SEARCH

SEARCH BY CITATION

Keywords:

  • Flk-1;
  • VEGF;
  • sinusoid;
  • angiogenesis;
  • hepatoblasts;
  • liver morphogenesis

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Early morphogenesis of hepatic sinusoids was histochemically and experimentally analyzed, and the importance of VEGF-Flk-1 signaling in the vascular development was examined during murine liver organogenesis. FITC-gelatin injection experiments into young murine fetuses demonstrated that all primitive sinusoidal structures were confluent with portal and central veins, suggesting that hepatic vessel development may occur via angiogenesis. At 12.5–14.5 days of gestation, VEGF receptors designated Flk-1, especially their mature form, were highly expressed in endothelial cells of primitive sinusoidal structures and highly phosphorylated on their tyrosine residues. At the same time, VEGF was also detected in hepatoblasts/hepatocytes, hemopoietic cells, and megakaryocytes of the whole liver parenchyma. Furthermore, the addition of VEGF to E12.5 liver cell cultures significantly induced the growth and branching morphogenesis of sinusoidal endothelial cells. Therefore, VEGF-Flk-1 signaling may play an important role in the growth and morphogenesis of primitive sinusoids during fetal liver development. Developmental Dynamics 239:386–397, 2010. © 2009 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The vascular system is the first functional organ in the developing vertebrate embryo and its continuous growth is required for embryonic development and organogenesis. Endothelial cell–specific mitogen vascular endothelial growth factor-A (hereafter, VEGF) is a key mediator of normal and abnormal angiogenesis (Ferrara and Henzel,1989; Roskoski,2008; Shibuya,2006). Even the loss of a single allele of the VEGF gene results in defective vasculature and embryonic lethality in mice (Ferrara et al.,1996). VEGF binds and activates two tyrosine kinase receptors, Flt-1 (VEGFR-1) and Flk-1 (VEGFR-2) (de Vries et al.,1992; Shibuya,2006; Shibuya and Claesson-Welsh,2006; Shibuya et al.,1990), both of which are expressed in endothelial cells. These receptors regulate physiological as well as pathological angiogenesis. Flk-1 has strong tyrosine kinase activity, and transduces the major signals for angiogenesis (Olsson et al.,2006; Shibuya,2006). Flt-1 has much higher affinity than Flk-1, and plays a dual role, a negative role in angiogenesis in the embryo most likely by trapping VEGF, and a positive role in adulthood in a tyrosine kinase–dependent manner (Shibuya,2006). Mouse embryos homozygous for a targeted mutation in the Flt-1 locus formed endothelial cells in both embryonic and extra-embryonic regions, but assembled these cells into abnormal vascular channels and died in utero at mid-gestation stages (Fong et al.,1995). Those homozygous for a targeted mutation of the Flk-1 locus die in utero between 8.5 and 9.5 days of gestation, as a result of an early defect in the development of hemopoietic and endothelial cells (Shalaby et al.,1995).

During murine liver development, endothelial cells expressing Flk-1 receptors are detected very close to the liver primordium in early developmental stages (Matsumoto et al.,2001; Nonaka et al.,2007; Shiojiri and Sugiyama,2004), but their developmental origin remains to be demonstrated. The omphalomesenteric veins and common or posterior cardinal veins, which hepatic cords invade from the hepatic diverticulum, may generate primitive sinusoids (Shiojiri et al.,2006). The septum transversum mesenchyme or hemangioblasts colonizing the liver primordium might generate endothelial cells (Du Bois,1963; Gouysse et al.,2002; Severn,1972). It is also unknown whether the hepatic vascular system is confluent from early developmental stages, which may give insights into the type of hepatic vascular development, angiogenesis, or vasculogenesis. Differentiation of hepatic blood vessels, including portal veins, central veins, and sinusoids, already proceeds in the E11.5 liver in terms of connexin and Jagged1 expression (Shiojiri et al.,2006). Functional differentiation such as SE-1 antigen expression, which is one of sinusoidal endothelial cell–specific markers, takes place with straight channel formation after midgestational stages, which finally leads to adult-type sinusoid architecture (Morita et al.,1998; Yoshida et al.,2007). Although the importance of VEGF-Flk-1 signaling in the liver primordium morphogenesis has already been demonstrated by using Flk-1-deficient mice, which lack endothelial cells (Matsumoto et al.,2001), and VEGF is also required for growth and survival in neonatal mice (Gerber et al.,1999; Carpenter et al.,2005), it remains to be revealed when and what kinds of vascular development this signaling involves during fetal liver development. In Flk-1 mutant embryos, hepatic specification occurs, but liver morphogenesis fails prior to mesenchyme invasion (Matsumoto et al.,2001). It is not clear which cells produce VEGF, or how VEGF-producing cells localize in the fetal liver in the context of vascular development.

Takahashi and Shibuya (1997) reported that Flk-1 was initially synthesized as a 150-kDa protein and rapidly glycosylated to a 200-kDa intermediate form, and then further glycosylated to a mature 230-kDa protein expressed on the cell surface in rat endothelial cells and NIH/3T3 cells expressing Flk-1. Only the 230-kDa form is rapidly and transiently phosphorylated on tyrosine residues in the presence of VEGF. Immunobloting analyses for this processing of Flk-1 proteins may reveal when and how Flk-1 is activated during liver development.

