Tesis sobre el tema "Liver sinusoidal endothelial cell"

Owing to its strategic position in the liver sinusoid, pathologic and morphologic alterations of the Liver Sinusoidal Endothelial Cell (LSEC) have far-reaching repercussions for the whole liver and systemic metabolism. LSECs are perforated with fenestrations, which are pores that facilitate the transfer of lipoproteins and macromolecules between blood and hepatocytes. Loss of LSEC porosity is termed defenestration, which can result from loss of fenestrations and/ or decreases in fenestration diameter. Gram negative bacterial endotoxin (Lipopolysaccharide, LPS) has marked effects on LSEC morphology, including induction LSEC defenestration. Sepsis is associated with hyperlipidemia, and proposed mechanisms include inhibition of tissue lipoprotein lipase and increased triglyceride production by the liver. The LSEC has an increasingly recognized role in hyperlipidemia. Conditions associated with reduced numbers of fenestrations such as ageing and bacterial infections are associated with impaired lipoprotein and chylomicron remnant uptake by the liver and consequent hyperlipidemia. Given the role of the LSEC in liver allograft rejection and hyperlipidemia, changes in the LSEC induced by LPS may have significant clinical implications. In this thesis, the following major hypotheses are explored: 1. The Pseudomonas aeruginosa toxin pyocyanin induces defenestration of the LSEC both in vitro and in vivo 2. The effects of pyocyanin on the LSEC are mediated by oxidative stress 3. Defenestration induced by old age and poloxamer 407 causes intrahepatocytic hypoxia and upregulation of hypoxia-related responses 4. Defenestration of the LSEC seen in old age can be exacerbated by diabetes mellitus and prevented or ameliorated by caloric restriction commencing early in life

Doctor of Philosophy(PhD)<br>Owing to its strategic position in the liver sinusoid, pathologic and morphologic alterations of the Liver Sinusoidal Endothelial Cell (LSEC) have far-reaching repercussions for the whole liver and systemic metabolism. LSECs are perforated with fenestrations, which are pores that facilitate the transfer of lipoproteins and macromolecules between blood and hepatocytes. Loss of LSEC porosity is termed defenestration, which can result from loss of fenestrations and/ or decreases in fenestration diameter. Gram negative bacterial endotoxin (Lipopolysaccharide, LPS) has marked effects on LSEC morphology, including induction LSEC defenestration. Sepsis is associated with hyperlipidemia, and proposed mechanisms include inhibition of tissue lipoprotein lipase and increased triglyceride production by the liver. The LSEC has an increasingly recognized role in hyperlipidemia. Conditions associated with reduced numbers of fenestrations such as ageing and bacterial infections are associated with impaired lipoprotein and chylomicron remnant uptake by the liver and consequent hyperlipidemia. Given the role of the LSEC in liver allograft rejection and hyperlipidemia, changes in the LSEC induced by LPS may have significant clinical implications. In this thesis, the following major hypotheses are explored: 1. The Pseudomonas aeruginosa toxin pyocyanin induces defenestration of the LSEC both in vitro and in vivo 2. The effects of pyocyanin on the LSEC are mediated by oxidative stress 3. Defenestration induced by old age and poloxamer 407 causes intrahepatocytic hypoxia and upregulation of hypoxia-related responses 4. Defenestration of the LSEC seen in old age can be exacerbated by diabetes mellitus and prevented or ameliorated by caloric restriction commencing early in life

Liver sinusoidal endothelial cells (LSEC) constitute a unique population of endothelial cells with specialised liver-specific morphologic features and functions. LSEC are the only endothelial cells with fenestrations and which lack an organised basement membrane. They are involved in hepatic stellate cell (HSC) quiescence, endocytosis of small particles, selective transfer of substances from the blood, in the hepatic sinusoid, to the parenchymal cells and in liver regeneration. As the group of cells that form the inner lining of the capillaries of the liver sinusoids, and being the first to be in contact with blood-borne particles, pathogens, and xenobiotics, they are prone to the deleterious effects of these. The aims of this thesis were to investigate the unique features of human liver sinusoidal endothelial cells (HLSEC) in comparison with endothelial cells from other vascular beds, evaluate the sensitivities of HLSEC to a range of hepatotoxic drugs, including small-molecule receptor tyrosine kinase inhibitors (RTKIs), such as regorafenib, and to explore the role of HLSEC in a triculture human liver microtissue. Results obtained from this study showed that HLSEC expressed phenotypic features of vascular and lymphatic endothelial cells, particularly vascular endothelial growth factor receptor 2 (VEGFR-2) which could be activated by VEGF-A to stimulate cell proliferation, migration and tubular morphogenesis. HLSEC also expressed functional VEGFR-3. Transcriptomic analysis indicated that HLSEC expressed specialised genes, such as plasmalemma vesicle associated protein (PLVAP), that support its liver-specific structure and functions. HLSEC were more sensitive to a range of small-molecule receptor tyrosine kinase inhibitors than other hepatic cells. (primary human hepatocytes [PHH] and human hepatic fibroblasts [HHF]) and endothelial cells from other vascular beds (human dermal microvascular endothelial cells and human dermal lymphatic endothelial cells). Regorafenib inhibited the activation of VEGFR-2 thereby abrogating cell proliferation, migration, tubular morphogenesis as well as upregulation of angiocrine factors involved in liver regeneration following activation by vascular endothelial growth factor A (VEGF-A). Regorafenib also caused a disruption of cytoskeletal structure of HLSEC and induced apoptosis via activation of caspase 3. Triculture liver microtissues formed with PHH, HLSEC and HHF were vascularised with higher expression of liver-specific drug-metabolising enzymes in comparison with the same combination of cells cultured as a monolayer. However, metabolic competence of triculture liver microtissues was significantly lower than in their monoculture counterparts (consisting of PHH only). This study has further confirmed the uniqueness of HLSEC as a specialised endothelial cell adapted to its anatomical role, which could respond to a range growth factors to initiate endothelial cell-specific functions. It has also been demonstrated that HLSEC are a direct target of hepatotoxic drugs. Triculture liver microtissues generated with PHH, HLSEC and HHF showed less metabolic competence than their PHH-only counterparts. Future studies need to investigate the role of RTKIs in vascular toxicity using in vivo models of sinusoidal obstruction syndrome (SOS) and liver regeneration. Finally, it would be informative to investigate the possibility of identifying HLSEC-specific biomarkers of liver toxicity.

The majority of liver studies focus on the hepatocyte however the work of this thesis investigates the vital role of the liver sinusoidal endothelial cell (LSEC). LSECs line the liver sinusoids forming a protective barrier between the blood and hepatocytes. The LSEC cytoplasm resembles a sieve, perforated with thousands of transcellular pores of approximately 50-150 nm in diameter called fenestrations, and is underlined by a very sparse extracellular matrix. This facilitates the virtually unimpeded passage of fluid and substances smaller than fenestrations from the blood such as drugs and nutrients, and size-dependent filtration of lipoproteins, to and from hepatocytes. Fenestrations are dynamic structures, in that their size and number can be modulated by hormones, drugs, hepatotoxins, and diseases. Reduction of LSEC fenestration size and number (defenestration) is associated with ageing and pathological states, and is also a cause of hyperlipidemia and reduced drug clearance, thus changes in LSEC morphology can affect the entire organism. This thesis aims to broaden knowledge of the role and ultrastructure of the LSEC in physiological and toxicological states by investigating: whether there is fenestration modulation during fasting that could facilitate increased nutrient exchange between the blood and hepatocytes; whether changes to LSEC ultrastructure during acetaminophen hepatotoxicity are consistent with exacerbation of liver injury and/or with the facilitation of liver regeneration after severe necrosis; whether a substance that targets the LSEC could have a therapeutic benefit in acetaminophen hepatotoxicity by protecting the microvasculature from damage; whether isolation and culture of LSECs from ageing rats maintain the ageing (defenestrated) phenotype, and thus whether it is a valid method to study therapeutic substances in vitro that could reverse defenestration-related ailments associated with normal ageing.

