Vascular endothelial growth factorA regulator of vascular morphogenesis in the Japanese quail embryo.код для вставкиСкачать
THE ANATOMICAL RECORD PART A 272A:403– 414 (2003) Vascular Endothelial Growth Factor: A Regulator of Vascular Morphogenesis in the Japanese Quail Embryo ERIC B. FINKELSTEIN AND THOMAS J. POOLE* Department of Cell and Developmental Biology, SUNY Upstate Medical University, Syracuse, New York ABSTRACT Experiments in mouse embryos indicate that a critical level of VEGF is required for normal vascular development, as mice lacking a single VEGF allele die at midgestation. Thus VEGF concentration may be a determinant of the size and location of major blood vessels during formation of the primary capillary plexus. Ectopic VEGF delivery was used to examine the effect of VEGF concentration on early vascular patterning in the quail embryo. VEGF was delivered by implanting VEGF-soaked heparin chromatography beads at three rostral-caudal locations in embryos with six somite pairs, which allowed us to study the effect of VEGF on different cellular activities. Ectopic VEGF resulted in signiﬁcant changes in the vascular pattern at three rostral-caudal levels. Quantitation demonstrated an increased vascularity in the area of the implanted VEGF bead compared to the vascular pattern of embryos with control beads. Areas lateral to the dorsal aortae that are normally avascular became vascularized, and there was an apparent fusion between the dorsal aorta and lateral capillary plexus. Anat Rec Part A 272A:403– 414, 2003. © 2003 Wiley-Liss, Inc. Key words: VEGF; vascular development; vasculogenesis; quail embryo; heparin bead Vascular development begins in situ during early somite stages in the avian embryo when angioblasts (endothelial cell precursors) emerge parallel to the embryonic midline (Pardanaud et al., 1987; Cofﬁn and Poole, 1988; Dieterlen-Lievre and Pardanaud, 1998). Angioblasts, apparently induced by ﬁbroblast growth factor-2 (FGF-2) (Flamme and Risau, 1992; Krah et al., 1994; Cox and Poole, 2000), begin to express vascular endothelial growth factor receptor-2 (VEGFR2) (Eichmann et al., 1993, 1996; Yamaguchi et al., 1993). The primary capillary plexus forms from the induced angioblasts by a poorly understood morphogenetic process inﬂuenced by FGF, vascular endothelial growth factor (VEGF), adhesion molecules, and the extracellular matrix (Tallquist et al., 1999; Poole et al., 2001). In a mature blood vessel, smooth muscle cells and pericytes surround an endothelial cell layer. This mature vessel structure forms following remodeling of the primary capillary plexus (reviewed in Darland and D’Amore, 1999; Conway et al., 2001). There are two major mechanisms of vascular development: vasculogenesis and angiogenesis (Pardanaud et al., © 2003 WILEY-LISS, INC. 1989; Poole and Cofﬁn, 1989, 1991). Vasculogenesis is the de novo formation of blood vessels through angioblast migration and cohesion, and is the mechanism by which the paired dorsal aortae form ventrolaterally to the somites on either side of the embryonic midline (Cofﬁn and Poole, 1988; Poole and Cofﬁn, 1988, 1989; Poole et al., 2001). Angiogenesis is the formation of new blood vessels from preexisting vessels, and is the mechanism by which E.B. Finkelstein’s current address is Schepens Eye Research Institute, 20 Staniford St., Boston, MA 02114. *Correspondence to: Dr. Thomas J. Poole, Department of Cell and Developmental Biology, SUNY Upstate Medical University, 766 Irving Ave., Syracuse, NY 13210. Fax: (315) 464-8535. E-mail: email@example.com Received 25 July 2002; Accepted 15 December 2002 DOI 10.1002/ar.a.10047 404 FINKELSTEIN AND POOLE the intersomitic arteries form by sprouting from the dorsal aortae (Poole and Cofﬁn, 1989, 1991; Poole et al., 2001). Vasculogenesis predominates in the quail embryo until the seven-somite stage (Hamburger-Hamilton (HH) stage 9 (Hamburger and Hamilton, 1951)), after which both mechanisms contribute to vascular development as intersomitic arteries branch from the rostral dorsal aortae, and free angioblasts undergo cohesion to expand the vascular plexus in caudal regions. One growth factor involved in vascular development is VEGF (also known as VEGF-A) (Conway et al., 2001; Poole et al., 2001). VEGF is identiﬁed by its mitogenic effect on endothelial cells, and its ability to induce endothelial cell chemotaxis (Ferrara, 1999) and localized edema (Senger et al., 1983). All embryonic cell types, except endothelial cells, express VEGF. A particularly high level of VEGF expression is observed in the embryonic endoderm (Flamme et al., 1995a; Aitkenhead et al., 1998). In contrast, VEGF receptors are expressed exclusively on endothelial cells and angioblasts, often adjacent to areas of VEGF expression (Breier et al., 1995; Flamme et al., 1995a; Eichmann, 1996). VEGF and its receptors are essential for vascular development (reviewed in Poole et al., 2001). Mice that lack a single VEGF allele die at embryonic day 11.5 (Carmeliet et al., 1996; Ferrara et al., 1996), which suggests that regulation of VEGF concentration is required during development. Severe defects have also been reported in embryos that are homozygous-null for VEGF, with death on day 10.5 (Carmeliet et al., 1996). In VEGF-null embryos, endothelial cell numbers are reduced and a vascular pattern fails to form. Embryos that are deﬁcient for VEGF receptors are also embryonically lethal, with a disorganized vascular pattern (Fong et al., 1995, 1999; Shalaby et al., 1995). The quail embryo has been used extensively for studies of early events in vascular development. This has been facilitated by the use of a monoclonal antibody, QH-1, which speciﬁcally recognizes quail endothelial cells and angioblasts (Pardanaud et al., 1987). By using the quail embryo as an “in vivo vasculogenesis assay” (Drake et al., 1992), growth factors can be assayed for their effects on the vascular pattern. In one study (Drake and Little, 1995), microinjection of recombinant human VEGF in the segmental plate of quail embryos with ﬁve to six somite pairs resulted in hypervascularization. Other studies using retroviral delivery demonstrated that four isoforms of quail VEGF could affect the vasculature by increasing blood vessel permeability without altering vessel patterns. To further examine the role of VEGF in quail vascular pattern formation, microsurgical techniques were used to deliver recombinant human VEGF165 (rhVEGF165) to day 2 quail embryos. Since VEGF is a heparin-binding protein, chemically linked heparin chromatography beads were used for VEGF delivery. Human VEGF165 and its quail counterpart, VEGF166, are 75% similar (Yue and Tomanek, 2001), so it was expected that the effects of human VEGF165 would be similar to that with native quail VEGF. The use of beads to deliver growth factors has several advantages over microinjection, including focal delivery of growth factors, a slow-release source of growth factor, and a traceable delivery location. This method has been used to deliver FGF-2 (Cox and Poole, 2000) and BMP-4 (Schmidt et al., 1998) to precise locations in avian embryos, and to deliver VEGF to explanted lung tissue (Healy et al., 2000). The present study focuses on morphogenesis of the paired dorsal aortae, which proceeds in a rostral to caudal sequence in three distinct stages. The ﬁrst stage is the migration and cohesion of single angioblasts along the embryonic midline to form a solid cord. The solid cord then differentiates into an endothelial-lined tube as a lumen forms (the mechanisms of which are unclear). Finally, vessels branch from the tubular dorsal aorta to form the intersomitic arteries. Data presented here show that bead-delivered VEGF can affect all three stages of dorsal aorta morphogenesis. Ectopic VEGF also resulted in increased vascularization lateral to the dorsal aortae, as well as apparent fusion between the dorsal aortae and the capillary plexus lateral to bead implantation sites. MATERIALS AND METHODS Microsurgery Methods Fertile Japanese quail eggs (Coturnix coturnix japonica) were purchased from the Cornell University Poultry Science Labs (Ithaca, NY). Eggs were incubated in a forcedair humidiﬁed incubator at 38°C and 60% relative humidity (Hamburger, 1960) for approximately 32 hr, until six or seven somite pairs formed (HH stage 8-9 (Hamburger and Hamilton, 1951)). Throughout this work, the number of formed somite pairs (6S ⫽ six somite pairs) will be used to indicate the developmental stage of the embryos. This is more precise than using the HH staging system for early development, since each HH stage contains a range of somite numbers (for example, HH stage 8 has four to six somites while HH stage 9 has seven to nine somites). Bead Implantation Lyophilized recombinant human VEGF 165 (rhVEGF165; PeproTech, Rocky Hill, NJ) was reconstituted at a concentration of 1.0 mg/ml in sterile phosphate-buffered saline (PBS). An equal volume of growth factor and PBS-washed heparin-Toyopearl chromatography beads (Supelco, Bellefonte, PA) was mixed and incubated for 1 hr at room temperature for approximately 0.5 mg/ml VEGF, as previously described for other growth factors (Crossley et al., 1996; Schmidt et al., 1998; Cox and Poole, 2000). This high soaking concentration was used to saturate the beads with growth factor (Eichele et al., 1984; Crossley et al., 1996; Schmidt et al., 1998; Cox and Poole, 2000). After the beads were soaked (referred to here as VEGF beads), they were rinsed in sterile Howard’s saline (0.12 M NaCl, 1.53 mM CaCl2, 4.96 mM KCl) (Packard et al., 2000). Beads 50 – 60 m in diameter were visually selected and implanted into somites of embryos with six or seven somite pairs. Somites were used for bead delivery because of their close proximity to the developing dorsal aortae, and because somites give rise to fewer angioblasts than do other mesodermally-derived tissues (Wilting et al., 1995; Cox and Poole, 2000). Beads were implanted into the somites with an electrolytically sharpened tungsten needle by making a small incision in the dorsal or lateral wall of the somite and pushing a bead into the somite, without removing any tissue. Beads soaked in sterile PBS without growth factor were implanted into embryos as controls (referred to as control beads). The VEGF and control beads were implanted in three locations along the rostralcaudal axis to study the effect of ectopic VEGF on vascular development. Somites within the embryo are at different developmental stages along the rostral-caudal axis. They 405 VEGF AND VASCULAR PATTERN FORMATION can be precisely staged using a nomenclature system in which Roman numerals are used to number the somites rostrally from the segmental plate (I being the most recently formed), and an Arabic numeral is used to indicate the total number of somites in the embryo (Christ and Ordahl, 1995; Stockdale et al., 2000). In embryos with six somite pairs, beads were implanted in somites I/6 (referred to as the last-formed somite), IV/6 (referred to as the middle somite), and VI/6 (referred to as the ﬁrstformed somite) (Fig. 1A). In embryos with seven somite pairs, beads were implanted in somites I/7 (referred to as the last-formed somite), IV/7 (referred to as the middle somite), and VI/7 (a similar stage as the ﬁrst-formed somite in an embryo with six somite pairs). For all experiments, the blastoderm was dissected from the yolk in sterile Howard’s saline and cultured dorsal side up on agar/albumen dishes (Packard et al., 2000). Whole-egg supernatant with a 1:100 dilution of 29.2 mg/ml L-glutamine, 10,000 U/ml penicillin-G, and 10,000 mcg/ml streptomycin (Irvine Scientiﬁc, Santa Ana, CA) added was used for media. After incubation for 4 or 7 hr at 38°C with 60% relative humidity, the embryos were removed from the culture dishes, rinsed in PBS, and ﬁxed for analysis. Immunostaining of Whole-Mount Embryos For QH-1 whole-mount immunostaining, embryos were ﬁxed in 10% formalin as previously described (Cofﬁn and Poole, 1988; Cox and Poole, 2000). Nonspeciﬁc antibody binding was blocked with 3% bovine serum albumin (BSA; Fraction V powder; Sigma, St. Louis, MO), in PBT (PBS with 1% BSA, 0.1% Triton X-100). Embryos were then incubated with QH-1 monoclonal antibody tissue culture supernatant (Developmental Hybridoma Studies Bank, Iowa City, IA) diluted 1:10 in 3% BSA/PBT, and rocked overnight at 4°C. After the primary antibody incubation, the embryos were washed with PBT, blocked and incubated overnight with a cy2-conjugated anti-mouse IgG F(ab⬘)2 secondary antibody (Jackson Immunoresearch, West Grove, PA) in BSA/PBT at 4°C. The embryos were washed in PBS for 4 hr or overnight at 4°C and dehydrated through graded ethanols, cleared in toluene, and mounted on glass slides ventral side up with Entellan (EM Sciences, Fort Washington, PA). Image Collection and Analysis Embryos stained as whole mounts were imaged using laser scanning confocal microscopy with a 1024ES confocal imaging system (BioRad, Hercules, CA) mounted on a Nikon Eclipse E600 microscope (Nikon, Melville, NY). Images were collected with a 10⫻ objective at 5.0-m intervals through the entire stained embryo as a z-series with Laser Sharp Image Acquisition software (BioRad). Each z-series consisted of 30 – 40 sections depending on thickness of embryos when mounted on slides. Individual zseries were collapsed into vertical projections by the use of Laser Sharp Processing and Confocal Assistant software (BioRad). A vertical projection combined all images collected through a three-dimensional embryo into a single one-dimensional image. The locations of beads and the stages of embryos at ﬁxation were conﬁrmed using differential interference contrast (DIC) microscopy. Changes in vascular patterns that resulted from ectopic growth factor delivery were quantiﬁed. The Image Pro- cessing Tool Kit (Reindeer Games, Asheville, NC), a set of plug-ins for Adobe Photoshop, was used for all analyses. Color images were converted to grayscale, and then the images were inverted such that stained areas were dark on a light background. Peaks in the image histogram were separated