In the present study, vascular development was analyzed during mouse liver development by using immunohistochemistry for Flk-1 and VEGF, immunoblotting of Flk-1 molecular species, FITC-gelatin injection experiments, and primary culture techniques. We obtained data supporting that hepatic sinusoids may emerge from the omphalomesenteric veins or common or posterior cardinal veins through their angiogenesis, and found that VEGF-Flk-1 signaling plays an important role in immature sinusoidal growth and morphogenesis.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Vascular Development During Hepatic Organogenesis

We visualized vascular development during mouse liver organogenesis by immunohistochemistry for PECAM-1, which is known as a marker of vascular endothelial cells (Matsumoto et al.,2001). PECAM-1-positive endothelial cells were developed in the interface of the liver diverticulum and septum transversum mesenchyme at E9.5. The large vessels running nearby the liver primordium, which might have been omphalomesenteric veins or common or posterior cardinal veins, branched into the septum transversum mesenchyme (Fig. 1A,B). In the E12.5 liver, many primitive sinusoidal structures were well developed in the parenchymal regions, and hepatic veins and portal veins, which had comparatively large diameters, were also recognized. At this stage, the liver also contained many hemopoietic cells that had nuclei densely stained with hematoxylin. Anti-PECAM-1 antibodies were also reactive with megakaryocytes and some hemopoietic cells (Fig. 1C,D). Sinusoidal structures anastomosed to form PECAM-1-positive longer vascular channels in sections by E17.5. These endothelial cells appeared to be longer and flattened as compared to those in the early stages (Fig. 1E, F). Thereafter, sinusoids were regularly arranged in the liver parenchyma, and the hepatic vascular system consisting of the sinusoids and portal and central veins established from the newborn stage through the adult stage (Fig. 1G–J). The expression of PECAM-1 was downregulated in sinusoids in neonatal and adult stages. Portal veins and hepatic veins expressed PECAM-1 at high levels throughout development. Endothelial cells of the hepatic artery were recognized in the perinatal periods and were PECAM-1-positive. PECAM-1-positive cells also existed around bile ducts (Fig. 1J).

thumbnail image

Figure 1. Development of vascular system during mouse hepatic organogenesis. A: H-E staining. C, E, G, I: Immunohistochemistry for cytokeratin (green) using anti-cow keratin antiserum (blue, DAPI staining). B, D, F, H, J: Immunohistochemistry for PECAM-1 (red). A, B: E9.5 embryo. C, D: E12.5 liver. E, F: E17.5 liver. G, H: Neonatal liver. I, J: Adult liver. PECAM-1-positive endothelial cells (arrow) develop mainly along the E9.5 hepatic diverticulum (A, B). Primitive sinusoids (arrows) are well developed at E12.5, but are short in section (C, D). Megakaryocytes (asterisk) are also PECAM-1-positive. Sinusoid structures (arrows) become elongated at E17.5 in section (E, F). In neonatal liver, PECAM-1 signals are lower in sinusoidal endothelial cells (G, H). Cytokeratin-positive biliary epithelial cells are detectable around portal vein (G). Although the expression of PECAM-1 is perfectly downregulated in sinusoids (arrows) in the adult stage (I, J), endothelial cells of the hepatic artery and biliary vessels (arrowheads) are strongly PECAM-1-positive in addition to those of portal and central veins (inset in J). bd, bile duct; cv, central vein; ha, hepatic artery; ld, liver diverticulum; ov, omphalomesenteric vein; pv, portal vein; stm, septum transversum mesenchyme. Bars = 100 μm.

Download figure to PowerPoint

Confluent Vascular System of the Liver Primordium in Early Developmental Stages

Although a dissection microscope three-dimensionally revealed well-developed capillary structures in the E12.5 liver (Fig. 2A), we did not know whether the primitive sinusoid structures were well connected with one another and other veins such as portal veins and hepatic veins.

thumbnail image

Figure 2. Confluent vascular system in an E12.5 mouse liver and E13.5 rat liver. A: Photograph of E12.5 mouse liver made using a dissection microscope. Primitive sinusoids are well developed. B–H: Localization of FITC-gelatin conjugate (green) in E13.5 rat tissue injected through the umbilical vein. B: Skin stained for isolectin B4 (red). C: Lung stained for isolectin B4 (red). D: Liver immunostained for cytokeratin 8/18 (red). E, F: Liver immunostained for PECAM-1 (red). G: Liver immunostained for desmin (red). H: Liver immunostained for type IV collagen (red). All capillaries in the dermis (B) and lung mesenchyme (C) contain the FITC-gelatin conjugate. The FITC-gelatin conjugate is distributed in the primitive sinusoids among the cytokeratin-positive parenchyma (D). The lumina of PECAM-1-positive sinusoids are FITC-positive (E, F; arrows). Primitive sinusoids filled with the FITC-gelatin conjugate (arrows) are surrounded by desmin-positive stellate cells (G) and type IV collagen (H). ep, epidermis; le, lung epithelial cells. Bars = 250 μm (A), 50 μm (B–D).

Download figure to PowerPoint

Flk-1-positive flattened cells mainly located along the liver diverticulum at E9.5, which was the same as the localization of PECAM-1-positive cells (Figs. 1B, 3A–C). In immunohistochemical analysis of Flk-1 expression in serial sections of the liver primordium at E9.5 and E10.5, isolated Flk-1-positive endothelial cells were not observed and all Flk-1-positive cells were well connected with one another (Fig. 3D–O). These cells around the liver diverticulum were also connected with nearby large vessels (the omphalomesenteric veins and common or posterior cardinal veins). However, we could not find Flk-1-positive endothelial cells in the main regions of the septum transversum mesenchyme like PECAM-1-positive cells (Fig. 3).

thumbnail image

Figure 3. Flk-1-positive cells along the E9.5 liver diverticulum are connected with the omphalomesenteric veins. A: Flk-1 immunohistochemistry (red). B: Cytokeratin 8/18 immunohistochemistry (green) in A. C: Merged photograph of A and B. D–O: Flk-1 immunohistochemistry in serial sections of the E9.5 liver primordium. Endothelial cells located along the liver diverticulum express Flk-1 receptor (A–C). All Flk-1 positive endothelial cells along the hepatic diverticulum are well connected with each other (white arrows and yellow arrows) and with the omphalomesenteric veins. Flk-1-positive cells are not observed in the main body of the septum transversum mesenchyme (D–O). ld, liver diverticulum; stm, septum transversum mesenchyme; ov, omphalomesenteric vein. Bars = 100 μm.

