In conclusion, the present study successfully demonstrate BJ as a potent anti-pancreatic cancer herb; BD is the main ingredient for its cytotoxic and apoptotic effects on the pancreatic cancer cells through activation of the redox-sensitive p38-MAPK signaling pathway and reduction of anti-apoptotic activity by inhibition of NF-kappaB activation in pancreatic cancer cells. The in vivo efficacy and low toxicity of BD render this chemical compound to be a potential for its further development into an anti-pancreatic cancer agent. / In recent decades, the application of Chinese herbal medicine has become an increasingly popular approach and alternative to treating cancer. Moreover, Chinese herbal medicine is the source for the discovery of novel anti-cancer drugs. For example, irinotecan and topotecan, the analogues of camptothecin which is isolated from the bark and stem of Camptotheca acuminate are found to be effective in ovarian, lung and colon cancers. Given that Chinese medicine is commonly used in the treatment of cancers, we postulate that Chinese herbs are a valuable source to possess anti-pancreatic cancer compounds. Accordingly, the aims of the present project are: (1) to screen Chinese medicinal herbs which has the most potent cytotoxic activity in pancreatic cancer cells in vitro; (2) to isolate and identify the effective compound in Brucea javanica (BJ) which mediates apoptosis in pancreatic cancer cell lines; (3) to study the mechanistic pathways involved in brucein D - (BD, a quassinoid found in abundance in BJ) mediated apoptosis in pancreatic cancer in vitro; and (4) to evaluate the efficacy of BD in pancreatic cancer using an xenograft animal model of pancreatic cancer. / In vivo study demonstrated that daily administration of BD through intravenous injection for ten days in nude mice bearing pancreatic cancer cells effectively reduced tumor growth in terms of tumor weight and size, while showing no significant toxicity in heart, liver and kidney tissues of the mice. / Nine Chinese medicinal herbs were selected for the screening experiment and, among them, BJ exhibited the most potent cytotoxic action on the three pancreatic adenocarcinoma cell lines, namely PANC-1, SW-1990 and CAPAN-1, with IC50 values of 2.5mug/ml, 5.1mug/ml and 1.5mug/ml, respectively. BD, one of the main chemical compounds found in BJ was found to possess strong apoptogenic effect in PANC-1 cells, as evidenced by DNA condensation and fragmentation, sub-G1 phase formation, proteolytic activation of caspase 3, 8 and 9, and attenuation of bcl-2 activity. Further mechanistic studies revealed that the apoptotic signals generated by BD were transduced from the cell membrane to nucleus via the mediation of p38-MAPK signaling pathway while the reactive oxygen species (ROS) was found to accumulate in BD-treated PANC-1 cells. The activation p38-MAPK phosphorylation was inhibited by pretreatment with an antioxidant. However, the inhibition of NF-kappaB activity and downregulation of anitapoptotic genes in BD-treated cells was independent of the ROS changes. / Pancreatic cancer is the forth and sixth leading cause of cancer deaths in the United States and Hong Kong, respectively. The morbidity of pancreatic cancer is almost equal to its mortality rate. Poor diagnosis and intrinsic resistance to chemotherapy are the major characteristics for pancreatic cancer. Therefore, new therapeutic strategy is urgently warranted to overcome the drug-resistance challenge in the management of pancreatic cancer. / Lau Sin Ting Cynthia. / Source: Dissertation Abstracts International, Volume: 73-03, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2009. / Includes bibliographical references (leaves 228-271). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

Currently, the only cure for liver failure is orthotopic liver transplantation. However, there are insufficient donor organs available to treat every patient on the transplant list and many die before they are able to receive a liver transplant. The bioartificial liver (BAL) device is a potential extracorporeal treatment strategy utilising hepatocytes or hepatocyte-like cells (HLCs) within a bioreactor to recapitulate normal liver function and therefore ‘bridge’ a patient with liver failure until they receive a transplant. The work in this thesis utilised tissue engineering methods to develop novel approaches to BAL device design through development and characterisation of a polymer membrane scaffold (“PX”) for hollow fibre bioreactor (HFB) culture and a HLC source generated from the transdifferentiation of pancreatic AR42J-B13 (B13) cells. A flat sheet membrane model was used for the development of asymmetrical, hydrophobic polystyrene (PS) phase inversion membranes. Oxygen plasma significantly increased PS membrane surface wettability through addition of oxygen functional groups to create an environment conducive for cell culture. The treated membrane was henceforth referred to as “PX”. The culture medium HepatoZYME+ was investigated for its ability to induce transdifferentiation of B13 cells to HLCs and maintain the hepatic phenotype. Overall, HepatoZYME+-cultured cells experienced viability loss. A diluted version, “50:50”, showed induction of the hepatic markers carbamoylphosphate synthetase-1 (CPS-1) and HNF4α, as well as a change towards a HLC morphology. When using 50:50 as a maintenance medium, transdifferentiated HLCs retained loss of pancreatic amylase and also induction of hepatic markers, with comparable serum albumin secretion to the established Dex + OSM treatment. However, culture viability in 50:50 was still compromised. Therefore, HepatoZYME+ based media were deemed unsuitable for induction and maintenance compared to Dex-based protocols. PX flat sheet membranes were able to support culture of B13 cells and also the human osteosarcoma cell line, MG63, demonstrating improved cell attachment over non-surface treated PS membranes. PX membranes supported transdifferentiation of B13 cells to HLCs, presenting with loss of pancreatic amylase, induction of the hepatic markers transferrin, GS and CPS-1 and serum albumin secretion. Furthermore, PX showed no change in mass or loss of culture surface area over 15 days in culture conditions. Together, the novel membrane material and the media formulation and feeding regime developed have strong potential to be translated to a HFB setting and guide future BAL device design.

Type 1 diabetes is one of the most common chronic conditions in childhood and adolescence. Despite ongoing development of more physiological insulin preparations, recent advancements in insulin pump technology and more accurate blood glucose monitoring, in clinical practice it remains challenging to achieve normoglycaemia whilst reducing the risk of hypoglycaemia, particularly in young people with type 1 diabetes. Closed-loop insulin delivery (the artificial pancreas) is an emerging technology gradually progressing from bench to clinical practice. Closed-loop systems combine glucose sensing with computer-based algorithm informed insulin delivery to provide real-time glucose-responsive insulin administration. The key objective of my thesis is to evaluate the safety, efficacy and utility of closed-loop insulin delivery in children and adolescents with type 1 diabetes outside of the research facility setting. Results of five clinical trials are presented in the main chapters of this thesis. In a mechanistic study, the impact of glucose sensor operation duration on efficacy of overnight closed-loop was investigated comparing closed-loop performance on day 1 of sensor insertion to day 3 to 4 of sensor. Twelve adolescents with type 1 diabetes attended the research facility for two overnight visits. The sequence of the interventions was random. Despite differences in sensor accuracy, overnight CL glucose control informed by sensor glucose on day 1 or day 3-4 after sensor insertion was comparable. The model predictive controller appears to mitigate against sensor inaccuracies. In home settings, overnight closed-loop application was evaluated over three months in 25 children and adolescents with type 1 diabetes aged six to 18 years. The study was conducted at three centres in the UK and adopted a randomised cross-over design. Compared to sensor-augmented pump therapy, overnight home use of closed-loop increased the proportion of time sensor glucose was in target, and reduced mean glucose and hypoglycaemia. Two randomised crossover studies evaluated the safety and efficacy of day-and-night hybrid closed-loop insulin delivery in young people with type 1 diabetes aged 10 to 18 years over seven days, and 21 days, respectively. A total of 24 subjects were enrolled in this single centre trial. Free-living home use of day-and-night closed-loop in suboptimally controlled adolescents with type 1 diabetes was safe, and improved glucose control without increasing the risk of hypoglycaemia. Finally, closed-loop technology was assessed in five very young children (aged one to seven years) with type 1 diabetes in a two-period, crossover study. Closed-loop was used during both 3-week intervention periods, either with standard strength insulin (U100), or with diluted insulin (U20). The order of intervention was random. Free-living home use of day-and-night hybrid closed-loop in very young children with type 1 diabetes was feasible and safe. Glucose control was comparable during both intervention periods. Thus, use of diluted insulin during closed-loop insulin delivery might not be of additional benefit in this population. In conclusion, studies conducted as part of my thesis demonstrate that use of hybrid closed-loop insulin delivery systems in children and adolescents aged one to 18 years in free daily living without remote monitoring or supervision is feasible, safe and effective. My work supports the progression of this technology from research to mainstream clinical practice.

