Role of RAGE as an Amphoterin Receptor : From Development to Disease

R ^{OLE OF} RAGE ^{AS AN} A ^MPHOTERIN R ^ECEPTOR:

F ^ROM D EVELOPMENT TO D ^ISEASE

H ^ENRI H ^UTTUNEN

I

NSTITUTE OF

B

IOTECHNOLOGY

D

EPARTMENT OF AND

B

IOSCIENCES,

D

IVISION OF

B

IOCHEMISTRY

U

NIVERSITY OF

H

^ELSINKI

H

^ELSINKI

G

^RADUATE

S

^CHOOL

IN

B

IOTECHNOLOGY AND

M

^OLECULAR

B

^IOLOGY

A

^CADEMIC

D

ISSERTATION

H

^ELSINKI

2002

Role of RAGE as an Amphoterin Receptor:

From Development to Disease

Henri Huttunen

Institute of Biotechnology Department of Biosciences,and

Division of Biochemistry University of Helsinki Helsinki Graduate School in Biotechnology and Molecular Biology

Academic Dissertation

To be presented for public criticism, with the permission of the Faculty of Science, University of Helsinki, in the auditorium 1041 at Viikki Biocenter, Viikinkaari 5, Helsinki,

on October 18

^th

, 2002, at 12 o’clock noon

Helsinki 2002

Supervised by:

Professor Heikki Rauvala Institute of Biotechnology and Department of Biosciences University of Helsinki

Reviewed by:

Professor Jorma Keski-Oja Department of Virology, Haartman Institute, Biomedicum Helsinki University of Helsinki and

Docent Erkki Koivunen Department of Biosciences, Division of Biochemistry University of Helsinki

Opponent:

Professor Ann Marie Schmidt College of Physicians and Surgeons Columbia University

New York, U.S.A.

 Henri Huttunen ISSN: 1239-9469

ISBN: 952-10-0315-4 (Printed version) ISBN: 952-10-0316-2 (PDF-version)

http://ethesis.helsinki.fi/julkaisut/eri/biote/vk/huttunen/

Dark Oy, Vantaa 2002

TABLE OF CONTENTS

List of original publications. . . Abbreviations . . . 1. Summary . . . 2. Review of literature . . . . 2.1. Cell motility . . . 2.1.1. How do cells move? . . . 2.1.2. Mechanisms regulating cell motility

. . . 2.1.2.1. Extracellular signals . . . 2.1.2.2. Signal transduction and cytoskeleton . . . . 2.1.2.3. Proteolytic mechanisms . . . 2.2. Neuronal development . . . 2.2.1. General aspects . . . .

2.2.1.1. Determination of neuronal cell fate . . . . 2.2.1.2. Factors regulating neuronal proliferation and differentiation

. . . 2.2.1.3. Neuronal migration. . . 2.2.1.4. Neuronal cell death

. . . . 2.2.2. Formation of neuronal networks . . . . 2.2.2.1. Mechanisms controlling growth cone guidance . . . 2.2.2.2. Target recognition and synapse formation . . . . 2.3. Amphoterin. . . 2.3.1. Characteristics of amphoterin . . . . 2.3.1.1. Structure of amphoterin . . . 2.3.1.2. Expression and secretion of amphoterin . . . . 2.3.2. Amphoterin binding proteins and carbohydrates

. . . 2.3.2.1. Proteoglycans and glycolipids . . . 2.3.2.2. RAGE(Receptor for Advanced Glycation End Products)

. . . 2.3.2.3. IL-1 receptors . . . .

2.3.2.4. Plasminogen and plasminogen activators . . . 2.3.2.5. Borna disease virus phosphoprotein p24 . . . 2.3.3. Functions of amphoterin . . . .

2.3.3.1. Nuclear functions of amphoterin (HMGB1)

. . . 2.3.3.2. Process extension and cell migration . . . 2.3.3.3. Cell differentiation . . . 2.3.3.4. Cytokine-like functions . . . . 2.4. RAGE (Receptor for Advanced Glycation End Products) . . . . 2.4.1. Characteristics of RAGE . . .

2.4.1.1. Structure of RAGE . . . . 2.4.1.2. Expression of RAGE . . . . 2.4.2. Ligands of RAGE . . . 2.4.2.1. Advanced Glycation End Products (AGE)

. . . 2.4.2.2. Amyloids. . . . 2.4.2.3. Amphoterin . . . 2.4.2.4. S100 proteins . . . 2.4.3. Signalling mechanisms of RAGE . . . . 2.4.3.1. Activation of multiple parallel signalling pathways by RAGE . . . . 2.4.3.2. Regulation of gene expression by RAGE . . . 2.4.4. RAGE in pathophysiology . . . . 2.4.4.1. RAGE and diabetes . . . . 2.4.4.2. RAGE and amyloid diseases . . . 2.4.4.3. RAGE and inflammation . . . . 2.4.4.4. RAGE and cancer

. . . 3. Aims of the study . . . 4. Experimental procedures . . . 5. Results . . . 5.1. RAGE signalling in neuronal development (I, II, III) . . . 5.1.1. Signalling mechanism of RAGE-mediated neurite outgrowth . . . . 5.1.2. Amphoterin and S100 proteins coregulate neurite outgrowth and

cell survival through their binding to RAGE . . . . 5.1.3. Identification of novel RAGE-regulated genes . . .

56 78 88 89 109 1010 1211 1212 1313 1415 1515 1718 1818 1919 1919 1920 2021 2222 2324 2425 2525 2627 2728 2930 3031 3133 3435 3535

3536

5.1.4. RAGE induces neuronal differentiation . . . 5.2. Amphoterin-RAGE interaction regulates invasive migration (IV, V) . . . 5.2.1. Amphoterin is a regulator of cell migration . . . . 5.2.2. Amphoterin-RAGE interaction and invasive migration . . . 5.2.2.1. Amphoterin-RAGE interaction regulates transendothelial

migration of tumor cells . . . . 5.2.2.2. Inhibition of amphoterin-RAGE interaction by a C-terminal

amphoterin peptide suppresses formation of metastases . . . 5.3. Ligand recognition by RAGE (V, VI) . . . 5.3.1. A C-terminal motif in amphoterin binds to RAGE . . . . 5.3.2. Homology between S100 proteins and the RAGE-binding

motif in amphoterin . . . 5.3.3. Novel N-glycans influence amphoterin binding to RAGE . . . 6. Discussion . . . 6.1. Signalling mechanisms of RAGE . . . 6.2. Amphoterin-RAGE interaction as a regulator of cell motility

. . . . 6.3. Common mechanistic features in neurite outgrowth and tumor cell invasion

. . . 6.4. Is there a RAGE-binding structural motif? . . . 6.5. Therapeutic applications

. . . 7. Acknowledgements

. . . References

. . . .

3637 3737

38 3839 39 4040 4141 4344 4547 4849

LIST OF ORIGINAL PUBLICATIONS

This thesis is based on the following original articles, which are referred to in the text by their Roman numerals, and on unpublished results presented in the text.

I Henri J. Huttunen, Carole Fages and Heikki Rauvala (1999).

Receptor for Advanced Glycation End Products (RAGE)-mediated Neurite Outgrowth and Activation of NF-κB Require the Cytoplasmic Domain of the Receptor but Different Downstream Signalling Pathways.

