ACTAUNIVERSITATISUPSALIENSISUPPSALA2007

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 235

Modulation of Angiogenesis byLaminins and Heparan Sulfate

LARS JAKOBSSON

ISSN 1651-6206ISBN 978-91-554-6818-7urn:nbn:se:uu:diva-7666

No man is an island

List of papers

This thesis is based on the following papers, which will be referred to in the text by their Roman numerals:

I Dixelius, J., Jakobsson, L., Genersch, E., Bohman, S., Ekblom, P., and Claesson-Welsh, L. (2004). Laminin-1 promotes angio-genesis in synergy with fibroblast growth factor by distinct regu-lation of the gene and protein expression profile in endothelial cells. J. Biol. Chem. 279, 23766-23772.

II Jakobsson, L., Kreuger, J., Holmborn, K., Lundin, L., Eriksson, I., Kjellén, L., and Claesson-Welsh, L. (2006). Heparan sulfate in trans potentiates VEGFR-mediated angiogenesis. Dev. Cell 10,625-634

III Jakobsson, L., Domagatskaya, A., Tryggvason, K., Edgar, D., Claesson-Welsh, L. (2007). Laminins regulate vascular lumen diameter. Submitted.

IV Jakobsson, L., Kreuger, J., Claesson-Welsh, L., (2007). Build-ing blood vessels; Stem cell models in vascular biology. Review. Submitted.

Contents

Introduction...................................................................................................11The vascular system .................................................................................11

Functional anatomy and histology.......................................................11Vasculogenesis ....................................................................................12Angiogenesis .......................................................................................13

-In health .........................................................................................13-in disease........................................................................................15

Molecular interplay in angiogenesis.........................................................16The VEGF signaling system................................................................16

VEGF receptors...............................................................................16VEGF-A..........................................................................................18Receptor activation and signal transduction ...................................19Receptor inactivation and degradation............................................21VEGF coreceptors...........................................................................21

Fibroblast growth factor receptor-1 and its ligands .............................23The basement membrane .....................................................................23

Laminins and the assembly of the BM............................................24The integrins ...................................................................................28

Glycobiology .......................................................................................28Heparan sulfate and heparin............................................................29

Embryonic stem cells, embryoid bodies and vascular biology............33

Present investigations and future perspectives..............................................34Paper I ......................................................................................................34

Laminin-1 promotes angiogenesis in synergy with fibroblast growth factor by distinct regulation of the gene and protein expression profile in endothelial cells ...............................................................................34Aim ......................................................................................................34Results .................................................................................................34Discussion............................................................................................35Future perspectives ..............................................................................36

Paper II .....................................................................................................36Heparan sulfate in trans potentiates VEGFR-mediated angiogenesis .36Aim ......................................................................................................36Results .................................................................................................36

Discussion............................................................................................37Future perspectives ..............................................................................39

Paper III....................................................................................................39Laminins regulate vascular lumen diameter ........................................39Aim ......................................................................................................39Results .................................................................................................39Discussion............................................................................................40Future perspectives ..............................................................................40

Paper IV ...................................................................................................41Building blood vessels: Stem cell models in vascular biology............41Aim ......................................................................................................41Results .................................................................................................41Future perspectives ..............................................................................41

Concluding remarks ......................................................................................42

Sammanfattning på svenska..........................................................................43

Acknowledgements.......................................................................................45

References.....................................................................................................47

Abbreviations

BM Basement membrane EB Embryoid body EC Endothelial cell ECM Extracellular matrix ESC Embryonic stem cell FGF Fibroblast growth factor FGFR Fibroblast growth factor receptor GAG Glycosaminoglycan HIF-1 Hypoxia inducible factor-1 HRE Hypoxia responsive element HS Heparan sulfate HSPG Heparan sulfate proteoglycan LM Laminin MMP Matrix metalloprotease NDST N-deacetylase/N-sulphotransferase PDGF Platelet-derived growth factor PECAM Platelet-endothelial cell adhesion molecule PKC Protein kinase C PlGF Placental growth factor RTK Receptor tyrosine kinase VEGF Vascular endothelial growth factor VPF Vascular permeability factor

11

Introduction

All species within the animal kingdom depend on oxygen for development and survival. In vertebrates and higher organisms, the delivery of oxygen and nutrients is mediated by the circulatory system, with the heart pumping blood to and from the tissues via the blood vessels. Also, cells require oxy-gen and nutrients to divide; hence blood vessel formation (angiogenesis) is a prerequisite for tissue expansion. In a variety of pathologies such as cancer, diabetic retinopathy and rheumatoid arthritis, vessel growth contributes to the progression of the disease. In other conditions, for instance in diabetic wounds, the situation is the opposite with insufficient circulation. The data presented in this thesis aim to increase our understanding of the regulation of angiogenesis.

The vascular system

Functional anatomy and histology The blood vascular system, first described by William Harvey 1628 (Harvey, 1628), carries the blood from the heart via the arteries, arterioles, capillaries and back through the venules and veins. The essential exchange of gases, nutrients, antibodies and metabolic waste products between tissue and blood takes place in the capillaries. In addition, this is the site for extravasation of leukocytes to inflamed or infected tissue. As diffusion of oxygen is limited to 100-200 µm, there is a need of a capillary in close vicinity to every cell.

The capillaries are 7-30 µm in diameter and consist of an inner lining of approximately 1-4 endothelial cells (ECs)/diameter resting on a basement membrane (BM). This BM is shared by mesenchymal smooth muscle cells, denoted pericytes (Fig. 1). The pericytes do not completely cover the capil-laries but cling on to them like a hand grabbing a racket. These specialized cells and their interaction with the ECs are of great importance for the func-tion of the vessel (Hellström et al., 2001; Leveen et al., 1994; Soriano, 1994).

Larger vessels serve to regulate blood pressure and to transport the blood to the capillaries. Since these vessels are exposed to high pressure they are surrounded by layers of elastic tissue, continuous smooth muscle cells and

12

connective tissue. ECs fulfill different tasks depending on their anatomical position. In some tissues, where there are needs for a fast interchange be-

tween blood and tissues (for example in kidney glomeruli and intestine), small gaps, fenestrae, are interrupting the plasma membrane of the ECs. The fenestrae will allow macromolecules to pass without crossing the lipid bi-layer. In muscles and connective tissue, on the other hand, capillaries lack fenestrae. However, to prevent free diffusion of liquid and lipids, ECs have intercellular connections denoted tight junctions. In addition, they connect via adherens junctions of which the vascular endothelial calcium dependent cell adhesion molecule (VE-cadherin) is the main protein (for a review see Dejana, 2004). Regardless of their organ location, ECs share most properties such as protein expression and location within the vessel.

VasculogenesisDuring gastrulation in mammals, both embryonic and extraembryonic meso-dermal cells are formed. Vasculogenesis is the de novo formation of vascular structures within the mesoderm (Risau and Flamme, 1995). The cardio-vascular system is the first organ to develop. In the mouse embryo, this process is initiated at embryonic day 7 (E7) in the primitive streak by the formation of clusters of pluripotent cells, hemangioblasts (Choi et al., 1998). It has been suggested that a fraction of the hemangioblasts stay within the embryo proper whereas some migrate to the yolk sac, where they differenti-ate into hematopoietic precursor cells (primitive erythroid cells) and endo-thelial precursor cells, denoted angioblasts (Fig. 2). Together these cells form multi-cellular structures called blood islands. The angioblasts differen-tiate into ECs that make up a primitive vascular network which later on reor-ganizes through sprouting angiogenesis. In the embryo proper, blood islands are lacking. Here, hemangioblasts are found as individual cells in close vi-

Figure 1. Cross section of a capillary. ECs are surrounded by pericytes and rest on a BM.

13

cinity of the floor of the developing aorta (for review see Wilting et al., 2003). At E8.5, certain hemangioblasts have become angioblasts that mi-

grate to outline the future dorsal aorta and the cardinal veins. The heman-gioblasts have also been suggested to have the capacity to differentiate into supportive smooth muscle cells (Yamashita et al., 2000). The cardinal vein and the dorsal aorta are formed through maturation and connection of the ECs, without involvement of sprouting angiogenesis. The smaller branches that connect the arterial and venous tubes, such as the intersomitic vessels, are in contrast generated through sprouting angiogenesis. However, not all EC precursors seem to originate from the hemangioblast, but are established as individual angioblasts that can not differentiate into hematopoietic cells. Nevertheless, it is difficult to exclude the possibility that all EC and hemato-poietic cells stem from a common precursor that might be hard to detect, due to a short life span.

Angiogenesis -In health Angiogenesis occurs in two fundamentally different ways, either by sprout-ing angiogenesis or by intussusceptive growth (Burri and Djonov, 2002). Developmental sprouting angiogenesis has been very well characterized, mainly by live imaging in the zebra fish (Lawson and Weinstein, 2002) and by studies of the developing retina (Uemura et al., 2006). In sprouting an-giogenesis, ECs that are integrated in the vessel wall break loose and start to migrate upon a stimulus of growth factors, dominated by vascular endothe-lial growth factor (VEGF) (Habeck et al., 2002; Leung et al., 1989; Tischer et al., 1989) (first described as vascular permeability factor; VPF (Keck et al., 1989; Senger et al., 1983)). The migrating ECs organize into a solid cord

Figure 2. Schematic outline of the vasculogenic process.

14

which subsequently opens up to form a lumenized vessel. The sequential steps of lumen formation are not completely clarified. However, developing intersegmental vessels in zebra fish, display fusion of intracellular vacuoles which connects to vacuoles of adjacent ECs to form a continuous lumen (Kamei et al., 2006). As the tip cell encounters another sprout or vessel it connects and circulation is established. Further maturation involves recruit-ment of smooth muscle cells and/or pericytes to the vessels. Dependent on the tissue, platelet derived growth factor (PDGF) -BB and its tyrosine kinase receptor PDGF receptor are main regulators of this process (Hellström et al., 1999; Lindahl et al., 1997). For example, in the retina, protruding endo-thelial tip cells produce the majority of PDGF-BB which in turn attracts pericytes by activation of PDGF receptor (Gerhardt et al., 2003; Hammes et al., 2002).

Mature ECs are completely enclosed by layers of interconnected extracel-lular matrix proteins (Fig. 1 and 6). This envelope is degraded by various proteases including matrix metalloproteases (MMPs) which are released for example during inflammation and as a result of VEGF-receptor (VEGFR) activation (Lamoreaux et al., 1998; Pettersson et al., 2000). This degradation is considered vital for the initiation if angiogenesis. As the ECs and the sur-rounding pericytes loose their connections to the BM, the ECs start to mi-grate towards the source of VEGF. Most information about sprouting angio-genesis in the adult has been retrieved by analysis of fixed tissue. This repre-sents a snapshot of a given state of the dynamic angiogenic process. It is therefore difficult to draw clear conclusions about the order of events, since “before and after” can not be distinguished. Microscopic techniques are con-stantly improved, providing tools for high resolution live imaging. Maybe such studies will unravel the plasticity of the vascular bed in adult and patho-logical tissue.

Intussusceptive angiogenesis is characterized by longitudinal splitting of capillaries by insertion of tissue pillars resulting in two new vessels parallel to the mother vessel (Burri and Djonov, 2002). In contrast to sprouting an-giogenesis, blood flow is maintained throughout the process. Intussusception has not been as extensively studied as developmental sprouting. Few exam-ples of live imaging to visualize this process have been presented. Although sprouting and intussusception are considered two separate processes it is possible that they may combine to facilitate angiogenesis. When sprouts are induced, the EC start to migrate, in turn generating a lateral force to its mother vessel. By doing so, the vessel might be flattened, thereby facilitating the creation of lateral pillars. Again, careful live imaging analyses might resolve such a possibility in the future.

Angiogenesis is mostly induced by cells exposed to low oxygen pressure, denoted hypoxia. Nucleated cells deprived of oxygen will upregulate the transcription factor hypoxia inducible factor-1 (HIF-1 ). HIF-1 in turn binds HIF-1 , allowing translocation of the complex into the nucleus. The

15

HIF-1 /1 complex binds to a specific oligonucleotide stretch, the hypoxia responsive element (HRE). Many genes involved in angiogenesis (for exam-ple vegfa and vegfr1) harbour the HRE sequence in their promoter region, and are thereby induced by hypoxia (for a review see Pugh and Ratcliffe, 2003). VEGF enhances the formation of vessel-structures by its specific mitogenic and migratory effects on ECs and by increasing vascular perme-ability (see Ferrara, 1999 for a review). Several other positive regulators of angiogenesis exist, such as fibroblast growth factors 1 and 2 (FGF) (Shing et al., 1984), transforming growth factors alpha and beta, hepatocyte growth factor, tumor necrosis factor alpha, angiogenin, interleukin-8, etc. (Folkman and Shing, 1992; Tonini et al., 2003).

Although formation of blood vessels in the adult is not abundant, it is a necessity for life and reproduction. All types of tissue expansion like muscle growth, wound healing, endometrial growth and ovulation is accompanied by angiogenesis.

-in disease Angiogenesis is a critical aspect of a number of pathological conditions such as diabetic retinopathy, rheumatoid arthritis, psoriasis, and cancer. The for-mation of new blood vessels is crucial for the progression of most growing tumors. Without vascularization the tumor is dependant on diffusion for the exchange of nutrients and oxygen. In such a poor environment cell division is suppressed. This knowledge has generated a great interest in the possibil-ity to directly target tumor angiogenesis, thereby preventing tumor growth. Also, tumor vessels are known to undergo constant remodeling, a property that would make them susceptible to anti-angiogenic treatment. Research has been intense, trying to find inhibitors of angiogenesis. So far, three drugs that target VEGF function have been approved by the federal drug admini-stration in the USA to be used clinically; 1. Bevacizumab (Avastin™; a VEGF neutralizing antibody). Treatment with bevacizumab in combination with chemotherapy resulted in about four months’ prolonged survival in patients suffering from colon cancer. 2. Pegaptanib, (Macugen™; an oli-gonucleotide that binds VEGF) is used for treatment of wet age-related macular degeneration (AMD), by intravitreal injections (Gragoudas et al., 2004). 3. Ranibizumab (Lucentis™; an antibody fragment that binds all VEGF isoforms) is another molecule used for treatment of wet AMD. Treatments with these substances have generated side effects that can be linked to the interruption of VEGF-signaling. For a review on VEGF-inhibition in clinical trials see (Jain et al., 2006).