Ce travail scientifique a abordé la problématique du vieillissement des îlots pancréatiques et l’effet de la senescence endothéliale et des microparticules (MPs) sur la fonction des îlots. Nous avons exploré l’impact du vieillissement du pancréas sur la morphologie, le devenir et la fonction de l’îlot pancréatique par analyse comparative entre pancréas de rats jeunes et d’âge moyen et le rôle des MPs endothéliales pro-sénescentes sur la fonction des îlots et leur sénescence prématurée. Nos résultats in vivo montrent que le pancréas est un organe précocement sensible au stress oxydant s’accumulant avec l’âge. Il conduit à la surexpression des marqueurs procoagulants et de senescence sans apparition d’apoptose. In vitro, les MPs de cellules endothéliales sénescentes ont un effet pro-sénescent sur les îlots pancréatiques isolés de rats jeunes avec une activité SA-β-galactosidase caractéristique, la surexpression des marqueurs p53, p21 et p16 et la réduction de la capacité de la sécrétion d’insuline en réponse au glucose. L’ensemble de nos résultats in vivo et in vitro désigne la contribution de la sénescence endothéliale comme une cause probable à la dysfonction de greffon. / This scientific work has tackled the question of the pancreatic islets aging and the effect of endothelial senescence and microparticles (MPs) on islet function. We investigated the impact of aging on pancreas morphology, fate and on the function of the pancreatic islet by comparative analysis between pancreas in young and middle-aged rats, as well as the role of pro-senescent endothelial MPs on islet function and their premature senescence. Our in vivo data show that the pancreas is an early sensitive organ to oxidative stress accumulating with age and leading to overexpression of the procoagulant and senescence markers without appearance of apoptosis. In vitro, MPs of senescent endothelial cells have a pro-senescent effect on pancreatic islets isolated from young rats with characteristic SA-β-galactosidase activity, overexpression of p53, p21 and p16 markers and reducing the ability of insulin secretion in response to glucose. Altogether, our in vivo and in vitro data indicate the contribution of endothelial senescence as a possible contributor to graft dysfunction.

胰腺癌是一種死亡率極高的癌症, 據統計在所有的種族和性別中能達到五年存活的胰腺癌患者僅有5.5%。在現有醫學的治療方法中，除了手術切除之外，化學療法依然是主要的應對之策。接受胰腺癌切除術後的無瘤患者的生存期中位數和以現代一線化療藥吉西他濱治療的胰腺癌患者的生存期分別為13.4和6.9月。因此, 臨床上迫切需求更有效治療胰腺癌的新藥物。 / 我們從中藥鴉膽子中分離出了10種不同的化學單體。經過在人類胰腺癌細胞Capan-2上進行的細胞毒性篩選後, 發現Brucein D （BD）擁有最強的胰腺癌細胞毒性作用。我們目前的體外與體內實驗的目標是對BD可能具備的抗胰腺癌活性進行深入評估, 並進一步揭示其作用機理。 / 體外實驗研究表明BD可以極大程度上抑制Capan-2細胞生長, 同時對於人類肝細胞WRL68和人類胰腺幹細胞PPC僅存在很輕微的毒性作用。BD的抑制細胞生長作用和喜樹堿相當, 但顯著強於一線抗胰腺癌藥吉西他濱作。實驗中我們發現在BD作用的Capan-2細胞的線粒體膜電位被減弱, 其減弱程度與BD的濃度存在一定的劑量依賴性。另外, 被BD處理後的Capan-2細胞中的Bcl-2表達減弱, 與此同時capase 9和caspase 3的表達呈顯著性加強。除此之外, BD可以導致基因DNA破碎, 增加Capan-2細胞處於細胞凋亡期的數量, 而且處於凋亡期細胞的數量與BD存在劑量依賴性。 / 我們建立起原位型胰腺癌裸鼠模型並利用其進行體內實驗研究。研究結果顯示, BD治療組裸鼠的存活率遠遠大於吉西他濱治療組。此外, 與磷酸鹽緩衝鹽水注射組比較, BD治療組可以極大程度的減輕腫瘤的重量和減小腫瘤的體積。與此同時, 血液生化分析結果表明BD可以明顯降低CA19-9在血液中的表達。螢光免疫檢驗法結果揭示BD能夠調低CA19-9和Ki-67在胰腺腫瘤組織中的表達。蛋白質印記分析的結果也顯示BD治療後可以增強胰腺腫瘤組織中caspase 3, 8, 9的表達, 而減弱IKKα和NF-κB p65的表達。另外, 通過ELISA分析後顯示, BD治療明顯降低了NF-κB p65在細胞質與細胞核中的表達, 其表達程度與BD的濃度成反比。 / 綜上所述, 我們目前的體外和體內研究表明, BD作為一種存在于天然中藥中的化學單體具有很好的抗胰腺癌的潛質, 值得進一步研究和開發, 使之成為臨床治療胰腺癌的一種安全有效的新藥物。 / Pancreatic adenocarcinoma has a high morbidity and mortality rate in cancers as it possesses only 5.5% of 5-year survival rate for all races and both sexes. The median disease-free survival following complete resection of the pancreatic tumor and adjuvant chemotherapy with the first-line chemotherapeutic agent gemcitabine is 13.4 and 6.9 months, respectively. There issued an urgent need for alternative effective agents to producing a better clinical outcome for the management of this deadly disease. / Previous studies in our research group have shown that the fruit of Brucea javanica L. exhibited potent anti-pancreatic cancer activity. In the current project, ten chemical compounds were isolated from this Chinese herb and screened for their cytotoxicity against cultured Capan-2 cells, a human pancreatic adenocarcinoma cell line. Among these compounds, Brucein D (BD) exhibited the most potent cytotoxic activity. Further in vitro and in vivo studies were conducted to evaluate the potential anti-pancreatic cancer activity of BD and elucidate its underlying mechanisms of action. / In the In vitro study, BD was found to significantly inhibit the growth of Capan-2 cells, while exerting only modest cytotoxicity on human hepatocyte WRL68 cells and human pancreatic progenitor PPC cells. The anti-proliferative effects of BD were comparable to those exhibited by camptothecin and gemcitabine. We found a dose-dependent decrease of the mitochondrial membrane potential in BD-treated Capan-2 cells. In addition, BD exposure was able to attenuate the expression of Bcl-2 and significantly accentuate the expression of both caspase 9 and caspase 3. Moreover, BD was capable of inducing the fragmentation of genomic DNA while increasing the percentage of Capan-2 cells in the apoptotic phase and the quantity of apoptosis cells was observed in a dose- dependent manner. / A mouse model of orthotopic pancreatic cancer was established for the in vivo experiments. The results demonstrated that the BD-treated groups had a higher survival rate than that the gemcitabine-treated groups. Moreover, it was found that BD treatments significantly reduced the tumor weight and volume when compared with those of PBS injected group. Meanwhile, blood biochemistry analyses showed that BD significantly decreased the expression of CA19-9 (a tumor mark). Immunofluorescence study also revealed that BD could down-regulate the expression of both CA19-9 and Ki-67 in pancreatic tumor tissues. Furthermore, Western blot analysis showed that BD treatments could accentuate the expression of caspases 3, 8, 9 and decreased the expression of IKKα and NF-κB p65 in total. Moreover, BD attenuated the expression of NF-κB p65 in both cytoplasmic and nuclear factions of the tumor tissues as detected by ELISA kit, and the expression rate was inversely proportional to the doses of BD used. / Taken these data together, our in vitro and in vivo studies have successfully demonstrated that BD, a naturally occurring chemical compound from Fructus Bruceae, is a promising anti-pancreatic cancer agent worthy of further development into pharmaceutical agent for pancreatic cancer. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Liu, Ling. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2013. / Includes bibliographical references (leaves 217-253). / Abstracts also in Chinese. / ABSTRACT --- p.I / 摘要 --- p.IV / PUBLICATIONS --- p.VI / ACKNOWLEDGEMENTS --- p.X / TABLE OF CONTENTS --- p.XI / LIST OF FIGURES --- p.XIX / LIST OF TABLES --- p.XXIII / LIST OF ABBREVIATIONS --- p.XXV / Chapter CHAPTER 1 --- GENERAL INTRODUCTION / Chapter 1.1 --- Pancreas --- p.2 / Chapter 1.1.1 --- Gross anatomy --- p.2 / Chapter 1.1.2 --- Microscopic anatomy --- p.5 / Chapter 1.1.2.1 --- Acini cells --- p.7 / Chapter 1.1.2.2 --- Duct cells --- p.7 / Chapter 1.1.2.3 --- Stroma --- p.14 / Chapter 1.1.2.4 --- Islets cells --- p.15 / Chapter 1.1.3 --- Pancreatic diseases --- p.16 / Chapter 1.2 --- Pancreatic Cancer --- p.30 / Chapter 1.2.1 --- Epidemiology --- p.30 / Chapter 1.2.2 --- Risk factors --- p.32 / Chapter 1.2.3 --- Clinical symptoms, diagnosis and staging --- p.34 / Chapter 1.2.4 --- Types of pancreas tumor --- p.42 / Chapter 1.3 --- Treatment of Pancreatic Cancer --- p.47 / Chapter 1.3.1 --- Treatment for localized disease --- p.49 / Chapter 1.3.2 --- Treatment for locally advanced disease --- p.50 / Chapter 1.3.3 --- Treatment for metastatic disease --- p.52 / Chapter 1.4 --- Molecular Targets for Pancreatic Cancer Therapy --- p.56 / Chapter 1.4.1 --- Mechanisms of apoptosis --- p.56 / Chapter 1.4.2 --- Roles of mitochondrial pathway in apoptosis --- p.58 / Chapter 1.4.3 --- NF-κB activation on cancers --- p.59 / Chapter 1.4.4 --- CA19-9 as a therapeutic target for Pancreatic Cancer --- p.62 / Chapter 1.4.5 --- Ki-67 is associated with for cellular proliferation --- p.64 / Chapter 1.5 --- Applications of Chinese Medicine in Cancer Treatment --- p.66 / Chapter 1.5.1 --- Background of traditional Chinese medicine --- p.66 / Chapter 1.5.2 --- Chinese medicine herbs commonly used for cancer treatment --- p.67 / Chapter 1.6 --- Mouse Models of Pancreatic Cancer --- p.75 / Chapter 1.6.1 --- Anatomy of pancreas in mouse --- p.75 / Chapter 1.6.2 --- Pancreatic cancer models --- p.77 / Chapter 1.7 --- Hypothesis and Objectives of the Study --- p.83 / Chapter CHAPTER 2 --- ANTI-PANCREATIC CANCER EFFECTS OF TEN CHEMICAL COMPOUNDS DERIVED FROM FRUCTUS BRUCEAE ON CULTURED CAPAN-2 CELLS / Chapter 2.1 --- Introduction --- p.86 / Chapter 2.1.1 --- Brucea javanica L. Merr --- p.86 / Chapter 2.1.2 --- The fruit of Brucea javanica --- p.88 / Chapter 2.1.3 --- Used of Fructus Bruceae to treat cancers by Chinese medicine practitioners --- p.90 / Chapter 2.1.4 --- Biological activities of some chemical compounds from Brucea javanica --- p.90 / Chapter 2.1.5 --- Chemical structure of ten compounds isolated from Fructus Bruceae --- p.92 / Chapter 2.2 --- Materials and Methods --- p.96 / Chapter 2.2.1 --- Plant material --- p.96 / Chapter 2.2.2 --- Extratcion, fractionation, isolate and characterization --- p.96 / Chapter 2.2.3 --- General procedures on structural elucidation and phytochemical work --- p.100 / Chapter 2.2.4 --- Preparation of solutions of tern chemical compounds derived from Fructus Bruceae --- p.101 / Chapter 2.2.5 --- General cell culture methods --- p.101 / Chapter 2.2.6 --- Selection of appropriate seeding density of Capan-2 cells --- p.102 / Chapter 2.2.7 --- Cytotoxicity evaluation by SRB assay --- p.102 / Chapter 2.2.8 --- Statistical analyses --- p.103 / Chapter 2.3 --- Results --- p.105 / Chapter 2.3.1 --- Seletion of appropriate seeding density of Capan-2 cells --- p.105 / Chapter 2.3.2 --- IC₅₀ values of ten tested compounds and chemical structures --- p.107 / Chapter 2.4 --- Discussion --- p.111 / Chapter CHAPTER 3 --- INVOLVEMENT OF THE MITOCHONDRIAL PATHWAY IN BRUCEIN D-INDUCED APOPTOSIS IN CAPAN-2 CELLS / Chapter 3.1 --- Introduction --- p.116 / Chapter 3.2 --- Materials and Methods --- p.117 / Chapter 3.2.1 --- General cell culture --- p.117 / Chapter 3.2.2 --- Cytotoxicity assay --- p.119 / Chapter 3.2.3 --- Proliferation assay --- p.120 / Chapter 3.2.4 --- Hoechest fluorescence staining for morphological evaluation --- p.121 / Chapter 3.2.5 --- Cell cycle analysis by flow cytometry --- p.122 / Chapter 3.2.6 --- Quantitative analysis of apoptosis by Annexin V-PI staining assay --- p.122 / Chapter 3.2.7 --- Estimation of the changes of MMP on BD-treated Capan-2 cells --- p.123 / Chapter 3.2.8 --- Western blot analysis --- p.124 / Chapter 3.2.9 --- Statistical analyses --- p.125 / Chapter 3.3 --- Results --- p.126 / Chapter 3.3.1 --- BD significantly inhibited the proliferation of Capan-2 cells --- p.126 / Chapter 3.3.2 --- BD was less cytotoxic on cultured WRL68 and PPC cells than that of controls --- p.128 / Chapter 3.3.3 --- BD induced DNA condensation in Capan-2 cells --- p.131 / Chapter 3.3.4 --- BD induced an increase in the percentage of subG1 phase (apoptotic cells) --- p.133 / Chapter 3.3.5 --- BD dose-dependently induced cellular apoptosis to Capan-2 cells --- p.136 / Chapter 3.3.6 --- The MMP of Capan-2 cells were significantly attenuated by BD treatment --- p.139 / Chapter 3.3.7 --- BD increased the expression of apoptotic caspases in Capan-2 cells --- p.142 / Chapter 3.4 --- Discussion --- p.144 / Chapter CHAPTER 4 --- BRUCEIN D SUPPRESSES PANCREATIC TUMOR GROWTH IN AN ORTHOTOPIC MOUSE MODEL THROUGH THE CASPASE 3, 8, 9 AND NF-κB PATHWAYS / Chapter 4.1 --- Introduction --- p.150 / Chapter 4.2 --- Materials and Methods --- p.153 / Chapter 4.2.1 --- Ethics statement and animal holdings --- p.153 / Chapter 4.2.2 --- Cell culture --- p.153 / Chapter 4.2.3 --- Establishment of an orthotopic pancreatic cancer mouse model --- p.154 / Chapter 4.2.4 --- Treatment of orthotopic pancreatic cancer mice with BD --- p.155 / Chapter 4.2.5 --- Necropsy procedure and histological