Refine
Year of publication
Document Type
- Doctoral Thesis (32)
Has Fulltext
- yes (32)
Is part of the Bibliography
- no (32)
Keywords
- Cytochromoxidase (2)
- Protonentransfer (2)
- Absorptionsspektroskopie (1)
- AlignMe (1)
- Alignment (1)
- Arzneimitteldesingn (1)
- Arzneimittelentwicklung (1)
- Atemwege (1)
- Bioenergetik (1)
- Cardiolipin (1)
Institute
- Biochemie und Chemie (22)
- Biochemie, Chemie und Pharmazie (5)
- Biowissenschaften (2)
- Physik (2)
- MPI für Biophysik (1)
- Pharmazie (1)
Membrane proteins play vital role in a variety of cellular processes, such as signal transduction, transport and recognition. In turn they are involved in numerous human diseases and currently represent one of the most prevalent drug targets. A comprehensive understanding of the mechanisms mediated by membrane proteins requires information about their structures at near-atomic resolution, although structural studies of membrane proteins remain behind those of soluble proteins. A bottleneck in the study of membrane proteins resides in the difficulties that are encountered during their high-level production in cell based systems. However, many toxic effects attributed to the over production of membrane proteins are eliminated by cell-free expression, as viable host cells are no longer required. Therefore, the objective of this study was to obtain adequate amounts of selected membrane transport proteins for their structural studies using a cell-free expression system. For the establishment of the cell-free system for membrane proteins, the transporters YbgR and YiiP from Salmonella typhimurium LT2, PF0558 and PF1373 from Pyrococcus furiosus, from the cation diffusion family (CDF), BetP from Corynebacterium glutamicum from the betaine/carnitine/choline transporter (BCCT) family and Aq-2030 from Aquifex aeolicus VF5 from the monovalent cation/proton antiporter-2 (CPA2) family were selected. An Escherichia coli S-30 extract based cellfree system was established by generating the best expression constructs of the target proteins, preparing T7 RNA polymerase and an S-30 extract with high translation efficiency. The functionality of the S-30 extract was shown by the cell-free expression of correctly folded Green Fluorescent Protein (GFP). Essential factors of the cell-free system such as the Mg2+ concentration, the bacterial S-30 extract proportion in the reaction mixture and the time-course of cell-free reactions have been optimized. For the cell-free production of membrane proteins in soluble form, the possibility to supplement cell-free reactions with detergents was explored. A wide range of non-ionic or zwitterionic detergents, were found to be compatible with cell-free synthesis, while ionic detergents and non-ionic detergents at high concentrations had an inhibitory effect. Moreover, high concentrations of polyoxyethylene-alkyl-ethers (Brij) detergents were found to have enhancing effect on the production levels as well as on the solubility of cell-free produced proteins. As membrane proteins tend to misfold and aggregate in a membrane-free translation system, the possibility to supplement the cell-free reactions with inner membrane vesicles (IMVs) to obtain correctly folded target transport proteins was explored. All the target proteins were successfully produced in the batch cell-free reactions and were found to be incorporated in the IMVs. A continuous exchange cell-free (CECF) system was established, where consumable substrates (amino acids, nucleotides and energy regenerating compounds) were supplied to the cell-free reaction mixture through a dialysis membrane, which in consequence resulted in high-level production of target proteins compared to the batch system. The osmosensing and osmoregulated sodium-coupled symporter BetP from C. glutamicum was chosen for the large scale production in CECF set-up. The protein is easily produced in E. coli and is functional as assayed by its transport activity, after purification and reconstitution in liposomes. It is therefore possible to compare in-vivo and cell-free production. High-level cell-free production of BetP was achieved in CECF mode in different forms: (i) as precipitate, (ii) as soluble form in detergent, and (iii) incorporated in IMVs. Cell-free production of BetP resulted in the yield of about 0.5 mg of purified BetP from 1 ml of CECF reaction. The yield of purified BetP was increased to 1.6 fold by addition of 1% polyoxyethylene-(20)-cetyl-ether (Brij58) detergent in the reaction mixture. Moreover, the high level cell-free production of BetP (0.5 mg purified BetP/ml reaction mixture) incorporated in IMVs was shown for the first time in this work.However, it was observed that oligomerization of BetP was not efficient in the cell-free system. Factors that can promote the folding of membrane proteins such as lipids and chaperones were investigated. Addition of lipids and molecular chaperone GroE facilitated correct folding of BetP resulting in increased yield and stability of cell-free produced BetP. The results obtained indicate that most of the cell-free produced BetP exists in functional oligomeric form. The possibility of obtaining milligram amounts of BetP, a 12 trans-membrane protein from the cell-free reactions holds promise for structural and functional studies of other membrane proteins. In any case, the strategies adapted in this study should prove extremely valuable for the production of membrane proteins in the E. coli cell-free expression system.
