Study of nutritional and genetic factors affecting lipid and glucose metabolism

Abstract

ABSTRACT

In my PhD thesis, I evaluated the impact of nutritional and genetic factors on glucose and lipid metabolism, whose alterations are strongly correlated to pathological conditions causally predisposing to cardiovascular diseases. The thesis is organized into three sections:

PART I: Impact of alcohol consumption on reverse cholesterol transport in mice

Background: epidemiological studies revealed that light to moderate alcohol consumption is linked to a reduced cardiovascular risk. On the contrary, binge alcohol consumption exerts detrimental effects on cardiovascular health1. The protective effect of moderate alcohol consumption seems to be related to a positive modulation of plasma lipoprotein profile2. In a recent study, Liu and colleagues demonstrated that a daily moderate alcohol intake causes an amelioration of lipid profile and reduces atherosclerotic plaque volume in Apo E -/- mice3. Atherosclerotic cardiovascular disease is inversely correlated with reverse cholesterol (RCT), the process promoting the removal of excess cholesterol from arterial wall4. Therefore, aim of my work was to evaluate whether moderate and binge alcohol consumption may differently impact RCT in an animal model of atherosclerosis-prone mice. Subsequently, in vitro investigations have been performed, to better clarify the mechanisms underlying the effect of alcohol consumption on the single RCT steps. Methods: RCT in vivo was measured through a standardized, radiolabelled technique in apolipoprotein E knockout mice: placebo group (n=9) received water, mimicking the abstainers; moderate group (n=10) received 0.8g/kg alcohol/day for 28 days, mimicking a moderate alcohol consumption; binge group (n=10) received 0.8g/kg alcohol/day for 5 days, followed by the administration of 2.8g/kg alcohol/day for 2 days/week, mimicking a binge alcohol consumption. Total cholesterol, HDL-C and triglycerides were measured through the Aries system for Clinical Chemistry (ILab International). Non HDL-C was calculated as the difference between total and HDL cholesterol. The cholesterol efflux capacity (CEC) of mouse sera was measured through a radiolabelled technique, by using murine peritoneal macrophages (MPM) as cell model, to emulate the conditions of the RCT in vivo. The direct effect of alcohol and acetaldehyde on cholesterol efflux was evaluated through a radiolabelled technique, using MPM as cell model. The gene expression of hepatic Sr-bi gene was performed by qPCR analysis. Concomitantly, the protein expression of SR-BI in liver was evaluated through western blotting and immunofluorescence assay. Results and conclusion: the results of RCT in vivo showed that moderate alcohol consumption slightly improved cholesterol mobilization along RCT pathway. In particular, it has been observed an increased radioactivity in livers of the moderate group compared to placebo group (4.2 % ± 0.8 vs 2.4 % ± 0.7, in moderate and placebo group respectively). Interestingly, binge group showed a reduced faecal cholesterol excretion at 48 hours compared to placebo group (0.89% ± 0.16 vs 1.17% ± 0.34, in binge and placebo group respectively). Moderate consumption did not produce any changes in plasma circulating lipoproteins. On the contrary, binge group evidenced an increase in total cholesterol, HDL-C and non HDL-C plasma levels compared to placebo (all p<0.0001 vs placebo). In vitro assays showed no difference in the capacity of sera of alcohol-treated mice to promote cholesterol efflux from cholesterol and acetylated LDL-enriched macrophages. Furthermore, MPM incubation with ethanol and acetaldehyde at concentrations comparable to the plasma concentration after moderate and heavy alcohol consumption did not impact the cholesterol efflux. The results of gene expression showed that binge group had a higher Sr-bi expression compared to placebo (p<0.05), whereas no difference in Sr-bi gene expression was observed between moderate and placebo groups. On the contrary, neither western blotting analysis nor immunofluorescence assay displayed statistically significant differences in hepatic SR-BI protein expression, although a tendency to increased expression was observed both in moderate and binge groups compared to placebo. Overall, these results demonstrated that binge alcohol consumption determines detrimental effects on lipoprotein profile and a slight reduction of faecal cholesterol excretion. On the contrary, moderate alcohol consumption causes a slight improvement of cholesterol mobilization along RCT pathway, exerting no detrimental effect on circulating lipids.

PART II: Role of sphingosine-1 phosphate (S1P) and its receptor S1P3 on atherogenic process of reverse cholesterol transport (RCT)

