INTRODUCTION: Lung cancer is the leading cause of cancer death, with low five-year survival rates and high percentage of local or metastatic recurrence. Surgical resection is integral for cure, however this procedure is associated with increased systemic inflammation and circulating tumour cells. This is further compounded by infectious postoperative complications, the most common of which is pneumonia, caused by both gram-positive and gram-negative bacteria. There is emerging clinical data to suggest that bacterial complications after lung cancer surgery may decrease survival, however the mechanism for this finding is not clear. Interactions between bacteria and host cells are mediated by specialized pathogen receptor proteins termed, Toll like receptors. Cancer cells have recently been shown to express Toll ike receptors, but their function in these malignant cells is unknown. The aim of this study is to investigate the mechanisms of bacterial infection facilitated lung cancer metastasis with a particular focus on the roles of Toll like receptors. METHODS: Murine H-59 Lewis lung carcinoma or Human A549 lung cancer cells were employed in a series of in vitro and in vivo experiments. Cancer cells were activated either directly with exposure to heat-inactivated S.pneumonia, E.coli or purified bacterial antigens, or indirectly with exposure to culture media from BEAS-2B bronchial epithelial cell incubation with heat-inactivated bacteria (S.pneumonia or E.coli) or bacterial antigens. In vitro assays include adhesion to extracellular matrix components fibronectin, collagen I and collagen IV, and migration/chemotaxis through a filter towards the BEAS-2B supernatant. In vivo assays include intravital microscopy to quantify in vivo cancer cell migration within the liver sinusoids, and intra-splenic hepatic metastasis assay. In some animals , a highly clinically relevant model of infection, cecal ligation and puncture (CLP) or sham control, was employed. To assess the role of Toll like receptors, we employed transgenic TLR4 knockout mice, monoclonal blocking antibodies, or small molecule inhibitors. RESULTS: Direct and indirect incubation of both H-59 and A549 cells with S.pneumonia and E.coli increased adhesion to extracellular matrix components approximately 2-8 fold and 2-4 fold, respectively. Presence of BEAS-2B culture media after infection with S.pneumonia and E.coli increased in vivo migration of H-59 cells approximately 2-4 fold over negative control, and into the bottom chamber approximately 4-5 fold. Direct and indirect incubation of H-59 cells with S.pneumonia and E.coli increased adhesion to liver sinusoids 2-4 fold. In vivo infection with CLP increased adhesion of H-59 cells to liver sinusoids 3-fold and increased number of liver metastases at 2 weeks post-injection approximately 12 fold. These effects were attenuated with TLR2 and TLR4 inhibition. CONCLUSION: Activation of murine and human lung cancer cells, either directly by S.pneumonia and E.coli, indirectly via activated bronchial epithelial cells, or via an in vivo infection, increases cancer cell adhesion, migration and metastasis, and this effect is mediated by toll like receptors. / INTRODUCTION: Le cancer du poumon est la principale cause de mortalité oncologique. La résection chirurgicale est essentielle dans le traitement et la rémission du cancer du poumon. Toutefois, la résection est aussi associée à une augmentation de l'inflammation systémique et des cellules tumorales circulantes. Cette situation est aggravée par des complications postopératoires de nature infectieuse, la plus fréquente étant la pneumonie causée par les bactéries gram-positives ou gram-négatives. De récentes recherches cliniques suggèrent que les complications de nature infectieuse après une résection chirurgicale d'un cancer du poumon diminuent la survie du patient. Cependant, la raison est présentement inconnue. Les interactions entre les bactéries et les cellules hôtes sont orchestrées par des récepteurs de reconnaissance de motifs moléculaires du nom de récepteurs de type Toll. Récemment, il a été démontré que les cellules cancéreuses expriment les récepteurs de type Toll, mais leur fonction dans les cellules cancéreuses restent inconnue. Le but de cette étude est d'étudier comment les infections bactériennes facilitent la métastase du cancer du poumon avec un regard particulier sur le rôle des récepteurs de type Toll. MÉTHODES: Des cellules cancéreuses pulmonaires de rat Lewis (H-59) ou humaines (A549) ont été employées dans une série d'expériences in vitro et in vivo. Les cellules cancéreuses ont été activées directement par S.pneumonia inactivé par la chaleur, E.coli inactivé par la chaleur ou des antigènes bactériens purifiés, ou indirectement en les exposant à un milieu de culture provenant de cellules épithéliales bronchiques BEAS-2B incubées avec des bactéries inactivées par la chaleur (S.pneumonia, E. coli) ou des antigènes bactériens. Les expériences in vitro réalisées incluent l'adhérence aux éléments de la matrice extracellulaire fibronectine, collagène de type I et collagène de type IV, et la migration/chimiotaxie à travers un filtre vers le surnageant de culture provenant des cellules BEAS-2B. Les expériences in vivo réalisées incluent la quantification de la migration des cellules cancéreuses dans les sinusoïdes hépatiques par microscopie intravitale et des métastases hépatiques suite à l'injection intra-splénique de cellules cancéreuses. Chez certains animaux, un modèle cliniquement pertinent de l'infection, soit la ligature et ponction du caecum, a été employé. Pour évaluer le rôle des récepteurs de type Toll, nous avons utilisé des souris transgéniques knockout pour le gène TLR4, des anticorps monoclonaux bloquant les récepteurs de type Toll, ou des inhibiteurs de petites molécules. RÉSULTATS: L'incubation directe et indirecte des cellules cancéreuses H-59 et A549 avec S.pneumonia et E. coli ont augmenté leur adhérence aux éléments de la matrice extracellulaire d'environ 2 à 8 fois et 2 à 4 fois respectivement. La présence de milieu de culture provenant de cellules BEAS-2B infectées par E. coli et S.pneumonia a augmenté la migration in vivo des cellules H-59 d'environ 2 à 4 fois comparativement au contrôle négatif, et dans la chambre inférieure d'environ 4 à 5 fois. L'incubation directe et indirecte des cellules H-59 avec S.pneumonia et E.coli ont augmenté leur adhérence aux sinusoïdes hépatiques d'environ 2 à 4 fois. L'infection in vivo par le modèle de la ligature et ponction du caecum a augmenté de 3 fois l'adhérence des cellules H-59 aux sinusoïdes hépatiques et a augmenté le nombre de métastases hépatiques d'environ 12 fois deux semaines après injection. Ces observations ont été atténuées par l'inhibition des récepteurs de type Toll TLR2 et TLR4. CONCLUSION: L'activation des cellules cancéreuses pulmonaires murines et humaines, soit directement par S.pneumonia et E. coli, soit indirectement par les cellules épithéliales bronchiques activées, ou soit via une infection in vivo, augmente leur adhérence, migration, et potentiel métastatique à travers les récepteurs de type Toll.