Die vorliegende Arbeit befaßte sich mit der Untersuchung der Protonenbewegung während des O-E Schrittes im katalytischen Zyklus der Cytochrom-c-Oxidase von P. denitrificans. Die Zuordnung der Protonenbewegung zu den einzelnen Schritten des katalytischen Zyklus der Cytochrom-c-Oxidase ist immer noch ein Gegenstand zahlreicher Kontroversen. Obwohl von Ruitenberg et al. (2000) durch Spannungsmessungen gezeigt wurde, daß die Reduktion von Häm a während des ersten Elektrontransfers in das oxidierte Enzyme eine schnelle Protonenaufnahme von der gegenüberliegenden Seite der Membran bewirkt, wurden diese Ergebnisse angezweifelt. Daher sollte mit einer unabhängigen und direkten Methode herausgefunden werden, ob Protonen bereits während des ersten Schrittes des katalytischen Zyklus aufgenommen werden. Dazu wurde ns-zeitaufgelöste Blitzlicht-Absorptionsspektroskopie in Kombination mit pH-sensitiven Farbstoffen genutzt, und zwar sowohl mit Fluorescein kovalent an der Proteinoberfläche gebunden als auch mit Phenolrot löslich im Medium vorliegend. Zur kovalenten Kopplung von thiolreaktiven Farbstoffen mußten zuerst die nötigen Voraussetzungen geschaffen werden. Dazu wurde in dieser Arbeit ein Mutagenesesystems für sowohl Untereinheit I als auch Untereinheit II etabliert und eine oberflächencysteinfreie Variante und elf Einzelcystein-Varianten hergestellt, exprimiert und aufgereinigt sowie die Enzymaktivitäten überprüft. Danach wurde ein Protokoll zur Kopplung der Einzelcysteinvarianten mit Iodoacetamidfluoresein ausgearbeitet und die Varianten Fluorescein-markiert. Dabei zeigte es sich, daß nur sieben Varianten erfolgreich mit IAF reagierten. Mittels dieser AF-markierten Varianten konnte die Pufferkapazität an der Oberfläche der Cytochrom-c-Oxidase bestimmt werden. Es zeigte sich, daß die Pufferkapazität des Enzyms in Lösung im Vergleich zu Bakteriorhodopsin dreimal so groß ist, an der Oberfläche sogar 10-15mal so groß. Dies deutet auf eine hohe Anzahl protonierbarer Gruppen um die für die Markierung ausgewählten Aminosäuren im Bereich der Eintrittsstellen der Protonen hin. Die gezielte Übertragung eines Elektrons auf die Cytochrom-c-Oxidase erfolgte durch Licht anregbare Rutheniumkomplexe. In unserem Meßsystem war die Elektronentransfereffizienz von [Ruthenium(2,2‘-bipyridin)2]2quarterpyridin am höchsten. Nach einer sorgfältigen Optimierung der Meßbedingungen wie pH-Wert, Ionenstärke und Energie des Lasers konnte eine 10-15 %ige Reduktion von Häm a mit einer Zeitkonstanten von t = 13,7 ± 2,4 µs nachgewiesen werden. Die Protonenkonzentrationsänderungen im Medium konnten durch Phenolrot verfolgt werden. Durch den Vergleich von Funktionsvarianten, bei denen jeweils einer oder beide Protoneneingangswege blockiert sind, konnte ein Modell für die Protonenaufnahme und -abgabe während der Einelektronen-Reduktion der Cytochrom-c-Oxidase entwickelt werden. Dies konnte durch Messungen an in Liposomen inkorporierter wt Cytochrom-c-Oxidase verifiziert werden. Die Nettoprotonenaufnahme von der N-Seite der Cytochrom-c-Oxidase beträgt somit 0,3 H+ für das im O-E Schritt aufgenommene Elektron. Die Variante CS-I302C-AF wurde dazu genutzt, die Oberflächenladungsdichte an der N-Seite der Cytochrom-c-Oxidase zu bestimmen. Die Oberflächenladungsdichte auf der N-Seite des Enzyms in der Nähe zum Eingang des K-Wegs ist negativ und beträgt 0,5 e-/1000 Å2.
G protein-coupled receptors (GPCRs) constitute an important class of integral membrane proteins that are involved in several signaling pathways. About 50% of the currently available drugs are targeted against these receptors and high-resolution structures of these receptors will be of immense importance from the perspective of designing specific and potent drugs. However, structure determination of these receptors and of membrane proteins in general, has been a very challenging task till date. A major limitation in the structure determination of these proteins is that they are present in minute amounts in the native tissues and therefore, they must be produced heterologously. Additionally, crystallization of GPCRs is difficult owing to their flexible nature and limited hydrophilic surface area available for crystal contacts. The aim of my Ph.D. thesis work is two fold, first, to address the problem of GPCR crystallization by using a fusion protein complex approach and second, to tailor Rhodobacter sphaeroides as an expression system for the heterologous production of GPCRs. In the first approach, R. sphaeroides was used as an expression system to generate a fusion protein complex of the photosynthetic reaction center (RC) with a GPCR, expecting that such a complex would be easier to crystallize than the receptor alone. The notion behind this approach is that the RC will act as a scaffold in providing surface area to create crystal contacts and at the same time, it will also reduce the flexibility of the receptor, hopefully without perturbing the functionality of the receptor. Based on the computational modelling experiments, two ways to generate a fusion complex were assigned. Long linkers were inserted between the subunits of the RC and the GPCR. The linkers were designed with a possibility of straightforward alteration of their length as they contained a number of restriction enzyme sites. A series of these constructs were designed and expressed in R. sphaeroides deletion strain, which did not possess the chromosomal RC genes. Though most of these fusion constructs could be successfully expressed, as analyzed by western blot, majority of them were not functional in terms of ligand binding of the GPCR component of the fusion complex. Interestingly, one of these constructs, where the M subunit of RC was directly fused to the human angiotensin II type 1a receptor (AT1aR), exhibited significant functional expression. Based on saturation binding analysis using [125I] iodotyrosyl4Sar1Ile8-angiotensin II (an AT1aR subtype specific antagonist), an expression level of 40+5 pmol/mg of total membrane protein was calculated. This expression level corresponds to approximately 0.3 mg of functional receptor per liter culture and it is significantly higher than the AT1aR expression in native tissues. Additionally, the binding affinity of the recombinant receptor for its endogenous ligand angiotensin II was found to be 1±0.1 nM, which is similar to that observed for the AT1aR in native tissues. More interestingly, the RC part of the fusion complex was structurally assembled in other words, properly