Download figure to PowerPoint

We next examined their confluence by injection of FITC-gelatin conjugates into E13.5 and E15.5 rat fetal livers, which corresponded to E11.5 and E13.5 mouse livers, respectively, through the umbilical vein. The FITC-gelatin conjugates distributed in all fine capillary networks of the dermis and lung, which were positive for isolectin B4 (Fig. 2B, C). The FITC-gelatin conjugates flowed not only into portal veins and hepatic veins but also into all narrow primitive sinusoidal structures in E13.5 and E15.5 rat livers (Fig. 2D). These primitive sinusoids were also supported by desmin-positive stellate cells and type IV collagen deposition (Fig. 2E–H).

Expression of Flk-1 During Liver Development

Endothelial cells of primitive sinusoidal structures also strongly expressed Flk-1 at E12.5, whereas those of portal veins weakly expressed Flk-1 and those of hepatic veins were almost Flk-1-negative (Fig. 4A–D). Flk-1 signals were observed on the overall cell membranes of the primitive sinusoidal endothelial cells. The positive signals in the sinusoidal structures significantly decreased in the E17.5 and newborn livers (Fig. 4E–H). Furthermore, the immunolocalization of Flk-1 in the endothelial cells dramatically changed from the overall staining of the cell membrane to dot-like staining. The dot-like staining was remarkable at the newborn stage (Fig. 4G, H). In the adult stage, Flk-1-positive signals were stronger than those of the newborn stage. Although some sinusoidal endothelial cells expressed Flk-1 in their overall cell membranes (Fig. 4K, L), the expression was weaker than that in the E12.5 liver, and dot-like staining of Flk-1 was observed in other sinusoids (Fig. 4I, J). These Flk-1-positive signals were perfectly abolished by competitive peptide absorption of the primary antibody (Fig. 4D).

thumbnail image

Figure 4. Immunolocalization of Flk-1 proteins changes during liver development. A–D: E12.5 liver. E, F: E17.5 liver. G, H: Neonatal liver. I–L: Adult liver. B, C: Magnified pictures of regions containing central vein and portal vein in A, respectively. D: Double immunofluorescence of Flk-1 (red) and vimentin (green). F, H, J, L: Magnified pictures of rectangles in E, G, I, and K, respectively. Endothelial cells of primitive sinusoids (arrows) are strongly immunostained for Flk-1 at E12.5 (A–D). Their whole cytoplasm or cell surface is positive. By contrast, endothelial cells of central veins are mostly Flk-1-negative (B), and those of portal veins are weakly Flk-1-positive (C). The Flk-1-positive signals are excluded in preabsorption experiments by competitive peptides (inset in D). Asterisk indicates a megakaryocyte. The staining of Flk-1 decreases at E17.5 (E, F), but its staining pattern is similar to that at E12.5. In neonatal liver, Flk-1 proteins are distributed as large positive dots (arrowheads), although the staining intensity significantly decreases (G, H). Dot-like staining and whole cell staining, both of which are weak, are detected in sinusoidal endothelial cells of adult liver (I–L). cv, central vein; pv, portal vein. Bars = 100 μm.

Download figure to PowerPoint

Immunoblotting analysis showed that 230- and 250-kDa Flk-1 forms were expressed in E12.5 and E14.5 livers as major bands (Fig. 5A). From E17.5 on, the downregulation of 230- and 250-kDa Flk-1 forms and upregulation of a 150-kDa Flk-1 form commenced. In the adult liver, a 150-kDa form was observed as a major band, though the 230- and 250-kDa bands were weakly detectable (Fig. 5A). The presence of competitive peptides in the primary antibody perfectly abolished the 230-, 250-, and 150-kDa bands in immunoblotting. Glycosidase treatment of fetal and adult liver lysates demonstrated that the 230- and 250-kDa bands disappeared and only a 180-kDa band was detectable (Fig. 5B). The phosphorylation of a tyrosine residue in the activation loop of Flk-1 molecules was detected from E12.5 through E17.5. Furthermore, the 250-kDa Flk-1 form was phosphorylated, but the 230- and 150-kDa forms were not (Fig. 5C). The 250-kDa form was not phosphorylated in neonatal and adult livers.

thumbnail image

Figure 5. Expression and activation of Flk-1 during liver development. A: Expression of Flk-1 protein. Liver cell lysates were separated on SDS-PAGE, and immunoblotted for Flk-1 proteins. Samples of each lane were derived from different animals. Very high expression of 230- and 250-kDa Flk-1 is detectable in both E12.5 and E14.5 livers. The expression of the 230- and 250-kDa form is downregulated at E17.5. On the other hand, 150-kDa-Flk-1 is upregulated from late gestational stages to the adult stage. Expression of the 150-kDa form peaks in the adult stage. B: The 230- and 250-kDa Flk-1 proteins are glycosylated. Liver lysates were treated with peptide-N-glycosidase F (PNGase) before immunoblotting. The molecular masses of the 230- and 250-kDa forms are decreased to 180 kDa by peptide-N-glycosidase F treatment in liver samples of all stages. C: Tyrosine phosphoylation (pTyr1054/1059) of 250-kDa Flk-1 proteins. Blots were incubated with anti-mouse VEGF receptor 2, phospho-specific (pTyr1054/1059) antibody. Tyrosine phosphoylation (pTyr1054/1059) is detected in E12.5, E14.5, and E17.5 livers, but not in neonatal and adult livers.