Journal of Biological Chemistry 274(28): 19919-19924.

II Henri J. Huttunen, Juha Kuja-Panula, Guglielmo Sorci, Anna Lisa Agneletti, Rosario Donato and Heikki Rauvala (2000).

Coregulation of Neurite Outgrowth and Cell Survival by Amphoterin and S100 Proteins through Receptor for Advanced Glycation End Products (RAGE) Activation.

Journal of Biological Chemistry 275 (51): 40096-40105.

III Henri J. Huttunen, Juha Kuja-Panula and Heikki Rauvala (2002).

Receptor for Advanced Glycation End Products (RAGE) Signalling Induces CREB-dependent Chromogranin Expression during Neuronal Differentiation.

Journal of Biological Chemistry 277 (41): 38635-38646.

IV Carole Fages, Riitta Nolo, Henri J. Huttunen, Eeva-Liisa Eskelinen and Heikki Rauvala (2000).

Regulation of Cell Migration by Amphoterin.

Journal of Cell Science 113 (4): 611-620.

V Henri J. Huttunen, Carole Fages, Juha Kuja-Panula, Anne J. Ridley and Heikki Rauvala (2002).

Receptor for Advanced Glycation End Products-binding COOH-terminal Motif in Amphoterin Inhibits Invasive Migration and Metastasis.

Cancer Research 62 (16): 4805-4811.

VI Geetha Srikrishna, Henri J. Huttunen, Lena Johansson, Bernd Weigle, Yu Yamaguchi, Heikki Rauvala and Hudson H. Freeze (2002).

N-Glycans on the Receptor for Advanced Glycation End Products Influence Amphoterin Binding and Neurite Outgrowth.

Journal of Neurochemistry 80 (6): 998-1008.

ABBREVIATIONS

aa amino acid

Aβ amyloid-β peptide

AGE advanced glycation end products Arp2/3 actin-related protein 2/3

BDV Borna disease virus

CML carboxymethyllysine

CNS central nervous system

CREB cyclic AMP (cAMP) response element binding protein CSF colony stimulating factor

∆cyto cytoplasmic domain deletion mutant ECM extracellular matrix

EN-RAGE extracellular newly identified RAGE binding protein (a.k.a. S100A12) ERK extracellular signal-regulated kinase

E embryonic day

ES cell embryonic stem cell

GAP GTPase (guanine triphosphate)ase activating protein GDI guanine nucleotide dissociation inhibitors

GEF guanine nucleotide exchange factor GFAP glial fibrillary acidic protein

GST glutathione S-transferase GTP guanosine triphosphate

HLH helix-loop-helix

HMG high mobility group protein

HMGB1 high-mobility group-box protein 1 (a.k.a. amphoterin) HUVEC human umbilical vein endothelial cell

ICAM intercellular adhesion molecule

IL interleukin

JNK c-Jun-NH₂-terminal kinase

kb kilobase

kD, kDa kilodalton

LIF leukemia inhibitory factor

LPS lipopolysaccharide

MAP kinase mitogen-activated protein kinase M-CSF macrophage-colony stimulating factor

MEK MAPK/ERK kinase

MLCK myosin light chain (MLC) kinase MMP matrix metalloproteinase mRNA messenger ribonucleic acid NCAM neural cell adhesion molecule

NF-κB nuclear factor-κB

NFL neurofilament light chain

NGF nerve growth factor

NMR nuclear magnetic resonance

P postnatal day

PKA protein kinase A

PKC protein kinase C

RA retinoic acid

RAGE receptor for advanced glycation end products

ROS reactive oxygen species

RPTPζ/β receptor-type protein tyrosine phosphatase ζ/β

Rsk ribosomal S6 kinase

SDS-PAGE sodium dodecylsulphate polyacrylamide gel electrophoresis sRAGE soluble ectodomain of RAGE

TGF-β transforming growth factor-β TNF-α tumor necrosis factor-α

t-PA tissue-type plasminogen activator

Tuj1 β-III-tubulin

u-PA urokinase-type plasminogen activator VCAM vascular cell adhesion molecule WASP Wiskott-Aldrich syndrome protein

1. SUMMARY

Cell migration plays a vital role in many physiological processes, such as embryonic development and immunological defense, but also in pathological processes, such as tumor cell invasion. Cues in the microenvironment of the cell influence cell motility by regulating intracellular signalling pathways that modulate cytoskeletal reorganization within the cell and proteolytic activity on the surface of the cell.

Amphoterin, also known as High-Mobility Group-Box protein 1 (HMGB1), is highly expressed in immature and malignant cells. Although it lacks a classical secretion signal, amphoterin is released from cells to regulate neurite outgrowth, cell migration and differentiation. Inhibition of extracellular amphoterin blocks cell migration to laminin, a pivotal component of the extracellular matrix, suggesting that secreted amphoterin is an important regulator of cell motility.

Many extracellular functions of amphoterin are mediated by cell surface receptors. Receptor for Advanced Glycation End products (RAGE), a multiligand receptor belonging to the immunoglobulin superfamily, has been shown to bind to amphoterin and mediate neurite outgrowth on amphoterin-coated surfaces. In the past few years, RAGE has attracted a lot of attention, as it appears to be involved in a variety of pathological conditions, such as diabetes, Alzheimer’s disease and cancer.

The current thesis concentrates on identifying molecular mechanisms through which RAGE mediates effects of extracellular amphoterin on cells. The cytoplasmic domain of RAGE was found to be of critical importance in mediating neurite outgrowth and cell migration upon stimulation of the cells with amphoterin. Identification of novel RAGE-activated signalling pathways involving the Rho-family small GTPases Cdc42 and Rac1, and CREB, a transcription factor, links RAGE to the regulation of cell migration and neuronal differentiation, respectively. In addition, RAGE activation was found to increase the expression of the antiapoptotic protein Bcl-2, suggesting that RAGE can also promote cell survival.

Mechanistic evidence suggests that RAGE signalling can promote both trophic and toxic effects on cells depending on the nature of the ligand, its concentration and duration of exposure.

As both amphoterin and RAGE are expressed in various tumor cells and are known to be efficient regulators of cell motility, they are intimately linked to invasive migration of tumor cells. In this study, a RAGE-binding motif was identified in the C-terminus of amphoterin that could be used to inhibit neurite outgrowth, transendothelial migration of tumor cells and formation of metastasis in a mouse model.

Interestingly, the RAGE-binding C-terminal motif in amphoterin shares homology with S100 proteins that are also known to interact with RAGE. This suggests that amphoterin and S100 proteins bind to RAGE through a common structural motif. Furthermore, RAGE was found to carry novel, carboxylated, N-linked glycans that modulate ligand binding to RAGE.

Altogether, these results suggest that amphoterin and RAGE serve as a ligand-receptor pair having important roles in the regulation of neuronal process growth, survival and differentiation, and in the regulation of tumor cell invasiveness.

