development: and molecular - CiteSeerX

14 downloads 0 Views 2MB Size Report
cell, inhibits. EC proliferation, induces mural cell differentiation, and stimulates matrix deposition. Further mural cell recruitment and proliferation, accompanied.
development:

Vascular

LAURENCE Program

Laboratory

ABSTRACT

BECK, JR. AND PATRICIA

in Biological

and Biomedical

for Surgical

The vascular of vasculogenesis

Research,

system

Children’s

and the Department Hospital,

forms through

a

be orchestrated in a complex sequence of steps that lead to the development of the adult vascular system. Thus, communication between the forming vasculature and the tissue parenchyma, as well as interactions among cells of the vascular wall, all appear to influence vascular development and growth. -Beck, L., Jr., D’Amore, P. A. Vascular development: cellular J. 11, 365-373

and molecular (1997)

Key Words: endothelial receptor

THE

REGULATION

cell

.

OF BLOOD

regulation.

tissue factor

VESSEL

.

FASEB

PDGF

GROWTH

.

TGF-

is currently

an exciting and important area in the study of developmental and vascular biology, and in investigations of a variety of pathologies, including diabetic retinopathy and tumor growth. The vasculature forms via two processes: vasculogenesis and angiogenesis. Although angiogenesis appears to be the sole means of neovascularization during processes such as wound healing, both vasculogenesis and angiogenesis are involved in the establishment of the vasculature

of the developing

embryo.

In vasculogenesis,

en-

dothelial cell (EC)2 precursors called angioblasts associate to form early vessel tubes. Tissues that are vascularized by this process are generally of endodermal origin and include the lung, pancreas, and spleen as well as the heart tube and dorsal aorta (1). In the process of angiogenesis,

0892-6638/97/0011

-0365/$01

.50 © FASEB

regulation

A. D’AMORE’

Sciences

and angiogenesis. In vasculogenesis, vessels form de novo via the assembly of endotheial precursors called angioblasts, whereas in angiogenesis new vessels arise by migration and proliferation of endothelial cells from preexisting vessels. Although the two processes are distinct in some respects, recent evidence suggests that they share a number of regulatory mechanisms. The identffication of a number of defined growth factors, observations of genetically manipulated mice, and the recognition of the importance of cellcell interactions have greatly expanded our understanding of the regulation of vascularization. The paracrine actions of a variety of polypeptide growth factors, including platelet-derived growth factor, vascular endothelial growth factor, transforming growth factor-beta, and the angiopoietins, appear to combination

and molecular

cellular

Boston,

of Pathology,

Massachusetts

Harvard

02115,

Medical

School,

USA

small blood vesselsform by budding and sprouting from larger, extant vessels. Tissues of ectodermal and mesodermal derivation such as the kidney and brain are thought to he vascularized primarily via angiogenesis. There is some controversy regarding vascularization of the retina. Since the retina is an extension of the brain, it was expected that, like the brain, it would be vascularized by angiogenesis. However, using immunohistochemical methods to visualize the vasculature, the presence of angioblast-like cells was documented in the developing dog retina (2). Common toboth angiogenesisand vasculogenesis is the process of remodeling. Remodeling is a poorly understood event that includes growth of new vessels and the regression of others. Vascular remodeling involves changes in lumen

diameter

and

vessel

wall

thickness

to suit

the local

tissue needs. Thus, the vasculature often begins as a plexus of primitive capillary tubes (Fig. 1) that is subsequently modified to generate the more complex vascular network of the adult (Fig. 2). The process of vascularization, whether by angiogenesis or by vasculogenesis, can be examined in a number of contexts. Although there are several fundamental differences between angiogenesis and vasculogenesis, these two processes also share a number of regulatory features. Many excellent

reviews

have

been

written

approaching

the sub-

ject from a variety of perspectives, including tumor biology and wound healing (e.g., see ref 3). The earliest context in which this subject can be examined is during embryonic development, and it is in this context that the molecular regulation of angiogenesis and vasculogenesis will be examined in this review. Until recently, information regarding the regulation of angiogenesis and vasculogenesis was largely descriptive and came from classic embryologic studies. For instance, chick-quail grafting systems were used to demonstrate that many mesodermal tissues harbor cells capable of differentiating into EC, whereas ectoderm-derived tissues such as the neural tube or neural crest do not (4). However, specific

only

molecules

recently Correspondence:

pital, 2

been

involved

in vascular

development

have

identified. Laboratory

for Surgical

Research,

300 Longwood Ave., Boston, MA 02115, USA. Abbreviations: EC, endothelial cells; PDGF.

Children’s

Hos-

platelet-derived

VEGF, vascular endothelial growth factor; TGF-, transforming growth factor-beta; bFGF. basic fibroblast growth factor; E, embryonic day; SMC, smooth muscle cells; IF, tissue factor; PAl-i, plasminogen activator inhibitor-i; HHT. hereditary hemorrhagic telangiectasia. growth

factor;

365

inductive

processes

that

allow

the

ventral

mesodermal

cells of the embryo to become hemangioblasts. When dissociated quail epiblasts were cultured in vitro, the cells would aggregate into blood island-likestructures and give rise to Flk-1-positive EC only in the presence of bFGF (and to a lesser extent, acidic FGF) (6). However, mice deficient in bFGF exhibit no overt developmental abnormalities (C. Basilico and S. Ortega, personal communication), but since maternal bFGF mRNA may be present in the early embryo and there is possible redundancy among the various FGF family members, a role for FGFs in the early stages of vascular development cannot be ruled out entirely. Figure 1. Visualization of the capillary plexus in a neonatal mouse retina. The vascular network of a newborn (P0) mouse retinawas visualized by perfusion of fixative containing 10% India ink. Note the uniform character of the primordial vascular tubes. (Micrograph from S. E. Connolly and P. A. D’Amore.)

