Refine
Has Fulltext
- yes (43)
Is part of the Bibliography
- yes (43)
Year of publication
- 2012 (43) (remove)
Document Type
- Doctoral Thesis (43)
Keywords
- Dendritische Zelle (5)
- Apoptosis (4)
- Maus (4)
- G-Protein gekoppelte Rezeptoren (3)
- Myc (3)
- Pneumolysin (3)
- Arzneimitteldesign (2)
- Bakteriengift (2)
- Dendritic cell (2)
- E. coli Nissle 1917 (2)
Institute
- Graduate School of Life Sciences (43) (remove)
Stimulatory or superagonistic (SA) CD28-specific monoclonal antibodies (mAbs) are potent polyclonal activators of regulatory T cells and have proven highly effective as treatment in a wide range of rodent models for autoimmune and inflammatory diseases. In these models, a preferential activation of regulatory T cells was observed by in vivo administration of CD28SA. In stark contrast, human volunteers receiving TGN1412, a humanized CD28-specific mAb, experienced a life-threatening cytokine release syndrome during the first-in-man trial. Preclinical tests employing human peripheral blood mononuclear cells (PBMC) failed to announce the rapid cytokine release measured in the human volunteers in response to TGN1412. The aim of this thesis project was to find an explanation of why standard PBMC assays failed to predict the unexpected TGN1412-induced "cytokine storm" observed in human volunteers. CD28 superagonists can activate T cells without T cell receptor (TCR) ligation. They do depend, however, on “tonic” TCR signals received by MHC scanning, signals that they amplify. PBMC do not receive these signals in the circulation. Short-term in vitro preculture of human PBMC at a high cell density (HDC) resulted in massive cytokine release during subsequent TGN1412 stimulation. Restoration of reactivity was cell-contact dependent, associated with TCR polarization and tyrosine-phosphorylation, and blocked by HLA-specific mAb. In HDC, both CD4 T cells and monocytes functionally mature in a mutually dependent fashion. However, only CD4 memory T-cells proliferate upon TGN1412 stimulation, and were identified as the main source of pro-inflammatory cytokines. Importantly, responses to other T-cell activating agents were also enhanced if PBMC were first allowed to interact under tissue-like conditions. A new in vitro protocol is provided that returns circulating T-cells to a tissue-like status where they respond to TGN1412 stimulation, and it might represent a more reliable preclinical in vitro test for both activating and inhibitory immunomodulatory drugs. Finally, the surprising observation was made that the IgG1 “sibling” of TGN1412, which is of the poorly Fc receptor-binding IgG4 isotype, has a much lower stimulatory activity. We could exclude steric hindrance as an explanation and provide evidence for removal of TGN1112 from the T-cell surface by trans-endocytosis.
Autoimmune diseases, unwanted overshooting immune responses against self antigens, are due to an imbalance in immunity and tolerance. Although negatively impacting cancer prognosis, myeloid derived suppressor cells (MDSC), with their potent suppressive capabilities, might be applicable in a more beneficial light when applied in to autoimmunity. As previous shown MDSC have protective roles in Experimental Autoimmune Encephalomyelitis (EAE) (Zhu et al., 2007), the established inducible mouse model for the autoimmune disease multiple sclerosis (MS). This decrease in disease severity indicates in vitro generated immature myeloid cells (IMC) from bone marrow (BM) as precursors of MDSC are promising candidates for cellular therapy. Important to any cellular therapy by adoptive transfer, the major questions regarding IMC efficacy was addressed within the thesis. This thesis attempts to elucidate how IMC operate in EAE. This thesis defines the factors within the autoimmune microenvironment that lead to the activation of MDSC, where IMC home once delivered in vivo, and the protective mechanisms BMIMC employ. To emulate BM cells when they first enter circulation through the blood, IMC were injected intravenously (i.v.). IMC are protective with no regard to the various routes delivered (i.v., i.p.). They protect to a lesser extent when pre-activated before injection. IMC suppress by causing a delay and/or by decreasing the severity of the disease via a mechanism yet determined. To understand the migration pattern of IMC after i.v. injection, in vivo kinetics experiments employing bioluminescence imaging were performed. This techinique allows for whole in vivo mouse imaging daily, allowing the tracking of cell migration over days within a single mouse. During steady-state, BMIMC circulate and appear to accumulate in the spleen by day 4 after injection, whereas they alternatively home to inflammatory sites (immunization site), draining lymph nodes, and the spleen within mice with low grade EAE. Visualization of CMDiI-labelled BMIMC by fluorescence microscopy could locate IMC injected cells outside the white pulp, as they were colocalizing in the regions stained with CD169 or outside, but not within the follicles of spleens on day 4. Consistant with these findings, the attempt to analyze the phenotype of these cells by flow cytometry was problematic as these cells seem to adhere strongly to collagen also indicating the cells are located in the collagenous area of the marginal zone and the red pulp.To determine factors influencing MDSC activation, we utilized different stimuli through a high throughput method detecting release of nitric oxide (NO). Extracts from yeast, fungi, and bacteria were observed to activate MDSC to produce nitric oxide. Surprisingly, material mimicking viral DNA (CpG) and RNA (poly I:C), and several self glycolipids, could not activate the MDSC to produce NO. Upon attempts to understand synergistic effects between microbial pathogens and host cytokines, IFNg was determined to boost the signal of pathogen stimuli, whereas IL17, another cytokine which causes pathology during EAE, and IFNb, a drug used in therapy to treat MS, did not cause any additional effects. Activation of MDSC was determined by the microbial pathogens components LPS, curdlan, and zymosan, to induce upregulation of B7H1 on the cell surface. MDSC did not increase any co-stimulatory markers, such as CD40, CD80, CD86, CD70, or the co-inhibitory marker, PDL2. On day 1 after EAE induction, endogenous MDSC populations when stimulated showed an increase in B7H1 expression and a downregulation of CD80. After further analysis, these cells were concluded to be mostly granulocytic cells (Ly6G+). As the B7H1 ligand PD1 is upregulated in chronic diseases and correlates to an exhausted phenotype, the PD1 : B7H1 interaction was a good candidate for the mechanism our cells may employ for their suppressive capacity. To investigate this interaction, fixed BM-IMC deficient in B7H1 were incubated with restimulated memory T cells. IMC deficient in B7H1 resulted in a significant loss of T cell suppression, as compared to the wildtype control BMIMC. To assess this interaction in vivo, we injected wildtype (WT) and B7H1-/- IMC into mice followed by induction of EAE to assess whether B7H1 mediated this suppression. The lack of B7H1 did not alter their suppressive capacity under these conditions, contrary to other findings which have described this interaction to be important in their suppressive capacity when administered post EAE induction (Ioannou et al., 2012). Interestingly, EAE mice pre-treated with IMC had similar amounts of cytokine production in the CNS after restimulation. Spleens from IMC injected mice had increased amounts of Arg-1 suggesting suppression is via oxidation or recruitment by soluble mediators may lead to this protection. We speculate this may inhibit T cell reactivation in the CNS.