2. REVIEW OF LITERATURE

2.1. Cell motility

Movement is a major characteristic of living organisms. Prokaryotic cells, such as bacteria, are able to swim within their environment with the aid of specialized appendages called cilia and flagella. Higher eukaryotic cells display a far more sophisticated motile repertoire. In multicellular organisms cell migration plays a crucial part in many normal physiological processes, such as embryonic development, immunological defense mechanisms and wound healing, but also in pathological processes, such as tumor cell invasion. Most cell types are capable of movement within their tissue environment by crawling over substrates, while some highly motile cell types are capable of penetrating tight physical barriers such as the vascular endothelium. In general, contacts between the cell and the surrounding tissue microenvironment mediated through molecules expressed on the cell surface largely dictate the nature of cell movement.

2.1.1. How do cells move?

Cells from a wide variety of tissues are capable of movement. A migratory cell placed in culture flattens and migrates randomly over the substrate.

This is typical for fibroblasts, a cell type widely used to study cell movement. A fibroblast migrates over a substrate by extending a broad, flat, leading edge or lamellipodium in conjunction with thinner, finger-like filopodia. These attach to the substrate at specific, transient attachment points allowing filopodia to explore the immediate surroundings of the cell. The cell then moves forward as a result of traction within the cytoplasm, and the cycle is completed by release and retraction of the tail of the cell (Fig. 1). Thus, migration of cells requires coordination of five cellular processes;

(1) extension of the leading edge; (2) adhesion to the matrix or to the surface of another cell;

(3) contraction of the cytoplasm; (4) release from contact sites; and (5) retraction of the tail of the cell along with recycling of membrane receptors from the rear to the front of the cell (reviewed in Sheetz et al., 1999).

Regulated cell movement can occur in many ways. Fibroblast migration on a substrate is the most simple, classical example of cell migration.

Cells can also migrate through tight physical barriers, as happens in the migration of leukocytes from the circulation through the vessel wall to the site of inflammation in tissue. Cells can also move cooperatively in big aggregates as happens during gastrulation when cells of the early embryo are

organized to establish the multilayered body plan of the organism. Furthermore, formation of neuronal networks during development is an example of cell migration where the cell soma stays still while a specialized cell compartment called the neuronal growth cone traverses long distances creating a network of axons and dendrites, a basis for neuronal signalling in the brain. Although all these different forms of cell migration display highly specialized features, they share the mechanistic basis involving cooperation of adhesive systems between the cells and their actin cytoskeletons.

2.1.2. Mechanisms regulating cell motility

The leading edge of a migrating cell contains a complex network of actin microfilaments (Fig.

1). Actin microfilaments are extremely dynamic, and their coordinated assembly nucleated at the membrane of the leading edge is believed to be responsible for extrusion of the leading edge of the cell. In the past ten years, significant progress has been made to understand intracellular signalling mechanisms mediating signals from cell surface receptors to molecules regulating cytoskeletal assembly. In addition to intracellular signalling and regulation of the cytoskeleton, pericellular proteolysis is thought to play a major role in the cell movement. The surrounding tissue and extracellular matrix often forms a physical barrier, which has to be broken down in a coordinated fashion to facilitate cell movement in tissue.

Figure 1. Schematic illustration of cell movement. Top view (A) and profile view (B) illustrate reorganizing actin cytoskeleton in the filopodia and lamellipodia of the lead- ing edge of a migrating cell.

Figure 2. Mechanistic example of extracellular signal- induced extension of a filopodium. An extracellular ligand binds to its membrane receptor which triggers formation of a complex of signaling molecules. The signaling com- plex activates Cdc42, a small GTPase which recruites N- WASP and Arp 2/3 complex to the membrane to initiate actin nucleation, a prerequisite for formation of filamen- tous actin to drive the protrusion of the filopodium.

2.1.2.1. Extracellular signals

The signals in the extracellular environment of the cell are the most important determinants of cell migration. These signals can be either (1) structural components of neighboring cells or the extracellular matrix, (2) gradients of soluble molecules, such as growth factors, cytokines or chemoattractants, or (3) gradients of extracellular matrix (ECM) components or molecules attached to the ECM. Chemotaxis refers to the unidirectional movement of a cell in response to a chemical gradient. The directed migration, or chemotaxis, of immune cells is an essential feature of the immune system. Leukocytes move from a low to a high concentration of chemoattractant.

Chemotaxis plays a crucial role in the recruitment of leukocytes into sites of inflammation and infection (reviewed in Downey, 1994). A large number of chemotactic factors and receptors are currently known, including chemokines, a large superfamily of endogenous chemotactic factors that regulate leukocyte trafficking through seven- transmembrane, G protein-coupled receptors (Zlotnik and Yoshie, 2000).

Extracellular matrix provides the physical microenvironment in which cells live, serves as a substrate for cell anchorage, transmits environmental signals to cells affecting all aspects of a cell’s life, including its proliferation, differentiation and death. The ECM is highly organized fibrillar meshwork composed of a variety of molecules, such as collagens, laminin, fibronectin, tenascin, thrombospondin and proteoglycans (reviewed in Aumailley and Gayraud, 1998). Although the extracellular matrix can hinder cell migration as a physical barrier, it can also serve as an important regulator of cell migration (reviewed in Geiger et al., 2001). Cell adhesion to the ECM is intrinsically associated with the regulation of cell migration (reviewed in Lackie et al., 1999).

Many of the effects of the ECM on cellular morphology, adhesion and motility are mediated by integrins. Integrins are a large family of heterodimeric transmembrane proteins, which function as cell adhesion molecules linking macromolecules of the ECM to the cell surface and cytoskeleton. Other cell surface receptor families that mediate contacts to the ECM and also to neighboring cells include cadherins, selectins, various proteoglycans and immunoglobulin superfamily cell adhesion molecules (IgCAMs)(reviewed in Juliano, 2002).

2.1.2.2. Signal transduction and cytoskeleton

Transmission of extracellular signals through the cell surface receptors to the actin

microfilaments is an integral event in cell migration (reviewed in Macieira-Coelho, 2000).

Attachment of the cell to the ECM is linked to the actin cytoskeleton via e.g. integrins, which are complexed with intracellular signalling molecules in focal adhesions and transmit signals bidirectionally across the cell membrane (reviewed in Geiger et al., 2001; Juliano, 2002). A key player in the signalling pathways regulating the dynamic properties of the actin cytoskeleton is the Rho family of small guanosine triphosphate-binding proteins (GTPases)(reviewed in Ridley, 2001).

One member of the Rho-family, Cdc42, regulates a signal transduction pathway leading to the formation of filopodia (Fig. 2), and another, Rac, controls formation of lamellipodia and membrane ruffles. Rho itself leads to the assembly of focal adhesions and actin-based stress fibers which are thought to interact with myosin motor proteins to produce the cytoplasmic contractions that result in the cell moving forwards (reviewed in Ridley, 2001).

The complex signal transduction pathways downstream of cell surface receptors culminate in

the activation of three classes of proteins regulating the cycling of Rho GTPases between active GTP- bound and inactive GDP-bound conformations.

These regulators include guanosine nucleotide exchange factors (GEFs), GTPase activating proteins (GAPs) and GDP dissociation inhibitors (GDIs)(reviewed in Schmitz et al., 2000). Numerous downstream effectors activated by Rho-family GTPases have been identified and some relay signals directly to the actin cytoskeleton (reviewed in Aspenstrom, 1999). These include WASP and WAVE family proteins that interact with Cdc42 and Rac, respectively (reviewed in Takenawa and Miki, 2001). WASP and WAVE proteins activate the Arp2/3 complex, a multifunctional protein complex that nucleates actin filaments and crosslinks them into cortical actin networks (reviewed in (Higgs and Pollard, 2001); Fig. 2).