In the past several defined

molecules,

years,

identification

in combination

with

netically manipulate the mouse vide insight into the molecular vessel

assembly.

involved leading

Although

in vessel

ability

of

to ge-

genome, has begun to promechanisms that regulate

our knowledge

formation

to the assembly

of a number the

of the molecules

and the sequence

of mature,

quiescent

of events

blood

vessels

is far from complete, observations do permit some speculation regarding how the various cells and factors contribute to the establishment of the vascular system. In the following discussion, we will review the available data that implicate particular molecules in the processes of vessel formation and maturation. Where possible, we will compare observations made both in vivo and in vitro on the same molecule in order to determine whether these data are consistentand to provide further insight into the rolesthe molecule may play in vessel formation. Factors that have been implicated in the regulation of vascular cells

include

platelet-derived

cular

endothelial

growth

factor

growth

growth

beta-i

factor

(TGF-1),

tor (bFGF), the Tie-i poietins, aJ33 integrin,

factor

(PDGF),

(VEGF),

basic

fibroblast

VASCULAR ENDOTHELIAL FACTOR (VEGF)

GROWTH

Of the many peptide growth factors known to act on the vasculature, the most compelling evidence is for VEGF as a key regulator of vasculogenesis and angiogenesis (for review,see ref 7). VEGF, particularly the VEGF16S isoform (8), is a potent

mitogen

and

chemoattractant

for EC. The

receptors for VEGF are the tyrosine kinases Flt-1 (VEGFR-i) and Flk-1/KDR (VEGFR-2), which are expressed primarily by the endothelium. VEGF is secreted by many mesenchymal and stromal cells, and the receptorligand pair shows a coordinate pattern of expression during development. In most developing organs, the forming vasculature stains positively for both VEGF receptors while the stromal cells immediately surrounding the nascent vessels express VEGF (9, 10). Embryos compromised in the expression of VEGF, as well as embryos to which exogenous VEGF has been added, show dramatic abnormalities in vascular patterning. Knockout mice have been created for VEGF as well as for both

of its receptors.

Flk-1-deficient

mice

die

in

vas-

transforming growth

fac-

and Tie-2 receptors, the angioand coagulation factors. Although

the interrelationships among the various factors are not entirely clear at this time, we will attempt to synthesize the observations

events that (Fig. 3).

BASIC (bFGF) Despite

to generate

lead

FIBROBLAST

numerous

of angiogenesis in review, see ref 5), plays a major role There are data to

366

Vol.11

a model

to the formation

reports

that may describe

of a new blood

GROWTH

that

bFGF

the

vessel

FACTOR

is a potent

stimulator

vitro and in several in vivo systems (for there is no convincing evidence that it in the development of the vasculature. suggest that bFGF may function in the

April 1997

Figure 2. Visualizationof the vasculatureof the adult mouse. Vessels were visualized by indirect immunofluorescent staining flat mounts of adult mouse retinas with antisera against type IV collagen. The primordial network has undergone remodeling and includes larger vessels as well as the arborizing pattern of microvessels. (Micrograph from S. E. Connoily and P. A. D’Amore.)

The FASEB Journal

BECK AND

D’AMORE

ANGIOGENESIS

VASCULOGENESIS

B)

A)

ANGIOBLASTS

MESODERM

VESSEL SPROUT VEGF



‘*-

bFGF P1k-I

C)

PDGE-BB

0#{149} Mesenchymal

DIFFERENTIATED VESSEL

Cell

Figure 3. Model of vascular development. New vessels form by angiogenesis (A) or vasculogenesis (B). The nascent vessel is seen in cross section in panel C, where EC-derived PDGF signals the recruitment and proliferation of nearby mesenchymal cells. In panel D, TGF-, activated by contact between EC and the mesenchymal cell, inhibits EC proliferation, induces mural cell differentiation, and stimulates matrix deposition. Further mural cell recruitment and proliferation, accompanied by establishment of a quiescent endothelium, leads to the fully differentiated vessel (E).For capillary formation, the vessel is surrounded by a single layer of pericytes, but the sequence of events involved in itsassembly is likely to be similar to that for larger vessels.

utero between embryonic day (E) 8.5 and E9.5, and are remarkable for their nearly complete lack of vascular structures (11). The networks of vessels in the yolk sac, the dorsal aortae, and other smaller vessels are all absent, confirming the absolute requirement for Flk-1 in the growth and/or establishment of the endothelial lineage. Flt-1 null mice also display embryonic lethality between E8.5 and E9.5 (12). These embryos have a grossly abnormal vasculature, but the defect appears to lie more in the organization of EC into tubelike structures than in the differentiation of the endothelium itself. Instead of surrounding a core of hematopoietic precursors as a monolayer, the EC of the yolk sac blood islands in Flk-1 null mice are disorganized and are present as aggregates and sheets within the lumen. Embryonic vessels such as the dorsal aortae, which arise by vasculogenesis, and the small vessels in the head mesenchyme, which form via angiogenesis, display a similar phenotype, with internally located EC. Animals deficient in VEGF have been generated by two independent laboratories and, somewhat surprisingly, reveal embryonic lethality in the heterozygous state (13, 14). The heterozygous mutant embryos die in utero between Eli and E12, and are characterized by defects in most of the early processes in vessel formation. These results suggest that VEGF exerts a dose-dependent effect during formation of the vasculature. Strict VEGF dose dependence is also suggested by studies that involve the overexpression of VEGF in developing avian embryos. Microinjection of VEGF16S into quail embryos, just before formation of the dorsal aortae and associated vascular plexus, led to the

VASCULAR

DEVELOPMENT

formation of a hyperfused network of vessels and the development of vessels in areas that are usually avascular (15). In another series of experiments, introduction of VEGF-expressing retroviruses into the developing chick limb bud resulted in local increases in vascular density, with no other phenotypic abnormalities (16). In light of the apparent sensitivity of the developing vascular system to VEGF (as reflected by the lethality of the embryos heterozygous for VEGF), the different outcomes in these two studies are likely due to local differences in VEGF concentrations. A series of studies has provided strong evidence in support of a role for hypoxic regulation of VEGF in retinal vascularization. Astrocytes that migrate into developing feline or murine retinas express elevated levels of VEGF, presumably in response to local “physiologic” hypoxia created by the rapidly differentiating neural elements (17). Once the retinal tissue is vascularized and the oxygen demand is met, the astrocyte production of VEGF is reduced. These observations have led to a viable explanation for an ocular pathology called retinopathy of prematurity in which excess oxygen leads to “vaso-obliteration” in the developing retinas of premature infants (18). It is speculated that the therapeutic oxygen administered to the premature infants leads to a down-regulation of VEGF, and thus interferes with the normal vascularization of the retina. Once the child is returned to room air, the retina experiences “relative hypoxia” and VEGF synthesis is induced, as has been shown in ischemic models (19), leading to the pathologic retinal neovascularization that characterizes this disease state. The authors demonstrate that

367

exogenous

administration

of VEGF

is sufficient

to “main-

suggestion

tain” developing vessels, suggesting that the continuous presence of VEGF is required during a particular (but unknown) window during vascular development.

eration

that factor

PDGF-BB

may

for EC during

act as an autocrine

prolif-

angiogenesis.

