The work presented in this thesis concerns the dynorphin neuropeptides, and dynorphin A (DynA) in particular. DynA belongs to the wider class of typical opioid peptides that, together with the opioid receptors, a four-membered family of GPCR membrane proteins, form the opioid system. This biological system is involved or implicated in several physiological processes such as analgesia, addiction and depression, and effects caused by DynA through this system, mainly through interaction with the kappa subtype of the opioid receptors (KOR), are called the opioid effects. In addition to this, non-opioid routes of action for DynA have been proposed, and earlier studies have shown that direct membrane interaction is likely to contribute to these non-opioid effects. The results discussed here fall into either of two categories; the interaction between DynA and a fragment of KOR, and the direct lipid interaction of DynA and two variant peptides.For the receptor interaction case, DynA most likely causes its physiological effects through binding its N-terminal into a transmembrane site of the receptor protein, while the extracellular regions of the protein, in particular the extracellular loop II (EL2), have been shown to be important for modulating the selectivity of KOR for DynA. Here we have focussed on the EL2, and show the feasibility of transferring this sequence into a soluble protein scaffold. Studies, predominantly by nuclear magnetic resonance (NMR) spectroscopy, of EL2 in this new environment show that the segment has the conformational freedom expected of a disordered loop sequence, while the scaffold keeps its native beta-barrel fold. NMR chemical shift and paramagnetic resonance enhancement experiments show that DynA binds with high specificity to EL2 with a dissociation constant of approximately 30 micro Molar, while binding to the free EL2 peptide is an order of magnitude weaker. The strength of these interactions are reasonable for a receptor recognition event. No binding to the naked scaffold protein is observed.In the second project, the molecules of interest were two DynA peptide variants recently found in humans and linked to a neurological disorder. Previously published reports from our group and collaborators pointed at very different membrane-perturbing properties for the two variants, and here we present the results of a follow-up study, where the variants R6W-DynA and L5S-DynA were studied by NMR and circular dichroism (CD) spectroscopy in solutions of fast-tumbling phospholipid bicelles, and compared with wild type DynA. Our results show that R6W-DynA interacts slightly stronger with lipids compared to wild type DynA, and much stronger compared to L5S-DynA, in terms of bicelle association, penetration and structure induction. These results are helpful for explaining the differences in toxicity, membrane perturbation and relationship to disease, between the studied neuropeptides.

The work presented in this thesis concerns the opioid peptide dynorphin A (DynA). DynA functions primarily as a neurotransmitter and belongs to the family of typical opioid peptides. These peptides are a part of the opioid system, together with the opioid receptors, a family of GPCR membrane proteins. The opioid system system is involved or implicated in several physiological processes such as analgesia, addiction, depression and other types of neurological disorders. In this thesis, two biologically relevant aspects of DynA have been investigated with biophysical methods. First, interactions between DynA and an opioid receptor, and second, the direct membrane interactions of DynA.

The DynA–receptor studies were focused on the selectivity-modulating second extracellular loop (EL2) of the kappa-opioid receptor (KOR). A protein engineering approach was used in which the EL2 was grafted onto a soluble protein scaffold. The results show that DynA binds with low affinity but high specificity to EL2 in the construct protein environment. The strength of the interaction is in the micromolar range, and we argue that this interaction is part of the receptor recognition event.

With bicelles as a mimetic, membrane interactions were probed for wild-type DynA and for two DynA peptide variants linked to a neurological disorder. R6W–DynA and L5S–DynA were shown to be very different in terms of bicelle association, penetration and structure induction. In these experiments, as well as in investigations of DynA dynamics in bicelles, the lipid environment was shown to have much larger effects on peptide dynamics than on structure; and both these properties depend on lipid charge.

Additionally, in a methodological project, DHPC/DMPC bicelle morphology as a function of total PC concentration was characterised by diffusion NMR in combination with two-way decomposition. The results may contribute to providing guidelines for the appropriate use of bicelles as a membrane mimetic.

Complex mixtures are at the heart of biology, and biomacromolecules almost always exhibit their function in a mixture, e.g., the mode of action for a spider venom is typically dependent on a cocktail of compounds, not just the protein. Information about diseases is encoded in body fluids such as urine and plasma in the form of metabolite concentrations determined by the actions of enzymes. To understand better what is happening in real living systems we urgently need better methods to characterize such mixtures. In this paper we describe a potent way to disentangle the NMR spectra of mixture components, by exploiting data that vary independently in three or more dimensions, allowing the use of powerful algorithms to decompose the data to extract the information sought. The particular focus of this paper is on NMR diffusion data, which are typically bilinear but can be extended by a third dimension to give the desired data structure.

The membrane interaction properties of two single-residue variants, R6W and L5S, of the 17-amino acid neuropeptide dynorphin A (DynA) were studied by circular dichroism (CD) and nuclear magnetic resonance (NMR) spectroscopy. Corresponding gene mutations have recently been discovered in humans and causatively linked to a neurodegenerative disorder. The peptides were investigated in buffer and in isotropic solutions of q = 0.3 bicelles with 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC) or DMPC (0.8) and 1,2-dimyristoyl-sn-glycero-3-phospho(1'-rac-glycerol) (DMPG) (0.2). The CD results and the NMR secondary chemical shifts show that R6W-DynA has a small a-helical fraction in buffer, which increases in the presence of bicelles, while L5S-DynA is mainly unstructured under all conditions studied here. R6W-DynA has an almost complete association with zwitterionic bicelles (similar to 90%, as probed by NMR diffusion experiments), similar to the behavior of wtDynA, while L5S-DynA has a weaker association (similar to 50%). For all peptides, the level of bicelle association is increased in negatively charged bicelles. The L5A-DynA peptide adopts a very shallow position in the headgroup region of the bicelle bilayer, as studied by paramagnetic spin relaxation enhancement experiments using paramagnetic probes. Similarly, the results show that R6W-DynA is more deeply buried in the bilayer, with only the C-terminal residues exposed to solvent, again more similar to the case of wild-type DynA. We suggest that the results presented here may explain the differences in cell toxicity of these disease-related neuropeptide variants.