studies --- p.156 / Chapter 4.2.6 --- Hematoxylin-eosin staining --- p.156 / Chapter 4.2.7 --- Determination of CA19-9 and Ki-67 by immunofluorescence staining --- p.160 / Chapter 4.2.8 --- CA 19-9 expression in blood --- p.161 / Chapter 4.2.9 --- Western blot analysis of Caspase 3,8,9, IKKα and NF-κB p65 --- p.162 / Chapter 4.2.10 --- Extraction of the nucleus and cytoplasm from pancreatic tumor tissues --- p.163 / Chapter 4.2.11 --- Detection of the expression of NF-κB p65 in both cytoplasm and nuclear parts of pancreatic cancer cells --- p.165 / Chapter 4.2.12 --- Statistical analyses --- p.166 / Chapter 4.3 --- Results --- p.167 / Chapter 4.3.1 --- BD treatment enhanced the survival rate of tumor-bearing mice and significantly attenuated the tumor weight and volume --- p.167 / Chapter 4.3.2 --- Histological evaluation of the pancreas and pancreatic tumor after BD treatment --- p.175 / Chapter 4.3.3 --- BD significantly decreased the expression of CA19-9 in the blood samples of the experimental mice --- p.178 / Chapter 4.3.4 --- BD down regulated the expression of CA19-9 in pancreatic tumor tissues --- p.180 / Chapter 4.3.5 --- BD down regulated the expression of Ki-67 in pancreatic tumor tissues --- p.183 / Chapter 4.3.6 --- BD accentuated the expression of Caspase3, 8, 9 and decreased the expression of NF-κB p65 --- p.186 / Chapter 4.3.7 --- BD decreased the expression of NF-κB p65 in both cytoplasm and nucleus of pancreatic tumor cells --- p.189 / Chapter 4.4 --- Discussion --- p.191 / Chapter CHAPTER 5 --- GENERAL DISCUSSION, CONCLUSIONS AND FUTURE STUDIES / Chapter 5.1 --- General Discussion --- p.200 / Chapter 5.2 --- General Conclusions --- p.209 / Chapter 5.3 --- Limitation of Study --- p.211 / Chapter 5.4 --- Clinical Significance of Study Results --- p.212 / Chapter 5.5 --- Future Studies --- p.214 / Chapter 5.5.1 --- Investigation of the possible synergistic effect of combination of BD with gemcitabine on orthotopic pancreatic cancer mouse model --- p.214 / Chapter 5.5.2 --- Testing BD on different animal models --- p.215 / REFERENCES / References by Alphabetical Order --- p.217

微小RNA是内源性小非编码RNA，在肿瘤生成中扮演重要角色。Brucein D(BD)是一种B. javanica果实提取物，已被报道在胰腺癌中具有抗肿瘤作用。在此研究中，我们证明了BD在体内和体外均可抑制肝癌细胞生长。为了研究BD是否通过调节微小RNA来执行其抗肿瘤功能，我们进行了一个肿瘤微小RNA定量PCR阵列谱分析。此阵列包括95个已被报道与肿瘤有关的微小RNA。通过对比BD处理前后微小RNA谱的变化，我们发现微小RNA-95在BD处理后被显著下调了。其后促凋亡的CUGBP2被确定为微小RNA-95的下游靶基因。 / 胰腺癌是一种预后很差的恶性肿瘤，常常在确诊时已发生转移。为了找出在胰腺癌转移过程中发挥决定性作用的基因，我们进行了全基因组范围的RNA干扰筛选。一个包含针对全部人类基因的shRNA文库被导入胰腺癌细胞系capan-2.然后将这些细胞移植到裸鼠的胰腺中来建立一个原位胰腺癌小鼠模型。我们的假设是下调某个基因会促使低转移潜力的capan-2细胞转移到肝脏。通过从肝转移结节中回收shRNA模板，我们找到了几个推定的转移抑制基因。其中之一，SOX9，通过体内实验验证，证明下调SOX9基因的表达可促进胰腺癌转移。 / 化疗适用于进展期胰腺癌病人。然而他们对一线化疗药吉西他滨的反应并不乐观，这进一步使胰腺癌的预后变差。我们展开了一个全基因组范围的RNA干扰筛选来确定一些在化疗耐药过程中起关键作用的基因。携带上述shRNA文库的capan-2细胞被用于吉西他滨药物处理之下的筛选。通过微阵列分析，一些基因被筛选成为可影响癌细胞对药物敏感性的潜在的靶基因。通过进一步验证，LLGL1基因被确定为在调节癌细胞对化疗敏感性过程中起重要作用的基因。 / MicroRNAs (miRNAs) are endogenous small non-coding RNAs that have been shown to play important roles in tumorigenesis. Brucein D (BD), a chemical compound isolated from Brucea javanica fruit, has previously been reported to have anti-cancer effect in pancreatic cancer. In this study, we showed that BD also inhibited the growth of liver cancer cells both in vitro and in vivo. To investigate whether BD exerts its anti-cancer effect through regulation of miRNAs, we performed a cancer miRNA qPCR array profiling. From the profiling, miR-95 was found to be significantly down-regulated after BD treatment. Subsequently, a pro-apoptotic gene CUGBP2 was identified as a direct downstream target of miR-95. These findings suggested BD suppressed liver cancer cell growth through down-regulation of miR-95 and reinforcing CUGBP2. / Pancreatic cancer is an aggressive malignancy with extremely poor prognosis. It is usually diagnosed when metastases are already present. To identify genes that play critical roles in the processes of pancreatic cancer metastasis, a whole genome RNAi screening was performed. An shRNA library targeting all human genes was introduced into a human pancreatic cancer cell line capan-2. The infected cells were then transplanted into the pancreas of nude mice. Because capan-2 is of low metastatic potential, we hypothesized that knocking down of metastasis suppressor genes would facilitate capan-2 cells to spread to the liver. By retrieving shRNA templates from the liver metastatic nodules, several candidate genes were found. One of them, SOX9, has been validated as metastasis suppressor gene in vivo, implying that loss of expression of SOX9 promotes pancreatic cancer metastasis. / Chemotherapy is recommended for patients of pancreatic cancer in advanced stage. However, their response to the first-line chemotherapy drug gemcitabine is not satisfactory. A genome-wide RNAi screening was conducted to identify genes that were critical in chemotherapy resistance. Capan-2 cells containing the above shRNA library were applied for the screening under gemcitabine treatment. Through microarray analysis, a number of genes were screened as potential gemcitabine sensitivity genes. Validation experiments implied that the gene LLGL1 may play an important role in modulating pancreatic cancer cells’ sensitivity to gemcitabine. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Xia, Tian. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 125-134). / Abstracts also in Chinese. / Chapter Abstract --- p.I / Chapter Acknowledgements --- p.V / Chapter Abbreviations --- p.VI / Chapter List of Figures --- p.XV / Chapter List of Tables --- p.XVI / Chapter Part I: --- Brucein D-modulated microRNA-95 expression inhibits hepatocellular carcinoma cell growth --- p.1 / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Hepatocellular carcinoma --- p.1 / Chapter 1.1.1 --- Definition and classification --- p.1 / Chapter 1.1.2 --- Epidemiology --- p.1 / Chapter 1.1.3 --- Etiology --- p.3 / Chapter 1.1.4 --- Molecular pathogenesis of HCC --- p.4 / Chapter 1.1.4.1 --- Genomic instability --- p.4 / Chapter 1.1.4.2 --- Deregulation of key signaling pathways --- p.5 / Chapter 1.1.4.3 --- Epigenetic changes of HCC --- p.6 / Chapter 1.1.4.4 --- Two models of HCC pathogenesis --- p.7 / Chapter 1.1.5 --- Therapeutic methods and prognosis of HCC --- p.8 / Chapter 1.2 --- Apoptosis --- p.9 / Chapter 1.2.1 --- Types of cell death --- p.9 / Chapter 1.2.2 --- Apoptosis --- p.10 / Chapter 1.2.3 --- Morphological features of apoptosis --- p.10 / Chapter 1.2.4 --- Molecular mechanisms of apoptosis --- p.11 / Chapter 1.2.5 --- Apoptosis and cancer --- p.13 / Chapter 1.2.5.1 --- Imbalance of pro-apoptotic proteins and anti-apoptotic proteins --- p.13 / Chapter 1.2.5.2 --- Impaired caspases activity --- p.14 / Chapter 1.2.5.3 --- Deregulated death receptor signaling --- p.15 / Chapter 1.2.6 --- Cancer therapy targeting apoptotic defects --- p.15 / Chapter 1.3 --- microRNA --- p.16 / Chapter 1.3.1 --- Overview --- p.16 / Chapter 1.3.2 --- Biogenesis and maturation of microRNA --- p.17 / Chapter 1.3.3 --- Gene silencing by microRNA --- p.18 / Chapter 1.3.4 --- MicroRNA and cancers --- p.19 / Chapter 1.3.5 --- MicroRNA’s involvement in HCC --- p.21 / Chapter 1.3.6 --- Involvement of miR-95 in cancer --- p.22 / Chapter 1.4 --- Brucein D --- p.22 / Chapter 1.5 --- Aims of study --- p.23 / Chapter 2 --- Materials and Methods --- p.25 / Chapter 2.1 --- Cell Culture --- p.25 / Chapter 2.1.1 --- Mammalian Cell Culture --- p.25 / Chapter 2.1.2 --- Preparation of cell stock --- p.25 / Chapter 2.1.3 --- Cell recovery from liquid nitrogen stock --- p.26 / Chapter 2.1.4 --- Preparation of drugs for treatments --- p.26 / Chapter 2.1.5 --- Drug treatment --- p.26 / Chapter 2.1.6 --- Transfection of siRNA --- p.27 / Chapter 2.1.7 --- MTT Assay --- p.28 / Chapter 2.1.8 --- Luciferase reporter assays --- p.28 / Chapter 2.1.9 --- Annexin V Assay --- p.29 / Chapter 2.2 --- In vivo mouse model --- p.29 / Chapter 2.3 --- Tunel Assay (Terminal deoxynucleotide transferase dUTP Nick End Labeling Assay) --- p.30 / Chapter 2.4 --- RNA manipulation --- p.31 / Chapter 2.4.1 --- RNA Isolation --- p.31 / Chapter 2.4.2 --- Synthesis of cDNA from miRNA --- p.32 / Chapter 2.4.3 --- Synthesis of cDNA from RNA and quantitative PCR --- p.33 / Chapter 2.4.4 --- miRNA qPCR array --- p.34 / Chapter 2.5 --- DNA manipulation --- p.34 / Chapter 2.5.1 --- Agarose gel electrophoresis and purification of DNA --- p.34 / Chapter 2.5.2 --- Restriction enzymes digestion --- p.35 / Chapter 2.5.3 --- Ligation of DNA fragments --- p.36 / Chapter 2.5.4 --- Polymerase chain reaction --- p.36 / Chapter 2.5.5 --- Preparation of competent E. coli cells --- p.37 / Chapter 2.5.6 --- Transformation of E. coli cells --- p.37 / Chapter 2.5.7 --- Small scale plasmid isolation from E. coli (mini-prep) --- p.38 / Chapter 3 --- Results --- p.39 / Chapter 3.1 --- Brucein D inhibited the growth of HCC cells both in vitro and in vivo --- p.39 / Chapter 3.2 --- BD induced apoptosis in HCC cells --- p.43 / Chapter 3.3 --- miR-95 is an target of BD to modulate cell growth --- p.46 / Chapter 3.4 --- Identification of CUGBP2 as a downstream target of miR-95 --- p.55 / Chapter 4 --- Discussion --- p.60 / Chapter Part II: --- Genome-wide RNAi screening identifies tumor metastasis suppressor genes and drug sensitivity genes in pancreatic cancer --- p.65 / Chapter 1 --- Introduction --- p.65 / Chapter 1.1 --- Pancreatic cancer --- p.65 / Chapter 1.1.1 --- Overview --- p.65 / Chapter 1.1.2 --- Pancreatic ductal adenocarcinoma (PDAC) --- p.67 / Chapter 1.1.3 --- Molecular basis of PDAC --- p.67 / Chapter 1.1.3.1 --- KRAS --- p.67 / Chapter 1.1.3.2 --- TP53 --- p.68 / Chapter 1.1.3.3 --- CDKN2A --- p.69 / Chapter 1.1.4 --- Gemcitabine treatment in PDAC --- p.69 / Chapter 1.2 --- Metastasis --- p.71 / Chapter 1.2.1 --- Overview --- p.71 / Chapter 1.2.2 --- The stepwise process of metastasis --- p.72 / Chapter 1.2.3 --- Metastasis of pancreatic cancer --- p.74 / Chapter 1.3 --- SOX9 --- p.75 / Chapter 1.4 --- Aims of study --- p.77 / Chapter 2 --- Materials and Method --- p.78 / Chapter 2.1 --- Cell culture --- p.78 / Chapter 2.1.1 --- Mammalian Cell Culture --- p.78 / Chapter 2.1.2 --- MTT Assay --- p.78 / Chapter 2.1.3 --- Colony formation assay --- p.79 / Chapter 2.1.4 --- Wound healing assay --- p.79 / Chapter 2.1.5 --- Transwell migration chamber assay --- p.80 / Chapter 2.1.6 --- Immunocytochemistry --- p.80 / Chapter 2.1.7 --- Transient transfection of siRNA --- p.81 / Chapter 2.2 --- Establishment of in-vivo and in-vitro models --- p.82 / Chapter 2.2.1 --- shRNA library introduction --- p.82 / Chapter 2.2.2 --- Establishment of the orthotopic pancreatic cancer mouse model --- p.82 / Chapter 2.2.3 --- Package of lentivirus expressing shRNA --- p.83 / Chapter 2.2.4 --- Generation of stable cell line expressing shRNA --- p.84 / Chapter 2.3 --- DNA manipulation --- p.84 / Chapter 2.3.1 --- Large scale plasmid isolation from E. coli (maxi-prep) --- p.84 / Chapter 2.4 --- Analysis of Protein --- p.85 / Chapter 2.4.1 --- Preparation of protein cell lysates --- p.85 / Chapter 2.4.2 --- Protein concentration determination --- p.86 / Chapter 2.4.3 --- SDS-PAGE --- p.86 / Chapter 2.4.4 --- Immunoblotting (Western blotting) --- p.87 / Chapter 2.5 --- RNA manipulations --- p.88 / Chapter 2.5.1 --- RNA Isolation --- p.88 / Chapter 2.5.2 --- Synthesis of cDNA from RNA and quantitative PCR --- p.89 / Chapter 2.6 --- Analysis of Clinical Samples --- p.90 / Chapter 2.6.1 --- Clinical specimens --- p.90 / Chapter 2.6.2 --- Immunohistochemistry --- p.90 / Chapter 3 --- Results --- p.92 / Chapter 3.1 --- Genome-wide RNAi screening identifies genes as metastasis suppressors in an orthotopic pancreatic cancer mouse model --- p.92 / Chapter 3.2 --- SOX9 is a metastasis suppressor gene in pancreatic cancer --- p.97 / Chapter 3.3 --- Investigation into cellular functions of SOX9 --- p.102 / Chapter 3.3.1 --- SOX9’s effect on cell growth --- p.102 / Chapter 3.3.2 --- SOX9’s effect on cell migration --- p.105 / Chapter 3.4 --- Implication of SOX9 in human pancreatic cancer samples --- p.109 / Chapter 3.5 --- Genome-wide RNAi screening for the identification of gemcitabine sensitivity genes --- p.113 / Chapter 4 --- Discussion --- p.120 / Chapter General conclusions --- p.125

