Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1347

Functions of Heparan Sulfateduring Mouse Development

Studies of Mice with Genetically AlteredHeparan Sulfate Biosynthesis

BY

MARIA RINGVALL

ACTA UNIVERSITATIS UPSALIENSISUPPSALA 2004

Front cover picture: Blastocyst injection for generation of a genetically modified mouse strain.A normal (wild-type) blastocyst (3.5 days old mouse embryo) is held with a holder-pipette (to the left) and is injected with about 15 embryonic stem (ES) cells with an injection-pipette (to the right). The ES cells have been genetically modified in vitrobefore injection. The diameter of a blastocyst and an ES cell is about 100 m and 15 m respectively. Several injected blastocysts will be reimplanted into the uterus of a surrogate mother where they will develop to chimeric mice with both normal cells and genetically modified cells. If the experiment has been successful, the chimeric mice will produce genetically modified germ cells and the genetic modification will be inherited by the offspring.

List of Papers

This thesis is based on the following papers:

I Forsberg, E., Pejler, G., Ringvall, M., Lunderius, C., Tomasini-

Johansson, B., Kusche-Gullberg, M., Eriksson, I., Ledin, J.,

Hellman, L., and Kjellén, L. (1999). Abnormal mast cells in mice

deficient in a heparin-synthesizing enzyme. Nature 400, 773-776.

II Ringvall, M., Ledin J., Holmborn K., van Kuppevelt T., Ellin F.,

Eriksson I., Olofsson A.M., Kjellén L., and Forsberg E. (2000)

Defective heparan sulfate biosynthesis and neonatal lethality in

mice lacking N-deacetylase/N-sulfotransferase-1. J. Biol. Chem.

275, 25926-30.

III Ledin, J., Ringvall, M., Eriksson, I., Wilén, M., Thuveson, M.,

Kusche-Gullberg, M., Forsberg, E., and Kjellén, L. In vivo

regulation of liver heparan sulfate biosynthesis in NDST1 and

NDST2 deficient mice. Submitted.

IV Ringvall, M., Forsberg, E., and Kjellén, L. Several processes in

skeletal development are affected by altered heparan sulfate

structure. Manuscript.

Reprints were made by permission of the publishers.

Contents

Introduction.....................................................................................................1

Background.....................................................................................................2

Model organisms for in vivo studies...........................................................2

Establishment of genetically altered mouse strains ...............................2

Mouse development...............................................................................3

Lung development and maturation....................................................4

Liver development ............................................................................4

Skeletal development ........................................................................5

Mast cells ..........................................................................................5

Glycobiology..............................................................................................7

Heparan sulfate, heparin and their core proteins ........................................7

Glycosaminoglycans..............................................................................7

Heparan sulfate and heparin ..................................................................7

Heparan sulfate proteoglycans...............................................................8

Cell surface heparan sulfate proteoglycans.......................................8

Extracellular matrix proteoglycans .................................................10

Serglycin — the only characterized heparin-proteoglycan .............11

HS biosynthesis — a multienzyme machinery.........................................11

Four N-deacetylase/N-sulfotransferase isoforms.................................12

What regulates HS biosynthesis?.........................................................13

Protein interactions...................................................................................15

Heparan sulfate in physiological processes..............................................17

Drosophila mutants .............................................................................17

HS in mammalian systems...................................................................19

Mouse models with defective HS biosynthesis...............................20

Defective GAG biosynthesis — a cause of human disorders .........21

A mouse model with increased HS degradation .............................22

Defects in the extracellular matrix ..................................................22

Lack of glypican-3 causes multiple defects ....................................23

Altered syndecan expression gives subtle phenotypes in mice.......23

Core protein versus side-chains ......................................................24

HS during development............................................................................25

HS in signaling and morphogen gradients...........................................25

Dynamic alterations in HS structure during development ..............27

A new mechanism for HS structure modification...........................28

HS degradation and pathology ........................................................28

The impact of HS on fundamental developmental processes ..............29

No HS — no gastrulation................................................................29

Development without 2-O-sulfate groups.......................................29

HS involvement in cell migration and branching morphogenesis ..30

Present investigation .....................................................................................32

Aim of study ........................................................................................32

Results...........................................................................................................33

Paper I — NDST2 deficient mice lack heparin and have a severe mast

cell defect.............................................................................................33

Paper II — NDST1 deficient mice die neonatally and produce

undersulfated HS .................................................................................34

Paper III — the liver develops normally despite low HS-sulfation

levels....................................................................................................35

Paper IV — skeletal development is dependent on HS structure ........36

Discussion.....................................................................................................38

NDST2 — crucial in mast cells but redundant in other cells?.............38

NDST1 and embryonic development ..................................................39

Future perspectives .......................................................................................42

Concluding remarks ......................................................................................44

In vivo veritas ......................................................................................44

Populärvetenskaplig sammanfattning ...........................................................45

Proteiner — genprodukter med många funktioner ..............................45

Heparansulfat-proteoglykaner — proteiner med sulfaterade

sockerkedjor ....................................................................................45

Lärdom från genetiskt modifierade möss ............................................46

Heparin—ett måste för mastceller ..................................................46

HS — viktigt för normal embryoutveckling ...................................47

Acknowledgement ........................................................................................49

References.....................................................................................................51

Abbreviations

BM Basement membrane

BMP Bone morphogenetic protein

C5epi C5-epimerase

CNS Central nervous system

CS Chondroitin sulfate

DS Dermatan sulfate

E Embryonic day

ECM Extracellular matrix

ES Embryonic stem

EXT Exostosin

EXTL Exostosin like

FGF Fibroblast growth factor

FGFR Fibroblast growth factor receptor

GAG Glycosaminoglycan

GlcA Glucuronic acid

GlcN Glucosamine

GlcNAc N-acetylglucosamine

GlcNSO3 N-sulfateglucosamine

Hh Hedgehog

HME Hereditary multiple exostoses

HS Heparan sulfate

HSPG Heparan sulfate proteoglycan

ICM Inner cellmass

IdoA Iduronic acid

Ihh Indian hedgehog

NDST N-deacetylase/N-sulfotransferase

OST O-sulfotransferase

PAPS 3’-phosphoadenosine-5’-phosphosulfate

PG Proteoglycan

SGBS Simpson-Golabi-Behmel syndrome

TE Trophectoderm

UTR Untranslated region

VEGF Vascular endothelial growth factor

Wg Wingless

1

Introduction

In a very early embryo all cells are equal and have the potential to

differentiate into any kind of cell. During development of the embryo, cells

will become more specialized and contribute to the different tissues and

organs of the body. To gain specific qualities, each cell has to receive

specific signals from the surrounding environment. Signaling takes place

between cells and between cells and the extracellular matrix (ECM) either by

direct contact or via soluble molecules.

Located at the surface of most cell types and in the ECM, heparan sulfate-

proteoglycans (HSPGs) are important in many signaling events: as co-

receptors and to facilitate the establishment and maintenance of morphogen

gradients. HSPGs are also important structural elements in the ECM and

their HS chains can also store and protect growth factors and other signaling

molecules from proteolytic degradation.

The sulfation degree and pattern of the HS chains are of importance for

the capacity to bind proteins in a specific way. The sulfate groups also make

HS chains negatively charged, thereby giving them the possibility to interact

with positively charged molecules in a less specific way. The most highly

sulfated form of HS is called heparin, and can only be found inside

connective tissue type mast cells.

Much of the knowledge about the functions of HS and HSPGs is derived

from biochemical and cell culture experiments. Studies of Drosophila

mutants with defective HS-biosynthesis or lack of HSPGs have in addition

given important in vivo knowledge.

The studies presented in this thesis were performed to further clarify the

in vivo functions of HS and heparin in mammals, and how different

HS/heparin-biosynthetic enzymes contribute to the structure of these

polysaccharides. This was performed by generation and analyses of

genetically modified mice. By gene targeting in embryonic stem (ES) cells,

two HS/heparin biosynthetic enzymes, N-deacetylase/N-sulfotransferase-1

and -2, were specifically inactivated in two mouse models, and the

phenotypes were analyzed.

2

Background

Model organisms for in vivo studies Rodents are often used as model organisms for studies of mammalian

development, pathology and other physiological events. Laboratory mouse

strains are well suited for these purposes and specific, inheritable genetic

modifications are fairly easy to perform in these animals.

Establishment of genetically altered mouse strains

Gene targeting, defined as the introduction of site-specific modifications into

the mouse genome by homologous recombination, is a valuable tool to study

gene function in vivo. The discovery that homologous recombination

between exogenous DNA and chromosomal DNA can take place in

mammalian cells (Smithies et al., 1985), although at a low frequency,

opened new possibilities to specifically alter the mouse genome. By this

mechanism, in vitro genetically altered material can be stably integrated into

the genome (Torres and Kühn, 1997). This kind of stably genetically altered

mouse strains are often referred to as “knockout” and “knockin” mice.

ES cells are pluripotent cells derived form the inner cell mass of early

embryos (Evans and Kaufman, 1981; Martin, 1981). In vitro genetically

modified ES cells are conjugated with a normal wild-type early embryo and

reimplanted in the uterus of a surrogate mother. The modified ES cells will

integrate with the wild-type cells of the normal embryo and give rise to a

chimeric embryo containing both wild-type and modified cells. Due to their

pluripotency, the ES cells are capable of forming different cell types,

including the germ line. Hence, the genetic alteration will be inherited by

subsequent generations (Hogan et al., 1994; Torres and Kühn, 1997).

This technique can be used to completely turn off a gene, alter expression

pattern and level, and change the functional properties of a gene product.

In the studies presented in this thesis (paper I-IV), so called classical

knockout mouse strains have been used where the specific gene of interest

has been turned off in all cells in the whole organism.

3

Mouse development

The gestational period for a mouse is approximately 19.5 days but might be

slightly longer or shorter depending on the strain. Cleavage and blastulation

of the fertilized egg occur between embryonic day 0 (E0) and E5, with the

time of fertilization designated as E0. At E3, the morula is transformed into a

blastocyst, two distinct cell lineages are formed during this transformation:

the trophectoderm (TE) and the inner cell mass (ICM). The TE will only

form extraembryonic tissues while the cells in the ICM are still pluripotent

and will contribute to all cell-types in the embryo. After implantation at

E4.5, gastrulation starts and the three primary germ layers are formed and

the basic bodyplan is established. The early organogenesis is started with the

establishment of organ primordia after which the organogenesis period takes

over at E10. This period lasts until E14 and fetal growth and further

development continue until birth (Fig. 1) (Hogan et al., 1994).

Figure 1 Time course of mouse development. (Adapted from Hogan et al., 1994) The fertilized egg (E0) will start to divide and forms after four to five cell-divisions an early blastocyst (E3.5). The blastocyst expands further, and implantation into the uterine wall takes place at E4.5 followed by gastrulation (E6.5) and initiation of the early organogenesis phase. When the primordial organs are established, organo-genesis starts at about E10. After E14 the fetus mainly grows in size and the organs develop further and become more mature until birth at E19.5.

The first differentiation step is when the morula divides its cells into TE and

ICM to form the blastocyst. The cells will further differentiate into ectoderm,

mesoderm and endoderm. From the cells of these three primary germ layers,

all organs and cell lineages of the body will form (Hogan et al., 1994).

Mesenchymal-epithelial interactions are known to be critical during

organogenesis (Saxen and Thesleff, 1992); epithelium can develop from any

germ layer whereas mesenchyme mainly is derived from the mesoderm

(Gilbert, 1997). At different stages of development, cells at various locations

in the embryo will express alternative sets of genes. The specific gene

4

expression is necessary for cells to differentiate into specific cell types and

to form ordered spatial arrangements, a process often referred to as

patterning (Gilbert, 1997).

An important factor in development is the ECM (Zagris, 2001), a

macromolecular meshwork rich in proteoglycans that can regulate

differentiation directly by interactions with cell surface receptors and more

indirectly by facilitating distribution of signaling molecules (Bernfield et al.,

1999; Esko and Selleck, 2002; Schonherr and Hausser, 2000). Basement

membranes (BMs) are thin sheets of highly specialized ECM that surround

muscle, fat and peripheral nerve cells, underlie endothelial cells and are

present at the epithelial-mesenchymal interface of most tissues (Erickson and

Couchman, 2000).

Lung development and maturation

At E10 (Fig. 1) the lung arises from the ventral foregut as two primary lung

buds. The inner epithelium is dictated by the surrounding mesenchyme to

undergo a typical branching morphogenesis. The primitive lungs are

vascularized and the epithelium branches further to establish the terminal

respiratory tracts. The epithelial cells lining the prospective terminal sacks

flatten out and close contacts are formed between blood vessels and the

developing respiratory units to mediate oxygen exchange. The capillary

endothelial cells and the pulmonary epithelium are separated by a continuous

BM. Before birth, the so-called type II pneumocytes produce and secrete

surfactant. Pulmonary surfactant is a mixture of lipids and proteins with the

main function to reduce surface tension at the air/liquid interface in the

postnatal lung (Bloom and Fawcett, 1975; Kaufman and Bard, 1999; Weaver

et al., 2003).

Many signaling molecules are involved in the growth and maturation of

the lung. Different fibroblast growth factors (FGFs) are, as an example,

important in branching morphogenesis while maturation is mainly

hormonally regulated (Bloom and Fawcett, 1975; Warburton et al., 2000).

Liver development

The early stages of liver differentiation are recognizable at E9-E9.5 (Fig. 1).

Large blood vessels are acquired and the liver will start to form a more right

sided structure. At about the same time, an enormous increase in size starts.

This enlargement is mainly due to increased hematopoietic activity in the

liver, which is the main embryonic source of red blood cells. Smaller blood

vessels are formed and the liver mass is full of sinusoids and is invaded by

mesenchyme. Many different blood cells appear intermingled with the liver

cells and in the hepatic blood vessels (Kaufman and Bard, 1999; Rugh,

1991).

Hepatic sinusoids are thin-walled vessels, larger and more irregularly

shaped than capillaries. The lining endothelial cells in the sinusoids are

5

separated from the underlying tissue by an incomplete BM. These

endothelial cells will by time become fenestrated and discontinuous, which

make them extremely permeable to many substances, which can then reach

the parenchymal cells of the liver, the hepatocytes (Rugh, 1991).

As in the case of lung development many different growth factors and

morphogens, such as FGFs, bone morphogenetic proteins (BMPs) and

hepatocyte growth factor, are important in liver development (Duncan,

2003).

Skeletal development

Skeletogenesis starts by the formation of mesenchymal condensations at

about E10.5 (Fig. 1). Condensations are either osteogenic or chondrogenic

depending on the skeletal element they are initiating. Most of the skeletal

parts arise from chondrogenic condensations. The cells differentiate to

chondrocytes, which produce a specific cartilage ECM that will serve as a

template for future bones. At the same time, mesenchymal cells surrounding

the cartilage differentiate into osteoblasts. The cartilage is invaded by blood

vessels along with osteoblasts, osteoclasts and hematopoietic cells.

Osteoblasts deposit bone matrix, thereby forming mineralized ossification

centers. The ossification centers successively grow and many can be clearly

visualized at E14.5 (Fig. 1). This kind of ossification process is called

endochondral ossification. Bones grow by proliferation of chondrocytes

located close to the border of ossification centers. Cranial bones arise from

osteogenic condensations that directly give rise to bone tissue, without any

cartilage template. The condensed mesenchymal cells differentiate directly

into osteoblasts that will deposit bone matrix to form ossification centers.

This kind of ossification is called intramembranous ossification (Horton,

2003; Kaufman and Bard, 1999; Rugh, 1991).

Most of the skeleton derives from the mesoderm while some craniofacial

bones arise from neural crest cells from the ectoderm. FGFs, BMPs and

hedgehog (Hh) proteins are among the factors important in skeletal

development (Hoffmann and Gross, 2001; Iwamoto et al., 1999; Marie,

2003).

Mast cells

Mast cells are important in the defense of parasite and bacterial infections.

They originate from hematopoietic stem cells and enter the blood circulation

to reach different tissues where they mature. The mast cells harbor numerous

cytoplasmic granules containing mediators such as histamine, cytokines and

proteases co-stored with proteoglycans (PGs). Activation of mast cells is IgE

mediated by crosslinking of IgE receptors on the cell surface. Upon

activation the mast cells degranulate and thereby release the stored content

into the extracellular environment. Mast cells are thought to be important in

the initiation and intensification of the inflammatory process. This cell type

6

is also the effector-cells in allergic reactions (Beil et al., 2000; Metcalfe et

al., 1997; Yong, 1997).

Rodent mast cells can be divided into two main groups depending on the

glycosaminoglycan (GAG) and mediator composition in their storage-

granules. In connective tissue type mast cells the mediators are co-stored

with heparin, while the mucosa type mast cells instead contain highly

sulfated chondroitin sulfate (CS). Connective tissue type mast cells are

mainly found in the skin and in the peritoneal cavity and mucosa type mast

cells are found beneath the mucosal surface of the gastrointestinal and

respiratory tracts. It is believed that the microenvironment in the tissue

decides whether a mast cell will mature into a connective tissue type or a

mucosa type mast cell (Beil et al., 2000).

7

GlycobiologyGlycobiology can be described as the study of the structure, biosynthesis and

biology of all the saccharides occurring in nature. Proteins often carry a

carbohydrate moiety; depending on what saccharide units included and by

what amino acid in the protein the carbohydrate is attached to, they can be

subdivided into different groups. Glycoproteins are proteins that carry one or

more oligosaccharide chains covalently attached to an aspargine, serine or

threonine residue. Proteoglycans are instead having one or more

glycosaminoglycan (GAG) chains attached to a serine or, in rare cases, a

threonine residue. The oligosaccharide on a glycoprotein is composed of

monosaccharide units and can be branched while a GAG is linear and

composed of repeated disaccharide units (Varki et al., 1999)

Heparan sulfate, heparin and their core proteins

Glycosaminoglycans

Each disaccharide unit of the long unbranched GAG chains consists of one

amino sugar and one hexuronic acid. Based on sugar composition, the GAGs

are classified into four groups: HS/heparin, chondroitin sulfate/dermatan

sulfate (CS/DS), keratan sulfate (contains galactose instead of hexuronic

acid) and hyaluronan. All GAGs except hyaluronan are synthesized

covalently linked to a serine residue of a core protein in the Golgi apparatus

and are commonly sulfated. Hyaluronan is unsulfated and synthesized at the

cell membrane, not bound to any core protein but in a free form.

Heparan sulfate and heparin

HS belongs to the GAG family of polysaccharides and is the most widely

spread polysaccharide among cell-types in the mammalian body. The

repeated disaccharide unit in HS is composed of N-acetylglucosamine

(GlcNAc) and glucuronic acid (GlcA). HS is an important component of the

cell surface and in the extracellular environment. The biosynthetic

machinery that produces HS is complex and during its way through this

machinery, the polysaccharide chain is modified by e.g. sulfation at different

positions. The end product can vary significantly in its structure depending

on by what cell type and at what time point the HS is produced (Lindahl and

Kjellén, 1991; Lindahl et al., 1998; Salmivirta et al., 1996).

8

HS in the ECM is believed to function as a scaffold for growth factors

and protect them from proteolytic degradation. Furthermore, HS binds to

different ECM proteins. At the cell surface, HS is a part of some cell

signaling systems and might work both as a co-receptor and as a tool to

catch ligands from the surrounding environment. HS is also considered

important for shaping and maintaining concentration gradients of

morphogens during embryonic patterning and development (Bernfield et al.,

1999; Esko and Selleck, 2002; Nybakken and Perrimon, 2002)

Heparin can be regarded as a fully sulfated form of HS and is thereby the

most negatively charged compound found in a mammalian cell. Heparin is

well known for its interactions with antithrombin and has been used as an

anticoagulant drug for a long time (Petitou et al., 2003). Heparin is also

known to interact with an array of other proteins (table 1), how many of

these interactions that have any physiologic relevance is not known (Esko

and Selleck, 2002). In vivo, heparin can only be found in connective tissue

type mast cells (Kolset et al., 2004).

