Refine
Year of publication
- 2005 (562) (remove)
Document Type
- Article (215)
- Working Paper (79)
- Doctoral Thesis (58)
- Part of a Book (51)
- Preprint (43)
- Part of Periodical (40)
- Conference Proceeding (36)
- Report (22)
- Book (11)
- Review (3)
Language
- English (562) (remove)
Has Fulltext
- yes (562) (remove)
Keywords
- Artikulation (13)
- Phonetik (13)
- Artikulatorische Phonetik (12)
- Englisch (11)
- Artikulator (8)
- Deutsch (7)
- Europäische Union (7)
- Geldpolitik (7)
- Bedeutungswandel (6)
- Computerlinguistik (6)
Institute
- Physik (72)
- Center for Financial Studies (CFS) (42)
- Wirtschaftswissenschaften (39)
- Biochemie und Chemie (32)
- Medizin (24)
- Frankfurt Institute for Advanced Studies (FIAS) (20)
- Geowissenschaften (17)
- E-Finance Lab e.V. (16)
- Extern (15)
- Biowissenschaften (14)
Introduction: Reactive oxygen species (ROS) have been implicated in neurodegeneration and seem to be involved in the physiology and pathophysiology of several diseases, including normal aging and Alzheimer’s disease (AD). Enhanced ROS production in aging or AD is not restricted to the brain, but can also been seen in several peripheral tissues. The objective of the present study was to evaluate whether the mechanisms involved in the generation of oxidative stress in normal senescence and Alzheimer’s disease are identical or not. Methods: We analysed intracellular basal levels of ROS in lymphocytes from AD patients and healthy young and aged not-demented subjects as well as ROS levels following stimulation with d-ribose and staurosporine in all three groups. ROS levels were measured by flow cytometry using the intracellular fluorescence dye dihydrorhodamine123 (DHR123). Results: Our study shows that AD lymphocytes have increased basal levels of ROS, low susceptibility to ROS stimulation by 2-deoxy-D-ribose (dRib) and an increased response to staurosporine when compared with age-matched controls. Discussion: The data suggest that the defect(s) responsible for enhanced ROS production in AD may involve different or additional biological pathways than those involved in enhanced ROS generation during aging.
Studies in particular of the last decade showed that active neurogenesis continuously takes place in the subventricular zone (SVZ) of the lateral ventricles of the adult rodent brain. Neurogenesis in the SVZ leads to migration of neuroblasts within the rostral migratory stream (RMS) and mature neuron formation mainly in the olfactory bulb (OB). According to present understanding, glial cells with astrocytic properties represent the actual adult neural stem cells. The cell types representing the various cellular transition states leading to the formation of mature neurons as well as the mechanisms controlling adult neurogenesis and neuroblast migration are poorly understood. A previous study from this laboratory demonstrated that the ATP-hydrolyzing enzyme nucleoside triphosphate diphosphohydrolase 2 (NTPDase2) is associated with type B cells, the presumptive neural stem cells. NTPDase2 is a protein of the plasma membrane with its catalytic site facing the extracellular space. It hydrolyzes extracellular nucleoside triphosphates to their respective nucleoside diphosphates. This raises the possibility that the signaling pathway via extracellular nucleotides is involved in the control of adult neurogenesis. Neurons as well as glial cells express several subtypes of receptors (P2 receptors) that are responsive to the nucleotides ATP, ADP, UTP, or UDP. P2X receptors are ATP-gated Na+, K+ and Ca2+ permeable ion channels, P2Y receptors are coupled to trimeric G-proteins. In order to probe for a functional role of nucleotides in adult neurogenesis, the present study referred to an in vitro system (neurospheres). Neurospheres produced from isolates of the mouse SVZ and cultured in the presence of EGF and bFGF expressed the neural stem cell marker nestin and also GFAP, S100β, NTPDase2 and tissue non-specific alkaline phosphatase. Neurospheres generated from the cells of the subventricular zone were multipotenital. This was revealed by immunostaining of differentiated cells with markers for astrocytes, neurons and oligodendrocytes. The presence of ecto-nucleotidase was verified by analyzing the free phosphate released from nucleotides. The tissue non-specific form of alkaline phosphatase was the predominant enzyme. Both NTPDase2 and TNAP could be identified by immunocytochemistry and Western blotting. Hydrolysis was not observed for p-nitrophenyl thymidine monophosphate, a substrate of members of the ectonucleotide pyrophosphatase/phosphodiesterase family (NPP1 to NPP3). Since ecto-nucleotidases control the availability of extracellular nucleotide agonists, neurospheres were studied for the potential expression and functional role of nucleotide receptors. Neurospheres responded to extracellular nucleotides with a transient rise in Ca2+ (ATP = ADP > UTP). The rise in Ca2+ was due to P2Y receptors. The Ca2+ response was unaltered in the absence of extracellular Ca2+ and strongly reduced by thapsigargin, a blocker of internal Ca2+ stores. The P2Y1 antagonist MRS2179 strongly reduced the ATP- or ADP-induced increase in Ca2+, suggesting the involvement of a P2Y1 receptor. In addition, suramin and PPADS, non-selective antagonists for P2 receptors, inhibited most of the Ca2+ response. The agonistic activity of UTP and the lack of response to UDP implied the additional presence of a P2Y2 and/or a P2Y4 receptors and the absence of a functional P2Y6 receptor. RT-PCR experiments demonstrated that neurospheres expressed P2Y1 and P2Y2 receptors but not P2Y4 receptor. That the majority of the Ca2+ response to ATP was mediated via P2Y1 receptors was also confirmed by analysis of P2Y1 knockout mice and by application of the P2Y1 receptor-specific antagonist MRS2179. In addition, agonists of P2Y1 and P2Y2 receptors and low concentrations of adenosine augmented cell proliferation inspite of the presence of mitogenic growth factors. Neurosphere cell proliferation was attenuated after application of MRS2179 and in neurospheres from P2Y1 receptor knockout mice. These results infer a nucleotide receptor-mediated synergism that augments growth factor-mediated cell proliferation. Taken together these results suggest that P2Y-mediated nucleotidergic signalling is involved in neurosphere function and possibly also in adult neurogenesis in situ.