In addition to the Rho-family small GTPases, other signalling pathways are also involved in the regulation of cell migration. For example, two mitogen-activated protein (MAP) kinases, namely extracellular signal regulated kinases 1 and 2 (ERK1/2) have been shown to regulate cell motility by phosphorylating and enhancing myosin light chain kinase (MLCK) activity leading to phosphorylation of myosin light chains (MLC)(Klemke et al., 1997; Cho and Klemke, 2000). Coordinated cell movement also requires dynamic interaction between the actin cytoskeleton and other cytoskeletal systems, especially the microtubules (reviewed in Waterman-Storer and Salmon, 1999).

2.1.2.3. Proteolytic mechanisms

Cell migration is critically dependent on interplay between forces of attachment and detachment. In order to regulate this delicate balance, cells are known to focus proteinases at their leading edge, where proteolysis can promote detachment, break-down of ECM barriers, inactivate inhibitory matrix components and direct cell migration (reviewed in Murphy and Gavrilovic, 1999). Pericellular proteolysis might also affect cell migration by releasing growth factors and generating modulatory neo-epitopes (reviewed in Taipale and Keski-Oja, 1997). The plasmin system driven by urokinase-type plasminogen activator (u-PA) and tissue-type plasminogen activator (t-PA) colocalizes with integrins in focal contacts and at the leading edge of migrating cells to generate localized proteolytic activity (reviewed in Dano et al., 1985; Johnsen et al., 1998). Another central group of proteolytic enzymes involved in the regulation of cell migration are the matrix metalloproteinases (MMP), a large family of secreted and transmembrane proteolytic enzymes

that process or degrade numerous pericellular substrates (reviewed in Sternlicht and Werb, 2001).

2.2. Neuronal development

2.2.1. General aspects

Undoubtedly the most challenging of all biological enigmas is the orderly construction of the human brain, an organ that coordinates a consious mind powerful enough to question its own creation. It has been estimated that during 15 weeks of human embryonic development

~200,000 neurons are formed per minute to make roughly 100 billion (10¹¹) neurons. However, it is not only the number of neurons but also the immense number and plasticity of synaptic connections between the neurons that underlies the power of the brain as an information processor. Each cortical neuron may have up to 10 000 synapses giving a total of ~10¹⁵ synapses in the whole network of neuronal connections (Zigmond et al., 1999). It is not easy to begin to understand the construction of such a complex system. However, during the last two decades the combination of classical developmental biology with the powerful methods of molecular biology has enabled rapid progess in understanding the basic mechanisms responsible for the development of the nervous system (Cowan et al., 1997).

In humans, neurogenesis begins at the third week of gestation and continues to 18 weeks whereas in mice it begins at embryonic day 9 and lasts until around embryonic day 20 (Gilbert, 1997; summarized in Table 1). The construction of vertebrate central nervous system begins after gastrulation in which the embryo has been divided into an internal endodermal layer, an intermediate mesodermal layer and an external ectoderm. In neurulation, the mesoderm directs the ectoderm overlying it to invaginate and form a hollow tube called neural tube. This newly formed neuroectoderm, the rudiment of the central nervous system, then undergoes patterning of the neurogenic region as it is divided into three primary brain vesicles, the areas that will later become forebrain, midbrain and hindbrain. Further subdivision results in formation of secondary vesicles giving rise to subregions of the brain, called telencephalon, diencephalon, mesencephalon, metencephalon and myelencephalon (Gilbert, 1997).

The original neural tube is composed of a rapidly dividing cell population called germinal neuroepithelium. As these cells divide the neural tube thickens and acquires a three-

Table 1. Key steps in the development of the nervous system.

layered structure, in which the original germinal neuroepithelium forms the ventricular zone (later becoming the ependyma). Cells migrating from the ventricular zone form a new layer called the intermediate zone (also called mantle zone,

“the gray matter”) in which most cells will later differentiate into glial cells and neurons sending axons forward to form the topmost layer called the marginal zone (“the white matter”). In the cerebral cortex the marginal zone is further organized into cortical plate containing six layers (Gilbert, 1997).

2.2.1.1. Determination of neuronal cell fate

Differential proliferation required for brain morphogenesis is followed by determination of cell fate, neuronal differentiation and migration. The cells that remain in the ventricular zone become ependymal cells that give rise to the precursors of neurons and glial cells. There is a wide variety of different neuronal and glial cell types. Combinations of distinct genetic subprograms are thought to be responsible for generation of more than 100 different neuronal cell types (Cowan et al., 1997).

Some components of neuronal identity, such as expression of synaptic vesicle and cytoskeletal proteins, are shared by all neurons. Others are

restricted to particular classes of neurons, such as expression of various neurotransmitters and their receptors.

Extrinsic and intrinsic cues act in concert to determine the fate of neuronal precursor cells.

Complex spatiotemporal expression patterns and cascades of different transcription factors, especially proneural basic helix-loop-helix (bHLH) and paired homeodomain proteins, switch on and off genes eventually determining neuronal identity.

These programs of gene expression operate during neurogenesis at the early embryonic period (E8.5-E13 in mouse).

As the expression of specific transcription factors begins within different regions of the developing brain, subsets of neurons begin to acquire a dorsal or ventral identity. The establishment of the dorso-ventral (DV) axis is concurrentwith the expression of transcription factors that mark specificcell types within the dorsal and ventral areas. Locally acting peptide growth factors induce cells towards dorsal cell fates, and Sonic Hedgehog protein (Shh) towards ventral fates (Cowan et al., 1997). By virtue of theirlocation in the dorsal or ventral aspect of the neuraxis,cells become specified toward cortical and subcortical fates.

2.2.1.2. Factors regulating neuronal proliferation and differentiation

Neural stem cells can give rise to both neurons and glia (reviewed in Blau et al., 2001).

Neuroblasts are a group of further differentiated precursor cells that can only give rise to neurons. In general, to begin differentiation cells are withdrawn from the cell cycle. Thus, the factors that induce differentiation often induce also proliferation arrest.

The vitamin A derivative retinoic acid (RA) is essential for normal development, especially in the developing CNS where it acts in a paracrine manner to regulate patterning of the nervous system and differentiation of progenitor cells to neurons (reviewed in Maden, 2001). Retinoic acid is also a potent factor regulating neuronal differentiation of embryonic stem (ES) cells in vitro (reviewed in Guan et al., 2001). Several transcription factors, intracellular signalling molecules, cytoplasmic proteins and extracellular molecules are known to be necessary for RA-induced differentiation (Maden, 2001). Another important regulator of neuronal differentiation is the family of homeobox genes. There are over 25 known homeobox genes expressed in a specific spatiotemporal pattern in the vertebrate forebrain that regulate cell proliferation and neuronal differentiation in addition to regional specification (Cowan et al., 1997). From extrinsic molecules regulating neuronal differentiation the most important are neurotrophic factors, such as nerve growth factor (NGF), and polypeptide growth factors, such as basic fibroblast growth factor (bFGF), epidermal growth factor (EGF), Sonic Hedgehog (Shh) and bone morphogenetic protein-4 (BMP-4) (Zigmond et al., 1999; Blau et al., 2001).

