Integrative spatial analysis reveals a multi-layered organization of glioblastoma.

Glioma contains malignant cells in diverse states. Here, we combine spatial transcriptomics, spatial proteomics, and computational approaches to define glioma cellular states and uncover their organization. We find three prominent modes of organization. First, gliomas are composed of small local environments, each typically enriched with one major cellular state. Second, specific pairs of states preferentially reside in proximity across multiple scales. This pairing of states is consistent across tumors. Third, these pairwise interactions collectively define a global architecture composed of five layers. Hypoxia appears to drive the layers, as it is associated with a long-range organization that includes all cancer cell states. Accordingly, tumor regions distant from any hypoxic/necrotic foci and tumors that lack hypoxia such as low-grade IDH-mutant glioma are less organized. In summary, we provide a conceptual framework for the organization of cellular states in glioma, highlighting hypoxia as a long-range tissue organizer.

Thymic mimetic cells function beyond self-tolerance.

Development of immunocompetent T cells in the thymus is required for effective defence against all types of pathogens, including viruses, bacteria and fungi. To this end, T cells undergo a very strict educational program in the thymus, during which both non-functional and self-reactive T cell clones are eliminated by means of positive and negative selection1.Thymic epithelial cells (TECs) have an indispensable role in these processes, and previous studies have shown the notable heterogeneity of these cells2-7. Here, using multiomic analysis, we provide further insights into the functional and developmental diversity of TECs in mice, and reveal a detailed atlas of the TEC compartment according to cell transcriptional states and chromatin landscapes. Our analysis highlights unconventional TEC subsets that are similar to functionally well-defined parenchymal populations, including endocrine cells, microfold cells and myocytes. By focusing on the endocrine and microfold TEC populations, we show that endocrine TECs require Insm1 for their development and are crucial to maintaining thymus cellularity in a ghrelin-dependent manner; by contrast, microfold TECs require Spib for their development and are essential for the generation of thymic IgA+ plasma cells. Collectively, our study reveals that medullary TECs have the potential to differentiate into various types of molecularly distinct and functionally defined cells, which not only contribute to the induction of central tolerance, but also regulate the homeostasis of other thymus-resident populations.

ICAM-1 on Breast Cancer Cells Suppresses Lung Metastasis but Is Dispensable for Tumor Growth and Killing by Cytotoxic T Cells.

Breast tumors and their derived circulating cancer cells express the leukocyte ß2 integrin ligand Intercellular adhesion molecule 1 (ICAM-1). We found that elevated ICAM-1 expression in breast cancer cells results in a favorable outcome and prolonged survival of breast cancer patients. We therefore assessed the direct in vivo contribution of ICAM-1 expressed by breast cancer cells to breast tumorigenesis and lung metastasis in syngeneic immunocompetent mice hosts using spontaneous and experimental models of the lung metastasis of the C57BL/6-derived E0771 cell line, a luminal B breast cancer subtype. Notably, the presence of ICAM-1 on E0771 did not alter tumor growth or the leukocyte composition in the tumor microenvironment. Interestingly, the elimination of Tregs led to the rapid killing of primary tumor cells independently of tumor ICAM-1 expression. The in vivo elimination of a primary E0771 tumor expressing the ovalbumin (OVA) model neoantigen by the OVA-specific OVA-tcr-I mice (OT-I) transgenic cytotoxic T lymphocytes (CTLs) also took place normally in the absence of ICAM-1 expression by E0771 breast cancer target cells. The whole lung imaging of these cells by light sheet microscopy (LSM) revealed that both Wild type (WT)- and ICAM-1-deficient E0771 cells were equally disseminated from resected tumors and accumulated inside the lung vasculature at similar magnitudes. ICAM-1-deficient breast cancer cells developed, however, much larger metastatic lesions than their control counterparts. Strikingly, the vast majority of these cells gave rise to intravascular tumor colonies both in spontaneous and experimental metastasis models. In the latter model, ICAM-1 expressing E0771- but not their ICAM-1-deficient counterparts were highly susceptible to elimination by neutrophils adoptively transferred from E0771 tumor-bearing donor mice. Ex vivo, neutrophils derived from tumor-bearing mice also killed cultured E0771 cells via ICAM-1-dependent interactions. Collectively, our results are a first indication that ICAM-1 expressed by metastatic breast cancer cells that expand inside the lung vasculature is involved in innate rather than in adaptive cancer cell killing. This is also a first indication that the breast tumor expression of ICAM-1 is not required for CTL-mediated killing but can function as a suppressor of intravascular breast cancer metastasis to lungs.