for further processing using an equalize function with linear emphasis. Stained vascular areas were selected and copied to manually exclude background staining in the neural tube and somites. Finally, multitoned images were converted to binary images using a set threshold value. Binary images were visually inspected to ensure that they accurately reﬂected the staining patterns in the original confocal images (Fig. 1B and C). Measurements were made from the binary images with a sample area approximately 250 m on a side, extending from the medial dorsal aorta to the lateral edge of the capillary plexus with the bead-implant site centered rostral-caudally. The use of this sample area enabled the stained areas to be measured reproducibly between embryos. This sample area was optimal for measuring maximal effects at a radius of 125 m from implanted beads. The same sample area was used for measurements on the bead-implanted and contralateral sides of the embryos, ensuring that the measurements were at the same rostralcaudal level. The stained area fraction was measured utilizing the ‘‘global’’ measurement command in the plug-in. The stained area fraction is the area occupied by stained pixels (black) relative to the total sample area, expressed as a percent: Area fraction (%) ⫽ Stained area . Total sample area The stained area fraction was recorded for the experimental and control sides of each embryo, and these data were used in all subsequent analyses. In about 30% of the embryos with beads implanted, a hole resulted from the operation. This hole is a region devoid of tissue due to beads protruding through the ventral side of the embryo, and can be visualized as an unstained area. For embryos with holes, the stained area fraction was measured using the standard sample area, with areas of the holes subtracted out to yield a modiﬁed area fraction: Stained area Modiﬁed area fraction (%) ⫽ [Total sample area- area of hole]. The area of the holes was approximately 10% of the sample area measured. The same area for the holes was subtracted from sample areas on both sides of the embryos. Statistical Analysis For analysis, stained area fractions as measured were grouped by embryo starting stage and bead location. From the grouped measurements, the mean, standard deviation (S.D.), and standard error of the mean (S.E.M.) were calculated. The total number of embryos with control beads was seven in the ﬁrst-formed somite, nine in the middle somite, and ﬁve in the last-formed somite. The total number of embryos with VEGF beads used in analyses was 11 in the ﬁrst-formed somite, 18 in the middle somite, and 12 in the last-formed somite. For statistical analysis, P ⬍ 0.01 was designated to be signiﬁcant. The bead-implanted 406 FINKELSTEIN AND POOLE Fig. 1. Somite numbering and quantitation methods. A: Confocal image of a normal (unoperated) embryo with six somite pairs (6S) to show the extent of vascularization at the stage the VEGF was delivered. QH-1 labeling was performed without Triton X-100 in the washes to slightly increase background and provide a clear view of the somites. Along the left side, the somites are numbered as described in Materials and Methods. CP, capillary plexus; DA, dorsal aortae (cord beginning to form); SOM, somites; NT, neural tube; PSM, presegmental mesoderm. B-C: Quantitation of changes in vascular patterns resulting from delivery of ectopic VEGF on heparin-Toyopearl chromatography beads. Parts B and C are the same images as in Fig. 3C and D, with VEGF beads implanted in the middle somite of embryos with six somite pairs. Em- bryos were incubated for 7 hr and stained with the QH-1 monoclonal antibody as whole mounts. Row 1 depicts grayscale confocal images. Row 2 depicts the same images after vascular areas were selected, copied, and converted to binary images. Note that this matches the gray staining in row 1. The boxed areas are those selected for quantitation of QH-1 staining, shown in isolation in row 3. Row 3 shows the regions of embryos in which the QH-1 stained area fractions (the proportion of stained area relative to total selected area) around the bead were measured on the bead-implanted and contralateral sides of the embryos. [Color ﬁgure can be viewed in the online issue, which is available at www.interscience.wiley.com.] VEGF AND VASCULAR PATTERN FORMATION Fig. 2. Ectopic delivery of VEGF at the level of the last-formed somite. Beads were incubated in (B) PBS (control) or in (C and D) recombinant human VEGF165 (VEGF) and implanted in the last-formed of 6S quail embryos, followed by a 7-hr incubation. All embryos were stained as whole mounts with the QH-1 monoclonal antibody, followed by confocal microscopy. A: A schematic of an embryo to indicate a bead implanted in the last-formed somite (dark somite). B: An embryo with a control bead, ﬁxed at the 11S stage. C: An embryo with a VEGF bead, ﬁxed at the 12S stage. D: An embryo with a VEGF bead, ﬁxed at the 10S stage. Note the increased vascularization on (C) the bead-implanted side and (D) both sides, compared to (A) an embryo with a control bead. These are ventral views, with the rostral end up. Asterisks indicate bead locations. AVA, avascular area; DA, dorsal aortae; NT, neural tube; CP, capillary plexus; SOM, somites. The scale bar in D is representative of 100 m and applies to B–D. sides of embryos with beads on the left were compared to those with beads on the right by means of Student’s t-test. If differences were not statistically signiﬁcant, data for beads on both the left and right sides were combined for further analyses. Complete analysis was carried out for embryos with VEGF or control beads. The paired t-test was used to compare bead-implanted and contralateral sides within the same embryo. Student’s t-test was used to compare bead-implanted sides of embryos with VEGF or control beads at the same location with the same starting stage. Student’s t-test was also used to compare contralateral sides of embryos with VEGF or control beads. All 407 Fig. 3. Ectopic VEGF delivery at the level of the middle somite. (C and D) VEGF or (B) control beads were implanted in the middle somite of 6S stage quail embryos, followed by a 7-hr incubation. All embryos were stained as whole mounts with the QH-1 monoclonal antibody, followed by confocal microscopy. A: The location of beads when implanted in the middle somite (dark somite), shown schematically. B: An embryo with a control bead, ﬁxed at the 11S stage. A normal avascular area is present lateral to the dorsal aortae on either side of the embryo. C: An embryo with a VEGF bead, ﬁxed at the 10S stage. Note the reduced avascular area adjacent to the bead, with numerous blood vessel branches between the dorsal aorta and capillary plexus, compared to the contralateral side. D: An embryo with a VEGF bead, ﬁxed at the 12S stage. The normally avascular area adjacent to the bead is extensively vascularized. These are ventral views, with the rostral end up. Asterisks indicate bead locations. AVA, avascular area; DA, dorsal aortae; NT, neural