Medical treatments with statins are prescribed to patients with increased risk of cardiovascular events. The benefits from statin treatment are well documented in clinical trials, but long-term adherence in patients is low, indicating that patients have an uncertainty about the necessity and benefits of treatment. The aims of this thesis were to investigate how patients and doctors view different aspects of statin treatment. Further aims were to investigate if the cardiovascular risk level in patients affects their views about different aspects of statin treatment. Yet further aims were to compare health behaviours and views about risk factors in patients using statins to a non-treated population. Data was obtained from patients (n = 829), doctors (n = 330) and a population sample (n = 720) using postal questionnaires. Views about the effect of statin treatment were assessed in different ways for patients and doctors. Patients based their assessments on their own situation, and doctors’ treatment decisions and assessments of anticipated effect of treatment were based on two hypothetical patient cases. The results indicate that patients greatly overestimate the general effect of statins, compared to efficacy results reported from clinical trials. Patients’ previous coronary heart disease or high overall risk were factors not associated with their views and expectations of treatment effect. Statin users with an internally perceived health control and patients satisfied with their doctor’s treatment explanation reported higher beliefs in treatment necessity and benefits. Statin users reported having better health behaviours and generally rated risk factors as more important than the non-treated population. Doctors had suboptimal understanding of the number of patients expected to benefit following five years of statin treatment and had a varying understanding of statins’ ability to prolong life. Overall the results illustrate that patients and doctors have different perspectives and views of the benefits from statin treatment which puts emphasis on how statin treatment is discussed in the clinical setting.

Experiments were conducted to investigate the role of free radicals in exercise induced cardiac adaptations and to determine if statin administration would adversely affect cardiac adaptations to exercise. In the first experiment myocardial antioxidant enzymes, cardiac function and cardiac hypertrophy were assessed following a chronic exercise protocol previously used by our lab. MPG effectively reduced myocardial oxidative stress and activation of the signaling proteins Akt and S6 following an exercise bout. Skeletal muscle mitochondria content increased to similar levels in E and E+MPG. Similar increases (P<0.05) in both exercised groups were observed for heart wt and heart wt to body wt ratio. Cardiac function at the high workload improved in E vs S as indicated by higher (P<0.05) peak systolic pressure (SP), cardiac output (CO), coronary flow, COxSP and mechanical efficiency (COxSP/VO2). MPG prevented these exercise-induced functional improvements. This study provides evidence that free radicals do not play a role in the development of exercise-induced cardiac hypertrophy, however, they are involved in functional cardiac adaptations, which may be mediated through the PI3K/Akt pathway. In the second experiment a similar exercise protocol was used to determine if statins which have been shown to prevent pathological forms of cardiac hypertrophy, would be detrimental to exercise induced cardiac adaptations. In addition to the sedentary and exercise groups sedentary+statin and exercise+statin groups were assessed. Hearts were isolated and perfused and assessed for function at low and high workloads. Exercise treatment resulted in cardiac hypertrophy in absolute and relative terms to a similar extent in statin-treated and untreated exercised rats. Additionally it resulted in significant functional increases for SP, CO, COxSP, VO₂, and EFF in both exercised groups. In conclusion, these studies provide evidence that exercise in the cold is a valid model for physiological cardiac hypertrophy and that pathological and physiological cardiac hypertrophy signal through different pathways due to the fact that two well established treatments (mpg and statins) that prevent pathological cardiac hypertrophy did not affect exercise induced cardiac hypertrophy. / text