Heparan sulfate proteoglycans

PGs are most frequently substituted with HS and CS/DS polysaccharide

chains and depending on what cell type they are expressed by, the HS/CS

ratio can vary. The combination of different core proteins with different

GAGs, which can vary in length, number, and structure (e.g. sulfation

pattern), results in a large structural diversity among PGs. This in turn

contributes to a large variety of biological functions.

HSPGs are ubiquitously found on the surfaces of most cell types and in

the ECM. HSPGs have been suggested to be involved in many cellular

processes and in cell-cell and cell-matrix contact and signaling. As an

example HSPGs in the ECM can work as structural elements and the highly

negatively charged HS chains may act as a filtering device in the kidney.

Cell surface bound HSPGs are of importance in growth factor signaling and

can also work as a docking device for several pathogens (Bernfield et al.,

1992; Spillmann, 2001). Shedding of cell surface HSPGs might be an

important regulatory mechanism in a variety of physiological reactions

(Bernfield et al., 1992; Park et al., 2000).

The structure of the HS chains appears not to depend on what core protein

it is attached to; rather, the structure is correlated to the cell type responsible

for its production (Kato et al., 1994; Tumova et al., 2000; Zako et al., 2003).

Cell surface heparan sulfate proteoglycans

There are two major families of membrane bound HSPGs: the syndecans and

the glypicans (Fig. 2). All four syndecans that have been identified in

mammals (syndecan-1 to -4), have a short cytoplasmic domain and an

extended extracellular domain with GAG attachment sites that position the

9

HS chains distant from the plasma membrane. Syndecan-1 and -4 can in

addition to HS carry CS chains (Carey, 1997). Syndecans are involved in

cell-cell and cell-matrix adhesion and signaling, thereby affecting cell

morphology and migration (Rapraeger, 2000; Yoneda and Couchman, 2003).

Clustering of syndecans at the cell surface might be required to trigger

signaling. Syndecans also facilitate growth factor signaling, primarily by

exposure of HS chains that bind growth factors and their receptors,

regulating their assembly into signaling complexes. The cytoplasmic region

of syndecans can interact with cytoskeletal and intracellular signaling

molecules either directly or via other molecules. The expression of

syndecans is regulated in a cell and tissue specific manner and the most

dramatic changes in expression occur during development, associated with

cell differentiation and changes in tissue organization (Salmivirta and

Jalkanen, 1995).

Figure 2 HS/heparin-proteoglycans. HS-proteoglycans are found on the surface of virtually all cells (syndecan, glypican, betaglycan and CD44) and in the ECM (perlecan, agrin and collagen XVIII). Heparin is only present in the storage granules of connective tissue type mast cells, in the form of the heparin-PG serglycin.

10

The glypicans, with six known members in mammalia (glypican-1 to -6)

are covalently linked to the cell surface via a glycosylphosphatidylinositol-

anchor (De Cat and David, 2001). In contrast to the syndecans, the glypicans

have their GAG attachment sites located in close proximity to the cell

surface. Glypicans are thought to always be substituted with HS side-chains

in vivo. It has been suggested that glypicans are targeted to special

membrane domains referred to as lipid rafts (Brown and Rose, 1992; Simons

and Ikonen, 1997). These rafts are rich in sphingolipids, cholesterol, Src

family kinases, G-proteins and molecules involved in Ca2+ influx. The

localization of glypicans to such domains may facilitate interactions with

specific intracellular signaling molecules despite the absence of cytoplasmic

domain (Ilangumaran et al., 2000). Recycling by endocytosis may be a

regulatory mechanism to internalize HS bound molecules and redirect

glypican expression to different cell surface areas (Fransson, 2003). During

recycling, the HS chains are remodeled and the glypican distribution on the

cell surface might be determined by the GAG moiety of the molecule

(Mertens et al., 1996). Glypicans are mainly expressed during development.

The expression levels differ in a spatiotemporal manner, suggesting

glypicans to be involved in the regulation of morphogenesis (De Cat and

David, 2001; Song and Filmus, 2002).

The cell surface receptors CD44 and betaglycan are so called part-time

PGs since they exist both with and without GAG side-chains. The GAGs can

be either HS or CS. CD44 is a lymphocyte homing cell surface receptor and

has also been described as a receptor for hyaluronan (Aruffo et al., 1990).

Betaglycan mainly works as a co-receptor for TGF that can be bound both

by the core protein and GAG side-chains (Delehedde et al., 2001).

Extracellular matrix proteoglycans

Perlecan (Fig. 2) is the best-characterized extracellular HSPG and is present

in virtually all BMs in the body and also in other ECMs (Hassell et al., 2002;

Iozzo et al., 1994). Perlecan binds many ECM molecules and is thought to

be an important stabilizer both in matrix organization and cell-matrix

interactions. The perlecan core protein has an elongated structure divided in

five distinct domains. The most N-terminal domain may contain three HS or

CS side-chains. There is also an additional HS/CS attachment site in the

most C-terminal domain. The greatest deposition of perlecan during

embryonic development is in cartilage undergoing endochondral ossification

(Handler et al., 1997).

Agrin (Fig. 2) (Groffen et al., 1998) and collagen XVIII (Halfter et al.,

1998; Saarela et al., 1998) are also HSPGs found in different ECMs. Agrin

can harbor at least three HS side-chains and is best known as the organizer in

the assembly of the postsynaptic apparatus. There are several different splice

variants of agrin expressed in a tissue specific manner. Agrin has also

recently been implicated in the “immunological synapse”, the specialized

11

connection between T lymphocytes and antigen presenting cells in the

immune system (Bezakova and Ruegg, 2003). Agrin may act by its ability to

trigger rearrangement of the cytoskeleton and to aggregate lipid rafts in the

cell membrane.

Collagen XVIII is the only known member of the collagen family that

carries HS side-chains and is expressed in most BMs (Halfter et al., 1998).

Collagen XVIII exists in three different splice variants with tissue-specific

expression pattern (Muragaki et al., 1995), all with several potential HS side-

chain attachment sites (Halfter et al., 1998). Collagen XVIII might be

involved in the regulation of signaling between mesenchyme and epithelium

during organ morphogenesis (Lin et al., 2001).

The angiogenesis inhibitor endostatin is a small proteolytic fragment of

collagen XVIII. As collagen XVIII is abundantly expressed in the BMs

underlying blood vessel endothelium, cleavage of the collagen molecule to

produce endostatin may be a control mechanism to locally regulate

angiogenesis (Zatterstrom et al., 2000).

Serglycin — the only characterized heparin-proteoglycan

Serglycin found in connective tissue type mast cells is the only known

heparin-PG. As the core protein contains repetitive serine-glycine sequences,

attachment of multiple GAG chains is possible. Serglycin can in other cell

types instead of being substituted with heparin carry CS side-chains (Kolset

et al., 2004; Kresse et al., 1993).

A heparin-like, almost fully N- and O-sulfated, HS is also produced by

cultured oligodendrocyte-type-2 astrocyte progenitor cells (Stringer et al.,

1999). The core protein to which this heparin-like polysaccharide is attached

has not been identified and it is not known whether this polysaccharide also

can be found in vivo.

HS biosynthesis — a multienzyme machinery The biosynthesis of HS takes place in the Golgi apparatus and involves a

complex machinery of glycosyltransferases, an epimerase and different

sulfotransferases (Fig. 3) (Lindahl et al., 1998). Several of the enzymes exist

in multiple isoforms. All except one 3-O-sulfotransferase isoform are, like

other resident Golgi proteins, typical type II transmembrane proteins with a

short N-terminal cytoplasmic tail, a single transmembrane spanning domain

and the C-terminal part of the protein in the luminal space.

HS chains are synthesized covalently attached to a serine residue in the

core protein. Initially a tetrasaccharide linkage region is formed by different

glycosyltransferases. This linkage region is the starting point of both HS and

CS biosynthesis. The committing step for HS biosynthesis is the addition of

a GlcNAc residue to the linkage tetrasaccharide by glycosyl-transferases of

12

the exostosin-like (EXTL) family (Kim et al., 2001; Kitagawa et al., 1999).

Polymerization of the HS chain is carried out by enzymes encoded by genes

of the exostosin (EXT) gene family by stepwise addition of GlcA and

GlcNAc units from the corresponding UDP-sugars. HS polymerases exist in

at least two isoforms, EXT1 and EXT2, each of them harboring both GlcA

and GlcNAc transferase activity (Senay et al., 2000).

During the polymerization process, the chain also undergoes different

modifications (Lidholt and Lindahl, 1992). The first step is N-deacetylation

and N-sulfation of GlcNAc units, carried out by the bi-functional enzyme

glucosaminyl N-deacetylase/N-sulfotransferase (NDST). Four different

NDST-isoforms have been identified and cloned. This modification is

performed in a more or less block-wise pattern, resulting in N-acetylated and

N-sulfated regions, respectively, so called NA and NS domains. A minor

proportion of N-unsubstituted GlcN residues are also found (Norgard-

Sumnicht and Varki, 1995; van den Born et al., 1995). How these residues

are formed is not known. All further modifications generally take place in or

adjacent to N-sulfated regions, as the enzymes appear to depend on N-

sulfated glucosamine residues for substrate recognition (Salmivirta et al.,

1996). Recent findings however suggest that N-sulfation is not always a

prerequisite for the subsequent modifications to occur (Holmborn et al.,

unpublished). The next modifying enzyme, GlcA C5-epimerase (C5epi),

converts GlcA to IdoA. After this step, three different O-sulfation reactions

can take place. IdoA residues are the prime targets for 2-O-sulfation, but also

some of the remaining GlcA residues will become 2-O-sulfated. Only one 2-

O-sulfotransferase (2-OST) has been cloned (Kobayashi et al., 1997). The

next step is 6-O-sulfation of GlcNSO3 and GlcNAc residues; the latter is

generally only sulfated if located adjacent to an N-sulfated GlcN residue. 6-

O-sulfated GlcNAc residues in N-unsulfated HS have however been found

recently (Holmborn et al., unpublished). Three different 6-OST isoforms

have been cloned (Habuchi et al., 2000). Although 3-O-sulfation of GlcN

units is a rare event at least six isoforms of 3-O-sulfotransferases are known

(Kusche-Gullberg and Kjellén, 2003; Shworak et al., 1997; Shworak et al.,

1999). In all sulfation steps 3’-phosphoadenosine-5’-phosphosulfate (PAPS)

is used as sulfate donor. Some of the possible targets are, by unknown

mechanisms, left unmodified and this incomplete modification of the

residues contributes to the structural complexity of the polysaccharide chain.

Four N-deacetylase/N-sulfotransferase isoforms

NDSTs are bi-functional, single polypeptide enzymes that catalyze the N-

deacetylation of GlcNAc followed by N-sulfation of the same sugar

(Eriksson et al., 1994; Kjellén et al., 1992; Pettersson et al., 1991; Wei et al.,

1993). These enzymes have been designated a key regulatory role in the

production of heparin/HS, since subsequent modification reactions occur in

13

the vicinity of N-sulfate groups (Salmivirta et al., 1996). NDSTs are, like the

other HS biosynthetic enzymes typical type II transmembrane proteins. The

large catalytically active domain in the Golgi lumen harbors both N-

deacetylase and N-sulfotransferase activity.

Proteins with glucosaminyl N-sulfotransferase activity were first purified

from rat liver (NDST1) (Brandan and Hirschberg, 1988) and from mouse

mastocytoma (NDST2) (Pettersson et al., 1991). About ten years later two

additional NDST-isoforms, NDST3 and NDST4, were discovered and

cloned (Aikawa and Esko, 1999; Aikawa et al., 2001). The different NDSTs

show 66-80% amino acid identity and NDST3 and -4 are more closely

related to each other than to NDST1 or -2.

Both NDST1 and -2 show a strong uniform expression during embryonic

development (Aikawa et al., 2001). Both enzymes are also expressed in most

adult tissues with the highest levels of NDST1 in lung, liver and kidney and

NDST2 in liver, kidney and testis (Aikawa et al., 2001; Kusche-Gullberg et

al., 1998). In mast cells NDST2 is the only expressed isoform. NDST3 is

expressed during embryonic development in low amounts with a peak at

about E11 while NDST4 is expressed at very low amounts in the embryo. In

the adult mouse NDST4 expression is limited to the brain whereas NDST3

additionally is expressed in heart (Aikawa et al., 2001).

The role of the different NDST isoforms in heparin and HS biosynthesis

is not fully known. It was early suggested that NDST2 primarily is involved

in the biosynthesis of heparin (Eriksson et al., 1994). Comparative studies

made on cell lines that overexpress NDST1 and -2 (Pikas et al., 2000) further

supported this idea. HS from both cell lines showed increased N-sulfation

and the NDST2 overexpressing cell line showed the most dramatic increase

of N-sulfation resulting in a heparin like N-sulfation pattern. The wide

distribution pattern of NDST2 in tissues that do not produce heparin

however indicated that NDST2 also participates in HS production (Aikawa

et al., 2001; Kusche-Gullberg et al., 1998). It has now, by in vivo

experiments in NDST2 deficient mice, been verified that NDST2 mainly

works in heparin production (Forsberg et al., 1999 (paper I); paper III).

What regulates HS biosynthesis?

Little is known about the regulatory mechanisms in HS biosynthesis.

Suggestions have been made that dissimilarities in the transmenbrane region

of various enzyme isoforms might locate them to different Golgi

compartments. Some enzymes, like NDST2, may require specific co-factors

to operate (Eriksson et al., 1994). Different isoforms might prefer alternative

sets of other enzymes to cluster with, to form a “GAGosome”. These

enzyme-clusters, containing different sets of isoforms, might then produce

HS with different structures.

14

Figure 3 HS/heparin biosynthesis. Biosynthesis of HS/heparin takes place in the Golgi apparatus. The polymerization of the sugar backbone starts by synthesis of a tetrasaccharide linkage region covalently attached to a serine residue in a core protein. EXT polymerases elongate the sugar chain by alternating addition of N-acetylglucosamine (GlcNAc) and glucuronic acid (GlcA). Concomitant to polymerization, the sugar residues undergo several modification steps carried out by different enzymes. The first modifying step is N-deacetylation and N-sulfation of GlcNAc units carried out by N-deacetylase/N-sulfotransferase (NDST) enzymes. All further modifications (C5-epimerization, 2-O-, 6-O- and 3-O-sulfation) take place in the vicinity of N-sulfate groups. Since not all possible sites are modified, there is a high variability in HS structure.

15

The various NDST-isoforms have turned out to differ a lot in their N-

deacetylase and N-sulfotransferase activity ratio (Aikawa et al., 2001). This,

and variations in charge distribution in the substrate binding cleft of

sulfotransferase domains in different NDSTs, have brought about

suggestions that different isoforms have different substrate preferences. One

isoform might start the N-deacetylation/N-sulfation process and another

isoform continue or one isoform might deacetylate and another N-sulfate the

same HS chain (Bengtsson et al., 2003).

Both the pattern and level of expression are possible steps of regulation,

but why is then the NDST2 transcript so abundant when it seems to only

participate in the biosynthesis of heparin but not of HS (paper I and III)?

There are suggestions made that HS biosynthetic enzymes in general are not

transcriptionally but rather translationally regulated (Grobe and Esko, 2002).

This theory arose from translational studies of different NDST isoforms in

cell culture and other in vitro translation systems, in combination with

studies of the 5’-untranslated region (5’-UTR). The different NDST isoforms

have an unusually long and structured 5’-UTR that differs a lot between the

isoforms while the intraspecies similarity is high. Studies of the 5’-UTR of

other HS biosynthetic enzymes suggest that translational control might be a

general regulatory mechanism in HS biosynthesis. Several HS binding

growth factors and morphogens also have structured 5’-UTRs. Therefore

cells might be able to coordinate regulation of these factors and HS

production.

Additional posttranslational regulatory mechanisms, like phosphorylation

or glycosylation, may also be involved in the regulation of the available

amount of active enzyme.

Protein interactions A large number of proteins bind to HS/heparin including enzymes, enzyme

inhibitors, growth factors and ECM molecules (Bernfield et al., 1999) (Table

1). The high negative charge and the high variability in structural motives in

HS/heparin contribute to the large number of proteins that are able to bind to

them. Some interactions have been proven to be highly dependent on

specific HS sequences, whereas other interactions might simply depend on

electrostatic forces (Lindahl et al., 1994). It is important to bear in mind that

most interactions are only tested for in vitro under non-physiological

conditions that may for example cause protein unfolding. Further, heparin

and not HS has often been used. Many of the molecules with heparin binding

properties will never encounter this extreme form of HS in vivo. Interactions

may anyhow be valid also in the case of HS since the NS domains can

contain the specific sequence needed, or exhibit high enough negative

charge, to bind proteins with heparin binding capacity.

16

Table 1 HS/heparin binding proteins (partial list)a

Protein function Protein

ECM components Fibrin, Fibronectin, Interstitial collagens, Laminins,

Pleiotropin, Thrombospondin, Tenascins, Vitronectin

Morphpogens/Morphogen

binding

Activin, BMP2, -4, Chordin, Frizzled-related peptides,

Sonic Hh, Sprouty peptides, Wnt1-13

Growth factors/Growth

factor binding EGF family, FGFs, IGF-II, PDGF-AA, TGF , VEGF,

Follistatin, IGF BP-3, -5, TGF BP

Chemokines C-C, CXC

Cytokines IL-2-5, -7, -12, GM-CSF, Interferon , TNF- ,

Anti-angiogenic factors Angiostatin, Endostatin

Cell adhesion L-selectin, MAC-1, N-CAM, PECAM-1

Coagulation Antithrombin III, Factor Xa, Leuserpin, Tissue factor

pathway inhibitor, Thrombin

Proteinases Neutrophil elastase, Cathepsin G,

Tissue remodeling factors Tissue plasminogen activator, Plasminogen activator

inhibitor, Protease nexin I

Energy metabolism Agouti-related protein, ApoB, ApoE, Lipoprotein lipase,

Triglyceride lipases

Antimicrobial peptides Bac-5, -7, PR-39 aAdapted from Bernfield et al. 1999

HS binding can modulate the action of a protein in many ways. An HS chain

can bring molecules bound to it together and thereby increase the likelihood

for interaction between the molecules (Lander, 1998), like enzyme and

enzyme inhibitor, ligand and receptor. The binding to HS can also confer a

conformational change to the bound protein as in the case of antithrombin

activation. In growth factor signaling, HS might work both as a co-receptor

and by facilitating receptor dimerization (Bernfield et al., 1999; Esko and

Selleck, 2002; Nybakken and Perrimon, 2002). HS may also serve as a

retention molecule of growth factors in the ECM, where they are stored

protected from proteolytic degradation and released by cleavage of the HS

chain when needed (Flaumenhaft et al., 1989; Saksela et al., 1988). Cell

surface HS can then catch the released molecules and bring them together

with their receptors or simply enhance the concentration of a specific

molecule at the cell surface. Depending on the structure of the

polysaccharide, concentration gradients of soluble molecules can also be

built up and maintained by HS (Lander et al., 2002; Nybakken and Perrimon,

2002).

FGFs are among the most well studied HS/heparin binding proteins

(Ornitz, 2000; Pellegrini, 2001). HS can alter the diffusion of FGF, and

heparin has been shown to protect FGF from proteolytic degradation and

stabilize the protein when subjected to high temperatures. Crystal structure

studies of FGF-FGFR-heparin complexes have suggested that heparin

facilitates FGF-FGFR dimerization by increasing the affinity of FGF for its

17

receptor and by stabilizing the complex. Different sulfate groups are

important for high affinity binding of the ligand and the receptor and one

defined HS structure can act differentially on different FGFs.

