Protein Interaction Studies with Low Molecular Weight Ligands: Applications for Drug Discovery, Basic Research and Diagnostic Tool Design

(1)

UNIVERSITATISACTA UPSALIENSIS

Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 1007

Protein Interaction Studies with Low Molecular Weight Ligands

Applications for Drug Discovery, Basic Research and Diagnostic Tool Design

TONY CHRISTOPEIT

ISSN 1651-6214 ISBN 978-91-554-8566-5

(2)

Dissertation presented at Uppsala University to be publicly examined in B21, BMC, Husargatan 3, Uppsala, Thursday, February 14, 2013 at 13:15 for the degree of Doctor of Philosophy. The examination will be conducted in English.

Abstract

Christopeit, T. 2013. Protein Interaction Studies with Low Molecular Weight Ligands:

Applications for Drug Discovery, Basic Research and Diagnostic Tool Design. Acta Universitatis Upsaliensis. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 1007. 34 pp. Uppsala. ISBN 978-91-554-8566-5.

In this thesis, the interactions between different proteins and small ligands were characterized by surface plasmon resonance spectroscopy (SPR) and fluorescence resonance energy transfer (FRET) based assays.

For the C-reactive protein (CRP), a new type of artificial binder was identified which allows designing diagnostic assays superior to commonly used standard assays. Furthermore, an interaction study with the endogenous ligand phosphocholine revealed the importance of the avidity of pentameric CRP for the distinction of different types of lipid membranes. The interaction study with calcium showed how SPR based assays can be used to study ion-protein interactions despite the low atomic weight of ions.

The transmembrane protease BACE1, an important drug target for Alzheimer’s disease, was immobilized to an SPR biosensor surface and embedded into a lipid membrane. An interaction study with a set of known BACE1 inhibitors showed that the transmembrane region has only minor effects on the interactions. Furthermore the pH-dependencies of the interactions were investigated and revealed new important conclusions for inhibitor design. Computer aided modelling showed that the protonation state of the aspartic dyad is dependent on the interacting inhibitor which offers new perspectives for in silico screenings.

The SPR assay developed for BACE1 was adapted to a more complex membrane protein, the pentameric β3 GABAA receptor. The assay allowed the pharmacological characterisation for histaminergic and GABAergic ligands and gave further evidence for cross-talk between the two signal transduction pathways. This study shows that the immobilisation method used for BACE1 and the ß3 GABAA receptor has the potential to become a standard method for handling membrane proteins.

The identification of new drug leads from natural sources is a common strategy for drug discovery. A combination of SPR and FRET based activity assays were explored to increase the efficiency of this process. For HIV-1 protease, secreted aspartic protease (SAP) 1, 2 and 3 extracts from a marine vertebrate were identified containing potent inhibitors which interacted with the active site of the enzymes.

The studies in this thesis show that the investigation of protein interactions is crucial for understanding protein functions and can help to develop novel drugs for the treatment of different diseases.

Tony Christopeit, Uppsala University, Department of Chemistry - BMC, Box 576, SE-751 23 Uppsala, Sweden.

urn:nbn:se:uu:diva-188328 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-188328)

(3)

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Tegler, L.T., Nonglaton, G., Büttner, F., Caldwell, K., Christopeit, T., Danielson, U.H., Fromell, K., Gossas, T., Larsson, A., Longati, P., Norberg, T., Ramapanicker, R., Rydberg, J. and Baltzer, L.

(2011) Powerful protein binders from designed polypeptides and small organic molecules - a general concept for protein recogni- tion. Angew. Chem. Int. Ed. 50(8), 1823-7

II Christopeit, T., Gossas, T., and Danielson, U. H. (2009) Characteri- zation of Ca²⁺ and phosphocholine interactions with C-reactive protein using a surface plasmon resonance biosensor. Anal. Bio- chem. 391, 39-44.

III Dominguez, J. L., Christopeit, T., Villaverde, M. C., Gossas, T., Otero, J. M., Nystrom, S., Baraznenok, V., Lindstrom, E., Danielson, U. H., and Sussman, F. (2010) Effect of the protonation state of the titratable residues on the inhibitor affinity to BACE-1. Bio- chemistry 49, 7255-7263.

IV Christopeit, T., Stenberg, G., Gossas, T., Nyström, S., Baraznenok, V., Lindström, E. and Danielson, U.H. (2011) A surface plasmon based biosensor with full length BACE1 in a reconstituted mem- brane. Anal. Biochem. 414(1), 14-22

V Seeger, C., Christopeit, T., Fuchs, K., Grote, K., Sieghart, W. and Danielson, U.H. (2012) Histaminergic pharmacology of homo- oligomeric β3 γ-aminobutyric acid type A receptors character- ized by surface plasmon resonance biosensor technology. Bio- chem. Pharmacol. 84(3), 341-51

VI Christopeit, T., Øverbø, K., Danielson, U.H. and Nilsen, I.W. Effi- cient screening of marine extracts for protease inhibitors by combining surface plasmon resonance spectroscopy and FRET based activity assays. (Manuscript)

Reprints were made with permission from the respective publishers.

(4)

(5)

Abbreviations

AIDS Acquired immunodeficiency syndrome

BACE1 β-site amyloid precursor protein cleaving enzyme 1

CRP C-reactive protein

C. albicans Candida albicans

DNA Deoxyribonucleic acid

E. coli Escherichia coli

ELISA Enzyme-linked immunosorbent assay

FRET Förster (fluorescence) resonance energy transfer

GABA γ−Aminobutyric acid

GABA_A γ−Aminobutyric acid type A

HCMV Human cytomegalovirus

HPLC-MS High performance liquid chromatography-mass spec- trometry

HIV Human immunodeficiency virus

HIV-1 Human immunodeficiency virus type 1 IC₅₀ Half maximal inhibitory concentration ka Association rate constant

k_d Dissociation rate constant

KD Equilibrium dissociation constant

kDa Kilodalton NMR Nuclear magnetic resonance

PDB ID Protein Data Bank identification code

RNA Ribonucleic acid

SAP Secreted aspartic protease

SPR Surface plasmon resonance spectroscopy

Amino acids are referred to by three letter abbreviations as recommended by IUPAC and IUBMB.¹

(8)

(9)

Introduction

Since the beginning of the 19^th century cells have been considered to be the basic unit of all living organisms and a main goal of scientist has been to understand their function. This goal is strongly connected to the understanding of proteins, the “molecular machines” of a cell. Proteins participate in all processes of living cells and interact non-covalently with many different types of macromolecules. For example during translation and transcription they interact with DNA and RNA. Interactions with other proteins such as receptors are essential for signal transduction or formation of macromolecu- lar complexes, like ion-channels or the proteasome. Enzymes, for example, interact with their substrates to carry out chemical reaction necessary for the catabolism and anabolism. The interaction with metal ions, e.g. calcium or magnesium, is a common way to induce conformational changes and thereby influence or change the activity of proteins. Furthermore, the pathological mechanisms of many diseases are based on altered protein interactions, e.g.

in autoimmune diseases.² Hence understanding protein interactions is a key to understanding how cells work.