INTRODUÇÃO: O transplante de pâncreas é a opção definitiva para a manutenção do estado normoglicêmico permanente nos portadores de diabetes mellitus tipo 1. O meio de contraste de ultra-som é uma metodologia capaz de avaliar a perfusão tecidual, mas não há um estudo para a avaliação do padrão de perfusão do pâncreas transplantado normal e patológico. A importância da avaliação da perfusão do enxerto e a aplicação do meio de contraste por microbolhas foram os motivos para a realização desta pesquisa. Este estudo, desenvolvido no Instituto de Radiologia do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo no período de novembro de 2004 a setembro de 2006 objetivou: A) estabelecer os padrões de vascularização do enxerto nos pacientes normais e naqueles com suspeita de complicações; B) comparar os achados de exames laboratoriais, uso ou não de insulina exógena e situação clínica dos pacientes com os achados da ultra-sonografia modo-B e com meio de contraste por microbolhas; C) Estabelecer o valor da ultra-sonografia com meio de contraste de microbolhas (USMCM) na avaliação do pâncreas transplantado. CASUÍSTICA e MÉTODOS: Vinte e seis pacientes foram submetidos ao exame de ultra-sonografia com contraste em um total de trinta estudos, 20 (66,7%) em homens e 10 (33,3%) em mulheres. A idade dos pacientes variou de 25 a 51 anos, com média de idade de 40 anos (+ 7,3 anos). O tempo médio da realização do transplante até a realização do estudo variou de um dia a 63 meses (mediana de 24 meses). Ao modo-B, o pâncreas foi analisado quanto a ecogenicidade, dimensões (avaliação qualitativa e quantitativa) e contornos. Após a administração do meio de contraste, observou-se o tempo de chegada do meio de contraste no pâncreas, o padrão e a intensidade de realce. Após esta avaliação foi realizada uma classificação baseada nos achados do modo-B e contraste nas seguintes possibilidades: padrão de perfusão normal, alterações agudas (rejeição, pancreatite ou trombose) ou alterações crônicas (rejeição crônica). RESULTADOS: Observou-se associação estatisticamente significante entre a ecogenicidade e a situação clínica (p=0,010); ecogenicidade e uso de insulina (p= 0,021); dimensões (avaliação qualitativa) e situação clinica (p=0,011); dimensões (avaliação qualitativa) e uso de insulina (p=0,028); padrão de realce (p=0,024) e intensidade do realce com a situação clínica (p=0,039). Houve associação estatisticamente significante também entre o uso ou não de insulina exógena com a perfusão do enxerto (p=0,014) e a hipótese diagnóstica (ultra-som) (p=0,001). CONCLUSÃO: Os padrões de vascularização do pâncreas transplantado normais e naqueles com suspeita de complicações foram estabelecidos. Os critérios de ecogenicidade e a avaliação qualitativa das dimensões do pâncreas ao ultra-som modo-B se mostraram adequados na diferenciação entre estudos normais e alterados. A ultra-sonografia com meio de contraste de microbolhas foi útil na diferenciação entre estudos normais e alterados do pâncreas transplantado, utilizando os critérios de padrão do realce, intensidade do realce e perfusão do enxerto na fase arterial. / INTRODUCTION: Pancreatic transplantation is the long-term therapeutic option for maintaining normoglycemic status in patients with type 1 diabetes mellitus. Sonographic contrast medium allows to evaluate tissue perfusion but there are no studies comparing this parameter in normal and pathologic pancreatic grafts in the literature at the time of this writing. The importance of assessment of graft perfusion and the potential role of contrast-enhanced sonography with microbubbles in this regard were the reasons for this research. The study conducted at the Instituto de Radiologia do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo between November, 2004 and September, 2006 had the following objectives: A) to establish patterns of graft vascularization in normal patients and in those with suspected abnormalities; B) to compare grayscale mode and contrastenhanced sonographic findings with clinical status, laboratory results, and need for exogenous insulin; C) to define the role of contrast-enhanced sonography with microbubbles in the evaluation of pancreatic transplantation.MATERIALS AND METHODS: Twenty-six patients underwent a total of 30 sonographic examinations with microbubbles, 20 (66.7%) men and 10 (33.3%) women. Age range was 25 to 51 years, with a mean of 40 years (+ 7.3 years). Time elapsed between transplantation and imaging ranged from 1 day to 63 months (median, 24 months). Using grayscale mode, pancreas was assessed for echogenicity, dimensions (both quantitatively and qualitatively) and contours. Following intravenous contrast medium administration, time for enhancement of the pancreatic graft, pattern and intensity of enhancement were documented. A classification system based on grayscale and contrast-enhanced sonographic findings was designed, with the following categories: normal perfusion pattern, acute changes (rejection, pancreatitis or thrombosis) and chronic changes (chronic rejection). RESULTS: There was a statistically significant association between echogenicity and clinical status (p=0.010); echogenicity and need for exogenous insulin (p=0.021); dimensions (qualitative criteria) and clinical status (p=0.011); dimensions (qualitative criteria) and need for exogenous insulin (p=0.028); pattern (p=0.024) and intensity of enhancement versus clinical status (p=0.039). There was also statistically significant association between need for exogenous insulin and graft perfusion (p=0.014), and sonography-based diagnosis (p=0.001). CONCLUSION: The study provided patterns of vascularization in normal pancreatic grafts and in patients with suspected abnormalities. Distinction of normal and abnormal pancreatic grafts was possible using echogenicity and qualitative analysis of graft size on grayscale mode. Contrast-enhanced sonography with microbubbles also contributed to differentiation between normal and abnormal pancreatic grafts, using pattern and intensity of enhancement and perfusion of the graft during the arterial phase.