The Na+/proline transporter of E. Coli (PutP) is responsible for the uptake of proline which is subsequently used not only as a carbon and nitrogen source and a constituent of proteins but also as a particularly effective osmoprotectant. However, for a long time there was little known about the single steps in the reaction cycle of this transporter and only few details about its structure-function relationship are available. Aim of the present work was to achieve a deeper understanding about the kinetic properties of the Na+/proline transporter and to get insights into the structure-function relationship of the substrate binding. To answer these questions different techniques were used. By using the novel SSM technique combining the preparation of PutP proteoliposomes it was possible to demonstrate for the first time the electrogenic substrate binding to PutP transporter. Due to rapid solution exchange measurements on the SSM it was additionally possible to obtain time resolved information about the kinetic details of the cytoplasmic substrate binding sites which were not available by previous steady state and equilibrium binding measurements. Pre-steady-state charge translocation was observed after rapid addition of one or both of the cosubstrates Na+ and/or proline to the PutP-WT proteoliposomes adsorbed on the SSM. Thereby it was possible to link the observed electrical signals with the binding activity of PutP. The observed Na+ and/or proline induced charge displacement were assigned to an electrogenic Na+ and/or proline binding process at the cytoplasmic face of the enzyme with a rate constant of k > 50 s-1 proceeding the rate limiting step of the reaction cycle. Furthermore, based on the kinetic analysis of the electrical signals obtained from the measurements of PutP on SSM, the following characteristics of the substrates binding in PutP were deduced: (1) both Na+ and proline can bind individually to the transporter. Under physiological conditions, an ordered binding mechanism prevails; while at sufficiently high concentrations, each substrate can bind in the absence of the other; (2) substrate binding is electrogenic not only for Na+, but also for the uncharged cosubstrate proline. The charge displacement associated with Na+ binding and proline binding is of comparable size and independent of the presence of the respective cosubstrate. In addition, it was concluded that Na+ accesses its binding site through a high-field access channel resulting in a charge translocation, whereas the binding of the electroneutral proline induces a conformation alteration involving the displacement of charged amino acid residue(s) of the protein; (3) Na+ and proline binding sites interact cooperatively with each other by increasing the affinity and/or the speed of binding of the respective cosubstrate; (4) proline binding proceeds in a two step process: low affinity (~ 0.9 mM) electroneutral substrate binding followed by a nearly irreversible electrogenic conformational transition; (5) membrane impermeable PCMBS inhibits both Na+ and proline binding to the inside-out orientated PutP transporter, indicating that rather than selectively blocking a specific binding site, PCMBS probably locks the enzyme in an inactive state. The possible targets for this SH-reagent are cysteines 281 and 344 located close to the cytoplasmic surface of the protein. Beyond it, transient electrical currents of PutP were also observed on the BLM after rapid addition of proline in the presence of Na+. This was possible by combining the conventional BLM technique with high-speed flash-photolysis of caged-proline. Indeed the signals on the BLM indicate the detection of a different underlying reaction process in comparison to the data achieved by the SSM technique. This has paved the way for supplemental information about the reaction cycle since it was possible to assign the flash-photolysis BLM signals to the proline binding step followed by the internalization of Na+ and proline into the liposome. Thereby it was found, that the presence of Na+ is indispensable and the time constant for the process is ~ 63 ms. Moreover, structure-function information about the Na+ and proline binding sites of PutP was obtained by investigating the functionally important amino acid residues Asp55, Gly63 and Asp187 with site-directed mutagenesis and the combined SSM technique. One finding is that the mutated proteins PutP-D55C and PutP-G63C showed no activity on the SSM. Therefore, it can be assumed that either both Asp55 and Gly63 are crucial for the structure of PutP protein, or they are located at or close to the Na+ and proline binding sites. Furthermore, the results obtained from PutP-D187N and PutP-D187C mutants on SSM suggest that Asp187 of PutP is likely to be involved in the Na+ binding at the cytoplasmic side of the backward running carrier. Taken together the results of the present work have substantially broadened the known picture of the Na+/proline transporter PutP thereby several steps of the reaction cycle were elucidated, and moreover, valuable insights into the structure-function relationship of the transporter have become available.
The technique of site-specific fluorescence labelling with Tetramethylrhodaminemaleimide (TMRM) in combination with two electrode voltage-clamp technique (TEVC), an approach that has been named voltage clamp fluorometry (VCF), has been used in this work to study the Na,K-ATPase. The TMRM dye has the ability to attach covalently to cysteine residues and it responds to changes in the hydrophobicity of its local environment. We exploited this property using a construct of the Na-pump in which the native, extracellularly accessible cysteines were removed and cysteine residues were introduced by site-directed mutagenesis in specific positions of the Na-pump. In this way it was possible to detect site-specific conformational rearrangements of the Na-pump in a time-resolved fashion within a native membrane environment. In particular this technique allows to resolve reactions with low electrogenicity that cannot be satisfactorily analyzed with purely electrophysiological techniques and to identify the conformations of the enzyme under specific ionic composition of the measuring buffers. We used VCF to study the influence that several cations like Na+, K+, NMG+, TEA+ and BTEA+ exert on the distribution of the Na,K-ATPase between several enzymatic intermediates and on some of the reactions related to cation transport. To this end we utilized the mutants N790C in the loop M5-M6 and the mutant E307C, T309C, L311C and E312C in the loop M3-M4. From the correspondence of the fluorescence changes with the activation and inhibition of pumping current, by K+ and ouabain respectively, and from the fact that in Na+/Na+ exchange conditions the voltage distribution of charge movement and fluorescence changes evoked by voltage jumps are in reasonable agreement we conclude that through the fluorescence signals measured from these mutants, we can indeed monitor conformational changes linked to transport activity of the enzyme. For the mutants N790 and L311, it was found that the Na+ dependence of the amplitude and kinetics of the fluorescence signal associated with the E1P-E2P transition is in agreement with the prediction of an access channel model describing the regulation of the access of extracellular Na+ to its binding site. In particular for the mutants E307 and T309 it was found that in Na+/Na+ exchange conditions, the conformational change tracked by the fluorescence was much slower than the charge relaxation at hyperpolarized potentials while the kinetics was very similar at depolarized potentials. This implies that at hyperpolarized potentials the conformational change connected to the E1P-E2P transition does not give a large contribution to the electrogenicity of the process which is also consistent with the access channel model. On the mutant N790C it was found that the external pH does not seem to have any effect on the E1P-E2P equilibrium even if it seems to modulate the fluorescence quantum yield of the dye. Fluorescence quenching experiments with iodide and D2O indicate that at hyperpolarized potentials the local environment of the mutant N790C, experiences a small change in the accessibility to water without major changes in the local electrostatic field ...