DestVI identifies continuums of cell types in spatial transcriptomics data.

Most spatial transcriptomics technologies are limited by their resolution, with spot sizes larger than that of a single cell. Although joint analysis with single-cell RNA sequencing can alleviate this problem, current methods are limited to assessing discrete cell types, revealing the proportion of cell types inside each spot. To identify continuous variation of the transcriptome within cells of the same type, we developed Deconvolution of Spatial Transcriptomics profiles using Variational Inference (DestVI). Using simulations, we demonstrate that DestVI outperforms existing methods for estimating gene expression for every cell type inside every spot. Applied to a study of infected lymph nodes and of a mouse tumor model, DestVI provides high-resolution, accurate spatial characterization of the cellular organization of these tissues and identifies cell-type-specific changes in gene expression between different tissue regions or between conditions. DestVI is available as part of the open-source software package scvi-tools ( https://scvi-tools.org ).

Reduced Lamin A/C Does Not Facilitate Cancer Cell Transendothelial Migration but Compromises Lung Metastasis.

The mechanisms by which the nuclear lamina of tumor cells influences tumor growth and migration are highly disputed. Lamin A and its variant lamin C are key lamina proteins that control nucleus stiffness and chromatin conformation. Downregulation of lamin A/C in two prototypic metastatic lines, B16F10 melanoma and E0771 breast carcinoma, facilitated cell squeezing through rigid pores, and reduced heterochromatin content. Surprisingly, both lamin A/C knockdown cells grew poorly in 3D spheroids within soft agar, and lamin A/C deficient cells derived from spheroids transcribed lower levels of the growth regulator Yap1. Unexpectedly, the transendothelial migration of both cancer cells in vitro and in vivo, through lung capillaries, was not elevated by lamin A/C knockdown and their metastasis in lungs was even dramatically reduced. Our results are the first indication that reduced lamin A/C content in distinct types of highly metastatic cancer cells does not elevate their transendothelial migration (TEM) capacity and diapedesis through lung vessels but can compromise lung metastasis at a post extravasation level.

TP53 missense mutations in PDAC are associated with enhanced fibrosis and an immunosuppressive microenvironment.

Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal cancer, which is refractory to all currently available treatments and bears dismal prognosis. About 70% of all PDAC cases harbor mutations in the TP53 tumor suppressor gene. Many of those are missense mutations, resulting in abundant production of mutant p53 (mutp53) protein in the cancer cells. Analysis of human PDAC patient data from The Cancer Genome Atlas (TCGA) revealed a negative association between the presence of missense mutp53 and infiltration of CD8+ T cells into the tumor. Moreover, CD8+ T cell infiltration was negatively correlated with the expression of fibrosis-associated genes. Importantly, silencing of endogenous mutp53 in KPC cells, derived from mouse PDAC tumors driven by mutant Kras and mutp53, down-regulated fibrosis and elevated CD8+ T cell infiltration in the tumors arising upon orthotopic injection of these cells into the pancreas of syngeneic mice. Moreover, the tumors generated by mutp53-silenced KPC cells were markedly smaller than those elicited by mutp53-proficient control KPC cells. Altogether, our findings suggest that missense p53 mutations may contribute to worse PDAC prognosis by promoting a more vigorous fibrotic tumor microenvironment and impeding the ability of the immune system to eliminate the cancer cells.

Spatiotemporal Proteomic Analysis of Stress Granule Disassembly Using APEX Reveals Regulation by SUMOylation and Links to ALS Pathogenesis.

Stress granules (SGs) are cytoplasmic assemblies of proteins and non-translating mRNAs. Whereas much has been learned about SG formation, a major gap remains in understanding the compositional changes SGs undergo during normal disassembly and under disease conditions. Here, we address this gap by proteomic dissection of the SG temporal disassembly sequence using multi-bait APEX proximity proteomics. We discover 109 novel SG proteins and characterize distinct SG substructures. We reveal dozens of disassembly-engaged proteins (DEPs), some of which play functional roles in SG disassembly, including small ubiquitin-like modifier (SUMO) conjugating enzymes. We further demonstrate that SUMOylation regulates SG disassembly and SG formation. Parallel proteomics with amyotrophic lateral sclerosis (ALS)-associated C9ORF72 dipeptides uncovered attenuated DEP recruitment during SG disassembly and impaired SUMOylation. Accordingly, SUMO activity ameliorated C9ORF72-ALS-related neurodegeneration in Drosophila. By dissecting the SG spatiotemporal proteomic landscape, we provide an in-depth resource for future work on SG function and reveal basic and disease-relevant mechanisms of SG disassembly.