Several molecules with anti-angiogenic properties have been identified in human tissue. These are denoted endogenous inhibitors, and most of them are proteolytic fragments of extracellular matrix (ECM) proteins or plasma proteins, exemplified by endostatin (a fragment of collagen XVIII), endore-pellin (the C-terminal part of perlecan), angiostatin (a fragment of plasmino-

16

gen) and tumstatin (a fragment of the collagen IV 3-chain) (for a review see Bix and Iozzo, 2005). Several other inhibitors have been very promising both in vitro and in animal models, however only a few have had any effect in the clinic. There are a number of plausible reasons for this discrepancy between murine and human studies. Most preclinical studies have been per-formed on immuno-compromised mice using ectopic and fast growing tu-mors. These differ from most human tumors in many ways. Also, some tu-mors gain resistance to hypoxia, with blood vessel independence as a conse-quence. In the majority of clinical trials, the patients have been suffering from a late stage terminal cancer which could not be cured by conventional therapy. Anti-angiogenic treatment in combination with conventional che-motherapy has therefore been initiated as a last attempt to cure the patient. It is possible that the potential of anti-angiogenic substances are greater at an earlier stage of disease progression. In spite of the intense research in the vasculo- and angiogenesis field, many questions remain unanswered con-cerning the treatment with anti-angiogenic substances. It is therefore of great interest to develop new models for such studies, in which genetic, as well as other manipulations easily can be made.

Molecular interplay in angiogenesis

The VEGF signaling system The family of VEGFs and their receptors are major players in angiogenesis (for reviews see Ferrara, 2002 and Olsson et al., 2006). The receptors, VEGFR-1, -2, and -3 are transmembrane tyrosine kinases that dimerize and become activated upon ligand-binding. VEGFR-1, -2 and -3 are primarily expressed on hematopoietic cells, blood ECs and lymphatic ECs, respec-tively. All three receptors are needed for vascular development, since dele-tion of any of the receptors in mouse leads to vascular abnormalities and death at E8.5-10.5 (Dumont et al., 1998; Fong et al., 1995; Shalaby et al., 1995). In nonvertebrates, only one VEGFR-related protein is known, indicat-ing diversification of the receptors through gene duplications during evolu-tion of early vertebrates (Shibuya, 2002). The VEGF receptors bind six dif-ferent ligands, VEGF-A, -B, -C, -D, placental growth factor (PlGF) and a virus encoded VEGF-E, in a distinct pattern (reviewed by Cross et al., 2003 and Olsson et al., 2006) (Fig. 3).

VEGF receptors VEGFR-1 (also denoted fms-like tyrosine kinase-1 (Flt-1) in mouse) is mainly expressed on blood vascular endothelium, macrophages/monocytes

17

and hematopoietic stem cells and exists in two forms: 1. The full length pro-tein with a ligand-binding extracellular domain, a transmembrane domain and an intracellular kinase domain. 2. A secreted protein, composed of only the extracellular domain, encoded by an alternatively spliced variant (Kendall and Thomas, 1993). Several different functions for this receptor have been described. The full length form mediates monocyte recruitment to inflamed tissues, a process dependent on receptor activation (known to be much weaker than VEGFR-2). The soluble receptor acts as a suppressor of full length VEGFR-1 and -2 by competing for ligand-binding. These find-ings are evident from phenotypes with either deleted full length receptor, resulting in death at E8.5-9 due to excessive EC growth (Fong et al., 1995),

or deleted tyrosine kinase domain, with reduced macrophage migration as the only reported phenotypic alteration (Hiratsuka et al., 1998). However, tumors grew slower in mice with deleted VEGFR-1 tyrosine kinase domain, maybe due to decreased monocyte/macrophage infiltration (Hiratsuka et al., 2001).

VEGFR-2 (also denoted Flk-1 (fetal liver kinase-1) in mouse and kinase domain region (KDR) in humans) is primarily expressed by EC precursors and blood ECs, but also by neuronal stem cells (Zachary, 2005). Deletion of the receptor in mouse is lethal at E8.5-9.5 with defective vascularization, illustrating its requirement for vascular development (Shalaby et al., 1995). The receptor was initially considered vital for angioblast commitment, but

Figure 3. The VEGFs bind the VEGFRs in an overlapping pattern. Arrows points at binding partners. Dashed lines indicate that prote-olytic processing is required for binding. Neuropilin-1, -2 and HSPGs are coreceptors for the VEGFs.

18

studies in zebra fish and differentiation of embryonic stem cells (ESCs) and careful analysis of embryos lacking the receptor have suggested the opposite (Childs et al., 2002; Schuh et al., 1999). It seems rather that the receptor is vital for migration, remodeling and survival of the EC progenitors. An im-portant role for VEGFR-2 is to regulate permeability of the adult endothe-lium. This leads to leakage of plasma proteins such as fibrinogen into the surrounding tissue which in turn has been suggested to promote angiogenesis and tumor growth (Dvorak et al., 1987).

The VEGFR-2 extracellular domain has seven immunoglobulin-like stretches of which the second and third compose the ligand-binding sites. In addition, the extracellular part may interact with heparan sulfate proteogly-cans (HSPGs) and possibly with neuropilins that both function as corecep-tors for the VEGFs (discussed below) (Dougher et al., 1997). The intracellu-lar part consists of a tyrosine kinase domain interrupted by a 70 amino-acid stretch. Within and around this kinase domain, 18 tyrosines are dispersed that may be phosphorylated upon dimerization of the receptor (the down-stream effectors will be discussed in detail below).

VEGFR-3 (alternatively denoted fms-like tyrosine kinase-4 (Flt-4) in mouse) is predominantly expressed on embryonic blood vasculature, lym-phatic ECs, monocytes and macrophages. Mice with deleted VEGFR-3 show vessel malformations and die at E9.5-10.5 due to cardiac failure (Dumont et al., 1998). The basic structure of VEGFR-3 is similar to VEGFR-1 and -2, apart from the substitution of the fifth immunoglobin-like domain by a disul-fide bridge.

VEGF-A The VEGFs are glycoproteins usually secreted as anti-parallel homo-dimers linked by cystein bridges. VEGF-A is transcribed into one pre-mRNA which

Figure 4. The major VEGF-A splice variants. Numbers indi-cate exons. Exon 3 encodes the VEGFR-1 binding site and exon 4 the VEGFR-2 binding site.

19

gives rise to at least six different human splice variants: 121, 145, 165, 183, 189 and 206, corresponding to the number of amino acids (Fig. 4) (reviewed by Robinson and Stringer, 2001). The murine forms are one amino acid shorter; VEGF-A120 etc. All isoforms are capable of binding both VEGFR-1 and -2, with distinct affinities. In general, VEGF-A165 seems to be the overall dominating form. However, it is not clear to what extent the respec-tive isoforms contribute to certain types of angiogenesis, i.e. pathological or developmental (Ng et al., 2001). The great importance of VEGF-A for vas-cular development was demonstrated in studies by Carmeliet et al. and Ferrara et al. 1996, where mice lacking only one vegf-a allele were able to form the dorsal aorta but showed abnormal vasculature and died at E11-12 (Carmeliet et al., 1996; Ferrara et al., 1996). Also, mice with homozygote deletion, generated by tetraploid aggregation, were even more severely af-fected with death at E9.5-10.5 (Carmeliet et al., 1996). The isoforms differ in their ability to bind heparin/heparan sulfate and neuropilin-1 and -2 with the longer forms displaying the highest affinity to these molecules (Gluzman-Poltorak et al., 2000; Jakobsson et al., 2006; Mamluk et al., 2002; Soker et al., 1998). Thus, VEGF-A121, that lacks the heparin/heparan sulfate binding domain, can travel further in tissue than VEGF-A189 since it does not stick to the extracellular matrix (ECM) (Park et al., 1993). However, VEGF-A121 and -A165 are equally potent in inducing VEGFR-2 activation, at least in porcine aortic ECs over-expressing the receptor (unpublished data). Previ-ously, VEGF-A121 has been suggested to be a poor activator of the receptor even in acute stimulations, which may be due to sub-optimal preparation of the different isoforms. Although equally potent in simplified in vitro models, large differences between the isoforms have been recorded in vivo. Mice genetically altered to express only VEGF-A120 display internal bleedings and deregulated patterning of the vasculature in the retina (table 1, p19-20). A combination of factors such as inability to shape gradients, or altered sig-naling properties due to lack of binding to HSPGs and neuropilin-1 and -2, might be reasons for these defects (Gerhardt et al., 2003; Jakobsson et al., 2006; Ruhrberg et al., 2002).

Receptor activation and signal transduction Ligand-binding to the second and third extracellular IgG-loops (out of seven) in VEGFR-2 induces dimerization of receptors leading to trans-phosphorylation of at least 8 tyrosines in the intracellular domain. Receptor activation is initiated by phosphorylation of tyrosines 1054, 1059 in the cata-lytic domain followed by phosphorylation of Y951 (Matsumoto et al., 2005), Y1175 and Y1214 (Takahashi et al., 2001) (Fig. 5). Each phosphorylated tyrosine constitutes a specific binding site for one or several downstream effectors. Knock-in mutation of tyrosine 1173 (1175 in human) to phenyla-lanine in mouse results in lethality at E9, which is similar to the effect of VEGFR-2 deletion itself, whereas exchange of Y1212 is of no consequence

20

(Sakurai et al., 2005). The phosphorylation of Y1173/1175 and the accom-panying down stream signaling may therefore be considered the prime path-way of VEGFR signaling during developmental angiogenesis. Phosphory-lated tyrosine 1175 binds and activates several effector molecules such as, phospholipase C- (PLC- ) (Takahashi et al., 2001), Shb (Holmqvist et al., 2004) and Sck (Warner et al., 2000). Out of these, PLC- seems to be the main regulator of EC proliferation by activation of protein kinase C and mi-togen-activated protein kinase (extracellular signal–regulated kinase 1 and 2 (ERK1/2)). Furthermore, phosphorylation of Y951 recruits VEGFR associ-ated protein (VRAP)/T-cell-specific adapter/Lad (Wu et al., 2000) involved in EC movement (Matsumoto et al., 2005). Other classical signal transduc-tion pathways involving ERK, focal adhesion kinase (FAK), phosphoinosi-tide 3 kinase (PI3K), and Src are also activated down stream of VEGFR-2. There is however growing evidence indicating that the cellular context dic-tates which pathway will be activated.

Figure 5. The intracellular part of VEGFR-2 and its major phos-phorylation sites. The outcome of each signaling cascade is indi-cated in the blue boxes. DAG, diacylglycerol; eNOS, endothelial nitric oxide synthase; FAK, focal adhesion kinase; HSP27, heat-shock protein-27; MAPK, mitogen-activated protein kinase; MEK, MAPK and ERK kinase; PI3K, phosphatidylinositol 3’ kinase;PKC, protein kinase C; PLC , phospholipase C- ; Shb, SH2 and -cells; TSAd, T-cell-specific adaptor. Adapted by permission from Macmillan Publishers Ltd: (Olsson et al., 2006), copyright 2006.

21

Receptor inactivation and degradation Even though activation of VEGFRs has been extensively studied, receptor down regulation and inactivation has received less attention. Naturally, this process is equally important since a continuously activated receptor would be unable to sense new signals. Internalization of RTKs other than the VEGFRs, like epidermal growth factor receptor (EGFR), is in contrast very well characterized and is known to require intrinsic receptor kinase activity. In brief, ligand-induced receptor activation leads to coated pit formation that facilitates the endocytosis of the complex. Depending on the state of ubiquit-ination, primarily mediated by c-Cbl E3 ubiquitin ligase in a phoshotyrosine specific manner (Duval et al., 2003), VEGFRs are either sorted to early en-dosomes, or alternatively to lysosomes for degradation. In the early en-dosome, the signaling complex is at least in part intact with persistent activ-ity (Lampugnani et al., 2006). Several reports suggest that signals from this compartment engage other pathways than those activated from the plasma membrane. Although it is likely that the basic mechanisms are shared by most RTKs, there are differences in the precise regulation; for example also inactive PDGFRs can be sorted to lysosomes (Wiley and Burke, 2001). The receptor activation can moreover be modulated or turned off by intracellular phosphatases (for a review see Kappert et al., 2005). The importance of phosphatases in the regulation of VEGFR signaling is however largely un-known.

VEGF coreceptors VEGF is the key activator of VEGFR-2 but the outcome of this activation is affected by the cellular context and the presence of coreceptors.

Neuropilin-1 is a transmembrane glycoprotein that function as a corecep-tor for VEGFR-2 (Soker et al., 2002; Soker et al., 1998). Embryos deficient in neuropilin-1 display reduced neuronal vascularization and development as well as disorganized yolk sac vascularization (Kawasaki et al., 1999; Kitsu-kawa et al., 1997). Ectopic overexpression of neuropilin-1 instead results in excessive angiogenesis and malformed hearts, in addition to abnormal ax-onal development (Kitsukawa et al., 1995).

HSPGs, which also function as coreceptors, will be discussed in detail be-low.

Table 1. Genetic modifications and their consequence in vascular biology

Genotype Phenotype Ref.

vegfr-2+/- Normal. (Shalaby et al., 1995)

22

vegfr-2-/- † E8.5–9.5, defective blood-island formation and vasculogenesis.

(Shalaby et al., 1995)

vegfr-3-/-

† E9.5-10.5, vasculogenesis and angiogenesis occurred, large vessels abnormally organized with defective lumens, fluid accumulation in the pericardial cavity and cardiovascular failure.

(Dumont et al., 1998)

vegfa+/-

† E11–12, defective vascular development. (Carmeliet et al., 1996; Ferrara et al., 1996)

vegfa-/-

† E9.5–10.5. Generated by aggregation of emb-ryonic stem cells with tetraploid embryos, more severe defects in vascular development than he-terozygote.

(Carmeliet et al., 1996)

vegfa188/188

Normal venular outgrowth but impaired arterial development. Defective vascularization sur-rounding the epiphysis. Dwarfism.

(Maes et al., 2004; Stal-mans et al., 2002)

vegfa164/164 Normal (Stalmans et al., 2002)

vegfa120/120

50% die shortly after birth owing to bleeding in multiple organs, and the remaining mice die be-fore postnatal day 14 owing to cardiac failure. Impaired myocardial angiogenesis, ischemic cardiomyopathy, skeletal defects, defects in va-scular outgrowth and patterning in the retina.

(Carmeliet et al., 1999)

pdgfb-/- Perinatal death, kidney defects and late stage-hemorrhages.

(Leveen et al., 1994)

pdgfr -/- Lethal shortly before birth, hemorrhagic etc. (Soriano, 1994)

LM- 5-null

† E14-17.5 with severe brain and limb defects and enlarged and disorganized placental vessels

(Miner et al., 1998)

LM- 4-null

Viable and fertile, but exhibit prenatal vascular bleedings, a phenotype that is gradually lost dur-ing development.