Background: in the last decades, different clinical trials failed in demonstrating clinical benefits of HDL-C increasing drugs in the reduction of cardiovascular risk5,6,7. These evidences suggested to focus the attention on HDL functions and HDL components. In fact, HDL exert its anti-atherogenic function by actively participating as main actors in RCT process. The evaluation of the entire cholesterol mobilization along RCT pathway and of cholesterol efflux allows to better evaluate atherosclerotic disease risk rather than the measure of HDL-C plasma levels. Among the various components of HDL, S1P is a bioactive sphingolipid, which exerts several anti-atherogenic functions in cardiovascular system, even though the mechanisms underlying them are partially unknown. The aim of this study was to demonstrate that the anti-atherogenic activity of S1P was related to the modulation of lipid metabolism. In details, we aimed to evaluate the role of macrophage S1P3 receptor, on the RCT process. Methods: to this purpose, S1P3 Lys-Cre mice, as macrophage S1P3 overexpressing mice and C57BL/6 mice, as control mice, were used. Cholesterol efflux from CTRL MPM and S1P3-Lys Cre MPM was evaluated through a radiolabelled technique. To evaluate the contribution of the different pathways involved in cholesterol efflux, MPM were treated with or without cAMP (0.3 mM) and with or without the combination of 22 hydroxycholesterol (5 μg/ml) and 9 cis retinoic acid (10μM) during the equilibration period. Cholesterol efflux was promoted to apolipoprotein AI (10 ug/ml), HDL (12.5 μg/ml) and different mouse plasma concentrations (0.1%, 2% v/v). Additionally, to evaluate a molecular link between S1P3 and ABCG1, TY52156 (10 μM), a specific S1P3 inhibitor was used all along cholesterol efflux experiment. MPM Abcg1 gene expression was measured through a real-time PCR assay. The protein expression of ABCG1 in C57BL/6 and S1P3 overexpressing macrophages was evaluated through Western blotting analysis. RCT in vivo was carried out in C57BL/6 mice receiving MPM from C57BL/6 mice (CTRL group), and S1P3 overexpressing MPM (S1P3 group) and measured through a radiolabelled technique. Results and conclusion: in acetylated LDL-loaded MPM, S1P3-Lys Cre MPM cholesterol efflux was higher compared to CTRL MPM, both to HDL (8.47%±0.63 vs 5.93%±0.46; p<0.001) and to mouse plasma (8.04% ±0.43 vs 10.49 ±1.2 and 22.99% ± 1.2 vs 37.09%± 5.43, to 0.5% and 2% mouse plasma concentrations, respectively). Furthermore, the stimulation with 22-hydroxycholesterol and 9-cys retinoic acid, an LXR/RXR agonist, caused an increased cholesterol efflux in S1P3-Lys Cre MPM compared to CTRL MPM both to HDL (7.45%± 1.36 vs 5.31%±0.37) and to mouse plasma (6.17%±1.17 vs 3.76%±0.46). Conversely, no differences in cholesterol efflux were observed between CTRL MPM and S1P3-Lys Cre MPM after cAMP stimulation, a specific ABCA1 transporter stimulator. Collectively, these results suggested a higher cholesterol efflux in S1P3 overexpressing macrophages mediated by ABCG1 transporter. The causal link between the macrophage S1P3 overexpression and the higher ABCG1-mediated cholesterol efflux was confirmed by the results of cholesterol efflux in presence of TY52156. In fact, in presence of this S1P3 inhibitor, S1P3-MPM had a lower cholesterol efflux compared to the same condition without TY52156 (2.20% ±0.14 vs 2.85%± 0.17 in basal condition; 2.95%± 0.64 vs 3.49%±.0.36 in the condition with acetylated LDL). Consistent with these results, S1P3-Lys Cre MPM showed an increased protein and gene expression of ABCG1, compared to CTRL MPM. RCT in vivo results evidenced an increased cholesterol mobilization in all the compartments of the RCT pathway in S1P3 group in comparison with CTRL group, as demonstrated by the higher radioactivity percentage found in plasma (0.99%±0.32 vs. 0.60%±0.12; p<0.05), liver (2.66%±0.41 vs 1.99%±0.35; p<0.01) and faeces (0.99%±0.19 vs 0.66%±0.10; p<0.01) in these mice. In conclusion, in the present work, we demonstrated that the endogenous S1P signalling amplification mediated by macrophage S1P3 overexpression improves the cholesterol excretion via RCT pathway. Overall, these findings evidence that the antiatherogenic function of endogenous S1P is also due to the modulation of lipid metabolism, through the interaction with the macrophage S1P3 receptor