Hepatitis C virus (HCV) infection is a major cause of global morbidity, causing chronic liver injury that can progress to cirrhosis and hepatocellular carcinoma. The liver is a large and complex organ containing multiple cell types, including hepatocytes, sinusoidal endothelial cells (LSEC), stellate cells, Kupffer cells and biliary epithelial cells. Hepatocytes are the major reservoir supporting HCV replication, however, the role of non-­‐parenchymal cells in the viral lifecycle remain largely unexplored. Endothelial cell hepatocyte co-­‐cultures were established to study the role of LSEC in HCV biology. Vascular endothelial growth factor (VEGF-­‐A) regulated transcripts were profiled by microarray to identify factors modulating HCV replication. The initial studies indicated that rather than transmitting HCV to permissive hepatocytes LSEC were protective in HCV infection. Co-­‐culture of epithelial and endothelial cell showed that LSEC limit hepatocyte permissivity to HCV infection via cell contact-­‐dependent mechanisms and by the expression of soluble mediator(s) that are regulated by VEGF-­‐A. Transcript analysis identified LSEC expression of bone morphogenetic protein 4 (BMP4), a novel proviral molecule that is negatively regulated by VEGF-­‐A via a VEGF receptor-­‐2 (VEGFR-­‐2) MAPK dependent pathway. Consistent with the in vitro data I observed increased BMP4 expression and reduced VEGFR-­‐2 activation in inflamed liver tissue. These studies show a novel role for LSEC and BMP4 in HCV infection and highlight BMP4 as a new therapeutic target for treating liver disease.

Ageing is associated with increased prevalence of metabolic syndrome, as well as impaired glucose metabolism, hyperinsulinemia and insulin resistance. The mechanism underlying these associations is poorly understood and is likely to be complex and multifactorial. The liver is the key target for insulin action and while the endothelium has been shown to influence insulin activity in muscle and fat, the role of the liver sinusoidal endothelium on the action of insulin in the liver is unknown. The liver sinusoidal endothelium is unique: it is perforated with transcellular pores called fenestrations that facilitate unimpeded passage of substrates between blood and hepatocytes. A constellation of age-related morphological changes in the liver sinusoidal endothelium known as pseudocapillarisation have been described in various species including rats, baboons and humans. During ageing, the liver sinusoidal endothelium thickens, there is basement membrane deposition, and the fenestrations are significantly reduced in size and number (defenestration). Age-related pesudocapillarisation has been shown previously to impede the transfer of lipoproteins and medications across the hepatic sinusoidal endothelium. This thesis tests the hypothesis that changes in the ageing liver contribute to age-related insulin resistance, with alterations of the liver sinusoidal endothelial cell leading to age-related impairment of insulin action and insulin resistance/glucose metabolism. This work aims to improve the understanding of the effects of ageing processes in the liver on insulin action and glucose metabolism. It investigates the role of age-related pseudocapillarisation and the acutely induced poloxamer 407 (P407) model of defenestration in hepatic disposition of insulin and glucose metabolism. This thesis also investigates the effect of P407 on the relationship between membrane rafts and fenestrations in SKHep1 cells, a cell line of liver endothelial origin and isolated LSECs. Finally, the effects of dietary macronutrients and calorie intake on fenestrations in old age are examined. The work contained in this thesis aims to examine the role of age-related pseudocapillarisation in one of the major causes of age-related disease and disability, insulin resistance. In doing so it explores the potential mechanisms involved in these changes and how we may alter the progression of ageing through nutritional intervention.

The cellular microenvironment in vivo consists of both mechanical and chemical signals, which drive cell function and fate. These signals include the composition, architecture, and mechanical properties of the extracellular matrix (ECM), signaling molecules secreted by cells into their surroundings, as well as physical interactions between neighboring cells. Cells are able to interact with their surroundings through a number of different mechanisms such as remodeling of the ECM through adhesion, contraction, degradation, and deposition of proteins, as well as the secretion of pro- or anti-inflammatory molecules. In diseased states, where homeostasis has been perturbed, inflammatory signals are secreted which can modify the cellular microenvironment. Diseased states such as cancer and fibrosis are often associated with the excessive production of ECM proteins that subsequently lead to an increase in tissue stiffness and changes to ECM architecture. Such changes to the mechanical properties of the cellular microenvironment affect the cytoskeletal arrangement, migration and adhesion of both the parenchymal cells, as well as immune response cells, which migrate to the sites of injury. Further understanding of the inflammatory responses and their relationships to tissue stiffness and ECM architecture could aid in the development of novel strategies to predict diseases as well as to target and monitor therapies. Since inflammation and mechanical properties of the affected tissue are closely interlinked, obtaining a detailed understanding of the interplay between the properties of the microenvironment and the cells that reside within it will be very beneficial to obtain physiologically relevant information. We have investigated the combinatorial effects of matrix stiffness, and architecture in the presence of co-cultures of cells to determine the overall effect on cellular responses and phenotypes. We have conducted studies on co-cultures of cells in 2D and 3D environments to identify how cellular behavior is affected by dimensionality.<br>PHD