Bone morphology is regulated modularly by global and regional genetic programs.

Bone protrusions provide stable anchoring sites for ligaments and tendons and define the unique morphology of each long bone. Despite their importance, the mechanism by which superstructures are patterned is unknown. Here, we identify components of the genetic program that control the patterning of Sox9+/Scx+ superstructure progenitors in mouse and show that this program includes both global and regional regulatory modules. Using light-sheet fluorescence microscopy combined with genetic lineage labeling, we mapped the broad contribution of the Sox9+/Scx+ progenitors to the formation of bone superstructures. Then, by combining literature-based evidence, comparative transcriptomic analysis and genetic mouse models, we identified Gli3 as a global regulator of superstructure patterning, whereas Pbx1, Pbx2, Hoxa11 and Hoxd11 act as proximal and distal regulators, respectively. Moreover, by demonstrating a dose-dependent pattern regulation in Gli3 and Pbx1 compound mutations, we show that the global and regional regulatory modules work in a coordinated manner. Collectively, our results provide strong evidence for genetic regulation of superstructure patterning, which further supports the notion that long bone development is a modular process.This article has an associated 'The people behind the papers' interview.

Fibroblast recruitment as a tool for ovarian cancer detection and targeted therapy.

Metastatic ovarian cancer, the most lethal of gynecologic malignancies, is typically managed by debulking surgery, followed by chemotherapy. However, despite significant efforts, survival rate remains low. We have previously demonstrated, in mouse models, a specific systemic homing of labeled fibroblasts to solid ovarian tumors. Here, we demonstrate the feasibility of utilizing this specific homing of genetically modified fibroblasts for detection and targeted therapy of orthotopic metastatic ovarian carcinoma model in immune-deficient mice. Using an in vivo metastatic mouse model for ovarian cancer, we demonstrated that fibroblasts expressing fluorescent reporters injected intra-peritoneally, were specifically recruited to peritoneal tumor nodules (resulting in 93-100% co-localization). We further used fibroblasts over expressing the soluble receptor variant of VEGFR1 (s-Flt1). Mice bearing tumors were injected weekly with either control or s-Flt1 expressing fibroblasts. Injection of s-Flt1 expressing fibroblasts resulted in a significant reduction in the ascites volume, reduced vascularization of adherent metastases, and improved overall survival. Using fluorescently labeled fibroblasts for tumor detection with readily available intra-operative fluorescence imaging tools may be useful for tumor staging and directing biopsies or surgical efforts during exploratory or debulking surgery. Fibroblasts may serve as a beacon pointing to the otherwise invisible metastases in the peritoneal cavity of ovarian cancer patients. Utilizing the recruited fibroblasts also for targeted delivery of anti angiogenic or antitumor molecules may aid in controlling tumor progression. Thus, these results suggest a novel approach for targeting ovarian tumor metastases for both tumor detection and therapy.

Multimodal Correlative Preclinical Whole Body Imaging and Segmentation.

Segmentation of anatomical structures and particularly abdominal organs is a fundamental problem for quantitative image analysis in preclinical research. This paper presents a novel approach for whole body segmentation of small animals in a multimodal setting of MR, CT and optical imaging. The algorithm integrates multiple imaging sequences into a machine learning framework, which generates supervoxels by an efficient hierarchical agglomerative strategy and utilizes multiple SVM-kNN classifiers each constrained by a heatmap prior region to compose the segmentation. We demonstrate results showing segmentation of mice images into several structures including the heart, lungs, liver, kidneys, stomach, vena cava, bladder, tumor, and skeleton structures. Experimental validation on a large set of mice and organs, indicated that our system outperforms alternative state of the art approaches. The system proposed can be generalized to various tissues and imaging modalities to produce automatic atlas-free segmentation, thereby enabling a wide range of applications in preclinical studies of small animal imaging.

Ovarian dendritic cells act as a double-edged pro-ovulatory and anti-inflammatory sword.