Another motivation for studying protein interactions is the development of new drugs. Most known drugs interact with proteins, so called drug targets, and thereby affect their function. For example, the inhibition of virus replication is often achieved by inhibiting essential enzymes of the virus assembly. Many psychoactive drugs, like antidepressant bind with high affinity to ion channels and receptors in the brain and thereby modify their performance.³ However, understanding the interactions between drugs and their targets is crucial for designing novel small molecules suitable as drugs.

Forces in protein interactions

Reversible protein interactions are based on the complementary match between the interaction partners and non-covalent forces primarily hydrogen bonds, hydrophobic effects and ionic interactions.

Hydrogen bonds are formed when a hydrogen atom is shared between two electronegative atoms, the hydrogen bond donor and the hydrogen bond acceptor (Figure 1). Their importance for protein interactions is immense since many amino acid side chains as well as the peptide bond in the protein backbone, are able to act as hydrogen bond acceptors and donors. Further-

(10)

more, many drugs contain groups that are able to participate in hydrogen bonds and in this way stabilize interactions with proteins.^{2, 4}

The hydrophobic effect is a result of the tendency of nonpolar molecules to form aggregates in aqueous solutions. This is mainly caused by entropic effects since nonpolar compounds disturb the network of hydrogen bonds between water molecules. Several amino acids, like phenylalanine, leucine or valine have a nonpolar character and can contribute in the formation of hydrophobic areas at protein surfaces. The interactions of such areas can significantly contribute to the binding affinity. This is also reflected in the structure of many enzyme inhibitors which have hydrophobic groups in dis- tinct positions of their structure (Figure 1).^{2, 4}

Figure 1. Left: General structure of a hydrogen bond. Right: Example of hydrogen bond donors, acceptors (circles) and hydrophobic groups (arcs) in the structure of Fluoxetine an antidepressant and inhibitor of the serotonin transporter.

Ionic interactions are formed through the attraction of opposite charges. This type of interaction occurs mainly between ions, but can also occur between atoms with partial charges induced by polarization. Several amino acids have charged or partial charged groups and can therefore participate in ion interactions. Furthermore, this is a common way for proteins to incorporate ions like magnesium or calcium into their structure.^{2, 5}

All these non-covalent forces are weak compared to covalent bonds, but together they can produce interactions with binding affinities in a pM range like the biotin-streptavidin interaction or antibody-antigen interactions.

Basic protein interaction kinetics

The non-covalent and reversible interaction of a protein (P) and an interaction partner (L) can be described by a simple equilibrium expression:

(11)

Scheme 1 The interaction can be described by the association rate constant, k_a and the dissociation rate constant, kd which together define the equilibrium dissocia- tion constant K_D:

ܭ_ୈൌ^௞_௞^ౚ

౗ ൌ^{ሾ௉ሿሾ௅ሿ}_ሾ௉௅ሿ (1)

It is worth noting that a lower value for KD corresponds to a higher affinity, and that K_D can be seen as the concentration of L which results in 50% of the protein to be in complex, providing the system is in the equilibrium.⁵

Methods to study protein interactions

There are many different methods to study protein interactions. A common way is to use purified proteins in co-precipitation based methods such as pull down assays or immunoprecipitation assays. Other methods are yeast-two hybrid or phage display, which are able to determine protein-protein interactions as well as protein-DNA interactions. To gain structural information about an interaction often X-ray crystallography or NMR spectroscopy are used. However, each method has their own strengths and weaknesses and to present them all here is beyond the scope of this introduction. Since this study focused on surface plasmon resonance spectroscopy (SPR) and Förster (fluorescence) resonance energy transfer (FRET) based enzyme activity assays only these methods are presented in detail.

FRET based enzyme activity assays

Measuring the activity of an enzyme can reveal important information about the interaction with its substrates or an inhibitor. Especially for the investigation of proteases, FRET based activity assays are widely used.⁶ Substrates for FRET assays normally consist of a peptide with an excitable fluorescent group and a quenching group covalently attached to the ends of the peptide (Figure 2). On excitation, the energy is transferred from the fluorescent to the quenching group which emits the energy as light with a special wave- length. The cleavage of the FRET substrate separates the two groups and hence changes the fluorescent properties of the substrate. This allows a time resolved measurement of product formation. In this way, the kinetic parameters of an enzyme can be determined and the influence of inhibitors can be investigated.^{5, 7}

P + L ka PL

kd

(12)

Figure 2. Schematic structure of a FRET substrate with two fluorescent groups.

Cleavage by a protease separates the groups and changes the fluorescent which allows the reaction to be followed in real time.

Surface plasmon resonance spectroscopy (SPR)

SPR is a powerful technology for the study of dynamic biomolecular interactions in real-time. It provides important information about the affinity, kinetics, chemodynamics and thermodynamics of an interaction.

A typical SPR-biosensor consists of a gold coated glass plate with an im- mobilized dextran matrix (Figure 3). Through a microfluidic system, it is possible to apply buffer or different reagents to the dextran matrix. The optical detection system is located on the opposite side of the glass plate. It allows the measurement of changes in the refractive index which are directly proportional to changes in mass on the gold surface due to surface plasmon resonance effects. However, other effects like changes in buffer conditions or charges close to the surface, has also been observed to influence the refractive index.

Figure 3. Schematic structure of an SPR biosensor.

(13)

The first interaction partner is covalently immobilized to the dextran matrix and the second interaction partner is solubilized in buffer and applied through the flow system. Upon an interaction the mass on the sensor surface increases and the refractive index changes are recorded in a response units to time plot, referred to as sensorgram (Figure 4).⁸

Figure 4. Set of typical sensorgrams for a concentration series of a 1:1 interaction.

Equilibrium values are calculated from a single point at steady state (grey marked area). The inset shows the corresponding steady state plot. The maximum signal was 100 response units.