La malaltia hepàtica per dipòsit de greix (MHDG) constitueix la manifestació hepàtica de la síndrome metabòlica. La seva incidència augmenta en les societats occidentals paral·lelament a la de l'obesitat. El dipòsit de greix intrahepàtic pot conduir al desenvolupament de dany hepatocitari, inflamació, fibrosi i cirrosi, però no es coneixen els mecanismes que promouen la progressió de la malaltia. No existeix un tractament farmacològic eficaç per a aquesta malaltia. La resistència a la insulina, el fet fisiopatològic subjacent a la síndrome metabòlica, condueix al desenvolupament de disfunció endotelial perifèrica. No obstant això, l'endoteli sinusoidal hepàtic és fenotípicament diferent al perifèric, i es desconeix si la resistència a la insulina genera disfunció endotelial en la circulació hepàtica. L'adequada funció de l'endoteli sinusoïdal és necessària per al manteniment d'un ambient antiinflamatori, antifibrogènic i antitrombòtic en el fetge. En aquest estudi hipotetitzem que a la MHDG, associada a resistència a la insulina, existeix disfunció endotelial sinusoïdal que contribueix a la progressió del dany hepàtic. Els objectius d'estudi van ser determinar en un model animal de EHDG si aquesta s'associa amb disfunció endotelial del sinusoïdal, els possibles mecanismes implicats, incloent la seva relació amb la resistència a la insulina, i si aquesta contribueix al desenvolupament de lesió hepàtica. Per respondre a aquestes preguntes es van caracteritzar les respostes vasculars de la circulació hepàtica mitjançant perfusió aïllada de fetge, i es van estudiar els mecanismes que determinen la biodisponibilitat de NO mitjançant tècniques moleculars en fetge complet. Es van estudiar, a més els efectes de la correcció de la insulino resistència sobre aquestes alteracions. En el primer estudi, després d'administrar a rates durant 3 dies una dieta rica en greix, aquestes van presentar esteatosi en absència d'inflamació. Això es va acompanyar d'un increment del contingut intrahepàtic de triglicèrids i àcids grassos lliures. Les rates amb fetges esteatòsics van presentar resistència vascular a la insulina i, més concretament, resistència a la insulina a nivell del endoteli sinusoidal hepàtic. Això es va acompanyar d'una menor capacitat de la insulina per fosforilar eNOS. Aquestes alteracions es van deure, si més no en part, per la regulació a l'alça de iNOS. El pre-tractament amb un fàrmac sensibilitzador de la insulina va atenuar la l'acumulació intrahepàtica de triglicèrids i àcids grassos lliures, va prevenir el desenvolupament de resistència a la insulina a nivell i va restaurar la capacitat de la insulina per fosforilar eNOS en rates alimentades durant 3 dies amb una dieta rica en greix. En el segon estudi, vam demostrar que l'administració durant un mes d'una dieta rica en greix provoca esteatosi en absència d'inflamació i fibrosi. Això es va acompanyar d'alteracions inclosa en la síndrome metabòlica, com l'obesitat, hipertrigliceridèmia, hiperglucèmia, hiperinsulinèmia i hipertensió arterial. Aquestes rates van presentar un augment de la pressió portal de perfusió, a causa de canvis funcionals, així com disfunció endotelial sinusoïdal. A nivell molecular això es va traduir en un una menor fosforilació de Akt i eNOS en els fetges de rates alimentades amb una dieta rica en greix. Aquests canvis funcionals es van produir en absència de canvis fenotípics en les cèl•lules endotelials sinusoïdals. Concloem d'aquest segon estudi que els resultats obtinguts en aquest estudi reforçarien la idea de considerar la disfunció endotelial hepàtica com un esdeveniment precoç que podria constituir una diana terapèutica per tractar aquesta malaltia.<br>Non-alcoholic fatty liver disease (NAFLD) is the hepatic manifestation of the metabolic syndrome. NAFLD incidence is increasing in western societies, paralleling the increase of obesity. The deposit of fat in the hepatocytes leads in some patients to the development of hepatocyte injury, inflammation, fibrosis and cirrhosis, but the mechanisms that govern the progression of liver injury are not well understood. There is not an effective pharmacological treatment for NAFLD. Insulin-resistance, the underlying pathophysiological feature of the metabolic syndrome, is associated with peripheral endothelial dysfunction. However, the sinusoidal endothelium is phenotypically different from the peripheral endothelium, and it is not known whether insulin resistance induces sinusoidal endothelial dysfunction. An adequate sinusoidal endothelial function is required to maintain an anti-inflammatory, antifibrogenic and antithrombotic environment in the liver. We hypothesize in this project that in NAFLD, which is associated with insulin-resistance, there is sinusoidal endothelial dysfunction, and that this is relevant to the development of liver injury. In the first study, we have demonstrated that after 3-days high fat feeding, rats developed steatosis without the presence of inflammation. Insulin resistance was present at the liver vasculature, and specifically, at the liver sinusoidal endothelium. This was accompanied by a reduction of the ability of insulin to phosphorylate eNOS. The conclusions of this first study were that vascular insulin resistance precedes inflammation, and can contribute to the progression of the disease. In the second study we demonstrate that rats fed for one month with a high fat diet presented steatosis without fibrosis or inflammation. This was accompanied by alterations resembling metabolic syndrome, endothelial dysfunction and a decreased liver eNOS and Akt phosphorylation. This was due to functional alterations (and no structural) at the sinusoidal endothelial cells, without phenotypical changes occurring at this level. We concluded that endothelial dysfunction is a primary event that precedes inflammation in a model of NAFLD, and might constitute a useful target for devising new therapies for this disease.

Ageing is associated with ultrastructural changes in the liver, including a reduction in hepatocyte mitochondrial number density and porosity of the liver sinusoidal cells. Histologically-verifiable changes include lipid and collagen accumulation, and increased perisinusoidal von Willebrand factor (VWF) expression. Sirtuin 1 (SIRT1), an NAD+-dependent deacetylase, is renowned for its range of anti-ageing actions. This study evaluated whether the anti-ageing effects of SIRT1 would extend to these liver ageing characteristics. Here the effects of whole-body SIRT1 overexpression on the livers of young (3-5 month) and old (13-20 month) SIRT1 transgenic mice were evaluated and compared to their wildtype controls. In addition, because NAD+ levels diminish with age, some old animals were supplemented with the NAD+ precursor, nicotinamide mononucleotide (NMN), to remove any limitation of naturally available NAD+ on SIRT1 overexpression. NAD+ is a SIRT1 cofactor. Old age is orthodoxly associated with reduced mitochondrial function and number in many tissues. However, transmission electron microscopy determination of liver mitochondrial numerical density (mitochondrial particles per cytosolic µm2) showed that there was no change with age. Additionally, there was no intervention effect of SIRT1 overexpression or NMN supplementation. Old age is also conventionally associated with reduced number and size of pores in the liver sinusoidal endothelial cell called fenestrae. However, analysis of the liver blood vessels with scanning electron microscopy was unable to confirm an effect of age on the number and size of fenestrae. Analysis did, however, reveal that SIRT1 overexpression was associated with increased sinusoidal porosity (%) in older but not younger mice (F(1,8) = 14.9, p < 0.01). Furthermore, although SIRT1-mediated increases in fenestrae diameter were not statistically verifiable in a 2-way ANOVA analysis, when plotting maximal fenestrae diameter size against groups, there was a trend suggesting a shift in the size distribution towards larger fenestrae for SIRT1-overexpressing animals. Histological examination of the mice livers showed that age was associated with an increase in lipid and collagen levels. However, there was no effect of age on the presence of activated stellate cells or perisinusoidal expression of VWF. Mice that were treated with NMN were 0.11 times less likely to have elevated α-SMA expression than controls, yet a combination of SIRT1 overexpression and NMN supplementation led to a 20.43 times greater likelihood of elevated α-SMA expression. The lack of canonical ageing-associated changes in a number of regards raises the question of whether the older mice were sufficiently old for such effects to emerge. Future studies should examine mice of at least 24 months of age. In summary, although SIRT1-overexpression shows promise in promoting liver sinusoidal endothelial porosity, the attendant risk related to stellate cell activation may be prohibitive in the use of SIRT1-elevating strategies with respect to the liver. The possibility for tissue- specific SIRT1-elevating strategies remains for other tissue types which do not incur this extra risk. Finally, NMN might reduce stellate cell activation by a SIRT1-independent pathway. Collectively, the results raise the possibility of a novel pathway leading to stellate cell activation and a new therapeutic approach to reducing stellate cell activation – both possibilities have important implications for the treatment of hepatic fibrosis. Keywords: sirtuin 1 (SIRT1), nicotinamide mononucleotide (NMN), nicotinamide adenine dinucleotide (NAD), ageing, anti-ageing, liver, gastrocnemius muscle, liver sinusoidal endothelial cells (LSECs), fenestrae, defenestration, pseudocapillarization, mitochondria, mitochondrial numerical density, fibrosis, steatosis, hepatosteatosis, lipid, collagen, α-smooth muscle actin (α-SMA), von Willebrand factor (VWF), scanning electron microscopy, transmission electron microscopy, histology, haematoxylin and eosin, Sirius Red, immunohistochemistry.