Ovulation and inflammation share common attributes, including immune cell invasion into the ovary. The present study aims at deciphering the role of dendritic cells (DCs) in ovulation and corpus luteum formation. Using a CD11c-EYFP transgenic mouse model, ovarian transplantation experiments, and fluorescence-activated cell sorting analyses, we demonstrate that CD11c-positive, F4/80-negative cells, representing DCs, are recruited to the ovary under gonadotropin regulation. By conditional ablation of these cells in CD11c-DTR transgenic mice, we revealed that they are essential for expansion of the cumulus-oocyte complex, release of the ovum from the ovarian follicle, formation of a functional corpus luteum, and enhanced lymphangiogenesis. These experiments were complemented by allogeneic DC transplantation after conditional ablation of CD11c-positive cells that rescued ovulation. The pro-ovulatory effects of these cells were mediated by up-regulation of ovulation-essential genes. Interestingly, we detected a remarkable anti-inflammatory capacity of ovarian DCs, which seemingly serves to restrict the ovulatory-associated inflammation. In addition to discovering the role of DCs in ovulation, this study implies the extended capabilities of these cells, beyond their classic immunologic role, which is relevant also to other biological systems.

Ovarian carcinoma: quantitative biexponential MR imaging relaxometry reveals the dynamic recruitment of ferritin-expressing fibroblasts to the angiogenic rim of tumors.

PURPOSE: To quantitatively monitor the dynamic perivascular recruitment of ferritin heavy chain (FHC)-overexpressing fibroblasts to ovarian carcinoma xenografts by using R2 mapping and biexponential magnetic resonance (MR) relaxometry. MATERIALS AND METHODS: In vivo studies of female mice were approved by the institutional animal care and use committee. In vitro analysis included MR-based R2 relaxation measurements of monkey kidney cell line (CV1) fibroblasts that overexpress FHC, followed by inductively coupled plasma mass spectrometry to assess cellular iron content. For in vivo analysis, CV1-FHC fibroblasts were either mixed with fluorescent human ovarian carcinoma cells before subcutaneous implantation (coinjection) or injected intraperitoneally 4 days after the cancer cells were injected (remote recruitment). Dynamic changes in tumor R2 were used to derive CV1-FHC cell fraction in both models. In coinjection tumors, dynamic contrast material-enhanced MR imaging was used to measure tumor fractional blood volume. Whole-body fluorescence imaging and immunohistochemical staining were performed to validate MR results. One-way repeated measures analysis of variance was used to assess MR and fluorescence imaging results and tumor volume, and one-way analysis of variance was used to assess spectrometric results, fractional blood volume, and immunohistochemical evaluation. RESULTS: CV1-FHC fibroblasts (vs CV1 fibroblasts) showed enhanced iron uptake (1.8 mmol ± 0.5 × 10(-8) vs 0.9 mmol ± 0.5 × 10(-8); P < .05), retention (1.6 mmol ± 0.5 × 10(-8) vs 0.5 mmol ± 0.5 × 10(-8), P < .05), and cell density-dependent R2 contrast. R2 mapping in vivo revealed preferential recruitment of CV1-FHC cells to the tumor rim in both models. Measurement of fractional blood volume was similar in all tumors (2.6 AU ± 0.5 × 10(-3) for CV1, 2.3 AU ± 0.3 × 10(-3) for CV1-FHC, 2.9 ± 0.3 × 10(-3) for CV1-FHC-ferric citrate). Dynamic changes in CV1-FHC cell fraction determined at MR relaxometry in both models were confirmed at immunohistochemical analysis. CONCLUSION: FHC overexpression, when combined with R2 mapping and MR relaxometry, enabled in vivo detection of the dynamic recruitment of exogenously administered fibroblasts to the vasculature of solid tumors.

Removable nanocoatings for siRNA polyplexes.

