Individuals with abdominal obesity are at higher risk for developing type 2 diabetes, predisposing cardiovascular events and insulin resistance. Low glycemic index (GI) diets may be beneficial in the management of insulin resistance. Insulin resistance is associated with increased intramyocellular lipid (IMCL) content as measured by proton nuclear magnetic resonance spectroscopy (1H-MRS). The primary objective of this thesis was to determine whether a low GI diet can improve insulin sensitivity by reducing IMCL of skeletal muscle. One hundred and twenty-one male and female participants aged 30 to 70 years (mean+SD, 53+10)) with abdominal obesity, entered a 4 to 6 week weight-maintaining, low-fat dietary advice run-in phase. Of the 121 eligible participants, 95 completed the run-in phase and were randomly assigned to either a low-GI (LGID, n=48) or high-GI diet (HGID, n=47) for 24 weeks. Participants underwent a 75g oral glucose tolerance test (OGTT) and had soleus-muscle IMCL measured by 1H-MRS at the beginning and end of the intervention period. Insulin sensitivity was assessed by the homeostatic model assessment index (HOMA) and the insulinogenic index (ISI) was calculated for insulin secretion. At the end of the run-in phase, there were significant reductions in serum total-, LDL-, and HDL-cholesterol (all, p<0.0001) and an increase in fasting plasma glucose (p<0.05). In 57 participants who wore a continuous glucose monitoring system for 24 hours during the run-in period, a total of 30% (p<0.001) of the variation in the incremental area under the blood glucose curve after self-selected breakfast meals was explained by GI. After 24 weeks, diet GI was significantly lower in the LGID than HGID group (55.5+3.1 vs 63.9+3.1, p<0.0001). Plasma glucose 60 minutes after the OGTT was significantly lower on the LGID than at baseline (p<0.05) and there was a non-significant trend towards an increase in ISI (p=0.07). On the HGID, ISI increased significantly from baseline (p<0.01). It is concluded that the LGID reduced 60 minute plasma glucose but did not significantly affect IMCL or insulin sensitivity in individuals with abdominal obesity.

Individuals with abdominal obesity are at higher risk for developing type 2 diabetes, predisposing cardiovascular events and insulin resistance. Low glycemic index (GI) diets may be beneficial in the management of insulin resistance. Insulin resistance is associated with increased intramyocellular lipid (IMCL) content as measured by proton nuclear magnetic resonance spectroscopy (1H-MRS). The primary objective of this thesis was to determine whether a low GI diet can improve insulin sensitivity by reducing IMCL of skeletal muscle. One hundred and twenty-one male and female participants aged 30 to 70 years (mean+SD, 53+10)) with abdominal obesity, entered a 4 to 6 week weight-maintaining, low-fat dietary advice run-in phase. Of the 121 eligible participants, 95 completed the run-in phase and were randomly assigned to either a low-GI (LGID, n=48) or high-GI diet (HGID, n=47) for 24 weeks. Participants underwent a 75g oral glucose tolerance test (OGTT) and had soleus-muscle IMCL measured by 1H-MRS at the beginning and end of the intervention period. Insulin sensitivity was assessed by the homeostatic model assessment index (HOMA) and the insulinogenic index (ISI) was calculated for insulin secretion. At the end of the run-in phase, there were significant reductions in serum total-, LDL-, and HDL-cholesterol (all, p<0.0001) and an increase in fasting plasma glucose (p<0.05). In 57 participants who wore a continuous glucose monitoring system for 24 hours during the run-in period, a total of 30% (p<0.001) of the variation in the incremental area under the blood glucose curve after self-selected breakfast meals was explained by GI. After 24 weeks, diet GI was significantly lower in the LGID than HGID group (55.5+3.1 vs 63.9+3.1, p<0.0001). Plasma glucose 60 minutes after the OGTT was significantly lower on the LGID than at baseline (p<0.05) and there was a non-significant trend towards an increase in ISI (p=0.07). On the HGID, ISI increased significantly from baseline (p<0.01). It is concluded that the LGID reduced 60 minute plasma glucose but did not significantly affect IMCL or insulin sensitivity in individuals with abdominal obesity.

The prevalence of insulin resistance and type 2 diabetes mellitus (T2DM) in obese children and adolescents is growing at an alarming rate, especially in ethnic minorities. It is not clear whether young people of different ethnic backgrounds vary in their metabolic response to excessive adiposity. Differences in lipid partitioning in the abdominal fat compartments have been observed among different ethnic groups. The aim of this study was to evaluate whether there are ethnic differences in intramyocellular lipid (IMCL) levels that are related to differences in insulin sensitivity. Eighty-two obese children and adolescents underwent 1) 1H nuclear magnetic resonance (NMR) spectroscopy to non-invasively quantify IMCL levels in their soleus muscle, 2) an oral glucose tolerance test and (in a subset of subjects) a euglycemic-hyperinsulinemic clamp to assess insulin sensitivity, 3) a dual-energy X-ray absorptiometry (DEXA) scan to measure total percent body fat, and 4) magnetic resonance imaging to measure abdominal fat distribution. IMCL levels in Hispanic children and adolescents (1.50 ± 0.64%) were significantly greater than in their Caucasian (1.19 ± 0.40%) and African-American (1.09 ± 0.49%) peers. Visceral fat was significantly lower in African Americans (42.7 ± 18.8cm2) and were similar in Caucasians (70.9 ± 27.5cm2) and Hispanics (77.3 ± 41.9cm2). The three groups were not different with respect to insulin sensitivity. For the entire cohort, IMCL levels were inversely related to insulin sensitivity. There was a significant correlation between visceral fat and insulin resistance in Hispanics and Caucasians but not in African Americans. In conclusion, these data suggest that there are significant ethnic differences in lipid partitioning in both the muscle and abdominal compartment. These findings may explain ethnic differences in insulin sensitivity and further the understanding of the pathogenesis of insulin resistance and T2DM.