Invasive research procedures, such as biopsy for pharmacodynamic study, often have no value for patient-volunteers in terms of diagnosis or clinical management. Accordingly, their burdens are generally justifiable only by appeal to knowledge value (that is, the benefit to future patients) rather than direct benefit. This thesis is an exploration of knowledge value production as a justification for invasive research biopsy in cancer research. The premise of this thesis is that human investigations can only enable knowledge value insofar as they are reported in ways that enable a broader scientific community to use their findings for planning new investigations. We studied the interruption of knowledge value accrual through two empirically evaluable proxies of knowledge value: publication and reporting quality.The focus of this thesis is on the publication and reporting quality of pharmacodynamic (PD) sub-studies embedded within early phase cancer clinical trials. Early phase cancer trials are designed to measure the safety and toxic effects of an investigational agent and tumor response (i.e. tumor shrinkage). PD sub-studies within them investigate the effects of a drug on its 'targets' (i.e. the inhibition of an enzyme or enzymatic pathway). We found that 37% of early phase cancer trials utilizing biopsy for PD study result in the complete publication of all PD data. A survey of study authors revealed that the most commonly cited barriers to publication were "strategic considerations in publication," which included studies where results were dismissed as uninteresting, uninformative, contradictory or difficult to interpret, and where "scientific disagreement" prevented the publication of pharmacodynamic results (59% of respondents). Quality of reporting varied widely within and across studies, with some important quality assurance practices being sporadically reported, including results of all planned tests (78% trials reporting), use of blinded outcome assessment (43% trials reporting), biopsy dimensions (38% trials reporting), and description of patient flow through the PD portion of the trial (62% trials reporting). PD analysis as a primary endpoint and the use of mandatory biopsy were significantly and positively associated with better quality reporting. A preponderance of positive results (61% of the studies described positive PD results) suggests the possibility of publication bias.Based on our research findings, we recommend that study investigators and IRB members critically evaluate targets for patient recruitment, tissue collection, and PD assay validation. Further, IRBs and investigators should ensure that any laboratories proposing the collection and assay of biopsied tissues have the requisite resources (financial, human, and physical) to complete a validated PD study. The major recommendation of this thesis is that a formalized reporting guideline, similar to CONSORT and REMARK, should be developed for PD investigations. We further recommend that study authors and journal editors consider separate PD publication or the inclusion of supplementary materials in order to provide space for richer methodologic description of PD research. / Les procédés de recherche effractifs, tels que la biopsie dans le cadre d'études pharmacodynamiques, n'ont souvent aucun avantage pour le patient-bénévole en ce qui concerne le diagnostic ou la prise en charge clinique. Le recours à ces procédés n'est souvent fondé que sur l'idée de la valeur des connaissances (ce en quoi ils aideront d'éventuels patients futurs) plutôt que sur l'avantage direct pour le patient.Cette thèse propose d'étudier la justification du recours à des procédés de biopsie effractive dans la recherche sur le cancer par la production de connaissances. Nous proposons comme prémisse que la recherche sur les êtres humains ne peut produire de connaissances de valeur que dans la mesure où les résultats de la recherche sont présentés de manière à permettre à la communauté scientifique élargie de les utiliser à des fins de recherches ultérieures. Nous avons étudié l'interruption de l'accumulation de connaissances de valeur selon deux critères empiriques : la publication et la qualité des renseignements. Cette thèse se concentre particulièrement sur la publication et la qualité des renseignements dans les sous-études pharmacodynamiques contenues dans les essais cliniques de premières phases dans le domaine du cancer. Ces essais visent à mesurer l'innocuité et les effets toxiques potentiels d'un nouveau médicament de recherche, ainsi que la réaction de la tumeur. Les sous-études pharmacodynamiques évaluent les effets du médicament sur ses « cibles » (un enzyme ou une voie enzymatique). Nos résultats indiquent que pour 37% des essais cliniques de premières phases qui comprennent des sous-études pharmacodynamiques utilisant la biopsie, toutes les données pharmacodynamiques ont été publiées. À la suite d'une enquête auprès d'auteurs d'études, les obstacles les plus communs sont les « considérations d'ordre stratégique quant à la publication », dans le cas d'études où on juge que les résultats n'apportent rien de nouveau ou sont sans intérêt, contradictoires ou difficiles d'interprétation, et lorsque des « différends d'ordre scientifique » empêchent la publication de résultats pharmacodynamiques (59% des répondants). La qualité des renseignements varie grandement à travers les études et on remarque que certaines données importantes pour le contrôle de la qualité ne sont qu'irrégulièrement signalées, dont les résultats de tous les tests planifiés (78% des essais), l'évaluation des résultats à l'insu (43% des essais), les dimensions des tissus prélevés (38% des essais) et la description du cheminement du patient durant la section pharmacodynamique de l'étude (62% des essais). Lorsque l'analyse pharmacodynamique est le but premier de l'essai ou que celui-ci comprend un recours obligatoire à la biopsie, l'étude tend considérablement à être de meilleure qualité. La prépondérance de résultats positifs (61% des études) suggère peut-être un biais de publication. Compte tenu des résultats de notre recherche, nous proposons que les investigateurs d'études et les membres de comités d'éthique indépendants évaluent de manière critique les objectifs en matière de recrutement des patients, de prélèvement des tissus et de validation des essais pharmacodynamiques. De plus, les comités d'éthique indépendants et les investigateurs doivent s'assurer que tout laboratoire visant le prélèvement et l'analyse de tissus par la biopsie possèdent les ressources financières, humaines et matérielles nécéssaires pour compléter une étude pharmacodynamique validée. La recommandation principale de cette thèse consiste en l'élaboration d'une ligne directrice formalisée, comparable à CONSORT ou REMARK, pour la publication d'études pharmacodynamiques. Nous recommandons également que les auteurs d'études et les éditeurs de revues envisagent de publier indépendamment les résultats pharmacodynamiques ou d'inclure des matériaux supplémentaires afin de permettre une description méthodologique plus riche de la recherche pharmacodynamique.