To assist in overcoming the inherent instability of nucleic acid-containing polyplexes in physiological solutions, we have here set out to develop removable nanocoatings for modifying the surface of siRNA-based nanoparticles. Here, N-(2-hydroxypropyl)methacrylamide (HPMA) based copolymers containing carbonylthiazolidine-2-thione (TT) reactive groups in their side chains bound via disulfide spacers to the polymeric backbone were synthesized, and these copolymers were used to coat the surface of polyplexes formed by the self-assembly of anti-Luciferase siRNA with the polycations polyethylene imine (PEI) and poly(HPMA)-grafted poly(l-lysine) (GPL). The coating process was monitored by analyzing changes in the weight-average molecular weight (M(w)), the hydrodynamic radius (R(h)), and the zeta-potential (ζ) of the polyplexes, using both static (SLS) and dynamic (DLS) light scattering methods. The outlined methods resulted in the attachment of, on average, 28 polymer molecules to the surface of the polyplexes, forming a ∼5-nm-thick hydrophilic stealth coating. Initial efforts to develop RGD-targeted coated polyplexes are also described. Atomic force microscopy (AFM) showed an angular polyplex structure and confirmed the narrow size distribution of the coated nanoparticles. The stability of the polymer-coated and uncoated polyplexes was evaluated by gel electrophoresis and by turbidity measurements, and it was found that modifying the surface of the siRNA-containing polyplexes substantially improved their stability in physiological solutions. Using polymer-coated GPL-based polyplexes containing anti-Luciferase siRNA, we finally also obtained some initial proof-of-principle for time- and concentration-dependent target-specific gene silencing, suggesting that these systems hold significant potential for further in vitro and in vivo evaluation.

p53 status in stromal fibroblasts modulates tumor growth in an SDF1-dependent manner.

The p53 tumor suppressor exerts a variety of cell-autonomous effects that are aimed to thwart tumor development. In addition, however, there is growing evidence for cell nonautonomous tumor suppressor effects of p53. In the present study, we investigated the impact of stromal p53 on tumor growth. Specifically, we found that ablation of p53 in fibroblasts enabled them to promote more efficiently the growth of tumors initiated by PC3 prostate cancer-derived cells. This stimulatory effect was dependent on the increased expression of the chemokine SDF-1 in the p53-deficient fibroblasts. Notably, fibroblasts harboring mutant p53 protein were more effective than p53-null fibroblasts in promoting tumor growth. The presence of either p53-null or p53-mutant fibroblasts led also to a markedly elevated rate of metastatic spread of the PC3 tumors. These findings implicate p53 in a cell nonautonomous tumor suppressor role within stromal fibroblasts, through suppressing the production of tumor stimulatory factors by these cells. Moreover, expression of mutant p53 by tumor stroma fibroblasts might exert a gain of function effect, further accelerating tumor development.

Visualizing vascular permeability and lymphatic drainage using labeled serum albumin.

During the early stages of angiogenesis, following stimulation of endothelial cells by vascular endothelial growth factor (VEGF), the vascular wall is breached, allowing high molecular weight proteins to leak from the vessels to the interstitial space. This hallmark of angiogenesis results in deposition of a provisional matrix, elevation of the interstitial pressure and induction of interstitial convection. Albumin, the major plasma protein appears to be an innocent bystander that is significantly affected by these changes, and thus can be used as a biomarker for vascular permeability associated with angiogenesis. Traditionally, albumin leak in superficial organs was followed by colorimetry or morphometry with the use of albumin binding vital dyes. Over the last years, the introduction of tagged-albumin that can be detected by various imaging methods, such as magnetic resonance imaging and positron emission tomography, opened new possibilities for quantitative three dimension dynamic analysis of permeability in any organ. Using these tools it is now possible to follow not only vascular permeability, but also interstitial convection and lymphatic drain. Active uptake of tagged albumin by caveolae-mediated endocytosis opens the possibility for using labeled albumin for vital staining of cells and cell tracking. This approach was used for monitoring recruitment of perivascular stroma fibroblasts associated with tumor angiogenesis.

RGD-labeled USPIO inhibits adhesion and endocytotic activity of alpha v beta3-integrin-expressing glioma cells and only accumulates in the vascular tumor compartment.