INTRODUÇÃO: O transplante de pâncreas é a opção definitiva para a manutenção do estado normoglicêmico permanente nos portadores de diabetes mellitus tipo 1. O meio de contraste de ultra-som é uma metodologia capaz de avaliar a perfusão tecidual, mas não há um estudo para a avaliação do padrão de perfusão do pâncreas transplantado normal e patológico. A importância da avaliação da perfusão do enxerto e a aplicação do meio de contraste por microbolhas foram os motivos para a realização desta pesquisa. Este estudo, desenvolvido no Instituto de Radiologia do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo no período de novembro de 2004 a setembro de 2006 objetivou: A) estabelecer os padrões de vascularização do enxerto nos pacientes normais e naqueles com suspeita de complicações; B) comparar os achados de exames laboratoriais, uso ou não de insulina exógena e situação clínica dos pacientes com os achados da ultra-sonografia modo-B e com meio de contraste por microbolhas; C) Estabelecer o valor da ultra-sonografia com meio de contraste de microbolhas (USMCM) na avaliação do pâncreas transplantado. CASUÍSTICA e MÉTODOS: Vinte e seis pacientes foram submetidos ao exame de ultra-sonografia com contraste em um total de trinta estudos, 20 (66,7%) em homens e 10 (33,3%) em mulheres. A idade dos pacientes variou de 25 a 51 anos, com média de idade de 40 anos (+ 7,3 anos). O tempo médio da realização do transplante até a realização do estudo variou de um dia a 63 meses (mediana de 24 meses). Ao modo-B, o pâncreas foi analisado quanto a ecogenicidade, dimensões (avaliação qualitativa e quantitativa) e contornos. Após a administração do meio de contraste, observou-se o tempo de chegada do meio de contraste no pâncreas, o padrão e a intensidade de realce. Após esta avaliação foi realizada uma classificação baseada nos achados do modo-B e contraste nas seguintes possibilidades: padrão de perfusão normal, alterações agudas (rejeição, pancreatite ou trombose) ou alterações crônicas (rejeição crônica). RESULTADOS: Observou-se associação estatisticamente significante entre a ecogenicidade e a situação clínica (p=0,010); ecogenicidade e uso de insulina (p= 0,021); dimensões (avaliação qualitativa) e situação clinica (p=0,011); dimensões (avaliação qualitativa) e uso de insulina (p=0,028); padrão de realce (p=0,024) e intensidade do realce com a situação clínica (p=0,039). Houve associação estatisticamente significante também entre o uso ou não de insulina exógena com a perfusão do enxerto (p=0,014) e a hipótese diagnóstica (ultra-som) (p=0,001). CONCLUSÃO: Os padrões de vascularização do pâncreas transplantado normais e naqueles com suspeita de complicações foram estabelecidos. Os critérios de ecogenicidade e a avaliação qualitativa das dimensões do pâncreas ao ultra-som modo-B se mostraram adequados na diferenciação entre estudos normais e alterados. A ultra-sonografia com meio de contraste de microbolhas foi útil na diferenciação entre estudos normais e alterados do pâncreas transplantado, utilizando os critérios de padrão do realce, intensidade do realce e perfusão do enxerto na fase arterial. / INTRODUCTION: Pancreatic transplantation is the long-term therapeutic option for maintaining normoglycemic status in patients with type 1 diabetes mellitus. Sonographic contrast medium allows to evaluate tissue perfusion but there are no studies comparing this parameter in normal and pathologic pancreatic grafts in the literature at the time of this writing. The importance of assessment of graft perfusion and the potential role of contrast-enhanced sonography with microbubbles in this regard were the reasons for this research. The study conducted at the Instituto de Radiologia do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo between November, 2004 and September, 2006 had the following objectives: A) to establish patterns of graft vascularization in normal patients and in those with suspected abnormalities; B) to compare grayscale mode and contrastenhanced sonographic findings with clinical status, laboratory results, and need for exogenous insulin; C) to define the role of contrast-enhanced sonography with microbubbles in the evaluation of pancreatic transplantation.MATERIALS AND METHODS: Twenty-six patients underwent a total of 30 sonographic examinations with microbubbles, 20 (66.7%) men and 10 (33.3%) women. Age range was 25 to 51 years, with a mean of 40 years (+ 7.3 years). Time elapsed between transplantation and imaging ranged from 1 day to 63 months (median, 24 months). Using grayscale mode, pancreas was assessed for echogenicity, dimensions (both quantitatively and qualitatively) and contours. Following intravenous contrast medium administration, time for enhancement of the pancreatic graft, pattern and intensity of enhancement were documented. A classification system based on grayscale and contrast-enhanced sonographic findings was designed, with the following categories: normal perfusion pattern, acute changes (rejection, pancreatitis or thrombosis) and chronic changes (chronic rejection). RESULTS: There was a statistically significant association between echogenicity and clinical status (p=0.010); echogenicity and need for exogenous insulin (p=0.021); dimensions (qualitative criteria) and clinical status (p=0.011); dimensions (qualitative criteria) and need for exogenous insulin (p=0.028); pattern (p=0.024) and intensity of enhancement versus clinical status (p=0.039). There was also statistically significant association between need for exogenous insulin and graft perfusion (p=0.014), and sonography-based diagnosis (p=0.001). CONCLUSION: The study provided patterns of vascularization in normal pancreatic grafts and in patients with suspected abnormalities. Distinction of normal and abnormal pancreatic grafts was possible using echogenicity and qualitative analysis of graft size on grayscale mode. Contrast-enhanced sonography with microbubbles also contributed to differentiation between normal and abnormal pancreatic grafts, using pattern and intensity of enhancement and perfusion of the graft during the arterial phase.

