocrine loop mechanism. J. Cell Sci. 118, 771–780 (2005).
Ambade, A. S., Hassoun, P. M. & Damico, R. L. Basement membrane extracellular matrix proteins in pulmonary vascular and right ventricular remodeling in pulmonary hypertension. Am. J. Respir. Cell Mol. Biol. 65, 245–258 (2021).
Kuo, C. J. et al. Oligomerization-dependent regulation of motility and morphogenesis by the collagen XVIII NC1/endostatin domain. J. Cell Biol. 152, 1233–1246 (2001).
Moulton, K. S. et al. Loss of collagen XVIII enhances neovascularization and vascular permeability in atherosclerosis. Circulation 110, 1330–1336 (2004).
Xu, C. et al. Arteries are formed by vein-derived endothelial tip cells. Nat. Commun. 5, 5758 (2014).
Hasan, S. S. et al. Endothelial Notch signalling limits angiogenesis via control of artery formation. Nat. Cell Biol. 19, 928–940 (2017).
Tammela, T. et al. VEGFR-3 controls tip to stalk conversion at vessel fusion sites by reinforcing Notch signalling. Nat. Cell Biol. 13, 1202–1213 (2011).
Luo, W. et al. Arterialization requires the timely suppression of cell growth. Nature 589, 437–441 (2021).
Ferrara, N., Hillan, K. J., Gerber, H.-P. & Novotny, W. Discovery and development of bevacizumab, an anti-VEGF antibody for treating cancer. Nat. Rev. Drug Discov. 3, 391–400 (2004).
Flister, M. J. et al. Inflammation induces lymphangiogenesis through up-regulation of VEGFR-3 mediated by NF-κB and Prox1. Blood 115, 418–429 (2010).
Ayroldi, E. et al. Modulation of T-cell activation by the glucocorticoid-induced leucine zipper factor via inhibition of nuclear factor κB. Blood 98, 743–753 (2001).
Teuwen, L.-A. et al. Tumor vessel co-option probed by single-cell analysis. Cell Rep. 35, 109253 (2021).
Nwadozi, E., Rudnicki, M. & Haas, T. L. Metabolic coordination of pericyte phenotypes: therapeutic implications. Front. Cell Dev. Biol. 8, 77 (2020).
De Bock, K. et al. Role of PFKFB3-driven glycolysis in vessel sprouting. Cell 154, 651–663 (2013).
Cantelmo, A. R. et al. Inhibition of the glycolytic activator PFKFB3 in endothelium induces tumor vessel normalization, impairs metastasis, and improves chemotherapy. Cancer Cell 30, 968–985 (2016).
Månberg, A. et al. Altered perivascular fibroblast activity precedes ALS disease onset. Nat. Med. 27, 640–646 (2021).
Binet, F. & Sapieha, P. ER stress and angiogenesis. Cell Metab. 22, 560–575 (2015).
Edagawa, M. et al. Role of activating transcription factor 3 (ATF3) in endoplasmic reticulum (ER) stress-induced sensitization of p53-deficient human colon cancer cells to tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL)-mediated apoptosis through up-regulation of death receptor 5 (DR5) by zerumbone and celecoxib. J. Biol. Chem. 289, 21544–21561 (2014).
Ma, Y. & Zhou, X. Spatially informed cell-type deconvolution for spatial transcriptomics. Nat. Biotechnol. 40, 1349–1359 (2022).
Goswami, D. et al. Endothelial CD99 supports arrest of mouse neutrophils in venules and binds to neutrophil PILRs. Blood 129, 1811–1822 (2017).
Geldhof, V. et al. Single cell atlas identifies lipid-processing and immunomodulatory endothelial cells in healthy and malignant breast. Nat. Commun. 13, 5511 (2022).
Zheng, S. et al. Mesenchymal stromal cells rapidly suppress TCR signaling-mediated cytokine transcription in activated T cells through the ICAM-1/CD43 interaction. Front. Immunol. 12, 609544 (2021).
Ausprunk, D. H. & Folkman, J. Migration and proliferation of endothelial cells in preformed and newly formed blood vessels during tumor angiogenesis. Microvasc. Res. 14, 53–65 (1977).
