| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Embryo |
Center for Vascular Biology,3
Departments of Physiology,4
Genetics and Developmental Biology,5 University of Connecticut Health Center, Farmington, Connecticut 06030
Samuel Lunenfeld Research Institute,6 Mount Sinai Hospital, Toronto, Ontario, Canada M5G 1X5
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
developmental biology, early development, embryo, growth factors, kinases
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In mouse embryos, endothelial differentiation closely follows gastrulation, and endothelial progenitors are generated abundantly in the yolk sac mesoderm (mesenchyme) of mid- to late gastrulation stage embryos (E7.3–7.5) [5, 6]. The yolk sac mesenchyme is an excellent in vivo model for the study of endothelial cell differentiation for two reasons. First, yolk sac mesenchyme is where the earliest endothelial differentiation occurs in mice, and, second, its anatomic structure is very simple. In the prospective yolk sac of gastrulation stage embryos, the layer of multipotent mesenchymal cells is sandwiched between two cellular monolayers, the endoderm and the mesothelium [7]. Thus, the identification of the mesenchymal cell population, from which endothelial progenitors arise, is a relatively simple task.
To further investigate the role of VEGF-A in mouse development, we took advantage of a mutant mouse line whose Vegfa gene is mutated to a hypomorphic allele, Vegfalo [8]. The Vegfalo allele was created by inserting LacZ sequence into the 3' untranslated region of the Vegfa gene, leading to an inadvertent splicing event near the 3' end of the Vegfa open reading frame [8]. The resulting VEGF-A has significantly reduced activity because the last six amino acid residues are replaced by a stretch of unrelated amino acid residues [8]. Vegfalo/lo embryos die at E9.0 and display mutant phenotypes that are only slightly less severe than those of Vegfa null mutants, indicating that the loss of VEGF-A activity is very significant [8]. Nonetheless, Vegfa+/lo mice are viable and fertile, which allows the production of homozygous embryos by breeding. In this work, we made a surprising observation that endothelial specification is not significantly affected by a significant VEGF-A deficiency. Instead, impaired vascular development in VEGF-A-deficient mice is largely due to defects in two developmental events: the development of the yolk sac mesoderm and angiogenic sprouting from existing blood vessels.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Procedures for mouse handling were approved by the Animal Care Committee at the University of Connecticut Health Center in accordance with Animal Welfare Assurance. The initial derivations of Flt-1+/-, Flk-1+/-, and Vegfa+/lo mice have been reported [3, 4, 8]. All stocks were maintained by crossing heterozygous males with CD1 females, and heterozygous progenies were identified by previously established methods [3, 4, 8]. Since the targeted allele of Flt-1 carries a lacZ knock-in sequence, it will be also referred to as Flt-1lacZ allele when the use of its LacZ sequence is emphasized. Vegfa+/lo/Flt-1+/- double heterozygotes were obtained by crossing Flt-1+/- and Vegfa+/lo mice. The progenies resulting from these crosses were screened for Flt-1+/- and Vegfa+/lo alleles by previously published protocols [3, 8].
Embryos
To estimate the developmental stages of resulting embryos, mice were inspected for vaginal plugs in the morning. Noon of the day when a vaginal plug was observed was considered as 0.5 days of gestation. Wild-type, Vegfa+/lo and Vegfalo/lo embryos were obtained by Vegfa+/lo x Vegfa+/lo mating. Flt-1+/-, Vegfa+/lo/Flt-1+/- and Vegfalo/lo/Flt-1+/- embryos were obtained by Vegfa+/lo/Flt-1+/- x Vegfa+/lo mating. Flk-1+/- or Flk-1-/- embryos were obtained by Flk-1+/- x Flk-1+/- crosses as reported by Shalaby et al. [4].
At 7.0 to 8.7 days postcoitus (d.p.c.), pregnant females were killed by carbon dioxide, and embryos were dissected. After the ectoplacental cones were removed for cultures (see Genotype Analysis of Embryos), individual embryos (including embryo proper and extraembryonic membranes) were placed in 24-well dishes, fixed with appropriate fixatives (see Histology), and subjected to either X-gal staining or immunohistochemical (IHC) staining. Stained embryos were photographed and fixed with 3.7% formaldehyde overnight at 4°C before being processed for paraffin sections.