(Thyboll et al., 2002)

LM- 1-null

† E7 (Miner et al., 2004)

LM- 1-null

† E ~6 (Miner et al., 2004)

LM- 1-null

† E5.5 (Smyth et al., 1999)

fibronectin -null

† E11; defective notochord and somite forma-tion as well as vascular malfunctions

(George et al., 1997; George et al., 1993)

23

collagen IV-null

† E11; normal composition of the BM until E9.5, a mild vascular phenotype with irregular protrusion of capillaries into the neural layer

(Poschl et al., 2004)

Jagged-1-null

† E11; lack of obvious large vessels in the yolk sac, failure to remodel the primary plexus in the yolk sac, less intricate network and a reduced diameter of vessels in the head

(Xue et al., 1999)

Fibroblast growth factor receptor-1 and its ligands The fibroblast growth factor (FGF)-2 is a member of a family of 23 heparin-binding proteins (Yamashita et al., 2000). It is synthesized as a monomer but binds in a dimeric form, in complex with heparan sulfate, to the third IgG-loop (out of three) of FGFR-1 (McKeehan and Kan, 1994). The binding me-diates dimerization of the receptor, and thereby cross phosphorylation of tyrosine residues in the receptor intracellular domain. This in turn leads to conformational changes and binding of signaling proteins resulting in vari-ous cellular responses. Mice deficient in FGFR-1 die between E7.5 and E9.5 with reduced yolk sac mesoderm and reduced numbers of blood islands (Yamaguchi et al., 1994).

The basement membrane BMs are specialized ECM structures, 50-100 nm in thickness, that function not only as structural support and a barrier between different tissues, but also in mediation of signals affecting cellular behavior. The BM is a large and complex meshwork of interconnected proteins, such as type IV collagen, laminins (LMs), HSPGs (such as perlecan and agrin), nidogen/entactin and fibronectin (see Fig. 6). Both LMs and collagen IV have the ability to po-lymerize, and hence make up two separate sheets suggested to connect via the binding to nidogen (Smyth et al., 1999; Timpl and Brown, 1996). In ad-dition to interactions between these molecules they also bind a variety of cell surface molecules, thereby influencing attachment, differentiation, migra-tion, polarization or proliferation, depending on cell type and BM composi-tion (Timpl and Brown, 1996). The heterogeneity of BMs in tissue and dur-ing development is mostly distinguished by differences in LM isoforms pre-sent.

Although the vascular BMs have been structurally well characterized, there is much to learn about their function in developmental and adult angio-genesis. Blood capillaries are surrounded by a specialized BM, composed of collagen IV, LM-411, LM-511, perlecan, nidogen/entactin and fibronectin, which are deposited by ECs in conjunction with vessel formation. Study III

24

of this thesis asked at what stage this occurs and which components that may influence EC differentiation, organization and stability. Numerous studies have shown that proteins and proteolytic fragments of the BM can stimulate or inhibit the processes of both normal and pathological angiogenesis (Nyberg et al., 2005). This is rather expected since some of the main LM receptors (the family of integrins) are implicated in a variety of signaling pathways. For example, undifferentiated, pluripotent embryonic stem cells can be directed into different cell lineages depending on which matrix they are cultured on (Yamashita et al., 2000). In the following section each family of BM proteins will be described in more detail.

Laminins and the assembly of the BM The laminins are a family of heterotrimeric glycoproteins secreted by many cell types to become a main constituent of the BM. In mammals, there are 16 laminins (including major splice variants) known so far, each of them con-sisting of a particular combination of -, - and -chains (Aumailley et al., 2005) (Fig. 7).

The chains are joined in a trimeric -helical coiled-coil structure in the C-terminal tail (the long arm) that opens up in the N-terminal tail (the three short arms) to form a cross-shaped laminin of 400-900 kD in size. To date, 5

( 1-5), 3 ( 1-3) and 3 ( -3) chains, as well as additional splice vari-ants, have been identified (for review see Colognato and Yurchenco, 2000; Miner and Yurchenco, 2004). The C-terminal end of the -chain consists of five globular domains (LG1-5). These possess binding sites for several plasma membrane-associated molecules such as integrins, -dystroglycan,

Figure 6. Illustration of the BM on the cell surface. Polymerizing proteins form a sheet. Col-IV, Collagen IV; DG, / - Dystrogly-can; Nd, nidogen; Perl, perlecan. Adapted with permission (Li et al., 2003).

25

sulfated carbohydrates (S-CHO), fibulin-1 and sulfatides, but also for the ECM-associated molecule perlecan (for a review see Miner and Yurchenco, 2004). The coiled-coil domain has only been shown to interact with the se-creted proteoglycan agrin, via the LM- -chain. The N-terminal parts of all chains, except 3A and 4, harbor globular domains that bind similar do-mains on other LM molecules, in a calcium-dependent manner.

Figure 7. The major LM isoforms of the blood vascular endo-thelial BM. The trimers consist of one , one and one -chain.Binding partners and sites for interactions are outlined.

26

The multiple interactions mentioned above facilitate assembly of LMs on the cell surface, a property that plays a significant role in the formation of the complete BM. For example, in the absence of LMs, collagen IV is essen-tially excluded from the BM (Li et al., 2002 and study III). This may be due to that LMs normally bridge between the cell surface and the collagen net-work. However there is no direct connection between the LMs and colla-gens. Instead, these molecules bind independent sites on nidogens-1 and -2, which thereby may function as linkers (Smyth et al., 1999; Timpl and Brown, 1996). Mice with double gene deletions of nidogen-1 and -2 survive until birth. The BM formation is not severely affected in these mice, display-ing only slight reduction in LM, collagen IV and perlecan in capillary BMs (Bader et al., 2005). These results indicate that nidogens are not strictly needed for general embryonic BM formation, but may alter certain proper-ties of its function. This phenotype resembles mice with targeted deletion for the nidogen-binding site in LM- 1 (Willem et al., 2002). The HSPG perle-can, is also important for the integrity of the BM and has been shown to bind both LMs, nidogen-1 and -2, fibronectin and collagen IV in vitro (Costell et al., 1999; Hopf et al., 1999). It is also clear from studies of collagen IV-deficient mice that this particular collagen is dispensable for early embryonic BM assembly (Poschl et al., 2004). So far, LMs are the only BM molecules whose deletion severely affects the embryonic composition and assembly of the remaining BM constituents (Li et al., 2002; Smyth et al., 1998; Smyth et al., 1999). However, in study III in this thesis we show that deposition of fibronectin as well as of HSPGs occur in the vascular BM also in the ab-sence of LMs. This is in contrast to previous studies of an embryonic epithe-lial BM (Li et al., 2002). This discrepancy could be explained by differences in integrin expression on ECs and epithelial cells. Furthermore, addition of LM-111 resulted in a close to complete rescue of the epithelial BM assem-bly, whereas only a partial rescue was observed for the vascular BM (paper III). Li et al also demonstrated that LM-dystroglycan interactions are dispen-sable for LM assembly, but that interactions with HSPGs via LG4 are re-quired (Li et al., 2002). Interestingly, differentiating embryonic stem cells with a dominant negative mutation for fgfr are unable to deposit LMs, which results in failure of BM assembly (Li et al., 2001).

This exemplifies the interplay between cell signaling and BM protein se-cretion and assembly. Taken together these data suggest a role for LMs in cell differentiation, whereas collagen IV is more of a mechanical support.

In 2006 it was agreed upon a change of the LM nomenclature from the previous LN-1 through -15 to LMxyz, where x is the number of the -chain, y the number of the -chain and z the number of the -chain (Aumailley et al., 2005). For example proteins previously known as laminin-1, laminin-8 and laminin-10 is now denoted laminin-111, laminin-411 and laminin-511, respectively.

27

There are today many different antibodies directed against the specific LM chains, but none that specifically recognizes a complete LM-trimer. The present information on LM tissue localization is therefore based on com-bined immunostainings or in situ hybridizations for detection of the separate chains. It is therefore more accurate to refer to the LM-chains rather than to the complete LM trimer, if not further examined.

Laminins in vascular biology LM-111 is the first laminin to be detected in embryonic development with its

1- and 1-chains expressed at the two cell stage, and the 1-chain at the 8-16 cell stage (Aumailley et al., 2000). In the adult, the LM- -chain is re-stricted to BMs of epithelia, sinusoids in the liver and blood vessels within the central nervous system (Sasaki et al., 2002; Virtanen et al., 2000). Tar-geted deletion of the lama1 (encoding LM- 1) in mice leads to lethality at E7 due to failure in formation of Reichert’s membrane (Miner et al., 2004). The possible role of LM-111 in pathological angiogenesis is largely un-known, although several studies indicate that LM-111, or fragments thereof, affect EC function (study I, and reviewed by Hallmann et al., 2005)

LM- 4 (which combines with 1 and 1) is relatively restricted to vascu-lar BMs throughout development (Talts et al., 2000). Genetic deletion of this chain in mouse leads to vascular bleedings and hemorrhages during the pre-natal stage. One week after birth, the abnormalities are completely abolished and the mice are viable and fertile (Thyboll et al., 2002). It is believed that this phenotype is reversed by the deposition of LM- 5, a component of the other major vascular LM; LM-511. LM- 4 is also produced by various he-matopoietic cells with consequence for their function. For example, produc-tion of LM- 4 by leukocytes has been reported to enhance extravasation of neutrophils into the tissue, as demonstrated by reduced migration in the LM-

4-deficient mice (Wondimu et al., 2004). LM-411-release is furthermore stimulated by the inflammatory cytokines interleukin-1 and tissue necrosis factor- (Sixt et al., 2001). Knockdown of lama1 or lama4 (encoding LM- 1and – 4 respectively) in zebra fish embryos, did not affect intersomitic ves-sel growth. However, when both genes where simultaneously suppressed, the intersomitic sprouting was absent or severely delayed (Pollard et al., 2006). The authors suggest that the defects are due to absence of LM-containing substratum for EC migration in the intersomitic clefts. In accor-dance, LM-411 is known to enhance monocyte migration, which might ap-ply also to ECs (Wondimu et al., 2004). LM- 4 is truncated in the N-terminus and lacks the short arm. Its ability to polymerize is therefore re-duced, as compared to LM-111 and LM-511. Despite this fact, it assembles within the BM. Binding-partners are shown in Figure 7 (p23). LM- 4 has also been suggested to mediate EC survival in vitro, by activation of the 1integrins (DeHahn et al., 2004).

28

LM- 5 (component of LM-511) has been observed in murine BMs of large vessels from E13, and in capillaries, 3-4 weeks after birth (Frieser et al., 1997; Sorokin et al., 1997). Mice lacking LM- 5 die at E13-17.5 with severe brain and limb defects, enlarged and disorganized placental vessels, and failed renal glomeruli vascularization (Miner et al., 1998; Miner and Li, 2000). This chain is the only intact LM-chain known to bind integrin v 3,an interaction dependent on the globular domain L4b in the short arm (Sasaki and Timpl, 2001). In addition, a fragment of the 1-chain stimulates angiogenesis by binding integrins 5 1 and v 3 (Ponce et al., 2001).

The study on LMs in EC biology has been challenging due to the vast number of molecular interactions, the number of isoforms with overlapping functions, the potential functional contribution of proteolytic fragments and the variation between LM-preparations used (Wondimu et al., 2006). Despite such complications, the understanding of these molecules is now substantial. This has been facilitated in part by gene targeting in mouse (table 1, p19) and zebra fish and by the development of specific antibodies and high qual-ity recombinant proteins.

The integrins Integrins form a large family of dimeric transmembrane glycoproteins. At the time of writing, 18 - and 8 subunits have been recognized, composing 24 different integrins. As the extracellular domains bind components of the ECM such as collagens, LMs, and fibronectin, the integrin undergoes a con-formational change allowing recruitment and activation of intracellular bind-ing partners. ECs have been demonstrated to express at least integrins 1 1,

2 3 1, 5 1, 6 1, 8 1, 9 1, v 1, v 3, 6 4 and v 5 de-pending on EC type and state (for a review see Ruegg and Mariotti, 2003). Signals via integrins are dependent on integrin clustering which occurs in cell-cell, cell-matrix contacts (Clark and Brugge, 1995). All 24 integrins differ from each other in binding properties. On top of this complexity, the signals are influenced by the combination of the surrounding matrix and the cell type. For example the activation of FAK via integrins requires both ligand-binding and an intact cytoskeleton (Ruegg and Mariotti, 2003).

Glycobiology Sugars (saccharides) are essential for life by providing energy, structural support and by modulating cell signaling and protein structure. Paper II fo-cuses on a certain group of sugars that are members of the glycosaminogly-can (GAG) -family, and their role in cellular communication. These sugars are unbranched polysaccharides composed of repeated disaccharide units, present at high copy number on all cells. They are all (except for hyaluronan, se below) synthesized in the Golgi, attached to a core protein. The GAG chains and the protein, together make up a proteoglycan of either intracellu-

29

lar, trans-membrane, glycosyl phosphatidylinositol (GPI)-anchored or se-creted type. Frequent sulfate groups within the sugar backbone mediate bind-ing to a number of proteins, including growth factors and morphogens. De-pendent on the sugar composition, the GAGs are divided into four groups as follows: Heparin/heparan sulfate (HS), hyaluronan and keratan sulfate, chondroitin sulfate (CS) and dermatan sulfate. Since Paper II is focused on protein interactions with HS/heparin, the following sections will be dedi-cated to these molecules.

Heparan sulfate and heparin HS and heparin are unbranched polysaccharides of the GAG type, composed of repeats of up to 100 N- and O-sulfated disaccharide units (Fig. 8). HS/heparin are synthesized in the Golgi, covalently linked to a serine of a core protein. The number of sugar chains per protein can vary from one to

Figure 8. The synthesis and structure of heparan sulfate. Enzymes responsible for each step are indicated in the shaded boxes. EXT, exostosin family; NDST, N-deacetylase/N-sulphotransferase; OST, sulfotransferase.