folded as judged by the presence of the characteristic peaks at 760 nm, 800 nm and 850 nm by absorption spectroscopy. However, a slight change in the intensity of the peak at 800 nm was observed while comparing the spectra of native RC with that in the fusion protein complex. This slight variation might be due to the change in the protein environment. The fusion protein complex RC-AT1aR was functionally solubilized and purified using a decahistidine tag fused at the c-terminus of the AT1aR. Subsequently, the monodispersity and integrity of the complex was confirmed by size exclusion chromatography, which revealed a homogeneous peak. Additionally, it was also possible to solubilize and purify this complex in the presence of a fluorescein tagged angiotensin II ligand which provides a nice tool to judge the functionality of the AT1aR and integrity of the complex at the same time. The purified RC-AT1aR fusion complex was then subjected to three-dimensional (3-D) crystallization trials and it was possible to obtain reproducible crystals of this complex. The crystals were fluorescent (as the complex was purified in presence of fluorescently labelled angiotensin II) and needle or tetragonal in shape, but produced a powdery diffraction pattern. Further attempts to improve the crystallization condition and to optimize the cryo-conditions are underway. In addition, attempts are also being made to obtain the crystals of this complex with the antagonist (e.g. losartan) bound to the receptor. In view of several limitations in the heterologous expression of GPCRs, as the second part of my Ph.D. thesis, I decided to explore the possibilities of developing a novel expression system based on R. sphaeroides for production of recombinant GPCRs. The notion behind using this host is that lack of inclusion bodies and high concentration of membranes in R. sphaeroides would result in efficient functional overexpression of recombinant membrane proteins. For this purpose, a R. sphaeroides strain, modified by the deletion of the genes encoding the RC and the light harvesting proteins LH1 and LH2, was used. The genes for RC and LHs constitute about 85-90% of total membrane proteins in a R. sphaeroides cell. These membranes are normally housed in special membrane vesicles called intracytoplasmic membranes (ICMs) that can fill almost the entire cell volume under certain growth conditions. Synthesis of a heterologous protein under the control of the moderately strong photosynthetic superoperonic promoter should be coordinated with the synthesis of new membranes to harbour these proteins, thus acting as a natural induction system. Moreover, as most of the native membrane proteins are absent in this deletion strain, heterologously produced protein should not experience a shortage of molecular chaperones for proper folding and insertion. Additionally, the absence of inclusion bodies in this host should enhance the functional and homogenous population of the recombinant proteins. Three human GPCRs, namely the adenosine A2a receptor (A2a), the angiotensin II type 1a receptor (AT1aR) and the bradykinin subtype 2 receptor (B2R) were tested for expression and functionality in this system. Two different constructs were used to determine the optimal position and ribosome-binding site (RBS) in the superoperon for the highest expression level. Of these three receptors, the AT1aR and B2R were successfully produced, while the A2aR failed to express, producing green carotenoid free R. sphaeroides mutants, for unknown reasons. For the recombinant B2R, [3H] bradykinin binding analysis revealed a low functional expression level of 0.7-0.8 pmol/mg of total membrane protein. This expression level corresponds to 0.01 mg functional receptor per liter of culture and is not sufficient for large-scale expression of this receptor. However, for the recombinant AT1aR, [125I] iodotyrosyl4Sar1Ile8- angiotensin II binding analysis revealed an expression level of 12±1 pmol/mg of total membrane protein. This expression level corresponds to approximately 0.1 mg functional receptor per liter culture and this is significantly higher than the AT1aR expression in native tissues. This expression system is still in the nascent stages of development and there are several parameters, which are still to be assessed for the optimal use of this system for the production of GPCRs and other membrane proteins. In conclusion, my Ph.D. work presents a novel fusion protein complex based approach for obtaining crystallizable GPCRs and a novel expression system for producing heterologous GPCRs. It was possible, for the first time, to produce a functional RC-GPCR complex that could easily be crystallized, though further finetuning of the system is required. R. sphaeroides based novel expression system was successfully used to produce functional human GPCRs under the control of a moderately strong photosynthetic superoperonic promoter. This expression system represents a naturally induced system where the expression of a heterologous protein is coordinated with the synthesis of new membranes to harbour the recombinant protein. The fusion protein complex approach and the expression system presented here can hopefully be used as a general method to facilitate the expression and crystallization of other membrane proteins.
Membrane proteins are biological macromolecules that are located in a cell’s membrane and are responsible for essential functions within an organism, which makes them to prominent drug targets. The extraction of membrane proteins from the hydrophobic membrane bilayer to determine high-resolution crystal structures is a difficult task and only 2% of all solved proteins structures are membrane proteins. Computational methods may help to gain deeper insights into membrane protein structures and their functions. This study will give an overview of such computational methods on a representative set of membrane proteins and will provide ideas for future computational and experimental research on membrane proteins.
In a first step (chapter 2), I updated an earlier, manually-curated data set of homologous membrane proteins (HOMEP) to more recent versions in 2010 (HOMEP2) and 2013 (HOMEP3) using an automated clustering approach. High-resolution structures of membrane proteins listed in the PDB_TM database were structurally aligned and subsequently clustered using structural similarity scores. Both data sets were used as a standard gold reference set for subsequent work.