tube; CP, capillary plexus; SOM, somites. The scale bar in D is representative of 100 m and applies to B–D. statistical analyses were carried out with Excel 2000 (Microsoft, Redmond, CA). Immunostaining of Sections For QH-1 staining of sections, embryos were ﬁxed in 4% paraformaldehyde (PFA) in PBS, and processed for embedding in Paraplast (Monoject, St. Louis, MO) following standard procedures. Then 10-m transverse sections were mounted on albumen-subbed slides. For immunostaining, sections were deparafﬁnized and rehydrated following standard procedures, and blocked with 10% fetal 408 FINKELSTEIN AND POOLE TABLE 1. Control (PBS) beads implanted in six-somite stage quail embryos Bead location n Sidea Stained area fraction (%)b Comparison with control side First-formed somite Middle somite 7 Control 54.4 ⫾ 3.9 Last-formed somite 9 Bead 54.4 ⫾ 3.1 P⫽NS Control 53.6 ⫾ 2.1 5 Bead 53.1 ⫾ 1.6 P⫽NS Control 58.4 ⫾ 1.3 Bead 56.3⫾ 2.9 P⫽NS a Measurements were made on (bead)-implanted and contralateral (control) sides of embryos. The mean stained area fraction of all embryos with control (PBS) beads at this location (expressed as a percent ⫾ SEM: Standard Error of the Mean). NS, not signiﬁcant. b bovine serum (FBS; Summit Biotechnology, Fort Collins, CO) in PBS. Sections were then incubated with a 1:50 dilution of QH-1 tissue culture supernatant followed by rhodamine-red-X-conjugated anti-mouse IgG F(ab⬘)2 secondary antibody (Jackson Immunoresearch) each in 10% FBS/PBS, for 1 hr at room temperature. After a ﬁnal wash in PBS, sections were dehydrated through graded ethanols, cleared in xylene, and mounted with coverslips with Entellan (EM Sciences, Fort Washington, PA). Images were collected by use of a 20⫻ objective on a Nikon Eclipse E800 epiﬂuorescence microscope (Nikon, Melville, NY) equipped with a SPOT-RT digital camera (Diagnostic Instruments, Sterling Heights, MI), and processed with Adobe Photoshop 5.5 (Adobe Systems, San Jose, CA). RESULTS VEGF Can Affect Initial Events of Dorsal Aorta Development Beads implanted in the last-formed somite (I/6) were used to study the effect of VEGF on initial stages of dorsal aorta formation, including angioblast cohesion (Fig. 2A). A control bead (soaked in PBS) (Fig. 2B) did not have an observable effect on the vascular pattern within the embryo. Both sides of the embryo developed a normal dorsal aorta approximately 100 m in diameter. The capillary plexus had a reticular appearance and was separated from each dorsal aorta by an avascular area. Statistics conﬁrmed that the control bead implant produced no signiﬁcant change in the vascular pattern (Table 1). Therefore, the observed effects of the VEGF beads were due to ectopic growth factor delivery. VEGF beads implanted in the last-formed somite (I/6) of quail embryos with six somite pairs resulted in an increased degree of vascularization lateral to the dorsal aorta (Fig. 2C). This effect was observed directly lateral to the implanted beads, extending several hundred microns along the rostral-caudal axis. Regions of the embryo that are normally avascular became heavily vascularized in the majority of samples (92%; Table 2 and Fig. 2C). An apparent fusion occurred between the dorsal aorta and capillary plexus (Fig. 2C). In the remaining samples (8%), a VEGF bead produced an increased number of blood vessel branches near the bead site (Table 2; Fig. 2D). The effect of a VEGF bead was not always restricted to the side of an embryo where the bead was implanted. A decreased avascular area was sometimes observed on the unoperated sides of embryos with VEGF beads (Fig. 2D). No statistical difference between VEGF beads implanted on the left and right sides occurred in 6 somite stage embryos at the level of the last-formed somite (Table 3). Therefore, embryos with beads on the left or right side were pooled for further analysis. With beads implanted at the level of the last-formed somite, there was no statistical difference between QH-1-stained areas on bead-implanted and contralateral sides within individual embryos (Table 3). This was supported by observed morphological effects on the vasculature of both sides in half of the embryos analyzed (6/12; 50%), even though beads had been implanted on only one side in each embryo. The effect on both sides of the embryos was further examined by comparison of QH-1 staining on control sides of embryos with control or VEGF beads. Our analysis revealed a signiﬁcant difference between control sides of embryos with VEGF or control beads at the level of the last-formed somite (Table 3). A signiﬁcant difference between the bead-implanted sides of embryos with VEGF or control beads indicated that all observed effects resulted from ectopic VEGF (Table 3). Similar results were obtained for beads implanted in the last-formed somite of embryos with seven somite pairs (data not shown). Ectopic VEGF Has an Effect When Delivered at a Level of Active Lumen Formation No morphological effects were observed when control beads were implanted in the middle somite (Fig. 3B). A normal avascular area lateral to the dorsal aorta was crossed by only small-caliber blood vessels, and the capillary plexus was distinct from the dorsal aorta on the bead-implanted and control sides of the embryos (Fig. 3A). As with the control beads in the last-formed somite, no signiﬁcant difference resulted from implanting a control bead in the middle somite (as determined by a comparison of bead-implanted and control sides (Table 1)). When a VEGF bead was implanted in the middle somite of 6S embryos (Fig. 3C and D), observable changes in the vascular pattern included increased vascularization lateral to the dorsal aorta (Fig. 3C and D), with an apparent fusion between the dorsal aorta and capillary plexus (Fig. 3D). The dorsal aorta was expanded compared to that in an embryo with a control bead implanted at the same level (Fig. 3B). While loss of avascular areas was limited to bead-implanted sides, there was an observable dorsal aorta enlargement on control sides as well (Fig. 3D). In the most extreme examples (67%; Table 2 and Fig. 3D), areas normally devoid of angioblasts were completely vascularized. In other cases (33%; Fig. 3C) the number of blood vessel branches crossing these areas adjacent to the bead increased (Table 2). In addition to being more numerous, the branches on the bead-implanted sides were larger than those on the control sides (Fig. 3C). The effect ex- 409 VEGF AND VASCULAR PATTERN FORMATION TABLE 2. Changes in the QH-1- stained areas resulting from VEGF delivery in six-somite stage embryos Bead location n Sidea Stained area fraction (%)b Increased branching Lost avascular areas First-formed somite Middle somite 11 Control Lc57.9 ⫾ 5.8 Rd62.5 ⫾ 5.0 Last-formed somite 18 Bead 81.7 ⫾ 3.2 Control 68.6 ⫾ 3.1 55% (6/11) 45% (5/11) 12 Bead 80.3 ⫾ 3.5 Control 77.0 ⫾ 2.7 33% (6/18) 67% (12/18) Bead 77.8 ⫾ 3.1 8% (1/12) 92% (11/12) a Measurements were made on (bead)-implanted and (control) contralateral sides. Data represent the mean stained area fraction (expressed as a