Visitor n°

Top Used abreviations Uppsala University Project description Theoretical background Materials and methods Results ans discussion References

Characterisation of binding properties of a phage display antibody to heparin and heparan sulfate

Maarten Vanwildemeersch

	USED ABBREVIATIONS

Top Used abbreviations Uppsala University Project description Theoretical background Materials and methods Results ans discussion References

	UPPSALA UNIVERSITY

Top Used abreviations Uppsala university Project description Theoretical background Materials and methods Results ans discussion References

Uppsala university is the oldest university in the Nordic countries and contains three disciplinary domains (Arts & Social Sciences, Medicine & Pharmacy and Science & Technology) and eight faculties (Figure 1). The Department of Medical Biochemistry and Microbiology (Institutionen för medicinsk biokemi och mikrobiologi, IMBIM) is the biggest of the 11 departments of the Faculty of Medicine. This department is the merger of the former Department of Medical and Physiological Chemistry, the Department of Medical Immunology and Microbiology and the Department of Biomedical Laboratory Sciences. Besides education of medical students, biomedical students and lab technician students, several research groups are active in the Biomedical Centre, where IMBIM is located. One of these research groups, the one where I did my project, is the group of Prof. Ulf Lindahl. This group, also called the "heparin group", studies the structure, biosynthesis and biological activities of heparan sulfate and heparin. My project was situated in the sub-group of Dorothe Spillmann, my supervisor. In this group protein-heparan solfate interactions are studied.

The projects of the "heparin group" are financed by The European Commission, Uppsala University, the Swedish medical research council, the GLIBS program (Glycoconjugates in Biological Systems) sponsored by the Stiftelse för Strategisk Forskning and the Polysackarid forskning AB.

Figure 1.1 Structure of Uppsala University

	PROJECT DESCRIPTION

Top Used abreviations Uppsala University Project description Theoretical background Materials and methods Results ans discussion References

Heparan sulfate (HS) and heparin are linear sulfated and acetylated polysaccharides linked to protein cores. HS is found in virtually all tissues, at the cell surface or in the extracellular matrix (ECM). Due to complex enzymatic modifications during the biosynthesis of HS, long heterogeneous glycosaminoglycan (GAG) chains are formed. Several proteins are prone to bind selectively to specific sequences in the heterogeneous HS chain, important for many physiological processes.

By using phage display technology, it is possible to create a library of antibodies against HS. These antibodies can be used to isolate HS sequences, which can then be characterised and used as a "fisherman's hook" to catch other ligands, of which the characteristics and functions can be studied afterwards.

In my project, I focused mainly on the characterisation of a HS-antibody (EV3C3). Important steps were the affinity selection of HS sequences on immobilised phage display antibodies and the determination of the binding properties between the antibody and the HS or heparin chain, using both in-solution assays and assays with immobilised heparin or HS chains.

A scheme of the main goal and the several steps involved in creating specific HS sequences and using them to catch binding proteins is given in Figure 2.1.

Figure 2.1 The use of HS sequences to catch ligands

	THEORETICAL BACKGROUND

Top Used abreviations Uppsala University Project description Theoretical background Materials and methods Results ans discussion References

AN INTRODUCTION TO GLYCOBIOLOGY (Varki, 1999)

The essentials of contemporary molecular biology are based on the expression of biological information through translation of the DNA configuration into proteins, which form the cells that construct an organism. This theorem can thus be displayed as follows:

Although this theorem is correct and very often used, it's not complete. Two more major classes of molecules, lacking in this theorem, are indispensable to create a cell: lipids and carbohydrates. Especially carbohydrates are essential when it comes to creating multicellular organs and organisms, because they play an important role in cell-cell interactions and interactions with the extracellular matrix (ECM). Carbohydrates also mediate interactions between organisms (e.g. between host cells and bacteria, parasites or viruses (Rostand, 1997)). The main theorem can thus be extended as follows:

Regardless the importance of several polysaccharides in the clockwork of an organism, the studies of glycans lingered for a long time compared with the other major classes of biomolecules during the first part of the modern revolution in molecular biology. This is caused by the complex nature of glycans, the difficulty to sequence them and the fact that their biosynthesis could not be linked directly to the DNA-pattern. New technologies, however, made it possible to study the structure of sugar chains. This was the beginning of what Rademacher, Parekh and Dwerk in 1988 called "glycobiology". Glycobiology in its broadest sense is thus the study of structure, biosynthesis and biology of saccharides that are widely distributed in nature.