Heparan sulfate in physiological processes Due to the many interactions between HS and different soluble signaling

molecules, ECM components as well as other functional groups of proteins,

HS can be expected to play an important role in many physiological

processes. In 1991 cell culture experiments pointed towards a more active

role of HS in FGF signaling than previously thought (Rapraeger et al., 1991;

Yayon et al., 1991). Since then, the evidence for the in vivo necessity of HS

in physiological events has been growing

Our knowledge of the in vivo functions of HSPGs have come from studies

of animal mutant phenotypes and human disorders involving mutations of

proteoglycan core proteins or GAG biosynthetic enzymes. The first in vivo

evidences of HS and HSPGs playing an important role in developmental

processes came from studies of Drosophila mutants.

Drosophila mutants

Much of our knowledge about HS in vivo functions is derived from studies

in the fruit fly Drosophila melanogaster. Studies during larval stages and of

wing formation have proven HS to be an important effector for several

growth factors and morphogens (Table 2). Defects in HS and HSPG

production can affect a signaling system both by direct alterations of ligand-

receptor interactions as well as by alterations in the building of concentration

gradients.

The first in vivo evidence for HS being directly involved in a specific

signaling event came in 1997 with the sugarless (also called suppenkasper or

kiwi) mutant (Binari et al., 1997; Häcker et al., 1997; Haerry et al., 1997).

The sugarless gene, is a UDP-glucose dehydrogenase, necessary for UDP-

GlcA production. Lack of this enzyme abolishes both HS and CS

biosynthesis. Sugarless mutants exhibit embryo patterning defects very

similar to Wg (Wnt homologue) mutants, and overexpression of Wg could

partially rescue the phenotype (Häcker et al., 1997). Selective degradation of

GAGs in wild-type embryos showed that only specific degradation of HS

mimicked the sugarless phenotype (Binari et al., 1997). Further, the

phenotype could be rescued by injection of exogenous HS. HS was

suggested to affect Wg signaling by limiting its diffusion, thereby facilitating

the binding of Wg to its receptor.

18

Table 2 Drosophila HS biosynthetic enzymes and HSPGs

Drosophila gene Mammalian homolog Required for signaling

by Slaloma PAPS transporter Hh1

Wg1

Sugarlessb UDP-glucose dehydrogenase FGF2

Wg3-5

Beta4GalT7b Beta4-galactosyltransferase Dpp6

Hh6

Tout-velu (ttv) EXT1 Dpp7, 8, 11

Hh7-11

Wg7, 8, 11

Sister of ttv EXT2 Dpp7, 8, 11

Hh7, 8, 11

Wg7, 8, 11

Brother of ttv EXT-like 3 Dpp7, 8

Hh7, 8

Wg7, 8

Sulfateless NDST1 FGF2

Wg12

DHS6ST 6-OST FGF13

Dally Glypican Dpp14-16

Hh17

Wg12, 15, 16, 18

Dally-like Glypican Hh17, 19, 20

Wg18

Syndecan Syndecan Slit21

Trol Perlecan FGF22

Hh22

Dpp=Decapentplegic, TGF -homologue; FGF=Fibroblast growth factor; Hh=Hedgehog Wg=Winglessaaffects all sulfation processes baffects both HS and CS biosynthesis 1Luders et al., 2003 2Lin et al., 1999 3Häcker et al., 1997 4Haerry et al., 1997 5Binari et al., 1997 6Nakamura et al., 2002 7Takei et al., 2004 8Han et al., 2004a 9Bellaiche et al., 1998 10The et al., 1998 11Bornemann et al., 2004 12Lin and Perrimon, 1999 13Kamimura et al., 2001 14Jackson et al., 1997 15Tsuda et al., 1999 16Fujise et al., 2001 17Han et al., 2004b 18Giraldez et al., 2002 19Lum et al., 2003 20desbordes et al., 2003 21Steigemann et al., 2004 22Park et al., 2003

Sulfateless is the only known Drosophila NDST; this mutant gave the in vivo

proof of HS being involved in FGF-FGFR signaling (Lin et al., 1999). Since

then several Drosophila mutants with defective HS biosynthesis have been

examined, all with developmental phenotypes due to disturbed morphogen

gradient formation and/or signaling (Nybakken and Perrimon, 2002).

Sulfateless and Sugarless mutants have phenotypes similar to mutants

lacking functions of two Drosophila FGFRs (Lin et al., 1999). Mutations in

Tout-velu (ttv), sister of ttv or brother of ttv (EXT and EXTL homologues)

have recently been reported by two groups to disturb the morphogen gradient

19

of Hh, Dpp (BMP2/4 homologue) and Wg (Han et al., 2004a; Takei et al.,

2004). However, Takei and co-workers report the signaling of all three

morphogens to be disturbed while Han and co-workers came to the

conclusion that both Hh and Dpp signaling are disturbed in all three mutants

while Wg signaling is only disturbed in brother of ttv or ttv/sister of ttv

double mutants.

A gene called Pipe, with 2-O-sulfotransferase homology has also been

cloned (Sen et al., 1998; Sergeev et al., 2001). The Pipe gene has later been

found to have ten homologous exons that all encode sulfotransfease

domains. Pipe is required in the dorsoventral polarization mechanism in the

Drosophila embryo. Whether Pipe is involved in HS biosynthesis is however

not yet known.

Four Drosophila HSPGs have also been identified, Dally (Lin and

Perrimon, 1999) and Dally-like (Khare and Baumgartner, 2000) (glypican

homologs), Syndecan (Spring et al., 1994) and Trol (Voigt et al., 2002)

(perlecan homolog), and their involvement in signaling has been examined

(Table 2). Dpp, FGF, Hh and Wg are the systems that mainly seem to be

affected by HS and HSPGs but also slit/robo signaling has turned out to be

affected in a syndecan mutant (Steigemann et al., 2004). It has been shown

that HS can affect a signaling system mainly in two ways. HS can both

facilitate the establishment of morphogen gradients and thereby regulate the

concentration of the specific protein and/or directly be involved in the

interaction between the ligand and receptor. In the case of HSPG mutants it

is not easily distinguished whether it is the lack of the core protein, the HS

chains on this specific core protein or both in combination that generate the

phenotype.

Biochemical studies of GAGs produced by sugarless, tout-velu and

sulfateless mutant larvae have been performed (Toyoda et al., 2000). Both

HS and CS biosynthesis are severely compromised in sugarless mutants

whereas tout-velu and sulfateless mutants only show defective HS

biosynthesis, as expected. Tout-velu mutants do not produce any detectable

amounts of HS while sulfateless mutants produce almost wild-type levels of

unsulfated polysaccharide chains.

HS in mammalian systems

Lack of, or defective HS biosynthetic enzymes and HSPG core proteins have

turned out to cause several human syndromes. Several mouse strains with

defects in homologous genes have verified the involvement of HS-

biosynthetic enzymes and HSPGs in embryonic development and

pathological states.

20

Table 3 Mouse models with defective HS biosynthesis

agenetic background dependent 1Lin et al., 2000 2Inatani et al., 2003 3Fan et al., 2000 4Ringvall et al., 2000 (paper II) 5 Ledin et al., unpublished (paper III) 6Ringvall et al., unpublished (paper IV) 7Forsberg et al., 1999 8Humphries et al., 1999 9Li et al., 2003 10Bullock et al., 1998 11Merry et al., 2001 12McLaughlin et al., 2003 13HajMohammadi et al., 2003

Mouse models with defective HS biosynthesis

In total, six of the known enzymes involved in HS biosynthesis have been

deleted in mice (Table 3). The first mammalian model organism with proven

HS structure defects was the NDST1-/- mouse (Ringvall et al., 2000). The

majority of NDST1-/- mice die neonatally due to breathing difficulties (Fan et

al., 2000; Ringvall et al., 2000 (paper II)). Head, eye and skeletal defects were also

observed (paper IV and unpublished data). HS from different organs from

E18.5 NDST1-/- embryos have much reduced N-sulfation levels and the total

sulfation level is almost 50% lower than in wild-type (Ringvall et al., 2000 (paper II)).Earlier, a 2-OST deficient gene trap mouse had been published (Bullock et

al., 1998), but without any HS structural data. 2-OST-/- mice suffer from

bilateral renal agenesis, skeletal, eye and brain defects (Bullock et al., 1998;

McLaughlin et al., 2003). Biochemical analyses were later performed and

showed that HS produced by cultured fibroblasts from 2-OST-/- embryos has

an altered structure (Merry et al., 2001). This HS completely lacks 2-O-

sulfation but has an overall increased sulfation level due to elevated N- and

6-O-sulfation.

C5epimerase-/- (C5epi) mice (Li et al., 2003) have defects similar to both

the 2-OST-/- and NDST1-/- strains and also die neonatally. HS prepared from

Inactive gene HS/heparin status Phenotype

EXT11 No HS production Death by E8.5

Gastrulation defect

EXT1, CNS specific2 No HS production Neonatal death

Brain patterning and axonal

guidance defect

NDST13-6 Low N- and total sulfation4, 5 Embryonica and neonatal death

Lung3, 4, head, eye and skeletal

defects4, 6

NDST27, 8 HS normal, no heparin

production7Mast cell defect

C5-epimerase9 No IdoA Neonatal death

Renal agenesis, lung, eye and

skeletal defects

2-OST10-12 No 2-O-sulfation11 Neonatal death

Renal agenesis, skeletal, eye10

and brain defects12

3-OST113 Reduced levels of anticoagulant

HS

Postnatal deatha

Intrauterine growth retardationa

No obvious defecta

21

whole E17.5-18.5 C5epi embryos lacks IdoA and shows a lower amount of

2-O-sulfate groups that are redistributed to GlcA. The total sulfation level is

however increased due to elevated N- and 6-O-sulfation levels (Li et al.,

2003).

EXT1-/- mice have the earliest phenotype with death before E8.5 (Lin et

al., 2000). These embryos likely lack HS completely as no HS could be

prepared from EXT1 deficient ES cells. Heterozygous ES cells produce half

as much HS as wild-type ES cells, demonstrating biallelic expression of

EXT1. EXT1 has also been selectively disrupted in the central nervous

system (CNS) (Inatani et al., 2003). These mice exhibit several brain defects

and die perinatally. The HS status in the CNS of these mice is however not

fully known. No HS was found on syndecan-3 core protein, or by

immunostaining of cortical neurons with antibodies recognizing low-sulfated

HS. Small amounts of HS on core proteins other than syndecan-3, and not

possible to visualize with the antibodies used, might therefore still be present

as the deletion of EXT1 in CNS cells might not be complete with the used

strategy.

Lack of NDST2 has been shown to abolish heparin biosynthesis but does

not seem to affect HS biosynthesis (Forsberg et al., 1999 (paperI)). Two

NDST2-/- mouse strains exist, both with a severe connective tissue type mast

cell phenotype (Forsberg et al., 1999 (paper I); Humphries et al., 1999).

3-OST1 is known to be the most critical of the 3-OST isoforms for

production of antithrombotic HS sequences present on blood vessel

endothelial cells. This antithrombotic HS has been proposed to contribute to

the nonthrombogenic properties of blood vessels. 3-OST1-/- mice do

however not exhibit any procoagulant phenotype (HajMohammadi et al.,

2003). Depending on genetic background, 3-OST-/- mice are affected by

intrauterine growth retardation and postnatal lethality. The cause of lethality

and growth retardation has not yet been elucidated; altered hemostasis does

not appear to be the reason.

Defective GAG biosynthesis — a cause of human disorders

PAPS is the sulfate donor for synthesis of all sulfated GAGs and also for all

other known post-translational sulfation reactions. Mutations in PAPS

synthetases are causing disproportionate dwarfism in both mouse and man

(Schwartz and Domowicz, 2002; ul Haque et al., 1998). Biochemical

analysis of cartilage from such mice shows a significant undersulfation of

GAGs. Defects in the sulfate transport into cells also give rise to different

dwarfing syndromes (Schwartz and Domowicz, 2002; ul Haque et al., 1998).

Hereditary multiple exostoses (HME) is the only human syndrome that

has been directly linked to mutations in HS specific biosynthetic enzymes.

HME is an autosomal dominant disorder and is the most common cause of

benign skeletal tumors in man (McCormick et al., 1999). Heterozygous

mutations in the HS glycosyltransferases EXT1 or EXT2, have been shown

22

to be the cause of HME (Ahn et al., 1995; Lind et al., 1998; McCormick et

al., 1998; Stickens et al., 1996). Interestingly, EXT1+/- mice do not develop

exostoses (Lin et al., 2000).

A mouse model with increased HS degradation

The generation and characterization of transgenic mice overexpressing

heparanase in most tissues, have clearly demonstrated the importance of a

regulated HS degradation (Zcharia et al., 2004). Investigators specifically

studied embryo implantation, mammary gland morphogenesis, hair growth

and feeding behavior. Homozygous heparanase overexpressing mice are

fertile with normal litter size. However, the number of implanted embryos in

the uterus during first trimester of pregnancy was significantly increased.

Transgenic mammary glands showed overbranching, hyperplasia and

widening both in virgin mice and during pregnancy. Hair growth rate was

much higher in transgenic animals than in wild-type while food intake was

lower. Altered plasticity in organs might be facilitated by enhanced

vascularization and alterations in BM structures. Cell surface binding studies

on embryonic fibroblasts with FGF2, showed that overexpression of

heparanase reduces the amount and/or size of HS available for FGF2

binding. In conclusion, degradation of HS seems to be important in many

dynamic processes, at least in the adult mouse.

Defects in the extracellular matrix

Perlecan has been proven to be an important structural component in the

ECM since deficient mice display stress sensitive BMs and disorganized

cartilage matrix (Arikawa-Hirasawa et al., 1999; Costell et al., 1999).

Development of exencephaly in perlecan null embryos is thought to be

caused by weak BMs and many of the embryos die due to a leaking heart at

E10.5, when the heart starts to beat (Costell et al., 1999). The skeletal

malformations, with severe chondrodysplasia, seen in these mice are

suggested to be caused by structural defects of the ECM (Costell et al.,

1999). Incorrect FGFR3 signaling has also been suggested as a cause of the

phenotype (Arikawa-Hirasawa et al., 1999). Later publications report several

cardiac malformations in perlecan deficient embryos suggested to be caused

by defective interaction between endocardial- and neural crest-derived

mesenchymal cells (Costell et al., 2002; Gonzalez-Iriarte et al., 2003).

In human, mutations in the perlecan gene has been recognized as the

cause of at least two syndromes. The Silverman-Handmaker type of

dyssegmental dysplasia is a lethal form of short-limbed dwarfism caused by

functional null mutations in the perlecan gene (Arikawa-Hirasawa et al.,

2001). Affected individuals show features similar to perlecan-/- mice.

Schwarz-Jampel syndrome is the other human syndrome caused by

mutations in the perlecan gene. These mutations result in production of truncated

perlecan forms (Arikawa-Hirasawa et al., 2002). Schwarz-Jample patients also share

23

features of perlecan-/- mice with skeletal changes and myotonic myopathy.

Alterations in expression levels of perlecan have been observed in several

other pathological states in both human and animal models (Ailles et al.,

1993; Cohen et al., 1994; Conde-Knape, 2001; Kimura et al., 2000).

Another ECM HSPG involved in a human syndrome is collagen XVIII.

Mutations in the COL18A1 gene cause a syndrome known as the Knobloch

syndrome characterized by encephalocele and ocular defects (Conde-Knape,

2001). Collagen XVIII deficient mice show pathological alterations in the

BM underlying the retinal pigmented epithelium similar to age related

macular degeneration, the major cause of blindness in the Western world

(Marneros et al., 2004).

Differently spliced forms of agrin are expressed in distinct patterns (Bowe

and Fallon, 1995). Mice made completely devoid of an agrin-form

specifically expressed by neurons also showed very low levels of other

forms of agrin (Gautam et al., 1996). Homozygous embryos developed

normally until the day before birth, but died in utero or were stillborn. The

perinatal death was due to disruption of neuromuscular function with

disturbed formation of postsynaptic acetylcholine receptor aggregates.

Lack of glypican-3 causes multiple defects

The only cell surface HSPG implicated in any human syndrome this far is

glypican-3 (Pilia et al., 1996). The X-linked Simpson-Golabi-Behmel

syndrome (SGBS) is among other things characterized by pre- and postnatal

overgrowth, visceral and skeletal anomalies and congenital heart defects and

is also associated with an increased risk of developing embryonal tumors

(Neri et al., 1998). Mutational analysis has revealed that many SGBS

patients display different loss of function mutations in the glypican-3 gene

(Veugelers et al., 2000). Glypican-3 deficient mice exhibit several of the

clinical features observed in SGBS patients (Cano-Gauci et al., 1999).

Suggestions have been made that glypican-3 might be involved in insulin-

like growth factor signaling (Pilia et al., 1996). This theory has however

been rejected in a later publication (Chiao et al., 2002).

Altered syndecan expression gives subtle phenotypes in mice

Syndecans are considered to be of importance in growth factor stimulation,

and have also been suggested to be involved in binding to ECM and in focal

adhesion formation (Rapraeger, 2000; Yoneda and Couchman, 2003). By

crossing syndecan-1 deficient mice with transgenic mice that have ectopic

Wnt1 expression, syndecan-1 could be shown to take part in Wnt1 signaling

(Alexander et al., 2000). Syndecan-1 deficient mice were otherwise reported

to be normal. A later publication showed that syndecan-1-/- mice have

increased leukocyte adhesion in a retina perfusion experiment suggesting

syndecan-1 to be a negative regulator of leukocyte-mediated inflammatory

responses (Gotte et al., 2002).

24

Syndecan-3 is predominantly expressed in the nervous system. Studies of

syndecan-3 deficient mice and transgenic mice with syndecan-1 expression

in the hypothalamus have shown that syndecans are involved in feeding

behavior (Reizes et al., 2001). Mice deficient in syndecan-3 are otherwise

healthy and fertile but additionally exhibit enhanced long term potentiation

and impaired hippocampal memory (Kaksonen et al., 2002; Pavlov et al.,

2002). Syndecan-3 may therefore be an important modulator of synaptic

plasticity. By binding via their HS chains syndecans are suggested to

potentiate the action of the feeding-regulatory agouti-related protein and

agouti signaling protein.

Studies on fibroblasts isolated from syndecan-4 deficient mice could only

vaguely, in cell culture during very specific conditions, confirm the

participation in focal adhesion formation (Ishiguro et al., 2000b). Syndecan-

4 deficient mice showed no macroscopic defects and reproduced normally.

In a later publication enhanced occurrence of fetal vessel degeneration in the

placenta, supposedly caused by increased thrombous formations, was

reported in the absence of syndecan-4 (Ishiguro et al., 2000a).

Core protein versus side-chains

It is far from clear what separate roles the core protein and the side-chains

play in different events. From the results of studies of PG functions in vivo

and in vitro it is most often not possible to read out the contribution of the

protein and GAG chains separately. It is, as for any other system studied,

always difficult to tell whether an absence of reaction is caused by any

compensatory mechanism in the system, or whether the impaired system

studied is not needed normally. A lost component might be well

compensated for in most situations but may be needed during very specific

conditions. When taking away one component in a system most certainly a

well-regulated balance is disrupted and nature struggles to compensate this

balance with another as soon as possible. Probably PGs must be looked at as

a unity, where both GAG chains and core protein are important.

Some attempts have been made to separate the function of the protein and

the GAG moieties. Syndecan-1 has been sequentially mutated to eliminate

one, two or all three HS chain attachment sites (Langford et al., 1998).