Sensorgrams are normally shown with an initial baseline representing the background signal from the continuous flow of buffer. During the injection of the second interaction partner, the signal increases until the interaction reaches equilibrium. After the injection is stopped, the second interaction partner dissociates from the surface and the signal changes back to the baseline. For a kinetic characterization, the second interaction partner is injected in different concentrations. A so produced set of sensorgrams as shown in Figure 4 can be used to calculate kinetic parameters like association rate constant, dissociation rate constant and equilibrium constant with help of computer aided non-linear regression analyses. Models used for such analyses can be for one as well as for several binding sites and a various number of mathematical factors can be included to compensate for effects like cooperativity, conformational changes or heterogeneous interaction partners. A different way to calculate the equilibrium dissociation constant for an inter- action is a steady state plot (Figure 4). In this case the equilibrium values are plotted against the concentration and it is possible to determine K_D through non-linear regression. Furthermore, the plot gives evidences about the number of binding sites and cooperativity.⁹

(14)

Proteins in this study

The proteins investigated in this thesis belong to different classes and are of different origins. Still all are important and established drug targets or diagnostic marker and their interactions with low molecular weight interaction partners are therefore interesting research subjects.

C-reactive protein (CRP)

CRP is a plasma protein, which consist of five identical subunits symmetri- cally arranged around a pore. Every subunit has a molecular weight of

~23 kDa and can bind two calcium ions. The binding of Ca²⁺ induces a strong conformational change and thereby enables the protein to bind to phosphocholine (Figure 5).¹⁰

Figure 5. The structure of pentameric CRP in complex with calcium (orange) and phosphocholine. The binding of calcium induces a strong conformational change where a mobile loop (yellow) moves into the protein (PDB ID: 1B09).

During the acute phase of an immune response, CRP is strongly up-regulated and it is assumed that it marks bacteria or disturbed cells for the complement system by interacting with the phosphocholine headgroups on the surface of lipid membranes. The mechanism how CRP distinguishes between different membranes is still elusive. However, due to the importance for the immune system CRP is a target for the development of novel anti-inflammatory drugs.¹¹ Furthermore, it has been shown that high CRP levels in blood are associated with cardiovascular diseases and hence it is an interesting diagnostic blood marker.^{10, 12}

(15)

HCMV protease

The human cytomegalovirus (HCMV) is a widespread virus and in healthy individuals the virus has no harmful effects. In contrast, infection of immunocompromised individuals can cause severe health problems. The virus encodes a serine protease with a molecular weight of 28 kDa. It is essential for the viral replication and is hence a potential drug target. The protein is synthesized as a precursor protein and undergoes autocatalytic activation during virus assembly. Also every monomer contains a catalytic triad consisting of one serine and two histidine residues the protease is only fully active as a homodimer since the substrate interacts with both subunits. Despite the strong interest in the HCMV protease as a drug target, so far no inhibitors are in clinical use.^{13, 14}

GABA

A

receptor

γ−Aminobutyric acid type A (GABAA) receptors are ligand-gated ion channels which are involved in many different neurological processes like modulation of sleep, depression or recovery from stroke.¹⁵ Native receptors normally consist of two α, two β and one γ subunit, which are arranged around an ion pore (Figure 6).¹⁶

Figure 6. Schematic structure of a pentameric GABAA receptor and molecular structures of ligands interacting with GABAA receptors.

Today, 19 different genes encoding for GABA_A receptors have been identified, leading to a large number of differently assembled receptors. Due to the difficulty in the recombinant expression of hetero-pentameric GABA_A receptors, homo-oligomeric receptors are often used as model systems. Although the endogenous main ligand is γ-aminobutyric acid (GABA), the receptors undergo complicated interactions with a large collection of compounds.

Many of them are clinically relevant drugs, like benzodiazepines and general anaesthetics, like propofol and etomidate. However, the whole spectrum of interaction partners for the receptors is still unknown.^{15, 17, 18}

(16)

Aspartic proteases

Aspartic proteases represent an enzyme family where all members catalyse the cleavage of a peptide bond by a mechanism involving two catalytic as- partic residues (Figure 7). All aspartic proteases have an acidic pH optimum and are inhibited by the active site inhibitor pepstatin.¹⁹ In this study the aspartic proteases β-site amyloid precursor protein cleaving enzyme 1 (BACE1), HIV-1 protease and secreted aspartic protease (SAP) 1, 2 and 3 were investigated and are therefore described in detail below.

Figure 7. Catalytic mechanism of aspartic proteases.²⁰ BACE1

BACE1 is an aspartic membrane bound protease and produces the Aβ- peptide, a main constituent of amyloid plaques and the hallmark of Alz- heimer’s disease. Hence, BACE1 is an interesting drug target and many ef- forts have been done to develop inhibitors suitable as drugs. The normal function of BACE1 is still unknown, but a participation in signal transduction is suggested. Furthermore, many details about the interaction with the substrate or inhibitors are still unclear which complicates the design of novel potent inhibitors.²¹

HIV-1 protease

The human immunodeficiency virus (HIV) causes the acquired immunodeficiency syndrome (AIDS) an infection that is spread worldwide. During the last decades, treatments for AIDS have been developed, but they only slow the course of the disease and a cure is still not available. Many of the clinically used HIV drugs aim for the inhibition of the HIV-1 protease. This ho- modimeric enzyme cleaves a polyprotein and creates thereby the mature protein components for building new virus particles. Clinically used HIV-1 protease inhibitors e.g. saquinavir or atazanavir, reduces enzyme activity by binding to the active site of the enzyme.²² However, resistance against the used inhibitors emerge rapidly and the development of new drugs is necessary.²³

Secreted aspartic protease (SAP)

Candida albicans is a fungal pathogen and the major cause of candidiasis, which can be life threatening for immunocompromised patients. The genome

(17)

of C. albicans encodes for at least ten different SAPs, which differ in expres- sion patterns, substrate specificities and enzymology. During infection, SAPs are released into the surrounding tissue and help C. albicans to acquire im- portant nutrients as well as to defeat the host defence. Hence it is assumed that the inhibition of SAPs can lead to new drugs against candidiasis.^{24, 25} In this thesis the focus was on SAP1, 2 and 3, since it was shown that they are important for the virulence of C. albicans.^{26, 27}

(18)

Aim of the present investigation

The aim of the present work was to increase the knowledge about protein interactions as well as drug-drug target interactions and to improve present methods for this purpose. In detail, this included:

• Development of an assay to study the interaction of CRP with natural and artificial ligands.

• Characterization of the interaction between truncated BACE1 and different inhibitors.

• Developing a surface plasmon resonance based biosensor with membrane embedded full length BACE1 to study the interactions with different inhibitors.

• Developing a surface plasmon based biosensor for the homo-oligomeric β3 GABAA receptor and characterization of the receptor interactions with small molecules

• Increasing the efficiency of screening natural sources for novel drug leads by combining SPR and FRET based assays

(19)

Results

Interactions between CRP and artificial protein binders – A tool for designing new diagnostic assays (Paper I)

CRP is an important diagnostic marker for inflammation and cardiovascular disease and there is a clear need for detection assays able to determine CRP concentrations in body fluids like blood.^{10, 12} Antibody based assays are mainly used for this purpose today, although they have several disadvantages associated with the complex structure of the antibodies e.g. expensive pro- duction and low stability.²⁸ In this study we tested a new type of CRP-binder and its suitability for application in detection assays.