30

10. The synthesis is initiated by the attachment of a xylose-galactose-galactose-glucuronic acid sequence to a serine of the protein. Subsequently, the HS polymer is generated by the action of HS polymerases (glycosyl transferases of the exostosin family (EXT)) that catalyze alternating addi-tions of N-acetylglucoseamin and glucuronic acid to the linkage region (for reviews see Esko and Lindahl, 2001; Hacker et al., 2005). This chain will then undergo a number of modifications: 1. In certain regions, N-acetyl groups are removed and replaced by sulfate groups. This is carried out by either of the four enzymes N-deacetylase/N-sulphotransferases (NDSTs)-1-4. 2. Transformation of glucuronic acid to iduronic acid (epimerization) pref-erably in N-sulfated regions. 3. 2O-, 3O- and 6O-sulfotransferases (OST) attach additional sulfate groups in the positions indicated by the names of the enzymes. As the sulfate groups are negatively charged these modifications will result in a highly negatively charged, hydrophilic molecule. The degree and pattern of sulfation, that depend on the activity of the different enzymes described, are highly variable, in a tissue-, cell-, and age-dependent manner (Dennissen et al., 2002; Feyzi et al., 1998). Heparin undergoes heavy modi-fications resulting in a higher degree of sulfation than is the case for HS.

Proteins with attached HS-chains are denoted HS proteoglycans (HSPGs). HSPGs can be divided into four main types depending on the core protein (Fig. 9.): 1. The intracellular, stored in granules (serglycin, carrying heparin in connective tissue-type mast cells (Kolset and Gallagher, 1990)). 2. The glypicans (six members, glypican 1-6), which are linked to the cell surface

Figure 9. The different forms of HSPGs are indicated. 1. Intracellu-lar, lower, 2. Membrane-bound glypicans (GPI-anchored), left, 3.Transmembrane. syndecans, right (the dashed lines indicate chon-droitin sulfate chains), 4. Secreted forms, upper.

31

by a glycosyl phosphatidylinositol (GPI)-anchor. 3. The transmembrane syndecans (four members, syndecan 1-4), and 4. Soluble HSPGs (three ma-jor members, perlecan, agrin and collagen XVIII) which are released from cells and deposited in the extracellular matrix. The ectodomains of both syn-decans (Fitzgerald et al., 2000) and glypicans can be shed from the cell sur-face by various enzymes. Several studies indicate that shedding might be of great importance for cell communication (Kreuger et al., 2004). In accor-dance, Kato et al. present indications for diverse signaling properties of soluble contra immobilized HS. The immobilised ectodomain of syndecan-1, prepared from wound exudates, stimulated heparin-mediated FGF-induced cell proliferation whereas the soluble ectodomain opposed this action (Kato et al., 1998). Such differences are further demonstrated in paper II in which only cell surface HS was able to rescue VEGFR-2-signaling in HS-deficient ECs.

HS/protein interactions and lessons from knockout studies Almost all cells in the body express HSPGs. HS or heparin have been shown to interact with over 100 proteins, including several growth factors and morphogens, such as VEGF, FGF, Wnt and Hedgehog (Burri and Djonov, 2002; Chen et al., 2004). The impact of these molecular interactions on cell biology is diverse. Formation of morphogen gradients is essential for proper embryogenesis. Here, HSPGs of both the glypican and syndecan families are known to play an important role in restricting diffusion of these proteins (Belenkaya et al., 2004; Han et al., 2005; Kirkpatrick et al., 2004; Kreuger et al., 2004; The et al., 1999). HSPGs also take part in formation of signaling complexes, modulating their stability (Yayon et al., 1991) and membrane trafficking (paper II: Jakobsson et al., 2006). Furthermore, HSPGs might function to protect proteins from proteolytic degradation (Rosengart et al., 1988). Whether this HS-protein binding is dependent on specific patterns of sulfation of the saccharide backbone, or if it simply relies on charge density, has been heavily debated (reviewed by Kreuger et al., 2006). In vivo data in favor of the net charge theory were presented by Kaminura and colleagues (Kamimura et al., 2006). By genetic alterations in Drosophila they illustrate that FGF-function depends on the HS net charge, regardless of whether it carries 6O- or 2O-sulfation. Also, Kreuger et al. demonstrated equal binding of FGF-2 to oktasaccharides irrespective of the positioning of sulfate groups, given that the number of sulfate groups was constant (Kreuger et al., 2005). There was on the other hand a strict correlation to the degree of sulfation. Although several studies favor the critical role of charge-density depend-ence, we cannot exclude the possibility of sequence-dependence in situations not yet investigated.

It is unclear however to what extent the core proteins themselves contrib-ute to signaling. The importance of the core protein was illustrated by a mu-

32

tation of syndecan-1 to eliminate certain HS-attachment sites (Langford et al., 1998). The syndecan-1-dependent cell-cell interactions and cell adhesion were affected even when only one single site was mutated. In contrast, no clear phenotype was observed in mice with deleted syndecan-1 under normal conditions (Alexander et al., 2000). In paper II, we show that HS-mediated signaling via the core protein (syndecans) in ECs, is dispensable for vasculo- or angiogenesis. Taken together, these data suggest that these particular HSPGs primarily function as co-receptors. However, targeting syndecan-2 by morpholinos or overexpression of a syndecan-2 with a truncated intracel-lular domain, result in defective angiogenic sprouting in zebra fish develop-ment (Chen et al., 2004). This suggests a role for the core protein itself in signaling.

The requirement for HS in development is exemplified by a number of genetically altered mice. Gene targeting of the glycosyl transferase EXT1 results in death at E8.5 with total absence of HS but increased chondroitin sulfate synthesis (Lin et al., 2000). Inactivation of the Ndst1 gene leads to a reduction in the degree of N-sulfation of HS. Most embryos lacking this enzyme survive birth but die shortly thereafter in a condition resembling respiratory distress syndrome (Ringvall et al., 2000). The Ndst2-/- mice sur-vive and are fertile (Forsberg et al., 1999). Elimination of Ndst2-/- results in reduction of heparin sulfation with abnormal mast cell function as a conse-quence. Combined deletion of Ndst-1 and -2 is accompanied by lethality at E5.5 (unpublished data, Kjellén and Holmborn). Embryonic stem cells gen-erated from blastocysts derived from these mice, synthesized HS without N-sulfation but with low amounts of 6O-sulfation (Holmborn et al., 2004). The early lethality of EXT1 and Ndst1/2 knockout mice prohibited analysis of the general HS contribution to vascular development. In paper II, we addressed this issue employing embryonic stem cells deficient in both NDST-1 and -2 expression. These embryonic stem cells cannot, unlike wild type cells, re-spond to VEGF by the formation of vascular structures. However, presenta-tion of HS to the HS-deficient ECs by adjacent cells, rescues vascular devel-opment and VEGF-induction. This will be further discussed below (see pre-sent investigation).

It is clear that invaluable information can be retrieved from gene dele-tions. It is however well known that defects due to gene deletions can be compensated by upregulation of other genes. It should therefore be empha-sized that care should be taken in the analysis and interpretations of knock out models. Also, most genes are involved in multiple actions that might affect a variety of cells. Given this, one should be aware of possible secon-dary effects.

33

Embryonic stem cells, embryoid bodies and vascular biologyIn 1981, Bradley et al. and Martin et al. presented protocols for the culture of ESCs in vitro, without loss of their pluripotency (Bradley et al., 1984; Evans and Kaufman, 1981; Martin, 1981). A few years later, targeted deletion of genes by homologous recombination in ESCs was described (Thomas and Capecchi, 1987). These cells were soon thereafter shown to give germ line transmission, allowing generation of transgenic mice (Thompson et al., 1989). These techniques have been invaluable to all fields of cell biology, and since then, numerous knockout animals have been demonstrated. Gene targeted ESCs can be isolated from blastocysts of such mice, and cultured in vitro. If these cells are let to differentiate in suspension, they will form sphe-roids, denoted embryoid bodies. Within the embryoid bodies cells will dif-ferentiate to a variety of cell lineages, including ECs. Development in em-bryoid bodies has been shown to constitute a useful model for the study of vasculo- and angiogenesis (Bloch et al., 1997; Byrd et al., 2002; Doetschman et al., 1985; Feraud et al., 2001; Ferrara, 2002; Magnusson et al., 2004; Vit-tet et al., 1996). Vascular development in the embryoid bodies share many features with development in vivo. For example, the order of appearance of a number of vascular markers (VEGFR-2, CD31, VE-cadherin, Tie-1 and Tie-2) resembles that observed in vivo (Vittet et al., 1996). Furthermore, EC structures are surrounded by supportive smooth muscle cells and/or pericytes (paper II-IV; Lindskog et al., 2006). At later stages of embryoid body differ-entiation (day 10-12 after initiation), the ECs form lumenized vessels en-closed in a BM composed of LMs, fibronectin, collagen IV and HSPGs (pa-per III). In paper I-III we have cultured embryoid bodies either in a rather two-dimensional state on a tissue culture dish (vasculogenesis), or in a three-dimensional collagen-I matrix (invasive angiogenesis). When cultured in the two-dimensional setting vasculo-/angiogenesis are enhanced by addition of PDGF-BB, FGF-2, VEGF-A165 or VEGF-A121, each giving raise to a char-acteristic vessel morphology. In collagen however, only VEGF-A165 is able to induce angiogenic sprouting with the setup used in paper II and III, al-though induction by treatment with FGF-2 has been observed in a similar system (Feraud et al., 2001).

Mice lacking either VEGFR-2, NDST-1 and -2 or LM- 1 are embryoni-cally lethal before or during early vasculogenesis. We have however been able to study vascular development employing ESCs lacking these genes, in the embryoid body model. This demonstrates the potential of this in vitro differentiation model (paper II-IV).

34

Present investigations and future perspectives

Paper ILaminin-1 promotes angiogenesis in synergy with fibroblast growth factor by distinct regulation of the gene and protein expression profile in endothelial cells

AimRemodeling of the BM of ECs is a prerequisite for angiogenesis. A major constituent of BMs are the LMs, which have been shown to affect EC biol-ogy in various ways (Hallmann et al., 2005). Previous observations that LM-111 (denoted LN-1 in paper I) affected EC morphology in vitro had not been explained; we therefore wanted to investigate possible molecular mecha-nisms for this action. Also, overexpression of the LM- 1-chain in a human colon carcinoma results in abundant angiogenesis and enhanced tumor growth of xenotransplants in mice, indicative of LM involvement in patho-logical angiogenesis (De Arcangelis et al., 2001). We wished to further in-vestigate the role of LMs in EC signaling and differentiation.

ResultsWe tested the effect of LM-111 on the chorio-allantois membrane of the chicken. We applied filter discs soaked in LM-111 with or without FGF-2 and noted that LM-111 and FGF-2 induced angiogenesis to the same extent. Furthermore, differentiating embryonic stem cells were exposed to the same combination of proteins at various concentrations. At low concentrations, neither LM-111 nor FGF-2 induced an increase in endothelial structures. A combination of the two at the same concentration, however, promoted the relative EC population.

To exclude possible indirect effects by other cells in the cultures, we in-vestigated differentiation of an EC-line. Immortalized mouse brain endothe-lial (IBE) cells were seeded between two layers of collagen to induce forma-tion of tube-like structures. Inclusion of FGF-2 and LM-111 or both in the medium for seven hours, induced formation of tubules that were absent in the untreated control. After 24 hours, IBE cells treated with LM-111 alone

35

had undergone apoptosis, whereas treatment with FGF alone and together with LM-111 still allowed tube formation. The tubes co-treated with FGF and LM-111 were longer and less branched than with FGF alone, and seemed to have reached a higher degree of differentiation as cell-cell borders were hardly distinguishable. This implicated a role for LM-111 in promotion of differentiation, but not survival of ECs in this system. Only LM- 1 and -

5 were slightly enhanced after FGF-stimulation, which may exclude any major contribution of endogenous LMs to the initiation of EC differentiation. However, several LM-chains were detected, indicating that they still might be required for this process. Many studies have suggested that Notch4 and its ligand Jagged-1 play a central role in regulation of EC differentiation. We observed a transient upregulation of mRNA and protein-levels of Jagged-1 during tube formation of the IBE cells after treatment with LM-111 alone, FGF alone and combined. LM-111 had no effect on phosphorylation of ei-ther ERK or Akt/protein kinase B. On the other hand, LM-111 induced an up-regulation of FGF-2 and FGFR-1 transcripts. Co-treatment with LM-111 and FGF-2 induced a transient increase in the number of fgf2, vegfa, fgfr1and vegfr-2 transcripts at 3 h with a drop at 7 h. This indicates an intriguing interplay between cytokines and components of the BM during differentia-tion of ECs.

DiscussionFrom data presented in paper I, it is not possible to make a statement on whether Jagged-1 triggers differentiation, or if the up-regulation is rather a consequence of differentiation. The regulation of the Notch ligands Jagged-1 and Delta-like 4 is still largely unexplored although their involvement in EC biology is apparent (Xue et al., 1999). Interestingly, in a recent publication by Hellström et al., activation of Notch by Jagged-1 results in decreased number of tip cells in the developing retina (Hellström et al., 2007). Similar observations were made concerning intersomitic EC-sprouting in the devel-oping zebra fish (Siekmann and Lawson, 2007). It seems likely that Notch signaling contributes to vascular differentiation and maturation rather than to active growth, similar to ECs in collagen treated with LM, in paper I (for a review on the family of Notch receptors and ligands in EC biology see Iso et al., 2003).

Although LM-111/FGF-induced tube formation involves regulation of Jagged-1, signaling via integrins are likely to be of great importance, func-tioning either as upstream or down stream mediators. Here, at least 3- and

6-integrins seem not to be central to the differentiation, since blocking antibodies had little or no effect in the collagen assay.

It is known from previous studies that exogenous addition of LM-111 at least in part can rescue BM formation in cells lacking LM (Li et al., 2002; Li et al., 2001 and paper II). However, we do not know if exogenous addition

36

of LM to ECs in collagen contributes to BM formation or if it rather com-petes with endogenous polymerized proteins for the binding to integrins. FGF-2 clearly induced EC morphogenesis in all models used in paper I. Sur-prisingly however; embryoid bodies with deleted FGFR-1 display an in-creased EC population, despite the fact that FGF-2 induced vascular devel-opment in the wild type (Magnusson et al., 2004). It might be, that the base-ment membrane composition is altered in the FGFR-1-deficient cells, since FGFRs are needed for LM secretion (Li et al., 2001).

Future perspectives It is still not clear precisely at what stage during EC differentiation that Jag-ged-1/Notch signaling is critical. Such knowledge would increase our under-standing of EC differentiation.

Paper II Heparan sulfate in trans potentiates VEGFR-mediated angiogenesis

AimHSPGs bind numerous proteins, thereby modulating the signaling properties for some of them. Since certain VEGF-A isoforms are known to bind hepa-rin we wished to investigate if, and possibly how, HSPGs modulate VEGF-signaling in vasculo- and angiogenesis. To address these issues we employed embryonic stem cells deficient in the enzymes NDST-1 and 2 with severely reduced sulfation of HS and heparin as a consequence.