Although there have been many improvements in technical aspects of microsurgical reconstructive surgery using free tissue transfer, partial and complete flap failure due to ischemia reperfusion injury still remains a significant clinical problem. Our limited understanding of the pathophysiology of ischemia reperfusion injury has been a major impediment in developing interventions to prevent or minimize this injury in order to reduce patient morbidity associated with major reconstructive procedures. There has been profound interest in the role played by the gaseous free radical nitric oxide in ischemia reperfusion injury in cerebral, myocardial and intestinal tissue. The role of this molecule has not been clearly identified in ischemia reperfusion injured surgical flaps. It has been shown that skin exhibits less of a response to salvage strategies than does skeletal muscle. The objectives of this thesis were to characterize and compare the flow dynamics in reperfused skin and skeletal muscle flaps and to explore the role played by nitric oxide in attenuation of ischemia reperfusion injury in these flaps. In a series of three complementary experiments it was shown that (1) the effect of ischemia reperfusion injury on flow patterns varies significantly between flap skin and skeletal muscle. In flap skeletal muscle, an early hyperemic phase during reperfusion maintains a significant blood flow to all regions including the area of the flap that is destined for necrosis. In flap skin however, there is a marked decrease in flow rates. (2) Ischemia, reperfusion injury causes a significant decrease in the function and availability of constitutive nitric oxide synthase enzymes and (3) infusion of a nitric oxide donor during reperfusion significantly attenuates tissue injury.