A set of 16 artificial binders was synthesized, every binder consisting of phosphocholine, a spacer molecule and a 42-residue long polypeptide chain differing in total charge and the position of phosphocholine incorporation (Figure 8). An SPR biosensor assay with CRP was established and every binder was tested in a concentration series. The interactions were reversible and Ca²⁺ dependent, but due to their complexity it was not possible to determine equilibrium constants. However, an advantage of SPR is that interactions can be followed in real-time and it was therefore possible to choose a binder with the intended binding properties directly from comparing the sensorgrams. One of the artificial binders showed an extremely slow dissoci- ation and the estimated affinity was in the low nM range (Figure 8). Hence this binder was selected for further interaction studies.

Figure 8. A: General structure of the artificial binders consisting of phosphocholine, a polypeptide chain and a chromophore. B: Sensorgram for the interaction between CRP and the selected artificial binder. The applied binder concentrations were between 0.54 and 1200 nM.

(20)

A chromophore was linked to the binder and the SPR results were confirmed by fluorescence titration. The selectivity of an antibody and the selected binder was compared by a pull-down assay from patient serums with in- creased CRP concentration. Next to CRP, also several proteins of the complement system were isolated in this pull-down assay, probably due to association with CRP. However, the binder and the antibody extracted the same protein pattern, which indicates similar selectivity. Furthermore, the applica- bility in an ELISA assay was tested and the results were comparable with commercially available assays. This study showed that the selected binder has binding properties comparable to an antibody and hence offers an interesting alternative to antibody based detection assays.

Interactions between CRP and endogenous ligands – Insights into the regulation of lipid membrane binding (Paper II)

It is clear that CRP has an important function during the activation of the complement system, but the mechanism how disturbed cells or bacteria are recognized is still elusive.^{10, 12} The established SPR biosensor assay was therefore used to investigate the interactions of CRP with different endogenous ligands.

The interaction with phosphocholine was too fast for the association and dissociation rate constants to be determined. However, from the steady state plot it was possible to calculate a K_D of 5.0 µM. With a CRP concentration in blood much lower than the determined K_D, it is obvious that the avidity of the multi binding sites in a pentameric CRP creates the high affinity necessary for the tight interaction with lipid membranes.¹¹ Furthermore, the results show how CRP distinguishes between different types of lipid membranes.

To interact with CRP, a membrane has to have certain fluidity and the phosphocholine head groups have to be arranged in special way which allows binding to several head groups at the same time.

Calcium ions have only a low mass and detecting their interactions with proteins is therefore difficult by using SPR. However, injections of different Ca²⁺-concentrations induced signals approximately 10-fold higher as ex- pected from the mass of calcium (Figure 9). It is known that the interaction between Ca²⁺ and CRP induces a strong conformational change.^{29, 30} Fur- thermore, previously reports have been shown that also conformational changes can induce SPR-signals.^{14, 31-35} Hence, it was assumed that the de- tected signals were related to the conformational change.

(21)

Figure 9. A: Sensorgrams for the interaction between CRP and Ca²⁺. The corresponding steady state plot (insert) shows a double sigmoidal shape typically for two binding sites B: Structure of the calcium binding site of CRP in complex with two Ca²⁺ (PDB ID: 1B09).

It was not possible to calculate the association or dissociation rate constants for the Ca²⁺ interactions due to the fast kinetics. The saturation plot showed a double sigmoidal concentration dependency, as expected for a protein with two binding sites. The affinities were determined with a KD of 0.03 mM for the high affinity binding site and a K_D of 5.45 mM for the low affinity binding site. The saturation level for the low affinity binding site was higher than for the high affinity binding site, which indicates that the low affinity binding site induces the major conformational change. From the three dimensional structure of CRP it is clear that each calcium ion is coordinated by five oxygen atoms from different amino acids in the protein structure and one from the phosphate group of phosphocholine (Figure 9). In binding site 1 the oxygen atoms are not as closely orientated to the ideal structure of the octahedron, which suggests a lower affinity of the binding site. Furthermore, all residues of binding site 1 are part of a mobile loop, which makes it likely that this interaction induces the conformational change. Hence it can be assumed that binding site 1 is the low affinity binding site and binding site 2 the high affinity binding site. Previous reports were not able to reveal the different affinities of the two binding sites and it was assumed that CRP circulates completely saturated in the blood stream with a Ca²⁺ concentration of 1.2 mM.^{36, 37} In contrast, the determined affinities indicating that only one out of five CRP molecules are saturated with two ions and thereby can interact with phosphocholine. Hence under these conditions a pentameric CRP cannot bind with high affinity to a lipid membrane presenting phosphocholine head groups. On the other hand in areas with high Ca²⁺ concentration,

(22)

like during inflammation, pentameric CRP has high affinity for lipid membranes due to avidity effects.

This study showed that SPR assays are suitable for the investigation of the interactions between proteins and ions as long as a strong conformational change is provided, although theoretically the mass of ions is too low for detection. This offers new possibilities since it makes direct investigation of such interactions possible without the dependency of a protein activity. In this way new insights can be gained in the relationship between the function and the structure of ion binding motifs. Furthermore, the study suggests a mechanism for how CRP distinguishes between different lipid membranes and how this process is regulated. This knowledge can help to further understand the function of CRP and to design new anti-inflammatory drugs.

Interactions of BACE1 with small molecules – Lessons for finding new drug leads (Paper III and Paper IV)

Despite the strong research interest in BACE1 as a drug target for the Alz- heimer’s disease, so far only few inhibitors have reached clinical trial.³⁸ A reason for this could be that commonly used assays are suboptimal for identification and optimization of new drug leads. The aim of this study was to improve such assays and to increase our basic understanding of BACE1 as a drug target.

Comparing full length and truncated BACE1

Membrane proteins are an important class of proteins and include many interesting drug targets.³⁹ Due to their hydrophobic character, handling can be difficult and standard biochemical methods can often not be applied. A common way to overcome such problems is to remove the transmembrane region and to work with the truncated protein although the influence on the structure is often unclear. Since this strategy has also been applied to BACE1, we investigated if the truncated protein is a suitable model enzyme to identify new potent drug leads for full length BACE1.⁴⁰

His-tagged full length BACE1 was expressed in an insect cell line and the proteolytic activity was confirmed by a FRET based activity assay with a peptide derived from the natural substrate. For immobilization to the SPR sensor, BACE1 was directly caught from the cell lysate by an immobilized anti-polyHistidine antibody and a solubilized lipid mixture was used to re- build the membrane. In this way it was possible to immobilize the enzyme with the transmembrane region embedded into the lipid membrane (Figure 10). Truncated BACE was expressed and purified from E. coli and covalent- ly attached to the dextran matrix of a SPR sensor by amine coupling.