The phenotypes of PDGF- and PDGF receptor-deficient mice have provided much information about the role of this growth factor system during development. Both the PDGF-B and the PDGF receptor null mice reveal similar phenotypes (23, 24). Perhaps the most striking finding is an apparent absence of mesangial cells in the kidney and an aneurysmal, endothelium-lined sac that replaces the usual capillary tuft in the renal glomerulus. As mesangial cellsbear many structuraland functionalsimilarities to the capillarypenicytes(25),it is perhaps not surprising that a recent report indicates that these animals also lack pericytes throughout their vascular beds (C. Betsholtz, personal communication, and P. Lindahl, unpublished results). In several tissues of Ei4.5 to E17.5 PDGF-B -Imice, capillaries appeared to lack PDGF receptor-positive pericytes, although PDGF receptor-positive mural cells could be found surrounding arteries. The authors suggest that due to the lack of PDGF-BB, the mural cells adjacent to the arteriescannot be recruitedby the EC of the capillary sprouts, leading to the observed lack of pencytes. The absence of pericytes in the brains of PDGF-B -Imice is associated with capillary aneurysms; both the PDGF-B- and PDGF receptor-deficient mice die pertnatally, often with subcutaneous hemorrhages. Both ligand- and receptor-deficient mice also display some degree of vascular dilation, although the number of smooth

1

PLATELET-DERIVED (PDGF) The platelet-derived modimers

GROWTH

growth

(PDGF-AA

factor

FACTOR

family

consists

or -BB) or heterodimers

of ho-

(PDGF-AB)

made from the pairwiseassembly ofthe two related PDGF chains, A and B. The three PDGF isoforms often have differenteffectson a given cell type, primarily due to the cells’ repertoire of PDGF receptors.The receptors,in a manner similar to their ligands, consist of dimers of the u and 13subtypes of the PDGF tyrosine kinase receptor. The a receptor can bind both PDGF chains, whereas the

13

receptor

is selective

for the PDGF-B

chain.

Therefore,

aa

receptor dimers can bind all three isoforms of PDGF (AA, AB, and BB), #{128}43 heterodimers can bind AB or BB, and the 313receptor dimer binds only PDGF-BB. PDGF has been proposed to function during two aspects of vessel formation: the paracrine recruitment of mural cell precursorsto the vesselwall and the autocrine stimulation of EC. The expression patterns of PDGF and its receptors strongly support a role for endothelium-derived PDGF in the recruitment and proliferation of nearby cells, although the

evidence

for an autocrine

role

is more

controversial.

Much of the work demonstrating dual expression of PDGF and its receptors in EC has used in situ hybridization to investigate

capillary

mRNA

expression,

and there

are little

data demonstrating the presence of the protein. For example, in the developing human placenta, capillaries have been shown to express mRNA for PDGF-B as well as the PDGF

13receptor

(20).

In addition,

the light

micrographs

in this report failed to provide resolution of the smaller vessels that was sufficient to determine whether the PDGF

13receptor-positive cytes.

though

of the

In

cells placental

larger

still positive

13receptor;

13receptor

were

were EC or the associated vessels, the endothelium,

for PDGF-B,

has negligible

the only cells that expressed in the adjacent

mesenchyme.

penal-

expression

the PDGF In situ

hy-

bridization analysis of the mouse embryo has revealed that small vessels express the PDGF receptor in the endo-

13

thelium

or surrounding

vessels,

the

EC have

mesenchyme, little

whereas

or no receptor

in larger

(21).

There have also been mixed reports about the presence of the PDGF receptor on EC in vitro, although one study (22) raises an interesting point about the potential in-

13

ducibility of the receptor during vine aortic EC that spontaneously

angiogenesis. Using boundergo “angiogenesis”

or cord formation in vitro, EC growing shown to be negative for the PDGF

in a monolayer

13receptor,

were

while the cells that form cords beneath the monolayer were positive for receptor expression. These receptor-positive cells were also distinguished by a high mitotic index, supporting the

368

Vol. 11

April 1997

13

13

13

muscle

cells

(SMC)

associated

with

the larger

vessels

ap-

pears to be normal. The observation that defects in the PDGF-B- or receptor-deficient mice are confined to the microvasculature suggests that PDGF-A may be able to function in the recruitment and/or proliferation of SMC, but not penicytes. Our earlier observation that PDGF-AA is a better mitogen for SMC than for pericytes supports this possibility (26). Furthermore, findings in the Patch mouse, whose PDGF a receptor has been deleted as a result of a spontaneous mutation, also provide evidence that PDGF-A is crucial for the development of the cardiovascular system (27). Histological examination of these homozygous null animals reveals fewer layers of SMC surrounding a normal vascular endothelium, as well as a thinned myocardium. The authors comment that there are also infrequent myoendothelial contacts in the large vessels. The presence of pericytes was not assessed in these mice. Studies from our laboratory using an in vitro model of vessel development have revealed that PDGF-BB secreted by EC can chemotactically recruit and stimulate the proliferation of SMC (J. Gabriels and P. A. D’Amore, unpublished data) and other mesenchymal cells (K. K. Hirschi, et al., unpublished results and ref 28). As has been reported for SMC (29), PDGF-BB down-regulates the expression of a smooth muscle (SM) actin in 1OT1/2 cells induced with TGF-13 (28). This suggests that once mes-

13

enchymal

cells

expression ferentiate

must be suppressed into mural cells.

The FASEB Journal

are recruited

to the forming

vessel,

so that the cells

BECK AND

PDGF

can dif-

D’AMORE

TGF-13 -I-

MURAL CELL ORIGIN PDGF-BB has been implicated in the recruitment of mural cell precursors to the vessel wall, but what is the origin of these cells? Two candidate cell populations have been suggested: mesenchymal cells present in the tissue being vascularized and SMC of the parent vessel. During the process of vasculogenesis (e.g., in the developing dorsal aorta), local mesenchymal cells that appear to associate with the forming vessel are induced to differentiate toward a SMC lineage (30). In the developing quail embryo, it has been shown that mesodermally derived cells in the vicinity of the dorsal aorta become incorporated into the vessel wall and express aSM-actin and a novel SMC-specific marker. The differentiation process begins at the ventral surface and continues dorsally; the appearance of smooth muscle markers occurs in a radial pattern, commencing adjacent to the endothelium and proceeding outward. Different segments of the vasculature appear to recruit mural cells from distinct embryonic lineages. The medial SMC of the arteries proximal to the heart are derived from cardiac neural crest cells, and are therefore neuroectodermal in origin (31). In contrast, SMC in more distal segments of the aorta are mesodermally derived. Vessels arising via angiogenesis are likely to recruit local mesenchymal cells in a similar fashion. However, there is also evidence that newly forming vessels may derive their mural cells from SMC of the parent vessel. Using an in vitro model of angiogenesis in which new vessels sprout from rings of rat aorta, penicytes were shown to be recruited from subintimal SMC (32). In studies focused on angiogenesis in the developing mouse brain, PDGF receptor-positive pericytes appeared to migrate from the existing medial layer to associate with the forming vessels of PDGF-B +1+ mice, a process not evident in PDGF-B -Imice (C. Betsholtz, personal communication, and P. Lindahl, unpublished results). Furthermore, PDGF receptor-positive cells were not detected throughout the brain parenchyma, as would be expected if the vessels were able to recruit local surrounding mesenchymal cells.