ResultsEmbryonic stem cells lacking NDST-1 and 2 produce HS with deficient sulfation. We show that this HS is unable to bind VEGF-A165, in contrast to wild type-derived HS. Furthermore, these ESCs failed to differentiate to form vessels; a phenotype which could not be rescued by addition of heparin or HS. VEGFR-2-deficient ESCs are, like the Ndst1/2-/- cells, unable to form vascular structures during differentiation. However, when these cell-types were mixed at the stem cell stage to generate chimeric embryoid bodies, the formation of vessel structures was at least as prominent as for the wild type cells. In addition, the chimeric cultures responded to VEGF-A165 with the characteristic formation of a ring-like structure of ECs in the embryoid body

37

periphery, when cultured on a flat surface, and by sprouting when cultured in three-dimensional (3D) collagen gels. The endothelial sprouts in the collagen matrix expressed VEGFR-2 and were hence derived from the Ndst1/2-/- cells. This indicates that EC-derived HS is dispensable for vascular development in this setting. In the wild type cultures, HS was expressed by all cells as indicated by immuno-fluorescent staining. In the chimeric embryoid bodies, the ECs lacked sulfated HS, whereas adjacent pericytes expressed high lev-els. To further investigate the HS contribution to invasive angiogenesis, we introduced VEGF-A121 that was shown not to bind heparin or HS. VEGF-A121 failed to induce vascular sprouting in the collagen assay. Furthermore, treatment of embryoid bodies in collagen with heparinase I-III to cleave HS into small inactive fragments reduced VEGF-A165-induced sprouting of chimeric embryoid bodies. These data further indicate that ligand/HS bind-ing is required for proper signaling. We then investigated possible differ-ences in VEGFR-2 signaling dependent on whether HS was positioned either in trans or in cis to the receptor. Stimulation of wild type embryoid bodies with VEGF-A165 resulted in a transient phosphorylation of the receptor and a decrease in the total receptor levels after 1 h. In the chimeric embryoid bodies, with HS presented in trans, there was instead a potent phosphoryla-tion that persisted for at least 6 h. This was accompanied by an increase in the total receptor level over a 2 h time period. Despite considerably lower amounts of total receptors in chimeric cultures compared with wild type, the phosphorylated protein levels were in the same range. This indicates that a higher percentage of the receptors is activated and/or that each receptor is tyrosine phosphorylated at increased levels (or stochiometry) when HS is presented in trans.

These data suggest that the composition of HS, as well as its location in relation to VEGFR-2, modulate both signaling potency and kinetics. Thus, interaction of HS with the VEGF receptor and the co-receptor neuropilin-1 seems to be of greater interest than previously considered.

DiscussionIn vivo, adjacent cells may display differences regarding their respective HS epitopes (Dennissen et al., 2002). In paper II, we have investigated the situa-tion in which one cell (the pericyte) expresses high amounts of “normal” HS, whereas the neighboring cell (the EC) expresses HS with deficient sulfation. This setting will maximize the interactions in trans between HS and the ligand/receptor-complex, and prevent interactions in cis. In normal tissue, it is likely that there is a competition between HS on the adjacent cell, HS in matrix and HS on the same cell, for interaction with the signaling compo-nents. It should be emphasized that HS interacts not only with the ligand but also with the receptor itself and the neuropilins.

38

The importance of cell-cell or cell-matrix interactions in RTK signal transduction has received relatively little attention. This is not true for the cadherins, molecules that interact in a trans-cellular fashion via homophilic binding, which have been carefully studied in this aspect. Suyama et al. demonstrated reduced internalization of FGFR with potentiated signaling, in the presence of N-cadherin (Suyama et al., 2002). No mechanism responsi-ble for this phenomenon was described. However, it is tempting to speculate that the reciprocal binding of N-cadherins, which also interacts with FGFRs, is responsible for the decrease in internalization observed in the presence of N-cadherin. Also, VE-cadherin is known to regulate internalization of VEGFR-2 as demonstrated by (Lampugnani et al., 2006). When VE-cadherin is localized at junctions, VEGFR-2 is internalized at a lower rate. In the absence of VE-cadherin, VEGF-stimulation results in rapid internaliza-tion of the receptor into endosomes where it remains active (as measured by PLC- activity). This indicates that a prolonged localization at the plasma membrane not necessarily leads to enhanced signaling and that the outcome of receptor activation is highly dependent on the cellular context.

The duration of a ligand-bound receptor at the plasma membrane may be crucial for the outcome of the cellular response. When VEGFR-2 binds its ligand, the intracellular kinase domains will be brought in close proximity to their substrates and trans-phosphorylation of tyrosine residues will occur. The phosphorylated tyrosines will subsequently bind different molecules that each may initiate down-stream signaling that result in specific cellular re-sponses (migration, proliferation, survival etc.). The cellular response is known to be dependent on the cellular context. Since the phosphotyrosine-binding partners exist at different concentrations and with varying affinities for their binding-sites, it is likely that this binding is time-dependent. This would cause certain pathways to be initiated faster than others, meaning that activation of the slower pathways requires persistent activation of the recep-tor. If such time-dependency is of biological significance to RTK signaling remains to be investigated.

In conclusion, this implies that cell-cell contacts are central to signal transduction and that care should be taken regarding cell confluency, and model setup. Furthermore, differences in quality as well as potency of sig-nals may be missed when only one time point is analyzed.

In addition, matrix components and coreceptors are known to affect ligand/receptor affinities and possibly conformations which can modulate RTK signaling. HS/FGF/FGFR interaction is perhaps the most studied in this aspect. Heparin has been demonstrated to increase the ligand/receptor affin-ity and thereby decreasing the off-rate of the signaling complex (Ibrahimi et al., 2004; Yayon et al., 1991). In accordance, certain tyrosines in the receptor intracellular domain were phosphorylated only when the ligand was intro-duced together with heparin (Lundin et al., 2003).

39

The data presented in paper II suggest that interactions with HS potentiate VEGFR-signaling and hence minimizes the amount of VEGF needed to me-diate survival. As discussed previously, few antiangiogenic substances have proven valuable in the treatment of cancer. Notably, it has not been eluci-dated if the treatments fully inhibit VEGFR-2-mediated signaling. For even more efficient inhibition it might therefore be beneficial to include soluble molecules that could compete with cell- and matrix-bound HS for the inter-action with the ligand/receptor. However, it is likely that HS or heparin frag-ments that interact with VEGF and/or its receptor also would intervene with other signaling cascades, with possible side-effects as a consequence.

Future perspectives We do not know if the suggested mechanism for regulation of VEGFR-signaling by HS concerns also other heparin/HS-binding ligand/receptor systems, although this is tempting to suggest. Here, the PDGFRs and ligands are clearly of interest. The possible time-dependency for initiation of certain signaling cascades is very interesting and may add in yet another dimension in cell signaling. In the long run, concepts should be proven in vivo, and we will therefore try to generate chimeric embryos from vegfr2-/- and Ndst1/2-/-

ESCs.

Paper III Laminins regulate vascular lumen diameter

AimIt is known that LMs may affect EC differentiation, attachment, migration and survival in an isoform-specific manner. We aimed to study vascular development, including stability, survival, lumen formation and pericyte recruitment, in a LM-deficient model. We also intended to analyze potential LM-isoform-specific effects on EC development by addition of purified LMs.

ResultsCharacterization of vascular basement membranes in wild type embryoid bodies revealed the presence of LMs (including LM- 4- and LM- 5-chain), collagen IV, HSPGs and fibronectin. We generated embryoid bodies, de-rived from ESCs null for the LM- 1-chain, to study vascular development. It has been previously shown that no LM deposition occurs in the absence of

40

LM- 1 and as a consequence, BMs fail to assemble (Li et al., 2002). Vascu-lar development proceeded with mild defects in EC differentiation, which could be rescued by exogenous addition of LM-111. Immunostaining for LMs revealed only intracellular deposits in the absence of LM- 1. There was however an increase in fibronectin deposition. In contrast, collagen IV was nearly completely lost from the vascular BM. VEGFA-165-induced vascular sprouting in a 3D collagen matrix was similar to wild type regarding length and number of vascular sprouts, as well as pericyte coating. Moreover, it was evident that LMs were produced and deposited from the stalk to the filopo-dia of tip cells at the front of invading sprouts. Furthermore, the kinetics of vessel deterioration was similar in wild type and LM- 1 null embryoid bod-ies following VEGF withdrawal. The number of vessels with large (>36 µm) lumen diameter was increased 4-fold in the absence of LM-deposition, suggesting a regulatory function independent on pressure. We conclude that LM is deposited at all stages of angiogenesis and that this deposition is dis-pensable for vasculogenesis but required for fine-tuning of the 3D vessel; a mechanism independent of flow.

DiscussionDeletion of sleepy (analogous to lamc1 (encoding LM- 1) in mouse) in zebra fish results in defective intersomitic sprouting, suggesting a defect either in differentiation or migration capacity (Parsons et al., 2002). From data pre-sented in paper III it is evident that ECs can migrate and organize in the ab-sence of LMs. It might be that LMs that normally cover the edges of the somites are required for guidance, and function as a substrate for EC migra-tion. This type of strict patterning does not exist in the embryoid bodies. From paper III, it is apparent that LMs are not required for EC migration per se. Furthermore, pericytes are equally well attracted to ECs and disperse over the sprout length regardless of LM presence or absence. It is still possi-ble that there is a redundancy between some of the LMs and fibronectin, suggesting that the increase in fibronectin levels in the absence of LM- 1could compensate for the lack of LMs. In this paper we show that vessels grow wider in the absence of LM deposition although no flow exists. It is likely that the introduction of flow and pressure would add on to this effect.

Future perspectives We would like to investigate the mechanism responsible for the increase in vessel diameter using different matrices and integrin-blocking antibodies. Another interesting pursuit would be to generate animals from the lamc1-/-

ES cells by tetraploid aggregation, to investigate the impact of pressure and flow on vessels lacking a BM.

41

Paper IV Building blood vessels: Stem cell models in vascular biology

AimWe wished to summarize the current state of the use of embryonic stem cell differentiation as a model for vasculo- and angiogenesis, and to discuss ad-vantages and shortcomings in relation to other systems.

ResultsWe conclude that the embryoid bodies constitute a suitable model for the study of vascular development. Furthermore it allows examination of signal transduction with close resemblance to the in vivo context.

Future perspectives We will continue to refine the techniques for live imaging in the embryoid bodies. To facilitate this, we will use lentiviral-mediated introduction of various fluorescent reporters for cell lineage tracking. Moreover, we plan to generate stable cell lines from embryoid body cultures to allow expansion of certain cell types.

42

Concluding remarks

The intention of this introduction has been to present facts and data that are closely related to papers I-IV. The number of processes involved in angio-genesis is enormous, only a fraction could therefore be represented in this thesis.

The studies presented in this thesis have provided some answers, but also generated many questions concerning the meaning of the interplay between the molecules involved in angiogenesis. It will be very exciting to see what the scientific “evolution” will bring in the future.

43

Sammanfattning på svenska

Samtliga celler i kroppen behöver syre och näring för att överleva. För att tillgodose detta behov pumpas blodet ut till vävnaderna via blodkärlen. I de finast förgrenade blodkärlen, de så kallade kapillärerna, sker gas- och när-ingsutbytet mellan blodet och vävnaderna. Här kan även celler engagerade i immunförsvaret vandra ut igenom kärlväggen för att bekämpa en eventuell infektion. I och med att celler kräver energi för att leva och dela sig (prolife-rera) kan brist på blodkärl verka begränsande för vävnadstillväxt. Då en vävnad börjar växa, som till exempel efter kraftigt muskelarbete, kommer cellerna att uppleva ett visst mått av syrebrist. Detta sätter igång en produk-tion av signalmolekylen VEGF (vascular endothelial growth factor) som i sin tur stimulerar nybildning av blodkärl, en process benämnd angiogenes.

I många patologiska tillstånd såsom cancer, reumatism och diabetes reti-nopati, sker en överproduktion av blodkärl vilken i sig bidrar till utveckling av sjukdomen. I andra tillstånd, såsom diabetiska fot- och bensår, är situatio-nen den omvända med otillräcklig angiogenes och cirkulation. Dessa sjuk-domar är inte bara livshotande och smärtsamma för den enskilda patienten, utan slukar också en ansenlig del av vårdbudgeten. Det är därför viktigt ut-öka vår kunskap för att i förlängningen utveckla läkemedel och förbättra metoder för att såväl stimulera som hindra angiogenes. Framsteg har redan gjorts men än finns stora luckor i vår förståelse kring denna process.

Blodkärl är uppbyggda av endotelceller samt glatta muskelceller som är förankrade i ett basalmembran bestående av ett nätverk av bl a lamininer, kollagen IV, fibronektin och heparansulfatproteoglykaner (HSPG). Ett antal varianter (isoformer) av laminin har påvisats i kroppen, men olika isoformers eventuella roll i utvecklingen av det vaskulära systemet har i stort varit okänd. I arbete I visar vi att laminin-111, ensamt eller i samspel med den angiogena faktorn fibroblast growth factor-2, stimulerar mognadsprocessen av både isolerade endotelceller och endotelceller i embryonala stamcellskul-turer. I arbete III kan vi studera angiogenes i frånvaro av lamininer i basal-membranet och kan konstatera att lamininer inte är nödvändiga för angioge-nes, men att de påverkar kärlens tjocklek.

HSPG är proteiner kopplade till långa oförgrenade sockerkedjor (heparan-sulfat) som finns i tusental på i stort sett alla celler i kroppen. Det är känt sedan tidigare att HSPG kan påverka cellsignalering genom att förändra mo-lekylers affinitet till varandra. I detta arbete visar vi för första gången att sockrets position i förhållande till cellsignaleringskomponenterna har en

44

avgörande betydelse för styrkan och varaktigheten av signalen via ett tyro-sinkinasreceptorsystem. Detta är en ny mekanism för reglering av cellsigna-lering och poängterar vikten av cell-cellkontakter i cellsignalering.

Vi vet ytterst lite om basalmembranproteiners betydelse för endotelcells-differentiering, polarisering samt stabilitet och regeneration. Vi visar i arbete III att endotelceller kan differentiera, polarisera och bilda lumen i frånvaro av samtliga lamininisoformer. Vidare ser vi att lamininer produceras i alla stadier av den angiogena processen och att de reglerar kärldiametern obero-ende av tryck.