Radiotherapy is the second most important treatment modality after surgery in the treatment of cancer. Recent technical advancements, such as intensity-modulated radiation therapy (IMRT) or image-guided radiation therapy (IGRT), combined with new targeted drugs have significant promise for therapeutic outcome. However radiation treatment could result in disabling normal tissue injury and in the development of progressive fibrosis in a subset of sensitive patients and in long-term cancer survivors. The main feature of tissue fibrosis is excessive accumulation of abnormal and cross-linked collagen mainly composed of fibrillar and immature extracellular matrix (ECM) components.The organs that can be affected by this phenomenon are liver, skin, intestine, kidneys and lungs. From a clinical point of view, fibrosis can be seen as an irreversible condition, without solution. We and others recently showed that beside the activation of the canonical TGF-β/Smad pathway, other intracellular signaling cascades including the Rho/ROCK pathway are switched on in fibrotic tissues. Interestingly, the Rho/ROCK pathway seems differentially activated in radiation-induced intestinal fibrosis, thereby providing a rationale for a specific, targeted anti-fibrotic strategy. Pharmacological inhibition of Rho using statins indeed prevent and even reverse intestinal radiation fibrosis.In our studies, we showed the role of Statin (Pravastatin e Simvastatin) and a specific inhibitor ROCK inhibitors (Y-27632) in a mice model of pulmonary induced-fibrosis obtained by a pharmacological approach (Bleomycin - BLM). Indeed, we developed a model of lung fibrosis by complete irradiation of chest and tested Pravastatin action. Confirmation of the involvement of Rho/ROCK/CTGF pathway in lung fibrosis are shown by immunohistochemistry: Pravastatin-treament normalized the expression of three markers: RhoB, TGF-RII and CTGF.Then, in models of radiation induced gut and lung fibrosis, we analysed, from a immunohistological point of view, the underlying mechanisms of the antifibrotic action of Pravastatin via MMP2-TIMP2 axis. Interestingly we found a different impact on fibrolysis when Pravastatin was administered preventively or curatively.Finally, in vitro, we investigate by zymography the expression of Gelatinases (MMP2 and MMP9) in primary lung fibroblasts cultures exposure at the different radiation and Pravastatin doses. Metalloproteases would appear to be in turn involved in pro-fibrolytic mechanisms induced by statin.The multiplicity of actors involved in the pathogenesis of fibrotic lesions explains why the definition of an effective therapeutic strategy is so complex.Researches in mechanistic processes of normal tissue damage paved the way for new therapeutic approaches. These new targets include reduction of vascular activation, inflammation and thrombosis and new molecular targets definition. Effective strategies are multiple on preclinical models, but numerous efforts have to be made to achieve the complicated goal of protection of normal tissues from the side effects of radiation therapy.

La proprotéine convertase subtilisine/kexine-9 (PCSK9) a été identifiée comme le troisième locus impliqué dans l’hypercholestérolémie autosome dominante (ADH). Les deux autres gènes impliqués dans l’ADH encodent le récepteur des lipoprotéines de faible densité (LDLR) et l’apolipoprotéine B. La PCSK9 est une convertase qui favorise la dégradation du LDLR dans les hépatocytes et augmente le niveau plasmatique de cholestérol des LDL (LDL-C). Les mutations « gain de fonction » de la PCSK9 sont associées à un phénotype d’hypercholestérolémie familiale, tandis que les variantes « perte de fonction » sont associées à un LDL-C réduit et à un risque coronarien plus faible. Pour élucider le rôle physiologique de la PCSK9, nous avons étudié sa régulation génique. En utilisant le RT-PCR quantitatif dans des hépatocytes humains, nous avons analysé la régulation de PCSK9 sous différentes conditions modulant l’expression des gènes impliqués dans le métabolisme du cholestérol. Nous avons démontré que l’expression de la PCSK9 était induite par les statines de manière dose-dépendante et que cette induction était abolie par le mévalonate. De plus, le promoteur de PCSK9 contenait deux motifs conservés pour la régulation par le cholestérol : le sterol regulatory element (SRE) et un site Sp1. La PCSK9 circule dans le plasma sous des formes mature et clivée par la furine. Grâce à notre anticorps polyclonal, nous avons mis au point un test ELISA mesurant la PCSK9 plasmatique totale. Une étude transversale a évalué les concentrations plasmatiques de PCSK9 chez des sujets sains et hypercholestérolémiques, traités ou non par des statines ou une combinaison statine/ezetimibe. Chez 254 sujets sains, la valeur moyenne de PCSK9 (écart-type) était de 89,5 (31,9) µg/L. La concentration plasmatique de la PCSK9 corrélait avec celle de cholestérol total, du LDL-C, des triglycérides (TG), de la glycémie à jeun, l’âge et l’indice de masse corporelle. Le séquençage de PCSK9 chez des sujets aux extrêmes de la distribution des concentrations de PCSK9 de notre cohorte a révélé la présence d’une nouvelle variation « perte de fonction » : R434W. Chez 200 patients hypercholestérolémiques, la concentration de PCSK9 était plus élevée que chez les sujets sains (P<0,04). Elle a augmenté avec une dose croissante de statine (P<0,001), et a augmenté encore plus suite à l’ajout d’ezetimibe (P<0,001). Chez les patients traités, ceux présentant une hypercholestérolémie familiale (HF; due à une mutation du LDLR) avaient des concentrations plus élevées de PCSK9 que les non-HF (P<0,005), et la réduction de LDL-C corrélait positivement avec la concentration de PCSK9 atteinte de la même manière dans les deux sous-catégories (P<0,02 et P<0,005, respectivement). Par ailleurs, une incubation des cellules HepG2 (hépatocytes) et Caco-2 (entérocytes) avec de l’ezetimibe a provoqué une augmentation de l’ARNm de PCSK9 et de NPC1L1 de 1,5 à 2 fois (P<0,05), mais aucune variation significative de PCSK9 sécrétée n’a été observée, suggérant que ces lignées cellulaires ne sont pas un modèle idéal. Nous avons également mesuré le niveau de PCSK9 chez 1 739 Canadiens-français âgés de 9, 13 et 16 ans. La valeur moyenne (écart-type) de PCSK9 dans cette cohorte était de 84,7 (24,7) µg/L, légèrement plus basse que dans la cohorte d’adultes (89,5 (31,9) µg/L). Chez les garçons, la PCSK9 circulante diminuait avec l’âge, tandis que c’était l’inverse chez les filles. Il y avait des associations positives et significatives entre la PCSK9 et la glycémie à jeun, l’insulinémie, le HOMA-IR, et les paramètres lipidiques (TC, LDL-C, TG, HDL-C, apoAI et apoB). Dans l’analyse multivariée, une hausse de 10% de l’insulinémie à jeun était associée à une augmentation de 1 à 2% de PCSK9. La régulation de PCSK9 est typique de celle d’un gène impliqué dans le métabolisme des lipoprotéines et est probablement la cible du facteur de transcription «sterol regulatory element-binding protein » (SREBP-2). La concentration plasmatique de la PCSK9 est associée avec l’âge, le sexe, et de multiples marqueurs métaboliques chez les enfants et les adultes. La détection de la PCSK9 circulante chez les sujets HF et non-HF signifie que ce test ELISA spécifique à PCSK9 pourrait servir à suivre la réponse à la thérapie chez un grand éventail de sujets. PCSK9 semble être une cible thérapeutique prometteuse dans le traitement de l’hypercholestérolémie et de la maladie cardiovasculaire. / Proprotein convertase subtilisin/kexin type 9 (PCSK9) has been identified as the third locus implicated in autosomal dominant hypercholesterolemia (ADH). The two other known genes implicated in ADH encode the low-density lipoprotein receptor (LDLR) and apolipoprotein B. PCSK9 is a protein convertase that post-translationally promotes the degradation of the LDLR in hepatocytes and increases plasma LDL cholesterol concentration (LDL-C). Heterozygote “gain-of-function” mutations of PCSK9 are associated with the familial hypercholesterolemia phenotype, whereas “loss-of-function” variants are associated with reduced LDL-C concentrations and lower coronary risk. As an approach toward the elucidation of the physiological role(s) of PCSK9, we studied its transcriptional regulation. Using quantitative RT-PCR, we assessed PCSK9 regulation under conditions known to regulate genes involved in cholesterol metabolism in HepG2 cells and in human primary hepatocytes. We found that PCSK9 expression was strongly induced by statins in a dose-dependent manner and that this induction was efficiently reversed by mevalonate. The PCSK9 promoter contains two typical conserved motifs for cholesterol regulation: a sterol regulatory element (SRE) and an Sp1 site. PCSK9 circulates in plasma as mature and furin-cleaved forms. A polyclonal antibody against human PCSK9 was used to develop an ELISA that measures total plasma PCSK9 rather than only the mature form. A cross-sectional study evaluated plasma levels in normal and hypercholesterolemic subjects treated or untreated with statins or statin plus ezetimibe. In 254 healthy subjects, the mean plasma PCSK9 (SD) concentration was 89 (32) µg/L. PCSK9 levels correlated positively with plasma cholesterol, LDL-C, triglycerides, fasting glucose, age and body mass index. Sequencing PCSK9 from subjects at the extremes of PCSK9 plasma distribution revealed a new loss-of-function R434W variant. In 200 hypercholesterolemic patients, circulating PCSK9 was higher than in controls (P<0.04), increased with increasing statin dose (P<0.001), and further increased when ezetimibe was added (P<0.001). In treated patients (n = 139), those with familial hypercholesterolemia (FH; due to LDLR gene mutations) had higher PCSK9 values than non-FH (P<0,005), and LDL-C reduction correlated positively with achieved plasma PCSK9 levels to a similar extent in both subsets (P<0.02 and P<0.005, respectively). However, incubation with ezetimibe of HepG2 (hepatocytes) and Caco-2 (enterocytes) cells caused an increase in PCSK9 and NPC1L1 mRNA of 1.5 to 2-fold (P<0.05), but no significant rise in PCSK9 protein secretion, suggesting that these transformed cells are not an ideal model. We also studied PCSK9 levels in 1,739 French Canadian youth ages 9, 13, and 16 years old. The mean (SD) plasma PCSK9 concentration, measured by ELISA, was 84.7 (24.7) µg/L in the