13

13

TRANSFORMING

GROWTH

FACTOR-BETA

(TGF-13) The TGF-f3s are a large family of homodimeric peptides widely implicated in the regulation of cellular growth and differentiation(forreview, see ref 33). TGF-13 has been shown to influence vascular cell growth and function, including inhibition of EC and SMC growth (26, 34, 35) and migration (36), as well as alterations in EC matrix accumulation (37). In light of these observations, it was surprising that initial reports of TGF-131-deficient mice, which die at about 3 wk of age of a multifocal inflammatory disorder, did not note any effects on the development or function of the vasculature (38). Subsequent analysis, however, revealed that nearly 50% of

VASCULAR

DEVELOPMENT

and 25% of TGF-13 +1- mouse embryos die in utero and have defective hematopoiesis and vasculogenesis (39). Analysis of structures at E8.5 revealed no morphological defects, but by E9.5-E 10.5 there were obvious abnormalities of the extraembryonic yolk sac, ranging from delayed vasculogenesis to the total absence of vessels. There were dramatic defects in yolk sac vessel development, including incomplete contacts between endothelial and mesothelial layers, leading to distended capillary structures. TGF-13 has many effects, including regulation of integrin expression and matrix accumulation, which may be the basis for this incomplete interaction. There was no apparent difference in EC proliferation between TGF-J3 +1+ and -Imice, and the authors suggest that the defect is not due to premature endothelial differentiation but rather to “inadequate terminal differentiation, affecting endothelial tube formation and/or integrity.” Similar observations were made in animals deficient for TGF-13 receptor type 2 where embryonic lethal defects in hematopoiesis and vasculogenesis were noted (40). Further evidence of a role for TGF-13 in vascular development is provided by the finding that the mutations that lead to hereditary hemorrhagic telangiectasia (HHT), a vascular dysplasia, map to TGF-13 binding proteins. HHT was found to be the result of mutations in the genes that code for endoglin (41) and the activin receptor-like kinase 1 (42). Endoglin, an integral membrane protein that has been shown to be expressed specifically by vascular EC, binds TGF-13 and associates with the high-affinity TGF-13 receptors 1 and 2, presumably presenting the ligand to the receptors. HHT also mapped to the activin receptor-like kinase gene, a member of the serine-threonine kinase receptor family that is also expressed in EC. The investigators postulate that disruptions

in the

TGF-13

signaling

system

may

lead

to

abnormal vascular development, secondary to “altered repair of vascular endothelium and remodeling of the vascular tissue via changes in expression profiles of extracellular matrix proteins” (42). Consistent with these in vivo findings, we have shown that interactions between EC and undifferentiated mesenchymal cells (K. K. Hirschi et al., unpublished results, and ref 28) or between EC and SMC (or pericytes) (43) lead to the activation of TGF-13. The activated TGF-13, in turn, inhibits endothelial proliferation (44) and migration (36) and induces the differentiation of mesenchymal cells to a SMC/pericyte lineage (K. K. Hirschi et al., unpublished results, and ref 28). These in vitro observations may provide some explanation for the defective vasculogenesis observed in the mice null for TGF-131 and TGF-13 receptor 2. Epithelial-mesenchymal interactions in the developing vessel wall may lead to local activation of TGF-13. We speculate that the TGF-13 then acts to influence and stabilize vessel formation via a variety of its actions, including inhibition of EC proliferation and migration, induction of SMC/pericyte differentiation, and stimulation of matrix accumulation.

369

THE

Tie

RECEPTOR

FAMILY

AND

ANGIOPOIETIN Receptors The Tie receptors comprise a second class of receptor tyrosine kinases, in addition to receptors for VEGF, that appear to be specific for vascular endothelium. tie-i mRNA is first detected in the mouse at E8.5 in angioblasts of the head mesenchyme, in EC of the dorsal aortae, and in blood islands of the yolk sac (45). Expression of tie-i mRNA in vascular endothelium continues throughout embryogenesis,but in the adult animal expressionislimited to capillaries in the perialveolarsepta of the lung. Studies with Tie-i-deficient animals suggest that Tie-i is important for EC differentiation as well as for the establishment of vessel integrity (46). Tie-i null embryos die perinatally due to breathing difficulties, and display marked subcutaneous as well as internal hemorrhages and edema. Although the general vascular patterning is normal in these animals, ultrastructural analyses reveal that the vascular defect lies in the overall integrity of the EC. Capillary EC of Tie-i -Iembryos appear “electron light” with respect to their wild-type counterparts, and despite apparently intact cell-cell junctions, red blood cells were observed to extravasate through the endothelium. tie-i mRNA has been shown to be up-regulated in endothelium of two types of abnormal vessels, including the pericyte-deficient capillaries of PDGF-B -Imice (C. Betsholtz, personal communication, and P. Lindahl, unpublished results) and the endothelium of certain vascular malformations (47). Tiei-deficient animals were also generated by another group; in this case, the homozygous mutant embryos died at E14.5 with extensive hemorrhages and abdominal edema (48). The difference in the time of death between the two Tie-i null lines of mice may be due to differences in genetic background. The ability of Tie-i-deficient EC to contribute to the vasculature was assessed by creating chimeras from wildtype morulas and embryonic stem cells in which the tie-i allele had been disrupted with a lacZ reporter construct (49). At EiO.5, Tie-i-deficient cells contributed to all vessels in the embryos. In contrast, at E15.5, although the mutant cells were still found in many vessels, such as those of the heart and lung, they were underrepresented in vessels derived from angiogenic processes, such as those from the midbrain and kidney. In the adult there appeared to be further selection against the Tie-i-deficient EC, as none were found in the kidney, and there were reduced numbers in the vasculature of the lung. These results suggest a differential requirement for Tie-i in vascular beds derived by vasculogenesis vs. angiogenesis. Tie-2 (also known as Tek) is the second member of the family. It is detected first in the developing mouse embryo in the vasculature of the E7.5 yolk sac. tie-2 mRNA is detected in E8.0 embryos in the endocardium, in the dorsal aortae, in maternal decidual blood vessels, and in the yolk sac vasculature (50). Very little tie-2 mRNA is de-

370

Vol. 11

April 1997

tected in the endothelium of adult animals, although a strong signal was obtained with adult heart tissue by Northern blot analysis. Two approaches have been taken to eliminate Tie-2 function. Using a transgenic approach, a dominant negative Tie-2 was expressed under the control of a 7.2 kb tie2 promoter. Although the authors report that the mutant receptor displayed no kinase activity when expressed in vitro, the animals showed a variable, apparently nonspecific phenotype (51). In the same study, the authors deleted the translation start site and signal peptide to create Tie-2-deficient mice. The homozygous mutant embryos were all dead by E9.5, and displayed engorged and distended yolk sac vessels at E8.5. The dorsal aorta was ruptured and disorganized, and the hearts did not possess organized trabeculae. The authors note a general deficiency of EC. In an independent line of Tie-2 null mice (in which the tie-2 gene was disrupted in the second exon), all homozygous null embryos were dead at EiO.5, with a mild growth retardation (especially in the head and heart) apparent by E9.5 (46). The most prominent abnormalities in the mutant embryos were in the vasculature. There was little distinction between large and small vessels in the brain, suggesting a defect in vascular remodeling. In addition, the angiogenic capillary sprouts that usually protrude into the neuroectoderm were absent in the Tie-2 null embryos.