Subsequently, I have updated and applied the sequence alignment program AlignMe to determine protein descriptors that are suitable for detecting evolutionary relationship between homologous a-helical membrane proteins. Single input descriptors were tested alone and in combination with each other in different modes of AlignMe by optimizing gap penalties on the HOMEP2 data set. Most accurate alignments and homology models on the HOMEP2 data set were observed when using position-specific substitution information (P), secondary structure propensities (S) and transmembrane propensities (T) in the AlignMe PST mode. An evaluation on an independent reference set of membrane protein sequence alignments from the BAliBASE collection showed that different modes of AlignMe are suitable for different sequence similarity levels. The AlignMe PST mode improved the alignment accuracy significantly for distantly related proteins, whereas for closely-related proteins from the BAliBASE set the AlignMe PS mode was more suitable. This work was published in March 2013 in PLOS ONE. In order to allow also an easier usage of the AlignMe program, I have implemented a web server of AlignMe (chapter 4) that provides the optimized settings and gap penalties for the AlignMe P, PS and PST modes. A comparison to other recent alignment web server shows that the alignments of AlignMe are similar or even more accurate than those of other methods, especially for very distantly related proteins for which the inclusion of membrane protein information has been shown to be suitable. This work was published in the NAR web server issue in July 2014.
Although membrane-specific information has been shown to be suitable for aligning distantly related membrane proteins on a sequence level, such information was not incorporated into structural alignment programs making it unclear which method is the most suitable for aligning membrane proteins. Thus, I compared 13 widely-used pairwise structural alignment methods on an updated reference set of homologous membrane protein structures (HOMEP3) and evaluated their accuracy by building models based on the underlying sequence alignments and used scoring functions (e.g., AL4 or CAD-score) to rate the model accuracy (chapter 5). The analysis showed that fragment-based approaches such as FR-TM-align are the most useful for aligning structures of membrane proteins that have undergone large conformational changes whereas rigid approaches were more suitable for proteins that were solved in the same or a similar state. However, no method showed a significant higher accuracy than any other. Additionally, all methods lack a measure to rate the reliability of the accuracy for a specific position within a structure alignment. In order to solve these problems, I propose a consensus-type approach that combines alignments from four different methods, namely FR-TM-align, DaliLite, MATT and FATCAT and assigns a confidence value to each position of the alignment that describes the agreement between the methods. This work has been published 2015 in the journal “PROTEINS: structure, function and bioinformatics”.
Consensus alignments were then generated for each pair of proteins of the HOMEP3 data set and subsequently analyzed for single evolutionary events within membrane spanning segments and for irregular structures (e.g., 310- and p-helices) (chapter 6). Interestingly, single insertions and deletions could be observed with the help of consensus alignments in the conserved membrane-spanning segments of membrane proteins in four protein families. The detection of such single InDels might help to identify crucial residues for a proteins function.
The electron transport chain (ETC) is used by cells to create an electrochemical proton gradient which can be used by the ATP synthase to produce ATP. ETC, also called respiratory chain, is formed in mitochondria by four complexes (complex I-IV) and mediated by two electron carriers: cytochrome c and ubiquinone. Electrons are passed from one complex to another in a series of redox reactions coupling proton pumping from the negative (N) side of the membrane to the positive (P) side. Complex I can introduce electrons into the ETC by oxidizing NADH to NAD+ and reducing quinone (Q) to quinol (QH2). The process accomplishes pumping of four protons across the membrane. Complex II is another electrons entry point. It catalyzes the oxidation of succinate to fumarate while reducing Q to QH2. Complex III, also called cytochrome bc1 complex, can transfer the electrons from QH2 to cytochrome c and couple to proton pumping. In complex III the Q-cycle contributes four proton translocations: two protons are required for the reduction of one quinone to a quinol and two protons are released to the P side. Complex IV (cytochrome c oxidase), the terminal complex of the ETC, catalyzes the electron transfer to oxygen and pumps four protons to the P side. Structures of ETC complexes are available. However, the structure of a hyperthermophilic cytochrome bc1 complex has not been elucidated till now. Additionally, the dimeric crystal structure of cytochrome c oxidase from bovine has been discussed controversially.
To build up a functional complex, cofactors are required. The active site of A- and B-type cytochrome c oxidases contain the high spin heme a which is synthesized by the integral membrane protein heme A synthase (HAS). HAS can form homooligomeric complexes and its oligomerization is essential for the biological function of HAS. HAS is evolutionarily conserved among prokaryotes and eukaryotes. Despite its importance, little is known about the detailed structural properties of HAS oligomers.
During my PhD studies, I focused on the cytochrome c oxidase (AaCcO), the cytochrome bc1 complex (Aabc1) and the heme A synthase (AaHAS) from Aquifex aeolicus. This organism is one of the most hyperthermophilic ones and can live at extremely high temperatures, even up to 95 °C. Respiratory chain complexes provide energy for the metabolism of organisms, and their structures have been studied extensively in the past few years. However, there has been a lack of atomic structures of complexes from hyperthermophilic and ancient bacteria, so little is known about the mechanism of these macromolecular machines under hyperthermophilic conditions. Therefore, my PhD studies had four main objectives: 1) to structurally and functionally characterize AaCcO, 2) to reveal the mechanism of Aabc1 thermal stability based on its structure, 3) to determine the oligomerization of AaHAS, 4) to provide valuable insights into the relationship between function and oligomerization of AaHAS.