Dendritic cell-based vaccination is a well established technique for preventive and therapeutic instruction of the immune system where conservative vaccine formulations fail to cure or prevent diseases, respectively. Efficiency of this technique already was demonstrated in infectious diseases as well as for cancer in animal or human studies. Well controlled manipulation and antigen-loading of immature DC is most beneficial to this technique. But, time-consuming and cost-extensive procedures for preparation of DC precursors, expansion and stimulation of DC and inpatient administration are big disadvantages regarding vaccine development for pandemic infectious diseases that occur mainly in underdeveloped countries. Therefore vaccines are needed that are pathogen-tailored and able to induce equal immune responses as their DC-based vaccine models. For vaccination against Leishmania parasites such a DC-based vaccine is feasible and its efficacy to induce protective Th1-based immune responses was already demonstrated in several animal studies. But, one of our own studies indicated supportive activity of host cells exceeding the allocation of T cells to become activated by transferred DC. IL-12, an important cytokine for the induction of Th1-related immune responses, has to be produced by host cells. Therefore, the aim of this study was to investigate the mechanism of BMDC-based vaccination with regard to simplification of the vaccine formulation. Key questions that have been addressed are: Which cells process the information that is transferred by the injected DC and what are the key components of this information? Further more, it was looked at whether altered vaccine formulations are able to induce protective immunity and whether they share equal molecular mechanisms. The current paradigm of BMDC-based vaccination proposes direct interaction of transferred BMDC with host T cells. These BMDC have to be antigen-loaded for stimulation via antigen-peptide-MHC molecule-complexes and they have to be activated for proper co-stimulation of T cells. Here, this study demonstrates that neither activation for co-stimulation nor direct interaction with adequate MHC molecules is needed for the induction of protective immunity against infection with Leishmania-parasites. Disrupted antigen-loaded BMDC are able to induce protective immunity in BALB/c mice without pre-stimulation via CpG ODN. Beyond, if BMDC were used with a different MHC-background than recipient mice then the vaccine still would be efficient in terms of reduction of footpad swelling and parasite load in draining lymph nodes. Even more, DC-specific features are no key component that leads to protective immunity as vaccination with disrupted antigen-loaded MΦ shows equal properties than before mentioned vaccine formulations. Further more, it was found that host DC play a major role in transforming the incoming signal, received from transferred antigen-loaded DC, into Th1-related stimuli and Leishmania-antigen-specific T cell activation. Suspensions of disrupted antigen-loaded DC resemble a combination of laid off soluble molecules together with exosome-like vesicles that formed after disruption of membranes. Here it was shown that separation of the membranous and soluble fractions and subsequent transfer into BALB/c mice will lead to protection of these mice against infection with L. major promastigotes only if the membranous fraction is used as vaccine. More, this vaccine formulation takes advantage of easy storage at -80°C with no need of fresh production. This clearly demonstrates that the immunity-inducing principle of disrupted DC-based vaccination lies within the membrane enclosed fraction. On a molecular level, disrupted antigen-loaded DC induce Th1-related cytokines during vaccination and as response on pathogen encounter. In vivo assays revealed IL-12 production and antigen-specific T cell proliferation among splenocytes that were stimulated with disrupted antigen-loaded DC. Splenocytes of accordingly vaccinated mice produce tremendous amounts of IFNγ after stimulation with Leishmania parasites. In summary, disrupted antigen-loaded BMDC fulfil all characteristics of DC-based vaccination against Leishmania major. But, while purification of membranes of antigen-loaded DC and subsequent transfer to BALB/c mice leads to control of the disease in the animal model, only slight levels of Th1-related cytokines are seen in the in vivo assays. Whether this points towards a loss of vaccine activity on unseen levels or unknown sites where Th1-related immunity is induced by both, complete solution and purified membranes, still has to be determined.
The probiotic Escherichia coli strain Nissle 1917 (EcN) is one of the few probiotics licensed as a medication in several countries. Best documented is its effectiveness in keeping patients suffering from ulcerative colitis (UC) in remission. This might be due to its ability to induce the production of human beta defensin 2 (HBD2) in a flagellin-dependent way in intestinal epithelial cells. In contrast to ulcerative colitis, for Crohn´s disease (CD) convincing evidence is lacking that EcN might be clinically effective, most likely due to the genetically based inability of sufficient defensin production in CD patients. As a first step in the development of an alternative approach for the treatment of CD patients, EcN strains were constructed which were able to produce human alpha-defensin 5 (HD5) or beta-defensin 2 (HBD2). For that purpose codon-optimized defensin genes encoding either the proform with the signal sequence or the mature form of human alpha defensin 5 (HD5) or the gene encoding HBD2 with or without the signal sequence were cloned in an expression vector plasmid under the control of the T7 promoter. Synthesis of the encoded defensins was shown by Western blots after induction of expression and lysis of the recombinant EcN strains. Recombinant mature HBD2 with an N-terminal His-tag could be purified by Ni-column chromatography and showed antimicrobial activity against E. coli, Salmonella enterica serovar Typhimurium and Listeria monocytogenes. In a second approach, that part of the HBD2-gene which encodes mature HBD2 was fused with yebF gene. The resulting fusion protein YebFMHBD2 was secreted from the encoding EcN mutant strain after induction of expression. Presence of YebFMHBD2 in the medium was not the result of leakage from the bacterial cells, as demonstrated in the spent culture supernatant by Western blots specific for ß-galactosidase and maltose-binding protein. The dialyzed and concentrated culture supernatant inhibited the growth of E. coli, Salmonella enterica serovar Typhimurium and Listeria monocytogenes in radial diffusion assays as well as in liquid coculture. This demonstrates EcN to be a suitable probiotic E. coli strain for the production of certain defensins.
The sexual phase of Plasmodium falciparum begins with the differentiation of intraerythrocytic sexual stages, termed gametocytes, in the human host. Mature gametocytes circulate in the peripheral blood and are taken up by the mosquito during the blood meal. These stages are essential for the spread of the malaria disease and form gametes in the mosquito midgut within minutes. A highly conserved family of six secreted proteins has been identified in Plasmodium falciparum. They comprise multiple adhesive domains and are termed PfCCp1 through PfCCp5, and PfFNPA. It was revealed in this work that PfCCp multi-domain adhesion proteins form protein complexes in gametocytes and on the surface of newly emerged macrogametes by adhesion domain-mediated binding. Co-Immunoprecipitation assays with activated gametocyte lysates show interactions between PfCCp proteins and indicate surface association via Pfs230 and Pfs25. Pfs230 is connected with the plasma membrane of the parasite by its interaction partner Pfs48/45. This protein is linked to the plasma membrane by a GPI anchor and presumably retains the multi-protein complex on the surface of newly emerged macrogametes in the mosquito midgut. A WD40 domain containing protein was identified to be part of this protein complex. It might serve as platform for the assembly of the multi protein complex or mediate the interplay among proteins, as suggested from known functions of the WD40 domain repeats. During egress from the host erythrocyte, the emerging gametes become vulnerable to factors of the human complement, which is taken up with the blood meal. In this thesis it was found that the complement system is active for about one hour post feeding. Macrogametes defend against complement-mediated lysis by co-opting the human complement regulators Factor H and FHL-1 from the blood-meal. These serum proteins bind via its SCR domains 5-7 to the surface of macrogametes. Once bound, they trigger complement inactivation of the alternative pathway, which prevents induction of complement lysis on the surface of the malaria parasite. Antibodies against Factor H are able to impair the sexual development in vitro and are able to block transmission to the mosquito. Interaction studies on endogenous proteins and immobilized recombinant proteins revealed the PfGAP50 protein as binding partner of Factor H and FHL-1. This protein was hitherto described as a glideosome-associated protein in invasive parasite stages, but has not yet been characterized in gametes. First localization studies indicate a relocation of PfGAP50 from the inner membrane complex to the surface of macrogametes. Malaria still persists as one of the deadliest infectious diseases worldwide. Investigations on the essential transmissive stages, gametocytes and gametes of Plasmodium falciparum, stood in the background of research for a long time. This work deciphered details on protein interactions on the surface of the malaria parasite and provides first information about coactions between the parasite and the human complement in the mosquito midgut.