Sodium proton antiporters are ubiquitous membrane proteins found in the cytoplasmic and organelle membranes of cells of many different origins, including plants, animals and microorganisms. They are involved in cell energetics, and play primary roles in the homeostasis of intracellular pH, cellular Na+ content and cell volume. Adaptation to high salinity and/or extreme pH in plants and bacteria or in human heart muscles requires the action of such Na+/H+ antiporters. NhaA is the essential Na+/H+ antiporter for pH and Na+ homeostasis (at alkaline pH) in Escherichia coli and many other enterobacteria. NhaA is an electrogenic Na+/H+ antiporter that exchanges 2H+ for 1Na+ (or Li+). NhaA shares with many other prokaryotic and eukaryotic antiporters a very strong dependence on pH. In order to achieve three-dimensional structure of NhaA, the previously described NhaA protein preparation was modified: (i) the wild type bacterial strain (TA16) used for homologous over-expression of NhaA was replaced with a delta nhaA strain (RK20). As a result, the purity and homogeneity of the sample was significantly improved; (ii) the previously two-step purification procedure was shortened to a single step affinity chromatography purification; (iii) a wide-range screening of crystallisation conditions, more than 20,000, was performed; (iv) a Seleno-L-methionine (SeMet) NhaA derivative was produced in order to solve the phases during structure determination. In parallel, attempts of production and crystallisation of co-complexes composed of NhaA and antibody fragments have been made. Four different monoclonal antibodies were available against NhaA. Selected antibody fragments were produced and the stability of the complex analysed. Here, the crystal structure of the pH down-regulated secondary transporter NhaA of Escherichia coli is presented at 3.45 Å resolution. A negatively charged ion funnel opens to the cytoplasm and ends in the middle of the membrane at the putative ion-binding site. There, a unique assembly of two pairs of short helices connected by crossed, extended chains creates a balanced electrostatic environment. A possible mechanism is proposed: the binding of charged substrates causes electric imbalance inducing movements, which allow for a rapid alternating access mechanism. This ion exchange machinery is regulated by a conformational change elicited by a pH signal perceived at the cytoplasmic funnel entry. The structure represents a novel fold that provides two major insights: it reveals the structural basis for the mechanism of Na+/H+ exchange and its unique regulation by pH in NhaA and in many other similar antiporters. Furthermore, it is also important for the understanding of the architecture of membrane proteins in general. However, although many aspects of the ion-translocation mechanism and pH regulation are clarified by the NhaA structure, higher resolution structures with Li+ or Na+ bound are required for understanding the ligand binding and the translocation mechanism at the atomic level. The alkaline pH-induced conformation is essential to further understand the pH-control and proton access to the binding site.
In this thesis, we opened the door towards a novel estimation theory for homogeneous vectors and have taken several steps into this new and uncharted territory. Present state of the art for homogeneous estimation problems treats such vectors p 2 Pn as unit vectors embedded in Rn+1 and approximates the unit hypersphere by a tangent plane (which is a n-dimensional real space, thus having the same number of degrees of freedom as Pn). This approach allows to use known and established methods from real space (e.g. the variational approach which leads to the FNS algorithm), but it only works well for small errors and has several drawbacks: • The unit sphere is a two-sheeted covering space of the projective space. Embedding approaches cannot model this fact and therefore can cause a degradation of estimation quality. • Linearization breaks down if distributions are not highly concentrated (e.g. if data configurations approach degenerate situations). • While estimation in tangential planes is possible with little error, the characterization of uncertainties with covariance matrices is much more problematic. Covariance matrices are not suited for modelling axial uncertainties if distributions are not concentrated. Therefore, we linked approaches from directional statistics and estimation theory together. (Homogeneous) TLS estimation could be identified as central model for homogeneous estimation and links to axial statistics were established. In the first chapters, a unified estimation theory for the point data and axial data was developed. In contrast to present approaches, we identified axial data as a specific data model (and not just as directional data with symmetric probability density function); this led to the development of novel terms like axial mean vectors, axial variances and axial expectation values. Like a tunnel which is constructed from both ends simultaneously, we also drilled from the parameter estimation side towards directional/axial statistics in the second part. The presentation of parameter estimation given in this thesis deviates strongly from all known textbooks by presenting homogeneous estimation problems as a distinguished class of problems which calls for different estimation tools. Using the results from the first part, the TLS solution can be interpreted as the weighted anti-mean vector of an axial sample. This link allows to use our results from axial statistics; for instance, the certainty of the anti-mode (i.e. of the TLS solution!) can be described with a weighted Bingham distribution (see (3.91)). While present approaches are only interested in the eigenvector of the some matrix, we can now exploit the whole mean scatter matrix to describe TLS solution and its certainty. Algorithms like FNS, HEIV or renormalization were presented in a common context and linked to each other. One central result is that all iterative homogeneous estimation algorithms essentially minimize a series of evolving Rayleigh coefficients which corresponds to a series of (converging?) cost functions. Statistical optimization is only possible if we clearly identify every step as what it exactly is. For instance, the vague statement “solving Xp ... 0” means nothing but setting ˆp := arg minp pTXp pT p . We identified the most complex scenario for which closed form optimal solutions are possible (in terms of axial statistics: the type-I matrix weighted model). The IETLS approach which is developed in this thesis then solves general type-II matrix weighted problems with an iterative solution of a series of type-I matrix weighted problems. This approach also allows to built converging schemes including robust and/or constrained estimation – in contrast to other approaches which can have severe convergence problems even without such extensions if error levels are not low. Chapter 6 then is another big step forward. We presented the theoretical background of homogeneous estimation by introducing novel concepts like singular vector unbiasedness of random matrices and solved the problem of optimal estimation for correlated data. For instance, these results could be used for better estimation of local image orientation / optical flow (see section 7.2). At the end of this thesis, simulations and experiments for a few computer vision applications were presented; besides orientation estimation, especially the results for robust and constrained estimation for fundamental matrices is impressive. The novel algorithms are applicable for a lot of other applications not presented here, for instance camera calibration, factorization algorithm formulti-view structure from motion, or conic fitting. The fact that this work paved the way for a lot of further research is certainly a good sign.