(23)

The interaction of full length and truncated BACE1 with a set of known BACE1 inhibitors was studied at pH 4.5 and 7.4. It was possible to charac- terize the interactions by determining K_D as well as association and dissociation rate constants. A comparison revealed no major differences in interaction kinetics. Hence, it can be assumed that the transmembrane region has no or only little influence on the structure of the active site. Still the embedding of the protein into the lipid membrane had a stabilizing effect on the protein structure, since replacing of the membrane by detergent strongly reduced the stability of the surface. However, truncated BACE can be considered as an appropriate model enzyme to find potent inhibitors for endogenous BACE1.

Figure 10. Immobilisation of full length BACE1 to the dextran matrix of an SPR biosensor embedded into a lipid membrane.

Influence of inhibitor interactions on the protonation state of the catalytic dyad

BACE1 belongs to the group of aspartic proteases which have a characteris- tic aspartic dyad in the active site. During substrate binding the dyad is monoprotonated and the same state therefore was assumed to be dominant when active site inhibitors bind (Figure 7). In addition it can be expected that other residues that can occur in different charged states can be influenced by conditions and ligand binding.^41-43

The protonation states of all titratable residues for a set of BACE1 inhibitor complexes were calculated and the structures obtained were used to make an affinity ranking at pH 4.5 and 7.4. The ranking reflected the experimental determined affinities well, which validates the model and shows that it is useful for the computer aided design of new inhibitors. Furthermore, the calculated titration curves for the aspartic dyad revealed that the protonation state is strongly influenced by the bound inhibitor. The interaction with some

(24)

inhibitors induced a deprotonated state of the aspartic dyad, whereas other induces a complete ionisation. So far, in silico screenings have been per- formed under the assumption that the aspartic dyad is monoprotonated during interactions. Our results show that it is a strong advantage to include several BACE1 structures differing in the protonation state into a screening.

Importance of the pH-dependency for the pharmacokinetics of BACE1 inhibitors

A known problem with BACE1 inhibitors is that compounds showing promising result in enzyme based assays often fail to show a similar potency in cell based assays.^{38, 44} By comparing the enzyme based inhibitor affinities from the previous experiments with IC₅₀ values from cell based assays, we noticed that inhibitors with high affinity at both pH values also showed good results in the cell based assays. During intracellular trafficking, BACE1 is transported from the cell membrane into different cell compartments.²¹ An explanation for the observed correlation could be that only inhibitors with high affinity at the neutral environment of the cell membrane get transported together with the protein into the acid endocytic vesicle, where BACE cleaves its substrate. However, it seems that BACE1 inhibitors with high affinities at acid and neutral pH have smaller chance to fail later on in the drug discovery process.

Investigating the interaction of inhibitors with BACE1 at a neutral pH can be difficult since the enzyme has no detectable activity under these conditions and hence many assays cannot be used. In contrast, the SPR assay developed during this study allows investigating interactions independent from the activity which further increases the value of the assay for the drug development process.

Pharmacological characterization of the β3 GABA

^A

receptor (Paper V)

So far, only few assays are available to study membrane proteins by using SPR technology.^45-48 In a previous study we established an assay for BACE1, a protease with one transmembrane region (Paper IV). The aim of this study was to adopt the assay to a more complex membrane protein containing several transmembrane regions.

A His-tagged homo-oligomeric β3 GABAA receptor was expressed with the help of an insect cell line. The receptor was extracted by using detergent solubilisation and kept in detergent throughout the interaction analyses. Im- mobilization to the SPR sensor was successfully carried out by affinity capture with an anti-PolyHistidine antibody. The interactions with several GA-

(25)

BAergic ligands were investigated and the results were in agreement with the literature data showing that the receptor was functionally immobilized.^49,

50 Replacing the detergent by embedding the receptor into a lipid membrane did not show any improvement of the signal intensity and hence was not necessary as for BACE1. The histaminergic pharmacology of GABAA receptors has been demonstrated before and also the homo-oligomeric β3 GABAA

receptor interacted with histamine. Although the signal intensity was low due to the low molecular weight of histamine it was possible to determine K_D which was in a physiological relevant range and similar to the affinity of other histaminergic receptors (Figure 11).^{51, 52}

Figure 11. Sensorgram (left) and steady state plot (right) for the interaction between histamine and the β3 GABA^A receptor. Histamine was injected in a concentration series from 8 to 1000 µM. The calculated affinity constant was 98 µM.

Additionally, 16 ligands specific for histamine receptors interacted with the β3 GABAA receptor. Competition experiments showed that the GABAergic and histaminergic ligands bound to different binding sites. It has been specu- lated before that there is a cross talk between the histaminergic and GA- BAergic signal transduction networks and our results support this theory.⁵²

This study is a further example of how to apply SPR technology to transmembrane proteins and shows the potential of the experimental design to become a standard method which can significantly improve the drug development process for transmembrane targets.

Screening for protease inhibitors from marine sources (Paper VI)

Screening marine sources for bioactive compounds is a common strategy and in this way many novel drug leads have been isolated during the last years.⁵³ In a first step a crude extract is typically produced from the marine source, containing a complex mixture of chemical compounds.^{54, 55} The identification of extracts containing promising inhibitors can be difficult and

(26)

often is associated with false positive hits. Hence we explored a combination of an FRET based activity assay and an SPR assay to improve this process.

An aqueous extract was prepared from Atlantic herring, containing compounds with a molecular weight below 10 kDa. Further fractionation was carried out by differential solubility in methanol and by using solid phase extraction. The success of the fractionation was verified by HPLC-MS.

FRET based activity assays were used to determine the influence on the activity of the different proteases BACE1, SAP1, SAP2, SAP3, HCMV protease and HIV-1 protease. Several extracts reduced protease activities by more than 50 %. With help of an SPR assay and competition experiments with known inhibitors, the inhibition mechanisms were elucidated. The results indicated that the inhibition of SAP1-3 and HIV-1 protease is caused by interaction with the active site of the enzymes. The extracts with promising result from the FRET and the SPR assay were chosen for further experiments. Since this is an on-going project we currently try to isolate the inhibitors from the extracts by weak affinity chromatography and SPR. However, the combination of FRET assays and SPR assays provided information about the inhibition potency and the causing inhibition mechanism and thereby helps to avoid false positive hits early during the drug discovery process from natural sources.