Non-small cell lung cancer (NSCLC) is the deadliest form of lung cancer and has a poor prognosis due to its high rate of metastasis. Notably, metastasis is one of the leading causes of death among cancer patients. Despite the clinical importance, the cellular and molecular mechanisms that govern the initiation, establishment and progression of metastasis remain unclear. Moreover, knowledge gained on metastatic process was largely based on cultured or in vitro manipulated cells that were reintroduced into immune-compromised recipient mice. In the present study, a spontaneous metastasis mouse model for NSCLC was generated with a heritable fluorescent tag (DsRed) driven by CAG (combination of cytomegalovirus early enhancing element and chicken beta actin) promoter in alveolar type II cells (SpC-rtTA/TetO-Cre/LSL-DsRed). This approach is essential, keeping in mind the reprogramming nature of Myc oncogene (Rapp et al, 2009). Such genetic lineage tracing approach not only allowed us to monitor molecular and cellular changes during development of primary tumor but also led us to identify the different stages of secondary tumor development in distant organs. Upon combined expression of oncogenic C Raf-BXB and c-Myc (MYC-BXB-DsRed) in lung alveolar type II epithelial cells, macroscopic lung tumors arose comprising of both cuboidal and columnal cellular features. C Raf-BXB induced tumors (CRAF-DsRed) exhibit cuboidal morphology and is non-metastatic whereas Myc-BXB induced lung tumors (Myc-BXB-DsRed) present cuboidal-columnar cellular features and is able to undergo metastasis mainly in liver. Surprisingly, cystic lesions which were negative for SpC (Surfactant protein C) and CCSP (Clara cell secretory protein), strongly expressed DsRed proteins indicating its origin from lung alveolar type II cells. Moreover, early lung progenitor markers such as GATA4 (GATA-binding protein 4) and TTF1 (Thyroid Transcription Factor 1) were still expressed in these early cystic lesions suggesting metastasis as a faulty recapitulation of ontogeny (Rapp et al, 2008). Interestingly, mixed cystic lesions and metastatic tumors contained DsRed and SpC positive cells. These results demonstrate secondary tumor progression from cystic, mixed cystic to malignant transformation. Our results shed tremendous light on reprogramming of metastasizing cells during secondary tumor development. Moreover, such fluorescent tagged metastatic mice model can also be used to track the migration ability of metastatic cancer cell to different organs and its potential to differentiate into other cell types such as blood vessel or stromal cell within the primary tumor.
Escherichia coli Nissle 1917 (EcN) gehört zu den am besten untersuchten und charakterisierten probiotischen Bakterienstämmen. Seit Beginn des letzten Jahrhunderts wird er als Medikament eingesetzt, um verschiedene Darmerkrankungen wie z.B. Diarrhöe, entzündliche Darmerkrankungen und Verstopfung zu behandeln. Die Flagelle des EcN vermittelt Beweglichkeit und kann die Produktion von humanem β-Defensin 2 (hBD2) durch Epithelzellen induzieren. Somit ist dieses Organell direkt in die probiotische Funktion des EcN involviert. Es konnte gezeigt werden, dass die Flagellen anderer Bakterien, wie z.B. dem probiotischen Stamm Bacillus cereus CH oder den pathogenen Stämmen Pseudomonas aeruginosa und Clostridium difficile, die Adhäsion an intestinalen Mucus, welcher von Epithelzellen sekretiert wird, vermitteln. Allerdings blieb unklar, welcher Teil der Flagelle an welche Mucuskomponente bindet. Die Fähigkeit effizient an Wirtgewebe zu adhärieren wird als wichtiges Attribut eines probiotischen Stammes angesehen. Ex vivo Adhäsionsstudien mit Kryoschnitten humaner Darmbiopsien haben gezeigt, dass die Flagelle des EcN in die effiziente Adhäsion an humanes Darmgewebe involviert sein muss. Aus diesem Grund wurde in dieser Arbeit die Funktion der Flagelle des EcN als Adhäsin untersucht. Zunächst wurde die hyperflagellierte Variante EcN ATHF isoliert und durch verschiedene Experimente, z.B. Schwärmagartests und Elektronenmikroskopie, charakterisiert. Weitere ex vivo Adhäsionsstudien mit EcN ATHF zeigten eine höhere Adhäsionseffizienz dieser hyperflagellierten Variante und bestätigten damit die Rolle der Flagelle bei der effizienten Adhäsion von EcN an die Kryoschnitte der humanen Darmbiopsien. Interessanterweise fungierte die Flagelle in in vitro Studien mit den humanen Epithelzellen Caco-2 und T24 nicht als Adhäsin. Diese Unterschiede zwischen den in vitro und ex vivo Studien führten zu der Annahme, dass die Flagelle des EcN in vivo die Adhäsion an Mucus vermittelt, welcher von den Caco-2- und T24-Zellen nicht produziert wird, aber in den Kryoschnitten der Darmbiopsien nachgewiesen wurde. Diese Vermutung wurde durch in vitro Adhäsionsstudien mit der Mucin-produzierenden Epithelzelllinie LS174-T bestätigt, da die Flagellen für eine effektive Adhäsion an diese Zellen essentiell waren. Zudem reduzierte die Präinkubation flagellierter EcN-Stämme mit Mucin2 ihre Adhäsionseffizienz an Kryoschnitte humaner Darmbiopsien. Um die direkte Interaktion zwischen Flagellen des EcN Wildtyps und Mucus zu zeigen, wurde ein ELISA etabliert. Es konnte eine direkte konzentrationsabhängige Interaktion zwischen isolierten Flagellen des EcN Wildtyps und Mucin2, bzw. humanem Mucus (Kolon) beobachtet werden. Interessanterweise konnte keine Interaktion zwischen isolierten Flagellen des EcN Wildtyps und murinem Mucus (Duodenum, Ileum, Caecum, Colon) festgestellt werden. Dies weist darauf hin, dass die Mucuszusammensetzung zwischen verschiedenen Spezies variiert. Verschiedene Kohlenhydrate, welche bekannte Mucusbestandteile sind, wurden auf ihre Interaktion mit der Flagelle von EcN getestet und Gluconat wurde als ein Rezeptor identifiziert. Die Präinkubation isolierter Flagellen mit Gluconat reduzierte ihre Interaktion mit Mucin2, bzw. humanem Mucus signifikant. Zudem wurde die oberflächenexponierte Domäne D3 des Flagellins, der Hauptuntereinheit der Flagelle, als möglicher Interaktionspartner von Mucin2, bzw. humanem Mucus ausgeschlossen. Flagellen, die aus einer Domäne D3 Deletionsmutante isoliert wurden, zeigten sogar eine effizientere Bindung an Mucin2, bzw. humanen Mucus. Weiterhin konnte gezeigt werden, dass Änderungen des pH-Wertes signifikante Effekte auf die Interaktion zwischen Mucus und isolierten Flagellen hatten, vermutlich aufgrund von Konformationsänderungen. Zusammenfassend wurde in dieser Arbeit die Flagelle als neues und scheinbar wichtigstes Adhäsin in vivo für den probiotischen Stamm EcN identifiziert. Hierfür wurden sowohl eine hyperflagellierte Variante, eine ΔfliC Mutante, sowie der dazugehörige komplementierte Stamm verwendet. EcN ist zudem der erste probiotische Stamm für den eine direkte Bindung der Flagellen an humanen Mucus nachgewiesen werden konnte. Die Mucuskomponente Gluconat konnte dabei als wichtiger Rezeptor identifiziert werden. Da einige pathogene Bakterien ihre Flagelle zur Adhäsion an Wirtsgewebe nutzen, könnte dieses Organell EcN dazu befähigen, mit Pathogenen um die erfolgreiche Kolonisierung des Darms zu konkurrieren, was als wichtige Eigenschaft eines Probiotikums betrachtet wird.