cohort, slightly lower than in the adult cohort (89.5 (31.9) µg/L. In boys, plasma PCSK9 decreased with age, whereas the inverse was true for girls. There were significant positive associations between PCSK9 and fasting glucose, insulin, and HOMA-IR (homeostasis model assessment of insulin resistance). In multivariable analysis, a 10% higher fasting insulin was associated with a 1%-2% higher PCSK9 in both sexes. There were also positive associations between PCSK9 and total cholesterol, LDL-C, and triglycerides, as well as with HDL-C and apolipoproteins A1 and B. PCSK9 regulation is typical of that of the genes implicated in lipoprotein metabolism. In vivo, PCSK9 is probably a target of the transcription factor “sterol response element-binding protein” (SREBP)-2. The PCSK9 plasmatic concentration is associated with age, sex, and multiple metabolic markers in youth and adult samples. The detection of circulating PCSK9 in both FH and non-FH subjects means that this PCSK9 ELISA test could be used to monitor response to therapy in a wide range of patients. PCSK9 seems to be a promising drug target in the treatment of hypercholesterolemia and coronary heart disease.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

Overall survival of the metastatic forms of epithelial derived cancers, especially head and neck squamous cell carcinomas (HNSCC), has not significantly improved even with the application of aggressive combined modality approaches incorporating radiation and chemotherapy. Cumulative evidence implicates the epidermal growth factor receptor (EGFR) as an important therapeutic target in HNSCC. We have previously demonstrated that the combination of lovastatin, a potent inhibitor of the mevalonate pathway, with EGFR tyrosine kinase inhibitors induced robust synergistic cytotoxicity. However, the use of high dose statins in our clinical trial was associated with significant toxicities including higher than anticipated rate of muscle pathologies. Our goal was to uncover novel downstream targets of the mevalonate pathway that may enhance the efficacy or limit toxicities of this novel combination therapeutic approach. In this study we have demonstrated that GGTI-298, an inhibitor of protein geranylgeranylation, through its ability to disrupt the actin cytoskeleton, inhibits EGFR dimerization and cellular trafficking. This novel mechanism targeting the EGFR has clinical implications as GGTI-298 in combination with tarceva, a clinically relevant EGFR inhibitor, showed enhanced cytotoxicity and inhibitory effects on EGFR activation and its downstream signaling.

The statin family of hydroxymethylglutaryl coenzyme A reductase (HMGCR) inhibitors, used to control hypercholesterolemia, triggers apoptosis of various human tumour cells. HMGCR is the rate-limiting enzyme of the mevalonate (MVA) pathway, a fundamental metabolic pathway required for the generation of a number of biochemical end-products including cholesterol and isoprenoids, but the contribution of the MVA pathway to human cancer remains largely unexplored. Furthermore, as only a subset of tumour cells has been shown to be highly responsive to statins, the identification of appropriate subsets of patients will be required to successfully advance these agents as anticancer therapeutics. To this end, there were two major aims to this work: 1) Elucidate a molecular rationale for the observed therapeutic index of statin-induced apoptosis in normal and tumour cells; 2) Identify molecular determinants of sensitivity for statin-induced apoptosis in human tumour cells. To address the first aim we demonstrated that dysregulation of the MVA pathway, achieved by ectopic expression of either full length HMGCR (HMGCR-FL) or its novel splice variant lacking exon 13 (HMGCR-D13), increases transformation. Ectopic HMGCR promotes growth of transformed and non-transformed cells under anchorage-independent conditions or as xenografts in immunocompromised mice. We also show that high mRNA levels of HMGCR and four out of five other MVA pathway genes correlate with poor prognosis in primary breast cancer, suggesting the MVA pathway may play a role in the etiology of human cancers. To address the second aim, we show that dysregulation of the MVA pathway is a key determinant of sensitivity to statin-induced apoptosis in multiple myeloma. In a panel of 17 distinct myeloma cell lines, half were sensitive to statin-induced apoptosis and the remainder were insensitive. Interestingly, in sensitive cells, the classic feedback response to statin exposure is lost, a feature we demonstrated could distinguish a subset of statin-sensitive primary myeloma cells. We further illustrated that statins are highly effective and well tolerated in an orthotopic model of myeloma using cells harboring a dysregulated MVA pathway. Taken together, this work provides a molecular rationale and determinants of sensitivity for statin-induced apoptosis of human tumour cells.