Studies of cell-matrix and cell-cell adhesion and cell invasion into collagen-

gels demonstrated that all three HS attachment sites are needed for optimal

mediation of these activities and that the different attachment sites may have

a functional hierarchy. A mouse strain that expresses perlecan lacking most

of its HS shows an eye-degenerative phenotype but excluded perlecan HS as

being the most important HS in the kidney since no kidney dysfunction or

histological alterations could be observed (Rossi et al., 2003). As only the

three GAG attachment sites in the N-terminal domain, but not the attachment

site in the C-terminal domain, were excluded in the mutant perlecan, it may

still carry at least one HS chain. It must therefore be emphasized that the

25

GAG status was not exhaustively studied. In a later publication, growth

control of smooth muscle cells of this mouse strain was shown to be

disturbed (Tran et al., 2004). One possible explanation is that HS in the

ECM normally sequesters mitogens. ECM with mutant perlecan prepared

from the cultured smooth muscle cells showed that the HS moiety of the PG

is highly needed for efficient binding of FGF2.

HS during development The importance of HS in developmental processes becomes more and more

evident. As discussed above, all Drosophila mutants with defects in HS

biosynthesis or lack of HSPG core proteins suffer from serious

developmental defects, often with striking similarities to other mutants

known to lack signaling components such as Wg or FGF/FGFR. The studies

of mouse strains that lack HS-biosynthetic enzymes or core proteins further

show that HS is also important in mammalian development.

HS in signaling and morphogen gradients

Signaling during development must be strictly regulated for each cell to get

the right amount of signal at the right time. Signaling is a prerequisite for

cells to differentiate and for embryo and organ patterning. The supply of

ligand is almost always the limiting step in initiating a signaling process

(Freeman and Gurdon, 2002). The available amount of ligand can be

regulated by the level and location of ligand transcript, the spreading of

ligand from the producing cells and also by the spreading of signaling

modulators such as inhibitors.

Signaling can be divided into threshold and concentration-dependent

(morphogen) signaling (Freeman and Gurdon, 2002). In a threshold

signaling event, the receiving cell will respond in a specific predefined way

as soon as the threshold concentration is reached. In a concentration

dependent signaling event the receiving cell can instead react in several ways

depending on the ligand concentration.

HS can participate in all of the above mentioned steps that regulate ligand

availability. An HS dependent signal may alter transcription levels of a

ligand. HS can also alter the distance a ligand is transported away from the

producing cell and also regulate the local concentration of the molecule

available for receptor binding at the receiving cell. Further, HS can bind

signaling inhibitory molecules and also itself work as enhancer or inhibitor

in a signaling system depending on specific sulfate patterns (Bernfield et al.,

1999; Esko and Selleck, 2002; Nybakken and Perrimon, 2002; Viviano et al.,

2004).

26

Figure 4 Models of HS regulation of developmental signaling. (Adapted from Nybakken and Perrimon, 2002) (A) Dimerization: dimerization of ligands and/or receptors is sometimes a prerequisite for initiation of effective signaling and can be facilitated by HS. (B) Receptor-ligand stabilization: HS can stabilize the ligand-receptor complex and thereby promote and prolong effective signaling. (C) Two-dimensional sliding: HS can by low affinity binding of the ligand facilitate a quicker movement of the ligand than possible by free diffusion. HS also reduces the movement dimensionality to the plane of the cell surface. (D) Surface transport: ligands can be actively passed to neighboring cells by exchange of ligands via HS side-chains. Suggestions have also been made that a ligand-HSPG complex can move between cells by transfer of whole cell membrane regions, so called lipid rafts, to adjacent cells. (E) Transcellular transport: a ligand-HSPG complex can be transported in a vesicle from the surface of a ligand-producing cell to a receptor-producing cell. Fusion between the ligand-HSPG-containing vesicle and a receptor-containing vesicle can take place and signaling is activated.

27

The mechanisms whereby HSPGs affect the regulation of signaling

during development are far from defined and may work slightly different in

various signaling systems. Five main mechanisms have been suggested:

dimerization, receptor-ligand complex stabilization, two-dimensional

sliding, surface transport and transcellular transport (Fig. 4) (Nybakken and

Perrimon, 2002). By two-dimensional sliding, low affinity binding of a

molecule would allow it to slide along HS chains to reach target cells further

away in a shorter time than otherwise possible. In the transcellular model,

molecules would hitchhike on HSPGs and by vesicular transport be

relocated at the cell surface of the same cell or carried to distantly located

cells by moving through neighboring cells in a phagocytosis related way.

In all of these mechanisms, the specific HS sulfation pattern should be

highly important. A cell could by production of HS sequences with higher or

lower affinity for a molecule either attract or repel different molecules

thereby facilitate local concentration differences. By production of

sequences that promote dimer formation or complex formation, and by

production of sequences that inhibit these events, a cell could regulate the

signaling activity.

Dynamic alterations in HS structure during development

Structural changes in HS have turned out to be part of normal developmental

processes, during aging and also in many pathological states, such as cancer,

Alzheimer’s disease and diabetes. Studies have shown that the GAG

synthesis is altered in various tissues in diabetic patients and experimentally

diabetic rats. HS from the liver of diabetic rats has less sulfate groups

compared to HS from wild-type liver (Kjellén et al., 1983) and the NDST1

activity is lowered in hepatocytes from diabetic rats (Unger et al., 1991). An

age dependent increase in 6-O-sulfation can be seen in HS from human aorta

(Feyzi et al., 1998).

In developing murine neuroepithelium, a switch in HS production from a

structure that favors FGF2 binding and signaling to one that instead

promotes the FGF1 pathway is seen at a time point that coincides with the

start of FGF1 expression and of neuronal differentiation (Nurcombe et al.,

1993). Biochemical examination of HS from the different time points

showed altered sulfation pattern, chain length and number of sulfated

domains (Brickman et al., 1998a; Brickman et al., 1998b). Cell culture

mitogenic assays have in addition shown that specific HS sequences favor

signaling by either FGF1 or -2 (Guimond et al., 1993; Pye et al., 2000).

It has also been shown that different FGF-FGFR complexes require

various HS sequences to assemble. Assays have been performed to probe for

in situ tissue-specific HS in mouse embryos at various stages (Allen et al.,

2001; Allen and Rapraeger, 2003). Different FGF-FGFR combinations

bound in spatiotemporally specific patterns. Specific HS requirements were

28

identified for complex formation and signaling for each FGF-FGFR

combination tested (Allen and Rapraeger, 2003).

The in situ expression pattern of different HS biosynthetic enzymes

uniformly suggest them to be strictly regulated to yield differences in the

fine structure of HS at different locations during development (Ford-Perriss

et al., 2002; Habuchi et al., 2003; Kamimura et al., 2001; Nogami et al.,

2004; Stickens et al., 2000). The expression of different HS sulfotransferases

during mouse brain development shows different profiles at various stages

that might underlie the differences seen in HS structure (Ford-Perriss et al.,

2002).

As it is thought that several HS biosynthetic enzymes may be

translationally regulated (Grobe and Esko, 2002), it therefore still remains

for a relationship to be established between transcriptional and translational

levels.

A new mechanism for HS structure modification

A recent discovery has shown that the modification of HS structure is not at

an end with the sulfation processes described above, but also involves

desulfation. An endosulfatase, Qsulf1, which in vitro can act both in the

Golgi apparatus and on the cell surface, removes 6-O-sulfate groups in the

most highly sulfated domains of HS (Ai et al., 2003; Viviano et al., 2004).

Qsulf1 was identified in a screen for Sonic Hh response genes activated

during somite formation in quail embryos and shows a developmentally

regulated expression pattern (Dhoot et al., 2001). Antisense inhibition of

Qsulf1 resulted in less expression of MyoD in muscle progenitors in the

somites. MyoD is Wnt-inducible, suggesting Qsulf1 to be of importance for

Wnt signaling in accordance with the earlier Drosophila studies. HS

inactivation of Wnt signaling is most likely performed by binding the ligand

with high affinity, thereby preventing it to bind its receptor (Ai et al., 2003).

Qsulf1 6-O-desulfation would convert the cell surface HS to bind to Wnt

with low affinity, thereby instead facilitate ligand-receptor interaction.

Qsulf1 has also been shown to regulate binding between Noggin, a BMP

antagonist, and HS (Viviano et al., 2004). Noggin needs N-, 6-O- and 2-O-

sulfate groups for efficient HS binding. Qsulf1 6-O-desulfation will release

Noggin from the cell surface thereby increasing the accessibility of BMP to

its receptor.

For the above reasons, Qsulf1 and possibly other related endosulfatases

may be important actors in the regulation of sulfation pattern heterogeneity

in HS.

HS degradation and pathology

Altered degradation of HS is also implicated in pathological processes.

Metastasis experiments have proven that elevated expression of the HS

degrading enzyme heparanase conveys high metastatic potential (Vlodavsky

29

et al., 1999). Furthermore, different pathological conditions collectively

known as mucopolysaccharidosis are characterized by accumulation of

GAGs intracellularly due to defective degradation (Kresse, 1973;

Yogalingam and Hopwood, 2001).

The impact of HS on fundamental developmental processes

No HS — no gastrulation

EXT1 deficient embryos die, as described above, by E8.5 due to defects in

mesoderm formation and failure of extraembryonic egg cylinder elongation

(Lin et al., 2000). Gastrulation fails and there is a general lack of organized

mesoderm and extraembryonic tissues. Studies of embryoid bodies showed

that the expression of several early embryonic differentiation markers such

as brachyury and BMPs, was disturbed in EXT1-/- bodies. It was further

shown that EXT1-/- embryos could not bind Ihh to the cell surface as a

consequence of lack of HS. Ihh is known to be involved in differentiation of

extraembryonic tissues among many other developmental events (Becker et

al., 1997).

Development without 2-O-sulfate groups

The 2-OST deficient mice (Bullock et al., 1998) discussed above were

actually not generated primarily to study the importance of this enzyme in

HS biosynthesis and the role of HS in developmental processes. These mice

resulted from a gene trap screen designed to identify genes expressed during

early organogenesis. One site of strong 2-OST expression is in the

embryonic mesenchyme, especially at the sites of epithelial-mesenchymal

interactions. The 2-OST expression was followed in kidney explants and

shown to be rapidly down-regulated as the mesenchyme starts to

differentiate. The renal agenesis observed in 2-OST-/- mice was shown to

originate from failure of mesenchymal condensation around the epithelial

ureteric bud and initiation of branching morphogenesis. The mechanisms for

this, and also for other phenotypes in 2-OST deficient embryos, were

speculated to origin from erroneous adhesive properties of the mesenchyme

in addition to disturbed signaling by FGFs and Wnts. However, even without

knowledge of the exact molecular mechanisms involved, this clearly shows

that HS plays a role in mesenchymal-epithelial interactions during

organogenesis, at least in some organs.

In a later publication 2-OST deficient mice were also shown to develop a

thin cerebral cortex due to a significant reduction in cell proliferation in it

(McLaughlin et al., 2003). Disturbed FGF and Wnt signaling were suggested

as plausible mechanisms.

30

Biochemical binding studies with HS prepared from 2-OST-/- embryonic

fibroblasts showed that binding of FGF1 and -2 was significantly lower than

to wild-type HS. Nevertheless, mutant cells could still potentiate a signaling

response (assayed by MAPK phosphorylation) to FGF1 and -2 as strong as

seen in wild-type cells. This might suggest that the 2-O-sulfate group is

indeed necessary for high affinity binding of FGF1 and -2 but that this strong

binding is not necessary to potentiate signaling, at least not for these

particular cells. There are however evidence that cells can potentiate a

response to FGF1 and -2 without access to sulfated GAGs but the kinetics of

at least FGF2 binding is altered and the expected cellular response (e.g.

proliferation) to the signal fails (Fernig et al., 2000).

HS involvement in cell migration and branching morphogenesis

HS has also been proposed to function as a guidance cue for migrating cells

and in branching morphogenesis. This has been verified by the branching

failure of the renal bud seen in 2-OST deficient mice (Bullock et al., 1998)

as well as by mice that specifically lack EXT1 in the CNS, which among

other defects show axonal guidance defects (Inatani et al., 2003). Lack of

major commissural tracts, where axons normally cross the midline of the

brain, were observed. Axons from the cortex failed to approach the midline

and instead extended ventrally. Misrouting of retinal axons at the optic

chiasm similar to Slit1/Slit2 double deficient mice was also observed.

Crossing of Slit2-/- mice, which show few guidance errors, and EXT1+/- mice

caused elevated axon misrouting in EXT1+/-/Slit2-/- offspring. It was

therefore suggested that HS plays a dose dependent role in Slit-mediated

retinal axonal guidance. O-sulfation of HS has earlier been suggested to be

critical for Slit binding (Ronca et al., 2001).

Vascular endothelial growth factor-A (VEGF-A) is expressed in at least

three splice forms that differ in their HS binding properties (Ng et al., 2001).

The variable ability to bind to HS can locate different isoforms of VEGF-A

to alternative areas. Experiments have been performed with mice that only

express one form of VEGF-A that lacks HS-binding properties (Ruhrberg et

al., 2002). This resulted in a disturbed VEGF-A concentration gradient and

altered the distribution of endothelial cells within the growing vasculature

with less branching as an effect. Opposing effects were seen in mice that

instead only expressed one VEGF-A form that binds HS. This suggests HS

to be important in regulation of vascular branching pattern.

The above results are also in accordance with the migration and branching

defects seen in sugarless and sulfateless Drosophila mutants caused by

disturbed FGF-FGFR signaling (Lin et al., 1999). Sugarless and sulfateless

mutants both give phenotypes very similar to the ones seen in the

Drosophila FGFR mutants heartless, breathless (FGFRs) and branchless

(FGF). These defects are thought to depend both on erroneous spatial

distribution of the ligand as well as compromised ligand-ligand and/or

31

ligand-receptor complex interactions. Interestingly, the expression pattern of

the only known Drosophila 6-OST gene clearly resembles that of breathless

in the developing tracheal system (Kamimura et al., 2001).

32

Present investigation

Aim of study

The general aim of this study was to investigate the in vivo functions of HS

and heparin by elucidating the effect of lack of different NDST-isoforms in

the mouse. The aims of the individual papers were to:

investigate the contribution of NDST2 to HS and heparin

biosynthesis and how the mammalian body develops and reacts

without this enzyme (paper I)

investigate the contribution of NDST1 to HS biosynthesis and how

the mammalian body develops and reacts without this enzyme

(paper II)

investigate the contribution of NDST1 and NDST2 to HS

biosynthesis in liver and increase the knowledge about the

regulation of HS biosynthesis. We also wanted to know how altered

HS structure affects liver development (paper III)

investigate the effect of altered HS structure on skeletal

development (paper IV)

33

Results

Paper I — NDST2 deficient mice lack heparin and have a severe mast cell defect

In this study, we demonstrated a physiological role of endogenous heparin in

the mammalian body. This was done by the generation of a mouse strain

devoid of NDST2.

NDST2 does not seem to be essential during development as shown by

the expected ratio of the different genotypes from heterozygous crossings.

Both male and female NDST2-/- mice were healthy and fertile and showed,

except for a mast cell phenotype, no obvious pathological changes even at 20

months when kept at clean conditions in a barrier facility.

Characterization of GAGs purified from peritoneal cells revealed that

sulfated heparin was absent in NDST2-/- cells. Still, there seemed to be a

production of unsulfated heparin precursor.

The fact that heparin biosynthesis was severely affected by the lack of

NDST2 could be expected since this cell type expresses high levels of

NDST2 but no or very little NDST1. Somewhat surprisingly, biochemical

analyses made on liver and kidney HS revealed no differences in N-sulfation

pattern. As NDST2 is expressed in both liver and kidney as well as in most

other organs (Kusche-Gullberg et al. 1998), it was expected that lack of this

enzyme would also cause alterations in HS structure.

Normal mast cells contain large amounts of secretory granules. An array

of proteases, histamine and other mediators are co-stored with heparin in

these granules and are released upon mast cell activation (Stevens and

Austen, 1989). Mast cell granules from NDST2-/- animals displayed an

altered morphology and contained severely reduced amounts of histamine

and no mast cell proteases. NDST2-/- mast cells contained vacuole-like

structures and a few smaller granules. There were hardly no detectable

protease activity or protein in the abnormal mast cells although the

transcriptional levels of mast cell proteases were the same as in wild-type

mast cells. Furthermore, the number of mast cells was decreased

approximately fourfold in NDST2-/- animals. Unexpectedly, in vivo

intraperitoneal injections of IgE and anti IgE showed that NDST2-/- mast

cells were still able to initiate an inflammatory response as measured by

neutrofil influx. Upon IgE and anti IgE challenge in vitro, the mast cells

34

from NDST2-/- animals released similar, or even higher, levels of histamine

per cell compared to wild-type cells. The conclusions from this study are:

NDST2 is indispensable for heparin biosynthesis

NDST2 is redundant in HS N-sulfation biosynthesis, at least in

kidney and liver

lack of NDST2 does not affect embryonic or postnatal

development, with the exception of connective tissue type mast

cells

heparin is needed to maintain connective tissue type mast cell

integrity

heparin depleted mast cells can still evoke an inflammatory

response

Paper II — NDST1 deficient mice die neonatally and produce undersulfated HS

In this study, we showed that NDST1 is important for HS biosynthesis and

that HS is important for normal development of the mouse embryo. HS can

be found in virtually every location in the body where its suggested

functions are many. Since the relative contribution of NDST1 to the overall

structure of HS was unknown, the expected phenotype of an NDST1

deficient mouse was hard to anticipate.

All NDST1-/- mice that are delivered at full term die shortly after birth,

probably due to respiratory failure. Genotyping of litters from heterozygous

crossings at different embryonic stages revealed that about one third of the

NDST1-/- embryos die somewhere between E14.5 and E18.5.

Newborn NDST1-/- pups make breathing efforts but turn into a cyanotic

bluish color as respiration fails. Histological sections of lungs from newborn

NDST1-/- pups appeared atelectatic and hyperplastic while lungs from wild-

type pups were air filled and had a thin-walled delicate structure.

Immunostainings of lung sections from E18.5 wild-type embryos were

performed and showed that both surfactant protein-A (SP-A) and -B were

produced and secreted into the airways in large amounts. Stainings of

NDST1-/- E18.5 embryos showed production of both SP-A and -B but no or

very little secretion. The number of surfactant producing type II

pneumocytes also seemed to be elevated. The mechanism for the

disturbances in SP secretion in NDST1-/- mice is not known but delayed

maturation of type II pneumocytes might be one reason.

Other not fully penetrant phenotypes concerning head and skeletal

development were also observed

Disturbances in the HS structure in BMs were visualized by

immunostainings of different tissues using antibodies that require N-sulfated

GlcN residues in HS for epitope recognition (Jenniskens et al., 2000;

35

Yoshida, personal communication). Ion exchange chromatography showed

that the overall charge density of HS purified from NDST1-/- fibroblast

cultures and different organs (lung and liver) was reduced compared to HS

from wild-type samples. The degree of N-sulfation, measured after gel

filtration of HS degradation products obtained by nitrous acid treatment at

pH1.5, was lowered from >40% in wild-type to <15% in homozygous

fibroblast cultures.

The conclusions from this study are:

NDST1 is important for correct HS sulfation

lack of NDST1 results in a systemic alteration in HS sulfation

pattern

NDST1 contributes quite uniformly to N-sulfation of HS produced

by different tissues/cells, at least in lung, liver and cultured

fibroblasts

N-sulfation occurs also in the absence of NDST1. Whether this N-

sulfation is the result of residual or compensatory activities of the

other NDST isoforms is not known

lack of NDST1 affects survival during late gestation and

postnatally

low N-sulfation of HS disturbs lung maturation but also other

developmental events such as head and skeletal formation

HS is important in mammalian development

Paper III — the liver develops normally despite low HS-sulfation levels

In this study we concluded that not only the N-sulfation levels are disturbed

by lack of NDST1 but also other modifications of the HS chain. The altered

HS structure does not seem to affect liver development essentially. Lack of

NDST2 does not affect liver HS biosynthesis either at E18.5 or in adult

mice.