Genotype Analyses of Embryos
Because of the small sizes of embryos used in this study, DNA samples used for genotype analyses were obtained from trophoblasts removed from individual embryos and cultured for extended periods of time. Briefly, ectoplacental cones from individual embryos were cultured for 12–14 days in DMEM supplemented with glutamine, 15% fetal bovine serum (FBS), and penicillin-streptomycin (PS) (DMEM, FBS, and PS all from Invitrogen, Carlsbad, CA). During this process, maternal blood cells gradually die, whereas trophoblasts continue to grow. Cultured trophoblasts were then rinsed three times with phosphate-buffered saline (PBS), lysed with lysis buffer (100 mM Tris-Cl, pH 7.5, 150 mM NaCl, 2 mM EDTA, 0.1% Triton X100, 50 µl per sample) containing 200 µg/ml proteinase K (Invitrogen) and incubated at 37°C overnight in a humidified incubator. DNA samples were diluted with 4 volumes of H2O, and proteinase K was inactivated by heating at 94°C for 10 min. DNA samples were then used for PCR analysis for Vegfalo or Flt-1+/- alleles according to protocols established earlier [3, 8].
Histology
To perform X-gal staining, embryos were fixed in 0.2% glutaraldehyde (Sigma, St. Louis, MO) rinsed several times in PBS, and stained in 0.1 M phosphate buffer (pH 7.4) containing X-gal (5-bromo-4-chloro-3-indoxyl-beta-D-galactopyranoside, Biosynth International Inc., Naperville, IL). Detailed conditions for X-gal staining have been described previously [9]. To obtain paraffin sections of X-gal-stained embryos, the specimens were dehydrated with increasing concentrations of ethanol. The specimens were further dehydrated with three changes of toluene. To avoid loss of embryos due to their small sizes, all embryos were processed manually in clear-glass vials. With the aid of a dissection microscope, dehydrated embryos were properly oriented and embedded in paraffin. Specimens were cut at 5 µm and counterstained with nuclear fast red.
For IHC staining with rat anti-mouse CD31 monoclonal antibody (clone Mec13.3, BD Biociences-Pharmingen, San Diego, CA) or rat anti-mouse Flk-1 monoclonal antibody (clone Avas 12
1, BD Biosciences-Pharmingen), a procedure previously reported by Sato et al. [10] was adapted for small embryos. Briefly, embryos were fixed for 3 h in 4% paraformaldehyde (Sigma) at 4°C, rinsed with PBS several times, and dehydrated with increasing concentrations of methanol. Embryos were treated with 5% hydrogen peroxide (Sigma) for 5 h and dehydrated. To block nonspecific antibody binding, embryos were rocked overnight in PBS containing 2% dry milk, 0.1% Triton X100 (PBSMT), supplemented with 3% normal goat serum. Primary antibody was diluted in PBSMT to a final concentration of 5 µg/ml, and embryos were incubated at 4°C overnight in PBSMT containing primary antibody (anti-CD31 or anti-Flk-1). The antibody solutions were removed under a dissection microscope, and embryos were rinsed several times with PBSMT. Embryos were then incubated overnight with F(ab)2 fragments of goat anti-rat IgG conjugated with horseradish peroxidase (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). After the embryos were rinsed extensively with PBSMT and PBS containing 0.1% Triton X100 and 0.1% BSA, they were developed in a cocktail containing 3,3' diaminobenzidine (DAB) (0.3 mg/ml), hydrogen peroxide (0.03%), and NiCl2 (0.5%) (DAB and NiCl2 were from Sigma).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To examine the impact of VEGF-A deficiency on endothelial differentiation, we took advantage of the Flt-1lacZ allele, whose expression specifically marks endothelial progenitors in the prospective yolk sac membrane [9]. We crossed Flt-1+/- and Vegfa+/lo mice and isolated resulting embryos at mid- to late gastrulation stages (E7.3–7.5). This stage of development was chosen for analysis because it is the stage when endothelial specification occurs. In Figure 1, we show that while Vegfa+/+/Flt-1+/- embryos exhibited abundant X-gal staining signal in the prospective yolk sac (Fig. 1B), X-gal staining in Vegfalo/lo/Flt-1+/- embryos was very weak (Fig. 1D). The staining intensity in Vegfa+/lo/Flt-1+/- embryos was also less than normal (Fig. 1C). Nevertheless, Vegfa+/lo/Flt-1+/- were viable and fertile, indicating that the remaining amount of endothelial cells was sufficient to maintain a functional vascular system in the yolk sac.
|
To confirm the results obtained by X-gal staining, we also performed immunohistochemical (IHC) staining with anti-Flk-1 antibody. During the early phase of gastrulation, Flk-1 is widely expressed in the yolk sac mesoderm, but its expression quickly becomes restricted to endothelial progenitors [4, 11]. By E7.5, only discrete clusters of cells fated to become blood islands are Flk-1 positive [4, 11]. The anti-Flk-1 IHC staining pattern in E7.5 embryos was consistent with the known expression pattern of Flk-1 (Fig. 1F), whereas no signal was detected when anti-Flk-1 antibody was omitted (Fig. 1E). These data confirmed that the anti-Flk-1 antibody used in this work was specific. In Figure 1G, we demonstrate that Vegfalo/lo embryos (left) have a significantly reduced level of anti-Flk-1 staining relative to the wild-type embryo (right), indicating that Vegfalo/lo embryos contain either a reduced amount of yolk sac mesoderm or a less number of endothelial progenitors or a combination of both. Overall, results obtained by anti-Flk-1 IHC staining were consistent with those from X-gal staining.