Contexte et hypothèse : La stéatohépatite non alcoolique (NASH) est définie par une accumulation excessive de lipides dans le foie (stéatose), une atteinte hépatocytaire et une inflammation hépatique avec ou sans fibrose. La NASH peut évoluer vers la cirrhose et le carcinome hépatocellulaire. De récentes études suggèrent que des altérations microvasculaires et une dysfonction endothéliale sinusoïdale précèdent l'atteinte fibreuse et inflammatoire de la NASH. L'autophagie est un processus cellulaire par lequel du matériel cytoplasmique rejoint les lysosomes pour dégradation. L'autophagie a été étudiée dans les différents types cellulaires du foie, mais le rôle de l'autophagie dans les cellules endothéliales sinusoïdales du foie (LSECs) n'a à ce jour jamais été évalué dans la NASH. Le but de mon projet de thèse a été d'évaluer le rôle de l'autophagie dans les LSECs dans la NASH. Matériel et méthodes : (a) échantillons humains : j'ai utilisé des biopsies hépatiques humaines de patients avec un foie histologiquement normal, de patients atteints de stéatose ou de patients atteints de NASH, afin d'évaluer par microscopie électronique l'autophagie dans les LSECs. (b) LSECs en culture : j'ai testé l'effet du TNFa et de l'IL6, aux concentrations rapportées dans le sang portal d'individus atteints de syndrome métabolique, sur l'autophagie dans une lignée de LSECs soumises à des contraintes de cisaillement. Enfin, j'ai caractérisé l'effet du défaut d'autophagie sur le phénotype endothélial en utilisant une lignée de LSECs que j'ai transduites avec un shARN ciblant ATG5 pour les rendre déficientes en autophagie. (c) souris transgéniques : J'ai évalué l'effet du défaut d'autophagie endothéliale sur les stades précoces de la NASH, en utilisant une ligné de souris déficientes en autophagie spécifiquement dans l'endothélium (Atg5lox/lox VE-cadhérine-Cre) soumises à un régime riche en matières grasses pendant 16 semaines, et sur les stades avancés de la fibrose hépatique en traitant les souris Atg5lox/lox VE-cadhérine-Cre au tétrachlorure de carbone (CCl4). Résultats : (a) échantillons humains : Le pourcentage de LSECs contenant des vacuoles d'autophagie était 2 fois moins important dans le foie des patients atteints de NASH comparativement aux témoins et aux patients atteints de stéatose. (b) LSECs en culture : La combinaison de TNFa et d'IL6 réduisait significativement l'autophagie dans les LSECs. Cette réduction passait par l'inhibition de l'AMPKa. Les LSECs déficientes en autophagie surexprimaient les gènes inflammatoires Mcp1 et Rantes et la protéine VCAM1. Le défaut d'autophagie dans les LSECs induisait une surexpression des marqueurs de transition endothélio-mésenchymateuse a-SMA, Collagène1a1, Collagène1a2 et Tgf-b1. (c) souris transgéniques : Comparativement aux souris contrôles, les souris Atg5lox/lox VE-cadhérine-Cre avaient une inflammation hépatique augmentée (surexpression hépatique des gènes inflammatoires Mcp1 et Rantes et de la protéine VCAM1). La déficience en autophagie endothéliale augmentait la fibrose (surexpression hépatique des gènes fibrogéniques Collagène1a2 et Tgf-b1, de la protéine a-SMA et de la surface de collagène). Les souris Atg5lox/lox VE-cadhérine-Cre traitées au CCl4 présentaient plus de fibrose que les souris contrôles soumises au même traitement (surexpression hépatique des gènes fibrogéniques a-SMA, Collagène 1a1, Collagène 1a2 et Tgf-b1, augmentation de la surface de collagène). Conclusion : Un défaut d'autophagie dans les LSECs se produit chez les patients atteints de NASH. Le TNFa et l'IL6 présents dans le sang portal de ces patients pourraient être responsables de ce défaut en altérant l'activité de l'AMPKa. Le défaut d'autophagie dans les LSECs favorise le développement de l'inflammation et de la fibrose hépatique aux stades précoces et avancées de la maladie. Cibler l'autophagie dans les LSECs constitue une stratégie thérapeutique intéressante pour le traitement de la NASH<br>Background and Aims: Non alcoholic steatohepatitis (NASH) is defined as the excessive lipids accumulation in the liver, hepatocellular injury and inflammation with or without fibrosis. NASH has the potential for cirrhosis and hepatocellular carcinoma. Recent studies suggest that microvascular alterations and sinusoidal endothelial dysfunction precede inflammation and fibrosis in NASH. Autophagy is a cellular process by which the dysfunctional cytoplasmic material joins lysosomes for degradation. The role of autophagy in hepatocytes, in hepatic stellate cells and in liver monocyte-macrophages has been studied but nothing is known about the role of autophagy in liver sinusoidal endothelial cells (LSECs) in NASH. The aim of my thesis work was to investigate the potential implication of autophagy in LSECs in NASH and liver fibrosis. Method: (a) Human samples: I used liver biopsies from patient without liver histological abnormalities, with simple steatosis or with NASH to analyze autophagy in LSECs by electron microscopy. (b) Cultured LSECs: I tested the effect of TNFa and IL6 (at concentrations present in the portal venous blood of patients with metabolic syndrome) on autophagy in LSECs exposed to shear stress. I characterized the effect of autophagy deficiency on the phenotype of LSECs by transducing transformed LSECs with a shRNA targeting ATG5. (c) Transgenic mice: I analyzed the effect of a defect in endothelial autophagy on early stages of NASH, by using mice deficient in autophagy specifically in endothelial cells (Atg5lox/lox-VE-CadherinCre), fed a high fat diet (HFD) for 16 weeks, and on advanced stages of liver fibrosis by treating Atg5lox/lox-VE-CadherinCre mice with carbon tetrachloride (CCl4). Results: (a) Human samples: Patients with NASH had twice less LSECs containing autophagic vacuoles than patients without liver histological abnormalities or patients with simple steatosis. (b) Cultured LSECs: The combination of TNFa and IL6 decreased autophagy level in LSECs. This reduction of autophagy involved the inhibition of AMPKa. LSECs deficient in autophagy overexpressed Mcp1 and Rantes genes and VCAM1 protein expression. Deficiency in autophagy in LSECs induced the expression of the endothelial to mesenchymal transition markers a-SMA, Collagen1a1, Collagen1a2 and Tgf-b1. (c) Transgenic mice: As compared to littermate controls, mice deficient in endothelial ATG5 fed a HFD had a more frequent nodular liver surface, a higher liver inflammation (increased liver gene expression of Mcp-1 and Rantes and VCAM1 protein) and more liver fibrosis (increased liver gene expression of Collagen1a2 and Tgf-b1, higher expression a-SMA protein and more collagen deposition). Mice deficient in ATG5 in endothelial cells treated with CCl4 had more liver fibrosis (increased liver gene expression of a-SMA, Collagen1a1, Collagen 1a2 and Tgf-b1 and more collagen deposition). Conclusion: Autophagy is defective in LSECs of patients with NASH. TNFa and IL6 at concentrations present in the portal blood of patients with NASH could be responsible for this defect through the impairment of AMPKa activity. Autophagy defect in LSECs contributes to the development of liver inflammation and fibrosis at early and advanced stages of the disease. Stimulating endothelial autophagy could be an attractive strategy for NASH treatment