[Formulae and special characters can only be approximated here. Please see the pdf version of the abstract for an accurate reproduction.] Atherosclerosis is a common degenerative disease in which the clinical manifestations are often through stroke or myocardial infarction. Some of the established risk factors for atherosclerosis include elevated plasma low-density lipoprotein (LDL)-cholesterol levels, obesity, diabetes mellitus (DM) and cigarette smoking. Of the risk factors, an elevation in plasma LDL is one of the most established and the most researched. This is partly a consequence of the deposition of cholesterol within arterial intima being a crucial step in the progression of atherosclerosis, combined with the finding that LDL particles are a major transporter of cholesterol in circulation. Recently there is increasing evidence showing a role of the other major transporter of cholesterol in circulation, chylomicron remnants, in the progression of atherosclerosis. The notion of atherosclerosis as a postprandial phenomenon has been further substantiated by the emergence of evidence showing a direct role of chylomicron remnants in arterial cholesterol deposition. Based on evidence that chylomicron remnants are proatherogenic, the suggestion arises that accumulation of postprandial lipoproteins in plasma may add another dimension of risk to the development of coronary artery disease (CAD). This thesis tests the general hypothesis that individuals with or at high risk of CAD have postprandial dyslipidaemia and that this metabolic abnormality is correctable with a class of lipid-lowering drugs called statins. To test the hypothesis, clinical studies were conducted in normolipidaemic CAD patients, heterozygous familial hypercholesterolaemia (FH) and postmenopausal women with type 2 DM. Determination of postprandial dyslipidaemia by comparison with control populations were conducted initially in each patient group (Studies 1, 3 and 5), followed by intervention studies investigating possible modulation of the dyslipidaemia with a statin (Studies 2, 4 and 6). Six observation statements based on case-control comparisons of postprandial lipaemia in patients with or at risk of CAD and the effects of statins on postprandial dyslipidaemia in the patient groups were derived from the general hypothesis. The observation statements were examined in the individual studies described below. Postprandial lipoprotein metabolism was assessed using a number of methods. For comparison of postprandial lipaemia in Studies 1 and 2, a classic oral fat challenge was utilised. As markers of chylomicrons and chylomicron remnants, retinyl palmitate and triglyceride were measured postprandially as well as apolipoprotein (apo) B48 concentrations, a specific marker of intestinal lipoproteins. ApoB48 was also measured in the fasting state and found to predict the postprandial responses of retinyl palmitate, triglyceride and apoB48. This suggested that fasting measurement of apoB48 could be used as a simple indicator of postprandial dyslipidaemia. Consequently for Studies 3 - 6, fasting apoB48 measurements were used as primary markers of postprandial dyslipidaemia. Other markers for chylomicrons and their remnants utilised were fasting plasma concentrations of remnant-like particle-cholesterol (RLP-C) and apoC-III. As well as these static markers, chylomicron remnant catabolism was measured using a stable isotope breath test. The breath test involves the intravenous injection of a chylomicron remnant-like emulsion labelled with ¹³C-oleate and measurement of enriched ¹³CO2 in expired breath by isotope ratio mass spectrometry. The fractional catabolic rate (FCR) of the injected emulsion was subsequently calculated using multi-compartmental modeling (SAAM II). The studies are presented in this thesis as published and unpublished works. In Study 1, postprandial lipoprotein metabolism was compared between 18 normolipidaemic CAD patients (cholesterol 4.54 ± 0.12 mmol/L, triglyceride 1.09 ± 0.16) with 13 asymptomatic healthy controls using an oral fat challenge. Normolipidaemic CAD patients had higher postprandial area-under-curve (AUC) for triglyceride (+34%, p=0.019), retinyl palmitate (+74%, p=0.032) and apoB48 (+36%, p<0.001). Fasting apoB48 was also higher (+41%, p=0.001) and found to correlate significantly with AUC of triglyceride (p=0.017), retinyl palmitate (p=0.001) and apoB48 (p<0.001). The data suggest that normolipidaemic CAD patients have increased concentrations of intestinal lipoproteins in the fasting and postprandial state. In addition to these findings, significant correlations of fasting apoB48 with postprandial markers (p<0.02) suggests the fasting marker to be a simpler surrogate marker for the degree of total postprandial lipaemia. Study 2 investigated the effect of atorvastatin treatment on postprandial dyslipidaemia found in the 18 near-normolipidaemic CAD patients from Study 1. The trial