In paper II we concluded that the HS structure is altered in most BMs.

This does not affect the ability of cells in NDST1-/- liver to lay down

extracellular matrices in a spatially unaltered manner as visualized by

immunostainings for the HSPG perlecan and the BM marker nidogen.

Further, the vascularisation seemed to be unaffected by lack of NDST1.

Immunostainings with antibodies against two endothelial markers, PECAM-

1 and von Willebrand factor, both showed the same staining pattern in

NDST1-/- and wild-type E18.5 embryos. Stainings for von Willebrand factor

also showed that the number of megakaryocytes were unaffected by lack of

NDST1. The same results were seen in NDST2-/- E18.5 liver.

The N-sulfation levels were lowered from 45% in wild-type to 17% in

NDST1-/- E18.5 liver. Disaccharide analysis revealed a dramatically lowered

36

2-O-sulfation level as well as a slight reduction in 6-O-sulfation. The

reduction of 6-O-sulfation was brought about by a reduction of 6-O-sulfated

IdoA containing disaccharides whereas the 6-O-sulfation of GlcA containing

disaccharides instead was increased. The overall total sulfation level was

lowered by almost 50% in NDST1-/- embryos. As expected, no alterations in

disaccharide composition could be detected in NDST2-/- liver HS.

We further wanted to know whether lack of NDST1 or NDST2

respectively induces any compensatory up-regulation of other HS modifying

enzymes. Northern blot analysis revealed that NDST1 and NDST2 are the

only NDST isoforms expressed in the liver both at E18.5 and in adult mice.

No compensatory up-regulation of either NDST or any other modifying

enzyme could be detected, since transcript levels were the same in both

mutants and wild-type. NDSTs have been proposed to be translationally

regulated, however, even though NDST2 apparently is translated in adult

wild-type liver, the impact on the final HS structure must be considered to be

minor.

The conclusions from this study are:

NDST1 is the major NDST isoform in liver HS biosynthesis both at

E18.5 and in adult life

NDST1 and NDST2 are the only NDST-isoforms normally

expressed in E18.5 and adult liver

there is no compensatory up regulation of other NDST isoforms or

other biosynthetic enzymes in the absence of NDST1 or NDST2,

nevertheless NDST2 can partially compensate for lack of NDST1

correct HS structure is not essential in liver development

correct HS structure is not needed for deposition of extracellular

matrix molecules at spatially correct locations

correct HS is not needed for the spatial distribution of endothelial

vessels

Paper IV — skeletal development is dependent on HS structure

In this study, we concluded that defective HS biosynthesis, by the lack of

NDST1, affects skeletal development in several ways. The many skeletal

phenotypes seen in NDST1-/- mice were described and the cause of different

defects discussed at a molecular level. Comparisons to other mouse strains

with skeletal defects indicate that multiple signaling pathways important in

skeletal development are affected by altered HS structure.

All NDST1-/- mice have skeletal defects. Skeletal preparations of wild-

type, NDST1+/- and NDST1-/- E18.5 embryos were performed and compared.

Some phenotypes were fully penetrant: less ossification of the whole or parts

of the skeleton, wide fontanels, erroneous curvature and/or angulation of

37

clavicles and failure to complete fusion of the sternum. Other not fully

penetrant phenotypes such as fusion of cervical vertebrae, misshaped ribcage

and asymmetric rib pairing were also observed. Some very rare, but

interesting defects were also seen. These defects include: one case of

syndactyly, two cases of rib fusion, two cases of severely underdeveloped

mandibula and one case of lack of several skull bones. Bones of both

endochondral and intramembranous origin are affected by altered HS

structure and both morphology of the skeletal parts as well as the ossification

process are disturbed in NDST1-/- mice.

The conclusions from this study are:

HS is involved in skeletal development

lack of NDST1 affects the development of bones of both

endochondral and intramembranous origin

the phenotypes seen in NDST1-/- E18.5 embryos are probably

caused by disturbances in several signaling systems; signaling via

the FGF-, BMP-, Hh- and Wnt-signaling systems are good

candidates

38

Discussion

NDST2 — crucial in mast cells but redundant in other cells?

By studies of NDST2 deficient mice we have been able to stipulate the

necessity of heparin for maintenance of storage-granules in connective tissue

type mast cells (Forsberg et al., 1999 (paperI)). The high negative charge of

heparin is supposedly needed to balance the high positive charge contributed

by the mast cell proteases and other inflammatory meditors. This theory has

been further supported by the fact that mice lacking histamine, have reduced

amounts of proteoglycans stored in the mast cell granules and this in turn

cause impaired storage of proteases (Kolset et al., 2004). Positively charged

mast cell mediators and negatively charged GAGs therefore seem to be

mutually dependent on each other for charge neutralization in the tightly

packed storage granules.

The correlation of NDST2 protein expression and its role in HS

biosynthesis remains enigmatic (paper III; Ledin et al., unpublished). Does

the lack of effect on HS production by the liver of NDST2 deficient mice

mean that this isoform is normally redundant in HS biosynthesis? One

possibility is that other isoforms in the absence of NDST2 fully compensate

for the lack of this enzyme. Since no transcriptional up-regulation of other

NDST isoforms could be observed in NDST2 deficient liver, the

compensatory mechanism must lie in elevated translation, by enhanced

amounts of active enzyme or an up-regulation in activity of remaining

NDSTs. In the liver, the compensatory enzyme should be NDST1 since

NDST3 and -4 are not expressed. These arguments require translational

regulation and/or post-translational modifications of the compensating

NDSTs, and possibly also a well-regulated access to activating co-factors.

However, NDST2 may be important for HS biosynthesis in some cell

populations, not yet identified. The amount of possibly defective HS

produced by these cells, might be too small to be detected during HS

characterization. Further, NDST2 may be important in HS biosynthesis in

more extreme situations not yet studied, such as wound healing and

pathogen defense.

NDST2 deficient mice have been used by other groups, as a model system

for neurogenic inflammation processes (Karlsen et al., 2004) and also for

39

tumor growth and peri-tumoral blood clotting studies (Samoszuk et al.,

2003; Samoszuk and Corwin, 2003). Cultured mast cells from these mice

have also been used to study the role of heparin on mast cell protease

processing and activation, and how this may influence processes such as

extravascular coagulation and fibronectin turnover (Henningsson et al.,

2002; Tchougounova et al., 2001; Tchougounova and Pejler, 2001). These

mice should further be an interesting model for studies of other

immunological issues such as IgE mediated allergic reactions and parasite

defense.

NDST1 and embryonic development

In light of the many possible tasks for HS in a developing embryo, it is

remarkable how well many NDST1 deficient individuals develop in utero

(Ringvall et al., 2000 (paper II)). Most individuals survive till birth and are

sometimes, without close examination or genotyping, impossible to separate

from their wild-type littermates prenatally. Why do such extensive

alterations in HS disaccharide composition not affect the development, in

relative terms, more severely than observed in NDST1 deficient individuals?

And why is the development of lung, head and skeleton more severely

affected by disturbed HS production than liver?

It is important how the remaining sulfates in HS from NDST1 deficient

embryos are arranged along the HS chain and whether there still are distinct

NS domains. A study performed by Bame and co-workers suggests that HS

with low sulfation, produced by a cell line with no or little NDST1, still

appears to be arranged in a wild-type like pattern. The NS domains were

fewer and all located at one end of the HS chains. This left some parts of the

chains wild-type like as the NS domains were normally spaced, while the

remaining part of the chains were devoid of sulfation (Bame et al., 2000).

Structural analysis of HS from NDST1 deficient embryos suggests the

remaining sulfates to be located to one region along the chain, leaving the

rest of it mainly unmodified (Ledin, J., personal communication). Thus,

enough wild-type like sequences may be left in this HS to facilitate a fairly

normal development of many tissues.

Production of HS with a specific structure may be more strictly regulated

in some organs, and different organs might vary in sensitivity to HS

alterations as they rely on alternative sets of growth factors and morphogens

during development. Depending on how reliant a signaling system is on

specific HS structures for correct function, an alteration in HS biosynthesis

can affect different organs to various degrees.

The severity and penetrance of some NDST1-/- phenotypes depend on

genetic background; a phenomenon often seen in genetically modified

mouse strains. With a mixed genetic background (C57Bl6/129SVJ), about

one third of the NDST1 deficient embryos die before birth (Ringvall et al.,

40

2000 (paper II)). A pure C57Bl6 background dramatically decreases the

frequency of prenatal death whereas eye and other head defects are more

frequently observed (paper IV; unpublished observations). Strains with

different genetic background may have small regulatory differences in some

systems. Thereby, the appearance and severity of a defect might be

dependent on the exact regulation of HS-biosynthesis and e.g. expression of

a specific growth factor at a certain time point.

The lung defect of NDST1 deficient mice is supposedly caused by a delay

in lung maturation (Ringvall et al., 2000 (paper II)). Lack of secreted

surfactant and the elevated number of surfactant producing cells can both be

signs of immature lungs. If this is the case, NDST1 deficient mice might be

rescued if the pregnant female is treated with corticosteroids to speed up

lung maturation. There are however reasons to speculate that NDST1

deficiency may confer difficulties of lung epithelial cells to develop into type

I pneumocytes. Differences in sulfation status of the BM underlying the lung

epithelial cells might be conclusive for the development of type I and type II

pneumocytes, since the HS in the BM underlying type I pneumocytes is

more highly sulfated than that of BM underlying type II pneumocytes

(Sannes, 1984; Van Kuppevelt et al., 1984). The two different kinds of

pneumocytes might arise from a common progenitor cell and type II

pneumocytes can differentiate into type I both in vitro and in vivo (Uhal,

1997). Thus, the elevated number of type II pneumocytes observed in

NDST1-/- lungs might be a consequence of difficulties to transform into type

I pneumocytes due to undersulfated BMs. Further, the elevated number of

type II pneumocytes can depend on enhanced proliferation and/or reduced

apoptosis of type II pneumocytes.

Delayed maturation may not be the cause of failed surfactant secretion.

Regulation of surfactant secretion might at some point be dependent on HS.

As an example, the exocytosis mechanism of lamellar bodies is dependent of

cytoskeletal components (Fehrenbach, 2001) and rearrangements of the

cytoskeleton can be influenced by syndecan signaling (Yoneda and

Couchman, 2003). Elevated cytoplasmic Ca2+ levels is known to trigger

surfactant secretion at least in vitro (Dietl et al., 2001). Disturbances in Ca2+

signaling might therefore cause surfactant secretion to fail, and since it has

been shown that cultured myotubes from NDST1 deficient mice have

disturbed Ca2+ kinetics (Jenniskens et al., 2003), disturbed Ca2+ signaling

may be found also in other systems such as type II pneumocytes in the lung.

Since NDST1 deficient mice might be devoid of type I pneumocytes, it is

interesting to note, that it has been suggested that this cell type can regulate

surfactant secretion. Type I pneumocytes supposedly receive external stimuli

and thereafter transduce a Ca2+ signal to type II pneumocytes for surfactant

release (Dietl et al., 2001).

Several HS interacting signaling molecules such as, FGF9, FGF10, Sonic

Hh, Wnt7b and BMP4, are known to be important in development of the

41

lung (Colvin et al., 2001; Shu et al., 2002; Warburton et al., 2000). In view

of this large number of potentially affected signaling pathways, NDST1

deficient mice exhibit surprisingly little disturbances in lung development; as

an example, no defects in branching morphogenesis have yet been observed

in these mice.

The liver of NDST1 deficient embryos appears to develop normally

despite production of undersulfated HS (paper III). No apparent changes in

the vasculature have been observed although several angiogenic factors such

as VEGF-A, FGF1, FGF2 and endostatin interact with HS (Zatterstrom et

al., 2000). Therefore, the HS status might be less important in liver

development and maturation than in other affected organs. However, we do

not yet know how the HS production is affected by lack of NDST1 at other

developmental stages than at E18.5. There is a possibility that other NDST

isoforms are able to compensate for the lack of NDST1 earlier during liver

development and HS sulfation may be of more importance for correct

development at these time points. Subtle changes in extracellular matrix

structure, microvasculature, some cell populations as well as functional

disturbances cannot be excluded.

Development of the skeleton also involves many signaling systems where

HS can be of great importance. Many of the observed skeletal phenotypes in

NDST1 deficient mice (paper IV) are indeed strikingly similar to defects

seen in mice that lack specific FGFs and BMPs, such as FGF18 (Liu et al.,

2002; Ohbayashi et al., 2002), BMP4 and BMP7 (Katagiri et al., 1998), or

have altered FGFR signaling (Ornitz and Marie, 2002). It is interesting to

note that several of the skeletal defects observed in NDST1 deficient mice

also have been noted in children born by diabetic mothers (Ewart-Toland et

al., 2000), and as mentioned above, correlations between HS biosynthesis

and diabetes have been seen.

The head phenotypes observed in NDST1 deficient mice, not covered by

the material presented in this thesis, are also very interesting. As it seems,

these defects typically arise from erroneous first branchial arch development

and some of them may be connected to alterations in neural crest cell-

migration.

42

Future perspectives

In a wider perspective, the increased understanding of HS biosynthesis

regulation and function in our body, will improve the chances to find cures

to pathological states where HS is involved. By the introduction of

genetically modified mouse strains in the HS research area, our possibilities

to explore HS in vivo functions have multiplied. Further studies of the

NDST1 and -2 deficient mouse strains presented in this thesis will contribute

with important knowledge. Many questions are waiting for an answer. Some

of these will be addressed in the near future:

Is the lack of secreted surfactant in the lungs of NDST1 deficient mouse

pups due to immature lungs? If so, will it be possible to accelerate lung

maturation by corticosteroid treatment of pregnant mice and thereby

rescue newborn NDST1 deficient pups? Will it be possible to rescue them

by oral administration of surfactant at birth?

Is the ultrastructure of different organs altered in NDST1 deficient

embryos? Are cells organized in a correct way and is the structure of

BMs, other ECMs and the microvasculature disturbed?

How dynamic are the fluctuations in HS structure in NDST1 deficient

embryos? Will FGF-FGFR in situ probing for different HS structures

reveal the same pattern in NDST1 deficient as in wild-type embryos? Are

there local variations in HS structure on different cell types and in

different ECM areas?

What are the molecular mechanisms causing the different phenotypes?

What signaling pathways are involved?

Is embryo patterning disturbed in the early NDST1 deficient embryos?

Establishment of NDST1 deficient ES cells will be highly valuable for future

studies of HS structure, molecular mechanisms and for in vitro

differentiation assays into various cell-types. The very early embryonic

lethality of embryos lacking both NDST1 and -2 (Holmborn et al.,

unpublished) indicates that NDST2 may have a role in HS biosynthesis

during development. In vitro differentiation of NDST2-/- ES cells may be a

43

useful method to investigate if this is the case. Studies of the HS produced

by ES cells that are double deficient for NDST1 and -2 have this far resulted

in a very interesting finding; these cells lack N-sulfation but are anyway 6-

O-sulfated (Holmborn et al, unpublished). This suggests HS biosynthesis to

be less hierarchic than previously considered.

44

Concluding remarks

In vivo veritas

From the phenotypes observed in mouse strains that lack different HS

biosynthetic enzymes we can make several conclusions. Lack of HS is

incompatible with embryonic development with the exception of the very

early stages, as we learned from EXT1-/- embryos (Lin et al., 2000). From

studies of mice with defective HS modification machinery we can see that

the development of some organs appears to be more sensitive to alterations

in HS structure than others. Depending on the nature of the HS structural

alterations, the development of different organs can respond in various ways.

Alterations of the HS structure by lack of any of NDST1, C5epi or 2-

OST, cause several developmental disturbances but is generally not lethal

until the perinatal period (Li et al., 2003; Merry and Wilson, 2002; Ringvall

et al., 2000); paper III; Ledin et al., unpublished). All three mouse strains

exhibit defects in skeletal and eye development even though they produce

HS with different structural alterations. The phenotypes and alterations in

HS structure seen in 2-OST-/- and C5epi-/- embryos suggest 2-O-sulfated

iduronic acid to be highly important in kidney development. A decrease in

total sulfation levels, mainly depending on less N-sulfation, as seen in

NDST1-/- embryos, cause altered lung development and also severe head

deformations. Both the NDST2-/- and 3-OST1-/- mouse strains are less

severely affected than the three other mouse strains. HS biosynthesis is, as

we have shown, not affected by lack of NDST2 (Forsberg et al., 1999 (paper

I); paper III; Ledin et al., unpublished), and lack of 3-OST1, being the last

modification of the HS chain, does not affect the HS structure in any

dramatic way (HajMohammadi et al., 2003).

In conclusion, HS is indeed highly important in mammalian development

but the exact modification pattern may have less impact than anticipated on

several signaling systems. Theory and reality are far from always in

accordance, and not least the results from studies of HS and phenotypes of

NDST2 and 3-OST1 deficient mice, show that it is always important to

distinguish between in vitro and in vivo results.

45

Populärvetenskaplig sammanfattning

Proteiner — genprodukter med många funktioner

I det tidiga embryot, under den första tiden efter att ett ägg befruktats av en

spermie, är alla celler lika. Efter ett tag kommer olika celler specialisera sig

för att kunna utföra olika uppgifter, för att ännu lite senare kunna bilda de

olika vävnader vår kropp är uppbyggd av. I var och en av dessa celler finns

vår arvsmassa, våra gener. En gen kodar för, dvs. fungerar som en mall för

produktionen av ett eller flera proteiner. Genens kod översätts till ett protein

via ett så kallat mRNA.

Proteiner har många olika funktioner inuti, på ytan av och mellan celler.

Olika typer av celler har olika gener påslagna och producerar därför olika

proteiner i bestämda mängder. En febril signaleringsverksamhet i och mellan

cellerna talar om för dem vilka olika gener som ska slås på och stängas av.

Denna signaleringsverksamhet måste vara mycket exakt reglerad för att

embryot ska kunna utvecklas på rätt sätt. Många av de molekyler som deltar

i regleringen av, och i själva signaleringen, är proteiner. Ett protein kan

skickas som signal från en cell till en annan för att den mottagande cellen

t.ex. ska dela sig eller förflytta sig. Den mottagande cellen tar emot

signalproteinet genom att fånga upp det med ett annat protein som sitter fast

på cellytan, en så kallad receptor. Ett signalprotein kan ibland färdas långa

sträckor och kan då passera genom den extracellulära matrisen. Denna matris

är ett stabilt nätverk av olika proteinkomponenter som sitter mellan cellerna

och som bland annat ger stöd och struktur till en vävnad.

Heparansulfat-proteoglykaner — proteiner med sulfaterade

sockerkedjor

Ofta sitter en eller flera sockerkedjor fast på proteinet; dessa sockerkedjor

är viktiga för proteinets funktion. Det finns många olika sorters

sockerkedjor, en av de vanligaste är heparansulfat (HS). Proteiner med HS

fästade till sig kallas HS-proteoglykaner. HS-proteoglykaner finns på

cellytan och mellan cellerna i den extracellulära matrisen. Varje enhet som

bygger upp HS-kedjan kan bära negativt laddade sulfatgrupper i olika

positioner. För att en cell ska kunna producera HS-kedjorna krävs ett

komplicerat maskineri bestående av många olika enzymer. Varje enzym har

sin speciella uppgift i produktionen av HS, t.ex. att placera en sulfatgrupp i

46

en viss position. Placeringen och mängden av sulfatgrupper gör att HS-

strukturen kan se olika ut beroende på vilken sorts cell och vid vilken

tidpunkt en cell producerar sockret. En extrem form av HS har sulfatgrupper

i nära nog alla de positioner som är möjliga, denna specialform av HS kallas

heparin. Heparin är mest känt för sin förmåga att hindra blodkoagulation och

har länge använts kliniskt för att hindra uppkomsten av blodproppar. Heparin

finns dock inte normalt i våra blodkärl.