The N-terminal domain (matrix protein or MA) of a retroviral Gag polyprotein precursor plays a critical role in several stages of the retrovirus life cycle. MA is involved in the effective membrane targeting, assembly and release of the immature viral particles from the infected cell. In order to understand the structural basis of these functions, the full length MA from Moloney Murine Leukemia Virus (MoMuLV) was purified and the solution structure of the MA MoMuLV was determined by means of heteronuclear high-resolution NMR spectroscopy and compared with that of the X-ray diffraction analysis as well as with the structures of several MA proteins from geterologous viruses. Structural features were also obtained from CD spectroscopy, dynamic light scattering, sedimentation velocity, differential scanning calorimetry and other methods. It was found that the MA MoMuLV globular core (residues 8-98) is comprised of 7 well-defined helices (five alpha-helices and two 310 helices), with the general fold typical for MA proteins from other retroviral species. The N-terminus (residues Met1-Leu7) and the C-terminal proline-rich part (residues Pro103-Tyr131) are not structured in solution. Although MA MoMuLV has a low sequence identity compared with other matrix proteins for which the three-dimensional structure is known, it was shown that its overall topology and pattern of secondary structural units is similar to other retroviral matrix proteins. The monomeric state is observed for the correctly folded MA MoMuLV in a variety of external conditions and protein concentrations, indicating that virion assembly starts with the plasma membrane targeting of the nascent Gag precursor. The denaturation of MA MoMuLV is irreversible and is connected with protein aggregation. For Moloney Murine Leukemia Virus (MoMuLV) a proteolytic processing of the R-peptide (last 16 amino acids from the C-terminus of the Envelope protein (Env)) has been described as a second mode of fusion and activation preceding the receptor contact between the viral particle and the cellular membrane. An interaction between the R-peptide and MA MoMuLV has been proposed, since the R-peptide and MA are localized at the inner part of the membrane. Therefore the interaction between 15N labelled purified MA MoMuLV and synthesized R-peptide has been investigated using high-resolution NMR. It was found that in water solution MA MoMuLV and R-peptide do not form a tight complex, but in a mature virion in the presence of membranes or other protein factors it might be possible. In the case of HIV-1 the cytoplasmic part (EnvC) of the Env protein is much longer than in other retroviruses and again as for MoMuLV little is known about the interaction between EnvC and HIV MA. Hence, the full length HIV MA, and the last 150 amino acids from HIV Env have been subcloned with suitable expression vectors, purified and analysed by native gel electrophoresis, a pull down assay and by high resolution NMR for the purpose to detect the complex formation of EnvC and HIV MA. Finally, after all those experiments, it was found that a stable complex is not formed, but a weak interaction between the two proteins can not be excluded.
I derive a general effective theory for hot and/or dense quark matter. After introducing general projection operators for hard and soft quark and gluon degrees of freedom, I explicitly compute the functional integral for the hard quark and gluon modes in the QCD partition function. Upon appropriate choices for the projection operators one recovers various well-known effective theories such as the Hard Thermal Loop/ Hard Dense Loop Effective Theories as well as the High Density Effective Theory by Hong and Schaefer. I then apply the effective theory to cold and dense quark matter and show how it can be utilized to simplify the weak-coupling solution of the color-superconducting gap equation. In general, one considers as relevant quark degrees of freedom those within a thin layer of width 2 Lambda_q around the Fermi surface and as relevant gluon degrees of freedom those with 3-momenta less than Lambda_gl. It turns out that it is necessary to choose Lambda_q << Lambda_gl, i.e., scattering of quarks along the Fermi surface is the dominant process. Moreover, this special choice of the two cutoff parameters Lambda_q and Lambda_gl facilitates the power-counting of the numerous contributions in the gap-equation. In addition, it is demonstrated that both the energy and the momentum dependence of the gap function has to be treated self-consistently in order to determine the imaginary part of the gap function. For quarks close to the Fermi surface the imaginary part is calculated explicitly and shown to be of sub-subleading order in the gap equation.
Background: The flavin in its FMN and FAD forms is a versatile cofactor that is involved in catalysis of most disparate types of biological reactions. These include redox reactions such as dehydrogenations, activation of dioxygen, electron transfer, bioluminescence, blue light reception, photobiochemistry (as in photolyases), redox signaling etc. Recently, hitherto unrecognized types of biological reactions have been uncovered that do not involve redox shuffles, and might involve the reduced form of the flavin as a catalyst. The present work addresses properties of reduced flavin relevant in this context. Results: N(5)-H exchange reactions of the flavin reduced form and its pH dependence were studied using the 15N-NMR-signals of 15N-enriched, reduced flavin in the pH range from 5 to 12. The chemical shifts of the N(3) and N(5) resonances are not affected to a relevant extent in this pH range. This contrasts with the multiplicity of the N(5)-resonance, which strongly depends on pH. It is a doublet between pH 8.45 and 10.25 that coalesces into a singlet at lower and higher pH values. From the line width of the 15N(5) signal the pH-dependent rate of hydrogen exchange was deduced. The multiplicity of the 15N(5) signal and the proton exchange rates are little dependent on the buffer system used. Conclusion: The exchange rates allow an estimation of the pKa value of N(5)-H deprotonation in reduced flavin to be ≥ 20. This value imposes specific constraints for mechanisms of flavoprotein catalysis based on this process. On the other hand the pK ≈ 4 for N(5)-H protonation (to form N(5)+-H2) would be consistent with a role of N(5)-H as a base.
G protein-coupled receptors (GPCRs) play regulatory roles in many different physiological processes and they represent one of the most important class of drug targets. However, due to the lack of three-dimensional structures, structure based drug design has not been possible. The major bottleneck in getting three-dimensional crystal structure of GPCRs is to obtain milligram quantities of pure, homogenous and stable protein. Therefore, during my Ph.D. thesis, I focused on expression, characterization and isolation of three GPCRs namely human bradykinin receptor subtype 2 (B2R), human angiotensin II receptor subtype 1 (AT1aR), and human neuromedin U receptor subtype 2 (NmU2R). These receptors were heterologously produced in three different expression systems (i.e. Pichia pastoris, insect cells and mammalian cells), biochemically characterized and subsequently solubilized and purified for structural studies The human bradykinin receptor subtype 2 (B2R) is constitutively expressed in a variety of cells, including endothelial cells, vascular smooth muscle cells and cardiomyocytes. Activation of B2R is important in pathogenesis of inflammation, pain, tissue injury and cardioprotective mechanisms. During this study, recombinant B2R was produced in methylotrophic yeast Pichia pastoris (3.5 pmol/mg), insect cells (10 pmol/mg) and mammalian cells (60 pmol/mg). The recombinant receptor was characterized in terms of [3H] bradykinin binding, G protein coupling, localization, and