Gridlock, a localized heritable vascular patterning defect in the zebrafish.

We are using the zebrafish, Danio rerio, to identify genes that generate and pattern the vertebrate vasculature. We have isolated a recessive mutation, gridlockm145 (grlm145) in which blood flow to the tail is impeded by a localized vascular defect. Using a novel microangiographic method, we show that the blockade is in the anterior trunk, where the paired lateral dorsal aortae normally merge to form the single midline aorta. Arterial-venous shunts and collateral vessels develop in most mutant embryos, bypassing the lesion and reconstituting caudal blood flow. The grl defect resembles coarctation of the aorta, a human congenital cardiovascular malformation of unknown aetiology, in the location of the lesion and its consequences and in the mutants' dependence on collateral vessels for survival.

A genetic screen for vascular mutants in zebrafish reveals dynamic roles for Vegf/Plcg1 signaling during artery development.

In this work we describe a forward genetic approach to identify mutations that affect blood vessel development in the zebrafish. By applying a haploid screening strategy in a transgenic background that allows direct visualization of blood vessels, it was possible to identify several classes of mutant vascular phenotypes. Subsequent characterization of mutant lines revealed that defects in Vascular endothelial growth factor (Vegf) signaling specifically affected artery development. Comparison of phenotypes associated with different mutations within a functional zebrafish Vegf receptor-2 ortholog (referred to as kdr-like, kdrl) revealed surprisingly varied effects on vascular development. In parallel, we identified an allelic series of mutations in phospholipase c gamma 1 (plcg1). Together with in vivo structure-function analysis, our results suggest a requirement for Plcg1 catalytic activity downstream of receptor tyrosine kinases. We further find that embryos lacking both maternal and zygotic plcg1 display more severe defects in artery differentiation but are otherwise similar to zygotic mutants. Finally, we demonstrate through mosaic analysis that plcg1 functions autonomously in endothelial cells. Together our genetic analyses suggest that Vegf/Plcg1 signaling acts at multiple time points and in different signaling contexts to mediate distinct aspects of artery development.

Zebrafish dracula encodes ferrochelatase and its mutation provides a model for erythropoietic protoporphyria.

Exposure to light precipitates the symptoms of several genetic disorders that affect both skin and internal organs. It is presumed that damage to non-cutaneous organs is initiated indirectly by light, but this is difficult to study in mammals. Zebrafish have an essentially transparent periderm for the first days of development. In a previous large-scale genetic screen we isolated a mutation, dracula (drc), which manifested as a light-dependent lysis of red blood cells [1]. We report here that protoporphyrin IX accumulates in the mutant embryos, suggesting a deficiency in the activity of ferrochelatase, the terminal enzyme in the pathway for heme biosynthesis. We find that homozygous drc(m248) mutant embryos have a G-->T transversion at a splice donor site in the ferrochelatase gene, creating a premature stop codon. The mutant phenotype, which shows light-dependent hemolysis and liver disease, is similar to that seen in humans with erythropoietic protoporphyria, a disorder of ferrochelatase.

Imaging blood vessels and lymphatic vessels in the zebrafish.

Blood vessels supply tissues and organs with oxygen, nutrients, cellular, and humoral factors, while lymphatic vessels regulate tissue fluid homeostasis, immune trafficking, and dietary fat absorption. Understanding the mechanisms of vascular morphogenesis has become a subject of intense clinical interest because of the close association of both types of vessels with pathogenesis of a broad spectrum of human diseases. The zebrafish provides a powerful animal model to study vascular morphogenesis because of their small, accessible, and transparent embryos. These unique features of zebrafish embryos permit sophisticated high-resolution live imaging of even deeply localized vessels during embryonic development and even in adult tissues. In this chapter, we summarize various methods for blood and lymphatic vessel imaging in zebrafish, including nonvital resin injection-based or dye injection-based vessel visualization, and alkaline phosphatase staining. We also provide protocols for vital imaging of vessels using microangiography or transgenic fluorescent reporter zebrafish lines.

