Apelin, the endogenous ligand to the apelin receptor, is a small peptide involved with cardiovascular regulation. Using nuclear magnetic resonance (NMR) spectroscopy, I demonstrate that at low temperature, residues R6-L9 and G13-F17 of apelin are more structured than the rest of the peptide. I also study the interactions of apelin with sodium dodecylsulphate (SDS), dodecylphosphocholine (DPC) and 1-palmitoyl-2-hydroxy-sn- glycero-3-[phospho-RAC-(1-glycerol)] (LPPG) micelles. Apelin binds to SDS micelles through residues R6-L9, with structure being induced in this region as well as the C- terminus of the peptide. The binding to micelles along with the corresponding change in structure make it likely that apelin binds to the apelin receptor following the membrane catalysis hypothesis. NMR spectroscopy was used to determine the structure of the N- terminal tail and first transmembrane segment of the apelin receptor (AR55) in DPC micelles. AR55 has two disrupted helices from D14-K25 and from A29-K57. The second helix is the membrane spanning region of AR55 and has a significant kink located at N46. Mutagenesis of the apelin receptor and functional assays indicate that G42, G45 and N46 are essential for the proper trafficking and function of AR. In the N-terminal tail, the functionally critical residues E20 and D23 form an anionic face that could take part in initial binding of apelin to AR. The structure of AR55 was also determined in SDS micelles, LPPG micelles and a 1:1 water: 1,1,1,3,3,3-hexafluoroisopropanol (HFIP) solution. Overall, the micelle spanning region of AR55 has a consistent structure with a kink near N46. The N-terminal tail of AR55 is more variable, having similar structures in the micelle conditions but being largely helical in 50% HFIP. NMR relaxation experiments indicate that the N-terminal tail of AR55 undergoes much more motion in LPPG micelles compared to SDS and DPC micelles. Finally, I created a program named MC-HELAN that characterizes the kinks that occur in protein helices. I used MC- HELAN to analyze all non-redundant membrane protein structures as of March 2010. Membrane protein helix kinks are remarkably common and diverse. Initial attempts to predict membrane protein kinks using only the protein sequence were unsuccessful.

Ph.D.

Cardiac hypertrophy and heart failure (HF) remains one of the major health problems in the UK and worldwide. However, advances in their management are limited because the underlying pathological mechanisms are not completely understood. Therefore, it is important to understand novel signalling pathways leading to HF. Myocardial hypertrophy is a crucial pathophysiological process that can lead to the development of HF. Signalling initiated by members of G-protein-coupled receptors (GPCRs) proteins plays an important role in mediating cardiac hypertrophy. One member of this family, the G-protein coupled receptor 99 (GPR99), may have a crucial role in the heart because it acts as a receptor for alpha-ketoglutarate, a metabolite that is elevated in heart failure patients. GPR99 is expressed in the heart, but its precise function during cardiac pathophysiological processes is unknown. The aim of this PhD study is to investigate the role of GPR99 during cardiac hypertrophy. In this study I used in vivo and in vitro approaches to investigate whether GPR99 is directly involved in mediating cardiac hypertrophy. Mice with genetic deletion of GPR99 (GPR99-/-) exhibited a significant increase in hypertrophy following two weeks of transverse aortic constriction (TAC) as indicated by heart weight/tibia length ratio (HW/TL). In addition, GPR99-/- mice displayed increased cardiomyocytes cross-sectional area (CSA) after TAC compared to wild-type (WT) littermates. Hypertrophic markers such as brain natriuretic peptide (BNP) and β-myosin heavy chain (β-MHC) were also elevated in GPR99-/- mice following TAC compared to WT mice. Although interstitial fibrosis was indistinguishable in both genotypes after TAC, a precursor of fibrosis, collagen, type III, alpha1 (COL3A1) was elevated in GPR99-/- mice compared to WT mice after TAC. The baseline cardiac function as indicated by ejection fraction (EF) and fractional shortening (FS) were reduced in GPR99-/- mice compared to WT littermates following TAC. Furthermore, left ventricular end-diastolic diameter (LVEDD), left ventricular end-systolic diameter (LVESD), interventricular septum wall thickness (IVS) and posterior wall thickness at diastole (PW) indicated profound wall thickening and enlargement of the left ventricular (LV) chamber in GPR99-/- mice compared to WT littermates after TAC. In an attempt to examine the mechanism through which GPR99 signals during hypertrophy, I performed molecular analyses based on the data from yeast two hybrid screening showing that GPR99 interacted with COP9 signalosome element 5 (CSN5). Using immunoprecipitation assay, I found that GPR99 formed a ternary complex with CSN5 and non-receptor tyrosine kinase 2 (TYK2). TYK2 is known as a regulator of pro-hypertrophic molecules including signal transducer and activation of transcription 1 (STAT1) and STAT3. I found that the activation of these molecules was increased in GPR99-/- mice following TAC and correspondingly, adenovirus-mediated overexpression of GPR99 in neonatal rat cardiomyocytes (NRCM) blunted TYK2 phosphorylation. In conclusion, my study has identified GPR99 as a novel regulator of pathological hypertrophy via the regulation of the STAT pathway. Identification of molecules that can specifically activate or inhibit this receptor may be very useful in the development of a new therapeutic approach for cardiac hypertrophy in the future.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第18675号 / 医博第3947号 / 新制||医||1007(附属図書館) / 31608 / 京都大学大学院医学研究科医学専攻 / (主査)教授 清水 章, 教授 渡邉 大, 教授 松田 道行 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