Man har länge studerat heparin och HS biokemiskt och funnit att dessa

sockerarter har stark bindningsförmåga till många proteiner som är viktiga

för t.ex. celldelning. Man har även sett i cellodlingsförsök att HS och HS-

proteoglykaner verkar vara viktiga för många cellulära processer. Troligtvis

beror mycket av bindningsförmågan på att sockerkedjorna tack vare sina

många sulfatgrupper har en hög negativ laddning och därför attraherar

positivt laddade molekyler. Det har dock visat sig att vissa bindningar är

beroende av att sulfatgrupperna sitter placerade på ett specifikt sätt.

Därigenom bildas speciella mönster med hög bindningsförmåga till ett visst

protein.

Lärdom från genetiskt modifierade möss

Genom att studera genetiskt modifierade möss kan man få mycket kunskap

om olika gener och deras proteinprodukters funktion. Genetiskt modifierade

möss kan även fungera som modeller för olika hos människan

förekommande sjukdomstillstånd. Denna avhandling har gjorts med syftet att

studera hur defekt HS/heparin-produktion påverkar en däggdjurskropp och

hur olika enzymer påverkar HS/heparin-strukturen.

Vi har hos möss specifikt stängt av funktionen hos två gener som kodar

för enzymerna NDST1 och NDST2. Dessa geners proteinprodukter är

viktiga för sulfateringen av HS/heparin kedjorna. Vi har därigenom tagit

reda på hur NDST1 respektive NDST2 bidrar till HS/heparin kedjornas

struktur och visat hur dessa förändringar påverkar organismen. NDST1 och

NDST2 är isoformer av samma typ av enzym. De kan utföra samma kemiska

reaktion men kan regleras var för sig, och utföra sina uppgifter med olika

effektivitet och i olika celler, beroende på om respektive gen är på- eller

avslagen.

Heparin—ett måste för mastceller

Vi har funnit att NDST1 påverkar sulfateringen av HS i hela kroppen och att

NDST2 är helt avgörande för heparinproduktionen. Heparin produceras

normalt endast i en sorts celler i hela kroppen, de så kallade bindvävs-

mastcellerna. Bindvävsmastceller är i normala fall fullpackade med heparin

och proteiner som är viktiga i vårt immunförsvar. I det första arbetet (paper

I) visar vi att möss som saknar NDST2 inte kan producera heparin, och att

deras bindvävsmastceller förlorar sin normala form och sitt proteininnehåll

47

samt är färre i antal. Det verkar alltså vara så att heparinets höga negativa

laddning behövs för att kunna packa cellerna fulla med alla de positivt

laddade proteinerna. Trots denna allvarliga defekt kan de kvarvarande

bindvävsmastcellerna fortfarande svara på vissa immunologiska

retningssignaler på ett relativt normalt sätt. Trots att genen för NDST2

normalt även är påslagen i många av kroppens HS-producerande celler

verkar avsaknad av NDST2 inte påverka HS-strukturen alls (paper I och III).

Förutom mastcell-defekten är möss som saknar NDST2 i övrigt friska och

fertila. Avsaknad av NDST2 verkar inte heller ha någon effekt under

embryoutvecklingen.

HS — viktigt för normal embryoutveckling

I arbete II och III (paper II och III) visar vi att avsaknad av NDST1 leder till

produktion av HS med ca hälften av den normala mängden sulfatgrupper.

Förändringar i HS-kedjornas sulfatmönster kan ses i hela kroppen hos

musungar som saknar NDST1. Detta leder i sin tur till en rad olika fel under

embryoutvecklingen. Några av de individer som saknar NDST1 dör redan

under embryostadiet medan resterande musungar dör strax efter födseln pga.

andningssvårigheter. För att kunna andas vid födseln måste ett ämne som

kallas surfaktant finnas i luftvägarna. Barn som föds mycket för tidigt saknar

surfaktant eftersom deras lungor är omogna. Musungar som saknar NDST1

lider av ett liknande problem. Vi har i vävnadsprover från lunga kunnat se att

surfaktant produceras i rikliga mängder även hos musungar med avsaknad av

NDST1, men att ämnet inte utsöndras i luftvägarna. Vi har dragit slutsatsen

att HS med ett korrekt sulfatinnehåll och mönster är nödvändigt för att

lungorna ska mogna på ett riktigt sätt. Kanske krävs även en korrekt HS-

produktion för att surfaktantproducerande celler ska kunna utsöndra ämnet i

luftvägarna.

Olika vävnader och organ verkar dock vara olika känsliga för de enorma

förändringar av HS-strukturen som avsaknad av NDST1 medför. Levern har

exempelvis inga uppenbara defekter trots den förändrade HS-strukturen

(paper III). Vi har även undersökt om leverns celler försöker kompensera

förlusten av NDST1 genom att öka produktionen av andra NDST-isoformer

samt övriga HS-sulfaterande enzymer. Så verkar inte vara fallet eftersom

mRNA nivåerna av alla de andra enzymerna är de samma i normal lever som

i lever som saknar NDST1 eller NDST2.

I det sista arbetet (paper IV) har vi visat att musungar som saknar NDST1

även har fel på skelettutvecklingen. Vi har även kunnat konstatera att

avsaknad av NDST1 ofta leder till felaktig utveckling av hjärnan och ögonen

samt att ansiktets mittlinje ibland inte sluter sig ordentligt. De flesta av

skelettets delar drabbas i varierande grad av felaktig HS produktion.

Generellt sett har skelettet hos möss som saknar NDST1 ett något

kompaktare och grövre utseende än hos normala individer. Deras skelett

består av mer brosk och mindre mineraliserat ben än normalt vilket kan tyda

48

på en försening av skelettutvecklingen. Många bendelar har felaktig form

och vissa delar kan saknas helt. Troligtvis beror den avvikande

skelettutvecklingen på fel i distribueringen av ett flertal signalproteiner samt

fel i signaleringen via ett flertal receptorer.

Genom studier av dessa två musstammar har vi alltså kunnat visa att den

fysiologiska funktionen för heparin främst är att upprätthålla

bindvävsmastcellernas förråd av proteiner viktiga för immunförsvaret.

NDST2 är i första hand viktigt för produktionen av heparin och verkar ej

påverka HS-produktionen. Vidare har vi slagit fast att NDST1 är mycket

viktigt för sulfateringen av HS och att HS är viktigt för normal

embryoutveckling.

49

Acknowledgement

The work presented in this thesis was performed at the former Department of

Animal Physiology (later Department of Cell and Molecular Biology) and at

the Department of Medical Biochemistry and Microbiology.

I would like to express my warmest gratitude to all the people involved in

the creation of this thesis and all people who have not contributed so very

much to the thesis but made life happy! Many thanks to:

Lena Kjellén, my supervisor. For scientific guidance and for making the

atmosphere so friendly and warm in our corridor! Thank you for letting

me keep on with my not so biochemical studies of HS and for

supporting my ideas. And off cause, for taking care of us when Erik

left!

Erik Forsberg, my former supervisor, today my co-supervisor. Hundreds of

thanks for giving me a fantastic start! I’m astonished by the “mouse-

centrifuge-trick”…and all other hilarious moments!

Ulf Lindahl, my other co-supervisor, for always being willing to discuss

science.

Peter Ekblom, for giving me the oportunity to start my PhD studies at

zoofys. Thank you for choosing me!

All persons in Lenas group: Anders, Håkan, Inger, Jenny, Johan, Katta, Lena, Mehdi and Pernilla. You’re great! I’m privileged to work in

such a nice group. Some extra thanks to:

Inger, for being a great technician and a very nice person!

Johan, for all great discussions, for being quick-witted, funny,

sarcastic and annoying!

Katta, the best room mate one can get! For listening to all my

babbling, for all the songs we’ve been singing, for being as

hypocondric as I, for interesting discussions about finger-bending,

tongue-wrinkling and other physical peculiarities...and now and

then some scientific discussions, and for being a great friend!

All other persons in the corridor: Anna, Anne, Anne-Mari, Birgitta, Fredrik, Jimmy, Josefine, Karin, Maria W, Maud, Mia, Pia, Qun, Stina, Sofie, Staffan, Teet, Ulla, Ulrika, Åsa H, Åsa M. Another

bunch of truly great people! I enjoy being at work; you definitely

contribute to that. Some extra thanks to:

50

Anna, for being a great friend. For always being full of ideas and

energy! For all great Stora Kalholmen visits and all other funny

excursions. For going miles (and standing still...) with me and

“Gamla Bettan” (the most unreliable of us!).

Anne-Mari, for being a great technician and such a helpful person.

Mia, for companionship during the final spurt.

All persons at IMBIM dealing with administrative issues, technical problems

etc.

The old “Zoofysgänget”, not least for all great parties! Extra thanks to:

Maria F, for being a zoofys-rookie at the same time as I, and for

becoming a very good friend!

All other former and present collaborators.

All friends (not already mentioned) outside BMC, especially:

Helena Alm and the rest of “Cirkus Alm” for all fun and for always

being there when most needed!

“Stora Kalholmen-gänget”, for all lazy days filled with hot-dogs,

wine, song and laughs, and for being great friends also when we

are not in this paradise!

Stig and Britt, for being the sweetest parents-in-law one can get!

Ulla and Hans, my parents, for always loving and supporting me. No matter

what I’ve been up to, you’ve always been there for me!

My most loved ones:

Anders for giving me endless tender, love and care. For being patient

and for being my life companion!

Oskar, for your unreserved love and happiness, for distracting my

mind after a hard days work and for making my life wonderful!

51

References

Ahn, J., Ludecke, H. J., Lindow, S., Horton, W. A., Lee, B., Wagner, M. J., Horsthemke, B. and Wells, D. E. (1995). Cloning of the putative tumour suppressor gene for hereditary multiple exostoses (EXT1). Nat Genet 11, 137-43.

Ai, X., Do, A. T., Lozynska, O., Kusche-Gullberg, M., Lindahl, U. and Emerson, C. P., Jr. (2003). QSulf1 remodels the 6-O sulfation states of cell surface heparan sulfate proteoglycans to promote Wnt signaling. J Cell Biol162, 341-51.

Aikawa, J. and Esko, J. D. (1999). Molecular cloning and expression of a third member of the heparan sulfate/heparin GlcNAc N-deacetylase/ N-sulfotransferase family. J Biol Chem 274, 2690-5.

Aikawa, J., Grobe, K., Tsujimoto, M. and Esko, J. D. (2001). Multiple isozymes of heparan sulfate/heparin GlcNAc N-deacetylase/GlcN N-sulfotransferase. Structure and activity of the fourth member, NDST4. J Biol Chem 276, 5876-82.

Ailles, L., Kisilevsky, R. and Young, I. D. (1993). Induction of perlecan gene expression precedes amyloid formation during experimental murine AA amyloidogenesis. Lab Invest 69, 443-8.

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

Allen, B. L., Filla, M. S. and Rapraeger, A. C. (2001). Role of heparan sulfate as a tissue-specific regulator of FGF-4 and FGF receptor recognition. J Cell Biol155, 845-58.

Allen, B. L. and Rapraeger, A. C. (2003). Spatial and temporal expression of heparan sulfate in mouse development regulates FGF and FGF receptor assembly. J Cell Biol 163, 637-48.

Arikawa-Hirasawa, E., Le, A. H., Nishino, I., Nonaka, I., Ho, N. C., Francomano, C. A., Govindraj, P., Hassell, J. R., Devaney, J. M., Spranger, J. et al. (2002). Structural and functional mutations of the perlecan gene cause Schwartz-Jampel syndrome, with myotonic myopathy and chondrodysplasia. Am J Hum Genet 70, 1368-75.

Arikawa-Hirasawa, E., Watanabe, H., Takami, H., Hassell, J. R. and Yamada, Y. (1999). Perlecan is essential for cartilage and cephalic development. Nat Genet 23, 354-8.

Arikawa-Hirasawa, E., Wilcox, W. R. and Yamada, Y. (2001). Dyssegmental dysplasia, Silverman-Handmaker type: unexpected role of perlecan in cartilage development. Am J Med Genet 106, 254-7.

Aruffo, A., Stamenkovic, I., Melnick, M., Underhill, C. B. and Seed, B. (1990). CD44 is the principal cell surface receptor for hyaluronate. Cell 61, 1303-13.

Bame, K. J., Venkatesan, I., Stelling, H. D. and Tumova, S. (2000). The spacing of S-domains on HS glycosaminoglycans determines whether the chain is a substrate for intracellular heparanases. Glycobiology 10, 715-26.

52

Becker, S., Wang, Z. J., Massey, H., Arauz, A., Labosky, P., Hammerschmidt, M., St-Jacques, B., Bumcrot, D., McMahon, A. and Grabel, L. (1997). A role for Indian hedgehog in extraembryonic endoderm differentiation in F9 cells and the early mouse embryo. Dev Biol 187, 298-310.

Beil, W. J., Schulz, M. and Wefelmeyer, U. (2000). Mast cell granule composition and tissue location--a close correlation. Histol Histopathol 15, 937-46.

Bellaiche, Y., The, I. and Perrimon, N. (1998). Tout-velu is a Drosophila homologue of the putative tumour suppressor EXT-1 and is needed for Hh diffusion. Nature 394, 85-8.

Bengtsson, J., Eriksson, I. and Kjellén, L. (2003). Distinct effects on heparan sulfate structure by different active site mutations in NDST-1. Biochemistry 42,2110-5.

Bernfield, M., Gotte, M., Park, P. W., Reizes, O., Fitzgerald, M. L., Lincecum, J. and Zako, M. (1999). Functions of cell surface heparan sulfate proteoglycans. Annu Rev Biochem 68, 729-77.

Bernfield, M., Kokenyesi, R., Kato, M., Hinkes, M. T., Spring, J., Gallo, R. L. and Lose, E. J. (1992). Biology of the syndecans: a family of transmembrane heparan sulfate proteoglycans. Annu Rev Cell Biol 8, 365-93.

Bezakova, G. and Ruegg, M. A. (2003). New insights into the roles of agrin. Nat Rev Mol Cell Biol 4, 295-308.

Binari, R. C., Staveley, B. E., Johnson, W. A., Godavarti, R., Sasisekharan, R. and Manoukian, A. S. (1997). Genetic evidence that heparin-like glycosaminoglycans are involved in wingless signaling. Development 124,2623-32.

Bloom, W. and Fawcett, D. W. (1975). A Textbook of Histology: W. B. Saunders Company.

Bornemann, D. J., Duncan, J. E., Staatz, W., Selleck, S. and Warrior, R. (2004). Abrogation of heparan sulfate synthesis in Drosophila disrupts the Wingless, Hedgehog and Decapentaplegic signaling pathways. Development.

Bowe, M. A. and Fallon, J. R. (1995). The role of agrin in synapse formation. Annu Rev Neurosci 18, 443-62.

Brandan, E. and Hirschberg, C. B. (1988). Purification of rat liver N-heparan-sulfate sulfotransferase. J Biol Chem 263, 2417-22.

Brickman, Y. G., Ford, M. D., Gallagher, J. T., Nurcombe, V., Bartlett, P. F. and Turnbull, J. E. (1998a). Structural modification of fibroblast growth factor-binding heparan sulfate at a determinative stage of neural development. JBiol Chem 273, 4350-9.

Brickman, Y. G., Nurcombe, V., Ford, M. D., Gallagher, J. T., Bartlett, P. F. and Turnbull, J. E. (1998b). Structural comparison of fibroblast growth factor-specific heparan sulfates derived from a growing or differentiating neuroepithelial cell line. Glycobiology 8, 463-71.

Brown, D. A. and Rose, J. K. (1992). Sorting of GPI-anchored proteins to glycolipid-enriched membrane subdomains during transport to the apical cell surface. Cell 68, 533-44.

Bullock, S. L., Fletcher, J. M., Beddington, R. S. and Wilson, V. A. (1998). Renal agenesis in mice homozygous for a gene trap mutation in the gene encoding heparan sulfate 2-sulfotransferase. Genes Dev 12, 1894-906.

Cano-Gauci, D. F., Song, H. H., Yang, H., McKerlie, C., Choo, B., Shi, W., Pullano, R., Piscione, T. D., Grisaru, S., Soon, S. et al. (1999). Glypican-3-deficient mice exhibit developmental overgrowth and some of the abnormalities typical of Simpson-Golabi-Behmel syndrome. J Cell Biol 146, 255-64.

53

Carey, D. J. (1997). Syndecans: multifunctional cell-surface co-receptors. Biochem J 327, 1-16.

Chiao, E., Fisher, P., Crisponi, L., Deiana, M., Dragatsis, I., Schlessinger, D., Pilia, G. and Efstratiadis, A. (2002). Overgrowth of a mouse model of the Simpson-Golabi-Behmel syndrome is independent of IGF signaling. Dev Biol243, 185-206.

Cohen, I. R., Murdoch, A. D., Naso, M. F., Marchetti, D., Berd, D. and Iozzo, R. V. (1994). Abnormal expression of perlecan proteoglycan in metastatic melanomas. Cancer Res 54, 5771-4.

Colvin, J. S., White, A. C., Pratt, S. J. and Ornitz, D. M. (2001). Lung hypoplasia and neonatal death in Fgf9-null mice identify this gene as an essential regulator of lung mesenchyme. Development 128, 2095-106.

Conde-Knape, K. (2001). Heparan sulfate proteoglycans in experimental models of diabetes: a role for perlecan in diabetes complications. Diabetes Metab Res Rev17, 412-21.

Costell, M., Carmona, R., Gustafsson, E., Gonzalez-Iriarte, M., Fassler, R. and Munoz-Chapuli, R. (2002). Hyperplastic conotruncal endocardial cushions and transposition of great arteries in perlecan-null mice. Circ Res 91, 158-64.

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

De Cat, B. and David, G. (2001). Developmental roles of the glypicans. Semin Cell Dev Biol 12, 117-25.

Delehedde, M., Lyon, M., Sergeant, N., Rahmoune, H. and Fernig, D. G. (2001). Proteoglycans: pericellular and cell surface multireceptors that integrate external stimuli in the mammary gland. J Mammary Gland Biol Neoplasia 6, 253-73.

Dhoot, G. K., Gustafsson, M. K., Ai, X., Sun, W., Standiford, D. M. and Emerson, C. P., Jr. (2001). Regulation of Wnt signaling and embryo patterning by an extracellular sulfatase. Science 293, 1663-6.

Dietl, P., Haller, T., Mair, N. and Frick, M. (2001). Mechanisms of surfactant exocytosis in alveolar type II cells in vitro and in vivo. News Physiol Sci 16,239-43.

Duncan, S. A. (2003). Mechanisms controlling early development of the liver. Mech Dev 120, 19-33.

Erickson, A. C. and Couchman, J. R. (2000). Still more complexity in mammalian basement membranes. J Histochem Cytochem 48, 1291-306.

Eriksson, I., Sandback, D., Ek, B., Lindahl, U. and Kjellén, L. (1994). cDNA cloning and sequencing of mouse mastocytoma glucosaminyl N- deacetylase/N-sulfotransferase, an enzyme involved in the biosynthesis of heparin. J Biol Chem 269, 10438-43.

Esko, J. D. and Selleck, S. B. (2002). Order out of chaos: assembly of ligand binding sites in heparan sulfate. Annu Rev Biochem 71, 435-71.