Tie-2

ligands

Ligands for Tie-2 have recently been identified and have been named angiopoietins (52). Angiopoietin-i is an activating ligand (i.e., it promotes tyrosine phosphorylation of the receptor). There is also an inhibitory ligand, angiopoietin-2, that apparently antagonizes the action of angiopoietin- 1 (G. D. Yancopoulos, personal communication). Angiopoietin-i is first expressed between E9 and Eli, most prominently in the heart myocardium surrounding the endocardium. Later in development the transcript is found more widely throughout the embryo, often associated with the mesenchyme adjacent to developing vessels. Angiopoietin-i-deficient mice have a phenotype similar to the Tie-2 -/mice (53). The angiopoietin-i null embryos die at Ei2.5 (ito 2 days later than the Tie-2 null embryos), and the most profound defects are in the heart where the endocardium and trabecular network are very immature. Similar to the Tie-2-deficient mice, the primary vascular plexus in the angiopoietin-i null animals does not fully remodel into large and small caliber vessels. A clue to the cellular basis of this remodeling defect comes from ultrastructural studies (53). In wild-type animals, existing vessels are split by “tissue folds” that contain organized collagen fibers and pericyte-like cells intimately associated with the endothelium. The tissue folds in the angiopoietin-1 null animals, by contrast, are less extensive, lack these pericytes, and contain scattered, disorganized collagen fibers. The EC in these areas are particularly

The FASEB Journal

BECK AND

D’AMORE

rounded and appear to have defective interactions with the underlying matrix. Recently, a dominantly inherited venous malformation has been mapped to the tie-2 locus (54). A single amino acid change has been found to lead to a 6- to iO-fold increase in autophosphorylation in a recombinantly expressed mutant receptor. The phenotype of the venous malformations is notable in that the vessels are dilated and display a paucity of SMC. The investigators hypothesize that the activating receptor mutation leads to a down-regulation of angiopoietin-i expression in the mesenchymal cells and, through an intracellular coupling pathway, to their decreased migration, proliferation, and differentiation. Another possibility is that angiopoietin-i normally induces the EC to release factors chemotactic for mesenchymal cells, but in the Tie-2 mutant cells the receptor is not truly activated in a physiological sense, so that this recruitment pathway is defective (55).

COAGULATION

FACTORS

Although the nature of their involvement is unclear, intriguing observations have also been made that link the coagulation factors to vascular development. Observations in mice with targeted disruption of the tissue factor (TF) and factor V gene, two proteins involved in coagulation, have led to the suggestion that these molecules may play a role in vessel formation.

Tissue

factor

TF, a high-affinity receptor and cofactor for factor VII/ VIIa, is the main cell-associated initiator of coagulation. Absence of TF leads to embryonic lethality between E9.5 and El0.5 (56). Although the defects in the TF -/embryos occurred with variable onset and penetrance, the phenotype of these mice includes an abnormal yolk sac vasculature characterized by a lack of uSM actin containing vascular wall cells, presumably the precursors for pericytes and SMC. A role for TF in regulating vessel growth is also suggested by studies in which overexpression of TF in a sarcoma cell line was shown to lead to larger and more vascularized tumors, whereas tumor cells transfected with antisense for TF formed smaller and less vascularized tumors (57). The phenotype of the TF-null embryos exhibits many similarities to those observed for the Tie-2 receptor/angiopoietin-i-deficient animals, with all having a notable lack of vascular remodeling. Yolk sac capillaries “appeared to fuse with each other in a disordered plexus,” a phenomenon the authors suggest is due to “defective formation of contacts between endodermal and mesodermal cell layers.” This defect also resembles that observed in TGF-131null mice, where defective adhesions between cell layers appeared to result in abnormal vasculogenesis (39). Although the similarities in these phenotypes are intriguing,

VASCULAR

DEVELOPMENT

there are no data to suggest where TF might pathways regulating vessel formation.

Factor

fit into the

V

Coagulation factor V has also been implicated in vasculogenesis,although the evidence to support this is less compelling than that for TF. About 50% of mice deficient for factor V die in utero between E9 and ElO due to abnormal development of the yolk sac vasculature (58). The remaining mice develop to term but die at birth, the result of massive hemorrhage. There are no data regarding the action of factor V on vascular cells outside of the coagulation pathway. However, the embryonic lethality of half of the embryos is similar to that observed in mice lacking the thrombin receptor, where 50% of the embryos die at E9.5 in the absence of any coagulation defects (59). Together, these results suggest that factor V-dependent generation of thrombin and its action through the thrombin receptor are necessary in early development and possibly during yolk sac vasculogenesis.

THE

INTEGRIN

cxf33

Vessel development, whether it be by vasculogenesis or angiogenesis, requires the migration of EC, a process that involves the function of cell adhesion molecules such as integrins. a433, an integrin expressed by EC, has been implicated in angiogenesis. Newly formed blood vessels in the granulation tissue of a healing cutaneous wound were found to express this integrin, whereas established vessels in normal skin were negative (60). This finding was later extended in studies in which a433 was localized to the tips of the EC in sprouting vessels (61). As the vessels matured, the a433 expression diminished to a level undetectable by immunohistochemistry. Perhaps more important, the function of the a,133integrin appears to be critical for the formation and/or maintenance of the newly formed vessels. Inclusion of a neutralizing monoclonal antibody specific for c133 in an angiogenesis assay on the chick chorioallantoic membrane led to the inhibition of neovascularization induced by bFGF and by human melanoma fragments (60); antibodies against a435, a related integrmn, had no effect. Blocking a133 had no apparent effect on the preexisting vessels of the chick chorioallantoic membrane, an observation that is consistent with the absence of a433 immunoreactivity in mature, quiescent vessels. Subsequent work has suggested that interfering with the interaction of the a.133integrin and its natural substrates (e.g. fibrin, fibronectin, vitronectin) induces apoptosis in the proliferating EC of the nascent vessels (62). The role of a133 in vasculogenesis has been addressed by microinjection of neutralizing antibodies into the developing quail embryo (63). In these embryos, the dorsal aortae, although positioned correctly, failed to form patent lumens. In addition, treated embryos displayed disconti-