A resistência à insulina está associada com o aumento do teor de gordura intramiocelular (GIMC) e com níveis séricos da adiponectina (ADP) diminuídos. A ADP por sua vez está envolvida na oxidação de gordura muscular. Entretanto, a relação entre ambas continua controversa. O objetivo deste estudo é explorar a relação entre a ADP e a GIMC em adultos não diabéticos, além de estudar o papel da rosiglitasona (RSG) sobre a distribuição da gordura entre os compartimentos musculares. Este estudo compreende duas fases: uma fase transversal (corte-transversal) e uma fase longitudinal, de intervenção terapêutica com uma droga, num desenho aberto. Laboratório de Pesquisas Clínicas e Experimentais em Biologia Vascular (Biovasc) - UERJ. Material e métodos Na fase transversal, 24 pacientes obesos, não diabéticos, com síndrome metabólica (SM) e 9 controles magros e saudáveis foram estudados. Foi realizada a Espectroscopia de Prótons por Ressonância Nuclear Magnética (1H-ERNM) para quantificar a gordura extramiocelular (GEMC) e a GIMC. Estas, associadas à ADP e aos parâmetros antropométricos e bioquímicos, foram avaliadas e comparadas nos dois grupos. Durante a fase longitudinal, 15 destes pacientes foram reestudados, através da 1H-ERNM, após o tratamento com RSG por 6 meses. Da mesma forma, as variáveis antropométricas e metabólicas foram reavaliadas. Fase transversal: os pacientes com SM apresentaram maior índice de massa corporal (IMC), cintura abdominal, relação cintura-quadril (RCQ), e níveis de glicemia, insulina e triglicerídeos e menores níveis de HDL-c, quando comparados com o grupo controle. Da mesma forma o HOMA-RI [3.25 (2.58-4.13) vs 1.02 (0.73-1.29); p<0.0001] e a GIMC [266.1 (189.9-296.3) vs 72.85 (55.3-109.4) unidades arbitrárias-UA, p<0.0001] estavam aumentados enquanto o QUICKI [0.32 (0.31-0.33) vs 0.38 (0.37-0.40); p<0.0001] e a ADP [8.6 (4.05-15.95) vs 21.1 (12.9-24.4) &#956;g/ml; p=0.02) estavam diminuídos. O teor de GIMC associou-se diretamente com a glicose, insulina, triglicerídeos e HOMA-RI e inversamente com o HDL-c, QUICKI e, mais importantemente, com a ADP (r = -0.41; p<0.05). Fase longitudinal: após o tratamento com RSG, o peso corporal e a circunferência do quadril aumentaram, respectivamente [100.9 (91.12-138.7) vs 107,0 (79.6-142.8) kg e 118 (107-126) cm vs 122 (110-131) cm]; enquanto a RCQ diminuiu [0.93 (0.87-1.00) vs 0.89 (0.82-0.97); P<0.001 para todos]. Adicionalmente, a glicemia, a insulina e o HOMA-RI diminuíram significativamente, enquanto a ADP aumentou mais de 3 vezes [9.7 (3.7-17.7) vs 38.0 (19.3-42.4) &#956;g/ml]. Finalmente, a GIMC não se modificou [267.54 (213.94-297.94) vs 305.75 (230.80-424.75) UA], mas a GEMC aumentou de forma significativa [275.53 (210.39-436.66) vs 411.39 (279.92-556.59) UA; P<0.01] diminuindo a razão GIMC sobre GEMC [GIMC/GEMC; 1.07 (0.78-1.23) vs. 0.71 (0.53-0.96); p<0.01]. A ADP correlacionou-se inversamente com o teor da GIMC em adultos obesos não diabéticos com SM. Este achado tem possíveis implicações para o papel da ADP na oxidação da gordura muscular, na RI e na SM. O tratamento com RSG aumentou a massa corporal e a circunferência do quadril e diminuiu a RCQ. Além disso, diminuiu a razão GIMC/GEMC, por aumentar a GEMC sem alterar significativamente a GIMC. Isto sugere que este medicamento pode prevenir a deposição da gordura no compartimento intramiocelular ao aumentar os depósitos periféricos e o extramiocelular. / Insulin resistance (IR) is associated with intramyocellular lipid (IMCL) content and low serum adiponectin (ADP) levels. ADP is also involved in muscle fat oxidation but the relationship between them is still controversial. We aimed to further explore the relationship between ADP and IMCL content in non-diabetic adults and the role of rosiglitazone (RSG) in muscle fat compartment distribution in an adult population of obese nondiabetic metabolic syndrome patients. This study comprises two phases: a cross-sectional and a longitudinal, open-label, drug-interventional one. Laboratory for Clinical and Experimental Research on Vascular Biology (Biovasc) at the State University of Rio de Janeiro. During the cross-sectional phase, 24 obese, nondiabetic patients with metabolic syndrome (MS) and 9 lean healthy controls were studied. Proton nuclear magnetic resonance spectroscopy (1H-NMRS) was performed to quantify IMCL, as well as extramyocellular lipid (EMCL) content. The latter plus serum ADP, anthropometrics and biochemical parameters were evaluated and compared in these two groups. During the longitudinal phase, fifteen of the MS patients were studied by means of 1HNMRS before and after treatment with 8mg/day of RSG for 6 months. Anthropometrical and metabolic variables were assessed. Measurements and main results cross-sectional phase: MS patients had higher body mass index (BMI), waist, waist-to-hip ratio (WHR), glucose, insulin and triglycerides and lower HDL-c as compared to controls. HOMA-IR (3.25 [2.58-4.13] vs 1.02 [0.73-1.29]; p<0.0001) and IMCL content (266.1 [189.9-296.3] vs 72.85 [55.3-109.4) AU, p<0.0001] were higher, and QUICKI (0.32 [0.31-0.33] vs 0.38 [0.37-0.40]; p<0.0001) and ADP (8.6 [4.05-15.95] vs 21.1 [12.9-24.4] &#956;g/ml; p=0.02) lower in MS compared to controls. IMCL content was directly associated with glucose, insulin, triglycerides and HOMAxiii IR and inversely to HDLc, QUICKI and, more importantly, with ADP (r = -0.41; p<0.05). Longitudinal phase: After RSG treatment, body weight and hip circumference increased [100.9 (91.12-138.7) vs 107,0 (79.6-142.8) kg and 118 (107-126) cm vs 122 (110-131) cm] respectively, while WHR decreased [0.93 (0.87-1.00) vs 0.89 (0.82-0.97); P<0.001 for all]. Additionally, fasting plasma glucose, insulin and HOMA-IR significantly decreased while adiponectin increased over 3 fold [9.7 (3.7-17.7) vs 38.0 (19.3-42.4) &#956;g/ml]. Finally, IMCL did not change [267.54 (213.94-297.94) vs 305.75 (230.80-424.75) arbitrary units (AU)] while EMCL increased [275.53 (210.39-436.66) vs 411.39 (279.92-556.59) AU; P<0.01] therefore decreasing IMCL to EMCL ratio (IMCL/EMCL) [1.07 (0.78-1.23) vs. 0.71 (0.53-0.96); p<0.01]. ADP is inversely related to IMCL content in non-diabetic adults. This finding has possible implications for the role of ADP in muscle fat oxidation, IR and MS. RSG treatment increased body weight and hip circumference decreasing WHR and decreased IMCL/EMCL ratio by increasing EMCL without any significant change on IMCL, thus suggesting that this drug may prevent IMCL fat deposition by increasing EMCL and peripheral deposits.

