New strategies for inhibiting tumor angiogenesis

Research project

One of the most innovative aspects of anti-cancer therapies concerns the possibility of inhibiting tumor growth by blocking blood supply. If starved, a tumor will not grow, but, on the contrary, will shrink and become more susceptible to chemotherapy and radiotherapy. Our research is aimed at understanding the mechanisms that regulate the formation of the vascular system in tumors in order to induce their regression. A large body of information derives from the study of embryonic development, since embryo and tumor vascularization are often regulated by the same molecular mechanisms.

• - Project 1: The role of adhesion proteins at endothelial junctions in angiogenesis
• - Project 2: Beta-catenin's role in vascular development
• - Project 3: Junctional proteins and leukocyte extravasation
• - Project 4: Definition of new genes responsible for endothelial cell differentiation
• - Project 5: Vascular development studies using zebrafish
• - Project 6: The organization of intercellular junctions in lymphatic endothelial cells

Project 1

fig.1: Dejana E,Nat Rev Mol Cell Biol. 2004 (Project 1) [+zoom]

During our research, we identified two novel proteins expressed in endothelial cells that are responsible for cell-to-cell adhesion. They are called vascular endothelial cadherin (VE-cadherin) and junctional adhesion molecule (JAM, see Project 3) (fig. 1). Inactivation of the gene encoding VE-cadherin, or inhibition of its adhesive properties by specific antibodies, inhibits new vessel formation and tumor progression in experimental systems. We observed that VE-cadherin, besides its adhesive properties, acts upon the regulation of cell growth and death and influences the ability of cells to form tubular structures. We identified a series of transmembrane and cytoplasmic partners that combine with VE-cadherin to transfer intracellular signals and affect the expression of many endothelial genes. Several of the proteins encoded by these genes regulate endothelial cell growth and apoptosis and constitute potential targets for drug interventions geared towards the inhibition of angiogenesis. Our future project deals with the study of the mechanism of action of these proteins and their reciprocal relationship.

ˆ top

Project 2

fig.2a: Wt embryo (Project 2) [+zoom]

fig.2b: KO embryo (Project 2) [+zoom]

Beta-catenin binds the cytoplasmic tail of membrane-bound cadherins, linking them to the actin cytoskeleton. When beta-catenin is released from junctions and stabilized in the cytosol, it can then translocate to the nucleus and control gene expression. To investigate the role of beta-catenin in mouse vasculature, we developed transgenic lines where the beta-catenin gene was inactivated only in endothelial cells. The defects are essentially of vascular origin, exhibiting frequent hemorrhages and alterations in head and yolk sac microcirculation. The embryos also show a defect in heart cushion formation, specifically a lack of septum development, which accounts for their lethality at mid-gestation. (fig. 2a, fig. 2b) Two other models are also under study: a) a gain-of-function model for beta-catenin transcriptional activity. These mice express a stabilized form of beta-catenin with high transcriptional activity; and b) an inducible, tissue-specific loss-of-function model where at a desired time the beta-catenin gene can be inactivated only in endothelial cells.

ˆ top

Project 3

fig 3: Altered neutrophil polarization in absence of JAM-A** (Project 3)
[+zoom]

Inflammatory reactions in the tumor stroma play an important role in tumor progression. In particular, infiltration of neutrophils and monocyte/macrophages can promote solid tumor progression by producing growth factors for tumor cells, lytic enzymes which may facilitate invasion and dissemination, and by inducing and sustaining angiogenesis.

fig 4: Altered neutrophil polarization in absence of JAM-A** (Project 3) [+zoom]