PROTEOGLYCANS AND GLYCOSAMINOGLYCANS

Proteoglycans and glycosaminoglycans

Proteoglycans (PGs) consist of a core protein with one or several glycosaminoglycans (GAGs) attached. GAGs are linear polysaccharides, assembled of a repetitive disaccharide unit. Each disaccharide contains one amino sugar (GlcNAc or GalNAc) and an uronic acid (GlcA or IdoA).

The existence of five GAGs is known by date, as shown in figure 3.1: hyaluronan (HA), chondroitin sulfate (CS), dermatan sulfate (DS), heparan sulfate/heparin (HS) and keratan sulfate (KS). This GAGs can be placed in two families: the glucosaminoglycans (HA and HS) and the galactosaminoglycans (CS and DS). KS is also a glucosaminoglycan, but it is in several ways a "false" glucosaminoglycan. KS contains neutral galactose instead of the charged

hexuronic acid, can be branched and can be N-glycosidically (KS type I) or O-glycosidically (KS type II) linked to the core protein.

Heparan sulfate (HS) and chondroitin sulfate (CS) contain a linkage sequence of four sugar units (GlcA - Gal - Gal - Xyl), which is O-glycosidically coupled to Ser or Thr in the core protein.

Different core proteins are known by date, with sizes between 14 and 580 kD (Kjellén, 1991). PGs can be divided into different families, according to their core proteins, their predominant GAG modification or according to location and function.

Figure 3.1 Classification of GAGs

Heparan sulfate and heparin

The history of heparin (Rodén, 1989)

In 1918, Howell and Holt described a new anticoagulant, called heparin. Because of the isolating procedure and the high amounts of phosphorus found in the first preparations, it was considered to be a phospholipid. In 1925 however, Howell performed a colorometric assay (Molish’ a-naphtol method) which indicated the presence of carbohydrate in purified heparin and in 1935 Jorpes proved that heparin was not a phospholipid, but a carbohydrate. The identification of the monosaccharides in heparin started in 1936, but lasted until 1964. During the same period, also other pieces of the puzzle were revealed, such as the sulfate and acetyl substituents, and in 1955 Foster and Huggard published a picture of heparin which started the discussion about the fine structure of heparin. Now we know that not all the assumptions made by Foster and Huggard were correct, but it lasted until the 1980’s until reliable information about the structure of heparin was available.

Structure and biosynthesis of heparin and heparan sulfate

Heparin and heparan sulfate are GAGs consisting of consecutively GlcNAc and GlcA or IdoA. During the chain growth, alternately GlcNAc and GlcA are attached to the non-reducing end of the chain by respectively a-GlcNAc Transferase II and a-GlcA Transferase II (Esko, 1999). During the heparin and heparan sulfate polymerisation, several enzymes modify the chain by epimerising GlcA and sulfating part of the monosaccharides (Figure 3.2 A). This modifications tend to occur in clusters, as a consequence of the substrate requirements of the modifying enzymes. This results in different domains in the HS/heparin chain: the highly sulfated NS-regions (presented as a zig-zag line in Figure 3.2 B), the barely modified NA-regions (presented as a straight line in Figure 3.2 B) and the mixed NA/NS-regions.

Figure 3.2 Polymer-modification reactions involved in the biosynthesis of heparin and HS (Lindahl, 1994).

Considering the biosynthetic constrains, the disaccharide species given in Figure 3.3 can be expected. Species between squared brackets have not been identified, but are theoretically tolerated.

Figure 3.3 Scheme illustrating disaccharide sequences identified in heparin and HS.

Modified from (Lindahl, 1989).

The difference between heparin and heparan sulfate(Esko, 1999)

Heparin and heparan sulfate are produced by the same enzymes with the major difference that heparin is the most extensively modified form. It therefore is characterised by a high degree of epimerised and sulfated saccharides. A list of some important differences is shown in Table 3.1.