was conducted in a randomised, placebo-controlled design, using oral fat challenges before and after 12-weeks atorvastatin/placebo treatment. Compared with the placebo group, atorvastatin decreased the total postprandial AUC for iii triglyceride (-22%, p=0.05) and apoB48 (-34%, p=0.013). Fasting markers of apoB48 (-35%, p=0.019) and RLP-C (-36%, p=0.032) also decreased significantly. Atorvastatin was also found to increase LDL-receptor activity by +218% (p<0.001) as reflected in binding studies. The data suggest atorvastatin reduces the fasting levels of intestinal lipoproteins as well as total postprandial lipaemia, but without acute dynamic changes in postprandial lipaemia. The reduction in fasting and total postprandial lipoprotein levels could be partly attributed to an increase in LDL-receptor mediated removal from circulation. In Study 3, postprandial lipaemia was compared in 15 heterozygous FH patients with 15 healthy controls. FH patients had higher fasting concentrations of apoB48 (+56%, p<0.001) and RLP-C (+48%, p=0.003). The elevation in these fasting markers of chylomicrons and their remnants suggests FH patients have postprandial dyslipidaemia due to an accumulation of these particles in plasma. Study 4 examined the effects of long- (> 6 months) and short-term (4 weeks) simvastatin treatment on modulating postprandial dyslipidaemia found in the 15 FH patients from Study 3. Short- and long-term simvastatin treatment decreased the fasting concentrations of apoB48 (-29% and 15% respectively, p<0.05) and RLP-C (both -38%, p<0.001), but did not significantly alter the FCR of the injected chylomicron remnant-like emulsion. The data suggest that in heterozygous FH both long- and short-term simvastatin treatments decrease the fasting markers of postprandial lipoproteins by mechanisms that may not be mediated via processes differentiated by the 13CO2 breath test. This implies that the effect on postprandial lipaemia may be from a decrease in production and/or a possible increase in catabolism of triglyceride-rich lipoproteins (TRLs). In Study 5, postprandial lipaemia was compared in 24 postmenopausal women age and body mass index matched with 14 postmenopausal women with type 2 DM. Postmenopausal diabetic women were found to have higher fasting concentrations of apoB48 (+21%, p=0.021) and apoC-III (+16%, p=0.042) as well as lower FCR of the chylomicron remnant-like emulsion (-50%, p<0.001). The data suggest that postmenopausal diabetic women have postprandial dyslipidaemia, and that this is due to delayed catabolism of chylomicron remnants. Study 6 was an hypothesis-generating exercise examining the effects of 4-weeks pravastatin treatment on postprandial dyslipidaemia found in 7 postmenopausal women with type 2 DM from Study 5. Although plasma LDL-cholesterol was reduced (-19%, p=0.028), there were no significant effects found on fasting apoB48 concentrations (-12%, p=0.116) or the FCR of the chylomicron remnant-like emulsion (+38%, p=0.345). A larger sample size of patients and/or treatment with a more potent statin at a dosage known to affect chylomicron remnant metabolism would be required to demonstrate a significant reduction in postprandial dyslipidaemia in postmenopausal women with type 2 DM. The results of the above mentioned studies combined support the general hypothesis that postprandial dyslipidaemia is a feature of patients with or at risk of CAD. This defect may be demonstrated using fasting apoB48 as an indicator of the degree of postprandial lipaemia. Postprandial dyslipidaemia may reflect a reduction in catabolism, as suggested with the breath test in type 2 DM, and/or an over overproduction of chylomicrons. Both these mechanisms would also increase competition for lipolysis and clearance pathways between hepatically and intestinally-derived lipoproteins. The exact mechanisms by which postprandial dyslipidaemia occurs are yet to be determined. Statins appear to improve defective postprandial lipaemia in patients with or at risk of CAD, which is in agreement with the general hypothesis. The effectiveness of a statin is dependant on their potency in inhibiting cholesterol biosynthesis and increasing receptor mediated clearance of LDL and chylomicron remnants. The studies conducted in this thesis show that postprandial dyslipidaemia can be reduced by statins but not to the extent demonstrated in controls. However, the demonstrated reduction in fasting and total postprandial lipaemia translates to a lowering in overall arterial exposure to circulating proatherogenic particles. The elevation in fasting and postprandial levels of proatherogenic chylomicron remnants found in the patient groups described in this thesis indicates another dimension to their risk of coronary disease. The reductions in the overall levels of proatherogenic particles in patients with or at high CAD risk, infers a possible reduction in the risk of coronary disease in these patients.