Die Einwanderung von T-Zellen in ein Gewebe wird durch selektive Wechselwirkungen mit vaskulären Endothelzellen kontrolliert. In der vorliegenden Arbeit wurde der Frage nachgegangen, ob Interaktionen zwischen Lebersinusendothelzellen (LSEC) und CD4+ T-Zellen die gewebespezifische Migration von CD4+ T-Zellen beeinflussen und damit Relevanz für den Verlauf spezifischer Immunantworten haben. Die Präsentation von Antigenen durch zytokinaktivierte LSEC erhöhte die Adhäsion und Transmigration antigenspezifischer CD4+ T-Zellen. Die Daten deuten auf eine Rolle des Lebersinusendothels bei der entzündungsinduzierten, antigenabhängigen Rekrutierung von CD4+ T-Zellen in das Lebergewebe hin. Eine antigenabhängige Aktivierung naiver CD4+ T-Zellen durch LSEC sowie deren Bereitstellung von Retinolsäure induzierte die Expression von darmspezifischen Homingrezeptoren auf CD4+ T-Zellen. LSEC-aktivierte CD4+ T-Zellen migrierten in das Darmgewebe von C57BL/6-Mäusen. Die Ergebnisse legen den Schluss nahe, dass LSEC einen darmspezifischen Homingphänotyp und damit die Migration von in der Leber aktivierten CD4+ T-Zellen in den Darm induzieren. Die Bereitstellung von Chemokinen durch LSEC mittels Transzytose und Immobilisierung verstärkte die Transmigration von CD4+ T-Zellen durch das Endothel. Die Gabe eines Inhibitors der endothelialen Chemokintranszytose während einer Concanavalin A-induzierten Autoimmunhepatitis supprimierte den Verlauf der Hepatitis und führte zu einer verminderten Migration von aktivierten CD4+ T-Zellen in das Lebergewebe. Diese Daten weisen dem Lebersinusendothel eine aktive Beteiligung in der chemokinabhängigen Rekrutierung von CD4+ T-Zellen in die Leber zu. In der vorliegenden Arbeit wurde die Modulation der gewebespezifischen Migration von CD4+ T-Zellen über Antigenpräsentation und Chemokinbereitstellung durch das Lebersinusendothel gezeigt und damit weitere spezifische Aspekte in der Funktion der Leber als immunologisches Organ beschrieben.<br>T-cell immigration into a tissue is controlled by selective interactions with vascular endothelial cells. The present study addressed the question if interactions between liver sinusoidal endothelial cells (LSEC) and CD4+ T cells influence the tissue-specific migration of CD4+ T cells and thus have relevance for the course of specific immune responses. Antigen presentation by cytokine-activated LSEC increased adhesion and transmigration of antigen-specific CD4+ T cells. These results indicate an involvement of LSEC in the inflammation-induced, antigen-specific migration of CD4+ T cells into the liver tissue. Antigen-specific activation of naive CD4+ T cells by LSEC and their supply of retinoic acid induced expression of gut-specific homing receptors on CD4+ T cells. LSEC-activated CD4+ T cells migrated into the intestine of C57BL/6 mice. The findings presented here imply that LSEC induce a gut-specific homing phenotype resulting in migration of liver-activated CD4+ T cells into the intestine. The active supply of chemokines by LSEC via transcytosis and immobilization enhanced transmigration of CD4+ T cells. Administration of an inhibitor of the endothelial chemokine transcytosis during Concanavalin A-induced autoimmune hepatitis suppressed hepatitis and resulted in reduced migration of activated CD4+ T cells into the liver tissue. The data show the impact of LSEC on the chemokine-dependent recruitment of CD4+ T cells into the liver. In the present study the modulation of the tissue-specific migration of CD4+ T cells by LSEC via antigen presentation and supply of chemokines was demonstrated. Thus, additional functional aspects concerning the immunologic functions of the liver were described.

The liver has a key role both from a metabolic and an immunological viewpoint. It is involved in the metabolism, or degradation of different molecules absorbed by the intestine, including xenobiotics, drugs and lipids, the synthesis and turnover of plasma proteins, the production of bile and the storage of glycogen and vitamin A. Unique amongst solid organs, the sinusoidal endothelium of the liver is perforated with pores, or fenestrations, and is not separated from hepatocytes by a basal lamina. It has been proposed that the fenestrations, also termed the ’liver sieve’, function as a bio-filter allowing free diffusion of small molecules (less than 100 nm in diameter) from the blood to the hepatocytes and Vice versa. Changes in these structures, as seen in ageing, could have important effects on hepatic function and in particular lipid metabolism. The liver also has unique immunological functions including T-cell activation and tolerance. This thesis explores potential new roles of the fenestrated sinusoidal endothelium in some aspects of liver immunology and ageing.

INTRODUCTION:Vascular injury and accumulation of red blood cells in the space of Disse (hemorrhage) is a characteristic feature of acetaminophen hepatotoxicity. However, the mechanism of nonparenchymal cell injury is unclear. Therefore, the objective was to investigate if either Kupffer cells or intracellular events in endothelial cells are responsible for the cell damage.RESULTS:Acetaminophen treatment (300 mg/kg) caused vascular nitrotyrosine staining within 1 h. Vascular injury (hemorrhage) occurred between 2 and 4 h. This paralleled the time course of parenchymal cell injury as shown by the increase in plasma alanine aminotransferase activities. Inactivation of Kupffer cells by gadolinium chloride (10 mg/kg) had no significant effect on vascular nitrotyrosine staining, hemorrhage or parenchymal cell injury. In contrast, treatment with allopurinol (100 mg/kg), which prevented mitochondrial injury in hepatocytes, strongly attenuated vascular nitrotyrosine staining and injury.CONCLUSIONS:Our data do not support the hypothesis that acetaminophen-induced superoxide release leading to vascular peroxynitrite formation and endothelial cell injury is caused by activated Kupffer cells. In contrast, the protective effect of allopurinol treatment suggests that, similar to the mechanism in parenchymal cells, mitochondrial oxidant stress and peroxynitrite formation in sinusoidal endothelial cells may be critical for vascular injury after acetaminophen overdose.

Auch unter physiologischen Bedingungen finden sich T-Zellen und andere Leukozyten nicht nur in den Sinusoiden, sondern auch im Parenchym der Leber. Da die Leber u. a. verschiedene Aufgaben für das Immunsystem übernimmt (z. B. Deletion aktivierter T Zellen, Induktion peripherer Toleranz), könnte die Akkumulation der T-Zellen in der Leber - neben der immunologischen Überwachung der Leber - Voraussetzung für ihre Modulation sein. In der vorliegenden Arbeit wurde der Einfluss von Leber-sinusoidalen Endothelzellen (LSEC), der Barriere zwischen Blut und Leber-Parenchym, auf CD4+ T-Zellen untersucht. Zum einen zeigte sich, dass die LSEC sowohl die spontane Transmigration der T-Zellen, als auch ihre Chemotaxis zu CXCL9 und CXCL12 effizienter unterstützen als andere Endothelien. Eine endotheliale Aktivierung durch die Chemokine wurde als Mechanismus ausgeschlossen. Dagegen schien eine effiziente Präsentation der Chemokine auf der luminalen LSEC-Oberfläche nach Aufnahme von abluminal für die gesteigerte Transmigration der T Zellen verantwortlich zu sein. Die LSEC könnten somit in vivo an der Rekrutierung von T-Zellen in die Leber beteiligt sein, indem sie eine rasche Wanderung der T-Zellen aus dem Blut ins Parenchym und möglicherweise auch zurück in die Zirkulation zulassen. Des Weiteren konnte gezeigt werden, dass die LSEC fähig sind, naive CD4+ T-Zellen in vitro Antigen-spezifisch zu aktivieren. Im Vergleich zu professionellen APZ war hierfür eine höhere Antigen-Dosis notwendig, die Expansion schwächer und es waren kaum Effektorzytokin-Produzenten detektierbar. Diese konnten jedoch durch Restimulierung mit professionellen APZ induziert werden (reversibler Phänotyp), was auf einen unreifen Differenzierungsstatus der T-Zellen schließen ließ. Es bleibt zu prüfen, in welchem Maße die Aktivierung naiver CD4+ T-Zellen durch LSEC in vivo stattfindet und diese durch LSEC aktivierten CD4+ T-Zellen funktionelle Bedeutung, z. B. regulatorische Kapazität, für das Immunsystem besitzen.<br>The liver plays a major role for the metabolism, but it is also of general importance for the immune system, e.g. for the deletion of activated T cells or the induction of peripheral tolerance. Under physiological conditions T cells and other leukocytes can be found in the liver, in the sinusoids as well as in the parenchyma. This hepatic accumulation of T cells might be due to immunosurveillance, but it would also be a prerequisite for modulation of T cells by hepatic cells. The present study investigated two different aspects of the interaction of liver sinusoidal endothelial cells (LSEC), the barrier between the sinusoidal lumen and the hepatic parenchyma, and CD4+ T cells. In the first part of the study it could be demonstrated that LSEC support the spontaneous transmigration of CD4+ T cells as well as their chemotaxis to CXCL12 and CXCL9 more efficiently than other endothelial cells. Whereas a direct endothelial activation by chemokines could be excluded the efficient chemokine presentation at the luminal LSEC surface (after abluminal uptake) might be responsible for the enhanced T cell transmigration. The findings suggest that LSEC might be involved in the recruitment of T cells by supporting a rapid transendothelial migration. The second part of the study focused on the characteristics of LSEC in the context of antigen presentation. LSEC were able to prime and expand naïve CD4+ T cells in vitro but less effective than professional APC as proven by weaker expansion of cells, a requirement for higher antigen concentration and the lack of cytokine producing T cells. The “immature effector” phenotype of the CD4+ T cells primed on LSEC was reversible since it could be overcome by restimulation on professional APC. In conclusion these data suggest that antigen presentation by LSEC results in activation but incomplete differentiation of CD4+ T cells.