ATP dependent chromatin remodeling complexes are multifactorial complexes that utilize the energy of ATP to rearrange the chromatin structure. The changes in chromatin structure lead to either increased or decreased DNA accessibility. SWI/SNF is one of such complex. The SWI/SNF complex is involved in both transcription activation and transcription repression. The ATPase subunit of SWI/SNF is called SWI2/SNF2 in yeast and Brahma, Brm, in Drosophila melanogaster. In mammals there are two paralogs of the ATPase subunit, Brm and Brg1. Recent studies have shown that the human Brm is involved in the regulation of alternative splicing. The aim of this study was to investigate the role of Brm in pre-mRNA processing. The model systems used were Chironomus tentans, well suited for in situ studies and D. melanogaster, known for its full genome information. Immunofluorescent staining of the polytene chromosome indicated that Brm protein of C. tentans, ctBrm, is associated with several gene loci including the Balbiani ring (BR) puffs. Mapping the distribution of ctBrm along the BR genes by both immuno-electron microscopy and chromatin immunoprecipitation showed that ctBrm is widely distributed along the BR genes. The results also show that a fraction of ctBrm is associated with the nascent BR pre-mRNP. Biochemical fractionation experiments confirmed the association of Brm with the RNP fractions, not only in C. tentans but also in D. melanogaster and in HeLa cells. Microarray hybridization experiments performed on S2 cells depleted of either dBrm or other SWI/SNF subunits show that Brm affects alternative splicing and 3´ end formation. These results indicated that BRM affects pre-mRNA processing as a component of SWI/SNF complexes. 1
Based on genetic association and functional imaging studies, reduced function of tryptophan hydroxylase-2 (TPH2) has been shown to be critically involved in the pathophysiology of anxiety-disorders and depression. In order to elucidate the impact of a complete neuronal 5-HT deficiency, mice with a targeted inactivation of the gene encoding Tph2 were generated. Interestingly, survival of Tph2-/- mice, the formation of serotonergic neurons and the pathfinding of their projections was not impaired. Within this thesis, I investigated the influence of 5-HT deficiency on the γ-amino butyric acid (GABA) system. The GABAergic system is implicated in the pathophysiology of anxiety disorders. Therefore, measurement of GABA concentrations in different limbic brain regions was carried out. These measurements were combined with immunohistochemical estimation of GABAergic cell subpopulations in the dorsal hippocampus and amygdala. In Tph2-/- mice GABA concentrations were increased exclusively in the dorsal hippocampus. In heterozygous Tph2+/- mice concentrations of GABA were increased in the amygdala compared to Tph2-/- and wt control mice, while the reverse was found in the prefrontal cortex. The changes in GABA concentrations were accompanied by altered cell density of GABAergic neurons within the basolateral complex of the amygdala and parvalbumin (PV) neurons of the dorsal hippocampus and by adaptational changes of 5-HT receptors. Thus, adaptive changes during the development on the GABA system may reflect altered anxiety-like and depressive-like behavior in adulthood. Moreover, chronic mild stress (CMS) rescues the depressive-like effects induced by 5-HT deficiency. In contrast, 5-HT is important in mediating an increased innate anxiety-like behavior under CMS conditions. This is in line with a proposed dual role of 5-HT acting through different mechanisms on anxiety and depressive-like behavior, which is influenced by gene-environment interaction effects. Further research is needed to disentangle these complex networks in the future.
Die MRT des Herzens wird aufgrund hoher Reproduzierbarkeit und geringer Variabilität als Referenzstandard für die Bestimmung der kardialen Funktion betrachtet. Auch in der präklinischen Forschung bietet die MRT eine ausgezeichnete Charakterisierung der kardialen Funktion und ermöglicht eine exzellente Analyse modellierter Krankheitsbilder. In beiden Fällen besteht jedoch weiterhin Optimierungsbedarf. Die klinische Herz-MRT stellt ein aufwendiges Verfahren mit relativ langer Messzeit dar und ist dadurch mit hohen Untersuchungskosten verbunden. In der präklinischen Kleintierbildgebung müssen zum Erreichen der notwendigen höheren Orts- und Zeitauflösung ebenfalls lange Aufnahmezeiten in Kauf genommen werden. Um die kardiale MRT dort routinemäßig in großen Studienkollektiven anwenden zu können, ist eine schnellere Bildgebung essentiell. Neben einer Verbesserung der Tomographen-Hardware und der Optimierung von Bildgebungssequenzen standen im letzten Jahrzehnt vermehrt informationstheoretische Ansätze zur Beschleunigung der MR-Datenakquisition im Fokus der Entwicklung. Während zu Beginn des Jahrtausends die Parallele Bildgebung (PI) einen Forschungsschwerpunkt repräsentierte, spielte sich in den letzten fünf Jahren vermehrt die von Donoho und Candès eingeführte Compressed Sensing (CS) Theorie in den Vordergrund. Diese ermöglicht eine Signalrekonstruktion aus unvollständig gemessenen Koeffizienten einer linearen Messung (z.B. Fouriermessung) unter Ausnutzung der Sparsität des Signals in einer beliebigen Transformationsbasis. Da sich die MRT hervorragend für den Einsatz von CS eignet, wurde die Technik in der Forschung bereits vielfach angewendet. Die zur Rekonstruktion unterabgetasteter Aufnahmen nötigen CS-Algorithmen haben jedoch eine signifikante Veränderung des Bildgebungsprozesses der MRT zur Folge. Konnte dieser zuvor in guter Näherung als linear und stationär betrachtet werden, so repräsentiert die CS-Rekonstruktion eine nichtlineare und nichtstationäre Transformation. Objektinformation wird nicht mehr ortsunabhängig und proportional zur Intensität in die Abbildung transportiert. Das Bild ist viel mehr das Ergebnis eines Optimierungsprozesses, der sowohl die Konsistenz gegenüber der unterabgetasteten Messung als auch die Sparsität des Signals maximiert. Der erste Teil dieser Dissertation beschreibt eine Methode, die eine objektive Einschätzung der Bildqualität CS-rekonstruierter MR-Bilder ermöglicht. Die CS-Beschleunigung verspricht eine Verkürzung der Messzeit ohne Verlust an Bildqualität, wobei letztere bisher größtenteils qualitativ bzw. quantitativ nur unzureichend beurteilt wurde. Konnte der Bildgebungsprozess der klassischen MRT (linear und stationär) durch die Bestimmung einer Punktspreizfunktion (PSF) robust und effektiv validiert und optimiert werden, erlauben die CS-Algorithmen aufgrund ihres nichtlinearen und nichtstationären Verhaltens ohne Weiteres keine äquivalente Analyse. Um dennoch eine entsprechende Evaluierung des CS-Bildgebungsprozesses zu ermöglichen, wurde die Anwendung einer lokalen Punktspreizfunktion (LPSF) für den in der Folge verwendeten Iterative Soft Thresholding Algorithmus untersucht. Die LPSF berücksichtigt die Ortsabhängigkeit der CS-Rekonstruktion und muss daher für jeden Ort (Pixel) eines Bildes bestimmt werden. Darüber hinaus wurde die LPSF im linearen Bereich der CS-Transformation ermittelt. Dazu wurde das zu bewertende Bild nach Anwenden einer kleinen lokalen Störung rekonstruiert. Die Breite des Hauptmaximums