Download figure to PowerPoint

Expression of VEGF During Liver Development

Many VEGF-positive cells were detected in E12.5 and E14.5 livers by immunohistochemistry (Fig. 6A–F, P). Double immunofluorescent analysis demonstrated that CK8/18-positive hepatoblasts expressed VEGF (Fig. 7A–C). In addition to hepatoblasts, cells expressing VEGF at higher levels were observed in these livers. These might have been CK8/18-negative hemopoietic cells. Megakaryocytes also expressed VEGF. The positional relationships between VEGF-positive cells and sinusoid development such as stronger VEGF expression near the sinusoidal structures were not observed in the fetal livers. The distribution of VEGF-positive cells was homogeneous in the liver. VEGF expression became almost negative in the liver after E17.5 (Fig. 6G–O).

thumbnail image

Figure 6. Immunohistochemical analysis of VEGF expression during liver development. A–C, P: E12.5 liver. D–F: E14.5 liver. G–I: E17.5 liver. J–L: Neonatal liver. M–O: Adult liver. Double immunohistochemistry for VEGF (red; A, D, G, J, M) and PECAM-1 (green; B, E, H, K, N). C, F, I, L, O: Merged photographs of A and B, D and E, G and H, J and K, and M and N, respectively. VEGF is abundantly and evenly expressed in livers at E12.5 and E14.5 (A–F). Megakaryocytes (asterisks) are VEGF-positive. E17.5 liver shows weaker immunoreaction for VEGF (G–I), and neonatal livers are almost negative except for megakaryocytes (asterisks) (J–L). VEGF is not expressed in adult liver (M–O). Competitive peptides perfectly abolished immunofluorescent VEGF signals in E12.5 liver (P). Bars = 50 μm.

Download figure to PowerPoint

thumbnail image

Figure 7. VEGF is expressed in hepatoblasts and hemopoietic cells at E12.5. A: VEGF immunohistochemistry. B: Cytokeratin 8/18 immunohistochemistry. C: Double exposure of A and B. Although cytokeratin 8/18-positive cells express VEGF (arrows), stronger signals are seen in cytokeratin 8/18-negative cells, which may be hemopoietic cells. Asterisk shows cytokeratin-negative but VEGF-positive cells. Bars = 50 μm.

Download figure to PowerPoint

VEGF Stimulates Sinusoidal Endothelial Proliferation In Vitro

The effects of VEGF on the growth of sinusoidal endothelial cells were analyzed using an in vitro culture system. Liver cells at E12.5, which consisted of hepatoblasts, stellate cells, endothelial cells, and hemapoietic cells as major cell types, were cultured with or without 10 ng/ml VEGF for 5 days. Although the growth patterns of cultured liver cells such as the shapes and sizes of colonies formed were not different between the experiments with addition of VEGF and controls in H-E staining (Fig. 8A, B), PECAM-1-positive endothelial cells showed significantly higher proliferation in the presence of VEGF than in the control culture (Fig. 8C, D). The PECAM-1-positive areas of cultures with VEGF were 2.5 times larger than those of control cultures (Fig. 8E). PECAM-1-positive endothelial cells cultured in the presence of VEGF also branched more frequently. When E17.5 liver cells were cultured in the absence of additional VEGF for 5 days, many endothelial cells showed branching morphogenesis. The addition of VEGF did not significantly stimulate growth and branching morphogenesis of endothelial cells in this case (Fig. 8F–J). The endothelial cells from E17.5 livers in culture formed thicker cord structures than those of E12.5 livers.

thumbnail image

Figure 8. VEGF stimulates growth and morphogenesis of sinusoid networks by E12.5 endothelial cells in vitro. VEGF (10 ng/ml) was added to the medium of E12.5 or E17.5 liver cell cultures for 5 days (A–D, F–I). PECAM-1-positive endothelial cells from E12.5 livers proliferate and have branching morphogenesis more extensively than the culture in the control experiment (C, D). PECAM-1-positive endothelial cells from E17.5 livers exhibit extensive growth and branching morphogenesis with and without VEGF (H, I). A, B, F, G: H-E staining. C, D, H, I: PECAM-1 immunohistochemistry. E, J: Quantitative data showing growth of E12.5 and E17.5 sinusoidal endothelial cells, respectively (PECAM-1-positive area). Data are expressed as means ± SD. The number in parenthesis denotes the number of samples analyzed. Bars = 50 μm.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The present study demonstrated that hepatic vascular development commenced with liver primordium formation, and that VEGF may be involved in the growth and morphogenesis of vascular endothelial cells, especially those of primitive sinusoids, in early stages of liver development. At E9.5, endothelial cells developing close to the liver diverticulum expressed both PECAM-1 and Flk-1, which is one of the VEGF receptors. This result agreed well with previous reports showing that endothelial cells around the liver diverticulum invade the liver bud after its growth (Matsumoto et al.,2001; Nonaka et al.,2007). Immunohistochemical analysis of Flk-1 and PECAM-1 expression in serial sections of the liver primordium revealed that all of the endothelial cells around the liver diverticulum were well connected with each other and also with large vessels, the omphalomesenteric veins, and common or posterior cardinal veins running on both sides of the liver primordium. PECAM-1- or Flk-1-positive endothelial cells did not invade the main body of the septum transversum mesenchyme. Isolated endothelial cells did not exist in the septum transversum mesenchyme. These data suggest that the origin of endothelial cells in the liver primordium may be omphalomesentric veins or common or posterior cardinal veins running close to it. Endothelial cells of these large vessels may grow along the liver diverticulum and then invade the septum transversum mesenchyme along with hepatic cords extending from the diverticulum. In the E12.5 mouse liver, a well-developed and connected vascular system was also observed by using a dissection microscope, and many primitive sinusoidal structures developed in their sections. We also injected FITC-labeled gelatin into E13.5 and E15.5 rats, which correspond to E11.5 and E13.5 mice, through the umbilical veins to visualize the hepatic vascular development at these stages. As a result, all blood vessels within the liver, including portal veins, central veins, and primitive sinusoid structures, were labeled with FITC-gelatin conjugates, suggesting that the primitive sinusoids have lumina and connect well with each other, and also with portal and central veins. The endothelial cells of sinusoidal structures were surrounded by an extracellular matrix of basal laminar components and desmin-positive hepatic stellate cells. These results showed that all primitive sinusoids at this stage were already functional vessels as channels for passage of blood cells. Because hepatic endothelial cells were well connected with each other at E9.5 in mice and E13.5 and E15.5 in rats, it is likely that the vascular development in the liver primordium takes place through angiogenesis from already existing blood vessels from early developmental stages.