percent ⫾ SEM: Standard Error of the Mean). c Mean stained area fraction for beads implanted on the left side (⫾ SEM: Standard Error of the Mean). d Mean stained area for beads implanted on the right side (⫾ SEM: Standard Error of the Mean). b TABLE 3. Analysis of the effect of ectopic VEGF at three rostral-caudal positions in six-somite stage embryos* Bead location Bead sidea n Comparison of VEGF-delivery sidesb Bead-implanted vs. contralateral sidesc Bead-implanted sidesd Contralateral sidese First-formed somite Left 6 Right 5 Middle somite Left 10 P ⬍ 0.01 Right 8 Last-formed somite Left 2 P ⫽ NS Right 10 P ⫽ NS P ⬍ 0.01 P ⫽ NS P ⬍ 0.01 P ⫽ NS P ⬍ 0.01 P ⫽ NS P ⬍ 0.01 P ⬍ 0.01 P ⫽ NS P ⫽ NS P ⬍ 0.01 P ⬍ 0.01 *For all analyses, embryos with beads implanted on left and right sides were combined if there was no statistical difference between the sides of bead implantation (P ⬎ 0.01). The side of an embryo where a bead was implanted. b Comparison of beads implanted on left or right sides of embryos. c Comparison between bead-implanted and contralateral sides within the same embryo. d Comparison of bead-implanted sides in embryos with control (PBS) or VEGF beads e Comparison of contralateral sides in embryos with control (PBS) or VEGF beads. NS, not signiﬁcant. Statistically signiﬁcant, P ⬍ 0.01. a tended 50 –75 m in both rostral and caudal directions from the bead sites. Following quantiﬁcation, the extent of changes in the vasculature was the same regardless of which side the bead was implanted on (Table 2). Therefore, all embryos with a bead implanted in the middle somite were combined for analysis. Implantation of VEGF beads resulted in an increased QH-1-stained area (Table 2), which was statistically signiﬁcant when experimental and control sides of the same embryo were compared (Table 2). The mean QH-1 stained area on the experimental VEGF sides was 80.3%, while on the control sides it was 68.6%. There was no observed effect resulting from control beads implanted in the middle somite (Table 1). However, there was a signiﬁcant difference between the experimental sides of the embryos with control or VEGF beads, as well as between their contralateral sides (Table 3). The effect of VEGF beads was also signiﬁcant when experimental and control sides were compared within the same embryo. For bead-implanted sides, signiﬁcance (Table 3) indicated that changes in the vascular pattern resulted from VEGF on the beads, rather than the beads themselves or the heparin bound to them. The effect on control sides of embryos can be observed by comparing unoperated sides in Fig. 3D and B. A control bead (Fig. 3B) did not affect the vascular pattern, so the vascular pattern on both sides of the embryo is essentially equivalent. With VEGF beads (Fig. 3C and D), the number and size of blood vessel branches increased on unoperated sides, so more anastomosing connections abnormally bridged the gap between the dorsal aorta and the lateral capillary plexus. With beads in the middle somite, fewer embryos (28%, 5/28) had observable effects on control sides than when the beads were implanted in the lastformed somite (50%; Table 2). A similar effect was observed with beads implanted in embryos with seven somite pairs (data not shown). Effect of Ectopic VEGF on Sprouting From the Dorsal Aortae Finally, to study the role of VEGF on sprouting from a lumenized dorsal aorta, beads were implanted into the ﬁrst-formed somite (Fig. 4A). At this anatomical level, the dorsal aorta ﬁrst becomes an endothelial cell tube. Intersomitic arteries begin to branch from this tube at the 8S stage. As with other anatomic locations, there were no observable effects from control beads in the ﬁrst-formed somite (Table 1; Fig. 4B). Normally avascular areas were traversed by only small-caliber blood vessels on the experimental and control sides of the embryos. Unlike other 410 FINKELSTEIN AND POOLE many embryos, the size of the dorsal aorta also increased (compare the embryo in Fig. 4C to the embryo in Fig. 4B). Unlike beads implanted at other locations, there was a signiﬁcant difference between the effect of VEGF on the left and right sides of the embryo (Table 2). Therefore, subsequent analyses were performed separately for embryos with beads implanted on the left or right. With VEGF beads on the left, the mean area fraction stained with QH-1 was 81.7% on the experimental side and 57.9% on the control side (P ⬍ 0.01; Table 2). With VEGF beads on the right, the mean area fraction stained with QH-1 was 63.6% on experimental sides and 62.5% on control sides (NS; Table 1). VEGF beads on the left resulted in signiﬁcant changes in QH-1-staining, while beads on the right did not (Table 2). Similar effects were observed for 7S embryos with beads implanted at the level of a rostral somite (somite VI/7 or VII/7; data not shown). Ectopic VEGF Delivery Results in an Enlarged, Patent Dorsal Aorta Fig. 4. Ectopic VEGF delivery at the level of the ﬁrst-formed somite. (B) Control or (C and D) VEGF beads were implanted in the ﬁrst-formed somite of 6S embryos, followed by a 7-hr incubation. All embryos were stained as whole mounts with the QH-1 monoclonal antibody, followed by confocal microscopy. A: Bead implantation in the ﬁrst-formed somite (dark somite) is shown schematically. B: An embryo with a control bead, ﬁxed at the 11S stage. There is a normal vascular pattern on both the bead-implanted and contralateral sides of this embryo. C: An embryo with a VEGF bead, ﬁxed at the 10S stage. Note the increased vascularization only on the bead-implanted side. D: An embryo with a VEGF bead, ﬁxed at the 9S stage. There is an increased vascularization only on the bead-implanted side, with numerous blood vessel branches between the dorsal aorta and capillary plexus. These are ventral views, with the rostral end up. Asterisks indicate bead locations. AVA, avascular area; DA, dorsal aortae; NT, neural tube; CP, capillary plexus; SOM, somites. The scale bar in D is representative of 100 m and applies to B–D. To further analyze the effect of VEGF on the developing vasculature, transverse sections of embryos with VEGF beads were stained with QH-1. On the experimental sides, the dorsal aorta was enlarged compared to the control sides (Fig. 5A and B). The dorsal aorta increased in size two- to threefold when VEGF beads were implanted in the middle somite (compare the left and right sides of Fig. 5A and B). The increase in dorsal aorta girth corresponded to the loss of normal avascular areas observed in whole mounts. The effect was observed for several sections from the bead in a rostral-caudal direction indicating that VEGF may diffuse several cell diameters. Fusion between the normal dorsal aortae and the capillary plexus may result in the observed dorsal aorta enlargement. In embryos with VEGF beads (Fig. 5A–C), the small vessels of the capillary plexus are absent, compared to normal embryos (Fig. 5D). From transverse sections, it can be concluded that delivery of ectopic VEGF does not produce a solid mass of QH-1-stained cells, but