Lee, H. W. et al. Role of venous endothelial cells in developmental and pathologic angiogenesis. Circulation 144, 1308–1322 (2021).
Rohlenova, K. et al. Single-cell RNA sequencing maps endothelial metabolic plasticity in pathological angiogenesis. Cell Metab. 31, 862–877 (2020).
Pitulescu, M. E. et al. Dll4 and Notch signalling couples sprouting angiogenesis and artery formation. Nat. Cell Biol. 19, 915–927 (2017).
Kanda, A., Hirose, I., Noda, K., Murata, M. & Ishida, S. Glucocorticoid-transactivated TSC22D3 attenuates hypoxia- and diabetes-induced Müller glial galectin-1 expression via HIF-1α destabilization. J. Cell. Mol. Med. 24, 4589–4599 (2020).
Bergers, G. & Song, S. The role of pericytes in blood-vessel formation and maintenance. Neuro Oncol. 7, 452–464 (2005).
Mayer, J. U. et al. Homeostatic IL-13 in healthy skin directs dendritic cell differentiation to promote TH2 and inhibit TH17 cell polarization. Nat. Immunol. 22, 1538–1550 (2021).
Dura, B. et al. scFTD-seq: freeze-thaw lysis based, portable approach toward highly distributed single-cell 3′ mRNA profiling. Nucleic Acids Res. 47, e16 (2019).
Zheng, G. X. Y. et al. Massively parallel digital transcriptional profiling of single cells. Nat. Commun. 8, 14049 (2017).
Klein, AllonM. et al. Droplet barcoding for single-cell transcriptomics applied to embryonic stem cells. Cell 161, 1187–1201 (2015).
Wen, L. et al. The blood flow-klf6a-tagln2 axis drives vessel pruning in zebrafish by regulating endothelial cell rearrangement and actin cytoskeleton dynamics. PLoS Genet. 17, e1009690 (2021).
Hao, Y. et al. Integrated analysis of multimodal single-cell data. Cell 184, 3573–3587 (2021).
Wolock, S. L., Lopez, R. & Klein, A. M. Scrublet: computational identification of cell doublets in single-cell transcriptomic data. Cell Syst. 8, 281–291 (2019).
Korsunsky, I. et al. Fast, sensitive and accurate integration of single-cell data with Harmony. Nat. Methods 16, 1289–1296 (2019).
Zheng, L. et al. Pan-cancer single-cell landscape of tumor-infiltrating T cells. Science 374, abe6474 (2021).
van den Brink, S. C. et al. Single-cell sequencing reveals dissociation-induced gene expression in tissue subpopulations. Nat. Methods 14, 935–936 (2017).
Xue, R. et al. Liver tumour immune microenvironment subtypes and neutrophil heterogeneity. Nature 612, 141–147 (2022).
He, Z., Brazovskaja, A., Ebert, S., Camp, J. G. & Treutlein, B. CSS: cluster similarity spectrum integration of single-cell genomics data. Genome Biol. 21, 224 (2020).
Yu, Q. et al. Charting human development using a multi-endodermal organ atlas and organoid models. Cell 184, 3281–3298 (2021).
He, Z. et al. Lineage recording in human cerebral organoids. Nat. Methods 19, 90–99 (2022).
Muhl, L. et al. Single-cell analysis uncovers fibroblast heterogeneity and criteria for fibroblast and mural cell identification and discrimination. Nat. Commun. 11, 3953 (2020).
Zhang, L. et al. Lineage tracking reveals dynamic relationships of T cells in colorectal cancer. Nature 564, 268–272 (2018).
Zhao, Q. et al. Single-cell transcriptome analyses reveal endothelial cell heterogeneity in tumors and changes following antiangiogenic treatment. Cancer Res. 78, 2370–2382 (2018).
Li, Z. et al. Single-cell transcriptome analyses reveal novel targets modulating cardiac neovascularization by resident endothelial cells following myocardial infarction. Eur. Heart J. 40, 2507–2520 (2019).
Goveia, J. et al. An integrated gene expression landscape profiling approach to identify lung tumor endothelial cell heterogeneity and angiogenic candidates. Cancer Cell 37, 21–36 (2020).