2.2.1.3. Neuronal migration

Embryonic neurons are typically born at some distance from theirsites of residence and action in the mature nervous system,and this is particularly true for peripheral neurons, and midbrain and hindbrain neurons in the CNS. The migratory pathways taken by newborn neurons are specifically dependent on substrate, cell surface and diffusibleguidance cues. Many of these guidance cues are provided by glial cells and theirprecursors (reviewed in Hatten, 1999). After primaryneurogenesis in the ventricular zones has begun, individualpostmitotic cells start migrating along the glialfiber system. Newborn neurons that emerge from the ventricular zone attach themselves to radial glial processes and then migrate along these processes toward the pial surface of the developing neocortex. An analogous process occursduring the migration of newborn

granule cells, along Bergmannglial processes, from the external to the internal granule celllayer of the developing postnatal cerebellum (reviewed in Hatten, 1999). In addition to the migration of cells from the ventricular zone towards the surface, tangential migration within the intermediate zone is known to occur. Most of these migratory events occur from midembryogenesis through early postnatal life.

A number of neuronal and glial receptor systems have been implicatedin the directed migration of CNS neurons along radial glialfibers.

These include neuregulin and its receptor erbB4, astrotactin and many ECM components, such as thrombospondin and tenascin (reviewed in Hatten, 1999).

2.2.1.4. Neuronal cell death

Lewis Thomas once wrote, ”By the time I was born, more of me had died than survived”

(Thomas, 1992). As surprising as it sounds this is true in the developing nervous system. In many parts of the developing vertebrate central and peripheral systems, over half of the neurons die by a mechanisms called apoptosis or programmed cell death. Neuronal apoptosis serves an important role in sculpting the maturing nervous system (Gilbert, 1997). An excess number of neurons differentiate and send axons to their targets but are submitted to competition for limiting amounts of target-derived survival factors to ensure that neurons only live when and where they are needed (Cowan et al., 1997). Many neuronal populations display dependency on at least one of the many neurotrophic factors. If neurons are deprived of trophic factors, they will undergo apoptosis.

The most important target-derived regulators of neuronal survival are the neurotrophic factors.

A vast number of proteins with neurotrophic properties have been characterized. The best characterized is the neurotrophin-family, including NGF, brain-derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3) and neurotrophin-4/5 (NT- 4/5)(Huang and Reichardt, 2001). Other proteins represent many common protein families, such as fibroblast growth factors (bFGF), transforming growth factor-β superfamily (glial cell line-derived neurotrophic factor; GDNF) and cytokines (ciliary neurotrophic factor; CNTF)(Cowan et al., 1997;

Gilbert, 1997).

A common feature for most neurotrophic factors is that their receptors activate signalling pathways which induce the expression of the anti- apoptotic proteins and repress the expression of proapoptotic proteins within the cell (Cowan et al., 1997). The Bcl-2 family of regulators of apoptosis and the caspase-family of proteases activated by

the apoptotic program are crucial targets of trophic factor signalling (Chao and Korsmeyer, 1998;

Grutter, 2000).

2.2.2. Formation of neuronal networks

Neuronal development is characterized by two major organizational periods. The first period, mainly reviewed above, includes the major histogenetic events neurulation, proliferation, migration, and differentiation. The second period is a time of reorganization in the cerebral cortex.

These events begin during gestation and continue postnatally, possibly through the second decade of human life. This stage is characterized by dendritic and axonal growth, synapse production, neuronal and synaptic pruning, and changes in neurotransmitter sensitivity. Although the initiation of these events is influenced by endogenous signals, further neural maturation is primarily influenced by exogenous signals.

Connecting approximately 100 billion neurons together through ~10¹⁵ synapses is a massive task. Neuronal connections are built mainly during embryonic development when each differentiating neuron sends out a neurite which migrates through the embryonicenvironment to its synaptic targets, laying down the extending axon or dendrite (Tessier-Lavigne and Goodman, 1996; Cowan et al., 1997). Growing neuronal processes often cannot be distinguished as axons or dendrites and

are thus collectively called neurites. The leading edge of the growing neurite contains a specialized apparatus named over a century ago by Ramón y Cajal as the growth cone (Ramón y Cajal, 1892) that transduces the extracellular signals in its environment to intracellular signals directing the growing neurite.

2.2.2.1. Mechanisms controlling growth cone guidance

The migration of growth cones to their targets occurs stepwise. Each small segment of the growing axon faces slightly different environment as the projecting axon traverses through layers of the embryonic nervous system. Navigation of the growing axon is influenced by cues falling to four basic categories: local cues and diffusible, long- range cues, each of which can be either attractive or repulsive to the growth cone (reviewed in (Tessier-Lavigne and Goodman, 1996; Cowan et al., 1997); summarized in Fig. 3). In addition, the growth cone can secrete autocrine factors such as proteases that facilitate penetratation of the growth cone through the tissue. These basic guidance forces likely act together, for example a growth cone could be simultaneously attracted by a long range cue, channeled to a path covered by local permissive and attractive forces and kept in this path by surrounding nonpermissive and repulsive factors.

Figure 3. Growth cone guidance mechanisms.

The traditional growth cone guidance forces include contact attraction (1), contact repulsion (2), chemoattraction (3) and chemorepulsion (4).

Attractive mechanisms include both permissive and attractive forces whereas repulsive mec- hanisms include both inhibitory and repulsive forces. In addition, autocrine mechanisms where the growth cone can promote its own movement by secreting molecules promoting growth cone migration have been included.

These include e.g. local proteolytic mechanisms that enhance growth cone movement by degrading the surround- ing extracellular matrix molecules.

A wide variety of developmentally regulated molecules involved in growth cone guidance has been identified (summarized in Table 2). Diffusible long range cues include the netrins, a family of secreted proteins related to the laminins (Yu and Bargmann, 2001) and various neurotrophins (Huang and Reichardt, 2001). Also some TGF-β superfamily members such as GDNF (Colavita et al., 1998; Saarma, 2000), and some neurotrophic cytokines (neuropoietins), such as CNTF (Murphy et al., 1997), might serve as diffusible guidance cues. Semaphorins are a family of secreted and transmembrane proteins mainly serving as repulsive guidance cues (Yu and Bargmann, 2001).

However, many of these molecules can serve as bifunctional guidance cues attracting some axons and repelling some axons away depending on the type of the neuron (Tessier-Lavigne and Goodman, 1996; Cowan et al., 1997).

Local modulators of axon guidance can act either on the cell surface or in the ECM, and can provide a variety of attractive and repulsive signals. Members of Slit and Ephrin families are membrane associated proteins that serve as the most important local repulsive guidance cues (Yu and Bargmann, 2001; Wilkinson, 2001). Laminin is the best known ECM molecule regulating axon growth and guidance (Cowan et al., 1997).