PART III: S1P/S1P1 axis and its role in insulin resistance

Background: numerous experimental studies evidenced that macrophages actively contribute to the development of chronic inflammation and insulin resistance (IR), two pathological conditions strongly related to type 2 Diabetes mellitus (T2DM)5. In addition to its well-defined anti-atherogenic function, HDL also exerts insulin-sensitive and anti-diabetic effects. HDL are complex macromolecules, carriers of lysosphingolipids such as sphingosine-1-phosphate (S1P), which has been demonstrated to directly contribute to the anti-inflammatory HDL activity. Differently from atherosclerosis, few studies investigated the role of HDL-S1P in the development of insulin resistance. Therefore, aim of my work was to investigate whether S1P/S1P1 axis amplification may attenuate the development of insulin resistance. In addition, it has been evaluated how the amplification of endogenous S1P signalling may impact on macrophage polarization. Methods: in vitro and in vivo studies were conducted in the Center for Laboratory Medicine at Universitatsklinikum of Munster (UKM) in collaboration with University of Leipzig. S1P1 Lys-Cre mice (n=8), characterized by a specific S1P1 overexpression in macrophages, and control mice (CTRL) (n=8) were fed with high-fat diet for sixteen weeks, to induce insulin resistance. Body weight was monitored during all the treatment. Glucose tolerance test (GTT) and insulin tolerance test (ITT) were performed one week before the end of the treatment, administrating intraperitoneally 1g/kg of glucose and 1U/kg of insulin, respectively. Glycated haemoglobin was quantified in plasma by a turbidimetric assay at the end of the treatment, while epididymis fat, used as adiposity index, was weighted after mice sacrifice together with ex vivo analysis. In detail, one part of murine peritoneal macrophages (MPM), withdrawn after mice sacrifice, was immediately processed for FACs analysis, to evaluate the expression of specific membrane receptors (Dectin-I, F4/80, MHC-II) and transcriptional factors (PU.1, IRF-8). Another part of MPM was seeded into 48-well plates, for the gene expression evaluation (Klf4, MafB, Blc6, Arg1, Irf-8, Pu.1), while the remaining macrophages were used for cytokines production quantification (IL-1RA and IL-10) by performing ELISA kits. Results and conclusion: S1P1-Lys Cre mice showed a reduced body weight gain during all the treatment in comparison with CTRL mice. These results were confirmed by an increased epididymis fat weight observed in S1P1-Lys Cre mice compared to CTRL mice (p<0.05). In addition, glycated haemoglobin analysis in plasma demonstrated higher levels of this haemoglobin form in S1P1-Lys Cre mice respect to CTRL mice (p<0.05), to prove a better glucose homeostasis in S1P1 overexpressing mice. The two functional tests, GTT and ITT, showed a reduced insulin resistance in S1P1-Lys Cre mice. GTT results showed an enhancement of glycaemia after 15 and 30 minutes from glucose injection more pronounced in CTRL mice than in S1P1-Lys Cre mice. Moreover, in CTRL mice, we also observed a slow decrease of glycaemia in 90 minutes following the injection. Consistent with these results, S1P1 Lys-Cre mice showed an increased reduction of glucose plasma levels, in response to insulin injection, in comparison with CTRL mice, to demonstrate an insulin dependent amelioration of glucose uptake in S1P1-Lys Cre mice. The characterization of macrophage polarization clearly revealed a M2 anti-inflammatory phenotype induced by S1P1 overexpression, as demonstrated by the results of qPCR and FACs that showed a higher expression of PU.1 and IRF-8, two transcriptional factors linked to M2-phenotype. A further demonstration of M2 phenotype polarization induced by S1P1 overexpression has been also provided by the higher expression of specific genes, upregulated by PU.1 and IRF-8 and related to M2 phenotype polarization (KLF-4, MafB, Bcl6 and Arg1) in S1P1 overexpressing macrophages. These results were confirmed by a more numerous macrophage population positive to dectin I and MHC-II, membrane receptors characterizing M2 phenotype, in S1P1-Lys Cre mice in comparison with the other group. Finally, the anti-inflammatory phenotype characterizing S1P1 overexpressing macrophages was also confirmed by the increased production of two anti-inflammatory cytokines (IL-1 RA and IL-10) from S1P1 overexpressing macrophages. Collectively, these results evidence that the amplification of S1P/S1P1 axis in macrophages reduces insulin resistance development. This effect may be caused by the anti-inflammatory function of S1P1 overexpressing macrophages, characterized by a defined M2 phenotype.

Bibliography: 1. Fernandez-Sola, J. Cardiovascular risks and benefits of moderate and heavy alcohol consumption. Nat. Rev. Cardiol. 12, 576–587 (2015). 2. Brinton, E. A. Effects of ethanol intake on lipoproteins. Curr. Atheroscler. Rep. 14, 108–114 (2012). 3. Liu, W., Redmond, E. M., Morrow, D. & Cullen, J. P. Differential effects of daily-moderate versus weekend-binge alcohol consumption on atherosclerotic plaque development in mice. Atherosclerosis 219, 448–54 (2011). 4. Rader, D. J., Alexander, E. T., Weibel, G. L., Billheimer, J. & Rothblat, G. H. The role of reverse cholesterol transport in animals and humans and relationship to atherosclerosis. J. Lipid Res. 50 Suppl, S189-94 (2009). 5. McArdle, M. A., Finucane, O. M., Connaughton, R. M., McMorrow, A. M. & Roche, H. M. Mechanisms of obesity-induced inflammation and insulin resistance: insights into the emerging role of nutritional strategies. Front. Endocrinol. (Lausanne). 4, 52 (2013). 6. Schwartz, G. G. et al. Effects of Dalcetrapib in Patients with a Recent Acute Coronary Syndrome. N. Engl. J. Med. 367, 2089–2099 (2012). 7. Rimm, E. B. et al. Effects of Torcetrapib in Patients at High Risk for Coronary Events. N. Engl. J. Med. 367, 1387–1396 (2012).