A eficácia das estatinas em reduzir o risco de eventos coronarianos não é completamente explicada por seus efeitos em diminuir colesterol de lipoproteína de baixa densidade (LDL-C). Um dos seus efeitos adicionais pode ser decorrente da modificação na concentração de lipoproteína de alta densidade (HDL), reconhecida como ateroprotetora, principalmente por seu papel no transporte reverso do colesterol (TRC). Os transportadores de membrana do tipo ATP-binding cassette, ABCA1 e ABCG1, e o scavenger receptor BI (SRBI) são proteínas importantes envolvidas no TRC e seus genes são regulados por vários fatores de transcrição, entre eles os liver-x-receptors (LXRs). Com a finalidade de avaliarmos os efeitos dos hipolipemiantes sobre expressão dos transportadores ABC e do receptor SRBI, a expressão de RNAm do ABCA1, ABCG1, SCARB1, NR1H3 (LXR&#945;) e NR1H2 (LRX&#946;) foi avaliada por PCR em tempo real em células das linhagens HepG2 (origem hepática) e Caco-2 (origem intestinal) tratadas com atorvastatina ou sinvastatina (10 µM) e/ou ezetimiba (até 5 µM) por até 24 horas. Além disso, a expressão desses genes também foi avaliada em células mononucleares do sangue periférico (CMSP) de 50 pacientes normolipidêmicos (NL) e 71 hipercolesterolêmicos (HC) tratados com atorvastatina (10mg/dia/4semanas, n=48) ou sinvastatina e/ou ezetimiba (10mg/dia/4 ou 8 semanas, n=23). A possível associação entre os polimorfismos ABCA1 C-14T e R219K e a expressão de RNAm em CMSP também foi avaliada por PCR-RFLP. O SCARB1 foi o gene mais expresso nas células HepG2 e Caco-2, seguido por NR1H2, NR1H3, ABCG1 e ABCA1 em HepG2 ou por ABCA1 e ABCG1 em Caco-2. O tratamento com estatinas (1 ou 10 µM) ou ezetimiba (5 µM), por 12 ou 24 horas, aumentou a expressão de RNAm do ABCG1, mas não de ABCA1 e SCARB1, em células HepG2. Ainda nesta linhagem, o aumento na transcrição dos genes NR1H2 e NR1H3 foi observado somente com a maior concentração de atorvastatina (10 µM) e, ao contrário, o tratamento com ezetimiba causou redução na transcrição de NR1H2, sem alteração de NR1H3. Em células Caco-2, o tratamento com atorvastatina ou sinvastatina por 12 ou 24 horas reduziu a quantidade do transcrito ABCA1 e não alterou a expressão do SCARB1 e do ABCG1, embora, para este último, tenha havido uma tendência à diminuição da expressão após tratamento com sinvastatina (p=0,07). Após tratamento com ezetimiba isolada (até 5 µM) nenhuma alteração de expressão de RNAm foi observada em células Caco-2; no entanto, após 24 horas de tratamento com sinvastatina e ezetimiba, foi reduzida a taxa de transcrição de ABCA1 e ABCG1, mas não de SCARB1. Ao contrário das linhagens celulares, em CMSP os genes NR1H2 e ABCG1 foram os mais expressos, seguidos pelos genes SCARB1 e ABCA1 e, finalmente, pelo NR1H3. Indivíduos HC tiveram maior expressão basal de NR1H2 e NR1H3, mas não de outros genes, quando comparados aos NL (p<0,05). Além disso, nos indivíduos HC, a expressão basal de ABCA1 foi maior em portadores do alelo -14T do polimorfismo ABCA1 -14C>T quando comparados aos portadores do genótipo -14CC (p=0,034). O tratamento com estatinas, com ezetimiba ou com a terapia combinada diminuiu a transcrição de ABCA1 e ABCG1. Para o SCARB1, NR1H2 e NR1H3, nenhuma alteração de expressão de RNAm em CMSP foi detectada após os tratamentos in vivo. Após todas as fases de tratamento, ABCA1 e ABCG1 e também NR1H2 e NR1H3 foram significativamente correlacionados entre si, mas nenhuma correlação com perfil lipídico sérico foi relevante. Coletivamente, esses resultados dão indícios de que os hipolipemiantes analisados (estatinas e ezetimiba) têm um importante papel na regulação da expressão de genes envolvidos no transporte reverso do colesterol e sugerem a existência de regulação tecido-específica para os dois transportadores ABC. Além disso, o efeito das estatinas ou da ezetimiba sobre a expressão do ABCA1, do ABCG1 ou do SCARB1 não sofreu influencia de alterações diretas da transcrição dos LXRs. / The efficacy of statins in reducing the risk of coronary events is not completely explained by their effects in decreasing cholesterol low-density lipoprotein (LDL-C). One of their additional effects may result from the change in concentration of high-density lipoprotein (HDL), recognized as atheroprotective, mainly for the role in reverse cholesterol transport (RCT). The membrane transporters, as ATP-binding cassette, ABCA1 and ABCG1, and scavenger receptor BI (SRBI) are important proteins involved in the RCT and their genes are regulated by various transcription factors, including the liver-X-receptors (LXRs) . In order to evaluate the effects of lipid lowering on expression of ABC transporters and SRBI receptor, the mRNA expression of ABCA1, ABCG1, SCARB1, NR1H3 (LXR&#945;) and NR1H2 (LRX&#946;) was assessed by real time PCR in HepG2 (hepatic origin) and