Street, K. et al. Slingshot: cell lineage and pseudotime inference for single-cell transcriptomics. BMC Genomics 19, 477 (2018).
Cao, J. et al. The single-cell transcriptional landscape of mammalian organogenesis. Nature 566, 496–502 (2019).
Stassen, S. V., Yip, G. G. K., Wong, K. K. Y., Ho, J. W. K. & Tsia, K. K. Generalized and scalable trajectory inference in single-cell omics data with VIA. Nat. Commun. 12, 5528 (2021).
Angerer, P. et al. destiny: diffusion maps for large-scale single-cell data in R. Bioinformatics 32, 1241–1243 (2016).
Qiu, X. et al. Mapping transcriptomic vector fields of single cells. Cell 185, 690–711 (2022).
Lange, M. et al. CellRank for directed single-cell fate mapping. Nat. Methods 19, 159–170 (2022).
Marxsen, J. H. et al. Hypoxia-inducible factor-1 (HIF-1) promotes its degradation by induction of HIF-α-prolyl-4-hydroxylases. Biochem. J. 381, 761–767 (2004).
Belaiba, R. S. et al. Hypoxia up-regulates hypoxia-inducible factor-1α transcription by involving phosphatidylinositol 3-kinase and nuclear factor κB in pulmonary artery smooth muscle cells. Mol. Biol. Cell 18, 4691–4697 (2007).
Han, H. et al. TRRUST v2: an expanded reference database of human and mouse transcriptional regulatory interactions. Nucleic Acids Res. 46, D380–D386 (2018).
Semenza, G. L. Hypoxia-inducible factor 1: oxygen homeostasis and disease pathophysiology. Trends Mol. Med. 7, 345–350 (2001).
Holland, C. H. et al. Robustness and applicability of transcription factor and pathway analysis tools on single-cell RNA-seq data. Genome Biol. 21, 36 (2020).
Badia, I. M. P. et al. decoupleR: ensemble of computational methods to infer biological activities from omics data. Bioinform. Adv. 2, vbac016 (2022).
Camp, J. G. et al. Multilineage communication regulates human liver bud development from pluripotency. Nature 546, 533–538 (2017).
Zhou, Y. et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat. Commun. 10, 1523 (2019).
Van den Berge, K. et al. Trajectory-based differential expression analysis for single-cell sequencing data. Nat. Commun. 11, 1201 (2020).
Shen, W.-K. et al. AnimalTFDB 4.0: a comprehensive animal transcription factor database updated with variation and expression annotations. Nucleic Acids Res. 51, D39–D45 (2023).
Browaeys, R., Saelens, W. & Saeys, Y. NicheNet: modeling intercellular communication by linking ligands to target genes. Nat. Methods 17, 159–162 (2020).
Bilous, M. et al. Metacells untangle large and complex single-cell transcriptome networks. BMC Bioinform. 23, 336 (2022).
Efremova, M., Vento-Tormo, M., Teichmann, S. A. & Vento-Tormo, R. CellPhoneDB: inferring cell-cell communication from combined expression of multi-subunit ligand-receptor complexes. Nat. Protoc. 15, 1484–1506 (2020).
Chu, T., Wang, Z., Pe’er, D. & Danko, C. G. Cell type and gene expression deconvolution with BayesPrism enables Bayesian integrative analysis across bulk and single-cell RNA sequencing in oncology. Nat. Cancer 3, 505–517 (2022).
Pan, X. et al. Codes for ‘Tumour vasculature at the single-cell resolution’. Zenodo https://doi.org/10.5281/zenodo.11188740 (2024).
Acknowledgements
This study is jointly supported by National Key R&D Programmes (NKPs) of China (grant no. 2022YFC3601800), the Natural Science Foundation of China General Program (grant no. 82073020), the Fundamental Research Funds for the Central Universities (grant no. 2023CDJYGRH-YB05), Science and Technology Innovation Key R&D Program of Chongqing (grant no. CSTB2023TIAD-STX0011), Chongqing Wanzhou Municipal Science and Health Joint Medical Research Project Key Program (grant no. wzstc-kw2023001), Chongqing Wanzhou PhD “through train” Research Project (grant no. wzstc20230402), Hunan Natural Science Foundation for Distinguished Young Scholars (grant no. 2021JJ10073), the Natural Science Foundation of China (grant no. 32200523 and grant no. 82072748), Chongqing Natural Science Foundation (grant no. CSTB2023NSCQ-MSX0658), the CAMS Innovation Fund for Medical Sciences (2021-12M-1-002,2023-12M-2-002) and National High Level Hospital Clinical Research Funding (2022-PUMCH-D-001 and 2022-PUMCH-A-082). We thank Q. Yan for suggestions on this study, Y. Zhao and J. Peng for supporting this project, and X. Hu for the illustrations. Graphs were also created using BioRender.