Breast cancer is the most common type of cancer among women in the United States and metastasis is the leading cause of mortality in patients diagnosed with malignant breast cancer. The receptor of thromboxane A2 (TxA2), TP, is a member of the G-protein coupled receptor family. Increased expression of TP at RNA level was found to correlate with a poor prognosis in breast cancer patients; however, it is unknown how TP expression and activities are involved in breast cancer progression. Here we report that TP is expressed in breast cancer cells at both RNA and protein levels. And further, activation of this receptor elicits rapid activation of small GTPase RhoA and cytoskeleton reorganization. We also found that knockdown of TP expression or inhibition of TP activation by SQ29548, a TP antagonist, reduces tumor cell motility, reduce tumor cell extravasation from the circulatory system, and most importantly, reduce breast cancer metastasis in vivo. These data provide compelling evidence suggesting that the TxA2-TP pathway plays an important role in breast cancer metastasis.

The CXC chemokine receptor 4 (CXCR4) and the atypical chemokine receptor 3 (ACKR3) are seven transmembrane receptors that are involved in numerous pathologies, including several types of cancers. Both receptors bind the same chemokine, CXCL12, leading to significantly different outcomes. While CXCR4 activation generally leads to canonical GPCR signaling, involving Gi proteins and β‐arrestins, ACKR3, which is predominantly found in intracellular vesicles, has been shown to signal via β‐arrestin‐dependent signaling pathways. Understanding the dynamics and kinetics of their activation in response to their ligands is of importance to understand how signaling proceeds via these two receptors. In this thesis, different Förster resonance energy transfer (FRET)‐based approaches have been combined to individually investigate the early events of their signaling cascades. In order to investigate receptor activation, intramolecular FRET sensors for CXCR4 and ACKR3 were developed by using the pair of fluorophores cyan fluorescence protein and fluorescence arsenical hairpin binder. The sensors, which exhibited similar functional properties to their wild‐type counterparts, allowed to monitor their ligand-induced conformational changes and represent the first RET‐based receptor sensors in the field of chemokine receptors. Additional FRET‐based settings were also established to investigate the coupling of receptors with G proteins, rearrangements within dimers, as well as G protein activation. On one hand, CXCR4 showed a complex activation mechanism in response to CXCL12 that involved rearrangements in the transmembrane domain of the receptor followed by rearrangements between the receptor and the G protein as well as rearrangements between CXCR4 protomers, suggesting a role of homodimers in the activation course of this receptor. This was followed by a prolonged activation of Gi proteins, but not Gq activation, via the axis CXCL12/CXCR4. In contrast, the structural rearrangements at each step of the signaling cascade in response to macrophage migration inhibitory factor (MIF) were dynamically and kinetically different and no Gi protein activation via this axis was detected. These findings suggest distinct mechanisms of action of CXCL12 and MIF on CXCR4 and provide evidence for a new type of sequential signaling events of a GPCR. Importantly, evidence in this work revealed that CXCR4 exhibits some degree of constitutive activity, a potentially important feature for drug development. On the other hand, by cotransfecting the ACKR3 sensor with K44A dynamin, it was possible to increase its presence in the plasma membrane and measure the ligand‐induced activation of this receptor. Different kinetics of ACKR3 activation were observed in response to CXCL12 and three other agonists by means of using the receptor sensor developed in this thesis, showing that it is a valuable tool to study the activation of this