der LPSF wurde schließlich verwendet, um ortsaufgelöste Auflösungsstudien durchzuführen. Es wurde sowohl der Einfluss typischer Unterabtastschemata für CS als auch der Einsatz diskreter Gradienten zur Sparsifizierung eines Phantombildes untersucht. Anschließend wurde die Prozedur zur Bestimmung der räumlichen und zeitlichen Auflösung in der Herzbildgebung getestet. In allen Beispielen ermöglichte das vorgeschlagene Verfahren eine solide und objektive Analyse der Bildauflösung CS-rekonstruierter Aufnahmen. Wurde zuvor meist ausschließlich auf Vergleiche mit einer vollständig abgetasteten Referenz zur Qualitätsbeurteilung zurückgegriffen, so stellt die vorgestellte Auflösungsbestimmung einen Schritt in Richtung einer standardisierten Bildanalyse bei der Verwendung der Beschleunigung mittels CS dar. Die Analyse der Abtastmuster zeigte, dass auch bei der Anwendung von CS die Berücksichtigung der nominell höchsten Frequenzen k_max unerlässlich ist. Frühere Publikationen schlagen Abtastfolgen mit einer teils starken Gewichtung der Messpunkte zum k-Raum-Zentrum hin vor. Die Ergebnisse der vorliegenden Arbeit relativieren ein derartiges Vorgehen, da zumindest bei den durchgeführten Untersuchungen ein Auflösungsverlust bei analoger Vorgehensweise zu verzeichnen war. Ebenso zeigten sich dynamische Aufnahmen, die unter Verwendung des x-f-Raums als sparse Basis rekonstruiert wurden, durchaus anfällig für zeitliches Blurring. Dieses resultiert aus der Unterdrückung hoher zeitlicher Frequenzen und konnte durch die ortsaufgelösten Auflösungskarten sichtbar gemacht werden. Neben der Auflösung ist für eine umfassende Analyse der Bildqualität auch die Untersuchung potentieller Aliasing-Artefakte sowie des Signal-zu-Rausch-Verhältnisses (SNR) notwendig. Während Aliasing mit Hilfe der Einträge der LPSF außerhalb des Hauptmaximums untersucht werden kann, wurde in Kap. 5 eine Modifikation der Multi-Replika-Methode von Robson et al. zur Rauschanalyse bei Verwendung nichtlinearer Algorithmen vorgestellt. Unter Einbeziehung aller genannten Qualitätsparameter ist eine robuste Bewertung der Bildqualität auch bei einer Verwendung von CS möglich. Die differenzierte Evaluierung ebnet den Weg hin zu einem objektiven Vergleich neuer Entwicklungen mit bisherigen Standard-Techniken und kann dadurch den Einzug von CS in die klinische Anwendung vorantreiben. Nach den theoretischen Betrachtungen der Bildqualität behandelt die Dissertation die erstmalige Anwendung von CS zur Beschleunigung der funktionellen Herzdiagnostik in der präklinischen MR-Kleintierbildgebung. Diese Studien wurden in Zusammenarbeit mit der British Heart Foundation Experimental Magnetic Resonance Unit (BMRU) der University of Oxford durchgeführt. Die Algorithmen für eine Beschleunigung mittels der CS-Theorie wurden anhand der dort am 9,4T Tomographen gemessenen (unterabgetasteten) Datensätze entwickelt und optimiert. Zunächst wurde eine Beschleunigung ausschließlich mittels CS untersucht. Dazu wurde die segmentierte, EKG- und Atemgetriggerte kartesische Cine-Aufnahme in Phasenkodierrichtung unterabgetastet und mittels CS rekonstruiert. Die sparse Darstellung wurde durch Ermitteln zeitlicher Differenzbilder für jede Herzphase erhalten. Durch Variation der Abtastmuster in der zeitlichen Dimension konnte ein vollständig abgetastetes zeitliches Mittelbild bestimmt werden, das anschließend von jedem einzelnen Herzphasenbild subtrahiert wurde. In einer Validierungsphase wurden an der Maus vollständig aufgenommene Cine-Akquisitionen retrospektiv unterabgetastet, um die maximal mögliche Beschleunigung mittels CS zu ermitteln. Es wurden u.a. funktionelle Herz-Parameter für jede Gruppe des jeweiligen Beschleunigungsfaktors bestimmt und mittels einer statistischen Analyse verglichen. Die Gesamtheit aller Ergebnisse zeigte die Möglichkeit einer dreifachen Beschleunigung ohne eine Degradierung der Genauigkeit der Methode auf. Die ermittelte Maximalbeschleunigung wurde in einer unterabgetastet gemessenen Bilderserie mit anschließender CS-Rekonstruktion validiert. Die Abtastschemata wurden dazu mit Hilfe der Transformations-Punktspreizfunktion weiter optimiert. In einer Erweiterung der Studie wurde zum Zweck einer noch höheren Beschleunigung die CS-Technik mit der PI kombiniert. Erneut fand eine Unterabtastung der Phasenkodierrichtung einer kartesischen Trajektorie statt. Die Messungen erfolgten mit einer 8-Kanal-Mäusespule an einem 9,4T Tomographen. Um das Potential beider Beschleunigungstechniken auszunutzen, wurden die Methoden CS und PI in serieller Weise implementiert. Für die PI-Beschleunigung wurde der vollständig abgetastete k-Raum zunächst gleichmäßig unterabgetastet. Auf dem resultierenden Untergitter wurde zusätzlich eine Unterabtastung nach Pseudo-Zufallszahlen durchgeführt, um eine Beschleunigung mittels CS zu ermöglichen. Die entwickelte Rekonstruktion erfolgte ebenfalls seriell. Zunächst wurde mittels CS das äquidistante Untergitter rekonstruiert, um anschließend mittels GRAPPA die noch fehlenden Daten zu berechnen. Um eine zusätzliche Messung zur Kalibrierung der GRAPPA-Faktoren zu umgehen, wurde das äquidistant unterabgetastete Untergitter von Herzphase zu Herzphase um je einen Phasenkodierschritt weitergeschoben. Dieses Vorgehen erlaubt die Ermittlung eines vollständig abgetasteten k-Raums mit einer geringeren zeitlichen Auflösung, der die notwendige Bestimmung der Wichtungsfaktoren ermöglicht. Folgende Kombinationen von Beschleunigungsfaktoren wurden mittels retrospektiver Unterabtastung eines vollständig aufgenommenen Datensatzes untersucht: R_CS x R_PI = 2 x 2, 2 x 3, 3 x 2 und 3 x 3. Die Analyse des Bildrauschens, des systematischen Fehlers und der Auflösung führte zu dem Schluss, dass eine sechsfache Beschleunigung mit Hilfe der hybriden Rekonstruktionstechnik möglich ist. Während mit steigender CS-Beschleunigung der systematische Fehler leicht anstieg, führte ein höherer PI-Beschleunigungsfaktor zu einer leichten Verstärkung des statistischen Fehlers. Der statistische Fehler zeigte jedoch ebenfalls eine Verringerung bei steigender Beschleunigung mittels CS. Die Fehler waren allerdings stets auf einem Niveau, das durchaus auch Beschleunigungen bis R_CS x R_PI =3 x 3 zulässt. Die LPSF-Analyse zeigte einen Verlust der räumlichen Auflösung von ca. 50 % bei R=6 sowie einen mittleren Verlust von 64 % bei R=9. Offensichtlich ging die ebenfalls beobachtete Minimierung des Bildrauschens durch den CS-Algorithmus im Falle der relativ stark verrauschten Kleintieraufnahmen zu Lasten der Bildauflösung. Die mit zunehmender Beschleunigung stärker geblurrten Grenzen zwischen Blutpool und Myokardgewebe erschweren die Segmentierung und stellen eine mögliche Fehlerquelle dar. Unter Beachtung aller Ergebnisse ist eine sechsfache Beschleunigung (R_CS x R_PI = 2 x 3, 3 x 2) vertretbar. Die Hinzunahme der PI ermöglicht somit im Vergleich zur alleinigen Verwendung von CS eine weitere Beschleunigung um einen Faktor von zwei. Zusammenfassend ermöglicht der Einsatz von CS in der präklinischen funktionellen Herzbildgebung am Kleintier eine deutliche Reduktion der Messzeit. Bereits ohne Vorhandensein von Mehrkanalspulen kann die notwendige Datenmenge ohne signifikante Beeinflussung der Messergebnisse auf ein Drittel reduziert werden. Ist der Einsatz von Spulenarrays