Author information
Authors and Affiliations
Contributions
M.Y. and X.P. conceived the presented idea. X.P. and X.L. designed and performed bioinformatic analyses. L.D. and M.Y. performed the mIHC staining and survival analysis. H. Shao and Y.W. constructed the animal model. M.Y., M.Z. and H. Sun collected tissue samples. T.L., L.Z., X.Z., L.H., W.S., Z.F., J.S. and Y.H. collected single-cell data. X.P., X.L., L.D. and M.Y. wrote the manuscript. M.Y. supervised this project. All of the authors discussed the results and contributed to the final manuscript.
Corresponding author
Ethics declarations
Competing interests
M.Y. is the CSO of Zhejiang Wenda Medical Technology. The other authors declare no competing interests.
Peer review
Peer review information
Nature thanks Joanna Kalucka and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Extended Data Fig. 1 Data integration of vascular cells.
a. UMAP plots showing the distribution of vascular cells following the integration steps (Original count to Z-scaled to CSS). After CSS processing, cells produced by different platforms were clustered obviously. b. UMAP plot showing selected marker gene expression of cell types. c. Boxplot showing the LISI (Local Inverse Simpson’s Index) values after each integration step (n = 116,958 cells). For boxplots, centre line represents median, box limits indicate upper and lower quartiles, and whiskers extend 1.5 times the interquartile range.
Extended Data Fig. 2 Data integration of vascular cells.
a. UMAP plots showing the joint density estimation of PECAM1-PDGFRB and PECAM1-DCN. Violin plot showing the expression levels of PECAM1, PDGFRB, and DCN across VECs, EndoMT cells, and MCs. b. Bar plot showing the total cell number (right) and percentages (left) of vascular cells in ANT and tumour tissues. c. Box plot showing the proportion of vascular cells in each sample. Dots represent single-cell samples (ANT, n = 99, Tumour, n = 338), coloured by cancer types. P-value was calculated by two-sided Mann–Whitney U test. d. Violin plot showing the expression levels of selected markers in VEC/LEC subpopulations with similar tissue origin. For boxplots, centre line represents median, box limits indicate upper and lower quartiles, and whiskers extend 1.5 times the interquartile range.
Extended Data Fig. 3 Tumour angiogenesis initiating from VenEC.
a. Visualization of angiogenesis trajectories inferred by Slingshot, VIA, and Monocle3. b. Velocity-based cell state inference supports VenEC as angiogenesis initiation. c. Boxplot showing the CytoTRACE scores among VenEC (n = 37,510), CapEC (n = 61,400), and ArtEC (n = 18,048). Less diff., less differentiated; More diff., more differentiated. d. Boxplot showing cell purity for VenEC (n = 37,510), CapEC (n = 61,400), and ArtEC (n = 18,048) by ROGUE. e. UMAP plot showing the tip-like, stalk-like, and RT/S scores across CapECs. Boxplot showing the distribution of RT/S values among CapEC subclusters. Six subclusters (V1 [n = 6,351], V7 [n = 5,376], V9 [n = 5,268], V12 [n = 4,831], V18 [n = 3,687], and V24 [n = 3,291]) were annotated as “tip-like” CapECs, three (V6 [n = 5,517], V16 [n = 4,399], and V23 [n = 3,361]) as “stalk-like” CapECs, and three (V0 [n = 8,194], V10 [n = 4,946], and V2 [n = 6,179]) in a “Transition” state between tip- and stalk-like cells. f. Box plot showing the mean proportions of each cancer type across VEC subtypes in ANT and tumour tissues, respectively. Dots represent cancer types (n = 31). Briefly, we first calculated the proportion of each cell type in each sample and then calculated the average proportion in each cancer. g. Boxplot showing the sprouting rate between ANT and tumour tissues. Dots represent cancer types (n = 31). Sprouting rate was expressed as the ratio of tip-like cell to VenEC. For boxplots, centre line represents median, box limits indicate upper and lower quartiles, and whiskers extend 1.5 times the interquartile range. Statistical analysis was performed using two-sided Mann–Whitney U test (d and g) and Kruskal-Wallis test (c and f).