Next, we attempted to compare endothelial differentiation in normal and VEGF-A-deficient embryos by examining transverse sections of prospective yolk sacs. Instead of finding a significant difference in endothelial differentiation, however, we found that the most obvious defect in VEGF-A-deficient embryos was dramatically impaired development of the yolk sac mesenchyme (Fig. 2). In normal embryos (Vegfa+/+/Flt-1+/-), the mesenchyme formed an easily recognizable layer of tissue between the endoderm and the mesothelial layer (Fig. 2A). In Vegfa+/lo/Flt-1+/- embryos, the mesoderm layer was significantly thinner (Fig. 2B). In Vegfalo/lo/Flt-1+/- or Vegfalo/lo embryos, the loss of yolk sac mesenchyme was even more drastic. There, few cells were present between the endoderm and mesothelium (Fig. 2, C and D).
|
In the yolk sac, endothelial progenitors are derived from multipotent mesenchymal cells. Thus, it is not surprising that defective mesenchyme development may significantly hamper subsequent production of endothelial progenitors. Since normal and VEGF-A-deficient embryos had very different amounts of yolk sac mesenchyme, it was not feasible to directly compare the total number of endothelial progenitors. Therefore, we evaluated relative frequencies of endothelial differentiation from a given number of yolk sac mesenchymal cells. In histological sections that transverse the prospective yolk sac, endothelial progenitors can be identified as Flt-1(X-gal)-positive cells located within mesenchymal layer. Although some nonendothelial progenitors also express Flt-1 [12], they are not located in the prospective yolk sac or embryo proper and so do not interfere with our analyses. While the absolute numbers of mesenchymal cells and endothelial progenitors vary greatly from one genotype to another, results summarized in Table 1 demonstrate that the ratio between endothelial progenitors and mesenchymal cells remained relatively constant among different embryos. We conclude that a significant VEGF-A deficiency leads to defective gastrulation but does not directly affect endothelial differentiation.
|
VEGF-A May Mediate Gastrulation Through Flk-1 but Not Flt-1
The VEGF-A dependence of mesoderm development in the yolk sac suggests that a VEGF-A receptor is also important for the development of the yolk sac mesoderm. To see if the expression patterns of Flt-1 or Flk-1 are consistent with a role in mediating the development of yolk sac mesoderm, we performed X-gal staining of E7.0 Flt-1+/- or Flk-1+/- embryos (Flk-1+/- mice also carry a lacZ insert in the targeted Flk-1 allele). During gastrulation, the interaction between the embryonic ectoderm and endoderm leads to the formation of the so-called primitive streak at the posterior end of the embryo proper. Cells of the primitive streak are loosely distributed and highly migratory epiblasts, and some of them migrate into the extraembryonic membranes to form extraembryonic mesoderm [7].
We found that while Flt-1 expression was undetectable in the primitive streak, Flk-1 expression in the primitive streak was strong (Fig. 3, A and B). These results indicate that Flk-1, but not Flt-1, mediates epiblast migration into the prospective yolk sac. Flk-1 expression in the primitive streak was transient, and the vast majority of X-gal-positive cells were found in the yolk sac by E7.7 (Fig. 3C), which is consistent with the fact that the gastrulation process ends at about E7.5. A role of Flk-1 in gasgtrulation is supported by Flk-1-/- phenotype. In Flk-1-/- embryos, X-gal-positive cells fail to reach the yolk sac but accumulate in the amnion [13]. A representative Flk-1-/- embryo at E7.7 is shown in Figure 3D to correlate Flk-1 expression in the primitive streak with its role in gastrulation.
|
Angiogenesis Is More Sensitive to VEGF-A Deficiency Than Vasculogenesis
Compromised mesenchyme development in the yolk sac explains why the development of the vascular bed in the yolk sac is more sensitive to VEGF-A deficiency than is the embryonic vasculature. On the other hand, the development of the embryonic vasculature is also partially defective in VEGF-A-deficient embryos. Although we have previously reported that embryonic vascular development was defective as a result of VEGF-A deficiency, detailed analysis was not performed to determine if vasculogenesis and angiogenesis had differential sensitivity to partial VEGF-A deficiency [8].