Encapsulation d'adn dans des liposomes contenant du lactosylceramide dont le sucre terminal est reconnu specifiquement par des recepteurs presents sur la membrane plasmique des cellules visees, c. A. D. , les hepatocytes et les cellules endotheliales du foie et egalement les lymphocytes de la rate. Injection par voie intraveineuse des liposomes. Role de l'endocytose, dans leur internalisation. Modele genetique constitue du gene de la preproinsuline i de rat insere dans des vecteurs retroviraux permettant l'expression du gene dans des celules non insulogenes

碩士<br>國立陽明大學<br>臨床醫學研究所<br>102<br>Non-parenchymal cells including liver sinusoidal endothelial cell (LSECs), kupffer cell and stellate cell, especially LSECs, have been known to respond to the microenvironment and promote liver regeneration through mediating the secretion levels of cellular secretomes such as hepatocyte growth factor (HGF), Wnt2, transforming growth factors (TGFs), tumour necrosis factor (TNF-α) and interleukin 6 (IL-6). Mesenchymal stem cells (MSCs) not only have multi-differentiation potential but also contribute to tissue regeneration both in vivo and vitro by paracrine signaling and our previous results have shown that MSCs can rescue carbon tetrachloride (CCl4)-induced liver injuries in mice. In this study, we have demonstrated that expression levels of mitogenic genes, WNT2 and HGF in LSEC were significant increased after co-cultured with MSCs and MSC condition medium which are important in liver. In addition, we found that MSC/LSEC co-cultured condition medium can not only promote the proliferation rate of primary hepatocytes but also increase the productions of albumin compare to hepatocyte growth medium. Although the detail mechanism of how MSCs increase the mitogenic marker expressions in LSECs is not clear, our data suggested that MSC can be used as a potential cell-based therapeutic resource in cooperation with LSEC which promotes liver regeneration.

Chronic immune insult and immunosuppressant-related toxicities have remained an enduring challenge in organ transplantation. Long-term survival of transplant patients has improved marginally in recent decades due to these challenges. To circumvent these issues, transplant investigators have researched immune tolerance mechanisms that demonstrate potential to induce immunosuppression and rejection-free survival in the clinic. One mechanism in particular, the liver tolerance effect, has already demonstrated this experimentally and clinically. Liver transplants in experimental models and human patients have exhibited the ability to become spontaneously accepted without being rejected by the recipient’s immune system. Research in recent decades has revealed that the liver parenchymal and non-parenchymal cell populations harbor potent immunomodulatory properties. In the context of liver transplantation, it has been found that two cell populations in particular, the mesenchyme-derived liver sinusoidal endothelial cells and hepatic stellate cells, mediate the induction of liver transplant tolerance through a mechanism known as mesenchyme-mediated immune control.

碩士<br>國立臺灣大學<br>生化科技學系<br>101<br>Hemophilia A is the most common type of hemophilia and is also known as factor VIII (FVIII) deficiency. FVIII is an essential blood clotting factor, and its defects result in the formation of fibrin deficient clots, causing bleeding. liver is a major source of FVIII and the liver sinusoidal endothelial cells (LSEC) are endothelial cells (EC) that line the hepatic microvasculature, sinusoids, which are the major cell type of FVIII production. Previous reports showed that transplantation of LSEC improved the hemophilia phenotype of mice deficient for FVIII. Thus, LSEC is expected to be one of the source for cell therapy of Hemophilia A. My research can be divided into two parts, the generation of human induced pluripotent stem cell (hiPSC) and the differentiation toward LSEC, which can express FVIII, from hiPSC. In the first part, I infect c-Myc, Klf4, Oct4, and Sox2 to human somatic cell line HS68 by lentivirus for human iPSC formation. However, the results from analysis of cell surface antigen and mRNA expression showed that there is no complete pluripotent-specific expressed in cells. LSEC is a specified EC which emerged into liver bud at early development stage. In my second part of thesis, on the basis of differentiation from human embryonic stem cell (hESC) toward EC, associated the previous study: the inhibition of TGFβ/activin signaling could generate LESC-like from mouse ESC. I designed the protocol for inducing hiPSC toward LSEC through conditional medium with specific cytokines such as BMP4, ActivinA, bFGF, VEGF and TGFβ receptor inhibitor (Tbr1ki2) at different stages. By adopting 253G1 into embryonic body (EB) form, I used the protocol for LSEC induction.16 days later, about 55%-differentiated cells expressed a LESC marker, stabilin-2, a major hyaluronan clearance receptor, and incorporated FITC-labeled hyaluronan. Most importantly, those cells produced FVIII as evidenced by mRNA expression and protein secretion. In conclusion, FVIII-producing LSEC can be derived from human iPS cells and those cells may be an excellent source for cell therapy of hemophilia and the application of regeneration medicine.