Extended Data Fig. 4 Tip-like and stalk-like cells in TVM.
a-b. Visualization of trajectories inferred by Slingshot, VIA, and Monocle3 in tip-like (a) and stalk-like (b) cells, respectively. c. Boxplot showing the CytoTRACE scores among differentiation stages of tip-like (top) and stalk-like (bottom) cells (tip cells: SI, n = 9,063, SII, n = 13,390, SIII, n = 6,351; stalk cells: SI, n = 4,399, SII, n = 3,361, SIII, n = 5,517). d. Violin plots showing the HIF1A activity at transcription levels (top), AUCell score of target genes (middle), and DoRothEA activity (bottom). Error bar represents mean ± standard deviation. e. Violin plots showing the expression of staged markers. f. Representative mIHC of APLN+ TipSI markers in LUAD from n = 10 independent biological replicates, GBM from n = 11 independent biological replicates, and HCC from n = 10 independent biological replicates. Scale bar, 50 µm. g. SOM (50 × 50 grid) constructed from single-cell transcriptomes of tip-like and stalk-like cells showing scaled metagene expression. Black lines demarcate eight overexpressed metagene-signatures comprising 971 overexpressed genes. h. Function enrichment of overexpressed metagenes interrogated by SOM, resulting in ten TAPFs. P-values are calculated based on the cumulative hypergeometric distribution. i-j. Tip-to-stalk communications in angiogenic stages. Heatmap showing the activities of top 20 ligands (i). Dot plot showing the stage-related LR pairs (j), sized by the pseudobulk Spearman’s correlation P-value and coloured by the interaction potential predicted using NicheNet. k. Violin plot showing the expression levels of genes associated with TAPFs. For boxplots, centre line represents median, box limits indicate upper and lower quartiles, and whiskers extend 1.5 times the interquartile range. Statistical analysis was performed using Kruskal-Wallis test (c).
Extended Data Fig. 5 Clinical implications of VECs.
a. Box plot showing the mean proportions of APLN+ TipSI in each cancer type between ANT and tumour tissues. Dots represent cancer types (n = 31). b. The median infiltration levels of APLN+ TipSI in each TCGA cancer type across ANT, stage I/II, and stage III/IV samples. GBM was graded by G2, G3, and G4. PRAD was graded by Gleason score (GS). We performed the two-sided Mann–Whitney U test on the mean infiltration levels of APLN+ TipSI between stage I/II and ANT samples in each cancer type. Different stages of the same cancer are connected using coloured lines when P < 0.05. c. Heatmap showing the effect of VEC infiltration on OS (right) and PFS (left). d. Variations of VEC infiltration levels following bevacizumab treatment. Boxplots showing differences between pre-treatment responders (R) and non-responders (NR) in UCPH-GBM (N, n = 6, NR, n = 10), KU Leuven-READ (N, n = 6, NR, n = 5), and SYSU-READ (n = 4). e. UMAP plot showing the cell types (left), treatment time points (middle), and patients (right) in SYSU-READ. Table showing the patients’ characteristics. For boxplots, centre line represents median, box limits indicate upper and lower quartiles, and whiskers extend 1.5 times the interquartile range. Statistical analysis was performed using two-sided Mann–Whitney U test (a and d).