Characterization of two frizzled8 homologues expressed in the embryonic shield and prechordal plate of zebrafish embryos.

We have isolated and characterized two complete cDNA clones, Zfz8a and Zfz8b, which encode zebrafish Frizzled (Fz) homologues. The predicted protein sequences, spanning 579 and 576 amino acid residues for ZFz8a and ZFz8b, respectively, were highly homologous (78%) to each other and contained an extracellular cysteine-rich domain and seven transmembrane domains that are well conserved in Fz receptor protein members. In comparison with other Fz family members, ZFz8a and ZFz8b showed the highest homology with mouse Fz8 (MFz8), sharing 84 and 76% amino acid identity, respectively. The presence of Zfz8a and Zfz8b transcripts was detected by in situ hybridization in zebrafish embryos from the 512 cell stage, and their appearance in the future dorsal region could be observed before embryos reached the 30% epiboly stage. At shield stage, Zfz8a transcripts were expressed in both epiblast and shield whereas expression of Zfz8b was only detected in the embryonic shield. During gastrula stages, both Zfz8a and Zfz8b transcripts were found in anterior dorsal regions of the involuting mesendoderm (future prechordal plate). By the 2- to 3-somite stage, expression of both Zfz8a and Zfz8b was restricted to the prechordal plate and prospective anterior neurectoderm, although expression of the Zfz8a gene was no longer present in the most anterior portion of the prechordal plate, the polster. In one-eyed pinhead mutant embryos, which lack prechordal plate, both Zfz8a and Zfz8b transcripts were reduced, confirming the prechordal plate specificity of Zfz8a and Zfz8b gene expression. These results provide an additional evidence supporting the role of Wnt signaling in organizer-mediated axial patterning.

Studying vascular development in the zebrafish.

The zebrafish, a genetically accessible vertebrate with an externally developing, optically clear embryo, is ideally suited for in vivo functional dissection of the embryonic development of the circulatory system. Here, we review the advantages of the zebrafish as a model system for studying vascular development, and describe genetic and experimental tools, methods and resources that have been developed to exploit these advantages. We also discuss briefly how some of these tools and methods can be brought to bear on problems of relevance to human health.

Fulfilling the social contract between medical schools and the public.

To gain a better understanding of the effects of medical schools related to transformations in medical practice, science, and public expectations, the Association of American Medical Colleges (AAMC) established the Advisory Panel on the Mission and Organization of Medical Schools (APMOMS) in 1994. Recognizing the privileges academic medicine enjoys as well as the power of and the strain on its special relationship with the American public, APMOMS formed the Working Group on Fulfilling the Social Contract. That group focused on the question: What are the roles and responsibilities involved in the social contract between medical schools and various interested communities and constituencies? This article reports the working group's findings. The group describes the historical and philosophical reasons supporting the concept of a social contract and asserts that medical schools have individual and collective social contracts with various subsets of the public, referred to as "stakeholders." Obligations derive implicitly from the generous public funding and other benefits medical school receive. Schools' primary obligation is to improve the nation's health. This obligation is carried out most directly by educating the next generation of physicians and biomedical scientists in a manner that instills appropriate professional attitudes, values, and skills. Group members identified 27 core stakeholders (e.g., government, patients, local residents, etc.) and outlined the expectations those stakeholders have of medical schools and the expectations medical schools have of those stakeholders. The group conducted a survey to test how leaders at medical schools responded to the notion of a social contract, to gather data on school leaders' perceptions of what groups they considered their schools' most important stakeholders, and to determine how likely it was that the schools' and the stakeholders expectations of each other were being met. Responses from 69 deans suggested that the survey provoked thinking about the broad issue of the social contract and stakeholders. Leaders on the same campuses disagreed about what groups were the most important stakeholders. Similarly, the responses revealed a lack of national consensus about the most important stakeholders, although certain groups were consistently included in the responses. The group concludes that medical school leaders should examine their assumptions and perspectives about their institutions' stakeholders and consider the interests of the stakeholders in activities such as strategic planning, policymaking, and program development.