The aims of this project were to: (1) successfully design two models of reconstructive tissue transplants in the primate, one with a purely sensory nerve supply, the other a mixed sensory and motor supply and (2) achieve long enough survival for reinnervation to have occurred, assuming it can take place in the presence of the immunosuppressants. / A neurovascular free flap comprised of the entire soft tissue coverage of the second digit and a hand transplant model were successfully designed in the baboon (Papio hamadryas anubis). Seven transplanted neurovascular free flaps and four hand transplants were undertaken. High dose Cyclosporin A was found to be necessary to prevent rejection. Steroids proved to be a necessary part of the immunosuppressive regime. Nine out of 11 transplants survived to or beyond 4 months. In most cases, the end point was determined by the date for evaluation of reinnervation by our neurophysiologist colleagues and not loss of the transplant due to rejection. / Only 3 out of 11 transplants survived with little or no signs of rejection. All others had significant episodes of rejection, most of which were successfully reversed or controlled by using our rejection protocol. In addition, all animals, to varying degrees, demonstrated some of the following side effects: anorexia, anemia, gingival hyperplasia, hepatotoxicity, hirsutism, lymphoma, nephrotoxicity, subcutaneous or intramuscular abscesses and tremors. (Abstract shortened by UMI.)

Secondary peritonitis is a significant cause of morbidity and mortality in the ICU and ICU patients as a group have the highest rate of nosocomial infections. Once recruited to the site of injury, the PMN interacts with endothelial cells (ECs) via rolling adhesion, firm adhesion, and transendothelial migration. Using a murine cecal ligation and puncture peritonitis model and either skin or cremaster injury as the secondary site in a two-front injury model, we examined the role of injury severity on the triage of PMNs to competing sites of injury. We demonstrated that a finite pool of PMNs was recruited to tissues in numbers correlating to the severity major injury. With intravital microscopy we demonstrated that numbers of PMNs involved in rolling adhesion, rolling velocity and stationary adhesion in the presence of one or more sites of injury were predictable, consistent and likely mediated by changes in surface adhesion molecule expression.

Distraction osteogenesis is a form of in vivo tissue engineering used to lengthen bone. Bone morphogenetic proteins (BMPs) are known mediators of bone formation during the distraction of enchondral bones. Their role in the distraction of membranous bones remains unknown. A rabbit model was used to characterize the expression of BMP-2, 4 and 7 during mandibular distraction osteogenesis. Fourteen animals were subjected to a defined distraction protocol. At weekly intervals, two rabbits were sacrificed and the generate bone harvested for radiographic, bone densitometric, histologic and immunohistochemical analyses. The results demonstrate that during distraction, bane forms primarily by intramembranous ossification. BMP-2 and 4 were maximally expressed in osteoblastic cells during the distraction period and in chondrocytes during the consolidation period. The pattern of mandibular BMP expression was different than previously described for long bones. Understanding the pattern of BMP expression may guide the strategic administration of recombinant BMPs to enhance the rate and quality of bone formation during distraction osteogenesis thereby shortening the time required for consolidation of the generate bone.

This study sought to maximize prenatal and postnatal interventions in order to accelerate lung growth and improve lung function in two animal models. Prenatal interventions consisted of fetal tracheal occlusion (TO), antenatal glucocorticoids and exogenous surfactant at birth (SURF) in an ovine model of congenital diaphragmatic hernia (CDH). CDH, CDH+TO, CDH+SURF, CDH+TO+SURF and unoperated twin control lambs were compared. Prenatal growth of both lungs was accelerated after fetal TO. Prophylactic surfactant did not improve gas exchange or ventilation but did increase lung compliance over 8 hours. The incidence of tension pneumothoraces was slightly decreased after exogenous surfactant. Fetal TO yields the best results in terms of overall postnatal lung function, likely acting via surfactant independent mechanisms. / Postnatal intervention involved perfluorocarbon (PFC) liquid distention of the right upper lobe in healthy newborn piglets. Postnatal lung growth, as measured indirectly by rates of DNA synthesis, was not accelerated after PFC distention.