Extended Data Fig. 6 LEC in TVM.
a. Pie chart showing the percentage of NLEC detected from ANT and tumour tissues. b. Box plot showing the proportion of NLEC in each sample. Dots represent single-cell samples (ANT, n = 83, Tumour, n = 253), coloured by cancer types. c. Visualization of the LEC trajectories inferred by VIA and Monocle3. d. Bubble plot showing the odds ratio of LEC subpopulations enriched in LEC lineages. Odds ratios are computed using Fisher’s exact test. e. Function enrichment of marker genes of apLEC, tip-like LEC, and stalk-like LEC. P-values are calculated based on the cumulative hypergeometric distribution. f. The activity of HIF1A across LEC subtypes (NLEC, n = 2,872; inter. LEC, n = 2,574; apLEC, n = 1,309; Stalk-like LEC, n = 975; Tip-like LEC, n = 1553). Error bar represents mean ± standard deviation. g. Violin plots showing the expression of PROX1, NFKB1, and TSC22D3 among LECs. h. Correlation of the infiltration levels of apLEC and neolymphatic vessels in the scRNA-seq dataset (left) and TCGA bulk-seq dataset (right). Pearson’s correlation coefficient and its P-value are shown. ****, P < 2.2×10−16. i. Representative examples of mIHC staining with anti-LYVE1 (yellow), anti-NFKB1 (red), and anti-TSC22D3 (green) in HCC from n = 12 independent biological replicates. Scale bar, 50 µm. j. Kaplan–Meier curves of OS for TCGA-HCC samples. Tumour samples were stratified by the median infiltration levels of tip-like LECs (left) and apLECs (right), respectively. k. Heatmap showing the effect of LEC infiltration and RT/ap on OS (left) and PFS (right). For boxplots, centre line represents median, box limits indicate upper and lower quartiles, and whiskers extend 1.5 times the interquartile range. Statistical analysis was performed using two-sided Mann–Whitney U test (b) and log-rank test (j).
Extended Data Fig. 7 Characteristics of PC phenotypes.
a. Boxplot showing the proportions of SMCs and PCs in ANT (n = 77) and tumour (n = 265) samples, respectively. b. Violin plot showing cell purity for SMC (n = 10,819) and PC (n = 29,572) by ROGUE. Error bar represents mean ± standard deviation. c. Bar plot showing the cell number (top) and percentages (bottom) of cancer types across MC subtypes. Box plot showing the mean proportions of each cancer type between ANT (n = 77) and tumour (n = 265) samples across MC subtypes. d. UMAP plot showing the density of quiescent score. MatPCs showing heterogeneous quiescent states. e. Heatmap showing the activities of energy metabolism pathways among PC subclusters. f. Boxplot showing cell purity of MC subtypes by ROGUE. SMC, n = 10,819, adiPC, n = 2,562, matPC, n = 10,230, myoPC, n = 5,599, vdPC, n = 11,181. g. Visualization of the matPC trajectories inferred by Slingshot, VIA, and Monocle3. h. Boxplot showing the CytoTRACE score among matPC subtypes. BASP1+ matPC, n = 2,060, inter. matPC, n = 4,211, matPCQ, n = 3,959. i. Loess regression-smoothened gene expression of the oxidative-stress genes (y-axis) associated with in pseudotime. Violin plot showing the expression levels of corresponding genes. j. Violin plot showing the SCENIC scores of ATF3 (right) and XBP1 (left). BASP1+ matPC, n = 2,060, inter. matPC, n = 4,211, matPCQ, n = 3,959. k-l. Loess regression-smoothened expression (y-axis in k) and SCENIC activity (y-axis in f) in pseudotime. Spearman’s correlation coefficient and its P-value are shown in f. ****, P < 0.0001. m. TF activity in BASP1+ matPC. ER-stress-related TFs were labelled. Shaded region represents the 95% confidence intervals (i, k, and l). For boxplots, centre line represents median, box limits indicate upper and lower quartiles, and whiskers extend 1.5 times the interquartile range. Statistical analysis was performed using two-sided Mann–Whitney U test (a, b, and c) and Kruskal–Wallis test (f, h, and j).