371

nuities along the length of the dorsal aortae and an aberrant pattern formation of the adjacent vascular plexus. The role of the a433 integrmn during angiogenesis may involve more than attachment between cells and their substrate. Two recent reports have provided evidence that may also interact with proteins involved in matrix degradation. cx43:t has been shown to colocalize in growing blood vessels with the active form of matrix metalloproteinase-2, and specific binding between the two proteins was demonstrated in vitro (64). Vitronectin’s binding sites for a33 and plasminogen activator inhibitor-i (PAl-i) have been reported to overlap, such that when PAl-i is bound to vitronectin the a433 binding site is not available (65). The authors propose that plasminogen activator, in addition to activating plasminogen, may bind PAl-i, displace it from vitronectin, and thus unmask a binding site for the migrating cells. It has been shown that VEGF can up-regulate aJ33 and plasminogen activator in EC (66), which is consistent with the abilityof VEGF to stimulate EC migration. It is also interesting to consider whether angiostatin, an endogenous inhibitor of angiogenesis that has been shown to be derived from plasminogen (67), may block with vessel formation by interfering with these pathways.

CONCLUDING

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

REMARKS

Significant progress has been made in our understanding of the development of the vasculature, which has as its foundation descriptive yet elegant classical embryological studies. With the ability to culture vascular cells in vitro and the elucidation of defined molecules important in vascular cell growth and function, a new era of vascular biology began. Many factors critical to vascular development and regulation have been discovered and/or better characterized via analysis of genetically altered mice. Future studies will build on this current base of knowledge by addressing poorly understood areas such as vascular remodeling, signaling pathways transduced by endothelialspecific receptors, and the regulation of vascular cell differentiation, to name a few.

19.

20.

21. 22. 23.

24. 25. 26. 27.

REFERENCES 1. Pardanaud,L.,Yassine,F.,and Dieterlen-Lievre, F. (1989) Relationship between vasculogenesis, angiogenesis and haemopoiesis during avian ontogeny. Development 105, 473-485 2. McLcod, D. S.. Lutty. G. A., Wajer. S. D., and Flower, R. W. (1987) Visualization of a developing vasculature. Microvasc. Res. 33, 257-269 3. Folkman, J. (1995) Angiogenesis in cancer, vascular, rheumatoid and other diseases. Nature Med. 1, 27-3 1 4. Noden. D. M. (1989) Embryonic origins and assembly of blood vessels. Am. Rev.Respir. Dis. 140. 1097-1103 5. Fernig, D. G., and Gallagher, J. T. (1994) Fibroblast growth factors and their receptors: an information network controlling tissue growth, morphogenesisand repair. Prog. Growth Factor Re.c. 3, 353-377 6. Flamme, I., and Risau,W. (1992) Induction of vasculogenesis and hematopoiesisin vitro. Development116, 435-439 7. Klagsbrun, M., and D’Amore, P. A. (1996) Vascularendothelial growth factor and its receptors. Cytokine & Growth Factor Rev. 7, 259-270

372

Vol. 11

April 1997

28. 29.

30.

31.

32.

Keyt, B. A., Berleau, L. 1., Nguyen, H. V., Chen, H., Heinsohn, H., Vandlen, R., and Ferrara, N. (1996) The carboxyl-terminal domain (111-165) of vascular endothelial growth factor is critical for its mitogenic potency. J.Bail.Chem. 271, 7788-7795 Millauer, B., Wizigmann-Voos, S., SchnUrch, H., Martinez, R., M#{248}ler, N. P. H., Risau, W., and UlInch, A. (1993) High affinity VEGF binding and developmental expression suggest FIk-1 as a major regulator of vasculogenesisand angiogenesis. Cell 72, 835-846 Dumont, D. J., Fong, G.-H., Pun, M. C., Gradwohl, G., Alitalo, K., and Breitman, M. L. (1995) Vascularization of the embryo: a study of flk-i, tek,tie,and vascular endothelial growth factor expression during development. Dev. Dyn. 203, 80-92 Shalaby, F., Rossant, J., Yamaguchi, T. P., Gertsenstein, M., Wu, X.-F., Breitman, M. L., and Schuh, A. C. (1995) Failure of blood-island formation and vasculogenesis in Flk-1-deflcient mice. Nature (London) 376,62-66 Fong, G.-H., Rossant, J., Gertsenstein, M., and Breitman, M. L. (1995) Role of the fit-i receptor tyrosine kinase in regulating the assembly of vascular endothelium. Nature (London) 376, 66-70 Carmeliet, P., Femera, V., Breier, G., Pollefeyt, S., Kieckens, L., Gertsenstein, M., Fahrig, M., Vandenhoeck, A., Harpal, K., Eberhardt, C., Declercq, C., Pawling, J., Moons, L., Collen, D., Risau, W., and Nagy, A. (1996) Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature (London) 380, 435-439 Ferrara, N., Carver-Moore, K., Chen, H., Dowd, M., Lu, L., O’Shea, K., Powell-Braxton, L., Hillan, K. J., and Moore, M. W. (1996) Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature (London) 380,439-442 Drake, C. J., and Little, C. D. (1995) Exogenous vascular endothelial growth factor induces malformed and hyperfused vessels during embryonic neovascularization. Proc.Nati. Acad. Sci. USA 92, 7657-7661 Flamme, I., von Reutem, M., Drexler, H. C. A., Syed-Ali, S., and Risau, W. (1995) Overexpression of vascular endothelial growth factor in the avian embryo induces hypervascularization and increased vascular permeability without alterations of embryonic pattern formation. Dev. Biol. 171,399-414 Stone, J., Itin, A., Alon, T., Pe’er, J., Gnessin, H., Chan-Ling, T., and Keshet, E. (1995) Development of retinal vasculature is mediated by hypoxia-induced vascular endothelial growth factor (VEGF) expression by neuroglia. J. Neurosci. 15, 4738-4747 Alon, T., Hemo, I., Itin, A., Pe’er, J., Stone, J., and Keshet, S. (1995) Vascular endothelial growth factor acts as a survival factor for newly formed retinal vessels and has implications for retinopathy of prematurity. Nature Med. 1, 1024-1028 Miller, J. W., Adamis, A. P., Shima, D. 1., D’Amore, P. A., Moulton, R. S., O’Reilly, M. S., Folkman, J., Dvorak, H. F., Brown, L. F., Berse, B., Yeo, T.-K., and Yeo, K-I. (1994) Vascular permeability factor/vascular endothelial cell growth factor is temporally and spatially correlated with ocular angiogenesis in a primate model. Am. J. Pathol. 145, 574-584 Holmgren, L., Glaser, A., Pfeifer-Ohlsson, S., and Ohlsson, R. (1991) Angiogenesis during human extraembryonic development involves the spatiotemporal control of PDGF ligand and receptor gene expression. Developmesu 113, 749-754 Shinbrot, E., Peters, K. C., and Williams, L. I. (1994) Expression of the platelet-derived growth factor 3 receptor during organogenesis and tissue differentiation in the mouse embryo. Dev. Dyn. 199, 169-175 Battegay, E. J., Rupp, J., Iruela-Arispe, L., Sage, E. H., and Pech, M. (1994) PDGF-BB modulates endothelial proliferation and angiogenesis in vitro via PDGF 3-receptors. J. Cell Bail. 125, 917-928 Leveen, P., Pekny, M., Gebre-Medhin, S., Swolin, B., Larsson, E., and Betsholtz, C. (1994) Mice deficient for PDGF B show renal, cardiovascular, and hematological abnormalities. Genes & Dev. 8, 1875-1887 Soriano, P. (1994) Abnormal kidney development and hematological disorders in PDGF 3-receptor mutant mice. Genes & Dev. 8, 1888-1896 Schlondorff, D. (1987) The glomerular mesangial cell: an expanding role for a specialized pencyte. FASEB J. 1, 272-281 D’Amore, P. A., and Smith, S. R. (1993) Growth factor effects on cells of the vascular wall: a survey. Growth Factors 8, 61-75 Schatteman, C. C., Motley, S. 1., Effmann, E. L., and Bowen-Pope, D. F. (1995) Platelet-derived growth factor receptor alpha subunit deleted patch mouse exhibits severe cardiovascular dysmorphogenesis. Teratology 51, 351-366 Rohovsky, S. A., Hirschi, K. K., and D’Amore, P. A. (1996) Growth factor effects on a model of vessel formation. Surg. Forwn 47. 390-391 Corjay, M. H., Thompson, M. M., Lynch, K. R., and Owens, C. K. (1989) Differential effect of platelet-derived growth factor-versus serum-induced growth on smooth muscle cm-actin and nonmuscle 13-actin mRNA expression in cultured rat aortic smooth muscle cells. J. Bail.C/tern. 264, 10501-10506 Hungerford, J. E., Owens, G. K., Argraves, W. S., and Uttle, C. D. (1996) Development of the aortic vessel wall as defined by vascular smooth muscle and extracellular matrix markers. Dev. Bail.178, 375-392 Waldo, K. L., Kumiski, D. H., and Kirby,M. L. (1994) Association of the cardiac neural crest with development of the coronary arteries in the chick embryo. Anat.Rec.239, 315-331 Nicosia, R. F., and Villaschi, 5. (1995) Rat aortic smooth muscle cells become pericytes during angiogenesis in vitro. Lab. Invest. 73, 658-666