rather a large blood vessel with lumen. VEGF Induces Changes in the Vasculature After Short Incubation Times Experiments were carried out to determine the time required for VEGF to affect the vasculature. Four-hour incubations with VEGF resulted in an effect similar to that observed after 7-hr incubations. Ectopic VEGF resulted in increased density of blood vessel branches or vascularization of normally avascular areas in approximately 40% of the samples (Fig. 6A and B). For other samples (Fig. 6C), no effect was observed after 4 hr. DISCUSSION delivery locations, there was no signiﬁcant difference between the control sides of the embryos with VEGF or control beads implanted in the ﬁrst somite. As with the VEGF beads at other locations, beads implanted in the ﬁrst-formed somite of embryos with six somite pairs resulted in morphological changes (Fig. 4B and D). The most dramatic effect (45%) was complete vascularization lateral to the dorsal aorta in areas normally devoid of large vessels (note the fusion of DA and CP in Fig. 5C). In other examples (55%), the density of blood vessel branches increased near bead locations (Fig. 4D). In Multiple stages of vascular development occur simultaneously along the rostral-caudal axis of an embryo (Cofﬁn and Poole, 1988; Poole and Cofﬁn, 1989, 1991; Poole et al., 2001). To determine the effect of ectopic VEGF on these stages, we implanted heparin chromatography beads soaked in recombinant human VEGF165 (VEGF beads) in three rostral-caudal locations of quail embryos with six somite pairs. To examine the effect of ectopic VEGF on initial angioblast cohesion, we implanted VEGF beads in the last-formed somite. Ectopic VEGF resulted in increased vascularization (apparently both vasculogenesis VEGF AND VASCULAR PATTERN FORMATION 411 Fig. 5. Ectopic VEGF results in an enlarged lumenized dorsal aorta. VEGF beads were implanted into quail embryos with six or seven somite pairs, followed by 7-hr incubations. Representative 10-m parafﬁn sections from embryos with VEGF beads stained with QH-1 are shown. A: A bead was implanted in somite 3 at the 6S stage and ﬁxed at the 12S stage. This section is 20 m caudal to the last with a visible bead. The area occupied by the bead is marked by a large arrow. B: A second embryo with a VEGF bead in somite 3 at the 6S stage, ﬁxed at the 11S stage. C: A VEGF bead was implanted in somite 2 at the 7S stage, and the embryo was ﬁxed at the 13S stage. D: A normal 12S stage embryo without a bead. Note the lateral blood vessels of the capillary plexus. In all sections, note the larger dorsal aorta on the bead-implanted side compared to the contralateral side. The paired dorsal aortae in panel D (normal embryo) are of uniform size. Asterisks indicate bead locations. S, somite; NT, neural tube; DA, dorsal aorta; CP, capillary plexus. The scale bars in C and D represent 100 m. Fig. 6. Four-hour incubations with ectopic VEGF affected the vascular pattern. Embryos in panels A–C were incubated for 4 hr following VEGF bead implantation. All embryos were stained as whole mounts with the QH-1 monoclonal antibody, followed by confocal microscopy. A: A VEGF bead was implanted in somite 4 at the 7S stage, and the embryo was ﬁxed with nine somites. There was an increased vascularization on the bead-implanted side compared with the contralateral side. B: A VEGF bead was implanted in somite 2 at the 7S stage, and the embryo was ﬁxed with nine somites. This bead resulted in blood vessel branches developing between the dorsal aorta and capillary plexus, which did not occur on the contralateral side. C: A VEGF bead was implanted in somite 4 at the 7S stage, and the embryo was ﬁxed with 10 somites. In this example, there was little effect from VEGF after 4 hr. These are ventral views, with the rostral end up. Asterisks indicate bead locations. AVA, avascular area; DA, dorsal aortae; NT, neural tube; CP, capillary plexus; SOM, somites. The scale bar in C is representative of 100 m and applies to A–C. and angiogenesis) lateral to the dorsal aortae. Changes in the vascular pattern were signiﬁcant on the bead-implanted and control sides of the embryos. The signiﬁcant change on the control sides suggests that VEGF from beads diffuses across the embryonic midline. VEGF beads were implanted in the middle somite so that we could examine the effect of VEGF on formation of a lumenized dorsal aorta from a solid endothelial cell cord. In this case, there was a signiﬁcant increase in vascular density lateral to the dorsal aortae. In some samples, areas normally devoid of vessels became ﬁlled with QH-1 stained tissue. In other samples, numerous blood vessel sprouts crossed normally avascular areas. Finally, we implanted beads in the ﬁrst-formed somite to examine the 412 FINKELSTEIN AND POOLE effect of VEGF on sprouting from the patent dorsal aorta. At this level, the major effect was an increased number of vessel sprouts between the capillary plexus and dorsal aorta. There were signiﬁcant effects of VEGF on control sides of embryos with VEGF beads in the middle or lastformed somites. This suggests that VEGF diffuses between bead-implanted and control sides of the embryo. A signiﬁcant effect on the vascular pattern was observed on only the left side of embryos with VEGF beads in the ﬁrst-formed somite. At all three anatomical levels, ectopic VEGF resulted in hypervascularization to different degrees, especially between the dorsal aorta and capillary plexus. The degree of VEGF effects along the rostralcaudal axis may be explained by the differentiation state of the somites in which the beads were implanted, or by the response of mesodermal cells to ectopic VEGF. Somites were used for bead-implant sites because they give rise to a relatively low number of angioblasts (Wilting et al., 1995) and closely appose the developing dorsal aortae. The developmental stage of the somites could affect the diffusion of VEGF in all directions from implanted beads. The last-formed somite pair has not yet developed an epithelial layer surrounding the somitocoel (Hirsinger, et al., 2000). Without this epithelial layer, VEGF could diffuse from implanted beads more readily than at other rostral-caudal levels. More mature somites become surrounded by a layer of extracellular matrix (ECM) (Hirsinger et al., 2000), to further reduce diffusion from beads. At the level of the ﬁrst-formed somite, further differentiation takes place (Hirsinger et al., 2000) and there are tighter cellular connections and additional matrix. This additional matrix may minimize diffusion of VEGF from the beads, and explain the reduced diffusion of VEGF at this axial level. The different effects of VEGF at three axial locations could also be explained by the response of angioblasts to VEGF. VEGF beads at all three axial locations studied resulted in signiﬁcant increases in QH-1 labeling when bead-implanted sides of embryos with VEGF or control beads were compared. At the level of the ﬁrst-formed somite, angioblasts on the right side of the embryo elicited no observable effect in response