In the present study, we also showed that different expression levels of Flk-1 were detectable among endothelial cells in fetal livers. Endothelial cells of primitive sinusoids strongly expressed Flk-1 while its expression in endothelial cells of portal veins and hepatic veins was weaker than that of the primitive sinusoids. These data suggest that mechanisms of sinusoid formation and growth factors required for sinusoid growth are different from those of other vessels in fetal livers from early developmental stages. During liver regeneration, Flk-1 is expressed specifically in the sinusoidal endothelial cells (Shimizu et al.,2001,2005). The gene expression of connexins and Jagged1 has also been shown to differ among endothelial cells of blood vessels in the young E11.5 liver (Shiojiri et al.,2006).

In terms of hepatic Flk-1 expression, three major forms of Flk-1 with molecular masses of 150, 230, and 250 kDa were noted. Takahashi and Shibuya (1997) reported that the 150-kDa protein was predicted to be a non-glycosylated form from the primary translational product of the Flk-1 gene, and that a 200-kDa-Flk-1 form might be produced by addition of glycosylation to the 150-kDa form. However, the molecule is an intermediate and immature protein that is not transferred to the cell surface. After further glycosylation to the resultant mature product, only the 230-kDa-Flk-1 form is found on the cell surface. The 150- and 230-kDa forms, and the 250-kDa form detected in the present study may correspond to the immature product and mature product distributed on the cell surface, respectively. Our glycosidase treatment generated a 180-kDa form from 230- and 250-kDa forms. During liver development, the mature 250-kDa Flk-1 form appearing in early stages decreased and the immature 150-kDa Flk-1 form increased after E17.5. The translational activity of the Flk-1 protein may be maintained at a constant level during fetal and perinatal liver development because the total amount of 150-, 230-, and 250-kDa proteins was not changed on a large scale as shown in the present study. However, this is true if the half life does not differ between the respective forms of Flk-1 receptor and if no difference in the half life exists during development. Flk-1 protein, especially the immature forms, was abundant in the adult liver. Processing from the 150-kDa form to the 250-kDa form may be suppressed after E17.5, though the mechanisms by which this could occur are not known at present. At least the posttranslational mechanisms may change to control the function of Flk-1 proteins during fetal liver development, which correlates well with remarkable growth of primitive sinusoidal structures and their maturation. It is also noteworthy that Flk-1 localization changed during liver development from overall distribution to dot-like localization in sinusoidal endothelial cells. This change may be closely related with the shift of Flk-1 molecular species during liver development. Furthermore, when tyrosine phosphorylation in the activation loop of Flk-1 proteins for their signaling was analyzed by immunoblotting in the present study, the phosphorylation was detected in the 250-kDa form, but not in the 150- and 230-kDa forms. This result agrees well with the report that the immature form of Flk-1 was not distributed at the cell surface (Takahashi and Shibuya,1997). The phosphorylation of Flk-1 in neonatal and adult livers was not detected, though 250-kDa Flk-1 was expressed even at these stages though at a low level. These data suggest that Flk-1 was activated during early liver development, when primitive sinusoidal structures grow and develop a fine capillary network in the whole liver. At the same time, strong expression of VEGF was detected in hepatoblasts/hepatocytes and hemopoietic cells in these stages. In addition, when VEGF was added to the E12.5 liver cell cultures, it remarkably promoted sinusoidal development, including extensive branching morphogenesis, which resembled that in vivo. The addition of VEGF did not have an effect on the growth and branching morphogenesis of endothelial cells from E17.5 livers. Furthermore, the addition of a Flk-1 inhibitor blocked growth and branching morphogenesis of endothelial cells in E12.5 liver cell cultures (Takabe et al., unpublished data). Proliferating endothelial cells in vitro do not express connexin40 or factor VIII, and are thought to be sinusoidal endothelial cells (Shiojiri et al.,2006). In any event, VEGF-Flk-1 signaling may be essential for proliferation of immature sinusoidal endothelial cells in early developmental stages. VEGF may stimulate not only the growth but also branching morphogenesis in primitive sinusoidal endothelial cells. As the tremendous growth of the liver and concomitant proliferation of sinusoidal endothelial cells proceed during late developmental stages, including postnatal stages, there may be different mechanisms for their growth such as angiopoietin/Tie or PEDF after E17.5 (Sato et al.,2001; Sawant et al.,2004). Sinusoidal endothelial cells express the angiopoietin/Tie family in the regenerating rat liver (Sato et al.,2001). PEDF may be a factor that regulates the formation and maintenance of normal vasculature in liver development (Sawant et al.,2004).

It has been reported that hepatocytes produce VEGF in the regenerating liver (Shimizu et al.,2001). In the present study, we demonstrated that VEGF was expressed in hepatoblasts/hepatocytes, hemopoietic cells, and megakaryocytes in fetal mouse livers. Immunolocalization of VEGF-positive cells did not have any special tendency in the fetal liver such as their localized expression in regions close to primitive sinusoids or in the periphery of the liver. Therefore, a high concentration of VEGF in the whole liver tissue, which is derived from hepatoblasts, hemopoietic cells, and megakaryocytes, may stimulate the sinusoidal growth.