atypical receptor and pharmacologically characterize ligands. No CXCL12‐induced G protein activation via ACKR3 was observed even when the receptor was re-localized to the plasma membrane by means of using the mutant dynamin. Altogether, this thesis work provides the temporal resolution of signaling patterns of two chemokine receptors for the first time as well as valuable tools that can be applied to characterize their activation in response to pharmacologically relevant ligands. / Der CXC Chemokin‐Rezeptor 4 (CXCR4) und der atypische Chemokin‐Rezeptor 3 (ACKR3) sind heptatransmembranäre Rezeptoren, die in zahlreichen Krankheitsbildern eine Rolle spielen, wie in einigen Krebsarten. Beide Rezeptoren werden zwar von dem gleichen Chemokin CXCL12 aktiviert, allerdings mit unterschiedlichen Signalweiterleitungsmustern. Die Aktivierung von CXCR4 führt zu kanonischer GPCR Signaltransduktion über Gi‐Proteine und β‐Arrestine. Die Signalweiterleitung des Rezeptors ACKR3 hingegen, welcher hauptsächlich in intrazellulären Vesikeln vorliegt, erfolgt über ß‐Arrestinabhängige Signalwege. Es ist von großer Wichtigkeit die Dynamik und Kinetik dieser beiden Rezeptoren hinsichtlich der Aktivierung durch ihre Liganden und der Signalweiterleitung zu verstehen. In dieser Arbeit wurden verschiedene Förster‐Resonanzenergietransfer (FRET) Anwendungen kombiniert, um die frühen Phasen der Signal‐Kaskade von CXCR4 und ACKR3 zu untersuchen. Zur genaueren Aufklärung der Rezeptoraktivierung wurden intramolekulare FRET‐Sensoren entwickelt, hierzu wurden die Fluorophore Cyan‐fluoreszierendes Protein und engl. fluorescence arsenical hairpin binder verwendet. Die generierten Sensoren zeigten ähnliche funktionelle Eigenschaften wie die unveränderten Rezeptoren. Liganden‐induzierte Änderungen der Rezeptorkonformation können mittels dieser Sensoren beobachtet werden und stellen die ersten RET‐basierten Sensoren auf dem Forschungsgebiet der Chemokin‐Rezeptoren dar. Weitere FRET‐basierte Methoden wurden zur Untersuchung von Interaktionen zwischen Rezeptor und G‐Protein, Neuanordnung von Dimeren, sowie der G‐Protein Aktivierung eingesetzt und für beide Chemokin‐Rezeptoren etabliert. CXCR4 zeigte einen komplexen Aktivierungsmechanismus nach Stimulation durch CXCL12, bei welchem zunächst eine Neuordnung der Rezeptor‐Transmembrandomäne gefolgt von Neuordnungen zwischen Rezeptor und G‐Protein und zuletzt eine Neuordnung zwischen CXCR4 Protomeren erfolgte. Dies impliziert, dass im Aktivierungsprozess des Rezeptors Homodimere eine Rolle spielen. Zudem wurde eine verlängerte Gi ‐Protein Aktivierung gegenüber der Gq‐Protein Aktivierung bei CXCL12 stimuliertem CXCR4 beobachtet. Hingegen zeigte eine Stimulierung mit dem Macrophage Migration Inhibitory Factor (MIF) bei jedem Schritt der frühen Singal‐Kaskade veränderte Dynamiken und Kinetiken im Vergleich zu CXCL12. Darüber hinaus konnte keine Gi ‐Protein Aktivierung festgestellt werden. Dieser Befund zeigt individuelle Mechanismen für MIF und CXCL12 am CXCR4‐Rezeptor und liefert Belege für eine neuer Art von sequenziellen Signalweiterleitungen an GPCRs. Eine wichtige Beobachtung dieser Arbeit für eine potentielle Medikamentenentwicklung ist das CXCR4 ligandenunabhängige Aktivität zeigt. Um die Aktivierung des ACKR3 Sensors messen zu können wurde durch eine Co‐Transfektion mit K44A Dynamin eine höhere Membranständigkeit erreicht. CXCL12 und drei weiteren Agonisten zeigten am hier entwickelten ACKR3‐Sensor unterscheidbare Kinetiken. Mit diesem wertvollen Werkzeug können Liganden an diesem atypischen Rezeptor pharmakologisch charakterisiert werden. Es konnte keine CXCL12‐induzierte G‐Protein Aktivierung gemessen werden, trotz der stärkeren Präsenz an der Plasmamembran mit Hilfe der Dynamin‐Mutante. In Summe liefert diese Arbeit zum ersten Mal eine zeitliche Auflösung von Signalweiterleitungsmustern von zwei Chemokin‐Rezeptoren sowie wertvolle Werkzeuge zur Charakterisierung der frühen Phase der Signal‐Kaskade durch andere pharmakologisch relevanten Liganden.