How medical schools can maintain quality while adapting to resource constraints.

To gain a better understanding of the effects on medical schools of ongoing transformations in medical practice, science, and public expectations, the Association of American Medical Colleges (AAMC) formed the Advisory Panel on the Mission and Organization of Medical Schools (APMOMS) in 1994. Six working groups were appointed to address different issues of importance. This article is a report of the findings and recommendations of the Working Group on Adapting to Resource Constraints. That group was charged to consider how leaders in academic medicine can respond to the challenges of external forces and the anticipated diminishing of resources, and to focus on medical schools and how they can maintain quality while reengineering to effect needed changes. The group members developed their thinking within four categories: size of the academic enterprise; organizational models and their relationships to the clinical enterprise; faculty tenure and compensation; and partnerships with capital-intensive entities. Three recommendations for action, to which the APMOMS unanimously agreed, were made to the AAMC, which has already acted upon them in ways described in the article. The group also developed a series of "ideas for consideration," which represent a range of the members' perspectives. The working group did not seek (and probably could not have obtained) unanimous agreement on many of the issues that these ideas focus upon. The ideas are presented as a series of resolutions designed to stimulate discussion and foster better-informed planning.

Distinct genetic interactions between multiple Vegf receptors are required for development of different blood vessel types in zebrafish.

Recent evidence indicates a specific role for vascular endothelial growth factor a (Vegfa) during artery development in both zebrafish and mouse embryos, whereas less is known about signals that govern vein formation. In zebrafish, loss of vegfa blocks segmental artery formation and reduces artery-specific gene expression, whereas veins are largely unaffected. Here, we describe a mutation in the zebrafish vegf receptor-2 homolog, kdra, which eliminates its kinase activity and leads to specific defects in artery development. We further find that Flt4, a receptor for Vegfc, cooperates with Kdr during artery morphogenesis, but not differentiation. We also identify an additional zebrafish vegfr-2 ortholog, referred to as kdrb, which can partially compensate for loss of kdra but is dispensable for vascular development in wild-type embryos. Interestingly, we find that these Vegf receptors are also required for formation of veins but in distinct genetic interactions that differ from those required for artery development. Taken together, our results indicate that formation of arteries and veins in the embryo is governed in part by different Vegf receptor combinations and suggest a genetic mechanism for generating blood vessel diversity during vertebrate development.

Situation analysis and strategic development in a public hospital.

Hospitals operate in a highly competitive and regulated market. Efforts are being made to shrink the health care system. This paper describes the experience of a public hospital in New York, one of the most regulated states in the country. This hospital, once a local charity care institution, evolved into the only regional medical center in its area. This was accomplished through support from local and state governmental agencies and implementation of new programs that did not compete with local community hospitals. Referral patterns were established to attract patients requiring tertiary care, who previously would have gone outside the region for these services. The paper describes the strategic planning process used to achieve institutional goals and identifies principles necessary to complete a reversal in institutional image, mission, market, and role successfully.

What guides early embryonic blood vessel formation?

Survival of vertebrate embryos depends on their ability to assemble a correctly patterned, integrated network of blood vessels to supply oxygen and nutrients to developing tissues. The arrangement of larger caliber intraembryonic vessels, specification of arterial-venous identity, and proper placement of major branch points and arterial-venous connections are all precisely determined. A number of recent studies in both mammalian and nonmammalian vertebrate species, reviewed here, have now begun to reveal the major role played by genetically predetermined extrinsic cues in guiding the formation of early embryonic blood vessels and determining the global pattern of the vasculature.

Building the vertebrate vasculature: research is going swimmingly.