(27)

Sammanfattning på svenska

Proteiner deltar i nästan alla processer i en cell, exempelvis DNA- replikation, katalys av kemiska reaktioner eller signaltransduktion. Vid dessa processer interagerar proteiner med andra molekyler för att kunna utföra sina uppgifter. Många läkemedel verkar även genom interaktioner med proteiner och påverkar därmed deras funktion. Att förstå molekylära interaktioner med proteiner är därför viktigt för att förstår hur celler fungerar och för att utveckla nya läkemedel. I detta arbete undersökts interaktioner mellan proteiner och mindre interaktionspartners genom användning av ytplasmon reso- nans spektroskopi (SPR) baserade biosensorer och fluorescens baserade aktivitets mätningar.

C-reaktivt protein (CRP) är ett blodplasmaprotein som markerar bakterie och skadade cellen för immunförsvaret. Dessutom är CRP en viktig in- flammationsmarkör och mätningar av koncentrationen i blod används vanligtvis i medicinisk diagnostik. För att analysera interaktioner mellan CRP och olika ligander har proteinet kopplats till en SPR biosensoryta. Interakt- ionsstudier med den endogena liganden phosphocholin belyste hur CRP känner igen cellmembran hos bakterier och skadade celler. Dessutom stude- rades interaktioner med flera artificiella ligander som kan användas för att utveckla nya förbättrade metoder för att mäter CRP koncentrationer i blod- prov. En interaktionsstudie med kalcium visade att biosensor analyser kan användas för att mäta jon-protein interaktioner. Denna typ av metod har knappt används förut och öppnar därmed upp för nya användningsområden för biosensor baserade analyser.

BACE1 är ett transmembran protein och ett viktigt mål för läkemedel mot Alzheimers sjukdom. Hitintills har det inte vart möjligt att hitta BACE hämmare som är lämpliga att använda som läkemedel. I denna studie för- sökts därför att förbättra metoder för att söka efter nya BACE1 hämmare.

Vanligtvis använder man en trunkerad och därmed löslig variant av BACE1, däremot är det inte klart att denna variant är lämplig för att identifiera och karaktärisera hämmare. En SPR biosensormetod utvecklades för både den trunkerade varianten och den membranbundna varianten i vilket enzymet var integrerat i ett lipid membran. Analysering av växelverkan med kända BACE1 hämmare visade att interaktionen inte blir påverkad av den transmembran regionen och att den trunkerade varianten är ett bra modellsystem.

Dessutom visade det sig att hämmare med en hög affinitet for BACE1 i ne- utrala och sura pH även är bra hämmare i cellbaserade analyser.

(28)

Samma metod som används för att koppla membranbundet BACE1 till en biosensor yta används också för en GABA_A receptor, en viktig jonkanal i det centrala nervsystemet och ett mål för många läkemedel. Karakterisering av dess interaktion med små molekyler såsom histamin eller GABA gav nya insikter i GABAA receptorernas funktion i hjärna. I framtiden kan denna metod användas för att hitta nya molekyler som interagerar med GABA_A receptorn och därmed finna nya läkemedel.

Att söka efter nya bioaktiva ämnen från havsdjurs extrakt är en vanlig strategi i marin bioprospektering. En kombination av SPR baserade metoder och fluorescens baserade aktivitets mätningar undersöks för att förbättra denna process. Det var möjligt att identifiera extrakt från sill som innehåller ämnen som reducerar aktiviteten hos olika proteaser. För HIV-1 proteasen och SAP1, 2 och 3 som är läkemedel mål för olika sjukdomar, var det möj- ligt med SPR baserade kompetitions analyser visar att denna hämning är med stor sannolikhet orsakat genom interaktion med enzymets aktiva säte.

Denna avhandling visar att det är viktig att studera protein interaktioner för att förstå proteiners funktion och för att utveckla nya läkemedel.

(29)

Acknowledgements

When I came to Sweden I planned to stay only for 6 months. Now I have been here for more than five years and I did not regret a single moment. I’m thankful for this memorable time and all the great people I meet. I’m sorry that it is not possible to mention all of you here but that would definitely go beyond the scope of this thesis.

I would like to thank my great supervisor, Helena Danielson, for accept- ing me as PhD student and giving me the possibility to stay in Uppsala. I also want to thank my cosupervisor Mikael Widersten, it was nice to see that also as professor and head of the department you still had time to show up at the lunch room for “fika”. Thanks to Gun Stenberg, Bengt Mannervik, Françoise Raffalli-Mathieu and Gunnar Johansson for sharing your bio- chemical knowledge and experience.

I also want to thank Inge W. Nilsen for giving me the opportunity to work at Nofima in Tromsø and to enjoy the great nature of Northern Norway.

Thanks to all people at Nofima for the friendly working atmosphere espacially Diana Lindberg, Kersti Øverbø, Bjørnar Myrnes and Jan Arne Arnesen.

I thank all my collaborators and co-authors for the contributions to this thesis. During the last years, I also had several project students which often were a great help.

Thanks to all the present and former members of the Department of Chemistry-BMC: Angelica, the ÄKTA expert, without you the lab was dis- turbingly calm during the last year and I’m sorry that I had to refuse all your kind offers. Sofia, how much I enjoy working with you is probably best re- flected in how I defended my office space in your room (now it is all yours Angelica). You have a great organizing talent and thanks to you we had sev- eral really nice group events. Johan, without you the lab would not have been running as smoothly as it did. Matthis and Thomas, your great introduc- tion to the Biacore made my time as a PhD student much easier. Eldar, you are a quite but invaluable person for the lab. Helena N., your Biacore exper- tise were a great help. Malin, you are a great discussion partner. Göran, you had always helpful tips in case an experiment was not working. Christian, it

(30)

was great to work with you and to have someone to talk with in germane.

Cissi, you have been a great friend at work and outside of the BMC, I will always remember the first time I met you. Åsa, thanks for the skiing lessons and a lot of fun discussion and new insights. Nisse, it was nice to have someone around who understands that it is fun climbing ice walls at -15°C.

Emilia, you have a great sense of humour which was greatly missed after you left BMC. Diana, we started together the adventure to work in Tromsø and I hope it will be as much fun in the future as it was in the past. Magnus B., thanks for open up your home in Kiruna and showing the beauty of Northern Sweden. Sara S., it is a pity that you started at the lab when I was already on my way out. I wish we could have worked more together. Johan- na J., you always helped when fast help was needed and not only that you know the latest interesting rumours, you also produced some by yourself.

Rikard, Christian, Sara N., thanks for many nice “fikas”

Kicki, Mangnus L., Patrik, Cissi, Emilia, Magnus B., thanks for distracted me from work with countless numbers of beers during the last years and introducing me to the student nation culture in Uppsala.

(31)

References

1 'Iupac-Iub Joint Commission on Biochemical Nomenclature (Jcbn).

Nomenclature and Symbolism for Amino Acids and Peptides.

Recommendations 1983', Biochem J, 219 (1984), 345-73.