1) Structure of AaCcO
Heme-copper oxidases (HCOs) catalyze the oxygen reduction reaction being the terminal enzymes in the plasma membranes in many prokaryotes or of the aerobic respiratory chain in the inner mitochondrial membrane. By coupling this exothermic reaction to proton pumping across the membrane to the P side, they contribute to the establishment of an electrochemical proton gradient. The energy in the proton electrochemical proton gradient is used by the ATP synthase to generate ATP. HCOs are classified into three major families: A, B and C, based on phylogenetic comparisons. The well-studied aa3-type cytochrome c oxidase from Paracoccus denitrificans (P. denitrificans) represents A-family HCOs. So far, the only available structure of the ba3-type cytochrome c oxidase from Thermus thermophilus represents the B-family of HCOs. This family contains a number of bacterial and archaeal oxidases. The C-family contains only cbb3-type cytochrome c oxidases.
The AaCcO is one of the ba3-type cytochrome c oxidases. Based on the genomic DNA sequence analysis, it has been revealed that A. aeolicus possesses two operons coding for cytochrome c oxidases (two different subunit I genes, two different subunit II genes and one subunit III gene). So far, only subunits CoxB2 and CoxA2 were identified. The presence of the additional subunit IIa was reported in 2012. Moreover, a previous paper reported that AaCcO can use horse heart cytochrome c and decylubiquinol as electron donors and the typical cytochrome c oxidase inhibitor cyanide does not block the reaction completely.
In the course of my PhD studies, I performed heterologous expression of AaCcO in Pseudomonas stutzeri (P. stutzeri) and co-expression with AsHAS in Escherichia coli, respectively. The subcomplex CoxA2 and CoxB2 can be purified from P. stutzeri, however, it lacks heme A. Additionally, a protocol for the heterologous production of cytochrome c555 from A. aeolicus was established. In parallel, I also purified the AaCcO from native membranes according to previously reported methods with some modifications. The activity of AaCcO with its native substrate, cytochrome c555, was 14 times higher than with horse heart cytochrome c.
To enable a detailed investigation and comparison of AaCcO and other cytochrome c oxidases, the cryo-EM structure of AaCcO was determined to 3.4 Å resolution. It shows that the three subunits CoxA2, CoxB2, and IIa are tightly bound together to form a dimer in the membrane. Surprisingly, CoxA2 contains two additional TMHs (TMH13 and TMH14) to enhance the protein stability. The cofactors heme a3, heme b, CuA and CuB are also identified. Interestingly, two molecules of 1,4-naphthoquinone and cardiolipin were observed in the dimer interface. Based on the structure analysis, the AaCcO possesses only the K-pathway for proton delivery to the active site and proton pumping.
...
The members of the multidrug/oligosaccharidyl-lipid/polysaccharide (MOP) transporter superfamily mediate export of a wealth of molecules of physiological and pharmacological importance. According to the Transporter Classification Database (TCDB), the MOP superfamily is mainly categorized into six distantly related families functionally characterized families: the multidrug and toxic compound extrusion (MATE), the polysaccharide transporter (PST), the oligosaccharidyl-lipid flippase (OLF), the mouse virulence factor (MVF) the agrocin 84 antibiotic exporter (AgnG), and the progressive ankylosis (Ank) family. Among these, the multidrug resistance MATE family transporters are most ubiquitous, being present in all domains of life: Archaea, Bacteria and Eukarya. As secondary active transporters, they utilize transmembrane electrochemical ion gradients of Na+ and/or H+ in order to drive the efflux of xenobiotics or cytotoxic metabolic waste products with specificity mainly for polyaromatic and cationic substrates. Active efflux of drugs and toxic compounds carried out by multidrug transporters is one of the strategies developed by bacterial pathogens to confer multidrug resistance. MATE proteins provide resistance to, e.g., fluoroquinolone, aminoglycoside antibiotics, and anticancer chemotherapeutical agents, thus serving as promising pharmacological targets for tackling a severe global health issue. Based on their amino acid sequence similarity, the MATE family members are classified into the NorM, the DNA-damage-inducible protein F (DinF), and the eukaryotic subfamilies. Structural information on the alternate conformational states and knowledge of the detailed mechanism of the MATE transport are of great importance for the structure-aided drug design. Over the past decade, the crystal structures of representative members of the NorM, DinF and eukaryotic subfamilies have been presented. They all share similar overall architecture comprising 12 transmembrane helices (TMs) divided into two domains, the N-terminal domain (TMs 1-6) and the C-terminal domain (TMs 7-12), connected by a cytoplasmic loop between TM6 and TM7 (Fig. II.1). Since all available MATE family structures are known only in V-shaped outward-facing states with the central binding cavity open towards the extracellular side, a detailed understanding of the complete transport cycle has remained elusive. In order to elucidate the underlying steps of the MATE transport mechanism, structures of distinct intermediates, particularly inward-facing conformation, are required.In my PhD project, structural and functional studies have been performed on a MATE family (DinF subfamily) transporter, PfMATE, from the hyperthermophilic and anaerobic archaeon Pyrococcus furiosus. This protein was produced homologously in Pyrococcus furiosus as well as heterologously in Escherichia coli, and used for the subsequent purification and crystallization trials by the vapor diffusion (VD) and lipidic cubic phase (LCP) method. To the best of my knowledge, PfMATE is the first