Caco-2 (intestinal origin) cells treated with atorvastatin or simvastatin (10 µM) and/or ezetimibe (up to 5 µM) for 24 hours. Furthermore, the expression of these genes was evaluated in peripheral blood mononuclear cells (PBMC) of 50 normolipidemic (NL) and 71 hypercholesterolemic (HC) patients treated with atorvastatin (10mg/d/4 weeks, n = 48) or simvastatin and/or ezetimibe (10mg/d/4 or 8 weeks, n = 23). The possible association between ABCA1 C-14T and R219K polymorphisms and mRNA expression in PBMC was also evaluated by PCR-RFLP. SCARB1 was the most expressed in HepG2 and Caco-2 cells, followed by NR1H2, NR1H3, ABCG1 and ABCA1 in HepG2 or by ABCG1 and ABCA1 in Caco-2. The treatment with statins (1 or 10 µM) or ezetimibe (5 µM) for 12 or 24 hours, increased mRNA expression of ABCG1 but not ABCA1 and SCARB1 in HepG2 cells. Moreover, in HepG2 cells, atorvastatin also upregulated NR1H2 and NR1H3 only at 10.0 &#181;M, meanwhile ezetimibe downregulated NR1H2 but did not change NR1H3 expression. In Caco-2 cells, atorvastatin or simvastatin treatment for 12 or 24 hours reduced the amount of ABCA1 transcript and did not alter the ABCG1 and SCARB1 expressions, despite the tendency to decrease ABCG1 mRNA expression after simvastatin treatment (p = 0.07). After treatment with ezetimibe alone (up to 5 &#181;M) no change in mRNA expression was observed in Caco-2 cells; however, after 24 hours- simvastatin and ezetimibe treatments decreased the transcription of ABCA1 and ABCG1, but not of SCARB1. Unlike cell lines, in PBMC, NR1H2 and ABCG1 were the most expressed, followed by SCARB1 and ABCA1 and finally by the NR1H3. HC patients showed higher NR1H2 and NR1H3 basal expressions, but not of other genes, compared to NL (p <0.05). Moreover, in HC individuals, the ABCA1 basal expression was higher in individuals carrying -14T allele of -14C> T polymorphism when compared with -14CC carriers (p = 0.034). Treatment with statins, ezetimibe, or combined therapy downregulated ABCA1 and ABCG1 expression. For SCARB1, NR1H2 and NR1H3, no change in mRNA expression in PBMC was detected after treatments. After all phases of treatment, ABCA1 and ABCG1 as well as NR1H2 and NR1H3 were significantly correlated, but no correlation with serum lipid profile was relevant. Collectively, these results provide evidences that the lipid lowering (statins and ezetimibe) have an important role in mRNA expression regulation of genes involved in reverse cholesterol transport and suggest the existence of tissue-specific regulation for the ABC transporters. Furthermore, the effect of statins or ezetimibe on ABCA1, ABCG1 or SCARB1 expression was not directly influenced by changes of LXR transcription.

Bien que les statines forment la classe d'hypolipidémiants la plus utilisée, une toxicité musculaire a été reportée, pouvant ainsi provoquer l’apparition d’une myopathie. Dans la première partie, nous avons montré chez l’Homme et l’animal que les statines inhibent directement la chaine respiratoire mitochondriale, et induisent la production de radicaux libres dérivés de l’oxygène (RLO), qui active les voies apoptotiques dans les muscles glycolytiques, alors que les muscles oxydatifs ne sont pas atteints. Nous avons ensuite montré in vitro que le stress réducteur peut engendrer une oxydation mitochondriale, pouvant conduire à une activation de la voie de biogenèse mitochondriale. De plus l’augmentation du contenu mitochondrial induite a permis de protéger les cellules contre l’apoptose induite par les statines. Enfin, nous avons montré in vivo que l’induction des voies de biogenèse mitochondriale est nécessaire à la tolérance des statines dans les muscles oxydatifs. En conclusion, le phénotype mitochondrial, tant au niveau quantitatif que qualitatif, semble être un facteur clé dans l’apparition de la myopathie aux statines. / Although statins are the most prescribed class of lipid-lowering agents, adverse muscular toxicity has been reported, which can lead to the appearance of a myopathy. In the first part, we showed in Humans and animals that statins inhibit directly the mitochondrial respiratory chain, and induce the production of reactive oxygen species (ROS), that trigger apoptotic pathways in glycolytic skeletal muscles, whereas oxidative muscles are not impaired. We then showed in vitro that reductive stress can provoke mitochondrial oxidation, that could lead to an activation of mitochondrial biogenesis pathways. Moreover, the consequent increase in mitochondrial content enabled to protect cells against statin-induced apoptosis. Finally, we showed in vivo that the induction of mitochondrial biogenesis is necessary for statin tolerance in oxidative skeletal muscles. In conclusion, mitochondrial phenotype, both quantitatively and qualitatively, seems to be a key factor in the appearance of statin myopathy.