The FASEB Journal

BECK AND

D’AMORE

33.

Kingsley, D. M. (1994) The TGF-13 superfamily: New members, new receptors. and new genetic tests of function in different organisms. Genes & Dev. 8, 133-146 34. Heimark, R. L., Twardzik, D. R., and Schwartz, S. M. (1986) Inhibition of endothelial regeneration by type-beta transforming growth factor from platelets. Science 233, 1078-1080 35. Battegay, E. J., Raines, E. W., Seifert, R. A., Bowen-Pope, D. F., and Ross, R. (1991) TGF- induces bimodal proliferation of connective tissue cells via complex control of an autocrine PDGF loop. Cell63, 515-524 36. Sato, Y., and Rilkin, D. B. (1989) Inhibition of endothelial cell movement by pericytes and smooth muscle cells: activation of a latent transforming growth factor-beta 1-like molecule by plasmin during co-culture. J. Cell Bail.109,309-315 37. Basson, C. T., Kocher, 0., Basson, M. D., Asis, A., and Madri, J. A. (1992) Differential modulation of vascular cell integrin and extracellular matrix expression in vitro by TGF-3i correlates with reciprocal effects on cell migration. I. Cell. Physail. 153, 118-128 38. Shull, M. M.,Ormsby, I., Kier. A. B., Pawlowski, S., Diebold, R. J., Yin, M., Allen, R., Sidman, C., Proetzel, C., Calvin, D., Annunziata, N., and Doetschman, T. (1992) Targeted disruption of the mouse transforming growth factor-1 gene results in multifocal inflammatory disease. Nature (London) 359, 693-699 39. Dickson, M. C., Martin, J. S., Cousins, F. M., Kulkarni, A. B., Karlsson, S., and Akhurst, R. J. (1995) Defective haematopoiesis and vasculogeneisis in transforming growth factor-1 knock-out mice. Development 121, 1845-1854 40. Oshima, M., Oshima, H., and Taketo, M. M. (1996) TCF-13 receptor type II deficiency results in defects of yolk sac hematopolesis and vasculogenesis. Dev. Bail.179,297-302 41. McAllister, K. A., Crogg, K. M., Johnson, D. W., Gallione, C. J., Baldwin. M. A., Jackson, C. E., Hembold, E. A., Markel, D. S., McKinnon, W. C., Murrell, J., McCormick, M. K., Pericak-Vance, M. A., Heutink, P., Oostra, B., A,, Haitjema, I., Westerman, C. J. J., Porteous, M. E., Cuttmacher, A. E., Letarte, M., and Marchuk, D. A. (1994) Endoglin, a TCF-13 binding protein of endothelial cells, is the gene for hereditary haemorrhagic telangietasia type 1. Nature Genet. 8,345-351 42. Johnson, D. W., Berg, J. N., Baldwin, M. A., Gallione, C. J., Marondel, I., Yoon, S.-J., Stenzel, I. I., Speer, M., Pericak-Vance, M. A., Diamond, A., Guttmacher, A. E., Jackson, C. E., Attisano, L., Kucherlapati, R., Porteous, M. E. M., and Marchuk, D. A. (1996) Mutations in the activin receptorlike kinase 1 gene in hereditary haemorrhagic telangiectasia type 2. NatureGenet. 13, 189-195 43. Antonelli-Orlidge, A., Saunders, K. B., Smith, S. R., and D’Amore, P. A. (1989) An activated form of transforming growth factor JI is produced by cocultures of endothelial cells and pencytes. Proc.Nail. Acad. Sci. USA 86,4544-4548 44. Orlidge, A., and D’Amore, P. A. (1987) Inhibition of capillary endothelial cell growth by pericytes and smooth muscle cells. I. Cell Biol. 105, 14551462 45. Korhonen, J., Polvi, A., Partanen, J., and Alitalo, K. (1994) The mouse tie receptor tyrosine kinase gene: expression during embryonic angiogenesis. Oncogene 9, 395-403 46. Sato, I., Tozawa, Y., Deutsch, U., Wolburg-Buchholz, K., Fujiwara, Y., Cendron-Maguire, M., Gridley, I., Wolburg, H., Risau, W., and Qin, Y. (1995) Distinct roles of the receptor tyrosine kinases Tie-i and Tie-2 in blood vessel formation. Nature (London) 376, 70 47. Hatva, E., Jaaskelainen, J., Hirvonen, H., Alitalo, K., and Haltia, M. (1996) Tie endothelial cell-specific receptor tyrosine kinase is upregulated in the vasculature of arteriovenous malformations. J. Neuropath.Expt. Neurol. 55, 1124-1133 48. Puri, M. C., Rossant, J., Alitalo, K., Bernstein, A., and Partanen, J. (1995) The receptor tyrosine kinase TIE is required for integrity and survival of vascular endothelial cells. EMBO 1. 14, 5884-5891 49. Partanen, J., Pun, M. C.. Schwartz,L, Fischer, K.-D., Bernstein, A., and Rossant, J. (1996) Cell autonomous functions of the receptor tyrosine ki-