The influence of lipid bicelles on the dynamics of the opioid peptide DynA has been investigated by Nuclear Magnetic Resonance. DynA exerts its opioid effects mainly through interactions with the κ subtype of the opioid receptors, but has also been demonstrated to have direct interactions with membranes. Among other properties, it has been shown that the peptide causes membrane disruption and may penetrate bilayers. Despite the fact that DynA appears to bind tightly to model lipid bilayers, no structure induction has been observed. To further study the effect of membrane interactions we have here therefore measured the fast local dynamics of DynA specifically labeled with ¹⁵N in three backbone amide sites (Gly2, Leu5 and Leu12) in fast-tumbling bicelles, both with and without the incorporation of the negatively charged dimyristoylglycerol. We also examined the amide exchange in the two bicelles. We find that despite the fact that DynA is largely unstructured in both types of bicelles, the peptide has restricted backbone dynamics, which depends on the presence of negatively charged lipids. Moreover we see that the lipid dependence is not uniform throughout the sequence, but is most noticeable for Leu5, which precedes an unusually basic stretch of amino acid residues. The findings indicate that this basic sequence may be of significance for bilayer recognition. Finally, we note that the dynamical behavior of the peptide is much more influenced by the lipid surroundings than what the structural properties are.

The molecular determinants for selectivity of ligand binding to membrane receptors are of key importance for the understanding of cellular signalling, as well as for rational therapeutic intervention. In the present study, we target the interaction between the kappa opioid receptor (KOR) and its native peptide ligand dynorphin A (DynA) using solution state NMR spectroscopy, which is generally made difficult by the sheer size of membrane bound receptors. Our method is based on 'transplantation' of an extracellular loop of KOR into a 'surrogate' scaffold; in this case, a soluble beta-barrel. Our results corroborate the general feasibility of the method, showing that the inserted receptor segment has negligible effects on the properties of the scaffold protein, at the same time as maintaining an ability to bind its native DynA ligand. Upon DynA binding, only small induced chemical shift changes of the KOR loop were observed, whereas chemical shift changes of DynA and NMR paramagnetic relaxation data show conclusively that the peptide interacts with the inserted loop. The binding interface is composed of a disordered part of the KOR loop and involves both electrostatic and hydrophobic interactions. Even so, simultaneous effects along the DynA sequence upon binding show that control of the recognition is a concerted event.

Mixtures of lipids and detergents are known to form bicelles at certain parameter ranges, but many uncertainties remain concerning the details of the phase behaviour of these mixtures and the morphology of the formed lipid assemblies. Here we used nuclear magnetic resonance (NMR) diffusion data in combination with the multivariate processing method speedy component resolution (SCORE) to analyse mixtures of 1,2-dihexanoyl-snglycero-3-phosphocholine (DHPC) and 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC) with the relative concentration q = [DMPC]/[DHPC] = 0.5 at total lipid concentrations ranging from 15 to 300 mM. With this approach we were able to resolve the heavily overlapping mixture spectra into component spectra and obtained reliable diffusion coefficients for lipid concentrations in the range 15 to 300 mM, although at high concentrations (250-300 mM), non-negativity constraints or overfactoring was required to successfully decompose the data. At 50-300 mM total lipid concentration, the radii estimated from the diffusion coefficient of DMPC indicate assemblies of the appropriate bicelle size, although small size variations exist, while at lower concentrations the morphology appears to change to larger assemblies. Taken together, the results suggest that for q = 0.5 DMPC/DHPC mixtures there is a relatively broad concentration range above 50 mM where bicelles may reliably be assumed to adopt the 'classical' bicelle morphology. The study clearly demonstrates the usefulness of our approach for accurately determining physical properties of complex mixtures such as bicelles. Both reliable diffusion coefficients and chemical shifts can be derived from overlapping data. This should prove useful for analysing the behaviour of other, more complex, lipid mixtures.

Mixtures of lipids and detergents are known to form bicelles at certain parameter ranges, but many un-certainties remain concerning the details of the phase be-haviour of these mixtures and the morphology of the formed lipid assemblies. Here we used NMR diffusion data in com-bination with the multivariate processing method SCORE to analyze mixtures of DHPC and DMPC with the relative concentration q=[DMPC]/[DHPC]=0.5 at total lipid con-centrations from 15 to 300 mM. With this approach we were able to resolve the heavily overlapping mixture spectra into component spectra and obtained reliable diffusion coeffi-cients for lipid concentrations in the range 15 to 200 mM. Between 200 and 300 mM, the similar diffusion coefficients in combination with substantial signal overlap makes it difficult to get very reliable spectra and diffusion coeffi-cients with standard processing parameters, but overfactoring provided useful diffusion coefficient estimates also at these concentrations. At 50–300 mM total lipid concentration, the radii estimated from the diffusion coeffi-cient of DMPC indicate assemblies of the appropriate bicelle size, although small size variations exist, while at lower concentrations the morphology appears to change to larger assemblies. Taken together, the results suggest that for q=0.5 DMPC/DHPC mixtures there is a relatively broad concentration range above 50 mM where bicelles may relia-bly be assumed to adopt the 'classical' bicelle morphology. At lower concentrations there is evidence for a more com-plex morphology with more than one type of lipid assembly in the sample.