Les cellules endothéliales (EC) constituent une première barrière physique à la dissémination de virus pléiotropiques circulant par voie hématogène mais leur contribution à la défense innée anti-virale est peu connue. Des dysfonctions des EC de la barrière hémato-encéphalique (BMEC) et des sinusoïdes hépatiques (LSEC) ont été rapportées dans des neuropathologies et des hépatites aiguës ou chroniques d’origine virale, suggérant que des atteintes à leur intégrité contribuent à la pathogenèse. Les sérotypes de coronavirus de l’hépatite murine (MHV), se différenciant par leur capacité à induire des hépatites et des maladies neurologiques de sévérité variable et/ou leur tropisme pour les EC, représentent des modèles viraux privilégiés pour déterminer les conséquences de l’infection des EC sur la pathogenèse virale. Lors d’infection par voie hématogène, le sérotype MHV3, le plus virulent des MHV, induit une hépatite fulminante, caractérisée par une réponse inflammatoire sévère, et des lésions neurologiques secondaires alors que le sérotype moins virulent, MHV-A59, induit une hépatite modérée sans atteintes secondaires du système nerveux central (SNC). Par ailleurs, le sérotype MHV3, à la différence du MHV-A59, démontre une capacité à stimuler la production de cytokines par la voie TLR2. Les variants atténués du MHV3, les virus 51.6-MHV3 et YAC-MHV3, sont caractérisés par un faible tropisme pour les LSEC et induisent respectivement une hépatite modérée et subclinique. Compte tenu de l’importance des LSEC dans le maintien de la tolérance hépatique et de l’élimination des pathogènes circulants, il a été postulé que la sévérité de l’hépatite et de la réponse inflammatoire lors d’infections par les MHV est associée à la réplication virale et à l’altération des propriétés tolérogéniques et vasculaires des LSEC. Les désordres inflammatoires hépatiques pourraient résulter d’une activation différentielle du TLR2, plutôt que des autres TLR et des hélicases, selon les sérotypes. D’autre part, compte tenu du rôle des BMEC dans la prévention des infections du SNC, il a été postulé que l’invasion cérébrale secondaire par les coronavirus est reliée à l’infection des BMEC et le bris subséquent de la barrière hémato-encéphalique (BHE). À l’aide d’infections in vivo et in vitro par les différents sérotypes MHV, chez des souris ou des cultures de BMEC et de LSEC, nous avons démontré, d’une part, que l’infection in vitro des LSEC par le sétotype MHV3, à la différence des variants 51.6- et YAC-MHV3, altérait la production du facteur vasodilatant NO et renversait leur phénotype tolérogénique en favorisant la production de cytokines et de chimiokines inflammatoires. Ces dysfonctions se traduisaient in vivo par une réponse inflammatoire incontrôlée et une dérégulation du recrutement intrahépatique de leucocytes, favorisant la réplication virale et les dommages hépatiques. Nous avons aussi démontré, à l’aide de souris TLR2 KO et de LSEC dont l’expression du TLR2 a été abrogée par des siRNA, que la sévérité de l’hépatite et de la réponse inflammatoire induite par le sérotype MHV3, dépendait en partie de l’induction et de l’activation préférentielle du TLR2 par le virus dans le foie. D’autre part, la sévérité de la réplication virale au foie et des désordres dans le recrutement leucocytaire intrahépatique induits par le MHV3, et non par le MHV-A59 et le 51.6-MHV3, corrélaient avec une invasion virale subséquente du SNC, au niveau de la BHE. Nous avons démontré que l’invasion cérébrale du MHV3 était associée à une infection productive des BMEC et l’altération subséquente des protéines de jonctions serrées occludine, VE-cadhérine et ZO-1 se traduisant par une augmentation de la perméabilité de la BHE et l’entrée consécutive du virus dans le cerveau. Dans l’ensemble, les résultats de cette étude mettent en lumière l’importance du maintien de l’intégrité structurale et fonctionnelle des LSEC et des BMEC lors d’infections virales aigües par des MHV afin de limiter les dommages hépatiques associés à l’induction d’une réponse inflammatoire exagérée et de prévenir le passage des virus au cerveau suite à une dissémination par voie hématogène. Ils révèlent en outre un nouveau rôle aggravant pour le TLR2 dans l’évolution de l’hépatite virale aigüe ouvrant la voie à de nouvelles avenues thérapeutiques visant à moduler l’activité inflammatoire du TLR2.<br>Endothelial cells (EC) act as a physical barrier against invasion by pleiotropic blood borne viruses but their contribution in innate antiviral defense is poorly known. Dysfunctions in blood-brain barrier EC (BMECs) and liver sinusoidal EC (LSECs) have been reported in viral neuropathologies and hepatitis, suggesting that loss of ECs integrity may contribute to the pathogenesis. Mouse hepatitis coronaviruses (MHV), differing in their ability to induce severe to subclinical hepatitis and neurological diseases and / or their tropism for ECs, are relevant viral models to study the consequences of EC infection in viral pathogenesis. Following hematogenous infection, the MHV3 serotype, the most virulent MHV, induces fulminant hepatitis, characterized by severe inflammatory response, followed by neurological damage whereas the less virulent MHV-A59 serotype induces milder hepatitis but does not invade the central nervous system (CNS). In addition, MHV3, in contrast to MHV-A59, shows ability to induce TLR2-dependent cytokine response. The attenuated MHV3 variants, 51.6-MHV3 and YAC-MHV3, are characterized by a weak tropism for LSECs and induce moderated and subclinical hepatitis respectively. Given the importance of LSECs in hepatic tolerance and the elimination of circulating pathogens, it has been postulated that the severity of hepatitis and inflammatory response induced by MHVs correlates with infection and alterations in vascular and tolerogenic properties of LSECs. Hepatic inflammatory disorders may result from differential activation of TLR2, rather than other TLRs and helicases, according to serotypes. Moreover, given the role of BMECs in preventing CNS infections, it has been postulated that secondary cerebral invasion by coronaviruses is related to infection of BMECs and subsequent breakdown of the blood-brain barrier (BBB). Through in vitro and in vivo infections of isolated BMECs, LSECs or mice with the different MHVs, we demonstrated, first, that in vitro productive infection of LSECs by the highly virulent MHV3 serotype, in contrast to 51.6- et YAC-MHV3 variants, altered their production of vasoactive factors and overthrew their intrinsic tolerogenic properties by promoting inflammatory cytokines and chemokines production. These disturbances were reflected in vivo by an uncontrolled inflammatory response and a deregulation of intrahepatic leukocyte recruitment, favoring viral replication and liver damages. We demonstrated, using TLR2 KO mice and LSECs treated with siRNA for TLR2 that the abnormal inflammatory response induced by MHV3 depended in part on preferential induction and activation of TLR2 by the virus on the surface of hepatic cells. Moreover, the severity of the primary viral replication in the liver and disorders in intrahepatic leucocyte recruitment induced by MHV3, but not by MHV-A59 and 51.6-MHV3, correlated with a subsequent brain invasion at the BBB level. Such invasion was related to productive infection of BMECs and subsequent IFN--dependent disruption of tight junction proteins occludin, VE-cadherin and ZO-1, resulting in an increase of BBB permeability and further viral entry into the CNS. Overall, the results of this study highlight the importance of structural and functional integrity of LSECs and BMECs during acute viral infections by MHVs to limit liver damages associated with viral-induced exacerbation of inflammatory response and prevent brain invasion by MHVs following viral spread through the bloodstream. They also reveal a new worsening role for TLR2 in the evolution of acute viral hepatitis paving the way for new therapies targeting TLR2-induced inflammatory activity.

Adhesion of malignant tumor cells under flow conditions to the endothelial monolayer lining the interior of the blood vessels is an important step in the metastatic cascade. This project examined the adhesive interactions of murine RAW117 large-cell lymphoma cells to murine hepatic sinusoidal endothelial cells (HSE). Flow cytometric analysis demonstrated constitutive-expression of VCAM-1, ICAM-1, PECAM-1 and beta1 integrin subunit on the surface of both HSE and RAW117 cells. Additionally, alpha4 and beta 7 integrin subunits and MAdCAM-1 were present on the RAW117 cell surface. The dynamic adhesion assay used employs a parallel-plate flow chamber coupled with video microscopy and digital image processing. It is capable of distinguishing initial attachment from adhesion stabilization. Using monoclonal antibodies to block the surface molecules mentioned above, we determined that an interaction of integrin alpha4beta1 on RAW117 cells with liver endothelial VCAM-1 occurs during the early stages of the adhesion process and may be important in liver metastasis. The second part of this project focused on the potential role of the cyclooxygenase (COX) pathway in the adhesive interactions of the cells studied. COX-1 isoform was detected in unstimulated RAW117 cells. Blocking it with COX inhibitors such as indomethacin, aspirin, sulindac and piroxicam resulted in increase in adhesion under static conditions. However, when lipoxygenase (LOX) inhibitors (such as ETI, NDGA and Esculetin) were used either alone or in combination with COX inhibitors, there was a significant decrease in adhesion. This observation suggested an involvement of the LOX pathway in the adhesion process. COX-2 isoform was detected in unstimulated HSE and was upregulated after stimulation with TNF-alpha. Blocking it in unstimulated HSE resulted in an increase in adhesion of RAW117. However, in the case of TNF-alpha stimulated HSE, NS-398 (a COX-2 specific inhibitor) caused a significant decrease in adhesion, indicating a potential role for COX-2 in the adhesion of RAW117 to stimulated HSE.