The vertebrate vasculature develops in remarkably similar fashion in all vertebrates. A cohort of unspecified mesodermal cells differentiates into primitive endothelial cells, which migrate to and occupy positions within the stereotypical blueprint of the primitive vasculature. Once in position, these cells coalesce and form cords, which lumenize and become ensheathed by supporting pericytes and smooth muscle cells. This primitive vascular network is extensively remodeled in some places, and expanded by sprouting in others. Various studies using the mouse, quail/chick, and frog have uncovered a number of signals that guide these complex processes but many gaps still exist in our understanding of the mechanisms by which the embryonic vasculature is built. Because many questions will require in vivo studies to be properly addressed, the zebrafish, with its unique accessibility to analysis by combined embryological, molecular, and genetic methods, should prove invaluable in identifying new molecules involved in blood vessel development and integrating pathways that influence embryonic blood vessel formation.

The vascular anatomy of the developing zebrafish: an atlas of embryonic and early larval development.

We have used confocal microangiography to examine and describe the vascular anatomy of the developing zebrafish, Danio rerio. This method and the profound optical clarity of zebrafish embryos make it possible to view the entire developing vasculature with unprecedented resolution. A staged series of three-dimensional images of the vascular system were collected beginning shortly after the onset of circulation at 1 day postfertilization through early- to midlarval stages at approximately 7 days postfertilization. Blood vessels in every region of the animal were imaged at each stage, and detailed "wiring patterns" were derived describing the interconnections between every major vessel. We present an overview of these data here in this paper and in an accompanying Web site "The interactive atlas of zebrafish vascular anatomy" online at (http://eclipse.nichd.nih.gov/nichd/lmg/redirect.html). We find a highly dynamic but also highly stereotypic pattern of vascular connections, with different sets of primitive embryonic vessels severing connections and rewiring in new configurations according to a reproducible plan. We also find that despite variation in the details of the vascular anatomy, the basic vascular plan of the developing zebrafish shows strong similarity to that of other vertebrates. This atlas will provide an invaluable foundation for future genetic and experimental studies of vascular development in the zebrafish.

Isolation and expression analysis of three zebrafish angiopoietin genes.

The Tie1 and Tie2 receptor tyrosine kinases and the Tie2 ligands, the angiopoietins, play critical roles in vertebrate vascular embryogenesis, helping to mediate the interaction between endothelial cells and the pericytes or vascular smooth muscle cells that envelop and support them. We have obtained full-length cDNA sequences for zebrafish orthologs of angiopoietin-1 (ang1), angiopoietin-2 (ang2), and angiopoietin-like-3 (angptl3). Ang1 is expressed in head ventral mesenchyme, in the ventromedial region of somites, in mesenchyme surrounding trunk axial vessels, and in the hypochord, a transient embryonic structure of endodermal origin that has been implicated in dorsal aorta assembly in both zebrafish and Xenopus. Ang2 is expressed in head and anterior trunk ventral mesenchyme and the developing pronephric glomeruli. Angptl3 is expressed in the yolk syncytial layer.

Vessel patterning in the embryo of the zebrafish: guidance by notochord.

We have cloned the zebrafish homolog of the receptor tyrosine kinase flk-1 to provide us with a tool to study normal vascular pattern formation in the developing zebrafish embryo and to compare it to mutants in which vascular pattern is perturbed. We find that during normal development the first angioblasts arise laterally in the mesoderm and then migrate medially to form the primordia of the large axial vessels, the dorsal aorta (axial artery) and the axial vein. Lumen formation occurs shortly before onset of circulation at 24 hr postfertilization. We examined the specification of vascular progenitors in the mutant cloche, which fails to form both vessels and blood. cloche lacks all flk-expressing cells and therefore appears to lack angioblasts. The axial vessels of the trunk form in close proximity to notochord and endoderm, which may provide cues for their formation. The dorsal aorta is normally just ventral to the notochord; the axial vein is just below the dorsal aorta and above the endoderm. floating head (flh) and no tail (ntl) mutants both have defects in the formation of notochord. Both are cell-autonomous lesions, flh abolishing notochord and ntl preventing its differentiation. In both mutants the dorsal aorta fails to form, while formation of the axial vein is less affected. Mosaic analysis of mutant embryos shows that transplanted wild-type cells can become notochord in mutant flh embryos. In these mosaic embryos flh cells expressing flk assemble at the midline, beneath the wild-type notochord, and form an aortic primordium. This suggests that signals from the notochord may guide angioblasts in the fashioning of the dorsal aorta. The notochord seems to be less important for the formation of the vein.