Table 3.1: Major differences between heparin and heparan sulfate (Esko, 1999)

Functions of proteoglycans

GAG-binding proteins

Most of the physiological functions of PGs known to date are very closely related to the ability of GAG chains to bind selectively to proteins. The most important form of interaction is electrostatic interaction between the negatively charged GAG chains and the positively charged proteins, but other interactions such as hydrogen binding and hydrophobic interactions also occur in nature (Lindahl, 1994) The most famous interaction of a GAG with a protein is the binding of heparin to antithrombin. To be able to bind to antithrombin, heparin or HS chains require a specific pentasaccharide, as given in Figure 3.4 (Lindahl, 1994). This was also the first known protein-GAG interaction with a physiological relevance (i.e. preventing the clogging of blood). By date, more then 100 GAG-binding proteins are known, of which the most important ones are given in Table 3.2. The binding of proteins to GAGs results in protein immobilisation, regulation of the enzymatic activity, binding of ligands to their receptors or protection of proteins against degradation. This interactions are of a huge importance in haemostasis, lipid transport, lipid adsorption, cell growth, cell division and cell migration. In some of the interactions the core proteins are also important, in others the proteins just interact with the GAG. Co-operation between the core protein and the GAG chain can occur in several ways. The most simple way of co-operation is that the core protein act as a frame to hold and immobilise the GAGs. The core protein can also act as an anchor to attach the GAGs to the cell surface or to another macromolecule. A third possibility is that the core protein has a physiological function. This means that the core protein is essential to get a certain reaction between the GAG chain and the binding protein. It is for example known that the core protein of HSPGs mediates the interaction of the PG with components of the ECM, such as fibronectin, several collagens and thrombospondin (Kjellén, 1991)

Table 3.2: Some examples of GAG-binding proteins and their biological activity (Esko, 1999)

Figure 3.4 The binding region for antithrombin in heparin. Modified from (Kjellén, 1991)

The role of PGs in microbial adherence

Beside their function as an attachment site for cellular adhesion molecules and matrix molecules, in order to create stable tissue structures and extracellular matrices, cell surface PGs are also used by micro-organisms as a primary attachment site to the infected cell. Although it is known that cell surface PGs act as adhesion receptors, their precise role in invasion is indistinct. Proteoglycans mediate probably in an early stage of adherence, because heparin can block initial interaction and dislodge freshly bound organisms (Rostand, 1997). The most important interactions between micro-organisms and proteoglycans in eukaryotic cells are given in Table 3.3.

Table 3.3: Micro-organisms bind proteoglycan receptors on eukaryotic cells (Rostand, 1997)

Turnover and degradation of PGs (Freeze, 1999)

Turnover of proteoglycans can occur in two ways, either by shedding from the cell surface or by endocytosis. Shedding from the cell surface happens through proteolysis of the core protein, resulting in free extracellular glycosaminoglycans (Bernfield, 1999). The endocytic way takes place in a highly organised manner. After internalisation, the core proteins are degraded. This is followed by a partial enzymatic cleavage of the GAG chains, which results in fragments of more or less 10 kD. This is important because these fragments are then digested by exoglycosidases, and it would take a long time to digest larger molecules. In most cases, the substitutes (SO3-) have to be removed first, because most exogycosidases do not degrade a substituted sugar.

STUDY OF HEPARAN SULFATE STRUCTURE AND FUNCTIONS

Background

Before describing the main techniques used in HS studies, it might be interesting to explain the goal of these studies. Why studying the HS structure?

The goal of all the experiments is to get specific HS sequences that can act as a tool to catch ligands. By using antibodies, specific antibody binding HS sequences are selected which then can be used as "fisherman’s hook" to catch other ligands. Once a ligand is catched to a certain sequence, one can try to find an answer to the final questions: What is the binding protein, what is its function and what is the role of the HS binding in its function?

In this chapter, both the main techniques used to make the samples I was working with as the main techniques I used to investigate my samples are discribed. The techniques I didn’t use myself are only given to make the picture complete. This means that they are not described into detail. The theoretical background behind the methods I did use is only given in function of the practical work. It is centainly not a complete description.

Preparation of HS and modified heparin fragments

Before HS or modified heparin fragments can be used, they have to be prepared. Figure 3.5 gives an overview of the several steps used to produce this samples.

Heparin and HS can be obtained from both tissues and cell cultures. If one needs a lot of material, tissue should be used as a source. If one wants to get metabolical radiolabeling, a cell culture can be used. The properties of the collected HS species depend on the tissues used as a source. HS from bovine lung for example is different from HS from porcine liver. Cell cultures on the other hand seem to have a more uniform HS pattern.