2 David L. Nelson, Albert L. Lehninger, and Michael M. Cox, Lehninger Principles of Biochemistry (New York: W.H. Freeman, 2008).

3 David E. Golan, Principles of Pharmacology : The Pathophysiologic Basis of Drug Therapy (Philadelphia, Pa.; London: Lippincott Williams & Wilkins, 2011).

4 Peter De Paula Julio Atkins, Physical ChemistryW. H. Freeman & Co, 2009).

5 Robert A. Copeland, Evaluation of Enzyme Inhibitors in Drug Discovery: A Guide for Medicinal Chemists and Pharmacolists (Hoboken, NJ: Wiley Interscience, 2005).

6 G. B. Kim, and Y. P. Kim, 'Analysis of Protease Activity Using Quantum Dots and Resonance Energy Transfer', Theranostics, 2 (2012), 127-38.

7 Robert Allen Copeland, Enzymes : A Practical Introduction to Structure, Mechanism, and Data Analysis (New York: Wiley, 2000).

8 M. A. Cooper, 'Optical Biosensors in Drug Discovery', Nat Rev Drug Discov, 1 (2002), 515-28.

9 Biacore AB, Biaevaluation Software Handbook. version 3.0 edn, 1997).

10 S. Black, I. Kushner, and D. Samols, 'C-Reactive Protein', J Biol Chem, 279 (2004), 48487-90.

11 M. B. Pepys, and G. M. Hirschfield, 'C-Reactive Protein: A Critical Update', J Clin Invest, 111 (2003), 1805-12.

12 J. E. Volanakis, 'Human C-Reactive Protein: Expression, Structure, and Function', Mol Immunol, 38 (2001), 189-97.

13 B. C. Holwerda, 'Herpesvirus Proteases: Targets for Novel Antiviral Drugs', Antiviral Res, 35 (1997), 1-21.

14 M. Geitmann, and U. H. Danielson, 'Studies of Substrate-Induced Conformational Changes in Human Cytomegalovirus Protease Using Optical Biosensor Technology', Anal Biochem, 332 (2004), 203-14.

15 U. Rudolph, and F. Knoflach, 'Beyond Classical Benzodiazepines: Novel Therapeutic Potential of Gabaa Receptor Subtypes', Nat Rev Drug Discov, 10 (2011), 685-97.

16 E. A. Barnard, P. Skolnick, R. W. Olsen, H. Mohler, W. Sieghart, G. Biggio, C. Braestrup, A. N. Bateson, and S. Z. Langer, 'International Union of Pharmacology. Xv. Subtypes of Gamma-Aminobutyric Acida Receptors:

Classification on the Basis of Subunit Structure and Receptor Function', Pharmacol Rev, 50 (1998), 291-313.

17 R. W. Olsen, and W. Sieghart, 'International Union of Pharmacology. Lxx.

Subtypes of Gamma-Aminobutyric Acid(a) Receptors: Classification on the Basis of Subunit Composition, Pharmacology, and Function. Update', Pharmacol Rev, 60 (2008), 243-60.

(32)

18 P. S. Miller, and T. G. Smart, 'Binding, Activation and Modulation of Cys- Loop Receptors', Trends Pharmacol Sci, 31 (2010), 161-74.

19 J. Eder, U. Hommel, F. Cumin, B. Martoglio, and B. Gerhartz, 'Aspartic Proteases in Drug Discovery', Curr Pharm Des, 13 (2007), 271-85.

20 Perry A. Frey, and Adrian D. Hegeman, Enzymatic Reaction Mechanisms (Oxford; New York: Oxford University Press, 2007).

21 C. E. Hunt, and A. J. Turner, 'Cell Biology, Regulation and Inhibition of Beta- Secretase (Bace-1)', FEBS J, 276 (2009), 1845-59.

22 A. Brik, and C. H. Wong, 'Hiv-1 Protease: Mechanism and Drug Discovery', Org Biomol Chem, 1 (2003), 5-14.

23 A. K. Ghosh, 'Harnessing Nature's Insight: Design of Aspartyl Protease Inhibitors from Treatment of Drug-Resistant Hiv to Alzheimer's Disease', J Med Chem, 52 (2009), 2163-76.

24 G. Koelsch, J. Tang, J. A. Loy, M. Monod, K. Jackson, S. I. Foundling, and X.

Lin, 'Enzymic Characteristics of Secreted Aspartic Proteases of Candida Albicans', Biochim Biophys Acta, 1480 (2000), 117-31.

25 M. Bein, M. Schaller, and H. C. Korting, 'The Secreted Aspartic Proteinases as a New Target in the Therapy of Candidiasis', Curr Drug Targets, 3 (2002), 351-7.

26 M. Schaller, M. Bein, H. C. Korting, S. Baur, G. Hamm, M. Monod, S.

Beinhauer, and B. Hube, 'The Secreted Aspartyl Proteinases Sap1 and Sap2 Cause Tissue Damage in an in Vitro Model of Vaginal Candidiasis Based on Reconstituted Human Vaginal Epithelium', Infect Immun, 71 (2003), 3227-34.

27 J. R. Naglik, S. J. Challacombe, and B. Hube, 'Candida Albicans Secreted Aspartyl Proteinases in Virulence and Pathogenesis', Microbiol Mol Biol Rev, 67 (2003), 400-28, table of contents.

28 N. Khuseyinova, A. Imhof, G. Trischler, D. Rothenbacher, W. L. Hutchinson, M. B. Pepys, and W. Koenig, 'Determination of C-Reactive Protein:

Comparison of Three High-Sensitivity Immunoassays', Clin Chem, 49 (2003), 1691-5.

29 M. A. Ramadan, A. K. Shrive, D. Holden, D. A. Myles, J. E. Volanakis, L. J.

DeLucas, and T. J. Greenhough, 'The Three-Dimensional Structure of Calcium-Depleted Human C-Reactive Protein from Perfectly Twinned Crystals', Acta Crystallogr D Biol Crystallogr, 58 (2002), 992-1001.

30 A. K. Shrive, G. M. Cheetham, D. Holden, D. A. Myles, W. G. Turnell, J. E.

Volanakis, M. B. Pepys, A. C. Bloomer, and T. J. Greenhough, 'Three Dimensional Structure of Human C-Reactive Protein', Nat Struct Biol, 3 (1996), 346-54.

31 J. E. Gestwicki, H. V. Hsieh, and J. B. Pitner, 'Using Receptor Conformational Change to Detect Low Molecular Weight Analytes by Surface Plasmon Resonance', Anal Chem, 73 (2001), 5732-7.