Cardiovascular development in the zebrafish. II. Endocardial progenitors are sequestered within the heart field.

We have examined the zebrafish embryo to ascertain the location of endocardial and myocardial progenitors prior to gastrulation, in an attempt to define the earliest stages of cardiac patterning. Currently there is uncertainty as to the spatial and lineage relationship of the progenitors for these two phenotypically distinct cell types that form the two concentric layers of the primitive heart tube. By single-cell injection and tracking, we distinguish a region in the early and midblastula which has the properties of a heart field, in that it defines a zone of cardiac progenitors within which there is a spatial gradient of propensity to generate heart cells, and which regulates, in the sense of adapting to the transplantation of pluripotential cells. This zone extends from the future ventral axis dorsally along the margin, with cardiogenic propensity tapering off laterally and dorsally. Myocardial progenitors are spread throughout this region, but endocardial precursors are restricted to the ventral marginal region. The cardiovascular progeny of the ventral cells include, in addition to endocardium and myocardium, cells in the endothelium and blood.

Cloche, an early acting zebrafish gene, is required by both the endothelial and hematopoietic lineages.

Endothelial and hematopoietic cells appear synchronously on the extra-embryonic membranes of amniotes in structures known as blood islands. This observation has led to the suggestion that these two ventral lineages share a common progenitor. Recently, we have shown in the zebrafish, Danio rerio, that a single cell in the ventral marginal zone of the early blastula can give rise to both endothelial and blood cells as well as to other mesodermal cells (Stainier, D. Y. R., Lee, R. K. and Fishman, M. C. (1993). Development 119, 31-40; Lee, R. K. K., Stainier, D. Y. R., Weinstein, B. M. and Fishman, M. C. (1994). Development 120, 3361-3366). Here we describe a zebrafish mutation, cloche, that affects both the endothelial and hematopoietic lineages at a very early stage. The endocardium, the endothelial lining of the heart, is missing in mutant embryos. This deletion is selective as evidenced by the presence of other endothelial cells, for example those lining the main vessels of the trunk. Early cardiac morphogenesis proceeds normally even in the absence of the endocardium. The myocardial cells form a tube that is demarcated into chambers, beats rhythmically, but exhibits a reduced contractility. This functional deficit is likely due to the absence of the endocardial cells, although it may be a direct effect of the mutation on the myocardial cells. Cell transplantation studies reveal that the endothelial defect, i.e. the endocardial deletion, is a cell-autonomous lesion, consistent with the possibility that cloche is part of a signal transduction pathway. In addition, the number of blood cells is greatly reduced in cloche mutants and the hematopoietic tissues show no expression of GATA-1 or GATA-2, two key hematopoietic transcription factors that are first expressed during early embryogenesis. These results show that cloche is involved in the genesis and early diversification of the endothelial and blood lineages, possibly by affecting a common progenitor cell population.

Notch signaling is required for arterial-venous differentiation during embryonic vascular development.

Recent evidence indicates that acquisition of artery or vein identity during vascular development is governed, in part, by genetic mechanisms. The artery-specific expression of a number of Notch signaling genes in mouse and zebrafish suggests that this pathway may play a role in arterial-venous cell fate determination during vascular development. We show that loss of Notch signaling in zebrafish embryos leads to molecular defects in arterial-venous differentiation, including loss of artery-specific markers and ectopic expression of venous markers within the dorsal aorta. Conversely, we find that ectopic activation of Notch signaling leads to repression of venous cell fate. Finally, embryos lacking Notch function exhibit defects in blood vessel formation similar to those associated with improper arterial-venous specification. Our results suggest that Notch signaling is required for the proper development of arterial and venous blood vessels, and that a major role of Notch signaling in blood vessels is to repress venous differentiation within developing arteries. Movies available on-line