A resistência à insulina está associada com o aumento do teor de gordura intramiocelular (GIMC) e com níveis séricos da adiponectina (ADP) diminuídos. A ADP por sua vez está envolvida na oxidação de gordura muscular. Entretanto, a relação entre ambas continua controversa. O objetivo deste estudo é explorar a relação entre a ADP e a GIMC em adultos não diabéticos, além de estudar o papel da rosiglitasona (RSG) sobre a distribuição da gordura entre os compartimentos musculares. Este estudo compreende duas fases: uma fase transversal (corte-transversal) e uma fase longitudinal, de intervenção terapêutica com uma droga, num desenho aberto. Laboratório de Pesquisas Clínicas e Experimentais em Biologia Vascular (Biovasc) - UERJ. Material e métodos Na fase transversal, 24 pacientes obesos, não diabéticos, com síndrome metabólica (SM) e 9 controles magros e saudáveis foram estudados. Foi realizada a Espectroscopia de Prótons por Ressonância Nuclear Magnética (1H-ERNM) para quantificar a gordura extramiocelular (GEMC) e a GIMC. Estas, associadas à ADP e aos parâmetros antropométricos e bioquímicos, foram avaliadas e comparadas nos dois grupos. Durante a fase longitudinal, 15 destes pacientes foram reestudados, através da 1H-ERNM, após o tratamento com RSG por 6 meses. Da mesma forma, as variáveis antropométricas e metabólicas foram reavaliadas. Fase transversal: os pacientes com SM apresentaram maior índice de massa corporal (IMC), cintura abdominal, relação cintura-quadril (RCQ), e níveis de glicemia, insulina e triglicerídeos e menores níveis de HDL-c, quando comparados com o grupo controle. Da mesma forma o HOMA-RI [3.25 (2.58-4.13) vs 1.02 (0.73-1.29); p<0.0001] e a GIMC [266.1 (189.9-296.3) vs 72.85 (55.3-109.4) unidades arbitrárias-UA, p<0.0001] estavam aumentados enquanto o QUICKI [0.32 (0.31-0.33) vs 0.38 (0.37-0.40); p<0.0001] e a ADP [8.6 (4.05-15.95) vs 21.1 (12.9-24.4) &#956;g/ml; p=0.02) estavam diminuídos. O teor de GIMC associou-se diretamente com a glicose, insulina, triglicerídeos e HOMA-RI e inversamente com o HDL-c, QUICKI e, mais importantemente, com a ADP (r = -0.41; p<0.05). Fase longitudinal: após o tratamento com RSG, o peso corporal e a circunferência do quadril aumentaram, respectivamente [100.9 (91.12-138.7) vs 107,0 (79.6-142.8) kg e 118 (107-126) cm vs 122 (110-131) cm]; enquanto a RCQ diminuiu [0.93 (0.87-1.00) vs 0.89 (0.82-0.97); P<0.001 para todos]. Adicionalmente, a glicemia, a insulina e o HOMA-RI diminuíram significativamente, enquanto a ADP aumentou mais de 3 vezes [9.7 (3.7-17.7) vs 38.0 (19.3-42.4) &#956;g/ml]. Finalmente, a GIMC não se modificou [267.54 (213.94-297.94) vs 305.75 (230.80-424.75) UA], mas a GEMC aumentou de forma significativa [275.53 (210.39-436.66) vs 411.39 (279.92-556.59) UA; P<0.01] diminuindo a razão GIMC sobre GEMC [GIMC/GEMC; 1.07 (0.78-1.23) vs. 0.71 (0.53-0.96); p<0.01]. A ADP correlacionou-se inversamente com o teor da GIMC em adultos obesos não diabéticos com SM. Este achado tem possíveis implicações para o papel da ADP na oxidação da gordura muscular, na RI e na SM. O tratamento com RSG aumentou a massa corporal e a circunferência do quadril e diminuiu a RCQ. Além disso, diminuiu a razão GIMC/GEMC, por aumentar a GEMC sem alterar significativamente a GIMC. Isto sugere que este medicamento pode prevenir a deposição da gordura no compartimento intramiocelular ao aumentar os depósitos periféricos e o extramiocelular. / Insulin resistance (IR) is associated with intramyocellular lipid (IMCL) content and low serum adiponectin (ADP) levels. ADP is also involved in muscle fat oxidation but the relationship between them is still controversial. We aimed to further explore the relationship between ADP and IMCL content in non-diabetic adults and the role of rosiglitazone (RSG) in muscle fat compartment distribution in an adult population of obese nondiabetic metabolic syndrome patients. This study comprises two phases: a cross-sectional and a longitudinal, open-label, drug-interventional one. Laboratory for Clinical and Experimental Research on Vascular Biology (Biovasc) at the State University of Rio de Janeiro. During the cross-sectional phase, 24 obese, nondiabetic patients with metabolic syndrome (MS) and 9 lean healthy controls were studied. Proton nuclear magnetic resonance spectroscopy (1H-NMRS) was performed to quantify IMCL, as well as extramyocellular lipid (EMCL) content. The latter plus serum ADP, anthropometrics and biochemical parameters were evaluated and compared in these two groups. During the longitudinal phase, fifteen of the MS patients were studied by means of 1HNMRS before and after treatment with 8mg/day of RSG for 6 months. Anthropometrical and metabolic variables were assessed. Measurements and main results cross-sectional phase: MS patients had higher body mass index (BMI), waist, waist-to-hip ratio (WHR), glucose, insulin and triglycerides and lower HDL-c as compared to controls. HOMA-IR (3.25 [2.58-4.13] vs 1.02 [0.73-1.29]; p<0.0001) and IMCL content (266.1 [189.9-296.3] vs 72.85 [55.3-109.4) AU, p<0.0001] were higher, and QUICKI (0.32 [0.31-0.33] vs 0.38 [0.37-0.40]; p<0.0001) and ADP (8.6 [4.05-15.95] vs 21.1 [12.9-24.4] &#956;g/ml; p=0.02) lower in MS compared to controls. IMCL content was directly associated with glucose, insulin, triglycerides and HOMAxiii IR and inversely to HDLc, QUICKI and, more importantly, with ADP (r = -0.41; p<0.05). Longitudinal phase: After RSG treatment, body weight and hip circumference increased [100.9 (91.12-138.7) vs 107,0 (79.6-142.8) kg and 118 (107-126) cm vs 122 (110-131) cm] respectively, while WHR decreased [0.93 (0.87-1.00) vs 0.89 (0.82-0.97); P<0.001 for all]. Additionally, fasting plasma glucose, insulin and HOMA-IR significantly decreased while adiponectin increased over 3 fold [9.7 (3.7-17.7) vs 38.0 (19.3-42.4) &#956;g/ml]. Finally, IMCL did not change [267.54 (213.94-297.94) vs 305.75 (230.80-424.75) arbitrary units (AU)] while EMCL increased [275.53 (210.39-436.66) vs 411.39 (279.92-556.59) AU; P<0.01] therefore decreasing IMCL to EMCL ratio (IMCL/EMCL) [1.07 (0.78-1.23) vs. 0.71 (0.53-0.96); p<0.01]. ADP is inversely related to IMCL content in non-diabetic adults. This finding has possible implications for the role of ADP in muscle fat oxidation, IR and MS. RSG treatment increased body weight and hip circumference decreasing WHR and decreased IMCL/EMCL ratio by increasing EMCL without any significant change on IMCL, thus suggesting that this drug may prevent IMCL fat deposition by increasing EMCL and peripheral deposits.