VASCULAR

DEVELOPMENT

50.

51.

52.

53.

54.

55. 56.

57.

58. 59. 60. 61. 62.

63. 64.

65. 66.

67.

nase TIE in a late phase of angiogenic capillary growth and endothelial cell survival during munne development. Development 122, 3013-3021 Sato, T. N., Qin, Y., Kozak, C., and Audus, K. L. (1993) tie-i and tie-2 define another class of putative receptor tyrosine kinase genes expressed in early embryonic vascular system. Proc.Nod. Acad.Sci.USA 90,93559358 Dumont, D. J., Gradwohi, C., Fong, C-H., Pun, M. C., Gertsenstein, M., Auerbach, A.. and Breitman, M. L. (1994) Dominant-negative and targeted null mutations in the endothelial receptor tyrosine kinase, tek, reveal a critical role in vasculogenesis of the embryo. Genes & Dev. 8, 1897-1907 Davis, S., Aldrich, I., H,, Jones, P. F., Acheson, A., Compton, D. L., Jam. V., Ryan, T. E., Bruno. J., Radziejewski, C., Maisonpierre, P. C., and Yancopoulos, C. D. (1996) Isolation of angiopoietin-i, a ligand for the TIE2 receptor, by secretion-trap expression cloning. Cell 87, 1161-1169 Sun, C., Jones, P. F., Patan, S., Bartunkova, S., Maisonpierre. P. C., Davis, S., Sato, T. N., and Yancopoulos, C. D. (1996) Requisite role of angiopoietin-1, a ligand for the TIE2 receptor, during embryonic angiogenesis. Cell87, 1171-1180 Vikkula,M., Boon, L M., Carraway. K. L., Calvert, J. I., Diamonti. A. J., Goumnerov, B., Pasyk, K. A., Marchuk, D. A., Warman. M. L, Cantley. L. C., Mulliken, J. B., and Olsen, B. R. (1996) Vascular dysmorphogenesis caused by an activating mutation in the receptor tyrosine kinase TIE2. Cell 87, ii8i-i190 Folkman, J., and D’Amore, P. A. (1996) Blood vessel formation: What is itsmolecularbasis?Cell87, 1153-1155 Carmeliet, P., Mackman, N., Moons, L., Luther, I., Gressens, P., Van Vlaenderen, I., Demunck, H., Kasper, M., Breier, C., Evrard, P., Muller, M., Risau, R., Edgington, I., and Collen, D. (i996) Role of tissue factor in embryonic blood vessel development. Nature (London) 383, 73-75 Zhang, Y., Deng, Y., Luther, T., Muller, M., Ziegler, R., Waldherr, R.. Stem, D. M., and Naawroth, P.P.(1994) Tissue factor controls the balance of angiogenic and antiangiogenic properties of tumor cells in mice. J. Clin. Invest. 94, 1320-1327 Cui, J., O’Shea, K. S., Purkayastha, A., Saunder, I. L., and Ginsburg, D. (1996) Fatal haemorrhage and incomplete block to embryogenesis in mice lacking coagulation factor V. Nature (London) 384, 66-68 Connolly, A. J., Ishihara, H., Kahn, M. L.. Farese,R. V., Jr., and Coughlin, S. R. (1996) Role of the thrombin receptor in development and evidence for a second receptor. Nature (London) 381, 516-519 Brooks, P. C., Clark, R. A. F., and Cheresh, D. A. (1994) Requirement of vascular integrin a, for angiogenesis. Science 264, 569-571 Clark, R. A. F., Tonnesen, M. C., Cailit, J., and Cheresh, D. A. (1996) Transient functional expression of a433 on vascular cells during wound repair. Am. J.Pathol. 148, 1407-1421 Brooks, P. C., Montgomery, A. M. P., Rosenfeld, M., Reisfeld, R. A., Hu, I., Klier, C., and Cheresh, D. A. (1994) Integrin a,3 antagonists promote tumor regression by inducing apoptosis of angiogenic blood vessels. Cell 79, 1157-1164 Drake, C. J., Cheresh, D. A., and Little, C. D. (1995) An antagonist of integnin a,f3., prevents maturation of blood vessels during embryonic neovascularization. J. CellSci.108, 2655-2661 Brooks, P. C., Stromblad, S., Sanders, L C., von Sehalscha, I. L, Aimes, R. I., Stetler-Stevenson, W. C., Quigley, J. P., and Cheresh, D. A. (1996) Localization of matrix metalloproteinase MMP-2 to the surface of invasive cells by interaction with integrin a.,3. Cell85, 683-693 Stefansson, S., and Lawrence, D. A. (1996) The serpin PAl-I inhibits cell migration by blocking integrin a., binding to vitronectin. Nature (London) 383,441-443 Pepper, M. S., Ferrara,N.,Orci,L.,and Montesano, R. (1991) Vascular endothelial growth factor(VEGF) induces plasminogen activators and plasminogen activator inhibitor-i in microvascular endothelial cells. Biochem. Biophys. Res.Commun. 181,902-906 O’Reilly, M. S., Holmgren, L., Shing, Y., Chen, C., Rosenthal, R. A.. Moses, M., Lane, W. S., Cao, Y., Sage, H. E., and Folkman. J. (1994) Angiostatin: a novelangiogenesisinhibitor that mediates the suppression of metastases by a Lewis lung carcinoma. Cell79. 315-328

373