möglich, kann die mit PI mögliche dreifache Beschleunigung um einen weiteren Faktor zwei mittels CS auf R=6 erweitert werden. Dementsprechend kann CS einen wesentlichen Beitrag dazu leisten, dass das Potential Herz-MRT am Kleintier in großen Studienkollektiven effektiver abgerufen werden kann. Im letzten Teil der Arbeit wurde eine Technik für die funktionelle klinische MR-Herzbildgebung entwickelt. Hier wurde eine Beschleunigung mittels CS verwendet, um die Aufnahme des gesamten Herzens innerhalb eines Atemstillstandes des Patienten zu ermöglichen. Bei der derzeitigen Standardmethode werden üblicherweise 10-15 2D-Schichten des Herzens akquiriert, wobei jede einzelne Aufnahme einen Atemstillstand des Patienten erfordert. Für die notwendige Beschleunigung wurde eine unterabgetastete 3D-Trajektorie verwendet. Durch Phasenkodierung einer Richtung sowie radiale Projektionen in den beiden anderen Dimensionen konnte eine effiziente Aufnahme unterhalb des Nyquist-Kriteriums erreicht werden. Die Sparsifizierung erfolgte, wie bereits in der beschriebenen präklinischen Anwendung, durch die Subtraktion eines zeitlichen Mittelbildes. In einer Simulation anhand eines retrospektiv unterabgetasteten Datensatzes konnte die theoretische Funktionalität der Rekonstruktionstechnik bei einer Beschleunigung bezüglich der Nyquist-Abtastung von R ~ 10 validiert werden. Die Unterschiede zum vollständig abgetasteten Datensatz waren vernachlässigbar klein, so dass die vorgeschlagene Abtastfolge am Tomographen implementiert wurde. Mit dieser Sequenz wurde anschließend eine funktionelle Bilderserie an einem gesunden Probanden mit vollständiger Herzabdeckung innerhalb eines Atemstopps aufgenommen. Fehlende Daten wurden analog zur Simulation mit Hilfe des vorgeschlagenen Algorithmus rekonstruiert. Im Vergleich zur Simulation ergaben sich aufgrund des Schichtprofils der 3D-Slab-Anregung zusätzliche Aliasing-Artefakte in den äußeren Partitionen. Die für radiale Aufnahmen typischen Streifenartefakte waren im rekonstruierten Bild, wenn auch mit sehr geringer Amplitude, noch erkennbar. Davon abgesehen wurde die Dynamik jedoch über das gesamte Herz hinweg gut dargestellt. Der hohe Kontrast zwischen Myokard und Blutpool bescheinigt den Bildern eine hervorragende Eignung für die Bestimmung funktioneller Herzparameter mittels einer Segmentierung. Zusammengefasst erlaubt die entwickelte Methode aufgrund der drastischen Reduktion der notwendigen Atemstopps des Patienten einen deutlich erhöhten Patientenkomfort sowie einen schnelleren Durchsatz aufgrund der verkürzten Messzeit.
Streptococcus pneumoniae (Pneumococcus) is one of the leading causes of childhood meningitis,pneumonia and sepsis. Despite the availability of childhood vaccination programs and antimicrobial agents, childhood pneumococcal meningitis is still a devastating illness with mortality rates among the highest of any cause of bacterial meningitis. Especially in low-income countries, where medical care is less accessible, mortality rates up to 50 % have been reported. In surviving patients, neurological sequelae, including hearing loss, focal neurological deficits and cognitive impairment, is reported in 30 to 50 %. Growing resistance of pneumococci towards conventional antibiotics emphasize the need for effective therapies and development of effective vaccines against Streptococcus pneumoniae. One major virulence factor of Streptococcus pneumoniae is the protein toxin Pneumolysin (PLY). PLY belongs to a family of structurally related toxins, the so-called cholesterol-dependent cytolysins (CDCs). Pneumolysin is produced by almost all clinical isolates of the bacterium. It is expressed during the late log phase of bacterial growth and gets released mainly through spontaneous autolysis of the bacterial cell. After binding to cholesterol in the host cell membranes, oligomerization of up to 50 toxin monomers and rearrangement of the protein structure, PLY forms large pores, leading to cell lysis in higher toxin concentrations. At sub-lytic concentrations, however, PLY mediates several other effects, such as activation of the classic complement pathway and the induction of apoptosis. First experiments with pneumococcal strains, deficient in pneumolysin, showed a reduced virulence of the organism, which emphasizes the contribution of this toxin to the course of bacterial meningitis and the urgent need for the understanding of the multiple mechanisms leading to invasive pneumococcal disease. The aim of this thesis was to shed light on the contribution of pneumolysin to the course of the disease as well as to the mental illness patients are suffering from after recovery from pneumococcal meningitis. Therefore, we firstly investigated the effects of sub-lytic pneumolysin concentrations onto primary mouse neurons, transfected with a GFP construct and imaged with the help of laser scanning confocal microscopy. We discovered two major morphological changes in the dendrites of primary mouse neurons: The formation of focal swellings along the dendrites (so-called varicosities) and the reduction of dendritic spines. To study these effects in a more complex system, closer to the in vivo situation, we established a reproducible method for acute brain slice culturing. With the help of this culturing method, we were able to discover the same morphological changes in dendrites upon challenge with sub-lytic concentrations of pneumolysin. We were able to reverse the seen alterations in dendritic structure with the help of two antagonists of the NMDA receptor, connecting the toxin´s mode of action to a non-physiological stimulation of this subtype of glutamate receptors. The loss of dendritic spines (representing the postsynapse) in our brain slice model could be verified with the help of brain slices from adult mice, suffering from pneumococcal meningitis. By immunohistochemical staining with an antibody against synapsin I, serving as a presynaptic marker, we were able to identify a reduction of synapsin I in the cortex of mice, infected with a pneumococcal strain which is capable of producing pneumolysin. The reduction of synapsin I was higher in these brain slices compared to mice infected with a pneumococcal strain which is not capable of producing pneumolysin, illustrating a clear role for the toxin in the reduction of dendritic spines. The fact that the seen effects weren´t abolished under calcium free conditions clarifies that not only the influx of calcium through the pneumolysin-pore is responsible for the alterations. These findings were further supported by calcium imaging experiments, where an inhibitor of the NMDA receptor was capable of delaying the time point, when the maximum of calcium influx upon PLY challenge was reached. Additionally, we were able to observe the dendritic beadings with the help of immunohistochemistry with an antibody against MAP2, a neuron-specific cytoskeletal protein. These observations also connect pneumolysin´s mode of action to excitotoxicity, as several studies mention the aggregation of MAP2 in dendritic beadings in response to excitotoxic stimuli. All in all, this is the first study connecting pneumolysin to excitotoxic events, which might be a novel chance to tie in other options of treatment for patients suffering from pneumococcal meningitis.