example of a successful homologous production of a membrane protein in P. furiosus. Due to the very low final amount of the purified protein from the native source, the heterologously produced PfMATE samples were typically used for the extensive structural studies. Crystal structures of PfMATE have been previously determined in an outward-facing conformation in two distinct states (bent and straight) defined on the arrangement of TM1. A pH dependent conformational transition of this helix regulated by the protonation state of the conserved aspartate residue Asp41 was proposed. However, it has been discussed controversially, leading to the hypothesis about TM1 bending to be rather affected by interactions with exogenous lipids (monoolein) present under the crystallization conditions. Based on these open questions, an experimental approach to investigate the role of lipids as structural and functional modulators of PfMATE has been taken in the course of my PhD project. The interplay between membrane proteins and lipids can affect membrane protein topology, structure and function. Considering differences between archaeal and bacterial lipid composition, cultivation of P. furiosus cells and extraction of its lipids was followed by the mass spectrometry (MS) based lipidomics for identification of individual lipid species in the archaeal extract. In order to assess the effects of lipids on PfMATE, different lipid molecules were used for co-purification and co-crystallization trials. This dissertation presents a workflow leading to the structure determination of a MATE transporter in the long sought-after inward-facing state, which has been achieved upon purification and crystallization of the heterologously produced PfMATE in the presence of lipids from its native source P. furiosus. Also, the PfMATE outward-facing state obtained from the crystals grown at the acidic pH conditions sheds light on the previously proposed pH-dependent structural alterations within TM1. It is interesting to note that the inward and outward-facing states of PfMATE were obtained from the crystals grown under similar conditions, but in the presence and absence of native lipids, respectively. This observation supports the hypothesis about physiologically relevant lipids to act as conformational modulators or/and a new class of substrates, expanding the substrate spectrum of the MATE family transporters. Comparative analysis of two PfMATE states reveals that transition from the outward to the inward-facing state involves rigid body movements of TMs 2-6 and 8-12 to form an inverted V, facilitated by a loose binding of TMs 1 and 7 to their respective bundles and their conformational flexibility. Local fluctuations within TM1 in the inward-facing structure, including bending and unwinding in the intracellular half of the helix, invoke its highly flexible nature, which is suitable for ion and substrate gating.
...
Im Mittelpunkt der vorliegenden Arbeit steht der humane β2-adrenerge Rezeptor, ein intrinsisches Membranprotein und Mitglied der Superfamilie G Protein-gekoppelter Rezeptoren (GPCRs).
Es wurden im Wesentlichen zwei Themenschwerpunkte bearbeitet. Den ersten Schwerpunkt bildeten die heterologe Produktion des β2-adrenergen Rezeptors in der methylotrophen Hefe Pichia pastoris sowie dessen chromatographische Reinigung und Charakterisierung. Hierbei konnten in allen wesentlichen Punkten signifikante Verbesserungen im Vergleich zu früheren Arbeiten erzielt werden: Die Ausbeute an heterolog produziertem, funktionellem Rezeptor konnte um das bis zu Fünffache gesteigert werden; Die Gesamtausbeute nach chromatographischer Reinigung wurde um 60% erhöht; Es gelang die Darstellung reiner, monodisperser Rezeptorpräparationen mit einer spezifischen Ligandenbindungsaktivität von 94% und einer Halbwertszeit der Ligandenbindungsaktivität von >50 Tagen bei 4 °C. Eine pharmakologische Charakterisierung des Rezeptors erfolgte sowohl in Membranen als auch in solubilisiert-gereinigter Form. Ferner konnte die in vitro-Interaktion des solubilisiert-gereinigten Rezeptors mit einer konstitutiv aktiven β-Arrestin Variante nachgewiesen werden.
Ziel des zweiten Themenschwerpunkts der vorliegenden Arbeit war die Auffindung und Charakterisierung von rezeptorspezifischen, nicht auf Immunglobulinen basierenden Bindeproteinen, um diese später vorrangig zur Kokristallisation einsetzen zu können. Mit der zellfreien Selektionsmethode des ribosome display ist es hierbei gelungen, aus einer naiven, auf Ubiquitin als Gerüstprotein basierenden Bibliothek hochaffine, spezifische Bindeproteine gegen den β2-adrenergen Rezeptor zu isolieren. Für Monomere dieser sog. Affiline® wurden Dissoziationskonstanten bis etwa 450 nM gefunden, die durch Homodimerisierung auf etwa 70 nM verringert werden konnten. Zusätzlich wurde mit einer Affinitätsmaturierung ausgewählter Varianten begonnen.
Die Superfamilie der G Protein-gekoppelten Rezeptoren (GPCRs) bildet die größte und auch diverseste der vier Hauptgruppen membranständiger Rezeptoren. Im Rahmen der Signaltransduktion vermitteln diese die Umwandlung eines extrazellulären Signals in eine spezifische intrazelluläre Reaktion und nehmen so eine Schlüsselposition bei der Kommunikation zwischen Zellen ein. Bei GPCRs erfolgt die Weitergabe der ligandeninduzierten Konformationsänderung vorwiegend über Kopplung an und Aktivierung von Guanin-Nukleotidebindenden Proteinen (G Proteine), die dann ihrerseits mit verschiedenen Effektorsystemen in Wechselwirkung treten. Da ein Großteil aller zurzeit auf dem Markt befindlichen Medikamente ihre Wirkung durch Bindung an einen GPCR entfaltet, ist diese Proteinfamilie von besonders großem pharmakologischem Interesse. Auch der humane β2-adrenerge Rezeptor (β2AR) zählt zu den kommerziell wichtigen drug targets und ist zudem nach bovinem Rhodopsin der erste GPCR, für dessen Struktur experimentell bestimmte, atomare Modelle vorgelegt werden konnten.