Many cell adhesion molecules belonging to the immunoglobulin (IgCAMs) superfamily or to the cadherin family also participate in axon guidance, mainly as attractive local cues (Walsh and Doherty, 1997; Ranscht, 2000). In the adult nervous system many components of the myelin sheets around axons, such as Nogo proteins, serve as potent inhibitors of axon growth complicating regeneration

of neuronal connections after injury (Brittis and Flanagan, 2001).

An important role in transducing the extracellular guidance cues to actual movement of the growth cone is played by the cell surface receptors that utilize the general mechanisms involved in the regulation of cell migration (reviewed in section 2.1.2.). These signalling pathways culminate in the regulation of the actin cytoskeleton within the growth cone. The Rho-family small GTPases Rac and Cdc42 are thought to be activated by attractive guidance cues to promote growth cone advance whereas Rho is activated by repulsive signals to inhibit growth by inducing retraction and collapse of the growth cone (Dickson, 2001).

2.2.2.2. Target recognition and synapse formation

Once axons have found their approximate target area using environmental molecular cues, growth cones recognize their specific targets using topographic maps of graded cues and unique labels marking distinct targets. After recognition of specific targets within the target area synapses are built between the axon and the target. Synapses are highly specialized subcellular structures built across intercellular junctions between neurons and partner cells, which utilize chemical neurotransmitters to pass electrical signals traveling along the axon to the target cell.

It has been estimated that in humans between the second month of gestation and two years after birth roughly 1.8 million synapses are formed per second.

Table 2. Molecules involved in growth cone migration and guidance.

The presynaptic active zone forms on the axon terminal, containing the synaptic vesicles carrying the neurotransmitters and specialized secretion machinery releasing the neurotransmitter upon stimulation. Across the synaptic cleft opposing the presynaptic density, a specialized postsynaptic site is formed containing densely positioned neurotransmitter receptors and signalling machinery ready to convert the neurotransmitter- mediated activation into an appropriate cellular response (reviewed in Jin, 2002).

In the final step of the wiring of the brain, patterns of neuronal activity drive the refinement of the initial connections into highly tuned circuits.

Neuronal connections are often adjusted so that neurons active at the same time (“Fire together”) increase the strength of their connection (“Wire together”). This process continues throughout the organism’s life and is largely responsible for the astonishing plasticity of the nervous system (Zigmond et al., 1999).

2.3. Amphoterin

Our studies have focused on two proteins, amphoterin and RAGE, operating as a ligand- receptor pair during the embryonic and neonatal development of the nervous system. Numerous studies on these proteins have shown that they also interact with other molecules and have a variety of other functions outside the nervous system.

2.3.1. Characteristics of amphoterin

Amphoterin is an ubiqitous, highly conserved molecule originally isolated from perinatal rat brain as a heparin-binding protein which promotes neurite outgrowth in brain neurons in vitro (Rauvala and Pihlaskari, 1987). Surprisingly, amphoterin sequence turned out to be identical to the sequence cloned for high mobility group 1 protein (HMG- 1) (Walker et al., 1980). The designation “HMG”

refers to non-histone components of chromatin.

In addition to HMG-1, amphoterin has also been

called as p30, sulphoglucuronyl carbohydrate binding protein-1 (SBP-1) and differentiation enhancing factor (DEF). The nomenclature for HMG protein family has recently been revised and amphoterin is now called HMGB1 (Bustin, 2001).

However, in our laboratory we have used the term

“amphoterin” referring to secreted, extracellularly active protein and “HMGB1” referring to nuclear protein serving as a structural component of chromatin. For reasons of clarity only the term amphoterin will be used in this review.

2.3.1.1. Structure of amphoterin

Amphoterin has a highly dipolar structure (designated therefore as amphoterin) consisting of a 185-amino acid basic region followed by a cluster of 30 acidic residues in the carboxy- terminus (Merenmies et al., 1991; Fig. 4 and Fig.

5A). The 185-amino acid basic part (~28% lysine or arginine residues) of amphoterin is subdivided in two homologous HMG boxes (Fig. 5A) each

~75 amino acids in length. Both HMG boxes have similar α-helical structure important in DNA-binding (Weir et al., 1993; Read et al., 1993; Fig. 5B). The 3-dimensional structure of the whole amphoterin molecule has not been resolved to date. However, it seems likely that the dipolar nature is the apparent reason for the formation of amphoterin dimers and oligomers under physiological conditions (Rauvala and Pihlaskari, 1987).

The N-terminus of amphoterin (residues 6-12 in Fig. 4) contains a consensus sequence found in a variety of heparin-binding proteins (Cardin and Weintraub, 1989). The region between the HMG box 2 and the acidic tail of amphoterin is the most distinctive feature of amphoterin as compared to the other members of the HMG protein family (residues 151-183 in Fig. 4). We have recently discovered that this part of amphoterin contains a motif responsible for RAGE binding (V; section 5.3.1.). Extracellular amphoterin has recently been reported to be proteolytically processed to a short 10 amino acid peptide active on erythroleukaemia cell differentiation (Sparatore et al., 2001); residues 130-139 in Fig. 4). The Figure 4. Amino acid sequence of rat amphoterin. Heparin-binding consensus sequence is shown by black text in a dark grey box, extracellular serine protease generated peptide by black text in a light grey box, the C-terminal RAGE- binding motif by white text in a dark grey box and the peptide used to produce antipeptide II antibody by white text in a black box.

location of these motifs important in the function of extracellular amphoterin are depicted in relation to the HMG box structures in Fig. 5B. The highly acidic C-terminus of amphoterin has been found to play an important role in the nuclear functions of amphoterin (Aizawa et al., 1994). In addition, it was recently reported that amphoterin contains a sequence motif (residues 12-27) homologous to amyloid-β peptide, and this peptide is capable of forming amyloid fibrils and binding to amyloid-β peptide in vitro (Kallijarvi et al., 2001).

Amphoterin is an extremely well conserved protein between species. For example, human, rodent and bovine amphoterin sequences are

>95% identical at amino acid level. Human and rodent amphoterin sequences are identical at 214 of 216 amino acids (>99%). As the differences between human and rodent sequences are glutamate/aspartate changes in the acidic C-

terminal tail of amphoterin, the human, mouse and rat proteins are virtually identical.

The human amphoterin gene has been mapped to chromosome 13q12 (chromosome 5 in mouse) and contains five exons and four introns (Ferrari et al., 1994; Gariboldi et al., 1995;

Bentley et al., 2001; Blake et al., 2002). At least seven retrotransposed pseudogenes have been found in the mouse genome (Gariboldi et al., 1995). Amphoterin gene has a very strong TATA- less promoter with a maximum activity 18-fold higher than the SV40 promoter (Lum and Lee, 2001). The promoter contains several binding sites for general transcription factors such as AP1 and Sp1. However, the promoter region of amphoterin gene contains a silencer element that holds amphoterin expression at basal levels under normal circumstances (Lum and Lee, 2001).

Figure 5. Secondary structure of amphoterin. A. Schematic illustration of amphoterin domain structure. The HMG boxes 1 and 2 are comprised of three α-helices each. The hypothetical “seventh helix” is shown between the helix 6 of the HMG box 2 and the Glu/Asp-tail. B. NMR solution structures of the HMG box 1 (Gly2-Glu84) and HMG box 2 (Phe89-Lys165) of amphoterin. Surfaces illustrate important sequences in the function of extracellular amphoterin.