glycosylation. Subsequently, it was solubilized and purified using affinity chromatography. Homogeneity and stability of purified B2R was monitored by gel filtration analysis. Milligram amounts of pure and stable receptor were obtained from BHK cells and Sf9 cells, which were used for three-dimensional crystallization attempts. The second receptor, which I worked on, is human angiotensin II receptor subtype 1 (AT1aR). AT1aR is distributed in smooth muscle cells, liver, kidney, heart, lung and testis. Activation of AT1aR is implicated in the regulation of blood pressure, hypertension and cardiovascular diseases. Recombinant AT1aR was produced at high levels in Pichia pastoris (167 pmol/mg), while at moderate levels in insect cells (29 pmol/mg) and mammalian cells (32 pmol/mg). The recombinant receptor was characterized in terms of [3H] angiotensin II binding, localization, and glycosylation. Subsequently, the receptor was solubilized and purified using affinity chromatography. Homogeneity and stability of purified AT1aR was monitored by gel filtration analysis. Milligram amounts of pure and stable receptor were obtained from Pichia pastoris, which were used for threedimensional crystallization attempts. In addition to B2R and AT1aR, I also attempted to produce and isolate the human neuromedin U receptor subtype 2 (NmU2R), which was deorphanized recently. It is found in highest abundance in the central nervous system, particularly the medulla oblongata, spinal cord and thalamus. The distribution of this receptor suggests its regulatory role in sensory transmission and modulation. During this study, recombinant NmU2R was produced in Pichia pastoris (6 pmol/mg) and BHK cells (9 pmol/mg). Recombinant receptor was characterized with regard to [125I] NmU binding, localization and glycosylation. Subsequently, the receptor was solubilized and purified using affinity chromatography. Due to its low expression level, further expression optimization is required in order to obtain milligram amounts for structural studies. The long-term goal of this study was to obtain three-dimensional crystal structure of recombinant GPCRs. However, 3-dimensional crystallization of human recombinant membrane proteins still remains a difficult task. On the other hand, recent advances in the solid-state NMR spectroscopy offer ample opportunities to study receptor-ligand systems, provided milligram quantities of purified receptor are available. Therefore, in parallel to 3-dimensional crystallization trials, purified B2R was also used for solid-state NMR analysis in order to investigate the receptor bound conformation of bradykinin. Preliminary results are promising and indicate significant structural changes in bradykinin upon binding to B2R. Further experiments are ongoing and will hopefully result in the structure of receptor bound bradykinin. One of the challenges in GPCR crystallization is the small hydrophilic surface area that is available to make crystal contacts. One possibility to overcome this problem can be the reconstitution of a GPCR complex with an interacting protein for cocrystallization. For this purpose, I coexpressed B2R and AT1aR, which form a stable heterodimer complex, in BHK cells. I could successfully isolate the heterodimer complex by using two-step affinity purification. Unfortunately, this complex was not stable over time and disassociates within three days of purification. However, during coexpression of B2R and AT1aR in BHK cells, I observed that B2R was localized in the plasma membrane in coexpressing cells while it was retained intracellularly when expressed alone. This coexpression of AT1aR with B2R resulted in a four-fold increase in [3H] bradykinin binding sites on the cell surface. In addition, these two receptors were cointernalized in response to their individual specific ligands. Interestingly, colocalization of B2R and AT1aR was also found in human foreskin fibroblasts (which endogenously express both receptors), in line with the possibility that heterodimerization may be required for surface localization of B2R in native tissues as well. This is the first report where surface localization of a peptide GPCR is triggered by a distantly related peptide GPCR. These data support the hypothesis that heterodimerization may be a prerequisite for cell surface localization of some GPCRs. A second approach that I followed to stabilize the purified B2R was to reconstitute the B2R-β-arrestin complex. β-arrestin is a cytosolic protein that participates in agonist mediated desensitization of GPCRs and therefore dampens the cellular responses initiated by the activation of GPCRs. I tried to reconstitute B2R-β-arrestin complex in vitro by mixing purified B2R and purified β-arrestin. But, no interaction of these two proteins was observed in the pull-down assays. However, a C-terminal mutant of B2R (where a part of the C-terminus of the B2R is exchanged with that of the vasopressin receptor) was found to interact with β-arrestin in vitro as revealed by pull-down assays. In conclusion, this work establishes the production, characterization and isolation of three recombinant human GPCRs. Recombinant receptors were produced in milligram amounts and therefore, pave the way for structural analysis. The heterodimer complex of B2R-AT1aR and B2R-β-arrestin complex can be of great help during crystallization. In addition, it was also found for the first time that the surface localization of a peptide GPCR can be triggered by heterodimerization with a distantly related peptide GPCR.
We evaluate the importance of the precautionary saving motive by relying on a direct question about precautionary wealth from the 1995 and 1998 waves of the Survey of Consumer Finances. In this survey, a new question has been designed to elicit the amount of desired precautionary wealth. This allows us to assess the amount of precautionary accumulation and to overcome many of the problems of previous works on this topic. We find that a precautionary saving motive exists and affects virtually every type of household. However, precautionary savings account for only 8 percent of total wealth holdings. Even though this motive does not give rise to large amounts of wealth, particularly for young and middle-age households, it is particularly important for two groups: older households and business owners. Overall, we provide strong evidence that we need to take the precautionary saving motive into account when modeling saving behavior. Klassifizierung: D91, E21, C21
In the classical Dirac equation with strong potentials, called overcritical, a bound state reaches the negative continuum. In QED the presence of a static overcritical external electric field leads to a charged vacuum and indicates spontaneous particle creation when the overcritical field is switched on. The goal of this work is to clarify whether this effect exists, i.e. if it can be uniquely defined and proved, in time-dependent physical processes. Starting from a fundamental level of the theory we check all mathematical and interpretational steps from the algebra of fields to the very effect. In the first, theoretical part of this thesis we introduce the mathematical formulation of the classical and quantized Dirac theory with their most important results. Using this language we define rigorously the notion of spontaneous particle creation in overcritical fields. First, we give a rigorous definition of resonances as poles of the resolvent or the Green's function and show how eigenvalues become resonances under Hamiltonian perturbations. In particular, we consider essential for overcritical potentials perturbation of eigenvalues at the edge of the continuous spectrum. Next, we gather various adiabatic theorems and discuss well-posedness of the scattering in the adiabatic limit. Then, we construct Fock space representations of the field algebra, study their