För att studera biologiska processer behöver vi modellsystem som är väl lämpade för respektive studie. Vi sammanfattar i arbete IV fördelar respekti-ve nackdelar med att använda embryonala stamceller som modell för studier av angiogenes. Vi kan se att utvecklingen av blodkärl i stamcellskulturer delar många egenskaper med utvecklingen in vivo. Exempelvis uppkommer proteiners uttryck i samma följd som in vivo. Dessutom bildas ett nätverk av endotelceller som är omgivna av ett basalmembran och pericyter. Så små-ningom bildas håligheter (lumen) i nätverket. Genom att förändra det gene-tiska materialet i stamcellerna är det möjligt att undersöka specifika geners roll i angiogenes. Ett annat sätt är att tillsätta isolerade substanser eller prote-iner för att sedan följa angiogenesen.

45

Acknowledgements

This work has been carried out at the Department of Genetics and Pathology at the Rudbeck laboratory at Uppsala University. I would like to thank eve-ryone that has supported me over the years with knowledge, enthusiasm and laughter. I would especially like to thank:

My supervisor, Lena Claesson-Welsh. You have given me insight into most areas in science, but I believe, also anything in its vicinity. Efficiency at its perfection and supportiveness beyond limits.

The friend who turned out to become my co-supervisor, Johan Kreuger. You make me jump into unknown waters (from 10 meters, was it?). Together the ideas keep popping up. I have now a reasonable amount of great poems on my phone, you might, like I do, consider them as a backup in case of system error. Thanks for the great collaborations and for your guidance.

Johan Dixelius; for previous engagement in the role as co-supervisor and laminin-expert. You really gave me a “godis” start in the lab. There is never a bad time for a scientific discussion with you.

Lena Kjellén, Katarina Holmborn and Inger Eriksson for a wonderful col-laboration on the “HSPGs across the way” (as well as other projects).

Peetra, for the brilliant introduction to the “body building” and immunostain-ing, and for showing that no upper speed limit exists. The tremendous joy of sitting back to back in the cell culture room, singing and maybe dancing…

Dan, for the knowledge and ideas you share with me: science, economics, reggae, matters of the heart and so much more. For being a true friend.

Chunsik, for all the fun, and for having the privilege of saving you from the frozen lake.

Lars L for discussions and ideas on HS and baking.

All former and present members of the vascular biology unit, for the scien-tific sharing, and for creating a warm and crazy international atmosphere.

46

Jan Grawé for excellent guidance and engagement in “time-lapse” and con-focal microscopy.

The co-authors for generous collaborations.

All people at Rudbeck and in particular the “innebandy” crowd. Maybe the best hour of the week.

Daniel Edgar, the scientific explorations at ÖG, the lab sessions, the moun-tain streaking and the tremendous fun at all times. For all the delicious din-ners, thanks also, Hanna.

All friends in “Sånggripen”. Endless joy and laughter, what a crowd of fan-tastic people. Joining the choir is one of the best experiments I have made.

My family, for love at all times.

My love, Christina/Nina/Kicki/Babylull, for your love and support that mean more than anything else to me. For all the everyday laughter.

47

References

Alexander, C.M., F. Reichsman, M.T. Hinkes, J. Lincecum, K.A. Becker, S. Cumberledge, and M. Bernfield. 2000. Syndecan-1 is required for Wnt-1-induced mammary tumorigenesis in mice. Nat Genet. 25:329-32.

Aumailley, M., L. Bruckner-Tuderman, W.G. Carter, R. Deutzmann, D. Edgar, P. Ekblom, J. Engel, E. Engvall, E. Hohenester, J.C. Jones, H.K. Kleinman, M.P. Marinkovich, G.R. Martin, U. Mayer, G. Me-neguzzi, J.H. Miner, K. Miyazaki, M. Patarroyo, M. Paulsson, V. Quaranta, J.R. Sanes, T. Sasaki, K. Sekiguchi, L.M. Sorokin, J.F. Talts, K. Tryggvason, J. Uitto, I. Virtanen, K. von der Mark, U.M. Wewer, Y. Yamada, and P.D. Yurchenco. 2005. A simplified laminin nomenclature. Matrix Biol. 24:326-32.

Aumailley, M., M. Pesch, L. Tunggal, F. Gaill, and R. Fassler. 2000. Altered synthesis of laminin 1 and absence of basement membrane compo-nent deposition in (beta)1 integrin-deficient embryoid bodies. J Cell Sci. 113 Pt 2:259-68.

Bader, B.L., N. Smyth, S. Nedbal, N. Miosge, A. Baranowsky, S. Mokka-pati, M. Murshed, and R. Nischt. 2005. Compound genetic ablation of nidogen 1 and 2 causes basement membrane defects and perinatal lethality in mice. Mol Cell Biol. 25:6846-56.

Belenkaya, T.Y., C. Han, D. Yan, R.J. Opoka, M. Khodoun, H. Liu, and X. Lin. 2004. Drosophila Dpp morphogen movement is independent of dynamin-mediated endocytosis but regulated by the glypican mem-bers of heparan sulfate proteoglycans. Cell. 119:231-44.

Bix, G., and R.V. Iozzo. 2005. Matrix revolutions: "tails" of basement-membrane components with angiostatic functions. Trends Cell Biol.15:52-60.

Bloch, W., E. Forsberg, S. Lentini, C. Brakebusch, K. Martin, H.W. Krell, U.H. Weidle, K. Addicks, and R. Fassler. 1997. Beta 1 integrin is essential for teratoma growth and angiogenesis. J Cell Biol.139:265-78.

Bradley, A., M. Evans, M.H. Kaufman, and E. Robertson. 1984. Formation of germ-line chimaeras from embryo-derived teratocarcinoma cell lines. Nature. 309:255-6.

Burri, P.H., and V. Djonov. 2002. Intussusceptive angiogenesis--the alterna-tive to capillary sprouting. Mol Aspects Med. 23:S1-27.

Byrd, N., S. Becker, P. Maye, R. Narasimhaiah, B. St-Jacques, X. Zhang, J. McMahon, A. McMahon, and L. Grabel. 2002. Hedgehog is re-quired for murine yolk sac angiogenesis. Development. 129:361-72.

48

Carmeliet, P., V. Ferreira, G. Breier, S. Pollefeyt, L. Kieckens, M. Gertsen-stein, M. Fahrig, A. Vandenhoeck, K. Harpal, C. Eberhardt, C. De-clercq, J. Pawling, L. Moons, D. Collen, W. Risau, and A. Nagy. 1996. Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature. 380:435-9.

Carmeliet, P., Y.S. Ng, D. Nuyens, G. Theilmeier, K. Brusselmans, I. Cor-nelissen, E. Ehler, V.V. Kakkar, I. Stalmans, V. Mattot, J.C. Per-riard, M. Dewerchin, W. Flameng, A. Nagy, F. Lupu, L. Moons, D. Collen, P.A. D'Amore, and D.T. Shima. 1999. Impaired myocardial angiogenesis and ischemic cardiomyopathy in mice lacking the vas-cular endothelial growth factor isoforms VEGF164 and VEGF188. Nat. Med. 5:495-502.

Chen, E., S. Hermanson, and S.C. Ekker. 2004. Syndecan-2 is essential for angiogenic sprouting during zebrafish development. Blood.103:1710-9.

Childs, S., J.N. Chen, D.M. Garrity, and M.C. Fishman. 2002. Patterning of angiogenesis in the zebrafish embryo. Development. 129:973-82.

Choi, K., M. Kennedy, A. Kazarov, J.C. Papadimitriou, and G. Keller. 1998. A common precursor for hematopoietic and endothelial cells. Devel-opment. 125:725-32.

Clark, E.A., and J.S. Brugge. 1995. Integrins and signal transduction path-ways: the road taken. Science. 268:233-9.

Colognato, H., and P.D. Yurchenco. 2000. Form and function: the laminin family of heterotrimers. Dev Dyn. 218:213-34.

Costell, M., E. Gustafsson, A. Aszodi, M. Morgelin, W. Bloch, E. Hunziker, K. Addicks, R. Timpl, and R. Fassler. 1999. Perlecan maintains the integrity of cartilage and some basement membranes. J Cell Biol.147:1109-22.

Cross, M.J., J. Dixelius, T. Matsumoto, and L. Claesson-Welsh. 2003. VEGF-receptor signal transduction. Trends Biochem Sci. 28:488-94.

De Arcangelis, A., O. Lefebvre, A. Mechine-Neuville, C. Arnold, A. Klein, L. Remy, M. Kedinger, and P. Simon-Assmann. 2001. Overexpres-sion of laminin alpha1 chain in colonic cancer cells induces an in-crease in tumor growth. Int J Cancer. 94:44-53.

DeHahn, K.C., M. Gonzales, A.M. Gonzalez, S.B. Hopkinson, N.S. Chan-del, J.K. Brunelle, and J.C. Jones. 2004. The alpha4 laminin subunit regulates endothelial cell survival. Exp Cell Res. 294:281-9.

Dejana, E. 2004. Endothelial cell-cell junctions: happy together. Nat Rev Mol Cell Biol. 5:261-70.

Dennissen, M.A., G.J. Jenniskens, M. Pieffers, E.M. Versteeg, M. Petitou, J.H. Veerkamp, and T.H. van Kuppevelt. 2002. Large, tissue-regulated domain diversity of heparan sulfates demonstrated by phage display antibodies. J. Biol. Chem. 277:10982-6.

Doetschman, T.C., H. Eistetter, M. Katz, W. Schmidt, and R. Kemler. 1985. The in vitro development of blastocyst-derived embryonic stem cell lines: formation of visceral yolk sac, blood islands and myocardium. J Embryol Exp Morphol. 87:27-45.

49

Dougher, A.M., H. Wasserstrom, L. Torley, L. Shridaran, P. Westdock, R.E. Hileman, J.R. Fromm, R. Anderberg, S. Lyman, R.J. Linhardt, J. Kaplan, and B.I. Terman. 1997. Identification of a heparin binding peptide on the extracellular domain of the KDR VEGF receptor. Growth Factors. 14:257-68.

Dumont, D.J., L. Jussila, J. Taipale, A. Lymboussaki, T. Mustonen, K. Paju-sola, M. Breitman, and K. Alitalo. 1998. Cardiovascular failure in mouse embryos deficient in VEGF receptor-3. Science. 282:946-9.

Duval, M., S. Bedard-Goulet, C. Delisle, and J.P. Gratton. 2003. Vascular endothelial growth factor-dependent down-regulation of Flk-1/KDR involves Cbl-mediated ubiquitination. Consequences on nitric oxide production from endothelial cells. J Biol Chem. 278:20091-7.

Dvorak, H.F., V.S. Harvey, P. Estrella, L.F. Brown, J. McDonagh, and A.M. Dvorak. 1987. Fibrin containing gels induce angiogenesis. Implica-tions for tumor stroma generation and wound healing. Lab Invest.57:673-86.

Esko, J.D., and U. Lindahl. 2001. Molecular diversity of heparan sulfate. JClin Invest. 108:169-73.

Evans, M.J., and M.H. Kaufman. 1981. Establishment in culture of pluripo-tential cells from mouse embryos. Nature. 292:154-6.

Feraud, O., Y. Cao, and D. Vittet. 2001. Embryonic stem cell-derived em-bryoid bodies development in collagen gels recapitulates sprouting angiogenesis. Lab Invest. 81:1669-81.

Ferrara, N. 1999. Molecular and biological properties of vascular endothelial growth factor. J Mol Med. 77:527-43.

Ferrara, N. 2002. Role of vascular endothelial growth factor in physiologic and pathologic angiogenesis: therapeutic implications. Semin Oncol.29:10-4.

Ferrara, N., K. Carver-Moore, H. Chen, M. Dowd, L. Lu, K.S. O'Shea, L. Powell-Braxton, K.J. Hillan, and M.W. Moore. 1996. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature. 380:439-42.

Feyzi, E., T. Saldeen, E. Larsson, U. Lindahl, and M. Salmivirta. 1998. Age-dependent modulation of heparan sulfate structure and function. J. Biol. Chem. 273:13395-8.

Fitzgerald, M.L., Z. Wang, P.W. Park, G. Murphy, and M. Bernfield. 2000. Shedding of syndecan-1 and -4 ectodomains is regulated by multiple signaling pathways and mediated by a TIMP-3-sensitive metallopro-teinase. J Cell Biol. 148:811-24.

Folkman, J., and Y. Shing. 1992. Angiogenesis. J Biol Chem. 267:10931-4. Fong, G.H., J. Rossant, M. Gertsenstein, and M.L. Breitman. 1995. Role of

the Flt-1 receptor tyrosine kinase in regulating the assembly of vas-cular endothelium. Nature. 376:66-70.

Forsberg, E., G. Pejler, M. Ringvall, C. Lunderius, B. Tomasini-Johansson, M. Kusche-Gullberg, I. Eriksson, J. Ledin, L. Hellman, and L. Kjel-len. 1999. Abnormal mast cells in mice deficient in a heparin-synthesizing enzyme. Nature. 400:773-6.

50

Frieser, M., H. Nockel, F. Pausch, C. Roder, A. Hahn, R. Deutzmann, and L.M. Sorokin. 1997. Cloning of the mouse laminin alpha 4 cDNA. Expression in a subset of endothelium. Eur J Biochem. 246:727-35.

George, E.L., H.S. Baldwin, and R.O. Hynes. 1997. Fibronectins are essen-tial for heart and blood vessel morphogenesis but are dispensable for initial specification of precursor cells. Blood. 90:3073-81.

George, E.L., E.N. Georges-Labouesse, R.S. Patel-King, H. Rayburn, and R.O. Hynes. 1993. Defects in mesoderm, neural tube and vascular development in mouse embryos lacking fibronectin. Development.119:1079-91.

Gerhardt, H., M. Golding, M. Fruttiger, C. Ruhrberg, A. Lundkvist, A. Abramsson, M. Jeltsch, C. Mitchell, K. Alitalo, D. Shima, and C. Betsholtz. 2003. VEGF guides angiogenic sprouting utilizing endo-thelial tip cell filopodia. J Cell Biol. 161:1163-77.

Gluzman-Poltorak, Z., T. Cohen, Y. Herzog, and G. Neufeld. 2000. Neu-ropilin-2 is a receptor for the vascular endothelial growth factor (VEGF) forms VEGF-145 and VEGF-165 [corrected]. J Biol Chem.275:18040-5.

Gragoudas, E.S., A.P. Adamis, E.T. Cunningham, Jr., M. Feinsod, and D.R. Guyer. 2004. Pegaptanib for neovascular age-related macular degen-eration. N Engl J Med. 351:2805-16.