2.3.1.2. Expression and secretion of amphoterin

Amphoterin expression can be detected at some level in almost all cell and tissue types and due to a strong promoter it can be expressed at very high levels (Lum and Lee, 2001). In addition, amphoterin expression is under strict developmental regulation. In rat brain the expression of amphoterin is high during the embryonic period and decreases after birth (Rauvala and Pihlaskari, 1987; Merenmies et al., 1991). The expression pattern of amphoterin in the developing rat nervous system appears to be rather widespread. Immunohistochemical studies have shown amphoterin to be expressed in the late embryonic period (E14-E21) in dividing and migrating neurons of cortical plate and subplate and in the ventricular zone of cerebral cortex.

During the perinatal period (P1-P15) amphoterin is expressed in neurons of external and internal granule cell layers of cerebellum (Nair et al., 1998; Zhao et al., 2000; Zhao et al., 2000; Chou et al., 2001). Expression of amphoterin in the hippocampus has also been reported (Hori et al., 1995). In E13 spinal cord, amphoterin is expressed in the dorsal root entry zone, the mantle layer of the ventral horn, the neuroepithelium surrounding the central canal, and in the dorsal root ganglia (Milev et al., 1998). In adult retina amphoterin immunoreactivity has been detected in the inner nuclear layer and in the outer plexiform layer located between the inner and outer nuclear layers (Milev et al., 1998). In general, amphoterin levels in the adult central nervous system are low.

In addition to neurons, amphoterin is also abundant in oligodendrocytes and Schwann cells prior to myelinogenesis during late embryonic development (Daston and Ratner, 1991; Daston and Ratner, 1994). In the peripheral nervous system, amphoterin expression in neurons persists into adulthood whereas the expression in Schwann cells is downregulated upon neuronal contact (Daston and Ratner, 1991). In the developing Xenopus nervous system abundant expression of HMG-X, an amphoterin homologue, has also been reported (Kinoshita et al., 1994).

In general, the abundance of amphoterin seems to correlate with an undifferentiated cell stage and early maturation. However, in the vascular system amphoterin is expressed in several mature cell types. Platelets contain amphoterin that is exported to the cell surface during platelet activation (Rouhiainen et al., 2000).

Amphoterin is highly expressed in mononuclear phagocytes and macrophages, and is secreted in response to proinflammatory cytokines (Wang et al., 1999; Andersson et al., 2000). Amphoterin

is also detectable in normal human serum (0.2 ng/ml) but not in plasma (Rouhiainen et al., 2000).

Increased amounts of amphoterin have been found in serum of patients suffering from systemic inflammation such as sepsis (Wang et al., 1999).

Interestingly, amphoterin has been suggested to function in the interface of neuroendocrine and immune systems as it is secreted by pituicytes upon stimulation with proinflammatory cytokines tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) (Wang et al., 1999).

Amphoterin expression is remarkably enhanced in transformed cells. Several transformed cell lines display abundant expression of amphoterin (Parkkinen et al., 1993; Taguchi et al., 2000). Furthermore, a large variety of tumor cells have been reported to express amphoterin (Takada et al., 2001; Flohr et al., 2001; Kuniyasu et al., 2002).

Subcellular localization of amphoterin depends on the cell type and the state of the cell.

As amphoterin has been reported to be a non- histone component of chromatin, it is regularly found in the nucleus. However, the majority of amphoterin can be found diffusely distributed in the cytoplasm (Bustin and Neihart, 1979). In motile cells amphoterin becomes strongly enriched at the leading edge and extending processes of the cell (Merenmies et al., 1991; Parkkinen et al., 1993).

The mechanism how subcellular localization of amphoterin is regulated is not well understood.

However, studies in our laboratory have suggested that localization of amphoterin mRNA to the cell periphery might be a key step in bypassing nuclear transport of amphoterin. Localization of amphoterin mRNA shares similar features with β-actin mRNA localization, and these two mRNA particles are often found in cell processes together with ribosomes suggesting local mode of translation (Punnonen et al., 1999; IV; Fig.

6). Thus, mechanisms controlling localization of amphoterin mRNA might also be important regulators of secretion of amphoterin.

Although amphoterin lacks a classical secretion signal peptide, it has repeatedly been reported to be secreted independently of cell damage (Rauvala and Pihlaskari, 1987; Rauvala et al., 1988; Passalacqua et al., 1997; Passalacqua et al., 1998; Wang et al., 1999). The classical secretory pathway involving endoplasmic reticulum (ER) and the Golgi apparatus is not involved in the secretion of amphoterin but intracellular increase of Ca²⁺ and protein kinase C may be required (Passalacqua et al., 1997). Amphoterin has been reported to be secreted in response to various stimuli. Secretion of amphoterin accompanies process extension and at least laminin directly induces secretion of amphoterin (IV). In addition, proinflammatory

cytokines TNF-α and IL-1β have been reported to induce secretion of amphoterin in monocytes, macrophages and pituicytes (Wang et al., 1999;

Wang et al., 1999).

In addition to regulated secretion, amphoterin can also be released from necrotic or damaged cells. It was recently reported that necrotic but not apoptotic cells release amphoterin (HMGB1) promoting local inflammation near necrotic cells (Scaffidi et al., 2002).

2.3.2. Amphoterin binding proteins and carbohydrates

Amphoterin is a rather sticky protein likely due to its highly charged nature. A wide range of different molecules has been shown to interact with amphoterin. These include certain conformations of DNA, various carbohydrate epitopes and proteins. Amphoterin-binding molecules can be divided roughly in two categories; the molecules relevant to the nuclear functions of amphoterin (discussed briefly in section 2.3.3.1) and the molecules relevant to the extracellular functions of amphoterin.

2.3.2.1. Proteoglycans and glycolipids

Amphoterin was originally isolated as a heparin-binding protein from perinatal rat brain (Rauvala and Pihlaskari, 1987). Heparin is a heterogeneous group of straight-chain highly anionic mucopolysaccharides having e.g.

anticoagulant properties. Heparin is composed mainly of uronic acid (usually α-L-iduronic acid) and α-D-glucosamine linked by α(1-4) bonds, and is polydisperse in chain length and heterogeneous in degree and type of sulphation (reviewed in (Sasisekharan and Venkataraman, 2000).

Modified heparin-type chains, heparan sulphates, are found in various proteoglycans where glycosaminoglycan chains are attached in core proteins through serine residues. In addition to heparan sulphate, proteoglycans may carry other types of glycosaminoglycans, e.g. chondroitin sulphate, dermatan sulphate and keratan sulphate (reviewed in Bovolenta and Fernaud-Espinosa, 2000). Proteoglycans are found on cell surfaces and as major constituents of the extracellular matrix. Various proteoglycans have been found to serve e.g. as modulators of neurite outgrowth (reviewed in Rauvala and Peng, 1997; Bovolenta and Fernaud-Espinosa, 2000).