In summary, all endothelial cells in early stages of liver development are well connected with each other, and the hepatic vascular system may develop via angiogenesis from the omphalomesenteric or common or posterior cardinal veins. VEGF-Flk-1 signaling may act on proliferation and branching of immature sinusoidal endothelial cells during early stages of liver development.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Animals

C3H/HeSlc strain mice (SLC, Shizuoka, Japan) and Wistar strain rats (SLC) were used. Animals were housed in a temperature-controlled environment with a 12-hr light/dark cycle and were mated during the night. If the vaginal plug in the mouse or sperm in the vaginal smears in the rat were found the next morning, noon was considered as E0.5. Mouse fetuses at E9.5, E10.5, E12.5, E14.5, E17.5, newborns (NB) (1 day old), and adult animals (8–12 weeks old) were used for histology, immunohistochemistry, and immunoblotting. E12.5 and E17.5 mouse fetuses were also used for cell culture. Rat fetuses at E13.5 and E15.5, which correspond to E11.5 and E13.5 mouse fetuses, respectively, were used for fluorescein isothiocyanate (FITC)-gelatin injection experiments. At least three animals were examined in all experiments. All animal experiments were carried out in compliance with the “Guide for Care and Use of Laboratory Animals” of Shizuoka University.

Immunohistochemistry and Histology

Mouse fetuses and tissues were frozen in liquid nitrogen. Frozen sections were cut 8 μm thick and fixed in cold acetone (−30°) for 10 min. Some mouse fetuses and tissues were fixed in a chilled mixture of 95% ethanol and glacial acetic acid (99:1 v/v) overnight, and embedded in paraffin after dehydration. Paraffin sections were cut 6 μm thick. Frozen sections, dewaxed sections, and culture samples after hydration were incubated with primary antibodies for 1 hr at room temperature or overnight at 4°. A rabbit anti-mouse Flk-1 antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), rat monoclonal anti-PECAM-1 antibody (eBioscience, San Diego, CA), guinea pig anti-bovine cytokeratin 8 and 18 (CK8/18) antiserum (Progen Biotechnik Gmbh, Heidelberg, Germany), rabbit anti-cow keratin antiserum (Dako, Carpinteria, CA), rabbit anti-mouse type IV collagen antiserum (Chemicon International, Temecula, CA), rabbit anti-chicken desmin antiserum (Medac Gmbh, Hamburg, Germany), goat anti-vimetin antiserum (Chemicon International), and goat anti-mouse VEGF antibody (Santa Cruz Biotechnology, Inc.) were used as primary antibodies. All primary antibodies except for rabbit anti-cow keratin antiserum were used at a 1/100 dilution with phosphate-buffered saline (PBS)-1% bovine serum albumin (BSA). Rabbit anti-cow keratin antiserum was used at a 1/300 dilution. For Flk-1 immunohistochemistry, retrieval of its antigenicity on paraffin-embedded tissue sections was performed by treatment in 10 mM Tris (pH8.0)/0.1 mM EDTA/0.05% Tween 20 (TE) for 10 min at 100° before incubation with the primary antibody. After washing with PBS, sections were incubated with fluorescein- (1/100 dilution) or Cy3-labeled (1/500 dilution) donkey anti-rabbit, rat, goat, or guinea pig IgG antibodies (Jackson ImmunoResearch Laboratory, West Grove, PA) for 1 hr at room temperature. They were then mounted in buffered glycerol containing p-phenylenediamine (Johnson and de C Nogueira Araujo1981). Control incubations were performed with PBS-BSA instead of the primary antibodies. To check the specificities of the anti-mouse Flk-1 and anti-mouse VEGF antibodies, peptide absorption experiments were performed. Antibodies were preabsorbed with fivefold excessive competitive antigenic peptides in weight-volume percentages for 30 min at room temperature. Dewaxed sections were also stained with hematoxylin and eosin (H-E) for histology.

Injection of FITC-Gelatin Conjugates Into Rat Fetuses

FITC-gelatin conjugation was carried out according to the method of Hashimoto et al. (1998). An aliquot of FITC (20 mg) (Dojin Laboratory, Kumamoto, Japan) was first dissolved in 0.5 ml of dimethylsulfoxide (Nacalai Tesque, Kyoto, Japan) and then mixed for conjugation in a 20% gelatin (Sigma-Aldrich Corp. St. Louis, MO)-water solution (pH 11.0) at 37° overnight. The FITC-conjugated gelatin was subsequently dialyzed in PBS at 37° in the dark for 18 days, after which it was stored at 4° in the dark until use. Because rat fetuses are much larger than mouse counterparts in the same stages and can be easily manipulated, the former was used for FITC-gelatin conjugate injection experiments in the present study. Following the removal of the uterus from the impregnated rats at E13.5 and 15.5 under ether anesthesia, the FITC-conjugated gelatin, diluted with PBS (1:2) at 37°, was gently perfused into the umbilical vein, and blood was drained from the umbilical artery. After complete perfusion (1–2 ml), the fetuses were fixed in cold 0.5 % paraformaldehyde (PFA) solution containing 15% picric acid for approximately 30 min. The fetal trunks were then removed and immediately postfixed in the same fixative solution at 4° overnight in the dark.

For the simultaneous visualization of primitive sinusoidal networks and immunohistochemical localization of basal laminar components, PECAM-1, desmin, or vimentin, frozen sections were prepared and immunostained as previously indicated. For histochemical detection of binding sites for isolectin B4, which is specifically expressed in endothelial cells (Gerhardt et al.,2003), fixed trunks were embedded in paraffin after dehydration. Isolectin B4 staining was performed to investigate the microstructures of endothelial cells according to the method of Nakakura et al. (2006). The sections were incubated with biotinylated isolectin B4 (Sigma-Aldrich Corp.) at a concentration of 20 μg/ml overnight, washed with PBS, and then incubated with Alexa 546-conjugated streptavidin (Molecular Probes, Eugene, OR) (1/200 dilution) for 2 hr.

Immunoblotting

Liver tissues were homogenized in a lysis buffer (1% Nonidet P40, 0.1% sodium dodecyl sulfate [SDS], 0.5% sodium deoxycholate, 150 mM NaCl, 10 mM Tris-HCl, pH7.8) containing 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, 1 μg/ml leupeptin, 1 μg/ml pepstatin, 1 mM sodium orthovanadate, and 1 mM NaF. After being placed on ice for 10 min, lysates were centrifuged at 11,000g for 10 min. The supernatant was collected and the concentration of protein was determined using a BCA protein assay reagent kit (Pierce Chemical Company, Rockford, IL).