No description available.

No description available.

Oocyte maturation (OM) in teleosts is under precise hormonal control by estrogens and progestins. We show here that estrogens activate an epidermal growth factor receptor (EGFR) signaling pathway through the G protein-coupled estrogen receptor (GPER) to maintain meiotic arrest of full-grown zebrafish (Danio rerio) oocytes in an in vitro germinal vesicle breakdown (GVBD) bioassay. A GPER- specific agonist decreased OM and a GPER-specific antagonist increased spontaneous OM, whereas specific nuclear estrogen receptor (ERα and ERβ) agonists did not affect OM, which suggests the inhibitory action of estrogens on OM are solely mediated through GPER. Furthermore, a peptide-bound estrogen, which cannot enter the oocyte, decreased GVBD, showing that these estrogen actions are mediated through a membrane receptor. Treatment of oocytes with actinomycin D, a transcription inhibitor, did not block the inhibitory effects of estrogens on OM, indicating that estrogens act via a nongenomic mechanism to maintain oocyte meiotic arrest. EGFR mRNA was detected in denuded zebrafish oocytes by reverse transcription polymerase chain reaction (RT-PCR). Therefore, the potential role of transactivation of EGFR in estrogen inhibition of OM was investigated. The matrix metalloproteinase inhibitor, ilomastat, which prevents the release of heparin-bound epidermal growth factor (HB-EGF), increased spontaneous OM. Moreover, specific EGFR1 (ErbB1) inhibitors and inhibitors of extracellular-related kinase 1 and 2 (ERK1/2) increased spontaneous OM. Previously, estrogens have been shown to increase 3’-5’-cyclic adenosine mono phosphate (cAMP) levels through GPER in zebrafish oocytes during meiotic arrest. Taken together these present results suggest that estrogens also act through GPER to maintain meiotic arrest through a second signaling pathway involving transactivation of EGFR and activation of ERK 1 and 2. / text

Lysophosphatidic acid (LPA) belongs to a class of extracellular lipid signaling molecules. In the vasculature, LPA may regulate platelet activation and modulate endothelial and smooth muscle cell function. LPA has therefore been proposed as a mediator of cardiovascular disease. The bulk of circulating LPA is produced from plasma lysophosphatidylcholine (LPC) by autotaxin (ATX), a secreted lysophospholipase D (lysoPLD). Early studies suggest that some of the production of circulating LPA is platelet-dependent. ATX possesses an N-terminal somatomedin B-like domain suggesting the hypothesis that ATX interacts with platelet integrins which may localize ATX to substrate in the membrane and/or alter the catalytic activity of ATX. Using static adhesion and soluble binding assays we found that ATX does indeed bind to platelets and cultured mammalian cells in an integrin-dependent manner which is blocked by integrin function-blocking peptides and antibodies. This binding increases both the activity of ATX and localization of its product, LPA, to the platelet/cell membrane. LPA is generally stimulatory to human platelets although platelets from a small population of donors are refractory to LPA stimulation. Likewise LPA is inhibitory to murine platelets. We previously found that LPA receptor pan-antagonists reduce agonist-induced platelet activation, and partial stimulation of LPA5 specifically increases platelet activation in humans. Since both LPA5 and LPA4 are present at significant levels in human platelets, we hypothesized that LPA4 is responsible for an inhibitory pathway and LPA5 is responsible for an inhibitory pathway. We used mice deficient in LPA4 to test this model. Isolated platelet function tests revealed no major difference between lpa4-/- mice compared with WT mice although lpa4-/- mice were more prone to FeCl3-induced thrombosis. Paradoxically, chimeric mice reconstituted with lpa4-/- deficient bone marrow derived cells were protected from thrombosis. These discrepancies may be explained by involvement of endothelial cells and the relative scarcity of LPA receptors in murine platelets compared with human platelets. Taken together, these results demonstrate two critical regulators of LPA signaling and open up new avenues to further our understanding of atherothrombosis.