碩士<br>國立成功大學<br>微生物暨免疫學研究所<br>91<br>Dengue virus (DV) infection causes dengue fever or severe life-threatening dengue hemorrhagic fever/dengue shock syndrome (DHF/DSS). However, the pathogenesis of DHF/DSS is still not well understood. Previous studies in our laboratory showed that the AST and ALT, but not BUN, in mouse sera increased after either active immunization with DV nonstructural protein 1 (NS1) or passive administration with anti-DV NS1 antibodies (Abs). Furthermore, anti-DV NS1 Abs could bind to mouse vessel endothelium in inferior vena. In this study, we investigate the potential pathogenic role of anti-DV NS1 Abs in the damaged organs. Mice actively immunized with DV NS1 protein revealed the presence of anti-DV NS1 Abs on vessel endothelium in the liver. However, in the kidney, mice actively immunized with DV NS1 protein, JEV NS1 protein or PBS show similar binding patterns. Studies using organ sections from normal mice showed that anti-DV NS1 Abs, but not anti-JEV NS1 Abs, could directly bind to vessel endothelium in the liver sections. However, no specific cross-reaction of anti-DV NS1 Abs with mouse kidney vessel endothelium. Interestingly, gross and histological examinations revealed tissue damage with the presence of macrophage infiltration in the liver of NS1-immunized mice. We speculate that anti-DV NS1-induced endothelial cell dysfunction might result in liver tissue damage, at least in part, by infiltrated cells. Previous in vitro studies in our laboratory showed immune activation including cytokine and chemokine production in endothelial cells after treatment with anti-DV NS1 Abs. In this study, we demonstrated the involvement of upstream regulator NF-kB in immune activation by EMSA. After treatment with NF-kB inhibitor pyrrolidine dithiocarbamate, the expressions of IL-6、IL-8 and MCP-1 which could be stimulated by anti-DV NS1 Abs were inhibited. Taken together, endothelial cell damage and abnormal activation after stimulation by anti-DV NS1 Abs have been examined both in vivo and in vitro. These results imply the involvement of anti-DV NS1 Abs in the pathogenesis of DV infection.

碩士<br>高雄醫學大學<br>藥學研究所碩士在職專班<br>92<br>Purpose: Vascular endothelial growth factor(VEGF) and E-cadherin have been examined and taken for indicators of angiogenesis and metastasis respectively. In our study, we evaluated the in vitro effects of IFN-α treatment on the tumorigenicity of liver cancer cell lines. From the changes of the expression of VEGF and E-cadherin, we expected to evaluate the prognostic markers. Materials and Methods: Two well differentiated (Hep G2, Hep 3B) and one poor differentiated (SK-Hep I) human liver cancer lines were used. Cells were incubated with Interferon-α to evaluate their metastatic potential. Hep G2, SK-Hep-I cells were incubated with IFN-α for different concentrations (5000 unit/ml; 1000 unit/ml) and different periods(3;6 days) of time. Then, the VEGF protein expression was assayed by ELISA test in those culture media. The Hep 3B cells were treated with fresh medium, that contained 1000, 5000, 20000 unit/ml INF-α. It were incubated for 3 days. After that, E-cadherin expression changes was confirmed by Western blotting and Immunocytochemistry. All three cell lines, treated with INF-α, their survival rates were measured by cell counting and MTT test . Results: In the angiogenesis study, both in group A(5000 unit/ml; 3days)and group B(1000 unit/ml, 6days), the VEGF protein expression was down-regulated by IFN-α. In Group A, the VEGF protein expression were, in Hep G2 (no IFN-α:124.4, with IFN-α: 76.8),and in SK-Hep-I (no IFN-α: 44.5, with IFN-α: 30.0). In Group B, VEGF protein expression were, in Hep G2 (no IFN-α: 225.5, with IFN-α:151.4) and SK-Hep-I (no IFN-α: 74.1, with IFN-α: 53.4). In the cell invasion study, Western blotting and Immunocytochemistry showed that E-cadherin expression were up-regulated by the increasing concentration of IFN-α. The declined cell number counts and survival rates of all three cell lines showed antiproliferative activity of IFN-α. Conclusion: IFN-α confers its antitumor activity, at least in part, by its antiangiogenic activity, which results from decreased VEGF expression, and by anti-invasive activity, which results from elevated E-cadherin expression and antiproliferative activity on cancer cell lines. The results may help us to design the treatment protocol and to define the chemopreventive role of IFN-α. Key words：Interferon-α，Angiogenesis，Invasion，Metastasis，VEGF，E-cadherin

The mechanisms behind cardiovascular disease (CVD) initiation and progression are not fully elucidated. It is hypothesized that blood flow patterns regulate endothelial cell (EC) function to affect the progression of CVDs. A system that subjects ECs to physiologically-relevant shear stress waveforms within microfluidic devices has not yet been demonstrated, despite the advantages associated with the use of these devices. In this work, a bioreactor was designed to fulfill this need. Waveforms from regions commonly affected by CVDs including were derived. Pump motion and fluid flow profiles were validated by actuator motion tracking, particle image velocimetry, and flowmeters. While several relevant waveforms were successfully replicated, physiological waveforms could not be produced at physiological frequencies owing to actuator velocity and accuracy limitations, as well as dampening effects in the system. Overall, this work lays the foundation for designing a system that provides insight into the role of shear stress in CVD pathogenesis.

The orally active microtubule-disrupting agent (S)-1-ethyl-3-(2-methoxy-4-(5-methyl-4-((1-(pyridin-3-yl)butyl)amino)pyrimidin-2- yl)phenyl)urea (CYT997), reported previously by us (Bioorg Med Chem Lett 19:4639-4642, 2009; Mol Cancer Ther 8:3036-3045, 2009), is potently cytotoxic to a variety of cancer cell lines in vitro and shows antitumor activity in vivo. In addition to its cytotoxic activity, CYT997 possesses antivascular effects on tumor vasculature. To further characterize the vascular disrupting activity of CYT997 in terms of dose and temporal effects, we studied the activity of the compound on endothelial cells in vitro and on tumor blood flow in vivo by using a variety of techniques. In vitro, CYT997 is shown to potently inhibit the proliferation of vascular endothelial growth factor-stimulated human umbilical vein endothelial cells (IC(50) 3.7 +/- 1.8 nM) and cause significant morphological changes at 100 nM, including membrane blebbing. Using the method of corrosion casting visualized with scanning electron microscopy, a single dose of CYT997 (7.5 mg/kg i.p.) in a metastatic cancer model was shown to cause destruction of tumor microvasculature in metastatic lesions. Furthermore, repeat dosing of CYT997 at 10 mg/kg and above (intraperitoneally, b.i.d.) was shown to effectively inhibit development of liver metastases. The time and dose dependence of the antivascular effects were studied in a DLD-1 colon adenocarcinoma xenograft model using the fluorescent dye Hoechst 33342. CYT997 demonstrated rapid and dose-dependent vascular shutdown, which persists for more than 24 h after a single oral dose. Together, the data demonstrate that CYT997 possesses potent antivascular activity and support continuing development of this promising compound.