PURPOSE: To investigate the biologic effect of arginine-glycine-aspartic acid (RGD)-labeled ultrasmall superparamagnetic iron oxide (USPIO) (referred to as RGD-USPIO) on human umbilical vein endothelial cells (HUVECs), ovarian carcinoma (MLS) cells, and glioblastoma (U87MG) cells and on U87MG xenografts in vivo. MATERIALS AND METHODS: All experiments were approved by the governmental review committee on animal care.USPIOs were coated with integrin-specific (RGD) or unspecific (arginine-alanine-aspartic acid [RAD]) peptides. USPIO uptake in HUVECs, MLS cells, and U87MG cells and in U87MG tumor xenografts was determined with T2 magnetic resonance (MR) relaxometry in 16 nude mice. Cells and tumors were characterized by using immunofluorescence microscopy. Trypan blue staining and lactate dehydrogenase assay were used to assess cytotoxicity. Statistical evaluation was performed by using a Mann-Whitney test or a linear mixed model with random intercept for the comparison of data from different experiments. Post hoc pairwise comparisons were adjusted according to a Tukey test. RESULTS: HUVECs and MLS cells internalized RGD-USPIOs significantly more than unspecific probes. Controversially, U87MG cells accumulated RGD-USPIOs to a lesser extent than USPIO. Furthermore, only in U87MG cells, free RGD and alpha(v)beta(3) integrin-blocking antibodies strongly reduced endocytosis of nonspecific USPIOs. This was accompanied by a loss of cadherin-dependent intercellular contacts, which could not be attributed to cell damage. In U87MG tumors, RGD-USPIO accumulated exclusively at the neovasculature but not within tumor cells. The vascular accumulation of RGD-USPIO caused significantly higher changes of the R2 relaxation rate of tumors than observed for USPIO. CONCLUSION: In glioma cells with unstable intercellular contacts, inhibition of alpha(v)beta(3) integrins by antibodies and RGD and RGD-USPIO disintegrated intercellular contacts and reduced endocytotic activity, illustrating the risk of inducing biologic effects by using molecular MR probes.

Transcriptional regulation of vascular endothelial growth factor C by oxidative and thermal stress is mediated by lens epithelium-derived growth factor/p75.

Vascular endothelial growth factor C (VEGF-C) plays a critical role in tumor lymphangiogenesis and lymph node metastasis. We report here that VEGF-C expression is regulated by microenvironmental stress including hyperthermia and oxidative stress. Furthermore, we show that this stress response is mediated by transcriptional activation mediated by lens epithelium-derived growth factor (LEDGF/p75). Ectopic expression of LEDGF/p75 in C6 rat glioma and in H1299 human non-small cell lung carcinoma induced VEGF-C expression in vitro, whereas in subcutaneous mouse tumor xenografts, LEDGF/p75 stimulated VEGF-C expression and augmented angiogenesis and lymphangiogenesis. Conversely, overexpression of a LEDGF/p75 native antisense or LEDGF/p75-targeted short interfering RNA downmodulated VEGF-C expression. LEDGF seemed to conferred this activity on binding to a conserved stress response element (STRE) located in the VEGF-C gene because mutating the STRE was sufficient for the suppression of basal and stress-induced activations of the VEGF-C promoter. Thus, the study reported here identified a role for LEDGF/p75 in stress-regulated transcriptional control of VEGF-C expression. These results provide a possible link for LEDGF/p75 in tumor lymphangiogenesis and cancer metastasis. Hence, our data suggest the LEDGF-VEGF-C axis as a putative biomarker for the detection of stress-induced lymphangiogenesis and LEDGF as a potential target for antimetastatic therapy.

In vivo imaging of the systemic recruitment of fibroblasts to the angiogenic rim of ovarian carcinoma tumors.

Tumor-associated stroma, in general, and tumor fibroblasts and myofibroblasts, in particular, play a role in tumor progression. We previously reported that myofibroblast infiltration into implanted ovarian carcinoma spheroids marked the exit of tumors from dormancy and that these cells contributed to vascular stabilization in ovarian tumors by expression of angiopoietin-1 and angiopoietin-2. Ex vivo labeling of fibroblasts with either magnetic resonance or optical probes rendered them detectable for in vivo imaging. Thus, magnetic resonance imaging (MRI) follow-up was feasible by biotin-bovine serum albumin-gadolinium diethylenetriaminepentaacetic acid or iron oxide particles, whereas labeling with near-IR and fluorescent vital stains enabled in vivo visualization by near-IR imaging and two-photon microscopy. Using this approach, we show here that prelabeled fibroblasts given i.p. to CD-1 nude mice can be followed in vivo by MRI and optical imaging over several days, revealing their extensive recruitment into the stroma of remote s.c. MLS human epithelial ovarian carcinoma tumors. Two-photon microscopy revealed the alignment of these invading fibroblasts in the outer rim of the tumor, colocalizing with the angiogenic neovasculature. Such angiogenic vessels remained confined to the stroma tracks within the tumor and did not penetrate the tumor nodules. These results provide dynamic evidence for the role of tumor fibroblasts in maintenance of functional tumor vasculature and offer means for image-guided targeting of these abundant stroma cells to the tumor as a possible mechanism for cellular cancer therapy.