Abstract Metabolic Syndrome (MetSyn) is a medical condition prevalent in Australia. MetSyn is diagnosed with a varying combination of visceral obesity, insulin resistance/ impaired glucose tolerance/ Type 2 diabetes, dyslipidaemia and hypertension. Obesity is a central feature of this syndrome that is characterised by abnormalities in glucose and lipid metabolism. An understanding of the cause of the metabolic derangement that occurs in obesity, and that contributes to MetSyn, would allow effective treatment and prevention strategies to be formulated. This is a priority in the current environment of highly prevalent overweight and obesity in Australian children and adults. Lipotoxicity of insulin-dependent tissues and ectopic fat depots are emerging as fundamental processes in the pathogenesis of MetSyn. Lifestyle intervention, such as increased physical activity, show great promise as agents for disrupting the disease progression and may act via direct or indirect mechanisms on the underlying pathology of MetSyn. This study aimed to determine if diagnostic markers of MetSyn exist in obese, prepubertal, Australian children and to assess the contribution of lifestyle factors on components of MetSyn. Further, this study sought to investigate the relationship between body fat patterning (total body fat, abdominal adipose depots, skeletal intramyocellular lipids, intrahepatocellular lipids) and markers of MetSyn. An experimental intervention was then employed to examine the effect of physical activity on body fat distribution, insulin sensitivity, and haemodynamic and biochemical markers of MetSyn, and additionally to determine if the effect of exercise on parameters of MetSyn was mediated by a change in body fat patterning. Data were collected in a group of 15 obese (mean BMI Z-score 2.51 ± 0.49), prepubertal children (6 male, 9 female) aged 5.1 – 11.4 years (mean age 7.82 yrs ± 1.83). Measures included insulin sensitivity, blood biochemistry (lipid, haemostatic and adipocyte activity markers), blood pressure, two-compartment body composition by hydrometry, and nuclear magnetic resonance scanning for abdominal adipose depots, intrahepatic lipids and skeletal intramyocellular lipids. Each child’s habitual nutrition and physical activity were also ascertained using multiple-pass 24-hr diet recalls and accelerometry respectively. Data collection was conducted pre and post a 12-week physical activity intervention which consisted of cardiorespiratory activity during instructor led sessions (60 mins, twice weekly) and family led sessions (>10 mins, 4 days/wk). There is no universally accepted definition of MetSyn in childhood. The International Diabetes Federation suggests that MetSyn should not be diagnosed in children aged 6 to < 10 years. Children can be identified to be at risk of MetSyn, however, based on waist circumference ≥90th percentile and family history1,2; all subjects in this study were at risk according to these criteria. Four definitions of paediatric MetSyn previously applied to a group of young, overweight Australian children3 were used to calculate the prevalence of MetSyn in the current sample and it was found to be 27-89% at baseline and 13-80% after the experimental intervention depending upon the definition used. Acanthosis nigricans and impaired glucose tolerance (IGT) were present in one female child. Post-intervention, IGT had resolved and the child was glucose tolerant. Habitual dietary intake (energy intake and macronutrients) measured over a 3-day period pre-intervention displayed a significant positive association between fasting glucose and energy intake, as well as a significant negative association between fasting glucose and the protein component of the diet. Following the physical activity programme, energy intake was significantly positively correlated with body fat percentage (% BF). There was no difference found in dietary intake assessed prior to and following cessation of the physical activity intervention, in terms of energy or % energy from macronutrients. Habitual physical activity was not related to MetSyn diagnostic indicators. A higher level of physical fitness, estimated by predicted O2max (ml•kg-1•min-1), was significantly correlated with a lower level of diastolic blood pressure at baseline. A greater fitness level ( O2max) was moderately correlated with a lower BMI Z-score following the 12-week intervention. There was no difference between pre- and post-intervention habitual physical activity. A trend towards less sedentary time and increased light intensity activity was found, but these did not reach significance. Physical fitness level showed a trend for improvement following the intervention (P = 0.060). Anthropometrically determined body composition and body fat distribution did not change following the intervention. Radiologically determined abdominal adipose tissue depots were not significantly different post-intervention. % BF was not different when assessed with bioelectrical impedance analysis. However, % BF did reduce significantly over the 12-week intervention period when quantified by hydrometry (42.3% ± 5.0 vs 36.9% ± 8.6, P = 0.022). Adipokines, the secretory products of adipocytes displaying pleiotropic metabolic action, were investigated for their relation to lipid depots and additionally for change post-intervention. Cardiovascular (CV) disease risk was investigated by proatherogenic and protective blood lipids. When examined at baseline, fasting blood triacylglycerols (TAG) were inversely associated with basal and stimulated insulin sensitivity. Post-intervention, a