Habeck, H., J. Odenthal, B. Walderich, H. Maischein, and S. Schulte-Merker. 2002. Analysis of a zebrafish VEGF receptor mutant reveals specific disruption of angiogenesis. Curr Biol. 12:1405-12.

Hacker, U., K. Nybakken, and N. Perrimon. 2005. Heparan sulphate pro-teoglycans: the sweet side of development. Nat Rev Mol Cell Biol.6:530-41.

Hallmann, R., N. Horn, M. Selg, O. Wendler, F. Pausch, and L.M. Sorokin. 2005. Expression and function of laminins in the embryonic and ma-ture vasculature. Physiol Rev. 85:979-1000.

Hammes, H.P., J. Lin, O. Renner, M. Shani, A. Lundqvist, C. Betsholtz, M. Brownlee, and U. Deutsch. 2002. Pericytes and the pathogenesis of diabetic retinopathy. Diabetes. 51:3107-12.

Han, C., D. Yan, T.Y. Belenkaya, and X. Lin. 2005. Drosophila glypicans Dally and Dally-like shape the extracellular Wingless morphogen gradient in the wing disc. Development. 132:667-79.

Harvey, W. 1628. Exercitatio Anatomica De Motu Cordis et Sanguinis in Animalibus.

Hellstrom, M., H. Gerhardt, M. Kalen, X. Li, U. Eriksson, H. Wolburg, and C. Betsholtz. 2001. Lack of pericytes leads to endothelial hyperpla-sia and abnormal vascular morphogenesis. J Cell Biol. 153:543-53.

Hellstrom, M., M. Kalen, P. Lindahl, A. Abramsson, and C. Betsholtz. 1999. Role of PDGF-B and PDGFR-beta in recruitment of vascular smooth muscle cells and pericytes during embryonic blood vessel formation in the mouse. Development. 126:3047-55.

Hellstrom, M., L.K. Phng, J.J. Hofmann, E. Wallgard, L. Coultas, P. Lind-blom, J. Alva, A.K. Nilsson, L. Karlsson, N. Gaiano, K. Yoon, J.

51

Rossant, M.L. Iruela-Arispe, M. Kalen, H. Gerhardt, and C. Bet-sholtz. 2007. Dll4 signalling through Notch1 regulates formation of tip cells during angiogenesis. Nature.

Hiratsuka, S., Y. Maru, A. Okada, M. Seiki, T. Noda, and M. Shibuya. 2001. Involvement of Flt-1 tyrosine kinase (vascular endothelial growth factor receptor-1) in pathological angiogenesis. Cancer Res.61:1207-13.

Hiratsuka, S., O. Minowa, J. Kuno, T. Noda, and M. Shibuya. 1998. Flt-1 lacking the tyrosine kinase domain is sufficient for normal develop-ment and angiogenesis in mice. Proc Natl Acad Sci U S A. 95:9349-54.

Holmborn, K., J. Ledin, E. Smeds, I. Eriksson, M. Kusche-Gullberg, and L. Kjellen. 2004. Heparan sulfate synthesized by mouse embryonic stem cells deficient in NDST1 and NDST2 is 6-O-sulfated but con-tains no N-sulfate groups. J. Biol. Chem. 279:42355-8.

Holmqvist, K., M.J. Cross, C. Rolny, R. Hagerkvist, N. Rahimi, T. Matsu-moto, L. Claesson-Welsh, and M. Welsh. 2004. The adaptor protein shb binds to tyrosine 1175 in vascular endothelial growth factor (VEGF) receptor-2 and regulates VEGF-dependent cellular migra-tion. J Biol Chem. 279:22267-75.

Hopf, M., W. Gohring, E. Kohfeldt, Y. Yamada, and R. Timpl. 1999. Re-combinant domain IV of perlecan binds to nidogens, laminin-nidogen complex, fibronectin, fibulin-2 and heparin. Eur J Biochem.259:917-25.

Ibrahimi, O.A., F. Zhang, S.C. Hrstka, M. Mohammadi, and R.J. Linhardt. 2004. Kinetic model for FGF, FGFR, and proteoglycan signal trans-duction complex assembly. Biochemistry. 43:4724-30.

Iso, T., Y. Hamamori, and L. Kedes. 2003. Notch signaling in vascular de-velopment. Arterioscler Thromb Vasc Biol. 23:543-53.

Jain, R.K., D.G. Duda, J.W. Clark, and J.S. Loeffler. 2006. Lessons from phase III clinical trials on anti-VEGF therapy for cancer. Nat Clin Pract Oncol. 3:24-40.

Jakobsson, L., J. Kreuger, K. Holmborn, L. Lundin, I. Eriksson, L. Kjellen, and L. Claesson-Welsh. 2006. Heparan sulfate in trans potentiates VEGFR-mediated angiogenesis. Dev Cell. 10:625-34.

Kamei, M., W.B. Saunders, K.J. Bayless, L. Dye, G.E. Davis, and B.M. Weinstein. 2006. Endothelial tubes assemble from intracellular vacuoles in vivo. Nature. 442:453-6.

Kamimura, K., T. Koyama, H. Habuchi, R. Ueda, M. Masu, K. Kimata, and H. Nakato. 2006. Specific and flexible roles of heparan sulfate modi-fications in Drosophila FGF signaling. J Cell Biol. 174:773-8.

Kappert, K., K.G. Peters, F.D. Bohmer, and A. Ostman. 2005. Tyrosine phosphatases in vessel wall signaling. Cardiovasc Res. 65:587-98.

Kato, M., H. Wang, V. Kainulainen, M.L. Fitzgerald, S. Ledbetter, D.M. Ornitz, and M. Bernfield. 1998. Physiological degradation converts the soluble syndecan-1 ectodomain from an inhibitor to a potent ac-tivator of FGF-2. Nat Med. 4:691-7.

52

Kawasaki, T., T. Kitsukawa, Y. Bekku, Y. Matsuda, M. Sanbo, T. Yagi, and H. Fujisawa. 1999. A requirement for neuropilin-1 in embryonic vessel formation. Development. 126:4895-902.

Keck, P.J., S.D. Hauser, G. Krivi, K. Sanzo, T. Warren, J. Feder, and D.T. Connolly. 1989. Vascular permeability factor, an endothelial cell mi-togen related to PDGF. Science. 246:1309-12.

Kendall, R.L., and K.A. Thomas. 1993. Inhibition of vascular endothelial cell growth factor activity by an endogenously encoded soluble re-ceptor. Proc Natl Acad Sci U S A. 90:10705-9.

Kirkpatrick, C.A., B.D. Dimitroff, J.M. Rawson, and S.B. Selleck. 2004. Spatial regulation of Wingless morphogen distribution and signaling by Dally-like protein. Dev Cell. 7:513-23.

Kitsukawa, T., M. Shimizu, M. Sanbo, T. Hirata, M. Taniguchi, Y. Bekku, T. Yagi, and H. Fujisawa. 1997. Neuropilin-semaphorin III/D-mediated chemorepulsive signals play a crucial role in peripheral nerve projection in mice. Neuron. 19:995-1005.

Kitsukawa, T., A. Shimono, A. Kawakami, H. Kondoh, and H. Fujisawa. 1995. Overexpression of a membrane protein, neuropilin, in chi-meric mice causes anomalies in the cardiovascular system, nervous system and limbs. Development. 121:4309-18.

Kolset, S.O., and J.T. Gallagher. 1990. Proteoglycans in haemopoietic cells. Biochim Biophys Acta. 1032:191-211.

Kreuger, J., P. Jemth, E. Sanders-Lindberg, L. Eliahu, D. Ron, C. Basilico, M. Salmivirta, and U. Lindahl. 2005. Fibroblast growth factors share binding sites in heparan sulfate. Biochem. J.

Kreuger, J., L. Perez, A.J. Giraldez, and S.M. Cohen. 2004. Opposing activi-ties of Dally-like glypican at high and low levels of Wingless morphogen activity. Dev Cell. 7:503-12.

Kreuger, J., D. Spillmann, J.P. Li, and U. Lindahl. 2006. Interactions be-tween heparan sulfate and proteins: the concept of specificity. J Cell Biol. 174:323-7.

Lamoreaux, W.J., M.E. Fitzgerald, A. Reiner, K.A. Hasty, and S.T. Charles. 1998. Vascular endothelial growth factor increases release of gelati-nase A and decreases release of tissue inhibitor of metallopro-teinases by microvascular endothelial cells in vitro. Microvasc Res.55:29-42.

Lampugnani, M.G., F. Orsenigo, M.C. Gagliani, C. Tacchetti, and E. Dejana. 2006. Vascular endothelial cadherin controls VEGFR-2 internaliza-tion and signaling from intracellular compartments. J Cell Biol.174:593-604.

Langford, J.K., M.J. Stanley, D. Cao, and R.D. Sanderson. 1998. Multiple heparan sulfate chains are required for optimal syndecan-1 function. J Biol Chem. 273:29965-71.

Lawson, N.D., and B.M. Weinstein. 2002. In vivo imaging of embryonic vascular development using transgenic zebrafish. Dev Biol. 248:307-18.

53

Leung, D.W., G. Cachianes, W.J. Kuang, D.V. Goeddel, and N. Ferrara. 1989. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science. 246:1306-9.

Leveen, P., M. Pekny, S. Gebre-Medhin, B. Swolin, E. Larsson, and C. Bet-sholtz. 1994. Mice deficient for PDGF B show renal, cardiovascular, and hematological abnormalities. Genes Dev. 8:1875-87.

Li, S., D. Edgar, R. Fassler, W. Wadsworth, and P.D. Yurchenco. 2003. The role of laminin in embryonic cell polarization and tissue organiza-tion. Dev Cell. 4:613-24.

Li, S., D. Harrison, S. Carbonetto, R. Fassler, N. Smyth, D. Edgar, and P.D. Yurchenco. 2002. Matrix assembly, regulation, and survival func-tions of laminin and its receptors in embryonic stem cell differentia-tion. J Cell Biol. 157:1279-90.

Li, X., Y. Chen, S. Scheele, E. Arman, R. Haffner-Krausz, P. Ekblom, and P. Lonai. 2001. Fibroblast growth factor signaling and basement mem-brane assembly are connected during epithelial morphogenesis of the embryoid body. J Cell Biol. 153:811-22.

Lin, X., G. Wei, Z. Shi, L. Dryer, J.D. Esko, D.E. Wells, and M.M. Matzuk. 2000. Disruption of gastrulation and heparan sulfate biosynthesis in EXT1-deficient mice. Dev Biol. 224:299-311.

Lindahl, P., B.R. Johansson, P. Leveen, and C. Betsholtz. 1997. Pericyte loss and microaneurysm formation in PDGF-B-deficient mice. Science.277:242-5.

Lindskog, H., E. Athley, E. Larsson, S. Lundin, M. Hellstrom, and P. Lin-dahl. 2006. New insights to vascular smooth muscle cell and peri-cyte differentiation of mouse embryonic stem cells in vitro. Arterio-scler Thromb Vasc Biol. 26:1457-64.

Lundin, L., L. Ronnstrand, M. Cross, C. Hellberg, U. Lindahl, and L. Claes-son-Welsh. 2003. Differential tyrosine phosphorylation of fibroblast growth factor (FGF) receptor-1 and receptor proximal signal trans-duction in response to FGF-2 and heparin. Exp Cell Res. 287:190-8.

Maes, C., I. Stockmans, K. Moermans, R. Van Looveren, N. Smets, P. Car-meliet, R. Bouillon, and G. Carmeliet. 2004. Soluble VEGF iso-forms are essential for establishing epiphyseal vascularization and regulating chondrocyte development and survival. J Clin Invest.113:188-99.

Magnusson, P., C. Rolny, L. Jakobsson, C. Wikner, Y. Wu, D.J. Hicklin, and L. Claesson-Welsh. 2004. Deregulation of Flk-1/vascular endothelial growth factor receptor-2 in fibroblast growth factor receptor-1-deficient vascular stem cell development. J Cell Sci. 117:1513-23.

Mamluk, R., Z. Gechtman, M.E. Kutcher, N. Gasiunas, J. Gallagher, and M. Klagsbrun. 2002. Neuropilin-1 binds vascular endothelial growth factor 165, placenta growth factor-2, and heparin via its b1b2 do-main. J Biol Chem. 277:24818-25.

Martin, G.R. 1981. Isolation of a pluripotent cell line from early mouse em-bryos cultured in medium conditioned by teratocarcinoma stem cells. Proc Natl Acad Sci U S A. 78:7634-8.

54

Matsumoto, T., S. Bohman, J. Dixelius, T. Berge, A. Dimberg, P. Magnus-son, L. Wang, C. Wikner, J.H. Qi, C. Wernstedt, J. Wu, S. Bruheim, H. Mugishima, D. Mukhopadhyay, A. Spurkland, and L. Claesson-Welsh. 2005. VEGF receptor-2 Y951 signaling and a role for the adapter molecule TSAd in tumor angiogenesis. Embo J. 24:2342-53.

McKeehan, W.L., and M. Kan. 1994. Heparan sulfate fibroblast growth fac-tor receptor complex: structure-function relationships. Mol Reprod Dev. 39:69-81; discusison 81-2.

Miner, J.H., J. Cunningham, and J.R. Sanes. 1998. Roles for laminin in em-bryogenesis: exencephaly, syndactyly, and placentopathy in mice lacking the laminin alpha5 chain. J Cell Biol. 143:1713-23.

Miner, J.H., and C. Li. 2000. Defective glomerulogenesis in the absence of laminin alpha5 demonstrates a developmental role for the kidney glomerular basement membrane. Dev Biol. 217:278-89.

Miner, J.H., C. Li, J.L. Mudd, G. Go, and A.E. Sutherland. 2004. Composi-tional and structural requirements for laminin and basement mem-branes during mouse embryo implantation and gastrulation. Devel-opment. 131:2247-56.

Miner, J.H., and P.D. Yurchenco. 2004. Laminin functions in tissue morphogenesis. Annu Rev Cell Dev Biol. 20:255-84.

Ng, Y.S., R. Rohan, M.E. Sunday, D.E. Demello, and P.A. D'Amore. 2001. Differential expression of VEGF isoforms in mouse during devel-opment and in the adult. Dev Dyn. 220:112-21.

Nyberg, P., L. Xie, and R. Kalluri. 2005. Endogenous inhibitors of angio-genesis. Cancer Res. 65:3967-79.