Evans, M. J. and Kaufman, M. H. (1981). Establishment in culture of pluripotential cells from mouse embryos. Nature 292, 154-6.

Ewart-Toland, A., Yankowitz, J., Winder, A., Imagire, R., Cox, V. A., Aylsworth, A. S. and Golabi, M. (2000). Oculoauriculovertebral abnormalities in children of diabetic mothers. Am J Med Genet 90, 303-9.

Fan, G., Xiao, L., Cheng, L., Wang, X., Sun, B. and Hu, G. (2000). Targeted disruption of NDST-1 gene leads to pulmonary hypoplasia and neonatal respiratory distress in mice. FEBS Lett 467, 7-11.

Fehrenbach, H. (2001). Alveolar epithelial type II cell: defender of the alveolus revisited. Respir Res 2, 33-46.

54

Fernig, D. G., Chen, H. L., Rahmoune, H., Descamps, S., Boilly, B. and Hondermarck, H. (2000). Differential regulation of FGF-1 and -2 mitogenic activity is related to their kinetics of binding to heparan sulfate in MDA-MB-231 human breast cancer cells. Biochem Biophys Res Commun 267, 770-6.

Feyzi, E., Saldeen, T., Larsson, E., Lindahl, U. and Salmivirta, M. (1998). Age-dependent modulation of heparan sulfate structure and function. J Biol Chem273, 13395-8.

Flaumenhaft, R., Moscatelli, D., Saksela, O. and Rifkin, D. B. (1989). Role of extracellular matrix in the action of basic fibroblast growth factor: matrix as a source of growth factor for long-term stimulation of plasminogen activator production and DNA synthesis. J Cell Physiol 140, 75-81.

Ford-Perriss, M., Guimond, S. E., Greferath, U., Kita, M., Grobe, K., Habuchi, H., Kimata, K., Esko, J. D., Murphy, M. and Turnbull, J. E. (2002). Variant heparan sulfates synthesized in developing mouse brain differentially regulate FGF signaling. Glycobiology 12, 721-7.

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

Fransson, L. A. (2003). Glypicans. Int J Biochem Cell Biol 35, 125-9. Freeman, M. and Gurdon, J. B. (2002). Regulatory principles of developmental

signaling. Annu Rev Cell Dev Biol 18, 515-39. Fujise, M., Izumi, S., Selleck, S. B. and Nakato, H. (2001). Regulation of dally, an

integral membrane proteoglycan, and its function during adult sensory organ formation of Drosophila. Dev Biol 235, 433-48.

Gautam, M., Noakes, P. G., Moscoso, L., Rupp, F., Scheller, R. H., Merlie, J. P. and Sanes, J. R. (1996). Defective neuromuscular synaptogenesis in agrin-deficient mutant mice. Cell 85, 525-35.

Gilbert, S. F. (1997). Developmental Biology: Sinauer Associates, Inc. Giraldez, A. J., Copley, R. R. and Cohen, S. M. (2002). HSPG modification by

the secreted enzyme Notum shapes the Wingless morphogen gradient. Dev Cell2, 667-76.

Gonzalez-Iriarte, M., Carmona, R., Perez-Pomares, J. M., Macias, D., Costell, M. and Munoz-Chapuli, R. (2003). Development of the coronary arteries in a murine model of transposition of great arteries. J Mol Cell Cardiol 35, 795-802.

Gotte, M., Joussen, A. M., Klein, C., Andre, P., Wagner, D. D., Hinkes, M. T., Kirchhof, B., Adamis, A. P. and Bernfield, M. (2002). Role of syndecan-1 in leukocyte-endothelial interactions in the ocular vasculature. Invest Ophthalmol Vis Sci 43, 1135-41.

Grobe, K. and Esko, J. D. (2002). Regulated translation of heparan sulfate N-acetylglucosamine N- deacetylase/n-sulfotransferase isozymes by structured 5'-untranslated regions and internal ribosome entry sites. J Biol Chem 277, 30699-706.

Groffen, A. J., Buskens, C. A., van Kuppevelt, T. H., Veerkamp, J. H., Monnens, L. A. and van den Heuvel, L. P. (1998). Primary structure and high expression of human agrin in basement membranes of adult lung and kidney. Eur J Biochem 254, 123-8.

Guimond, S., Maccarana, M., Olwin, B. B., Lindahl, U. and Rapraeger, A. C.(1993). Activating and inhibitory heparin sequences for FGF-2 (basic FGF). Distinct requirements for FGF-1, FGF-2, and FGF-4. J Biol Chem 268, 23906-14.

55

Habuchi, H., Miyake, G., Nogami, K., Kuroiwa, A., Matsuda, Y., Kusche-Gullberg, M., Habuchi, O., Tanaka, M. and Kimata, K. (2003). Biosynthesis of heparan sulphate with diverse structures and functions: two alternatively spliced forms of human heparan sulphate 6-O-sulphotransferase-2 having different expression patterns and properties. Biochem J 371, 131-42.

Habuchi, H., Tanaka, M., Habuchi, O., Yoshida, K., Suzuki, H., Ban, K. and Kimata, K. (2000). The occurrence of three isoforms of heparan sulfate 6-O- sulfotransferase having different specificities for hexuronic acid adjacent to the targeted N-sulfoglucosamine. J Biol Chem 275, 2859-68.

Häcker, U., Lin, X. and Perrimon, N. (1997). The Drosophila sugarless gene modulates Wingless signaling and encodes an enzyme involved in polysaccharide biosynthesis. Development 124, 3565-73.

Haerry, T. E., Heslip, T. R., Marsh, J. L. and O'Connor, M. B. (1997). Defects in glucuronate biosynthesis disrupt Wingless signaling in Drosophila. Development 124, 3055-64.

HajMohammadi, S., Enjyoji, K., Princivalle, M., Christi, P., Lech, M., Beeler, D., Rayburn, H., Schwartz, J. J., Barzegar, S., de Agostini, A. I. et al.(2003). Normal levels of anticoagulant heparan sulfate are not essential for normal hemostasis. J Clin Invest 111, 989-99.

Halfter, W., Dong, S., Schurer, B. and Cole, G. J. (1998). Collagen XVIII is a basement membrane heparan sulfate proteoglycan. J Biol Chem 273, 25404-12.

Han, C., Belenkaya, T. Y., Khodoun, M., Tauchi, M. and Lin, X. (2004a). Distinct and collaborative roles of Drosophila EXT family proteins in morphogen signalling and gradient formation. Development.

Han, C., Belenkaya, T. Y., Wang, B. and Lin, X. (2004b). Drosophila glypicans control the cell-to-cell movement of Hedgehog by a dynamin-independent process. Development 131, 601-11.

Handler, M., Yurchenco, P. D. and Iozzo, R. V. (1997). Developmental expression of perlecan during murine embryogenesis. Dev Dyn 210, 130-45.

Hassell, J., Yamada, Y. and Arikawa-Hirasawa, E. (2002). Role of perlecan in skeletal development and diseases. Glycoconj J 19, 263-7.

Henningsson, F., Ledin, J., Lunderius, C., Wilen, M., Hellman, L. and Pejler, G. (2002). Altered storage of proteases in mast cells from mice lacking heparin: a possible role for heparin in carboxypeptidase A processing. Biol Chem 383,793-801.

Hoffmann, A. and Gross, G. (2001). BMP signaling pathways in cartilage and bone formation. Crit Rev Eukaryot Gene Expr 11, 23-45.

Hogan, B., Beddington, R., Costantini, F. and Lacy, E. (1994). Manipulating the mouse embryo: Cold Spring Harbour Laboratory press.

Horton, W. A. (2003). Skeletal development: insights from targeting the mouse genome. Lancet 362, 560-9.

Humphries, D. E., Wong, G. W., Friend, D. S., Gurish, M. F., Qiu, W. T., Huang, C., Sharpe, A. H. and Stevens, R. L. (1999). Heparin is essential for the storage of specific granule proteases in mast cells. Nature 400, 769-72.

Ilangumaran, S., He, H. T. and Hoessli, D. C. (2000). Microdomains in lymphocyte signalling: beyond GPI-anchored proteins. Immunol Today 21, 2-7.

Inatani, M., Irie, F., Plump, A. S., Tessier-Lavigne, M. and Yamaguchi, Y.(2003). Mammalian brain morphogenesis and midline axon guidance require heparan sulfate. Science 302, 1044-6.

Iozzo, R. V., Cohen, I. R., Grassel, S. and Murdoch, A. D. (1994). The biology of perlecan: the multifaceted heparan sulphate proteoglycan of basement membranes and pericellular matrices. Biochem J 302, 625-39.

56

Ishiguro, K., Kadomatsu, K., Kojima, T., Muramatsu, H., Nakamura, E., Ito, M., Nagasaka, T., Kobayashi, H., Kusugami, K., Saito, H. et al. (2000a). Syndecan-4 deficiency impairs the fetal vessels in the placental labyrinth. DevDyn 219, 539-44.

Ishiguro, K., Kadomatsu, K., Kojima, T., Muramatsu, H., Tsuzuki, S., Nakamura, E., Kusugami, K., Saito, H. and Muramatsu, T. (2000b). Syndecan-4 deficiency impairs focal adhesion formation only under restricted conditions. J Biol Chem 275, 5249-52.

Iwamoto, M., Enomoto-Iwamoto, M. and Kurisu, K. (1999). Actions of hedgehog proteins on skeletal cells. Crit Rev Oral Biol Med 10, 477-86.

Jackson, S. M., Nakato, H., Sugiura, M., Jannuzi, A., Oakes, R., Kaluza, V., Golden, C. and Selleck, S. B. (1997). dally, a Drosophila glypican, controls cellular responses to the TGF-beta-related morphogen, Dpp. Development 124,4113-20.

Jenniskens, G. J., Oosterhof, A., Brandwijk, R., Veerkamp, J. H. and van Kuppevelt, T. H. (2000). Heparan sulfate heterogeneity in skeletal muscle basal lamina: demonstration by phage display-derived antibodies. J Neurosci 20,4099-111.

Jenniskens, G. J., Ringvall, M., Koopman, W. J., Ledin, J., Kjellén, L., Willems, P. H., Forsberg, E., Veerkamp, J. H. and Van Kuppevelt, T. H. (2003). Disturbed Ca2+ kinetics in N-deacetylase/N-sulfotransferase-1 defective myotubes. J Cell Sci 116, 2187-93.

Kaksonen, M., Pavlov, I., Voikar, V., Lauri, S. E., Hienola, A., Riekki, R., Lakso, M., Taira, T. and Rauvala, H. (2002). Syndecan-3-deficient mice exhibit enhanced LTP and impaired hippocampus-dependent memory. Mol Cell Neurosci 21, 158-72.

Kamimura, K., Fujise, M., Villa, F., Izumi, S., Habuchi, H., Kimata, K. and Nakato, H. (2001). Drosophila heparan sulfate 6-O-sulfotransferase (dHS6ST) gene. Structure, expression, and function in the formation of the tracheal system. J Biol Chem 276, 17014-21.

Karlsen, T. V., Iversen, V. V., Forsberg, E., Kjellén, L., Reed, R. K. and Gjerde, E. A. (2004). Neurogenic inflammation in mice deficient in heparin-synthesizing enzyme. Am J Physiol Heart Circ Physiol 286, H884-8.

Katagiri, T., Boorla, S., Frendo, J. L., Hogan, B. L. and Karsenty, G. (1998). Skeletal abnormalities in doubly heterozygous Bmp4 and Bmp7 mice. Dev Genet 22, 340-8.

Kato, M., Wang, H., Bernfield, M., Gallagher, J. T. and Turnbull, J. E. (1994). Cell surface syndecan-1 on distinct cell types differs in fine structure and ligand binding of its heparan sulfate chains. J Biol Chem 269, 18881-90.

Kaufman, M. H. and Bard, J. B. L. (1999). The Anatomical Basis of Mouse Development: Academic Press.

Khare, N. and Baumgartner, S. (2000). Dally-like protein, a new Drosophila glypican with expression overlapping with wingless. Mech Dev 99, 199-202.

Kim, B. T., Kitagawa, H., Tamura, J., Saito, T., Kusche-Gullberg, M., Lindahl, U. and Sugahara, K. (2001). Human tumor suppressor EXT gene family members EXTL1 and EXTL3 encode alpha 1,4- N-acetylglucosaminyltransferases that likely are involved in heparan sulfate/ heparin biosynthesis. Proc Natl Acad Sci U S A 98, 7176-81.

Kimura, S., Cheng, J., Ida, H., Hao, N., Fujimori, Y. and Saku, T. (2000). Perlecan (heparan sulfate proteoglycan) gene expression reflected in the characteristic histological architecture of salivary adenoid cystic carcinoma. Virchows Arch 437, 122-8.

57

Kitagawa, H., Shimakawa, H. and Sugahara, K. (1999). The tumor suppressor EXT-like gene EXTL2 encodes an alpha1, 4-N-acetylhexosaminyltransferase that transfers N-acetylgalactosamine and N-acetylglucosamine to the common glycosaminoglycan-protein linkage region. The key enzyme for the chain initiation of heparan sulfate. J Biol Chem 274, 13933-7.

Kjellén, L., Bielefeld, D. and Hook, M. (1983). Reduced sulfation of liver heparan sulfate in experimentally diabetic rats. Diabetes 32, 337-42.

Kjellén, L., Pettersson, I., Unger, E. and Lindahl, U. (1992). Two enzymes in one: N-deacetylation and N-sulfation in heparin biosynthesis are catalyzed by the same protein. Adv Exp Med Biol 313, 107-11.

Kobayashi, M., Habuchi, H., Yoneda, M., Habuchi, O. and Kimata, K. (1997). Molecular cloning and expression of Chinese hamster ovary cell heparan- sulfate 2-sulfotransferase. J Biol Chem 272, 13980-5.

Kolset, S. O., Prydz, K. and Pejler, G. (2004). Intracellular proteoglycans. Biochem J 379, 217-27.

Kresse, H. (1973). Mucopolysaccharidosis 3 A (Sanfilippo A disease): deficiency of a heparin sulfamidase in skin fibroblasts and leucocytes. Biochem Biophys Res Commun 54, 1111-8.

Kresse, H., Hausser, H. and Schonherr, E. (1993). Small proteoglycans. Experientia 49, 403-16.

Kusche-Gullberg, M., Eriksson, I., Pikas, D. S. and Kjellén, L. (1998). Identification and expression in mouse of two heparan sulfate glucosaminyl N-deacetylase/N-sulfotransferase genes. J Biol Chem 273, 11902-7.

Kusche-Gullberg, M. and Kjellén, L. (2003). Sulfotransferases in glycosaminoglycan biosynthesis. Curr Opin Struct Biol 13, 605-11.

Lander, A. D. (1998). Proteoglycans: master regulators of molecular encounter? Matrix Biol 17, 465-72.

Lander, A. D., Nie, Q. and Wan, F. Y. (2002). Do morphogen gradients arise by diffusion? Dev Cell 2, 785-96.

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

Li, J. P., Gong, F., Hagner-McWhirter, A., Forsberg, E., Abrink, M., Kisilevsky, R., Zhang, X. and Lindahl, U. (2003). Targeted Disruption of a Murine Glucuronyl C5-epimerase Gene Results in Heparan Sulfate Lacking L-Iduronic Acid and in Neonatal Lethality. J Biol Chem 278, 28363-6.

Lidholt, K. and Lindahl, U. (1992). Biosynthesis of heparin. The D-glucuronosyl- and N-acetyl-D- glucosaminyltransferase reactions and their relation to polymer modification. Biochem J 287, 21-9.

Lin, X., Buff, E. M., Perrimon, N. and Michelson, A. M. (1999). Heparan sulfate proteoglycans are essential for FGF receptor signaling during Drosophila embryonic development. Development 126, 3715-23.

Lin, X. and Perrimon, N. (1999). Dally cooperates with Drosophila Frizzled 2 to transduce Wingless signalling. Nature 400, 281-4.

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

Lin, Y., Zhang, S., Rehn, M., Itaranta, P., Tuukkanen, J., Heljasvaara, R., Peltoketo, H., Pihlajaniemi, T. and Vainio, S. (2001). Induced repatterning of type XVIII collagen expression in ureter bud from kidney to lung type: association with sonic hedgehog and ectopic surfactant protein C. Development128, 1573-85.

58

Lind, T., Tufaro, F., McCormick, C., Lindahl, U. and Lidholt, K. (1998). The putative tumor suppressors EXT1 and EXT2 are glycosyltransferases required for the biosynthesis of heparan sulfate. J Biol Chem 273, 26265-8.

Lindahl, U. and Kjellén, L. (1991). Heparin or heparan sulfate--what is the difference? Thromb Haemost 66, 44-8.

Lindahl, U., Kusche-Gullberg, M. and Kjellén, L. (1998). Regulated diversity of heparan sulfate. J Biol Chem 273, 24979-82.

Lindahl, U., Lidholt, K., Spillmann, D. and Kjellén, L. (1994). More to "heparin" than anticoagulation. Thromb Res 75, 1-32.

Liu, Z., Xu, J., Colvin, J. S. and Ornitz, D. M. (2002). Coordination of chondrogenesis and osteogenesis by fibroblast growth factor 18. Genes Dev 16,859-69.

Luders, F., Segawa, H., Stein, D., Selva, E. M., Perrimon, N., Turco, S. J. and Hacker, U. (2003). Slalom encodes an adenosine 3'-phosphate 5'-phosphosulfate transporter essential for development in Drosophila. Embo J 22,3635-44.

Lum, L., Yao, S., Mozer, B., Rovescalli, A., Von Kessler, D., Nirenberg, M. and Beachy, P. A. (2003). Identification of Hedgehog pathway components by RNAi in Drosophila cultured cells. Science 299, 2039-45.

Marie, P. J. (2003). Fibroblast growth factor signaling controlling osteoblast differentiation. Gene 316, 23-32.

Marneros, A. G., Keene, D. R., Hansen, U., Fukai, N., Moulton, K., Goletz, P. L., Moiseyev, G., Pawlyk, B. S., Halfter, W., Dong, S. et al. (2004). Collagen XVIII/endostatin is essential for vision and retinal pigment epithelial function. Embo J 23, 89-99.

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

McCormick, C., Duncan, G. and Tufaro, F. (1999). New perspectives on the molecular basis of hereditary bone tumours. Mol Med Today 5, 481-6.

McCormick, C., Leduc, Y., Martindale, D., Mattison, K., Esford, L. E., Dyer, A. P. and Tufaro, F. (1998). The putative tumour suppressor EXT1 alters the expression of cell- surface heparan sulfate. Nat Genet 19, 158-61.

McLaughlin, D., Karlsson, F., Tian, N., Pratt, T., Bullock, S. L., Wilson, V. A., Price, D. J. and Mason, J. O. (2003). Specific modification of heparan sulphate is required for normal cerebral cortical development. Mech Dev 120,1481-8.

Merry, C. L., Bullock, S. L., Swan, D. C., Backen, A. C., Lyon, M., Beddington, R. S., Wilson, V. A. and Gallagher, J. T. (2001). The molecular phenotype of heparan sulfate in the Hs2st-/- mutant mouse. J Biol Chem 276, 35429-34.

Merry, C. L. and Wilson, V. A. (2002). Role of heparan sulfate-2-O-sulfotransferase in the mouse. Biochim Biophys Acta 1573, 319-27.

Mertens, G., Van der Schueren, B., van den Berghe, H. and David, G. (1996). Heparan sulfate expression in polarized epithelial cells: the apical sorting of glypican (GPI-anchored proteoglycan) is inversely related to its heparan sulfate content. J Cell Biol 132, 487-97.