Therefore, we examined vasculogenesis and sprouting morphorgenesis (angiogenesis) in the headfold, where the two processes happen in different anatomic zones. In the headfolds of normal embryos, blood vessels first form by vasculogenesis in the head mesenchyme. Subsequently, these blood vessels extend toward the neuroepithelium and other tissues by angiogenesis. By E8.7, an extensive perineural vasculature had formed in the headfold of normal embryos (Fig. 4, A and C). In Vegfalo/lo embryos, blood vessels in the headfold failed to extend toward the neuroepithelium (Fig. 4, B and D). In the center area of the headfold, normal embryos also seemed to have more blood vessels than Vegfalo/lo mutants (Fig. 4, C and D). However, vascular beds shown in whole-mount images included not only blood vessels in the head mesenchyme but also those that had sprouted to invest other tissues overlaying the mesenchyme. To directly compare vascular development in head mesenchymes of normal and mutant embryos, we performed histological sections. We found that vascular development was apparently normal in the head mesenchyme of Vegfalo/lo embryos, although the perineural vasculature was essentially missing (Fig. 4F). These observations suggest that VEGF-A deficiency has little effect on vasculogenesis in the head mesenchyme but strongly affects angiogenic sprouting.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Vasculogenesis begins with differentiation of multipotent mesenchymal cells to form endothelial and hematopoietic progenitors [14–18]. As a result of VEGF-A deficiency, the number of multipotent mesenchymal cells is reduced. Thus, vascular abnormality in VEGF-A-deficient yolk sac is mostly a secondary defect to impaired gastrulation. Our data demonstrating that VEGF-A is important for gastrulation match those of Shalaby et al. [13], who reported a mesenchyme deficiency in Flk-1-/- yolk sacs. A similar lack of yolk sac mesoderm was not observed in Flt-1-/- embryos [3]. At E6.5, Vegfa is expressed in the visceral extraembryonic endoderm (corresponding to the prospective yolk sac endoderm at E7.0–7.5) [19]. By the onset of gastrulation (E7.0), Flk-1 is expressed in the primitive streak and the nascent mesoderm (Fig. 3B) [11]. In contrast, Flt-1 is not expressed in the primitive streak, suggesting that Flt-1 expression is activated only after cells in the primitive streak migrate into the prospective yolk sac.
Taken together, these data suggest that VEGF-A is a paracrine factor that positively regulates gastrulation by signaling through Flk-1. Since VEGF-A is initially expressed in the yolk sac endoderm but Flk-1 is expressed in the primitive streak and nascent mesoderm, it is reasonable to assume that a soluble form of VEGF-A diffuses from the endoderm to the primitive streak to activate Flk-1 signaling. However, further investigations are needed to determine precisely which VEGF-A isoform is important for its role in gastrulation and early vascular development. Similarly, it will also be important to learn whether haparan sulfate is important for VEGF-A functions in early embryos.
Our finding that endothelial specification is insensitive to VEGF-A deficiency raises the possibility that VEGF-A may not be the initial inducer of endothelial differentiation. To directly test this proposition, it is necessary to analyze embryos null for Vegfa expression. However, Vegfa-/ - embryos cannot be obtained because Vegfa+/- mice for the null mutation are nonviable [1, 2]. Even though "Vegfa-/-" embryos can be derived from Vegfa-/- ES cells by in vitro aggregation with tetraploid embryonic cells [2, 20], the yolk sac endoderm of resulting embryos also contain cells derived from tetraploid aggregation partners, which are Vegfa positive [20]. While it is technically challenging to obtain truly Vegfa-/- embryos, Flk-1-/- embryos are readily available because Flk-1+/- mice are viable. Using in vitro explant cultures of Flk-1-/- embryos, Schuh et al. [21] demonstrated that endothelial differentiation occurred in Flk-1-/- cultures, but Flk-1-/- endothelial cells could not be maintained during subsequent culturing. These data indicate that the VEGF-A/Flk-1 signaling pathway is not essential for endothelial specification but is critical for endothelial growth and survival.
Then, what signal initiates endothelial differentiation in normal embryos? While a definitive answer to this question remains elusive, some evidence suggests that basic fibroblast growth factor (FGF) may be involved in this process. In cultured quail blastodiscs, bovine FGF (bFGF) but not VEGF-A was able to induce endothelial differentiation [22]. In Xenopus animal caps, bFGF treatment was able to initiate endothelial specification [23]. Therefore, one possibility is that bFGF promotes endothelial specification, but VEGF-A is essential for continued endothelial differentiation once their developmental fate is determined. Further investigations will be necessary to definitively identify the initial signal that triggers the specification of a subset of mesenchymal cells to endothelial fate.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Guo-Hua Fong, Center for Vascular Biology, University of Connecticut Health Center, 263 Farmington Avenue, Farmington, CT 06030-3501. FAX: 860 679 1201; fong{at}nso2.uchc.edu
Received: 22 May 2003.
First decision: 11 June 2003.
Accepted: 16 July 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
