Our objective is to understand the cellular and molecular mechanisms responsible for the specification, patterning, and differentiation of internal organs during development, specifically, how the elaborate network of blood vessels arises during vertebrate embryogenesis. Many of our insights into mechanisms of blood vessel formation have come from developmental studies. Given the potential shown by antiangiogenic therapies for combating cancer, blood vessels have also become a subject of great clinical interest in recent years.

The zebrafish, a small tropical freshwater fish, possesses a unique combination of features that make it particularly well suited for studying blood vessels. The fish is a genetically tractable vertebrate with a physically accessible, optically clear embryo. These features facilitate the study of vascular development by permitting observation of every vessel in a living animal and simple, rapid screening for even subtle vascular-specific mutants. Major aims of the laboratory include developing new experimental tools and resources to enhance the zebrafish as an experimental model for studying vascular embryogenesis; studying the molecular basis for arterial-venous differentiation; studying the role of neuronal guidance factors in vascular guidance and patterning; and performing forward genetic analysis of blood vessel development by using vascular-specific mutants.

Molecular Dissection of Arterial-Venous Development
Kamei, Lawson, Vogel
We have uncovered a molecular pathway regulating the acquisition of arterial-venous identity. Although the fundamental distinction between these two types of blood vessel has been appreciated for thousands of years, the fact that arterial and venous endothelial cells have distinct molecular and functional identities has become apparent only very recently, and the mechanisms responsible for establishing this identity have not yet been elucidated. We have now shown that sonic hedgehog (SHH), vascular endothelial growth factor (VEGF), and notch signaling act in series to determine arterial-venous identity (Lawson et al., 2001; Lawson et al., 2002). We were able to do this by using different combinations of drug treatments, mutants, morpholinos, and mRNA injections to activate or repress the activity of each of these signals in vivo, either alone or in combination. Our surprising findings regarding the novel role of VEGF in arterial specification have been confirmed by a number of very recent publications from other laboratories that have described a similar activity for VEGF in mice. We are now exploring additional factors that might participate in the pathway we have uncovered, including the zebrafish DeltaC gene and a vascular-specific phosopholipase C-gamma gene we recently uncovered in a mutagenic screen (see below).

Analysis of the Roles of Neuronal Guidance Factors in Vascular Patterning
Berk, Torres-Vazquez, in collaboration with Kuwada, Li, Chien, Epstein
We have recently become interested in the role that well-known neuronal guidance factors might play in vascular guidance and vascular patterning. In particular, we are studying the roles of the Slit-Robo and Semaphorin-Plexin ligand-receptor pairs in vascular patterning. Both of these ligand-receptor pairs are known to play important roles in axonal pathfinding, mostly via repulsive interactions between Slit or Semaphorin ligands and Robo- or Plexin- bearing neuronal growth cones. In collaboration with researchers studying related murine genes, we have uncovered novel Robo and Plexin receptors expressed in zebrafish blood vessels and have examined their functional roles. The Robo4 receptor is expressed in both neuronal and vascular tissues in zebrafish. Vascular-specific expression of a dominant-negative truncated form of either the murine or zebrafish Robo4 genes in zebrafish embryos (via transient transgenesis) results in premature vascular sprouting, consistent with the pro-migratory activity of this construct in murine cell culture (Park et al., 2002, submitted). The PlexinD gene is expressed in an entirely vascular-specific fashion in both mice and zebrafish. Targeted knock-down of zebrafish PlexinD using antisense morpholine oligonucleotide injection results in aberrant pathfinding of trunk intersegmental vessels and other vessels. Further analysis of the functional roles and activities of both of these receptors and their ligands is in progress. Our work suggests that the mechanisms of axonal and vascular guidance and patterning have a great deal in common, including specific molecular guidance factors.

Isolation and Analysis of Vascular-Specific Mutants
Berk, Diamond, Kamei, Lawson, Mugford, Pham, Roman, Torres-Vazquez; in collaboration with Dawid, Lechleider, Liu, Moon
Genetic dissection of vascular development and the molecular pathways that regulate it is an important ongoing goal of the group. We employ unbiased, forward genetic mutational screening approaches to identify and then perform phenotypic and molecular characterizations of zebrafish mutants that affect the formation of the developing vasculature. We have already positionally cloned the defective genes from a number of previously identified vascular patterning mutants. Resulting from defects in a zebrafish ortholog of the TGF-beta superfamily receptor acvrl1 (Roman et al., 2002), violet beauregarde mutants have abnormal cranial vascular patterning and circulation. Defects in human acvrl1 cause hereditary hemorrhagic telangiectasia type 2 (HHT), an inherited vascular disorder characterized by arterial-venous malformations with a high incidence of hemorrhage and stroke. Another mutation, kurzschluss, has defects in the posterior aortic arches caused by defects in smap1, a chaperonin expressed in the mesenchyme surrounding the arch vessels that may be involved in regulating myosin assembly. We have also performed new screens for vascular-specific mutants by using transgenic zebrafish expressing green fluorescent protein (GPF) in blood vessels. We identified 11 new vascular mutants in a pilot screen of haploid progeny of NEU-mutagenized animals that was performed with the Dawid and Liu laboratories (members of these laboratories independently screened for hematopoietic and neuronal defects). Two of the new mutants have defects in a phospholipase C-gamma (plcg) gene expressed specifically in the vasculature. PLCGs act downstream of receptor tyrosine kinase signaling, including that induced by vascular endothelial growth factor (VEGF). Like animals in which VEGF itself is directly targeted, plcg mutants display defects in arterial endothelial differentiation. Molecular and phenotypic characterization of other mutants from this pilot screen is in progress. We have recently initiated a larger-scale F3 diploid screen of NEU-mutagenized animals (to be carried out with members of the Dawid and Chitnis laboratories) to screen for mutants affecting both intersegmental vessel formation and later vascular patterning events that cannot be examined in haploid animals. Our experience suggests that these ongoing mutant screenings should continue to yield a rich harvest of novel vascular-specific mutants and bring to light new pathways regulating the specification, differentiation, and patterning of the developing vertebrate vasculature.