Diabetes is a growing chronic disease and a worldwide problem. Without any available cure in sight for the public other methods needs to be applied to increase the life quality of diabetic patients. Artificial Pancreas (AP), a concept of having a closed loop system to control the glucose level on Type 1 Diabetes (T1D) patients has been introduced and is under development. In this thesis, Model Predictive Control (MPC) has been re implemented from scratch in MATLAB/SIMULINK with associated Kalman filter and prediction function. It was implemented in the latest version of the UVA/Padova Simulator which is a tool approved by FDA for simulating diabetes treatment in order to speed up the AP development. Different MPC cost functions where tested together with integral action on a simplified system using a linear approximation of a population model. It was implemented and tuned with a new simulation tuning method using Genetic Algorithm (GA). It showed that the quadratic cost function without integral action was the best with respect to performance and time efficiency. 3 hours was the best prediction horizon and was used for the individualized tuning using the University of Virginia (UVA)/Padova simulator. For the individualized MPC, models identified by the University of Padova were used. These simulations showed that an individualized model could be used for improved T1D treatment compared to an average population model even though the results were mixed. Almost all of the patients got improved treatment with the closed treatment and non hypoglycemic event occurred. The identification of better models is a great challenge for the future development of the AP MPC due to the excitation problems.

Introduction:Un des problèmes du cancer du Pancréas (CP) est le temps de latence entre la suspicion du CP et la mise en place des traitements. Les méthodes de biopsie liquide pourraient accélérer la mise en évidence d’éléments tumoraux et le diagnostic.Objectif :L’objectif principal de l’étude était de comparer la performance diagnostique de plusieurs techniques de biopsie liquide chez des patients atteint d’un CP résécable d’emblé. L’objectif secondaire était la corrélation avec le taux de récidive post-opératoire.Méthodes:Tout d'abord, nous avons testé 2 méthodes d'enrichissement CTC pour estimer la sensibilité de la détection CTC avec des expériences de cell-spiking de deux lignées de cellules tumorales pancréatiques dans des échantillons de sang de 24 volontaires sains en utilisant la méthode en gradient de densité OncoQuick® et la méthode de sélection négative RosetteSep™. De plus, les mutations KRAS ont été quantifiées dans l'ADN génomique de cellules purifiées par digital droplet PCR (dd-PCR) avec des amorces spécifiques des allèles.Nous avons conçu un essai clinique prospectif (NCT03032913) visant à détecter les cellules tumorales circulantes (CTC), l’ADN tumoral circulant (ADNct) et les onco-exosomes chez les patients atteint de CP et chez les patients d’un groupe témoin. Pour les CTCs : enrichissement et détection de CTCs par la méthode CellSearch©, méthode d’enrichissement de CTCs RosetteSep® et OncoQuick® puis quantification de l’ADN tumoral par dd-PCR. Les exosomes ont été isolés puis caractérisés avec le taux d’expression de Glypican-1. Tous les patients de l’étude ont eu un prélèvement de sang périphérique, les patients du groupe CP ont eu un prélèvement de sang portal peropératoire.Résultats:La sensibilité analytique était de 100 % pour OncoQuick®, quelle que soit la lignée cellulaire, et se situait entre 70 et 100 % pour RosetteSep™. Le taux moyen de récupération des cellules était de 56±23% pour OncoQuick® contre 39±27% pour RosetteSep™ (p<0,001). Les cellules tumorales de la population de cellules sanguines enrichies ont été détectées par dd-PCR après enrichissement par RosetteSep™ et OncoQuick® La détection des allèles K-RAS mutants par ddPCR après enrichissement de RosetteSepTM était 3 à 4 fois plus sensible qu'après OncoQuick®. Ainsi, RosetteSep™ est plus fiable en termes d'efficacité de récupération et de détection des mutants KRAS que OncoQuick®.De février à novembre 2017, 22 patients atteints de CP résécable et 28 patients témoins ont été inclus. Tous les patients ont été détectés positifs par au moins une méthode. Les CTCs ont été détectées chez 9 patients avec la méthode cellsearch (70% dans le sang portal exclusif) et 13 avec la méthode Rosettesep (60%). Les onco-exosomes ont été détecté chez 14 patients sur 22. L’ADNct n’a été détecté que chez deux patients métastatiques. La détection combinée des CTCs et des onco-exosomes était significativement corrélée à la survie sans récidive.Conclusion:Cette étude suggère que la biopsie liquide combinée peut être un outil prometteur à fois diagnostique et pronostique dans le CP à un stade précoce. / Introduction:One of the problems of pancreatic ductal adenocarcinoma (PC) is the latency time between the suspicion of PC and the initiation of treatments, especially neo-adjuvants that require histological evidence. Liquid biopsy methods could be a companion test for diagnosis.Objective :The main objective of the study was to compare the diagnostic performance of several liquid biopsy techniques in patients with resectable pancreatic without neo-adjuvant therapy cancer. The secondary objective was the correlation between the quantification of liquid biopsy parameters and clinic-pathologic features.Methods:First, we tested 2 CTC enrichment methods to estimate the sensitivity of CTC detection with cell spiking experiments of two pancreatic tumour cell lines in blood samples from 24 healthy volunteers using the onco-specific density gradient OncoQuick® and the negative selection enrichment method RosetteSep™. Additionally, KRAS mutations were quantified in genomic DNA of purified cells by digital droplet Q-PCR (dd-PCR) with allele specific primers.We designed a prospective clinical trial (PANC-CTC# NCT03032913) to detect circulating tumour cells (CTC), circulating tumour DNA (ADNct) and onco-exosomes in patients with pancreatic cancer and in patients in a control group using different methods. For CTCs, it was the enrichment and detection of CTCs by the CellSearch© method (reference method), the RosetteSep® and OncoQuick® CTC enrichment method and the quantification of tumor DNA by dd-PCR. Exosomes were isolated and characterized with the expression rate of Glypican-1. All patients in the study had a peripheral blood sample, patients in the PDAC group had a portal blood sample during surgery.Results:Analytical sensitivity was 100% for OncoQuick®, regardless of the cell line, and ranged between 70 and 100% for RosetteSep™. Mean recovery rate of cells was 56±23% for OncoQuick® versus 39±27% for RosetteSep™ (p<0.001). Molecular detection of mutant K-RAS alleles by ddPCR after RosetteSepTM enrichment was 3- to 4-fold more sensitive than after OncoQuick®. Thus, RosetteSep™ is more reliable in terms of recovery efficiency and KRAS mutant detection than OncoQuick®.From February to November 2017, 22 patients with resectable pancreatic cancer and 28 control patients were included. All patients were positive by at least one method. CTCs were detected in 9 patients with the cellsearch method (70% in the exclusive portal blood) and 13 with the Rosettesep method (59%), Onco-exosomes were detected in 14 out of 22(64%) patients in peripheral and/or portal blood. DNAct was detected in only two metastatic patients. The combined detection of CTCs with cellsearch and onco-exosomes was significantly correlated with progression free survival and overall survival when CTC cluster were found.Conclusion: This study suggests that combined liquid biopsy can be a promising tool for both diagnosis and prognosis in early pancreatic cancer.