From the cell culture or the tissue, the HS or heparin can be retrieved by quite a few extraction steps. The three main steps are removing proteins and glycoproteins, removing the core protein and removing the other glycans. Each of this steps is followed by a run on a DEAE-column. This is a beaded cellulose ion exchange column which can be used over a wide range of molecular weights. Due to its rather weak charge, it is possible to separate long and highly sulfated chains. If one wants to get rid of CS, one can digest CS by chondroitinase and run the mixture on the DEAE column again to separate HS from the disaccharides of the CS digestion. The HS can now be characterised and used in binding assays, or one can cleave the HS chain to obtain smaller fragments.

Cleavage of HS chain can be performed in 2 major ways: chemically or enzymatically. Chemical cleavage results in a sugar change at the reducing end, where anhydromannose is formed. This structure can be reduced, so it is possible to introduce a radioactive label there. Chemical cleavage with HNO2 at pH 1.5 is selective for N-sulfated GlcN. Chemical cleavage with HNO2 at pH 4.0 is selective for GlcNH2+, so it needs deacetylation first. Enzymatical cleavage is much more selective for certain sequences, and introduces a double bound at the non-reducing end of hexuronic acid.

Figure 3.5 From tissue or cell to HS or modified heparin fragments

If heparin is used, it is possible to mimic the HS structure (except for the domains) by modifying heparin, but also HS can be modified to change the configuration. This is accomplished by removing and adding certain functional groups on the chain. The possible approaches are N-deacetylation or N-desulfation folowed by N-sulfation or N-acetylation, 2-O-desulfation and 2-O-desulfation

Once the desired fragments are produced and radiolabeled, subfractionating is possible according to size, charge and affinity. The columns used for size and charge separation are given in Table 3.4.

Table 3.4 Columns used for size and charge separation of heparin and HS fragments

Affinity selection of heparin and HS fragments

Once heparin or HS fragments of a certain size are produced it is possible to select them on their affinity to proteins. A scheme of this approach is given in Figure 3.6. The goal of affinity selection is to separate HS or heparin species that bind to a certein protein from those that don’t bind. The theory behind the most important techniques I used for affinity selection are described in this chapter.

Figure 3.6 Affinity selection of HS and heparin fragments

Phage display antibodies

Phage display antibodies are artificially made antibodies, constructed completely in vitro without using hybridoma technology or immunised animals. In this way, it is possible to make high affinity antibodies against virtually every biological molecule outside the natural host.

In nature, B lymphocytes display immunoglobulin antibodies on their surface, where they act as antigen receptors. Which antibody is displayed on the surface of a B lymphocyte depends on the arrangement of the V genes, that code the variable domain of the Ig, during the development of the B-lymphocyte. Encounters of B cells, expressing each a single type of antibody, with all the antigens present in the body, will conserve this B cell and the respective antibody.

Phage display technology imitates the immune system by rearranging the V-genes, and expressing them on the surface of a bacteriophage instead of a B-lymphocyte. This makes it possible to produce whatever combination of antibodies, also such ones that would not be produced in an organism due to tolerance of own antigens. By screening the phages to immobilised antigens (Figure 3.7), one can separate the effective antibodies from the ineffective ones. One then infects a bacterium, such as E. coli, with the phage. Because the phage contains inside the genetic material to produce the antibody on its surface, huge amounts of soluble antibody can be made by the bacterium after infection. This process results in a phage display antibody, containing a light chain interlinked to a heavy chain, attached to a phage protein (Figure 3.8).

Figure 3.7 Screening of phage display antibodies tot immobilised antigens (Hoogenboom, 1997)

Figure 3.8 Scheme of a phage display antibody

Biomolecular interaction analysis (BIA)

If light goes from one medium into another medium with a lower refractive index, the light will be partially reflected. Above a critical angle, depending on the refractive indices of both media and the wavelength of the light, light is totally reflected. This phenomenon is called total internal reflection (TIR). When TIR occurs, a so called evanescent wave penetrates into the medium with the lower refractive index. If the interface between the two media is covered with a thin metal film, and if the light is monochromatic and p-polarised (the electrical vector component is parallel to the plane of incidence), another phenomenon can occur, called surface plasmon resonance (SPR). This is caused by free electron clouds in the metal film, which can resonate with the evanescent wave. If this happens, a minimum can be seen in the intensity of the reflected light beam at a certain angle, as showed in figure 3.9 (nsurf is the refractive index of the bulk solution near the sensor surface). The angle of resonance changes clearly if the refractive index near the metal film changes. This makes it possible to use SPR in a biosensor. If a ligand is immobilised on the sensor surface (i.e. the metal film), binding of a soluble ligate from the bulk solution to the ligand will change the refractive index near the surface and change the resonance angle.