Metcalfe, D. D., Baram, D. and Mekori, Y. A. (1997). Mast cells. Physiol Rev 77,1033-79.

Muragaki, Y., Timmons, S., Griffith, C. M., Oh, S. P., Fadel, B., Quertermous, T. and Olsen, B. R. (1995). Mouse Col18a1 is expressed in a tissue-specific manner as three alternative variants and is localized in basement membrane zones. Proc Natl Acad Sci U S A 92, 8763-7.

59

Nakamura, Y., Haines, N., Chen, J., Okajima, T., Furukawa, K., Urano, T., Stanley, P. and Irvine, K. D. (2002). Identification of a Drosophila gene encoding xylosylprotein beta4-galactosyltransferase that is essential for the synthesis of glycosaminoglycans and for morphogenesis. J Biol Chem 277,46280-8.

Neri, G., Gurrieri, F., Zanni, G. and Lin, A. (1998). Clinical and molecular aspects of the Simpson-Golabi-Behmel syndrome. Am J Med Genet 79, 279-83.

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

Nogami, K., Suzuki, H., Habuchi, H., Ishiguro, N., Iwata, H. and Kimata, K.(2004). Distinctive expression patterns of heparan sulfate O-sulfotransferases and regional differences in heparan sulfate structure in chick limb buds. J Biol Chem 279, 8219-29.

Norgard-Sumnicht, K. and Varki, A. (1995). Endothelial heparan sulfate proteoglycans that bind to L-selectin have glucosamine residues with unsubstituted amino groups. J Biol Chem 270, 12012-24.

Nurcombe, V., Ford, M. D., Wildschut, J. A. and Bartlett, P. F. (1993). Developmental regulation of neural response to FGF-1 and FGF-2 by heparan sulfate proteoglycan. Science 260, 103-6.

Nybakken, K. and Perrimon, N. (2002). Heparan sulfate proteoglycan modulation of developmental signaling in Drosophila. Biochim Biophys Acta 1573, 280-91.

Ohbayashi, N., Shibayama, M., Kurotaki, Y., Imanishi, M., Fujimori, T., Itoh, N. and Takada, S. (2002). FGF18 is required for normal cell proliferation and differentiation during osteogenesis and chondrogenesis. Genes Dev 16, 870-9.

Ornitz, D. M. (2000). FGFs, heparan sulfate and FGFRs: complex interactions essential for development. Bioessays 22, 108-12.

Ornitz, D. M. and Marie, P. J. (2002). FGF signaling pathways in endochondral and intramembranous bone development and human genetic disease. Genes Dev16, 1446-65.

Park, P. W., Reizes, O. and Bernfield, M. (2000). Cell surface heparan sulfate proteoglycans: selective regulators of ligand-receptor encounters. J Biol Chem275, 29923-6.

Park, Y., Rangel, C., Reynolds, M. M., Caldwell, M. C., Johns, M., Nayak, M., Welsh, C. J., McDermott, S. and Datta, S. (2003). Drosophila perlecan modulates FGF and hedgehog signals to activate neural stem cell division. Dev Biol 253, 247-57.

Pavlov, I., Voikar, V., Kaksonen, M., Lauri, S. E., Hienola, A., Taira, T. and Rauvala, H. (2002). Role of heparin-binding growth-associated molecule (HB-GAM) in hippocampal LTP and spatial learning revealed by studies on overexpressing and knockout mice. Mol Cell Neurosci 20, 330-42.

Pellegrini, L. (2001). Role of heparan sulfate in fibroblast growth factor signalling: a structural view. Curr Opin Struct Biol 11, 629-34.

Petitou, M., Casu, B. and Lindahl, U. (2003). 1976-1983, a critical period in the history of heparin: the discovery of the antithrombin binding site. Biochimie 85,83-9.

Pettersson, I., Kusche, M., Unger, E., Wlad, H., Nylund, L., Lindahl, U. and Kjellén, L. (1991). Biosynthesis of heparin. Purification of a 110-kDa mouse mastocytoma protein required for both glucosaminyl N-deacetylation and N-sulfation. J Biol Chem 266, 8044-9.

60

Pikas, D. S., Eriksson, I. and Kjellén, L. (2000). Overexpression of different isoforms of glucosaminyl N-deacetylase/N- sulfotransferase results in distinct heparan sulfate N-sulfation patterns. Biochemistry 39, 4552-8.

Pilia, G., Hughes-Benzie, R. M., MacKenzie, A., Baybayan, P., Chen, E. Y., Huber, R., Neri, G., Cao, A., Forabosco, A. and Schlessinger, D. (1996). Mutations in GPC3, a glypican gene, cause the Simpson-Golabi-Behmel overgrowth syndrome. Nat Genet 12, 241-7.

Pye, D. A., Vives, R. R., Hyde, P. and Gallagher, J. T. (2000). Regulation of FGF-1 mitogenic activity by heparan sulfate oligosaccharides is dependent on specific structural features: differential requirements for the modulation of FGF-1 and FGF-2. Glycobiology 10, 1183-92.

Rapraeger, A. C. (2000). Syndecan-regulated receptor signaling. J Cell Biol 149,995-8.

Rapraeger, A. C., Krufka, A. and Olwin, B. B. (1991). Requirement of heparan sulfate for bFGF-mediated fibroblast growth and myoblast differentiation. Science 252, 1705-8.

Reizes, O., Lincecum, J., Wang, Z., Goldberger, O., Huang, L., Kaksonen, M., Ahima, R., Hinkes, M. T., Barsh, G. S., Rauvala, H. et al. (2001). Transgenic expression of syndecan-1 uncovers a physiological control of feeding behavior by syndecan-3. Cell 106, 105-16.

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

Ronca, F., Andersen, J. S., Paech, V. and Margolis, R. U. (2001). Characterization of Slit protein interactions with glypican-1. J Biol Chem 276,29141-7.

Rossi, M., Morita, H., Sormunen, R., Airenne, S., Kreivi, M., Wang, L., Fukai, N., Olsen, B. R., Tryggvason, K. and Soininen, R. (2003). Heparan sulfate chains of perlecan are indispensable in the lens capsule but not in the kidney. Embo J 22, 236-45.

Rugh, R. (1991). The Mouse: Oxford University Press. Ruhrberg, C., Gerhardt, H., Golding, M., Watson, R., Ioannidou, S., Fujisawa,

H., Betsholtz, C. and Shima, D. T. (2002). Spatially restricted patterning cues provided by heparin-binding VEGF-A control blood vessel branching morphogenesis. Genes Dev 16, 2684-98.

Saarela, J., Rehn, M., Oikarinen, A., Autio-Harmainen, H. and Pihlajaniemi, T.(1998). The short and long forms of type XVIII collagen show clear tissue specificities in their expression and location in basement membrane zones in humans. Am J Pathol 153, 611-26.

Saksela, O., Moscatelli, D., Sommer, A. and Rifkin, D. B. (1988). Endothelial cell-derived heparan sulfate binds basic fibroblast growth factor and protects it from proteolytic degradation. J Cell Biol 107, 743-51.

Salmivirta, M. and Jalkanen, M. (1995). Syndecan family of cell surface proteoglycans: developmentally regulated receptors for extracellular effector molecules. Experientia 51, 863-72.

Salmivirta, M., Lidholt, K. and Lindahl, U. (1996). Heparan sulfate: a piece of information. Faseb J 10, 1270-9.

Samoszuk, M., Corwin, M., Yu, H., Wang, J., Nalcioglu, O. and Su, M. Y.(2003). Inhibition of thrombosis in melanoma allografts in mice by endogenous mast cell heparin. Thromb Haemost 90, 351-60.

61

Samoszuk, M. and Corwin, M. A. (2003). Acceleration of tumor growth and peri-tumoral blood clotting by imatinib mesylate (Gleevec trade mark ). Int J Cancer106, 647-52.

Sannes, P. L. (1984). Differences in basement membrane-associated microdomains of type I and type II pneumocytes in the rat and rabbit lung. J Histochem Cytochem 32, 827-33.

Saxen, L. and Thesleff, I. (1992). Epithelial-mesenchymal interactions in murine organogenesis. Ciba Found Symp 165, 183-93; discussion 193-8.

Schonherr, E. and Hausser, H. J. (2000). Extracellular matrix and cytokines: a functional unit. Dev Immunol 7, 89-101.

Schwartz, N. B. and Domowicz, M. (2002). Chondrodysplasias due to proteoglycan defects. Glycobiology 12, 57R-68R.

Sen, J., Goltz, J. S., Stevens, L. and Stein, D. (1998). Spatially restricted expression of pipe in the Drosophila egg chamber defines embryonic dorsal-ventral polarity. Cell 95, 471-81.

Senay, C., Lind, T., Muguruma, K., Tone, Y., Kitagawa, H., Sugahara, K., Lidholt, K., Lindahl, U. and Kusche-Gullberg, M. (2000). The EXT1/EXT2 tumor suppressors: catalytic activities and role in heparan sulfate biosynthesis. EMBO Rep 1, 282-6.

Sergeev, P., Streit, A., Heller, A. and Steinmann-Zwicky, M. (2001). The Drosophila dorsoventral determinant PIPE contains ten copies of a variable domain homologous to mammalian heparan sulfate 2-sulfotransferase. Dev Dyn220, 122-32.

Shu, W., Jiang, Y. Q., Lu, M. M. and Morrisey, E. E. (2002). Wnt7b regulates mesenchymal proliferation and vascular development in the lung. Development129, 4831-42.

Shworak, N. W., Liu, J., Fritze, L. M., Schwartz, J. J., Zhang, L., Logeart, D. and Rosenberg, R. D. (1997). Molecular cloning and expression of mouse and human cDNAs encoding heparan sulfate D-glucosaminyl 3-O-sulfotransferase. J Biol Chem 272, 28008-19.

Shworak, N. W., Liu, J., Petros, L. M., Zhang, L., Kobayashi, M., Copeland, N. G., Jenkins, N. A. and Rosenberg, R. D. (1999). Multiple isoforms of heparan sulfate D-glucosaminyl 3-O- sulfotransferase. Isolation, characterization, and expression of human cdnas and identification of distinct genomic loci. J Biol Chem 274, 5170-84.

Simons, K. and Ikonen, E. (1997). Functional rafts in cell membranes. Nature 387,569-72.

Smithies, O., Gregg, R. G., Boggs, S. S., Koralewski, M. A. and Kucherlapati, R. S. (1985). Insertion of DNA sequences into the human chromosomal beta-globin locus by homologous recombination. Nature 317, 230-4.

Song, H. H. and Filmus, J. (2002). The role of glypicans in mammalian development. Biochim Biophys Acta 1573, 241-6.

Spillmann, D. (2001). Heparan sulfate: anchor for viral intruders? Biochimie 83,811-7.

Spring, J., Paine-Saunders, S. E., Hynes, R. O. and Bernfield, M. (1994). Drosophila syndecan: conservation of a cell-surface heparan sulfate proteoglycan. Proc Natl Acad Sci U S A 91, 3334-8.

Steigemann, P., Molitor, A., Fellert, S., Jackle, H. and Vorbruggen, G. (2004). Heparan sulfate proteoglycan syndecan promotes axonal and myotube guidance by slit/robo signaling. Curr Biol 14, 225-30.

62

Stickens, D., Brown, D. and Evans, G. A. (2000). EXT genes are differentially expressed in bone and cartilage during mouse embryogenesis. Dev Dyn 218,452-64.

Stickens, D., Clines, G., Burbee, D., Ramos, P., Thomas, S., Hogue, D., Hecht, J. T., Lovett, M. and Evans, G. A. (1996). The EXT2 multiple exostoses gene defines a family of putative tumour suppressor genes. Nat Genet 14, 25-32.

Stringer, S. E., Mayer-Proschel, M., Kalyani, A., Rao, M. and Gallagher, J. T.(1999). Heparin is a unique marker of progenitors in the glial cell lineage. J Biol Chem 274, 25455-60.

Takei, Y., Ozawa, Y., Sato, M., Watanabe, A. and Tabata, T. (2004). Three Drosophila EXT genes shape morphogen gradients through synthesis of heparan sulfate proteoglycans. Development 131, 73-82.

Tchougounova, E., Forsberg, E., Angelborg, G., Kjellén, L. and Pejler, G.(2001). Altered processing of fibronectin in mice lacking heparin. a role for heparin-dependent mast cell chymase in fibronectin degradation. J Biol Chem276, 3772-7.

Tchougounova, E. and Pejler, G. (2001). Regulation of extravascular coagulation and fibrinolysis by heparin-dependent mast cell chymase. Faseb J 15, 2763-5.

The, I., Bellaiche, Y. and Perrimon, N. (1999). Hedgehog movement is regulated through tout velu-dependent synthesis of a heparan sulfate proteoglycan. Mol Cell 4, 633-9.

Torres, R. M. and Kühn, R. (1997). Laboratory protocols for Conditional Gene Targeting: Oxford University Press.

Toyoda, H., Kinoshita-Toyoda, A., Fox, B. and Selleck, S. B. (2000). Structural analysis of glycosaminoglycans in animals bearing mutations in sugarless, sulfateless, and tout-velu. Drosophila homologues of vertebrate genes encoding glycosaminoglycan biosynthetic enzymes. J Biol Chem 275 21856-61.

Tran, P. K., Tran-Lundmark, K., Soininen, R., Tryggvason, K., Thyberg, J. and Hedin, U. (2004). Increased intimal hyperplasia and smooth muscle cell proliferation in transgenic mice with heparan sulfate-deficient perlecan. CircRes 94, 550-8.

Tsuda, M., Kamimura, K., Nakato, H., Archer, M., Staatz, W., Fox, B., Humphrey, M., Olson, S., Futch, T., Kaluza, V. et al. (1999). The cell-surface proteoglycan Dally regulates Wingless signalling in Drosophila. Nature 400,276-80.

Tumova, S., Woods, A. and Couchman, J. R. (2000). Heparan sulfate chains from glypican and syndecans bind the Hep II domain of fibronectin similarly despite minor structural differences. J Biol Chem 275, 9410-7.

Uhal, B. D. (1997). Cell cycle kinetics in the alveolar epithelium. Am J Physiol 272,L1031-45.

ul Haque, M. F., King, L. M., Krakow, D., Cantor, R. M., Rusiniak, M. E., Swank, R. T., Superti-Furga, A., Haque, S., Abbas, H., Ahmad, W. et al.(1998). Mutations in orthologous genes in human spondyloepimetaphyseal dysplasia and the brachymorphic mouse. Nat Genet 20, 157-62.

Unger, E., Pettersson, I., Eriksson, U. J., Lindahl, U. and Kjellén, L. (1991). Decreased activity of the heparan sulfate-modifying enzyme glucosaminyl N-deacetylase in hepatocytes from streptozotocin-diabetic rats. J Biol Chem 266,8671-4.

van den Born, J., Gunnarsson, K., Bakker, M. A., Kjellén, L., Kusche-Gullberg, M., Maccarana, M., Berden, J. H. and Lindahl, U. (1995). Presence of N-unsubstituted glucosamine units in native heparan sulfate revealed by a monoclonal antibody. J Biol Chem 270, 31303-9.

63

Van Kuppevelt, T. H., Domen, J. G., Cremers, F. P. and Kuyper, C. M. (1984). Staining of proteoglycans in mouse lung alveoli. I. Ultrastructural localization of anionic sites. Histochem J 16, 657-69.

Varki, A., Cummings, R., Esko, J., Freeze, H., Hart, G. and Marth, J. (1999). Essentials of glycobiology: Cold Spring Harbor Laboratory Press.

Veugelers, M., Cat, B. D., Muyldermans, S. Y., Reekmans, G., Delande, N., Frints, S., Legius, E., Fryns, J. P., Schrander-Stumpel, C., Weidle, B. et al.(2000). Mutational analysis of the GPC3/GPC4 glypican gene cluster on Xq26 in patients with Simpson-Golabi-Behmel syndrome: identification of loss-of- function mutations in the GPC3 gene. Hum Mol Genet 9, 1321-8.

Viviano, B. L., Paine-Saunders, S., Gasiunas, N., Gallagher, J. and Saunders, S.(2004). Domain-specific Modification of Heparan Sulfate by Qsulf1 Modulates the Binding of the Bone Morphogenetic Protein Antagonist Noggin. J Biol Chem 279, 5604-11.

Vlodavsky, I., Friedmann, Y., Elkin, M., Aingorn, H., Atzmon, R., Ishai-Michaeli, R., Bitan, M., Pappo, O., Peretz, T., Michal, I. et al. (1999). Mammalian heparanase: gene cloning, expression and function in tumor progression and metastasis. Nat Med 5, 793-802.

Voigt, A., Pflanz, R., Schafer, U. and Jackle, H. (2002). Perlecan participates in proliferation activation of quiescent Drosophila neuroblasts. Dev Dyn 224, 403-12.

Warburton, D., Schwarz, M., Tefft, D., Flores-Delgado, G., Anderson, K. D. and Cardoso, W. V. (2000). The molecular basis of lung morphogenesis. Mech Dev 92, 55-81.

Weaver, M., Batts, L. and Hogan, B. L. (2003). Tissue interactions pattern the mesenchyme of the embryonic mouse lung. Dev Biol 258, 169-84.

Wei, Z., Swiedler, S. J., Ishihara, M., Orellana, A. and Hirschberg, C. B. (1993). A single protein catalyzes both N-deacetylation and N-sulfation during the biosynthesis of heparan sulfate. Proc Natl Acad Sci U S A 90, 3885-8.

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

Yogalingam, G. and Hopwood, J. J. (2001). Molecular genetics of mucopolysaccharidosis type IIIA and IIIB: Diagnostic, clinical, and biological implications. Hum Mutat 18, 264-81.

Yoneda, A. and Couchman, J. R. (2003). Regulation of cytoskeletal organization by syndecan transmembrane proteoglycans. Matrix Biol 22, 25-33.

Yong, L. C. (1997). The mast cell: origin, morphology, distribution, and function. Exp Toxicol Pathol 49, 409-24.

Zagris, N. (2001). Extracellular matrix in development of the early embryo. Micron32, 427-38.

Zako, M., Dong, J., Goldberger, O., Bernfield, M., Gallagher, J. T. and Deakin, J. A. (2003). Syndecan-1 and -4 synthesized simultaneously by mouse mammary gland epithelial cells bear heparan sulfate chains that are apparently structurally indistinguishable. J Biol Chem 278, 13561-9.

Zatterstrom, U. K., Felbor, U., Fukai, N. and Olsen, B. R. (2000). Collagen XVIII/endostatin structure and functional role in angiogenesis. Cell Struct Funct25, 97-101.

Zcharia, E., Metzger, S., Chajek-Shaul, T., Aingorn, H., Elkin, M., Friedmann, Y., Weinstein, T., Li, J. P., Lindahl, U. and Vlodavsky, I. (2004). Transgenic expression of mammalian heparanase uncovers physiological functions of heparan sulfate in tissue morphogenesis, vascularization, and feeding behavior. Faseb J 18, 252-63.

Acta Universitatis UpsaliensisComprehensive Summaries of Uppsala Dissertations

from the Faculty of Medicine

Editor: The Dean of the Faculty of Medicine

Distribution:Uppsala University Library

Box 510, SE-751 20 Uppsala, Swedenwww.uu.se, acta@ub.uu.se

ISSN 0282-7476ISBN 91-554-5955-2

A doctoral dissertation from the Faculty of Medicine, Uppsala University,is usually a summary of a number of papers. A few copies of the completedissertation are kept at major Swedish research libraries, while the sum-mary alone is distributed internationally through the series Comprehen-

sive Summaries of Uppsala Dissertations from the Faculty of Medicine.

(Prior to October, 1985, the series was published under the title “Abstracts ofUppsala Dissertations from the Faculty of Medicine”.)