Amphoterin is able to bind to syndecan- 1 (Salmivirta et al., 1992). Syndecans are transmembrane proteoglycans that are involved in regulation of cell behavior ranging from lipase

activity and anticoagulation to growth factor signalling and cell adhesion (reviewed in Woods and Couchman, 1998; Rapraeger, 2000). As amphoterin binds to syndecan-1 in a specific manner requiring heparan sulphate side chains of syndecan-1 (Salmivirta et al., 1992), it is likely that amphoterin also binds to heparan sulphate structures of syndecan-2 and syndecan-3 that are expressed in the nervous system (Carey et al., 1992; Kim et al., 1994). Syndecans can activate cell signalling pathways either as independent cell surface receptors or as coreceptors for integrins and growth factor receptors (reviewed in Rapraeger, 2000). In addition to syndecans, amphoterin can bind to chondroitin sulphate side chains of phosphacan, a splice variant containing the ectodomain of the receptor-type protein tyrosine phosphatase β/ζ (RPTPβ/ζ), with a rather high affinity (K_d≈0.3-0.8 nM)(Milev et al., 1998).

Sulphoglucuronyl carbohydrate (SGC) reacting with the monoclonal antibody HNK-1 is temporally and spatially regulated in the developing nervous system. Sulphoglucuronyl carbohydrates are expressed on several neural recognition molecules of the immunoglobulin superfamily, such as neural cell adhesion molecule NCAM, L1, Tag-1 (Axonin- 1) and contactin (F3/F11) (reviewed in Jungalwala, 1994). SGC is also expressed on certain chondroitin sulphate proteoglycans and on some sulphoglucuronyl glycolipids (SGGLs). Amphoterin has been reported to bind to glycolipids carrying the sulphoglucuronyl carbohydrate (Mohan et al., 1992; Nair and Jungalwala, 1997; Chou et al., 2001). Interestingly, SGC has been implicated in adhesion of neurons and astrocytes on laminin and in the outgrowth of neuronal and astrocytic processes (reviewed in Jungalwala, 1994).

Recently, it was reported that mice carrying a targeted mutation in the HNK-1 sulphotransferase gene, which encodes an enzyme needed in the synthesis of SGC, show no defects in cerebellar granule neuron migration in vivo and normal neurite outgrowth in response to amphoterin in vitro. However, amphoterin was shown to bind also to unsulphated glucuronyl glycolipids, although with a lower affinity, likely explaining the overall normal neuronal phenotype of these mice (Chou et al., 2002).

2.3.2.2. RAGE (Receptor for Advanced Glycation End Products)

Perhaps the most studied cell surface receptor for amphoterin is Receptor for Advanced Glycation End Products (RAGE; see section 2.4. below).

2.3.2.3. IL-1 receptors

Amphoterin has recently been found to bind to interleukin-1 (IL-1) receptors type I and II (Zetterström, 2001). Interleukin-1 refers to two molecules, IL-1α and IL-1β, which are key inflammatory cytokines activating innate and adaptive immune responses that prepare the body to deal with injury or infection. The actions of both IL-1α and IL-1β are mediated through the same receptor called IL-1 receptor type I (IL- 1RI) (reviewed in Sims, 2002). The extracellular, ligand binding portion of IL-1RI consists of three immunoglobulin-like domains and in this sense it resembles the extracellular part of RAGE. The

~200 amino acid long cytoplasmic domain of IL- 1RI shares homology with other members of the IL-1R family and the cytoplasmic domains of Toll family members (reviewed in Akira et al., 2001) comprising a novel cytoplasmic protein module called ”Toll/IL-1-receptor” (TIR) domain. IL-1 receptor type II is similar to IL-1RI but has a very short cytoplasmic domain incapable of activating signalling pathways. IL-1RII can be released from the cell surface by a metalloproteinase to generate a soluble inhibitor of IL-1 action. IL-1 binding to its receptors requires an auxilliary factor known as IL-1R accessory protein (AcP) (reviewed in Sims, 2002). However, amphoterin binding to IL-1RI and IL-1RII does not seem to require this auxilliary protein (Zetterström, 2001).

2.3.2.4. Plasminogen and plasminogen activators

Amphoterin binds to tissue-type plasminogen activator (t-PA), urokinase-type plasminogen activator (u-PA) and plasminogen, an inactive zymogen that produces lysine-specific serine protease plasmin upon activation (Parkkinen and Rauvala, 1991; Parkkinen et al., 1993). Interaction of amphoterin with t-PA and plasminogen enhances generation of active plasmin with similar capacity as soluble fibrin (Parkkinen and Rauvala, 1991). As amphoterin is a lysine-rich protein (~20% of amino acids) it is a sensitive target for plasmin itself. Thus, amphoterin enhances its own breakdown upon contact with t-PA/plasminogen system (Parkkinen and Rauvala, 1991; Parkkinen et al., 1993). Amphoterin and t-PA colocalize to distal tips of extending processes in motile cells (Parkkinen and Rauvala, 1991).

2.3.2.5. Borna disease virus phosphoprotein p24

Recently, amphoterin was reported to bind to Borna disease virus phosphoprotein p24

(Kamitani et al., 2001). Borna disease virus (BDV) is a nonsegmented, negative stranded RNA virus, belonging to the Bornaviridae family in the Mononegavirales order (de la Torre, 1994).

BDV causes central nervous system diseases characterized by behavioral abnormalities in a wide variety of animals (reviewed in Lipkin et al., 2001; Carbone et al., 2001). A recent study showing that cerebellar damage in rodents is observed only if infection occurs before postnatal day 15 suggests that BDV interferes with early events of postnatal brain development (Rubin et al., 1999). Interestingly, BDV infection - as well as recombinant p24 alone - inhibits neurite outgrowth and migration of neural cells. Furthermore, amphoterin localization to the leading edge and growth cones of neuronal cells was found to be disturbed in BDV-infected cells likely due to the interaction of cytoplasmic p24 with amphoterin in infected cells (Kamitani et al., 2001).

2.3.3. Functions of amphoterin

2.3.3.1. Nuclear functions of amphoterin (HMGB1)

Amphoterin (HMGB1) was originally found to bind cruciform DNA, a non-double helix form of DNA, that is generated e.g. as an intermediate in genetic recombination (Bianchi et al., 1989).

In general, amphoterin and its individual HMG boxes do not display sequence specificity but bind to DNA substrates with wide minor groove, e.g. four-way junctions (Hill et al., 1999), DNA crosslinked by cisplatin drugs (Locker et al., 1995) and supercoiled DNA (Stros and Reich, 1998). In addition, the acidic C-terminus of amphoterin has been reported to be essential for the enhancement of gene expression by amphoterin (Aizawa et al., 1994). Apparently amphoterin itself is not capable of enhancing transcription but interacts with several transcription factors (e.g. Hoxand Pou proteins, steroid hormone receptors p53 andTBP), some viral proteins, and the RAG1 protein (involved in V(D)J recombination)(reviewed in Bianchi and Beltrame, 2000). Recently, amphoterin was reported to bind several unrelated transcriptional regulators by recognizing short peptides in them (Dintilhac and Bernues, 2002). Altogether, these observations suggest roles for amphoterin in DNA recombination, repair, replication and gene transcription (reviewed in Bustin, 1999).

Interestingly, mice lacking amphoterin die only a few hours after birth, and it has been suggested that this is due to a defect in transcriptional enhancement of glucocorticoid receptor (Calogero et al., 1999). However, whether amphoterin serves as a transactivator, a quasi-transcription factor or