Lysates (20 μg) reduced by the method of Laemmli (1970) were electrophoresed on SDS-10% polyacrylamide gel. Proteins in the gel were transferred to polyvinylidene fluoride (PVDF) membranes (Immobilon-P) (Millipore Corp., Billeria, MA). The membranes were blocked with 5% skim milk in 0.01 M Tris-HCl (pH 7.4)-buffered saline containing 0.1% Tween 20 (TBST). The blocking for detection of phosphotyrosine of Flk-1 was performed in TBST-0.1%BSA. After washing with TBST, the membranes were incubated with the rabbit anti-mouse Flk-1 antibody (1/2,000 dilution) or anti-mouse VEGF receptor 2, phospho-specific (pTyr1054/1059) antibody (1/1,000 dilution) (Calbiochem, La Jolla, CA) overnight at 4°. Phosphorylation of Flk-1 at Tyr1054 and Tyr1059 within the activation loop is required for its activation and enhances its intrinsic tyrosine kinase activity (Kendall et al.,1999; Zeng et al.,2001). Control incubations were performed with 5% skim milk or TBST-0.1%BSA in place of the primary antibody. Peptide absorption experiments were also performed for specific detection of Flk-1. Following washing with TBST, a horseradish peroxidase (HRP)-labeled goat anti-rabbit IgG antibody (ICN Biochemicals, Aurora, OH) was used as a secondary antibody at a 1/2,000 dilution for 2 hr at room temperature. ECL detection was conducted with Amersham Pharmacia Biotech (Buckinghamshire, UK) reagents according to the manufacturer's recommendations. To determine whether the immunoreactive proteins were glycosylated, extracts from livers at various stages were treated for 24 hr at 37° with peptide-N-glycosidase F (Daiichi Pure Chemicals, Tokyo, Japan) before SDS-PAGE and immunoblotting, in accordance with the manufacturer's instructions.