equivalence and give a unitary implementer of all Bogoliubov transformations induced by unitary transformations of the one-particle Hilbert space as well as by the projector (or vacuum vector) changes as long as they lead to unitarily equivalent Fock representations. We implement in Fock space self-adjoint and unitary operators from the one-particle space, discussing the charge, energy, evolution and scattering operators. Then we introduce the notion of particles and several particle interpretations for time-dependent processes with a different Fock space at every instant of time. We study how the charge, energy and number of particles change in consequence of a change of representation or in implemented evolution or scattering processes, what is especially interesting in presence of overcritical potentials. Using this language we define rigorously the notion of spontaneous particle creation. Then we look for physical processes which show the effect of vacuum decay and spontaneous particle creation exclusively due to the overcriticality of the potential. We consider several processes with static as well as suddenly switched on (and off) static overcritical potentials and conclude that they are unsatisfactory for observation of the spontaneous particle creation. Next, we consider properties of general time-dependent scattering processes with continuous switch on (and off) of an overcritical potential and show that they also fail to produce stable signatures of the particle creation due to overcriticality. Further, we study and successfully define the spontaneous particle creation in adiabatic processes, where the spontaneous antiparticle is created as a result of a resonance (wave packet) decay in the negative continuum. Unfortunately, they lead to physically questionable pair production as the adiabatic limit is approached. Finally, we consider extension of these ideas to non-adiabatic processes involving overcritical potentials and argue that they are the best candidate for showing the spontaneous pair creation in physical processes. Demanding creation of the spontaneous antiparticle in the state corresponding to the overcritical resonance rather quick than slow processes should be considered, with a possibly long frozen overcritical period. In the second part of this thesis we concentrate on a class of spherically symmetric square well potentials with a time-dependent depth. First, we solve the Dirac equation and analyze the structure and behaviour of bound states and appearance of overcriticality. Then, by analytic continuation we find and discuss the behaviour of resonances in overcritical potentials. Next, we derive and solve numerically (introducing a non-uniform continuum discretization for a consistent treatment of narrow peaks) a system of differential equations (coupled channel equations) to calculate particle and antiparticle production spectra for various time-dependent processes including sudden, quick, slow switch on and off of a sub- and overcritical potentials. We discuss in detail how and under which conditions an overcritical resonance decays during the evolution giving rise to the spontaneous production of an antiparticle. We compare the antiparticle production spectrum with the shape of the resonance in the overcritical potential. We study processes, where the overcritical potentials are switched on at different speed and are possibly frozen in the overcritical phase. We prove, in agreement with conclusions of the theoretical part, that the peak (wave packet) in the negative continuum representing a dived bound state partially follows the moving resonance and partially decays at every stage of its evolution. This continuous decay is more intensive in slow processes, while in quick processes the wave packet more precisely follows the resonance. In the adiabatic limit, the whole decay occurs already at the edge of the continuum, resulting in production of antiparticles with vanishing momentum. In contrast, in quick switch on processes with delay in the overcritical phase, the spectrum of the created antiparticles agrees best with the shape of the resonance. Finally, we address the question how much information about the time-dependent potential can be reconstructed from the scattering data, represented by the particle production spectrum. We propose a simple approximation method (master equation) basing on an exponential, decoherent decay of time-dependent resonances for prediction of particle creation spectra and obtain a good agreement with the results of full numerical calculations. Additionally, we discuss various sources of errors introduced by the numerical discretization, find estimations for them and prove convergence of the numerical schemes.
This thesis is devoted to the study of Micro Structured Electrode (MSE) sustained discharges. Innovative approaches in this work are i) the implementation of MSE arrays for high-pressure plasma generation and ii) the use of diode laser atomic absorption spectroscopy for investigating sub-millimetric discharges. By means of MSE arrays the discharge gap is scaled down to the sub-millimetric range and accordingly the working pressure could be increased up to atmospheric. It should be underlined that besides the ease of use, since expensive vacuum equipment is not required, high-pressure discharges offer also a high density of active species. A MSE consists of holes, regularly distributed in a composite sheet made of two metal layers separated by an insulator. The electrodes and insulator thickness and the diameter of the holes are in the 100 micrometer range. Based on these microstructures stable non-filamentary DC discharges are generated in noble gases and gas mixtures at pressures up to 1000 mbar. The MSE sustained discharge can be considered as a normal glow discharge whereby the excitation and ionization efficiency is increased by the specific electrode configuration (hollow cathode geometry). Large area high-pressure plasma can be achieved by parallel operation of a large number of microdischarges. Parallel operation of up to 200 microdischarges without individual ballast was proven for pressures up to 300 mbar. Furthermore, arrays of resistively decoupled microdischarges were operated up to atmospheric pressure. Spectral investigations have revealed the presence of highly energetic electrons (20 eV), a large density of atoms in metastable states (1013 cm-3) and a high electron density (1015 cm-3). Although the plasma confined inside the hole of the MSE may reach gas temperatures up to 1000 K, the ambient gas temperature immediately above the microstructure exceeds only slightly the room temperature. The reactivity of the MSE sustained discharge was demonstrated in respect to waste gas decomposition and surface treatment. The MSE arrays are providing a non-equilibrium high-pressure plasma, which is very promising for surface processing, plasma chemistry and generation of UV radiation.
We present a detailed study of chemical freeze-out in p-p, C-C, Si-Si and Pb-Pb collisions at beam momenta of 158A GeV as well as Pb-Pb collisions at beam momenta of 20A, 30A, 40A and 80A GeV. By analyzing hadronic multiplicities within the statistical hadronization model, we have studied the parameters of the source as a function of the number of the participating nucleons and the beam energy. We observe a nice smooth behaviour of temperature, baryon chemical potential and strangeness under-saturation parameter as a function of energy and nucleus size. Interpolating formulas are provided which allow to predict the chemical freeze-out parameters in central collisions at centre-of-mass energies > 4.5 GeV and for any colliding ions. Specific discrepancies between data and model emerge in particle ratios in Pb-Pb collisions at SPS between 20A and 40A GeV of beam energy which cannot be accounted for in the considered model schemes.