Figure 3.9 Principle of an SPR biosensor (Schuck, 1997)

In the BiaCore®, sensor chips are used to apply SPR in a device for real-time BIA. Ligands, which don’t have to be can be bound to the chip surface (as showed in figure 3.10 for biotinilated ligands), and the sensor chips can be replaced quite easily. Once a chip is made with a certain ligand, it can be used to examine several ligates.

For the Biacore®, several chips are available with different properties. The basic principle, however, remains the same, but the immobilisation chemistry differs.

Figure 3.10 Immobilisation of a biotinilated ligand on a Biacore® sensor chip (modified from BIAapplications handbook)

Purification of antibodies on a Protein A Sepharose column

Protein A is an outer coat protein of Staphylococcus aureus and bind strongly with the intervariant region of immunoglobulins. Protein A also binds the phage display antibodies we used. If one immobilises protein A on a matrix (for example cross-linked agarose 4%, as in Pharmacia’s Protein A Sepharose), one can make a gel that binds specifically to the antibodies. If one runs a mixture of proteins through the gel, only the antibodies will remain in the gel. They can be eluted by lowering the pH after several washing steps. In this way, it is possible to purify the antibodies.

If one doesn’t elute the antibodies, the gel can be used as an affinity column. This is described in 3.3.3.4.

Affinity chromatography

As shown in figure 3.11, affinity chromatography requires 3 basic tools. First of all, a matrix is needed, to which the ligand (L) can be bound (for example Protein A Sepharose if an antibody is used as ligand). This can be packed into a column. If one let a solution which contains the sample flow through the column, only the samples (S) with a high affinity for the ligand will stay in the column. By eluting the samples, for example by increasing the salt concentration, the samples are separated based on their affinity. This method is often used to purify biomolecules, such as antibodies, receptors, DNA-binding proteins, etc.

Figure 3.11 Principle of affinity chromatography (modified from Pharmacia handbook)

Sequencing

Once a certain HS or heparin fragment is selected based on its affinity, it can be sequenced, in order to find out which saccharide sequences are responsible for the binding of the fragment to a certain protein. By date, several methods can be used, such as gel-based sequencing, mass spectrometry sequencing and sequencing of radiolabeled HS.

We used the last technique, which is based on partial cleavage of HS. By adding specific exoenzymes to the the mixture of the cleavage products, some structures are digested. By comparing the results of different digestions, one can determine the sequence.

Theorectical background of other used techniques

SDS-PAGE by Laemmli

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis, commonly known as SDS-PAGE, is one of the most used methods for protein analysis.

Before the samples are applied, they are denaturated by cooking them with SDS in the presence of a reducing agent, such as 2-mercaptoethanol or dithiothreitol (DTT). All the proteins then have a rod-like shape and a uniform charge to mass ratio due to the SDS surrounding the proteins.

SDS-PAGE is a discontinuous system. This means that the gel buffers have a composition different from the electrode buffer or running buffer, which has a higher pH and conductivity. The samples first enter the large pore stacking gel, in which they are concentrated into small bands. The lower gel or separation gel has smaller pores and a higher pH. Here the proteins are separated by molecular weight.

The pore size of the gel is by convention given in %T. This is the weight percentage of the total acrylamide monomer including the crosslinker (bis-acrylamide) in the gel. After crosslinking the polyacrylamide, a network of pores is created in which the proteins can be separated (Figure 3.12).

Figure 3.12 Model of polyacrylamide gel

Native PAGE

Native PAGE is based on the same principle as SDS-PAGE, but the buffers don’t contain SDS or other denaturating substances. In this way, proteins are not denaturated, which makes it possible to analyse their native structure.

	MATERIALS AND METHODS

Top Used abreviations Uppsala University Project description Theoretical background Materials and methods Results ans discussion References

	RESULTS AND DISCUSSION

Top Used abreviations Uppsala University Project description Theoretical background Materials and methods Results ans discussion References

	REFERENCES

Top Used abreviations Uppsala University Project description Theoretical background Materials and methods Results ans discussion References