Cell Culture

Livers were dissected out from E12.5 or E17.5 mouse fetuses and then diced (Sugiyama et al.,2007). Diced livers were treated with 1,000 U/ml dispase (Godo Shusei Co. Ltd., Tokyo, Japan) dissolved in Dulbecco's modified Eagle's medium (Gibco, Grand Island, NY) containing 10% fetal bovine serum (FBS) (Sigma Chemical Co., St. Louis, MO) for 30 min at 37° (Nitou et al.,2002). After gentle pipetting with a Pasteur pipet, undigested tissues were removed from cell suspension by filtration using a nylon mesh filter (132-μm pore size) (Nihon Rikagaku Kikai Co. Ltd., Tokyo, Japan). Cells were collected by centrifugation at 60g for 10 min and the resultant cellular pellet was washed twice with DM-160 (Kyokuto Seiyaku Co. Ltd., Tokyo, Japan) containing 10% FBS and 0.01% deoxyribonuclease I (Worthington Biochem. Corp., Freehold, NJ). Finally, cells were resuspended in DM-160 containing 10% FBS, 10−7M dexamethasone (Sigma-Aldrich Corp.), 100 U/ml penicillin G potassium (Meiji Seika Co. Ltd., Tokyo, Japan), and 100 μg/ml streptomycin sulfate (Meiji Seika) at a concentration of 1×106 cells/ml. The viability was checked by the trypan blue exclusion test (more than 95%). Cell suspensions of 70 μl were cultured for 5 days on Teflon-coated glass slides (AR Brown Co. Ltd., Tokyo, Japan) at 37° in a water-saturated atmosphere containing 5% CO2. The medium was changed at days 1 and 3. In some experiments, 10 ng/ml VEGF (Genzyme-Techne Corp., Boston, MA) was added to the culture medium. After 5 days, cultured cells were fixed in cold acetone (−30°) for 10 min for immunofluorescence of PECAM-1 and H-E staining. The growth of PECAM-1-positive endothelial cells in culture was quantified using NIH image 1.61/ppc after digitized photographs of PECAM-1-positive areas were input into a PC.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank Professor Emeritus Takeo Mizuno of the University of Tokyo and Prof. Nelson Fausto of the University of Washington for their interest in our study and their encouragement, and Mr. Kim Barrymore for his help in preparing the manuscript.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  • Carpenter B, Lin Y, Stoll S, Raffai RL, McCuskey R, Wang R. 2005. VEGF is crucial for the hepatic vascular development required for lipoprotein uptake. Development 132: 32933303.
  • de Vries C, Escobedo JA, Ueno H, Houck K, Ferrara N, Williams LT. 1992. The fms-like tyrosine kinase, a receptor for vascular endothelial growth factor. Science 255: 989991.
  • Du Bois AM. 1963. The embryonic liver. In: RouillerC, editor. The liver, morphology, biochemistry, physiology. New York: Academic Press. p. 139.
  • Ferrara N, Henzel WJ. 1989. Pituitary follicular cells secrete a novel heparin-binding growth factor specific for vascular endothelial cells. Biochem Biophys Res Commun 161: 851858.
  • 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: 439442.
  • 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: 6670.
  • Gerber HP, Hillan KJ, Ryan AM, Kowalski J, Keller GA, Rangell L, Wright BD, Radtke F, Aguet M, Ferrara N. 1999. VEGF is required for growth and survival in neonatal mice. Development 126: 11491159.
  • Gerhardt H, Golding M, Fruttiger M, Ruhrberg C, Lundkvist A, Abramsson A, Jeltsch M, Mitchell C, Alitalo K, Shima D, Betsholtz C. 2003. VEGF guides angiogenic sprouting utilizing endothelial tip cell filopodia. J Cell Biol 161: 11631177.
  • Gouysse G, Couvelard A, Frachon S, Bouvier R, Nejjari M, Dauge MC, Feldmann G, Hénin D, Scoazec JY. 2002. Relationship between vascular development and vascular differentiation during liver organogenesis in humans. J Hepatol 37: 730740.
  • Hashimoto H, Ishikawa H, Kusakabe M. 1998. Simultaneous observation of capillary nets and tenascin in intestinal villi. Anat Rec 250: 488492.
  • Johnson GD, de C Nogueira Araujo GM 1981. A simple method of reducing the fading of immunofluorescence during microscopy. J Immunol Methods 43: 349350.
  • Kendall RL, Rutledge RZ, Mao X, Tebben AJ, Hungate RW, Thomas KA. 1999. Vascular endothelial growth factor receptor KDR tyrosine kinase activity is increased by autophosphorylation of two activation loop tyrosine residues. J Biol Chem 274: 64536460.
  • Laemmli UK. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680685.
  • Matsumoto K, Yoshitomi H, Rossant J, Zaret KS. 2001. Liver organogenesis promoted by endothelial cells prior to vascular function. Science 294: 559563.
  • Morita M, Qun W, Watanabe H, Doi Y, Mori M, Enomoto K. 1998. Analysis of the sinusoidal endothelial cell organization during the developmental stage of the fetal rat liver using a sinusoidal endothelial cell specific antibody, SE-1. Cell Struct Funct 23: 341348.
  • Nakakura T, Yoshida M, Dohra H, Suzuki M, Tanaka S. 2006. Gene expression of vascular endothelial growth factor-A in the pituitary during formation of the vascular system in the hypothalamic-pituitary axis of the rat. Cell Tissue Res 324: 8795.
  • Nitou M, Sugiyama Y, Ishikawa K, Shiojiri N. 2002. Purification of fetal mouse hepatoblasts by magnetic beads coated with monoclonal anti-E-cadherin antibodies and their in vitro culture. Exp Cell Res 279: 330343.
  • Nonaka H, Tanaka M, Suzuki K, Miyajima A. 2007. Development of murine hepatic sinusoidal endothelial cells characterized by the expression of hyaluronan receptors. Dev Dyn 236: 22582267.
  • Olsson AK, Dimberg A, Kreuger J, Claesson-Welsh L. 2006. VEGF receptor signalling in control of vascular function. Nat Rev Mol Cell Biol 7: 359571.
  • Roskoski R Jr. 2008. VEGF receptor protein-tyrosine kinases: structure and regulation. Biochem Biophys Res Commun 375: 287291.
  • Sato T, El-Assal ON, Ono T, Yamanoi A, Dhar DK, Nagasue N. 2001. Sinusoidal endothelial cell proliferation and expression of angiopoietin/Tie family in regenerating rat liver. J Hepatol 34: 690698.
  • Sawant S, Aparicio S, Tink A, Lara N, Barnstable CJ, Tombran-Tink J. 2004. Regulation of factors controlling angiogenesis in liver development: a role for PEDF in the formation and maintenance of normal vasculature. Biochem Biophys Res Commun 325: 408413.
  • Severn CB. 1972. A morphological study of the development of the human liver. II Establishment of liver parenchyma, extrahepatic ducts and associated venous channels. Am J Anat 133: 85108.
  • 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: 6266.
  • Shibuya M. 2006. Differential roles of vascular endothelial growth factor receptor-1 and receptor-2 in angiogenesis. J Biochem Mol Biol 39: 469478.
  • Shibuya M, Claesson-Welsh L. 2006. Signal transduction by VEGF receptors in regulation of angiogenesis and lymphangiogenesis. Exp Cell Res 312: 549560.
  • Shibuya M, Yamaguchi S, Yamane A, Ikeda T, Tojo A, Matsushime H, Sato M. 1990. Nucleotide sequence and expression of a novel human receptor-type tyrosine kinase gene (flt) closely related to the fms family. Oncogene 5: 519524.
  • Shimizu H, Miyazaki M, Wakabayashi Y, Mitsuhashi N, Kato A, Ito H, Nakagawa K, Yoshidome H, Kataoka M, Nakajima N. 2001. Vascular endothelial growth factor secreted by replicating hepatocytes induces sinusoidal endothelial cell proliferation during regeneration after partial hepatectomy in rats. J Hepatol 34: 683689.
  • Shimizu H. Mitsuhashi N, Ohtsuka M, Ito H, Kimura F, Ambiru S, Togawa A, Yoshidome H, Kato A, Miyazaki M. 2005. Vascular endothelial growth factor and angiopoietins regulate sinusoidal regeneration and remodeling after partial hepatectomy in rats. World J Gastroenterol 11: 72547260.
  • Shiojiri N, Sugiyama Y. 2004. Immunolocalization of extracellular matrix components and integrins during mouse liver development. Hepatology 40: 346355.
  • Shiojiri N, Niwa T, Sugiyama Y, Koike T. 2006. Preferential expression of connexin37 and connexin40 in the endothelium of the portal veins during mouse liver development. Cell Tissue Res 324: 547552.
  • Sugiyama Y, Koike T, Shiojiri N. 2007. Immunohistochemical analyses of cell-cell interactions during hepatic organoid formation from fetal mouse liver cells cultured in vitro. Histochem Cell Biol 128: 521531.
  • Takahashi T, Shibuya M. 1997. The 230 kDa mature form of KDR/Flk-1 (VEGF receptor-2) activates the PLC-gamma pathway and partially induces mitotic signals in NIH3T3 fibroblasts. Oncogene 14: 20792089.
  • Yoshida M, Nishikawa Y, Omori Y, Yoshioka T, Tokairin T, McCourt P, Enomoto K. 2007. Involvement of signaling of VEGF and TGF-beta in differentiation of sinusoidal endothelial cells during culture of fetal rat liver cells. Cell Tissue Res 329: 273282.
  • Zeng H, Sanyal S, Mukhopadhyay D. 2001. Tyrosine residues 951 and 1059 of vascular endothelial growth factor receptor-2 (KDR) are essential for vascular permeability factor/vascular endothelial growth factor-induced endothelium migration and proliferation, respectively. J Biol Chem 276: 3271432719.