Extended Data Fig. 8 BASP1+ matPC is involved in tumour angiogenesis.
a. Correlation between the proportion of BASP1+ matPC and APLN+ TipSI. Shaded region represents the 95% confidence intervals. Spearman’s correlation coefficient and its P-value are shown. Dots coloured by cancer types. b. Co-localization between APLN+ TipSI and BASP1+ matPC at spatial positions. Scale bar, 1 mm. c. LR pairs between APLN+/TipSI and BASP1+ matPC. d. Violin plot showing the expression levels of VEGF across PC phenotypes. e. Genome browser view showing the binding of XBP1 and ATF3 in the genome region of VEGFA/VEGFB. ChIP-seq data were obtained from public datasets (XBP1: GSE49952 [BC cell lines: T47D, M231, and HS578T] and GSE157117 [OC cell line: A1847]; ATF3: ENCFF020THR [HCC cell line: HepG2], ENCFF341RJA [human induced pluripotent stem cell: WTC11], and ENCFF561SGX [leukaemia cell line: K562]). f. The median infiltration levels of BASP1+ matPC in each TCGA cancer type across ANT, stage I/II, and stage III/IV samples. GBM was graded by G2, G3, and G4. PRAD was graded by Gleason score (GS). We performed the two-sided Mann–Whitney U test on the mean infiltration levels of BASP1+ matPC between stage I/II and ANT samples in each cancer type. Different stages of the same cancer are connected using coloured lines when P < 0.05. g. Heatmap showing the effects of MC infiltration on OS (right) and PFS (left).
Extended Data Fig. 9 Cell interactions during tumour angiogenesis.
a. Dot plot showing the LR pairs between TME cells and APLN+ TipSI when APLN+ TipSI is a sender (top)/receiver (bottom). b. Circos plot showing the VEGF-VEGFR interactions between vascular cells and TME cells. c. Correlation between the proportion of SPP1+ TAM and tip-like VECs. Dots coloured by cancer types. Shaded region represents the 95% confidence intervals. Spearman’s correlation coefficient and its P-value are shown. d. Bar plow showing the number of interactions between SPP1+ TAM and tip cells. e. Co-localization among APLN+ TipSI, SPP1+ TAM, MDSC, myoCAF, matCAF, exhausted T cells, and Treg at spatial positions from ST data of BC (n = 1 case), NSCLC (n = 1 case), PDAC (n = 1 case), OC (n = 1 case), ccRCC (n = 1 case), and cSCC (n = 1 case). Scale bar, 1 mm. f. Representative mIHC of APLN+ tip cells and SPP1+ TAM in NSCLC from n = 7 independent biological replicates. Arrows depict the specific cell types.
Extended Data Fig. 10 LR pairs associated with APLN+ TipSI.
a. Spatial plot showing the expression intensity of LR pairs of PAC interactions. APLN+ TipSI can recruit PACs via CD99-PILRA, PGF-/PDGFB-related LR pairs. b. Representative mIHC of PDGFB-expressing APLN+ tip cells and PDGFRB-expressing fibroblasts in CSC from n = 8 independent biological replicates. Arrows depict the specific cell types. c. Spatial plot showing the expression intensity of LR pairs of Treg homing PODXL/CD34-SELL. d. Neovascularization can interact with T cells via GRN-TNFRSF1B, CD34/PODXL-SELL, ICAM1-SPN, and FAM3C-PDCD1/CLEC2D pairs. Expression intensity was calculated by the geometric mean of an LR pair (a and c).
Supplementary information
Supplementary Information
Supplementary Discussion, Supplementary Figs. 1–5 and Supplementary References.
Supplementary Table 1
Overview of datasets analysed in this study.
Supplementary Table 2
Marker genes of vascular cells.
Supplementary Table 3
Angiogenic-stage-related LR pairs between tip and stalk cells.
Supplementary Table 4
Deconvolution analysis of vascular cell types in TCGA cohort.
Supplementary Table 5
Overview of internal BC survival cohorts for VECs.
Supplementary Table 6
Overview of internal HCC survival cohorts for VECs.
Supplementary Table 7
Overview of internal NSCLC survival cohorts for VECs.
Supplementary Table 8
Overview of internal HCC survival cohorts for LECs.
Supplementary Table 9
Significant LR interactions between MCs and ECs.
Supplementary Table 10
Overview of internal GBM survival cohorts for PCs.
Supplementary Table 11
Significant LR interactions between vasculature cells and microenvironmental cells.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Pan, X., Li, X., Dong, L. et al. Tumour vasculature at single-cell resolution. Nature (2024). https://doi.org/10.1038/s41586-024-07698-1
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41586-024-07698-1
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