In a normal inflammatory response, the polymorphonuclear neutrophil (PMN) is recruited to a site of injury or infection through a sequence of events that include margination, rolling, adhesion, diapedesis and chemotaxis. These processes become dysregulated in SIRS, sepsis and MODS. Septic patients demonstrate decreased PMN exudation to peripheral sites and an absence of a C5a chemotactic gradient across the intravascular and extravascular environments. Using both in vitro and in vivo PMN transmigration assays, we evaluated the role of an intact C5a gradient on human PMN recruitment to peripheral inflammatory sites in sepsis. We demonstrated in vitro that a C5a gradient could induce both beta2-integrin dependent and independent transendothelial migration of septic PMN. In vivo, the exogenous re-establishment of the C5a gradient increased septic PMN exudation to normal baseline levels. These observations implicate an intact C5a gradient as an essential component to optimal PMN recruitment in sepsis.

Background. Data indicates that early xenoislet graft failure is due to non specific inflammatory mechanisms that occur prior to T cell mediated rejection. The host hepatic endothelial cell may be important in this process because it is essentially the first cellular barrier encountered by intraportally introduced islets. We have characterized the initial response of endothelium to xenogeneic islets by measuring the expression of 11-1alpha, TNF-alpha, IFN-gamma, and iNOS mRNA in an in vitro dog-to-pig model of xenoislet transplantation. / Materials & methods. Dog islets were co-cultured with either porcine hepatic endothelium or porcine aortic endothelium over a 24-hour period in serum free medium. RNA was extracted at eight time points. RT-PCR was performed in order to visualize II-1alpha, TNF-alpha, IFN-gamma, and BIOS expression. Bands were semiquantitated by comparison to an external standard (GAPDH) using band densitometry. / Results. Hepatic endothelium had early (one hour) expression of IL-1alpha, IFN-gamma, and iNOS transcript. IL-1alpha peaked at two hours, IFN-gamma at 12 hours, and iNOS at one and 12 hours. Aortic endothelium expressed low levels of IL-1alpha and TNF-alpha, but not IFN-gamma or NOS. / Conclusions. We have demonstrated that xenogeneic islets are able to activate host endothelial cells without serum or immune cells. The observed endothelial response corresponds with known islet toxic substances. Furthermore, the response differs between endothelial beds suggesting that these differences may be important in choosing suitable implantation sites for islets. Our findings suggest that host endothelium may play an important part in early injury of islet xenotransplants.

Arsenic trioxide (As2O3) is an effective treatment strategy for acute promyelocytic leukemia (APL), as sensitivity to As2O3 may correlate with the induction of oxidative stress. However, the use of As2O3 in other malignancies is limited at clinically achievable doses. To expand the clinical potential of arsenic to other malignancies, a promising novel arsenical, darinaparsin (S-dimethylarsino-glutathione, ZIO-101) is more potent in vitro than As2O3 in a variety of hematopoietic malignant cell lines. At equimolar concentrations, darinaparsin is significantly more potent than As2O3 in inducing signalling pathways known to be required for arsenic-induced apoptosis. Moreover, unlike As2O3, darinaparsin is a selective inducer of Nrf2 antioxidant responses, thereby generating unopposed oxidative stress response and apoptosis in APL. Interestingly, darinaparsin does not demonstrate sensitivity to inhibition of ABCC1 (Mrp1), a known transporter involved in resistance to arsenic, suggesting that darinaparsin may be an effective treatment in tumours that overexpress ABCC1. / L’anhydride arsénieux (As2O3) est un traitement efficace contre la leucémie aigue myéloide (LAM). La sensibilité envers l’As2O3 peut être attribué à l’induction de stress oxidative. Cependant, l’usage d’ As2O3 dans le cadre d’autres malignités est limité par la posologie clinique. Pour élargir le potential thérapeutique de l’arsenic, un nouveau composé arsenical, le darinaparsin (ZIO-101), est plus efficace in vitro que l’ As2O3 dans une variété de cellules hématopoïétiques malines. Dans des conditions d’équimolarité, le darinaparsin est beaucoup plus puissant que l’ As2O3 pour provoquer des voies métaboliques requises pour l’apoptose. De plus, contrairemet à l’ As2O3, le darinaparsin insite séléctivement l’antioxidant Nrf2. Ce dernier produit un stress oxidative incontesté qui permet l’apoptose dans le cadre de la LAM. Curieusement, le darinaparsin ne répond pas à l’inhibition de l‘ABCC1 (Mrp 1), un transporteur qui confère de la résistance contre l’arsenic. Ceci suggère que le darinaparsin peut être un traitement efficace contre des tumeurs qui possèdent une importante quantitée d’ABCC1.