One possible approach to study systematically the influence of the deformation regime on the geometry of geological structures like folds and boudins is analogue modelling. For a complete understanding of the resulting structures, consideration of the third dimension is required. This PhD study deals with scaled analogue modelling under constriction and plane-strain conditions to improve our knowledge of folding and boudinage of lower crustal rocks in space and time. Plasticine is an appropriate analogue material for rocks in the lower crust. Therefore, this material was used for the experiments. The macroscopic behaviour of most types of plasticine is quite similar to rocks undergoing strain-rate softening and strain hardening regardless of the different microscopic aspects of deformation. Therefore, if one is aware that the stress exponent and viscosity increase with increasing strain, the original plasticine types used with stress exponents ranging from 5.8 to 8.0 are adequate for modelling geologic structures. The same holds for plasticine/oil mixtures. Thus, plasticine and plasticine/oil mixtures can be used to model the viscous flow of different rock types in the lower crust. If climb-accommodated dislocation creep and associated steady-state flow is assumed for the natural rocks, the plasticine/oil mixtures should be used, which flow under steady-state conditions. Three different experimental studies of plane-strain coaxial deformation of stiff layers, with viscosity η2 and stress exponent n2, embedded in a weak matrix, with viscosity η1 and stress exponent n1, have been carried out. The undeformed samples (matrix plus layer) were cubes with an edge length of 12 cm. All experimental runs have been carried out at T = 25 ± 1°C and varying strain rates ė, ranging from 7.9 x 10 high -6 s high -1 to 1.7 x 10 high -2 s high -1, until a finite longitudinal strain of 30% – 40% was achieved. The first experimental study improved the understanding about the evolution of folds and boudins when the layer is oriented perpendicular to the Y-axis of the finite strain ellipsoid. The rock analogues used were Beck’s green plasticine (matrix) and Beck’s black plasticine (competent layer), both of which are strain-rate softening modelling materials with stress exponent n = ca. 8. The effective viscosity η of the matrix plasticine was changed by adding different amounts of oil to the original plasticine. At a strain rate ė of 10 high -3 s high -1 and a finite strain e of 10%, the effective viscosity of the matrix ranges from 1.2 x 10 high 6 to 7.2 x 10 high 6 Pa s. The effective viscosity of the competent layer has been determined as 4.2 x 10 high 7 Pa s. If the viscosity ratio is large (> ca. 20) and the initial thickness of the competent layer is small, both folds and boudins develop simultaneously. Although the growth rate of the folds seems to be higher than the growth rate of the boudins, the wavelength of both structures is approximately the same as is suggested by analytical solutions. A further unexpected, but characteristic, aspect of the deformed competent layer is a significant increase in thickness, which can be used to distinguish plane-strain folds and boudins from constrictional folds and boudins. In the second experimental study, the impact of varying strain rates on growing folds and boudins under plane strain have been investigated. The strain rates used range from 7.9 x 10 high -6 s high -1 to 1.7 x 10 high -2 s high -1. The stiff layer and matrix consist of non-linear viscous Kolb grey and Beck’s green plasticine, respectively, both of which are strain-rate softening modelling materials with power law exponents (n) and apparent viscosities (η) ranging from 6.5 to 7.9 and 8.5 x 10 high 6 to 7.2 x 10 high 6 Pa s, respectively. The effective viscosity (η) of the matrix plasticine was partly modified by adding oil to the original plasticine. At the strain rates used in the experiments the viscosity ratio between layer and matrix ranges between 3 and 10. Different runs have been carried out where the layer was oriented perpendicular to the principal strain axes (X>Y>Z). The results suggest a considerable influence of the strain rate on the geometry of the deformed stiff layer including its thickness. This holds for every type of layer orientation (S ┴ X, S ┴ Y, S ┴ Z). If the stiff layer is oriented perpendicular to the short axis Z of the finite strain ellipsoid, the number of the resulting boudins and the thickness of the stiff layer increase, whereas the length of boudins decreases with increasing strain rate. If the stiff layer is oriented perpendicular to the long axis, X, of the finite strain ellipsoid, enlargement of the strain rate results in increasing wavelength of folds, whereas the number of folds and the degree of thickening of the stiff layer decreased. If the stiff layer is oriented perpendicular to the intermediate Y-axis of the finite strain ellipsoid enlargement of the strain rate results in a decreasing number of boudins and folds associated with increasing wavelengths of both structures. The wavelength of folds is approximately half of the boudins wavelength. This is true for the case where folds and boudins develop simultaneously (S ┴ Y) and for cases where both structures develop independently (folds at S ┴ X and boudins at S ┴ Z). In the third experimental study, scaled analogue experiments have been carried out to demonstrate the growth of plane-strain folds and boudins through space and time. Previous 3D-studies are based only on finite deformation structures. Their results can therefore not be used to prove if both structures grew simultaneously or in sequence. Plane strain acted on a single stiff layer that was embedded in a weak matrix, with the layer oriented perpendicular to the intermediate Y-axis of the finite strain ellipsoid. Two different experimental runs have been carried out using computer tomography (CT) to analyse the results. The first run was carried out without interruption. During the second run, the deformation was stopped in each case at longitudinal strain increments of 10%. Every experiment was carried out at a temperature T of 25°C and a strain rate, ė, of ca. 4 x 10 high -3 s high -1 until a finite longitudinal strain of 40% was achieved with a viscosity contrast m of 18.6 between the non-linear viscous layer (Kolb brown plasticine) and the matrix (Beck’s green plasticine with 150 ml oil kg high -1). The apparent viscosity, η, and the stress exponent, n, for the layer at a strain rate ė = ca. 10 high -3 s high -1 and a finite strain e = 10% are 2.23 x 10 high 7 Pa s and n = 5.8 and for the matrix 1.2 x 10 high 6 Pa s and 10.5. These new data that result from incremental analogue modelling corroborate previous suggestions that folds and boudins are coeval structures in cases of plane-strain coaxial deformation with the stiff layer oriented perpendicular to the intermediate Y-axis of the finite strain ellipsoid. They will be of interest for all workers who are dealing with plane-strain boudins and folds, where the fold axes are parallel to the major axis (X) of the finite strain ellipsoid. As has been demonstrated by the first experimental study, coeval folding and boudinage under plane strain, with S ┴ Y, are associated with a significant increase in the thickness of the competent layer. The latter phenomenon does not occur in other cases of simultaneous folding and boudinage, such as bulk pure constriction. To study the impact of layer thickness on the geometry of folds and boudins under pure constriction, we carried out additional experiments using different types of plasticine for a stiff layer and a weaker matrix to model folding and boudinaging under pure constriction, with the initially planar layer oriented parallel to the Xaxis of the finite strain ellipsoid. The stiff layer and matrix consist of non-linear viscous Kolb brown and Beck’s green plasticine, respectively, both of which are strain-rate softening modelling materials. Six runs have been carried out using thicknesses of the stiff layer of 1, 2, 4, 6, 8 and 10 ± 0.2 mm. All experimental runs were carried out at a temperature T of 30 ± 2°C and a strain rate, ė, of ca. 1.1 x 10 high -4 s high -1 until a finite longitudinal strain of 40% was achieved with a viscosity contrast m of 3.1 between the stiff layer (Kolb brown plasticine) and the matrix (Beck’s green plasticine). The apparent viscosity, η, and the stress exponent, n, for the layer at a strain rate ė = ca. 10 high -3 s high -1 and a finite strain e = 10% are 2.23 x 10 high 7 Pa s and n = 5.8 and for the matrix 7.2 x 10 high 6 Pa s and 7.9. Our results suggest a considerable influence of the initial thickness of the stiff layer on the geometry of the deformed stiff layer. There is no evidence for folding in XY=XZ-sections if the initial thickness of the competent layer is larger than ca. 8 mm. If the initial thickness of the competent layer is set at ca. 10 ± 0.2 mm, both folds and boudins develop simultaneously. However, the growth rate of the boudins seems to be higher than the growth rate of the folds. A further expected, but characteristic, aspect of the deformed competent layer is no change in thickness of the competent layer, which can be used to distinguish plane-strain folds and boudins from constrictional folds and boudins. The model results are important for the analysis and interpretation of deformation structures in rheologically stratified rocks undergoing dislocation creep under bulk constriction. Tectonic settings where constrictional folds and boudins may develop simultaneously are stems of salt diapirs, subduction zones or thermal plumes. To make (paleo) viscosimetric statements possible, the rheological data of the different plasticine types were related to the geometrical data. When comparing the normalized dominant wavelength Wd obtained from the deformed layer of the models with the theoretical dominant wavelength (Ld) calculated using the Smith equation (1977, 1979), the latter probably also holds when folding and boudinage develop simultaneously (S ┴ Y) and when boudins develop independently (S ┴ Z), but can obviously not be applied at very low viscosity ratios as is indicated by the low-strain-rate experiments.