to ectopic VEGF. With beads implanted in the middle or last-formed somites, there was no difference in the effect of VEGF beads implanted on either the left or right side. At the level of the ﬁrst-formed somite, signiﬁcant effects of ectopic VEGF only on the left side may be explained by asymmetrical development of the heart just rostral to this somite (Sugi and Markwald, 1996). It was previously demonstrated that a higher level of QH-1 on the right side corresponds to more endocardial precursor cells on that side (Cofﬁn and Poole, 1991; Sugi and Markwald, 1996). As such, angioblasts at this level may be precommitted to the endocardial lineage and may not be capable of being incorporated in the dorsal aorta in response to ectopic VEGF. The three stages of dorsal aorta development involve distinct cellular behaviors of angioblasts along the rostralcaudal axis, including angioblast cohesion and migration, cellular changes of lumen formation, and sprouting angiogenesis. Ectopic VEGF has effects at all three rostralcaudal levels, which suggests that it is active in three distinct cellular behaviors during vascular development. Activity in early embryos contrasts with that in older embryos, in which VEGF induces edema (Flamme et al., 1995b; Schmidt and Flamme, 1998). Thus, VEGF effects may be reﬂective of the developmental stage of the vasculature, with edema occurring predominantly in more mature blood vessels. Morphologically, the number of QH-1 positive cells increased near areas of ectopic VEGF delivery. The area lateral to the dorsal aorta, which normally has a lower level of QH-1 staining compared to the remainder of the embryo, became highly vascularized, and the number and size of blood vessel branches increased. In cross-section, an enlarged dorsal aorta with a lumen was observed, which is suggestive of an expanded blood vessel rather than a mass of QH-1-positive cells formed by vasculogenesis. This may have resulted from an increase in angioblast numbers through induction, proliferation, or chemotaxis from the lateral plate mesoderm, an area rich in angioblasts (Ferrara, 1999; Zachary and Gliki, 2001). The observed effects are probably not a result of angioblast induction, since it has been demonstrated that VEGF has a different effect on the vasculature compared to ﬁbroblast growth factor-2 (FGF2), which can induce an angioblast phenotype (Cox and Poole, 2000; Poole et al., 2001). FGF-2 induces an angioblast phenotype from mesodermal tissues that normally do not have large numbers of angioblasts (Cox and Poole, 2000). Following induction, angioblasts are characterized by their expression of VEGF receptors, which must be present for VEGF activity (Eichmann et al., 1993; Yamaguchi et al., 1993). Therefore, angioblast induction occurs at an earlier developmental stage than does VEGF activity. Finally, ectopic VEGF resulted in changes in the vascular pattern after very short time periods, with similar results observed following 4- or 7-hr incubations. This suggests that VEGF does not upregulate angioblast and endothelial cell proliferation, since the cell cycle time in embryos with six or seven somite pairs is approximately 10 hr (Sanders et al., 1993). More likely, VEGF acts by mediating chemotaxis of angioblasts and endothelial cells during embryonic vascular development, and perturbing the balance between cell-cell cohesion and cell-substrate adhesion. Data (Table 3) indicate that ectopic VEGF can diffuse across the embryonic midline to affect the contralateral side. There is evidence of growth factor diffusion from beads implanted in embryos (Storey et al., 1998). Increasing the growth factor concentrations used to soak beads led to diffusion distances of 3– 4 cell diameters with 50 g/ml growth factor (Storey et al., 1998). The same is expected to hold true for VEGF. By image analysis, it was determined that observed effects were not due to embryo manipulations or the beads themselves. There were no observable effects on the vasculature resulting from control beads, indicating that heparin on beads does not remove endogenous VEGF from embryonic tissues. To investigate the mechanisms by which VEGF can affect the developing vasculature, focal delivery of VEGF on chromatography beads is superior to previous methods used in microinjection studies (Drake and Little, 1995). Combined with focal delivery of VEGF inhibitors, such as truncated forms of VEGF receptors (Drake et al., 2000), bead implantation will be useful for studying the mechanism of VEGF action in distinct cellular processes during vascular development. The embryos studied differed by a small developmental window, but VEGF delivery at different anatomical levels tested a broad window of activity in vascular development. The effect of ectopic VEGF was rapid (observed in as little VEGF AND VASCULAR PATTERN FORMATION as 4 hr), which suggests that a mechanism other than increased proliferation is involved. Ectopic VEGF can disrupt the normal vascular pattern, indicating that a critical level of VEGF is required for normal vascular development. This correlates with previous studies in mouse embryos, in which decreased (Carmeliet et al., 1996; Ferrara et al., 1996) or increased (Miquerol et al., 2000) VEGF levels led to abnormal development. The present data indicate that the role of precisely controlled VEGF levels within the embryo may be to determine which areas become vascularized and which remain avascular, by inﬂuencing angioblast migration and adhesion. ACKNOWLEDGMENTS We thank Dr. Douglas Robertson (SUNY Upstate Medical University, Syracuse, NY) for assistance with statistical analysis, and Dr. Christopher Cox (University of Arizona) for helpful discussion. The QH-1 monoclonal antibody developed by Pardanaud et al. (1987) was obtained from the Developmental Hybridoma Studies Bank under the auspices of the NICHD, and maintained by the Department of Biological Sciences, University of Iowa, Iowa City, Iowa. LITERATURE CITED Aitkenhead M, Christ B, Eichmann A, Feucht M, Wilson DJ, Wilting J. 1998. Paracrine and autocrine regulation of vascular endothelial growth factor during tissue differentiation in the quail. Dev Dyn 212:1–13. Breier G, Clauss M, Risau W. 1995. Coordinate expression of vascular endothelial growth factor receptor-1 (ﬂt-1) and its ligand suggests a paracrine regulation of murine vascular development. Dev Dyn 204:228 –239. Carmeliet P, Ferreira 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, Nagy A. 1996. Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature 380:435– 439. Christ B, Ordahl CP. 1995. Early stages of chick somite development. Anat Embryol (Berl) 191:381–396. Cofﬁn JD, Poole TJ. 1988. Embryonic vascular development: immunohistochemical identiﬁcation of the origin and subsequent morphogenesis of the major vessel primordia. Development 102:735– 749. Cofﬁn JD, Poole TJ. 1991. Endothelial cell origin and migration in embryonic heart and cranial blood vessel development. Anat Rec 231:383–395. Conway EM, Collen D, Carmeliet P. 2001. Molecular mechanisms of blood vessel growth. Cardiovasc Res 49:507–521. Cox CM, Poole TJ. 2000. Angioblast differentiation is inﬂuenced by the local environment: FGF-2 induces angioblasts and patterns vessel formation in the quail embryo. Dev Dyn 218:371–382. Crossley PH, Minowada G, MacArthur CA, Martin GR. 1996. Foles for FGF8 in the induction, initiation and maintenance of chick limb development. Cell 84:127–136. Darland DC, D’Amore PA. 1999. Blood vessel maturation: vascular development comes of age. J Clin Invest 103:157–158. Dieterlen-Lievre F, Pardanaud L. 1998. Ontogeny of the endothelial network analyzed in the avian model. In: Little CD, Mironov V, Sage EH, editor. Vascular morphogenesis: in vivo, in vitro, in mente. Boston: Birkhauser. p 35–50. Drake CJ, Davis LA, Little CD. 1992. Antibodies to beta 1-integrins cause alterations of aortic vasculogenesis, in vivo. Dev Dyn 193:83– 91. Drake CJ, LaRue A, Ferrara N, Little CD. 2000. VEGF regulates cell behavior during vasculogenesis. Dev Biol 224:178 –188. Drake CJ, Little CD. 1995. Exogenous vascular endothelial growth factor induces malformed and hyperfused vessels during embryonic neovascularization. Proc Natl Acad Sci USA 92:7657–7661. 413 Eichele G, Tickle C, Alberts BM. 1984. Microcontrolled release of biologically active compounds in chick embryos: beads of 200-microns diameter for the local release of retinoids. Anal Biochem 142:542–555. Eichmann A, Marcelle C, Breant C, LeDouarin NM. 1993. Two molecules related to the VEGF receptor are expressed in early endothelial cells during avian embryonic development. Development 42:33– 48. Eichmann A, Marcelle C, Breant C, LeDouarin NM. 1996. Molecular cloning of Quek 1 and 2, two quail vascular endothelial growth factor (VEGF) receptor-like molecules. Gene 174:3– 8. Ferrara N, Carver-Moore K, Chen H, Dowd M, Lu L, O’Shea KS, Powell-Braxton L, Hillan KJ, Moore MW. 1996. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 380:439 – 442. Ferrara N. 1999. Molecular and biological properties of vascular endothelial growth factor. J Mol Med 77:527–543. Flamme I, Risau W. 1992. Induction of vasculogenesis and hematopoiesis in vitro. Development 116:435– 439. Flamme I, Breier G, Risau W. 1995a. Vascular endothelial growth factor (VEGF) and VEGF receptor 2 (ﬂk-1) expressed during vasculogenesis and vascular differentiation in the quail embryo. Dev Biol 169:699 –712. Flamme I, Reutern M, Drexler HCA, Syed-Ali S, Risau W. 1995b. 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. Fong GH, Rossant J, Gertsenstein M, Breitman ML. 1995. Role of the Flt-1 receptor tyrosine kinase in regulating the assembly of vascular endothelium. Nature 376:66 –70. Fong GH, Zhang L, Bryce DM, Peng J. 1999. Increased hemangioblast commitment, not vascular disorganization, is the primary defect in ﬂt-1 knock-out mice. Development 126:3015–3025. Hamburger V. 1960. A manual of experimental embryology. Chicago: University of Chicago Press. 221 p. Hamburger V, Hamilton H. 1951. A series of normal stages in the development of the chick embryo. J Morphol 88:49 –92. Healy AM, Morgenthau L, Zhu X, Farber HW, Cardoso WV. 2000. VEGF is deposited in the subepithelial matrix at the leading edge of branching airways and stimulates neovascularization in the murine embryonic lung. Dev Dyn 219:341–352. Hirsinger E, Jouve C, Dubrulle J, Pourquie O. 2002. Somite formation and patterning. Int Rev Cytol 198:1– 65. Krah K, Mironov V, Risau W, Flamme I. 1994. Induction of vasculogenesis in quail blastodisc-derived embryoid bodies. Dev Biol (Orl) 164:123–132. Miquerol L, Langille BL, Nagy A. 2000. Embryonic development is disrupted by modest increases in vascular endothelial growth factor gene expression. Development 127:3941–3946. Packard DS, Cox CM, Poole TJ. 2000. Improved techniques for avian embryo culture, somite cell culture and microsurgery. In: Tuan RS, Lo CW, editors. Developmental biology protocols. Totowa, NJ: Humana Press. p 185–199. Pardanaud L, Altmann C, Kitos P, Dieterlen-Lievre F, Buck CA. 1987. Vasculogenesis in the early quail blastodisc as studied with a monoclonal antibody recognizing endothelial cells. Development 100: 339 –349. Pardanaud L, Yassine F, Dieterlen-Lievre F. 1989. Relationship between vasculogenesis, angiogenesis and hematopoiesis during avian ontongeny. Development 105:473– 485. Poole TJ, Cofﬁn JD. 1988. Developmental angiogenesis: quail embryonic vasculature. Scanning Microsc 2:443– 448. Poole TJ, Cofﬁn JD. 1989. Vasculogenesis and angiogenesis: two distinct morphogenetic mechanisms establish embryonic vascular pattern. J Exp Zool 251:224 –231. Poole TJ, Cofﬁn JD. 1991. Morphogenetic mechanisms in avian vascular development. In: Feinberg RN, Sherer GK, Auerbach R, editors. The development of the vascular system. Basel, Switzerland: S. Karger AG. p 25–36. 414 FINKELSTEIN AND POOLE Poole TJ, Finkelstein EB, Cox CM. 2001. The role of FGF and VEGF in angioblast induction and migration during vascular development. Dev Dyn 220:1–17. Sanders EJ, Varedi M, French AS. 1993. Cell proliferation in the gastrulating chick embryo: a study using BrdU incorporation and PCNA localization. Development 118:389 –399. Schmidt C, Christ B, Patel K, Brand-Saberi B. 1998. Experimental induction of BMP-4 expression leads to apoptosis in the paraxial and lateral plate mesoderm. Dev Biol 202:253–263. Schmidt M, Flamme I. 1998. The in vivo activity of vascular endothelial growth factor isoforms in the avian embryo. Growth Factors 15:183–197. Senger DR, Galli SJ, Dvorak AM, Perruzzi CA, Harvey VS, Dvorak HF. 1983. Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites ﬂuid. Science 219:983–985. Shalaby F, Rossant J, Yamaguchi TP, Gertsenstein M, Wu XF, Breitman ML, Schuh AC. 1995. Failure of blood-island formation and vasculogenesis in Flk-1-deﬁcient mice. Nature 376:62– 66. Stockdale FE, Nikovits Jr W, Christ B. 2000. Molecular and cellular biology of avian somite development. Dev Dyn 219:304 –321. Storey KG, Goriely A, Sargent CN, Brown JM, Burns HD, Abud HM, Heath JK. 1998. Early posterior neural tissue is induced by FGF in the chick embryo. Development 125:473– 484. Sugi Y, Markwald RR. 1996. Formation and early morphogenesis of endocardial endothelial precursor cells and the role of endoderm. Dev Biol 175:66 – 83. Tallquist MD, Soriano P, Klinghoffer RA. 1999. Growth factor signaling pathways in vascular development. Oncogene 18:7917–7932. Wilting J, Brand-Saberi B, Huang R, Zhi Q, Kontges G, Ordahl CP, Christ B. 1995. Angiogenic potential of the avian somite. Dev Dyn 202:165–171. Yamaguchi TP, Dumont DJ, Conlon RA, Breitman ML, Rossant J. 1993. Flk-1, an ﬂt-related receptor tyrosine kinase is an early marker for endothelial cell precursors. Development 118:489 – 498. Yue X, Tomanek RJ. 2001. Effects of VEGF(165) and VEGF(121) on vasculogenesis and angiogenesis in cultured embryonic quail hearts. Am J Physiol Heart Circ Physiol 280:H2240 –H2247. Zachary I, Gliki G. 2001. Signaling transduction mechanisms mediating biological actions of the vascular endothelial growth factor family. Cardiovasc Res 49:568 –581.