Plant-derived natural products and their analogs continue to play an important role in the discovery of new drugs for the treatment of human diseases. Potentially promising representatives of secondary metabolites are the naphthylisoquinoline alkaloids, which show a broad range of activities against protozoan pathogens, such as plasmodia, leishmania, and trypanosoma. Due to the increasing resistance of those pathogens against current therapies, highly potent novel agents are still urgently needed. Thus, it is worthy to discover new naphthylisoquinoline alkaloids hopefully with pronounced bioactivities by isolation from plants or by synthesis. The naphthylisoquinoline alkaloids are biosynthetically related to another class of plant-derived products, the naphthoquinones, some of which have been recently found to display excellent anti-multiple myeloma activities without showing any cytotoxicities on normal blood cells. Multiple myeloma still remains incurable, although remissions may be induced with co-opted therapeutic treatments. Therefore, more potent naphthoquinones are urgently required, and can be obtained by isolation from plants or by synthesis. In detail, the results in this thesis are listed as follows: 1) Isolation and characterization of naphthylisoquinoline alkaloids from the stems of a Chinese Ancistrocladus tectorius species. Nine new naphthylisoquinoline alkaloids, named ancistectorine A1 (60), N-methylancistectorine A1 (61), ancistectorine A2 (62a), 5-epi-ancistectorine A2 (62b), 4'-O-demethylancistectorine A2 (63), ancistectorine A3 (64), ancistectorine B1 (65), ancistectorine C1 (66), and 5-epi-ancistrolikokine D (67) were isolated from the Chinese A. tectorius and fully characterized by chemical, spectroscopic, and chiroptical methods. Furthermore, the in vitro anti-infectious activities of 60-62 and 63-66 have been tested. Three of the metabolites, 61, 62a, and 62b, exhibited strong antiplasmodial activities against the strain K1 of P. falciparum without showing significant cytotoxicities. With IC50 values of 0.08, 0.07, and 0.03 μM, respectively, they were 37 times more active than the standard chloroquine (IC50 = 0.26 μM). Moreover, these three compounds displayed high antiplasmodial selectivity indexes ranging from 100 to 3300. According to the TDR/WHO guidelines, they could be considered as lead compounds. In addition, seven alkaloids, 69-74 (structures not shown here), were isolated from A. tectorius that were known, but new to the plant, together with another fourteen known compounds (of these, only the structures of the three main alkaloids, 5a, 5b, and 78 are shown here), which had been previously found in the plant. The three metabolites ancistrocladine (5a), hamatine (5b), and (+)-ancistrocline (78) were found to show no or moderate activities against the MM cell lines. 2) Isolation and characterization of naphthylisoquinoline alkaloids from the root bark of a new, botanically yet undescribed Congolese Ancistrocladus species. An unprecedented dimeric Dioncophyllaceae-type naphthylisoquinoline alkaloid, jozimine A2 (84), as first recognized by G. Bauckmann from an as yet undescribed Ancistrocladus species, was purified and characterized as part of this thesis. Its full structural assignment was achieved by spectroscopic and chiroptical methods, and further confirmed by an X-ray diffraction analysis, which had never succeeded for any other dimeric naphthylisoquinoline alkaloids before. Structurally, the dimer is composed of two identical 4'-O-demethyldioncophylline A halves, coupled through a sterically hindered central axis at C-3',3'' of the two naphthalene moieties. Pharmacologically, jozimine A2 (84) showed an extraordinary antiplasmodial activity (IC50 = 1.4 nM) against the strain NF54 of P. falciparum. Beside jozimine A2 (85), another new alkaloid, 6-O-demethylancistrobrevine C (84), and four known ones, ancistrocladine (5a), hamatine (5b), ancistrobrevine C (86), and dioncophylline A (6) were isolated from the Ancistrocladus species, the latter in a large quantity (~500 mg), showing that the plant produces Ancistrocladaceae-type, mixed-Ancistrocladaceae/Dioncophyllaceae-type, and Dioncophyllaceae-type naphthyl- isoquinoline alkaloids. Remarkably, it is one of the very few plants, like A. abbreviatus, and A. barteri, that simultaneously contain typical representatives of all the above three classes of alkaloids. 3) Semi-synthesis of jozimine A2 (85), 3'-epi-85, jozimine A3 (93) and other alkaloids from dioncophylline A (6). The dimeric naphthylisoquinoline alkaloids, jozimine A2 (85) and 3'-epi-85, constitute rewarding synthetic targets for a comparative analysis of their antiplasmodial activities and for a further confirmation of the assigned absolute configurations of the isolated natural product of 85. They were semi-synthesized in a four-step reaction sequence from dioncophylline A (6) in cooperation with T. Büttner. The key step was a biomimetic phenol-oxidative dimerization at C-3' of the N,O-dibenzylated derivative of 89 by utilizing Pb(OAc)4. This is the first time that the synthesis of such an extremely sterically hindered (four ortho-substituents) naphthylisoquinoline alkaloid – with three consecutive biaryl axes! – has been successfully achieved. A novel dimeric naphthylisoquinoline, jozimine A3 (93), bearing a 6',6''-central biaryl axis, was semi-synthesized from 5'-O-demethyldioncophylline A (90) by a similar biomimetic phenol-oxidative coupling reaction as a key step, by employing Ag2O. HPLC analysis with synthetic reference material of 3'-epi-85 and 93 for co-elution revealed that these two alkaloids clearly are not present in the crude extract of the Ancistrocladus species from which jozimine A2 (85) was isolated. This evidences that jozimine A2 (85) is very specifically biosynthesized by the plant with a high regio- and stereoslectivity. Remarkably, the two synthetic novel dimeric naphthylisoquinoline alkaloids 3'-epi-85 and 93 were found to display very good antiplasmodial activities, albeit weaker than that of the natural and semi-synthetic product 85. Additionally, the two compounds 3'-epi-85 and 93 possessed high or moderate selectivity indexes, which were much lower than that of 85. However, they can still be considered as new lead structures. Two unprecedented oxidative products of dioncophylline A, the diastereomeric dioncotetralones A (94a) and B (94b), were synthesized from dioncophylline A (6) in a one-step reaction. Remarkably, the aromatic properties in the “naphthalene” and the “isoquinoline” rings of 94a and 94b are partially lost and the “biaryl” axis has become a C,C-double bond, so that the two halves are nearly co-planar to each other, which has never been found among any natural or synthetic naphthylisoquinoline alkaloid. Their full structural characterization was accomplished by spectroscopic methods and quantum-chemical CD calculations (done by Y. Hemberger). The presumed reaction mechanism was proposed in this thesis. In addition, one of the two compounds, 94a, exhibited a highly antiplasmodial activity (IC50 = 0.09 μM) with low cytotoxicity, and thus, can be considered as a new promising lead structure. Its 2'-epi-isomer, 94b, was inactive, evidencing a significant effect of chirality on the bioactivity. Of a number of naphthylisoquinoline alkaloids tested against the multiple-myeloma cell lines, the three compounds, dioncophylline A (6), 4'-O-demethyldioncophylline A (89), and 5'-O-demethyldioncophylline A (90) showed excellent activities, even much stronger than dioncoquinones B (10), C (102), the epoxide 175, or the standard drug melphalan. 