higher level of HDL-C was found to be associated with greater insulin sensitivity, although this was not apparent at baseline. The relation between TAG and insulin sensitivity discovered pre-intervention was no longer evident. All other biomarkers of CV risk were not associated with body composition, glucose homeostasis, and lifestyle factors pre- and post-intervention. The effect of the physical activity intervention on indicators of haemostasis, physical fitness, blood lipids and lipoproteins, systemic inflammation, and fibrinolytic activity were analysed for change. Both systolic and diastolic blood pressure were significantly reduced following the physical activity programme. There was no significant difference found in any other measured parameter of CV risk. Log[HOMA], a surrogate index of insulin resistance, was significantly decreased post-intervention indicating reduced insulin resistance. QUICKI, a surrogate index of insulin sensitivity, was significantly improved post-intervention. The remaining indicators of insulin resistance, insulin sensitivity and β-cell function based on fasting surrogates did not significantly change over the 12-week experimental period. Dynamic insulin sensitivity and β-cell function were investigated pre- and post-intervention using paired samples t-tests. Glucose and insulin area under the curve of the OGTT were significantly reduced and whole-body insulin sensitivity index (WBISI) was significantly increased hence showing an improvement in stimulated insulin sensitivity. AUCCP/AUCglu significantly declined also indicating an improved response to oral glucose stimulation. IGI and ΔCP30/ΔG30, as markers of β-cell insulin secretion, did not change. Disposition index, the interrelationship of insulin secretion (IGI) and insulin sensitivity (WBISI), was not changed pre- and post-intervention. Hepatic insulin extraction was increased post-intervention (4.3 ± 1.2 vs 4.8 ± 1.1, P = 0.022) possibly due to greater hepatic and/or peripheral insulin sensitivity. General linear modeling (GLM) showed the improvement in whole-body insulin sensitivity discovered following the intervention was independent of % BF, abdominal adipose tissue depots, and ectopic lipid depots. Intrahepatocellular lipids (IHCL) significantly decreased after the 12-week intervention (6.99% ± 9.41 vs 5.83% ± 8.54) whilst there was no significant change in the serum markers of liver inflammation. IHCL was positively and strongly associated with total abdominal adipose tissue, intra-abdominal adipose tissue and subcutaneous abdominal adipose tissue both before and after the intervention. IHCL was positively associated with %BF measured post-intervention; this relationship almost reached significance when measured pre-intervention (P = 0.060). IHCL was not associated with insulin sensitivity either pre- or post-intervention nor with circulating lipids at either timepoint. The change in IHCL was independent of % BF and abdominal adipose tissue tested by GLM. However, there was no significant difference found in IHCL post-intervention after adjustment for insulin sensitivity (WBISI) by GLM. Prior to intervention, 10 of 15 subjects had hepatic steatosis diagnostic of non-alcoholic fatty liver disease. Eight of the 10 subjects with clinically significant hepatic steatosis had reduction of fatty infiltrate following the exercise intervention. In the whole group it was demonstrated that physical activity attenuates lipid infiltration of the liver independent of body fat. To further investigate the pathophysiology of ectopic lipid depots, biomarkers of oxidative stress and anti-oxidant status were examined in relation to IHCL. Pre-intervention, there was no association found between pro-oxidative or anti-oxidative activity and IHCL. Post-intervention, an inverse association of plasma carotenoid:cholesterol ratio with IHCL was found. Skeletal intramyocellular lipids (IMCL) measured in the right soleus were significantly increased post-intervention (2.4 ± 1.1 vs 2.6 ± 1.2, P = 0.035). There was no association between IMCL and % BF when measured pre- or post-intervention. Abdominal adipose depots were associated with IMCL at baseline and following the intervention. IMCL was not related to IHCL at either timepoint. Pre-intervention, there was a trend for a relationship between IMCL and insulin. Post-intervention, IMCL was tightly and inversely correlated with insulin sensitivity (r = -0.85 P = 0.000). Linear regression between IMCL and WBISI run pre-intervention and post-intervention found the slopes were not significantly different whereas the intercepts were highly significantly different (P = 0.001), thus, as IMCL increased there was a corresponding decrease in insulin sensitivity. GLM found the increase in IMCL was independent of % BF and abdominal adipose tissue, but was not independent of WBISI. These data indicate the greater IMCL level found post-intervention was a non-pathologic training adaptation. To further investigate the pathophysiology of ectopic lipid depots, biomarkers of oxidative stress and anti-oxidant status were examined in relation to IMCL. Pre-intervention, there was a positive association between malondialdehyde and IMCL. Post-intervention, an inverse association was found between IMCL and both plasma total carotenoids and total carotenoid:free cholesterol ratio. In summation, this study found improved metabolic health in obese, prepubertal children following a 12-week physical activity intervention without dietary intervention or intentional weight loss. Body fat and fat distribution were not prime mediators for the effect of the intervention on parameters of the Metabolic Syndrome; whereas insulin sensitivity was discovered to be a mediator of the change shown in ectopic fat depots. Causality and directionality of these fascinating relationships cannot be determined from the present study, and further research is encouraged. This thesis offers an insight into the pathogenesis of MetSyn and the use of physical activity to improve MetSyn in the setting of paediatric obesity.