Zur funktionellen Überproduktion des β2AR hatte sich bereits die methylotrophe Hefe Pichia pastoris bewährt. In der vorliegenden Arbeit konnte gezeigt werden, dass durch Anwendung hier erarbeiteter, optimierter Bedingungen das Produktionsniveau aller acht für dieses Expressionssystem zur Verfügung stehenden bzw. in dieser Arbeit hergestellten Konstrukte gegenüber früheren Ergebnissen deutlich gesteigert werden konnte. Hierbei wurde, je nach Konstrukt, eine Verdopplung bis Verfünffachung der Produktion an aktivem, bindungsfähigem Rezeptor erzielt. Bei drei der acht Konstrukte konnte das Produktionsniveau auf ≥ 50 pmol/mg Membranprotein gesteigert werden, was bis zu 1,1 mg aktivem Rezeptor in Membranen pro Liter Schüttelkultur entspricht. Die im Rahmen der Produktionsoptimierung der verschiedenen Konstrukte gewonnenen Erkenntnisse über den Einfluss sowohl von Art als auch Lokalisation eines bestimmten Anhängsels innerhalb der Expressionskassette auf das zu erwartende Produktionsniveau konnten genutzt werden, um bei einem mit bestimmten Eigenschaften neu zu erstellenden Konstrukt – einem Baukastensystem gleich – jene Anhängsel mit den gewünschten (und bekannten) Auswirkungen zu kombinieren. Auf dies Weise konnten auf Anhieb sehr gute Produktionsniveaus für neu erstellte Konstrukte erzielt werden.
Durch Optimierung einer zweischrittigen affinitätschromatographischen Reinigung konnte zum einen die Gesamtausbeute an gereinigtem β2AR um 60% gegenüber früheren Ergebnissen gesteigert werden. Zum anderen gelang es, für derartige, nach SDS-PAGE und analytischer Gelfiltration als homogen und frei von jeglichen Aggregationen befundenen Präparationen, die spezifische Aktivität (Radioligandenbindung) auf 94 ± 4% zu steigern. Die Halbwertszeit der spezifischen Aktivität derartiger Präparationen bei 4 °C lag bei über 50 Tagen. Im Rahmen einer alternativen Reinigungsstrategie konnte der praktische Nutzen proteolytisch abspaltbarer Affinitätsanhängsel in Fällen mit inhomogener posttranslationaler Prozessierung des rekombinanten Proteins durch das Expressionssystem nachgewiesen werden.
Durch Integration einer inversen Affinitätschromatographie in das Reinigungsschema konnte bei bestimmten Konstrukten die Homogenität und Dispersität der Präparation verbessert werden, bei gleichzeitiger vollständiger Entfernung sowohl der abgetrennten Anhängsel als auch der verwendeten Protease.
Die Ligandenbindung des β2AR wurde sowohl in Membranen als auch in solubilisiertgereinigter Form charakterisiert. Die hierbei für den Rezeptor in Membranen mit dem tritiierten Liganden [5,7-3H]-(-)-CGP-12177 erhaltene Dissoziationskonstante (KD) von 4,1 ± 0,2 nM befindet sich in guter Übereinstimmung mit den Literaturwerten für dieses Expressionssystem.
Bei der Charakterisierung der Ligandenbindung des β2AR in solubilisiert-gereinigter Form wurde ein stark modulierender Effekt von Lipiden auf die KD beobachtet: Die Dissoziationskonstante des soubilisiert-gereinigten Rezeptors wurde zu 8 ± 0,6 nM bestimmt, ließ sich jedoch durch gleichzeitige Verwendung von solubilisiertem Phosphatidylcholin und Cholesterinhemisuccinat im Bindungstest auf ein Viertel (2,13 ± 0,2 nM) diese Wertes reduzieren. Der Effekt der Affinitätsmodulation war voll reversibel und die Gesamtmenge (Bmax) an bindungsfähigem Rezeptor blieb praktisch unbeeinflusst.
Des weiteren konnte die spezifische Interaktion des solubilisiert-gereinigten β2AR mit rekombinantem, konstitutiv aktivem β-Arrestin in vitro nachgewiesen werden. Dies ist sowohl für die Etablierung etwaiger funktioneller Assays als auch für Ansätze zur Kokristallisation des β2AR mit einem physiologischen Bindungspartner für von Interesse.
Im Rahmen des zweiten Themenkomplexes wurde die Methode des ribosome display (RD) erfolgreich eingesetzt, um Bindeproteine gegen den β2AR zu selektieren. Ziel war es hier, diese später zur Kokristallisation mit dem Zielprotein einsetzen zu können. Das RD als zellfreie (in vitro) Selektionsmethode findet für lösliche Proteine bereits seit längerem breite Anwendung.