4) Isolation and characterization of bioactive naphthoquinones from cell cultures of Triphyophyllum peltatum. Three new naphthoquinones, dioncoquinones C (102), D (103), and E (104), the known 8-hydroxydroserone (105), which is new to this plant, and one new naphthol dimer, triphoquinol A (107), were isolated from cell cultures of T. peltatum in cooperation with A. Irmer. Dioncoquinone C (102) showed an excellent activity against the MM cells, very similar to that of the previously found dioncoquinone B (10), without showing any inhibitory effect on normal cells. The other three naphthoquinones, 103105, were inactive or only weakly active. 5) Establishment of a new strategy for a synthetic access to dioncoquinones B (10) and C (102) on a large scale for in vivo experiments and for the synthesis of their analogs for first SAR studies. Before the synthesis of dioncoquinone B (10) described in this thesis, two synthetic pathways had previously been established in our group. The third approach described here involved the preparation of the joint synthetic intermediate 42 with the previous two routes. The tertiary benzamide 135 was ortho-deprotonated by using s-BuLi/TMEDA, followed by transmetallation with MgBr2▪2Et2O, and reaction with 2-methylallyl bromide (139). It resulted in the formation of ortho-allyl benzamide 140, which was cyclized by using methyl lithium to afford the naphthol 42. This strategy proved to be the best among the established three approaches with regard to its very low number of steps and high yields. By starting with 136, this third strategy yielded the related bioactive natural product, dioncoquinone C (102), which was accessed by total synthesis for the first time. To identify the pharmacophore of the antitumoral naphthoquinones, a library of dioncoquinone B (10) and C (102) analogs were synthesized for in vitro testing. Among the numerous naphthoquinones tested, the synthetic 7-O-demethyldioncoquinone C (or 7-O-hydroxyldioncoquinone B) (145), constitutes another promising basic structure to develop a new anti-MM agent. Furthermore, preliminary SAR results evidence that the three hydroxy functions at C-3, C-5, and C-6 are essential for the biological properties as exemplarily shown through the compounds 10, 102, and 145. All other mixed OH/OMe- or completely OMe-substituted structures were entirely inactive. By a serendipity the expoxide 175 was found to display the best anti-MM activity of all the tested isolated metabolites from T. peltatum, the synthesized naphthoquinones, and their synthetic intermediates. Toxic effects of 175 on normal cells were not observed, in contrast to the high toxicities of all other epoxides. Thus, the anti-MM activity of 175 is of high selectivity. The preliminary SAR studies revealed that the 6-OMe group in 175 is required, thus differed with the above described naphthoquinones (where 6-OH is a requisite in 10, 102, and 145), which evidenced potentially different modes of action for these two classes of compounds. 6) The first attempted total synthesis of the new naturally occurring triphoquinone (187a), which was recently isolated from the root cultures of T. peltatum in our group. A novel naphthoquinone-naphthalene dimer, 187a (structure shown in Chapter 10), was isolated in small quantities from the root cultures of T. peltatum. Thus, its total synthesis was attempted for obtaining sufficient amounts for selected biotestings. The key step was planned to prepare the extremely sterically hindered (four ortho-substituents) binaphthalene 188 by a coupling reaction between the two 2-methylnaphthalene derivatives. Test reactions involving a system of two simplified 2-methylnaphthylboron species and 2-methylnaphthyl bromide proved the Buchwald ligand as most promising. The optimized conditions were then applied to the two true - highly oxygenated - coupling substrates, between the 2-methylnaphthylboron derivatives 210, 211, 213, or 214 and the 2-methylnaphthyl iodides (or bromides) 215 (206), 215 (206), 212 (205), or 212 (205), respectively. Unfortunately, this crucial step failed although various bases and solvent systems were tested. This could be due to the high electron density of the two coupling substrates, both bearing strongly OMOM/OMe-donating function groups. Therefore, a more powerful catalyst system or an alternative synthetic strategy must be explored for the total synthesis of 187a. 7) Phytochemical investigation of the Streptomyces strain RV-15 derived from a marine sponge. Cyclodysidins A-D (216-219), four new cyclic lipopeptides with a- and ß-amino acids, were isolated from the Streptomyces strain RV15 derived from a marine sponge by Dr. U. Abdelmohsen. Their structures were established as cyclo-(ß-AFA-Ser-Gln-Asn-Tyr-Asn-Ser-Thr) by spectroscopic analysis using 2D NMR techniques and CID-MS/MS in the course of this thesis. In conclusion, the present work contributes to the discovery of novel antiplasmodial naphthylisoquinoline alkaloids and antitumoral naphthoquinones, which will pave the way for future studies on these two classes of compounds.
PTPN22 encodes the lymphoid tyrosine phosphatase Lyp that can dephosphorylate Lck, ZAP-70 and Fyn to attenuate TCR signaling. A single-nucleotide polymorphism (C1858T) causes a substitution from arginine (R) to tryptophan (W) at 620 residue (R620W). Lyp-620W has been confirmed as a susceptible allele in multiple autoimmune diseases, including type 1 diabetes (T1D). Several independent studies proposed that the disease-associated allele is a gain-of-function variant. However, a recent report found that in human cells and a knockin mouse containing the R620W homolog that Ptpn22 protein degradation is accelerated, indicating Lyp-620W is a loss-of-function variant. Whether Lyp R620W is a gain- or loss-of-function variant remains controversial. To resolve this issue, we generated two lines (P2 and P4) of nonobese diabetic (NOD) mice in which Ptpn22 can be inducibly silenced by RNAi. We found long term silencing of Ptpn22 increased spleen cellularity and regulatory T (Treg) cell numbers, replicating the effect of gene deletion reported in the knockout (KO) B6 mice. Notably, Ptpn22 silencing also increased the reactivity and apoptotic behavior of B lymphocytes, which is consistent with the reduced reactivity and apoptosis of human B cells carrying the alleged gain-of-function PTPN22 allele. Furthermore, loss of Ptpn22 protected P2 KD mice from spontaneous and Cyclophosphamide (CY) induced diabetes. Our data support the notion that Lyp-620W is a gain-of-function variant. Moreover, Lyp may be a valuable target for the treatment of autoimmune diseases.