Intramyocellular lipid (IMCL), an ectopic fat depot found within skeletal muscle fibers, is highly associated with obesity and strongly correlated with insulin resistance. IMCL accumulation in sedentary individuals may contribute to insulin resistance by interfering with insulin signaling in skeletal muscle, leading to inadequate glucose uptake by the cell. Lipid peroxidation is also associated with both obesity and insulin resistance, and with IMCL, but a relationship has yet to be established among all of these variables. The purpose of this project is to study for the first time the relationships among lipid peroxidation, IMCL content, and glucose uptake in skeletal muscle. Nine insulin-sensitive adults (IS), 13 insulin-resistant adults (IR), 10 diabetic (DB) and 8 subjects pre- and post- 12-week intervention with insulin-sensitizing thiazolinedione (TZD) were assessed for soleus IMCL with nuclear magnetic resonance, insulin sensitivity by both hyperinsulinemic-euglycemic clamp (GDR) and homeostasis model assessment index (HOMA1), and anthropometrics, including body mass index (BMI), percent fat by DEXA scan, and waist circumference. Vastus lateralis biopsies of all subjects were homogenized and analyzed by immunoblotting for post-translational protein modifications occurring from lipid-peroxidation (HNE). GDR and HOMA were significantly different among IS, IR, and DB groups, as expected, as were waist circumference and BMI. IMCL was significantly higher in DB than in IS and IR. HNE was also higher in DB than in IS, although it did not differ from IR. HNE was significantly correlated to GDR, HOMA1, and BMI, but not to IMCL, WAIST, or percent fat measures. IMCL showed a strong, negative correlation with GDR and was the primary, independent predictor of GDR in stepwise multiple regression. HNE was the primary, independent predictor of HOMA in stepwise multiple regression. Paired t-tests revealed improvements in insulin sensitivity measures after 12 weeks of TZD intervention, but no significant differences were observed in IMCL or HNE after intervention. These data show that skeletal muscle HNE and IMCL are both determinants of insulin resistance in obese, sedentary adults. HNE and IMCL are not related and therefore impact insulin resistance independently. These results reveal, for the first time, a negative relationship between skeletal muscle HNE and insulin sensitivity in sedentary individuals and underscore the importance of lipid peroxidation in insulin resistance.

Abstract Metabolic Syndrome (MetSyn) is a medical condition prevalent in Australia. MetSyn is diagnosed with a varying combination of visceral obesity, insulin resistance/ impaired glucose tolerance/ Type 2 diabetes, dyslipidaemia and hypertension. Obesity is a central feature of this syndrome that is characterised by abnormalities in glucose and lipid metabolism. An understanding of the cause of the metabolic derangement that occurs in obesity, and that contributes to MetSyn, would allow effective treatment and prevention strategies to be formulated. This is a priority in the current environment of highly prevalent overweight and obesity in Australian children and adults. Lipotoxicity of insulin-dependent tissues and ectopic fat depots are emerging as fundamental processes in the pathogenesis of MetSyn. Lifestyle intervention, such as increased physical activity, show great promise as agents for disrupting the disease progression and may act via direct or indirect mechanisms on the underlying pathology of MetSyn. This study aimed to determine if diagnostic markers of MetSyn exist in obese, prepubertal, Australian children and to assess the contribution of lifestyle factors on components of MetSyn. Further, this study sought to investigate the relationship between body fat patterning (total body fat, abdominal adipose depots, skeletal intramyocellular lipids, intrahepatocellular lipids) and markers of MetSyn. An experimental intervention was then employed to examine the effect of physical activity on body fat distribution, insulin sensitivity, and haemodynamic and biochemical markers of MetSyn, and additionally to determine if the effect of exercise on parameters of MetSyn was mediated by a change in body fat patterning. Data were collected in a group of 15 obese (mean BMI Z-score 2.51 ± 0.49), prepubertal children (6 male, 9 female) aged 5.1 – 11.4 years (mean age 7.82 yrs ± 1.83). Measures included insulin sensitivity, blood biochemistry (lipid, haemostatic and adipocyte activity markers), blood pressure, two-compartment body composition by hydrometry, and nuclear magnetic resonance scanning for abdominal adipose depots, intrahepatic lipids and skeletal intramyocellular lipids. Each child’s habitual nutrition and physical activity were also ascertained using multiple-pass 24-hr diet recalls and accelerometry respectively. Data collection was conducted pre and post a 12-week physical activity intervention which consisted of cardiorespiratory activity during instructor led sessions (60 mins, twice weekly) and family led sessions (>10 mins, 4 days/wk). There is no universally accepted definition of MetSyn in childhood. The International Diabetes Federation suggests that MetSyn should not be diagnosed in children aged 6 to < 10 years. Children can be identified to be at risk of MetSyn, however, based on waist circumference ≥90th percentile and family history1,2; all subjects in this study were at risk according to these criteria. Four definitions of paediatric MetSyn previously applied to a group of young, overweight Australian children3 were used to calculate the prevalence of MetSyn in the current sample and it was found to be 27-89% at baseline and 13-80% after the experimental intervention depending upon the definition used. Acanthosis nigricans and impaired glucose tolerance (IGT) were present in one female child. Post-intervention, IGT had resolved and the child was glucose tolerant. Habitual dietary intake (energy intake and macronutrients) measured over a 3-day period pre-intervention displayed a significant positive association between fasting glucose and energy intake, as well as a significant negative association between fasting glucose and the protein component of the diet. Following the physical activity programme, energy intake was significantly positively correlated with body fat percentage (% BF). There was no difference found in dietary intake assessed prior to and following cessation of the physical activity intervention, in terms of energy or % energy from macronutrients. Habitual physical activity was not related to MetSyn diagnostic indicators. A higher level of physical fitness, estimated by predicted O2max (ml•kg-1•min-1), was significantly correlated with a lower level of diastolic blood pressure at baseline. A greater fitness level ( O2max) was moderately correlated with a lower BMI Z-score following the 12-week intervention. There was no difference between pre- and post-intervention habitual physical activity. A trend towards less sedentary time and increased light intensity activity was found, but