32 T. Flatmark, A. J. Stokka, and S. V. Berge, 'Use of Surface Plasmon Resonance for Real-Time Measurements of the Global Conformational Transition in Human Phenylalanine Hydroxylase in Response to Substrate Binding and Catalytic Activation', Anal Biochem, 294 (2001), 95-101.

33 T. Mannen, S. Yamaguchi, J. Honda, S. Sugimoto, A. Kitayama, and T.

Nagamune, 'Observation of Charge State and Conformational Change in Immobilized Protein Using Surface Plasmon Resonance Sensor', Anal Biochem, 293 (2001), 185-93.

(33)

34 T. Zako, K. Harada, T. Mannen, S. Yamaguchi, A. Kitayama, H. Ueda, and T.

Nagamune, 'Monitoring of the Refolding Process for Immobilized Firefly Luciferase with a Biosensor Based on Surface Plasmon Resonance', J Biochem, 129 (2001), 1-4.

35 H. Sota, Y. Hasegawa, and M. Iwakura, 'Detection of Conformational Changes in an Immobilized Protein Using Surface Plasmon Resonance', Anal Chem, 70 (1998), 2019-24.

36 C. M. Kinoshita, S. C. Ying, T. E. Hugli, J. N. Siegel, L. A. Potempa, H. Jiang, R. A. Houghten, and H. Gewurz, 'Elucidation of a Protease-Sensitive Site Involved in the Binding of Calcium to C-Reactive Protein', Biochemistry, 28 (1989), 9840-8.

37 O. I. Egbuna, and E. M. Brown, 'Hypercalcaemic and Hypocalcaemic Conditions Due to Calcium-Sensing Receptor Mutations', Best Pract Res Clin Rheumatol, 22 (2008), 129-48.

38 A. K. Ghosh, M. Brindisi, and J. Tang, 'Developing Beta-Secretase Inhibitors for Treatment of Alzheimer's Disease', J Neurochem, 120 Suppl 1 (2012), 71- 83.

39 M. A. Yildirim, K. I. Goh, M. E. Cusick, A. L. Barabasi, and M. Vidal, 'Drug- Target Network', Nat Biotechnol, 25 (2007), 1119-26.

40 W. D. Mallender, D. Yager, L. Onstead, M. R. Nichols, C. Eckman, K.

Sambamurti, L. M. Kopcho, J. Marcinkeviciene, R. A. Copeland, and T. L.

Rosenberry, 'Characterization of Recombinant, Soluble Beta-Secretase from an Insect Cell Expression System', Mol Pharmacol, 59 (2001), 619-26.

41 H. Park, and S. Lee, 'Determination of the Active Site Protonation State of Beta-Secretase from Molecular Dynamics Simulation and Docking Experiment: Implications for Structure-Based Inhibitor Design', J Am Chem Soc, 125 (2003), 16416-22.

42 R. Rajamani, and C. H. Reynolds, 'Modeling the Protonation States of the Catalytic Aspartates in Beta-Secretase', J Med Chem, 47 (2004), 5159-66.

43 N. Yu, S. A. Hayik, B. Wang, N. Liao, C. H. Reynolds, and K. M. Merz, Jr., 'Assigning the Protonation States of the Key Aspartates in Beta-Secretase Using Qm/Mm X-Ray Structure Refinement', J Chem Theory Comput, 2 (2006), 1057-69.

44 I. D. Hills, and J. P. Vacca, 'Progress toward a Practical Bace-1 Inhibitor', Curr Opin Drug Discov Devel, 10 (2007), 383-91.

45 P. Stenlund, G. J. Babcock, J. Sodroski, and D. G. Myszka, 'Capture and Reconstitution of G Protein-Coupled Receptors on a Biosensor Surface', Anal Biochem, 316 (2003), 243-50.

46 M. A. Cooper, A. Hansson, S. Lofas, and D. H. Williams, 'A Vesicle Capture Sensor Chip for Kinetic Analysis of Interactions with Membrane-Bound Receptors', Anal Biochem, 277 (2000), 196-205.

47 O. P. Karlsson, and S. Lofas, 'Flow-Mediated on-Surface Reconstitution of G- Protein Coupled Receptors for Applications in Surface Plasmon Resonance Biosensors', Anal Biochem, 300 (2002), 132-8.

48 I. Navratilova, M. Dioszegi, and D. G. Myszka, 'Analyzing Ligand and Small Molecule Binding Activity of Solubilized Gpcrs Using Biosensor Technology', Anal Biochem, 355 (2006), 132-9.

49 A. Slany, J. Zezula, V. Tretter, and W. Sieghart, 'Rat Beta 3 Subunits Expressed in Human Embryonic Kidney 293 Cells Form High Affinity [35s]T- Butylbicyclophosphorothionate Binding Sites Modulated by Several Allosteric Ligands of Gamma-Aminobutyric Acid Type a Receptors', Mol Pharmacol, 48 (1995), 385-91.

(34)

50 J. R. Wooltorton, S. J. Moss, and T. G. Smart, 'Pharmacological and Physiological Characterization of Murine Homomeric Beta3 Gaba(a) Receptors', Eur J Neurosci, 9 (1997), 2225-35.

51 A. Saras, G. Gisselmann, A. K. Vogt-Eisele, K. S. Erlkamp, O. Kletke, H.

Pusch, and H. Hatt, 'Histamine Action on Vertebrate Gabaa Receptors: Direct Channel Gating and Potentiation of Gaba Responses', J Biol Chem, 283 (2008), 10470-5.

52 M. T. Bianchi, A. G. Clark, and J. L. Fisher, 'The Wake-Promoting Transmitter Histamine Preferentially Enhances Alpha-4 Subunit-Containing Gabaa Receptors', Neuropharmacology, 61 (2011), 747-52.

53 J. W. Blunt, B. R. Copp, R. A. Keyzers, M. H. Munro, and M. R. Prinsep, 'Marine Natural Products', Nat Prod Rep, 29 (2012), 144-222.

54 K. P. Mishra, L. Ganju, M. Sairam, P. K. Banerjee, and R. C. Sawhney, 'A Review of High Throughput Technology for the Screening of Natural Products', Biomed Pharmacother, 62 (2008), 94-8.

55 A. L. Harvey, and I. A. Cree, 'High-Throughput Screening of Natural Products for Cancer Therapy', Planta Med, 76 (2010), 1080-6.

(35)

(36)

Acta Universitatis Upsaliensis

Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 1007

Editor: The Dean of the Faculty of Science and Technology

A doctoral dissertation from the Faculty of Science and Technology, Uppsala University, is usually a summary of a number of papers. A few copies of the complete dissertation are kept at major Swedish research libraries, while the summary alone is distributed internationally through the series Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology.

UNIVERSITATISACTA

Protein Interaction Studies with Low Molecular Weight Ligands: Applications for Drug Discovery, Basic Research and Diagnostic Tool Design