Olsson, A.K., A. Dimberg, J. Kreuger, and L. Claesson-Welsh. 2006. VEGF receptor signalling - in control of vascular function. Nat Rev Mol Cell Biol. 7:359-71.

Park, J.E., G.A. Keller, and N. Ferrara. 1993. The vascular endothelial growth factor (VEGF) isoforms: differential deposition into the subepithelial extracellular matrix and bioactivity of extracellular ma-trix-bound VEGF. Mol Biol Cell. 4:1317-26.

Parsons, M.J., S.M. Pollard, L. Saude, B. Feldman, P. Coutinho, E.M. Hirst, and D.L. Stemple. 2002. Zebrafish mutants identify an essential role for laminins in notochord formation. Development. 129:3137-46.

Pettersson, A., J.A. Nagy, L.F. Brown, C. Sundberg, E. Morgan, S. Jungles, R. Carter, J.E. Krieger, E.J. Manseau, V.S. Harvey, I.A. Eckelhoe-fer, D. Feng, A.M. Dvorak, R.C. Mulligan, and H.F. Dvorak. 2000. Heterogeneity of the angiogenic response induced in different nor-mal adult tissues by vascular permeability factor/vascular endothe-lial growth factor. Lab Invest. 80:99-115.

Pollard, S.M., M.J. Parsons, M. Kamei, R.N. Kettleborough, K.A. Thomas, V.N. Pham, M.K. Bae, A. Scott, B.M. Weinstein, and D.L. Stemple. 2006. Essential and overlapping roles for laminin alpha chains in no-tochord and blood vessel formation. Dev Biol. 289:64-76.

55

Ponce, M.L., M. Nomizu, and H.K. Kleinman. 2001. An angiogenic laminin site and its antagonist bind through the alpha(v)beta3 and al-pha5beta1 integrins. Faseb J. 15:1389-97.

Poschl, E., U. Schlotzer-Schrehardt, B. Brachvogel, K. Saito, Y. Ninomiya, and U. Mayer. 2004. Collagen IV is essential for basement mem-brane stability but dispensable for initiation of its assembly during early development. Development. 131:1619-28.

Pugh, C.W., and P.J. Ratcliffe. 2003. Regulation of angiogenesis by hy-poxia: role of the HIF system. Nat Med. 9:677-84.

Ringvall, M., J. Ledin, K. Holmborn, T. van Kuppevelt, F. Ellin, I. Eriksson, A.M. Olofsson, L. Kjellen, and E. Forsberg. 2000. Defective heparan sulfate biosynthesis and neonatal lethality in mice lacking N-deacetylase/N-sulfotransferase-1. J Biol Chem. 275:25926-30.

Risau, W., and I. Flamme. 1995. Vasculogenesis. Annu Rev Cell Dev Biol.11:73-91.

Robinson, C.J., and S.E. Stringer. 2001. The splice variants of vascular en-dothelial growth factor (VEGF) and their receptors. J Cell Sci.114:853-65.

Rosengart, T.K., W.V. Johnson, R. Friesel, R. Clark, and T. Maciag. 1988. Heparin protects heparin-binding growth factor-I from proteolytic inactivation in vitro. Biochem Biophys Res Commun. 152:432-40.

Ruegg, C., and A. Mariotti. 2003. Vascular integrins: pleiotropic adhesion and signaling molecules in vascular homeostasis and angiogenesis. Cell Mol Life Sci. 60:1135-57.

Ruhrberg, C., H. Gerhardt, M. Golding, R. Watson, S. Ioannidou, H. Fuji-sawa, C. Betsholtz, and D.T. Shima. 2002. Spatially restricted pat-terning cues provided by heparin-binding VEGF-A control blood vessel branching morphogenesis. Genes Dev. 16:2684-98.

Sakurai, Y., K. Ohgimoto, Y. Kataoka, N. Yoshida, and M. Shibuya. 2005. Essential role of Flk-1 (VEGF receptor 2) tyrosine residue 1173 in vasculogenesis in mice. Proc Natl Acad Sci U S A. 102:1076-81.

Sasaki, T., R. Giltay, U. Talts, R. Timpl, and J.F. Talts. 2002. Expression and distribution of laminin alpha1 and alpha2 chains in embryonic and adult mouse tissues: an immunochemical approach. Exp Cell Res. 275:185-99.

Sasaki, T., and R. Timpl. 2001. Domain IVa of laminin alpha5 chain is cell-adhesive and binds beta1 and alphaVbeta3 integrins through Arg-Gly-Asp. FEBS Lett. 509:181-5.

Schuh, A.C., P. Faloon, Q.L. Hu, M. Bhimani, and K. Choi. 1999. In vitro hematopoietic and endothelial potential of flk-1(-/-) embryonic stem cells and embryos. Proc Natl Acad Sci U S A. 96:2159-64.

Senger, D.R., S.J. Galli, A.M. Dvorak, C.A. Perruzzi, V.S. Harvey, and H.F. Dvorak. 1983. Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites fluid. Science. 219:983-5.

Shalaby, F., J. Rossant, T.P. Yamaguchi, M. Gertsenstein, X.F. Wu, M.L. Breitman, and A.C. Schuh. 1995. Failure of blood-island formation and vasculogenesis in Flk-1-deficient mice. Nature. 376:62-6.

56

Shibuya, M. 2002. Vascular endothelial growth factor receptor family genes: when did the three genes phylogenetically segregate? Biol Chem.383:1573-9.

Shing, Y., J. Folkman, R. Sullivan, C. Butterfield, J. Murray, and M. Klags-brun. 1984. Heparin affinity: purification of a tumor-derived capil-lary endothelial cell growth factor. Science. 223:1296-9.

Siekmann, A.F., and N.D. Lawson. 2007. Notch signalling limits angiogenic cell behaviour in developing zebrafish arteries. Nature.

Sixt, M., B. Engelhardt, F. Pausch, R. Hallmann, O. Wendler, and L.M. So-rokin. 2001. Endothelial cell laminin isoforms, laminins 8 and 10, play decisive roles in T cell recruitment across the blood-brain bar-rier in experimental autoimmune encephalomyelitis. J Cell Biol.153:933-46.

Smyth, N., H.S. Vatansever, M. Meyer, C. Frie, M. Paulsson, and D. Edgar. 1998. The targeted deletion of the LAMC1 gene. Ann N Y Acad Sci.857:283-6.

Smyth, N., H.S. Vatansever, P. Murray, M. Meyer, C. Frie, M. Paulsson, and D. Edgar. 1999. Absence of basement membranes after targeting the LAMC1 gene results in embryonic lethality due to failure of endo-derm differentiation. J Cell Biol. 144:151-60.

Soker, S., H.Q. Miao, M. Nomi, S. Takashima, and M. Klagsbrun. 2002. VEGF165 mediates formation of complexes containing VEGFR-2 and neuropilin-1 that enhance VEGF165-receptor binding. J Cell Biochem. 85:357-68.

Soker, S., S. Takashima, H.Q. Miao, G. Neufeld, and M. Klagsbrun. 1998. Neuropilin-1 is expressed by endothelial and tumor cells as an iso-form-specific receptor for vascular endothelial growth factor. Cell.92:735-45.

Soriano, P. 1994. Abnormal kidney development and hematological disor-ders in PDGF beta-receptor mutant mice. Genes Dev. 8:1888-96.

Sorokin, L.M., F. Pausch, M. Frieser, S. Kroger, E. Ohage, and R. Deutz-mann. 1997. Developmental regulation of the laminin alpha5 chain suggests a role in epithelial and endothelial cell maturation. DevBiol. 189:285-300.

Stalmans, I., Y.S. Ng, R. Rohan, M. Fruttiger, A. Bouche, A. Yuce, H. Fuji-sawa, B. Hermans, M. Shani, S. Jansen, D. Hicklin, D.J. Anderson, T. Gardiner, H.P. Hammes, L. Moons, M. Dewerchin, D. Collen, P. Carmeliet, and P.A. D'Amore. 2002. Arteriolar and venular pattern-ing in retinas of mice selectively expressing VEGF isoforms. J Clin Invest. 109:327-36.

Suyama, K., I. Shapiro, M. Guttman, and R.B. Hazan. 2002. A signaling pathway leading to metastasis is controlled by N-cadherin and the FGF receptor. Cancer Cell. 2:301-14.

Takahashi, T., S. Yamaguchi, K. Chida, and M. Shibuya. 2001. A single autophosphorylation site on KDR/Flk-1 is essential for VEGF-A-dependent activation of PLC-gamma and DNA synthesis in vascular endothelial cells. Embo J. 20:2768-78.

57

Talts, J.F., T. Sasaki, N. Miosge, W. Gohring, K. Mann, R. Mayne, and R. Timpl. 2000. Structural and functional analysis of the recombinant G domain of the laminin alpha4 chain and its proteolytic processing in tissues. J Biol Chem. 275:35192-9.

The, I., Y. Bellaiche, and N. Perrimon. 1999. Hedgehog movement is regu-lated through tout velu-dependent synthesis of a heparan sulfate pro-teoglycan. Mol Cell. 4:633-9.

Thomas, K.R., and M.R. Capecchi. 1987. Site-directed mutagenesis by gene targeting in mouse embryo-derived stem cells. Cell. 51:503-12.

Thompson, S., A.R. Clarke, A.M. Pow, M.L. Hooper, and D.W. Melton. 1989. Germ line transmission and expression of a corrected HPRT gene produced by gene targeting in embryonic stem cells. Cell.56:313-21.

Thyboll, J., J. Kortesmaa, R. Cao, R. Soininen, L. Wang, A. Iivanainen, L. Sorokin, M. Risling, Y. Cao, and K. Tryggvason. 2002. Deletion of the laminin alpha4 chain leads to impaired microvessel maturation. Mol Cell Biol. 22:1194-202.

Timpl, R., and J.C. Brown. 1996. Supramolecular assembly of basement membranes. Bioessays. 18:123-32.

Tischer, E., D. Gospodarowicz, R. Mitchell, M. Silva, J. Schilling, K. Lau, T. Crisp, J.C. Fiddes, and J.A. Abraham. 1989. Vascular endothelial growth factor: a new member of the platelet-derived growth factor gene family. Biochem Biophys Res Commun. 165:1198-206.

Tonini, T., F. Rossi, and P.P. Claudio. 2003. Molecular basis of angiogenesis and cancer. Oncogene. 22:6549-56.

Uemura, A., S. Kusuhara, H. Katsuta, and S. Nishikawa. 2006. Angiogenesis in the mouse retina: a model system for experimental manipulation. Exp Cell Res. 312:676-83.

Warner, A.J., J. Lopez-Dee, E.L. Knight, J.R. Feramisco, and S.A. Prigent. 2000. The Shc-related adaptor protein, Sck, forms a complex with the vascular-endothelial-growth-factor receptor KDR in transfected cells. Biochem J. 347:501-9.

Wiley, H.S., and P.M. Burke. 2001. Regulation of receptor tyrosine kinase signaling by endocytic trafficking. Traffic. 2:12-8.

Willem, M., N. Miosge, W. Halfter, N. Smyth, I. Jannetti, E. Burghart, R. Timpl, and U. Mayer. 2002. Specific ablation of the nidogen-binding site in the laminin gamma1 chain interferes with kidney and lung development. Development. 129:2711-22.

Wilting, J., B. Christ, L. Yuan, and A. Eichmann. 2003. Cellular and mo-lecular mechanisms of embryonic haemangiogenesis and lymphan-giogenesis. Naturwissenschaften. 90:433-48.

Virtanen, I., D. Gullberg, J. Rissanen, E. Kivilaakso, T. Kiviluoto, L.A. La-itinen, V.P. Lehto, and P. Ekblom. 2000. Laminin alpha1-chain shows a restricted distribution in epithelial basement membranes of fetal and adult human tissues. Exp Cell Res. 257:298-309.

Vittet, D., M.H. Prandini, R. Berthier, A. Schweitzer, H. Martin-Sisteron, G. Uzan, and E. Dejana. 1996. Embryonic stem cells differentiate in vi-

58

tro to endothelial cells through successive maturation steps. Blood.88:3424-31.

Wondimu, Z., T. Geberhiwot, S. Ingerpuu, E. Juronen, X. Xie, L. Lindbom, M. Doi, J. Kortesmaa, J. Thyboll, K. Tryggvason, B. Fadeel, and M. Patarroyo. 2004. An endothelial laminin isoform, laminin 8 (al-pha4beta1gamma1), is secreted by blood neutrophils, promotes neu-trophil migration and extravasation, and protects neutrophils from apoptosis. Blood. 104:1859-66.

Wondimu, Z., G. Gorfu, T. Kawataki, S. Smirnov, P. Yurchenco, K. Tryggvason, and M. Patarroyo. 2006. Characterization of commer-cial laminin preparations from human placenta in comparison to re-combinant laminins 2 (alpha2beta1gamma1), 8 (al-pha4beta1gamma1), 10 (alpha5beta1gamma1). Matrix Biol. 25:89-93.

Wu, L.W., L.D. Mayo, J.D. Dunbar, K.M. Kessler, O.N. Ozes, R.S. Warren, and D.B. Donner. 2000. VRAP is an adaptor protein that binds KDR, a receptor for vascular endothelial cell growth factor. J Biol Chem. 275:6059-62.

Xue, Y., X. Gao, C.E. Lindsell, C.R. Norton, B. Chang, C. Hicks, M. Gen-dron-Maguire, E.B. Rand, G. Weinmaster, and T. Gridley. 1999. Embryonic lethality and vascular defects in mice lacking the Notch ligand Jagged1. Hum Mol Genet. 8:723-30.

Yamaguchi, T.P., K. Harpal, M. Henkemeyer, and J. Rossant. 1994. fgfr-1 is required for embryonic growth and mesodermal patterning during mouse gastrulation. Genes Dev. 8:3032-44.

Yamashita, J., H. Itoh, M. Hirashima, M. Ogawa, S. Nishikawa, T. Yurugi, M. Naito, and K. Nakao. 2000. Flk1-positive cells derived from em-bryonic stem cells serve as vascular progenitors. Nature. 408:92-6.

Yayon, A., M. Klagsbrun, J.D. Esko, P. Leder, and D.M. Ornitz. 1991. Cell surface, heparin-like molecules are required for binding of basic fi-broblast growth factor to its high affinity receptor. Cell. 64:841-8.

Zachary, I. 2005. Neuroprotective role of vascular endothelial growth factor: signalling mechanisms, biological function, and therapeutic poten-tial. Neurosignals. 14:207-21.

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 235

Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofMedicine. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Medicine”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-7666

ACTAUNIVERSITATISUPSALIENSISUPPSALA2007