these did not reach significance. Physical fitness level showed a trend for improvement following the intervention (P = 0.060). Anthropometrically determined body composition and body fat distribution did not change following the intervention. Radiologically determined abdominal adipose tissue depots were not significantly different post-intervention. % BF was not different when assessed with bioelectrical impedance analysis. However, % BF did reduce significantly over the 12-week intervention period when quantified by hydrometry (42.3% ± 5.0 vs 36.9% ± 8.6, P = 0.022). Adipokines, the secretory products of adipocytes displaying pleiotropic metabolic action, were investigated for their relation to lipid depots and additionally for change post-intervention. Cardiovascular (CV) disease risk was investigated by proatherogenic and protective blood lipids. When examined at baseline, fasting blood triacylglycerols (TAG) were inversely associated with basal and stimulated insulin sensitivity. Post-intervention, a higher level of HDL-C was found to be associated with greater insulin sensitivity, although this was not apparent at baseline. The relation between TAG and insulin sensitivity discovered pre-intervention was no longer evident. All other biomarkers of CV risk were not associated with body composition, glucose homeostasis, and lifestyle factors pre- and post-intervention. The effect of the physical activity intervention on indicators of haemostasis, physical fitness, blood lipids and lipoproteins, systemic inflammation, and fibrinolytic activity were analysed for change. Both systolic and diastolic blood pressure were significantly reduced following the physical activity programme. There was no significant difference found in any other measured parameter of CV risk. Log[HOMA], a surrogate index of insulin resistance, was significantly decreased post-intervention indicating reduced insulin resistance. QUICKI, a surrogate index of insulin sensitivity, was significantly improved post-intervention. The remaining indicators of insulin resistance, insulin sensitivity and β-cell function based on fasting surrogates did not significantly change over the 12-week experimental period. Dynamic insulin sensitivity and β-cell function were investigated pre- and post-intervention using paired samples t-tests. Glucose and insulin area under the curve of the OGTT were significantly reduced and whole-body insulin sensitivity index (WBISI) was significantly increased hence showing an improvement in stimulated insulin sensitivity. AUCCP/AUCglu significantly declined also indicating an improved response to oral glucose stimulation. IGI and ΔCP30/ΔG30, as markers of β-cell insulin secretion, did not change. Disposition index, the interrelationship of insulin secretion (IGI) and insulin sensitivity (WBISI), was not changed pre- and post-intervention. Hepatic insulin extraction was increased post-intervention (4.3 ± 1.2 vs 4.8 ± 1.1, P = 0.022) possibly due to greater hepatic and/or peripheral insulin sensitivity. General linear modeling (GLM) showed the improvement in whole-body insulin sensitivity discovered following the intervention was independent of % BF, abdominal adipose tissue depots, and ectopic lipid depots. Intrahepatocellular lipids (IHCL) significantly decreased after the 12-week intervention (6.99% ± 9.41 vs 5.83% ± 8.54) whilst there was no significant change in the serum markers of liver inflammation. IHCL was positively and strongly associated with total abdominal adipose tissue, intra-abdominal adipose tissue and subcutaneous abdominal adipose tissue both before and after the intervention. IHCL was positively associated with %BF measured post-intervention; this relationship almost reached significance when measured pre-intervention (P = 0.060). IHCL was not associated with insulin sensitivity either pre- or post-intervention nor with circulating lipids at either timepoint. The change in IHCL was independent of % BF and abdominal adipose tissue tested by GLM. However, there was no significant difference found in IHCL post-intervention after adjustment for insulin sensitivity (WBISI) by GLM. Prior to intervention, 10 of 15 subjects had hepatic steatosis diagnostic of non-alcoholic fatty liver disease. Eight of the 10 subjects with clinically significant hepatic steatosis had reduction of fatty infiltrate following the exercise intervention. In the whole group it was demonstrated that physical activity attenuates lipid infiltration of the liver independent of body fat. To further investigate the pathophysiology of ectopic lipid depots, biomarkers of oxidative stress and anti-oxidant status were examined in relation to IHCL. Pre-intervention, there was no association found between pro-oxidative or anti-oxidative activity and IHCL. Post-intervention, an inverse association of plasma carotenoid:cholesterol ratio with IHCL was found. Skeletal intramyocellular lipids (IMCL) measured in the right soleus were significantly increased post-intervention (2.4 ± 1.1 vs 2.6 ± 1.2, P = 0.035). There was no association between IMCL and % BF when measured pre- or post-intervention. Abdominal adipose depots were associated with IMCL at baseline and following the intervention. IMCL was not related to IHCL at either timepoint. Pre-intervention, there was a trend for a relationship between IMCL and insulin. Post-intervention, IMCL was tightly and inversely correlated with insulin sensitivity (r = -0.85 P = 0.000). Linear regression between IMCL and WBISI run pre-intervention and post-intervention found the slopes were not significantly different whereas the intercepts were highly significantly different (P = 0.001), thus, as IMCL increased there was a corresponding decrease in insulin sensitivity. GLM found the increase in IMCL was independent of % BF and abdominal adipose tissue, but was not independent of WBISI. These data indicate the greater IMCL level found post-intervention was a non-pathologic training adaptation. To further investigate the pathophysiology of ectopic lipid depots, biomarkers of oxidative stress and anti-oxidant status were examined in relation to IMCL. Pre-intervention, there was a positive association between malondialdehyde and IMCL. Post-intervention, an inverse association was found between IMCL and both plasma total carotenoids and total carotenoid:free cholesterol ratio. In summation, this study found improved metabolic health in obese, prepubertal children following a 12-week physical activity intervention without dietary intervention or intentional weight loss. Body fat and fat distribution were not prime mediators for the effect of the intervention on parameters of the Metabolic Syndrome; whereas insulin sensitivity was discovered to be a mediator of the change shown in ectopic fat depots. Causality and directionality of these fascinating relationships cannot be determined from the present study, and further research is encouraged. This thesis offers an insight into the pathogenesis of MetSyn and the use of physical activity to improve MetSyn in the setting of paediatric obesity.