While the molecular mechanisms regulating leukocyte adhesion to inflamed endothelial cells have been elucidated to a large extent, we have only a partial picture of how leukocytes open endothelial junctions and move into the underlying tissues in a polarized fashion. We have identified an adhesion molecule called junctional adhesion molecule-A (JAM-A) which is expressed in endothelial cells, epithelial cells, leukocytes, and platelets. This project focuses on the JAM-A mechanism of action on leukocyte diapedesis and infiltration into tumor stroma. Besides cell culture systems, we have developed genetically-modified mice in which the JAM-A gene has been inactivated.
Previous data indicate that JAM-A is required for correct leukocyte adhesion and polarization. In its absence, leukocytes remain attached to matrix proteins and show an altered morphology with long tails and actin mislocalization. As a consequence, leukocyte migration is severely affected. Time-lapse video microscopy shows that JAM-A -/- leukocytes move less effectively towards a chemotactic stimulus than wild-type leukocytes: they are completely unable to retract their uropod, which remains attached to the substratum and elongated, even when they move. We are now investigating JAM-A's mechanism of action in leukocyte trafficking. Experimental tumors have been induced in JAM-A null mice. These studies should tell us whether JAM-A may constitute a valuable target for inhibiting leukocyte infiltration in tumor stroma and whether this action is beneficial in limiting tumor growth.
(fig 3, fig 4: Altered neutrophil polarization in absence of JAM-A** - Immunofluorescence analysis of actin distribution in bone-marrow derived neutrophils from JAM-A -/- and +/+ mice. Cells were seeded onto fibronectin-coated coverslips and incubated with the chemotactic peptide WKYMVm. JAM-A -/- neutrophils show a more elongated morphology and long tails (Figure 3) compare to controls (Figure 4). JAM-A +/+ cells show actin clustering at the leading edge (Figure 4, arrow-heads) and a short uropod (arrow); in contrast to wild types, in JAM-A -/- neutrophils actin was absent or poorly organized at the leading edge (Figure 3, arrow-heads), while it was concentrated at the opposite site, at the basis of the uropod (arrows). Scale bars: 12.5 mm.)

ˆ top

Project 4

The mechanisms that regulate the development of the vascular system in mouse embryos are comparable with those that regulate angiogenesis in tumors. We use mouse embryonic stem cells (ES), induced to differentiate into blood and lymphatic endothelial cells in vitro, as a model to study the expression of genes important in vascular development. (fig. 5)
Affymetrix oligonucleotide arrays are used to define sets of known or unknown genes which may mediate ES cell differentiation towards the two endothelial cell lineages. The selected genes are studied by in situ hybridization of mouse embryos and tumor specimens and by ex vivo models of angiogenesis such as mouse corneal angiogenesis (fig. 6) and allantois culture. (fig. 7)

fig.5 (Project 4) [+zoom]

fig.6: Corneal angiogenesis assay (Project 4) [+zoom]

fig.7: Culture of murine allantoic explants (Project 4) [+zoom]

ˆ top

Project 5

fig.8: The vascular anatomy of the developing zebrafish* (Project 5) [+zoom]

Zebrafish (Danio rerio) has become an ideal model system for in vivo studies of vascular development. Zebrafish embryos develop externally, and, due to their transparency, their blood flow can be directly observed under the microscope. Moreover, several transgenic lines expressing fluorescent protein (e.g., EGFP) in the vasculature are available, making it possible to analyze in detail the formation of the angiogenic network over time. Other methodological approaches, such as the microangiography technique, allow a fine observation of the circulatory system and contribute to a better analysis of vascular function (ref. 1, fig. 8). We have analyzed an Affymetrix gene expression profile of mouse embryonic stem cells induced to differentiate into endothelial cells (see Project 4), as well as the transcriptome of endothelial cells with or without VE-cadherin. Through the analysis of these data generated in vitro, many genes with differential expression have been identified. We are validating these Affymetrix results using the zebrafish model. The project comprises the identification of zebrafish orthologue genes and the analysis of the expression pattern by in situ hybridization. We are studying the loss-of-function of genes of interest by injecting morpholino antisense oligonucleotides into the transgenic line Tg (fli1:EGFP)y1, which expresses EGFP in endothelial cells (ref. 2). In the long term, it may lead to the discovery of new targets for anti-angiogenic therapy.

*Isogai, S., Horiguchi, M., and Weinstein, B. M. The vascular anatomy of the developing zebrafish*: an atlas of embryonic and early larval development. Developmental Biology, Volume 230, pp. 278-301.

ˆ top

Project 6

fig.9: Immunofluorescence on mouse intestinal frozen sections, showing the partial co-localization of a new anti lynphatic endothelium Antibody with the known lymphatic marker LYVE-1. (Project 6) [+zoom]

The lymphatic vasculature complements the blood vasculature by transporting tissue fluid, proteins, and cells, including immune cells, from body tissues back into the circulatory system. Until recently, however, the lack of specific molecular markers has limited the characterization of the interactions between lymphatic endothelial cells. Initial genetic profiling has shown differences in junctional marker expression between lymphatic and blood endothelial cells. This prompted us to apply a differential immunization protocol to develop antibodies specifically directed towards lymphatic junctional molecules (fig. 9). This junctional identi-kit suggests that lymphatic junctional organization may be both qualitatively different and structurally unique from that of blood endothelial cells. The role of novel and known lymphatic junctional markers will be further assessed to determine how the lymphatic program induces their expression and how these markers are involved in lymphatic vasculature function.

ˆ top