Für Membranproteine wurde es bisher hingegen nur sehr selten erfolgreich verwendet und der Einsatz bei GPCRs ist als neu zu bezeichnen. Im Rahmen dieser Arbeit wurde eine hochdiverse (theoretische Diversität: 1013, funktionelle Größe 1010 – 1011), naive Bibliothek verwendet, die auf Ubiquitin als Gerüstprotein basiert und von SCIL Proteins, Halle unter der Bezeichnung Affilin® entwickelt wurde. Es handelt sich somit um alternative, d. h. nicht auf Antikörpern oder deren Fragmenten beruhenden Bindeproteine. Zielprotein der in dieser Arbeit durchgeführten Selektionen war solubilisiert-gereinigter β2AR. Sämtliche Arbeiten unter Beteiligung des Zielproteins fanden zur Bewahrung dessen Aktivität in Anwesenheit von Detergenz statt. Die verschiedenen Schritte des RD mussten daher zunächst an die speziellen Erfordernisse des Zielproteins angepasst und geeignete Bedingungen zur Durchführung einer Selektion gefunden werden. Durch entsprechende Vorversuche wurde dann verifiziert, dass der Rezeptor unter den gewählten Selektions- und Immobilisierungsbedingungen seine Ligandenbindugsaktivität beibehält. Es wurden insgesamt sechs RD-Selektionsrunden durchgeführt und der abschließend vorliegende Bindeprotein-Subpool war noch divers (keine Reduktion auf eine Konsensus-Sequenz). Zur Charakterisierung der angereicherten Varianten wurde eine Abfolge von Hochdurchsatzverfahren sowie abschließenden Einzelanalysen eingesetzt. Hierbei gelang es, hochaffine, spezifische Bindeproteine gegen den β2AR zu isolieren. Die beobachteten Dissoziationskonstanten (KD) der detailliert charakterisierten Affilin-Monomere reichen von etwa 15 μM bis hin zu 450 nM. Nach Homodimerisierung ausgewählter, zuvor charakterisierter Varianten wurde eine Verringerung der Dissoziationskonstanten bis auf etwa 70 nM beobachtet. Damit decken die selektierten Bindeproteine einen Affinitätsbereich ab, der verschiedenartige Anwendungsgebiete eröffnet: Vom Einsatz zur affinitätschromatographischen Reinigung des Zielproteins, über dessen immunologischen Nachweis bzw. dem Einsatz in Assaysystemen bis hin zur Kokristallisation mit dem Zielprotein. Die Spezifität der Bindeproteine wurde gegenüber verschiedenen Negativkontrollen wie etwa dem verwendeten Immobilisierungssystem aber auch gegenüber anderen GPCRs mit homologer Auswahl an Affinitätsanhängseln verifiziert. Durch geeignete Konditionierung (prepanning) der in den eigentlichen Selektionsprozess eingehenden ternären Komplexe war es gelungen, die Anreicherung unerwünschter, gegen das zur Immobilisierung verwendete Biotin/Streptavidin-System gerichteter Varianten wirkungsvoll zu verhindern.
Die selektierten Bindeproteine konnten in verschiedenen Maßstäben (von 1 ml bis 1 l) mit sehr hohen Ausbeuten in E. coli produziert werden. Nach Affinitätschromatographie und präparativer Gelfiltration wurden hochreine, monodisperse Präparationen erhalten. Die Gesamtausbeuten an gereinigtem Bindeprotein lagen bei etwa 15 mg/l Kulturvolumen.
Neben der Affinitätsbestimmung wurde die Bindung einzelner Affilin-Varianten an den β2AR sowohl mittels pull-down Assay als auch analytischer Gelfiltration verifiziert. Parallel zur Homo- und Hetero-Dimerisierung wurde eine Affinitätsmaturierung ausgewählter Varianten begonnen. Diese Affinitätsmaturierung der Monomere erfolgte in einem evolutiven Ansatz bestehend aus Zufallsmutagenese zuvor charakterisierter Varianten und anschließender Selektion im Rahmen eines erneuten RD.
Systematische Ansätze zur Kokristallisation des β2AR mit mehreren der aus der Selektion hervorgegangenen alternativen Bindeproteine wurden durchgeführt.
Cytochrome c oxidase is the terminal enzyme in the respiratory chain of mitochondria and aerobic bacteria. This enzyme ultimately couples electron transfer from cytochrome c to an oxygen molecule with proton translocation across the inner mitochondrial and bacterial membrane. This reaction requires complicated chemical processes to occur at the catalytic site of the enzyme in coordination with proton translocation, the exact mechanism of which is not known at present. The mechanisms underlying oxygen activation, electron transfer and coupling of electron transfer to proton translocation are the main questions in the field of bioenergetics. The major goal of this work was to investigate the coupling of electron transfer and proton translocation in cytochrome c oxidase from Paracoccus denitrificans. Different theoretical approaches have been used to investigate the coupling of electron and proton transfer. This thesis presents an internal water prediction scheme in the enzyme and a molecular dynamics study of cytochrome c oxidase from Paracoccus denitrificans in the fully oxidized state, embedded in a fully hydrated dimyristoylphosphatidylcholine lipid bilayer membrane. Two parallel molecular dynamics simulations with different levels of protein hydration, 1.125 ns each in length, were carried out under conditions of constant temperature and pressure using three-dimensional periodic boundary conditions and full electrostatics to investigate the distribution and dynamics of water molecules and their corresponding hydrogen-bonded networks inside cytochrome c oxidase. The average number of solvent sites in the proton conducting K- and D- pathways was determined. The highly fluctuating hydrogen-bonded networks, combined with the significant diffusion of individual water molecules provide a basis for the transfer of protons in cytochrome c oxidase, therefore leading to a better understanding of the mechanism of proton pumping. The importance of the hydrogen bonding network and the possible coupling of local structural changes to larger scale changes in the cytochrome c oxidase during the catalytic cycle have been shown.