Acute myeloid leukemia (AML) is characterized by the accumulation of a large number of abnormal, immature blast cells. Recently, histone deacetylase inhibitors (HDIs) received considerable interest on the ground of their ability to overcome the differentiation block in these leukemic blasts regardless of the primary genetic alteration, an effect achieved either alone or in combination with differentiating agents, such as all-trans retinoic acid (t-RA). Valproic acid (VPA), a potent HDI, is now under clinical evaluation owing to its potent differentiation effect on transformed hematopoietic progenitor cells and leukemic blasts from AML patients. Conversely, in a clinical study by Bug et al., the favorable effects of the combination treatment with t-RA/VPA in advanced acute myeloid leukemia patients were reported to be most likely due to an enhancement of nonleukemic myelopoiesis and the suppression of malignant hematopoiesis rather than enforced differentiation of the leukemic cells. Based on the hypothesis that VPA influences normal hematopoiesis, the effect of chromatin modeling through VPA on HSCs was investigated with respect to differentiation, proliferation as well as self-renewal in the present study. It has been shown that valproic acid increases both proliferation and self-renewal of HSC. It accelerates cell cycle progression of HSC accompanied by a down-regulation of p21cip-1/waf-1. Furthermore, valproic acid inhibits GSK3B by phosphorylation on Ser9 accompanied by an activation of the Wnt signaling pathway as well as by an up-regulation of HoxB4, a target gene of Wnt signaling. Both are known to directly stimulate the proliferation of HSC and to expand the HSC pool. To sum up, valproic acid, a potent histone deacetylase inhibitor known to induce differentiation and/or apoptosis in leukemic blasts, stimulates the proliferation and self-renewal of hematopoietic stem cells. Therefore, the data reported in this study suggest to reconsider the role of histone deacetylase inhibitors from a differentiation inducer to a coadjuvant factor for increasing the response to conventional therapy in acute myeloid leukemia.
Haematopoietic stem cells (HSCs) are regarded as the prime target for gene therapy of inherited and acquired disorders of the blood system, e.g. X-linked chronic granulomatous disease (X-CGD). The major reason for this is that HSCs posses the ability to self renew as well as the potential to differentiate into all lineage-specific cell types. However, the need to reach and to maintain sufficient therapeutic levels of genetically modified stem cells and their progeny after gene delivery still presents major challenges for current HSC gene therapy approaches. In particular, one of the main limitations for most genetic defects is the lack of a selective growth advantage of gene-modified cells after engraftment. In vitro and in vivo methods have been developed that focus on either positive or negative selection of HSCs. An artificial selection advantage can be conferred to transduced HSCs by incorporating a selection marker in addition to the therapeutic transgene. In the present study, two novel strategies for positive selection of murine gp91phox gene-modified haematopoietic stem cells were developed and tested, bearing in mind that with selective growth advantage, the possibility of uncontrolled proliferation arises. The first strategy to be investigated was based on the homeobox transcription factor HOXB4, which plays an important role in the control of haematopoietic stem cell proliferation and differentiation. Overexpression of a retroviral bicistronic construct containing the therapeutic gene gp91phox and HOXB4 in murine primary bone marrow cells led to a significant 3–4-fold expansion of transduced cells ex vivo. The numbers of transgene-expressing cells increased 2–3-fold after 2 weeks cultivation under cytokine stimulation. Furthermore, the clonogenic progenitor cell assay (CFU assay) demonstrated that the number of colony-forming cells had increased to levels 2-fold higher than those of mock-transduced cells after 1 week of culture, thereby augmenting the presence of a significant number of stem/progenitor cells in the selected cell population. However, in our experiments, HOXB4-overexpressing murine HSCs did not show any repopulating advantage in transplanted recipient mice over control construct-transduced HSCs. These results indicate that selective expansion of gp91phox gene-modified HSCs can be induced by the HOXB4 transcription factor ex vivo but not in vivo. This is possibly dependent on HOXB4 expression levels, which are too low in vivo to achieve selection. The second strategy made use of a chemically inducible dimerizer system consisting of the therapeutic gene gp91phox and a fusion protein, containing sequences from a growth factor receptor signalling domain (epidermal growth factor receptor, EGFR, or prolactin receptor, PrlR) and the drug binding protein FKBP12, as the selection cassette. This strategy aimed to allow inducible selection that could be easily switched off. The activity of these fusion proteins is controlled through the small molecular dimerizer AP20187. Transduction of BaF/3 cells with lentiviral vectors expressing the EGFR construct induced proliferation and led to complete selection within 18 days (99%). However, removing AP20187 could not turn off proliferation. This construct is, therefore, not suitable as a selection cassette for the expansion of gene-modified HSCs due to its oncogenic potential. Transduction of the construct containing the intracellular domain of PrlR caused significant selective expansion of AP20187-treated BaF/3 cells. Following expression in cells, the fusion protein, which lacks membrane-anchoring sequences, mainly localized to the cytoplasm. Evidence was found to indicate that activated STAT5 might be responsible for this effect. Upon expression of the prolactin construct, phosphorylation of STAT5 and its DNA-binding activity to a ß-casein promoter sequence was strongly increased. Importantly, the induced proliferation was reversible after removal of AP20187. Transduced Sca1+ bone marrow cells obtained from C57BL/6-CD45.1 mice could be expanded about 20–100-fold ex vivo in the presence of AP20187 and mSCF without losing progenitor cell features and the capability to contribute to all lineages of the haematopoietic system. To exclude oncogenic outgrowth of one single clone, the polyclonality of selected cells was proven by ligation-mediated PCR (LM-PCR) analysis. In mouse transplantation experiments, ex vivo-expanded cells repopulated the bone marrow of lethally irradiated mice suggesting that the ex vivo expansion took place at the level of haematopoietic stem and/or progenitor cells. Genomic gp91phox sequences were detected in the bone marrow, spleen and peripheral blood cells of transplanted animals, indicating that gp91phox-containing cells most likely contributed to the reconstitution of haematopoiesis in these mice.