The global epidemic of obesity is rapidly becoming a major public health problem in many parts of the world. Unhealthy diets and physical inactivity are two modifiable risk factors for prevention of obesity and its associated chronic diseases. Their influence on muscle energy metabolism and fat mass is not completely elucidated. A decreased capacity for fatty acid oxidation (FAO) may be a metabolic risk factor for weight gain and is found to be depressed in obese individuals; and exercise training may promote an increased capacity for FAO. In addition to the interest in whole-body FAO, the role of site specific lipid accumulation including visceral adipose tissue (VAT), intrahepatic lipids (IHL) and intramyocellular lipids (IMCL) has become a focus of interest because of their reported association with insulin resistance (IR), a key metabolic defect associated with obesity and type 2 diabetes mellitus (T2DM). However, ambiguity persists regarding the importance of IMCL as a metabolic substrate for energy production in obesity. A better understanding of the factors regulating FAO, body fat distribution and IMCL mobilisation is important for the development of interventions allowing effective treatment of conditions in which these are disturbed. The study of individuals with metabolic myopathies can give more information about the energy metabolism of muscle. McArdle disease (MD) affects glucose availability to muscle for energy production. Investigations into IMCL storage and mobilisation in MD have not been reported. The aims of this thesis are to investigate 1) the effects of weight-loss via dietary restriction plus modest but clinically-relevant exercise training on FAO, body fat distribution and mobilisation of IMCL during exercise in obese non-diabetic adults; 2) the effect of an exercise training intervention on IMCL storage and mobilisation in a subject with MD. All obese subjects underwent a 4 month lifestyle intervention with weekly meetings with a dietitian and an exercise physiologist. Of the 92 subjects, 73 completed the intervention. They showed significant decreases in body weight (8%), fat mass (14%) and total cholesterol (5%). The exercise prescription of 1500 kcal.week-1 resulted in variable compliance with the prescription (1224 ± 1085 kcal.week-1) measured by heart rate monitor. Those who did most exercise and also those who had less weekly variability in their exercise, had greater reductions in body weight and fat mass. The total activity energy expenditure measured by accelerometry did not change post-intervention but there was a reduction in low intensity activity and an increase in moderate and high intensity activity. A submaximal treadmill test and resting metabolic rate (RMR) using indirect calorimetry was measured before and after the intervention to investigate factors regulating FAO and energy expenditure. Subjects showed increases in FAO without change in energy expenditure for the same walking speed post intervention, but the volume of exercise completed during the intervention was not associated with these changes. To investigate body fat distribution in obesity, VAT, IHL and soleus muscle IMCL was measured in a sub-group of 18 males by magnetic resonance imaging (MRI) and spectroscopy (MRS) along with measurement of maximal aerobic capacity. Fitness increased significantly with significant decreases in VAT (29%) and IHL (54%), without significant change in IMCL. Subjects who had the greatest decrease in VAT were those who exercised for longer durations during the intervention. IHL was the only measure of excess lipid that correlated with IR. The measurement of IMCL before and after 1-hour of cycle ergometer exercise showed no significant mobilisation of IMCL either at baseline or after the lifestyle intervention. The intensity of the acute exercise was adjusted to correspond to each individual’s maximal fatty acid oxidation (MFAO) which increased by over 60% post intervention. In the subject with MD, an 8 week exercise training intervention without dietary intervention increased IMCL stores by 27%, but there was no marked change in IMCL with acute exercise at both time points. The findings of this thesis demonstrate that a clinically relevant and achievable lifestyle intervention incorporating weight loss through diet and objectively measured exercise can achieve improvements in blood lipid profile, body composition and FAO. The differential effects of the intervention on the various fat depots and their associations to metabolic markers suggest that individualised strategies may be required dependent upon body fat distribution. The non detection of mobilisation of IMCL by MRS suggests that these lipids may not be present as a substrate source in this population but rather an ectopic lipid depot related to increased energy consumption in diet. The relatively low capacity for FAO in both the obese and MD subjects may have affected the results. This thesis discusses implications for clinical practice, discusses novel findings related to the energy metabolism in obesity and MD and informs clinical and basic science about important future directions.

The global epidemic of obesity is rapidly becoming a major public health problem in many parts of the world. Unhealthy diets and physical inactivity are two modifiable risk factors for prevention of obesity and its associated chronic diseases. Their influence on muscle energy metabolism and fat mass is not completely elucidated. A decreased capacity for fatty acid oxidation (FAO) may be a metabolic risk factor for weight gain and is found to be depressed in obese individuals; and exercise training may promote an increased capacity for FAO. In addition to the interest in whole-body FAO, the role of site specific lipid accumulation including visceral adipose tissue (VAT), intrahepatic lipids (IHL) and intramyocellular lipids (IMCL) has become a focus of interest because of their reported association with insulin resistance (IR), a key metabolic defect associated with obesity and type 2 diabetes mellitus (T2DM). However, ambiguity persists regarding the importance of IMCL as a metabolic substrate for energy production in obesity. A better understanding of the factors regulating FAO, body fat distribution and IMCL mobilisation is important for the development of interventions allowing effective treatment of conditions in which these are disturbed. The study of individuals with metabolic myopathies can give more information about the energy metabolism of muscle. McArdle disease (MD) affects glucose availability to muscle for energy production. Investigations into IMCL storage and mobilisation in MD have not been reported. The aims of this thesis are to investigate 1) the effects of weight-loss via dietary restriction plus modest but clinically-relevant exercise training on FAO, body fat distribution and mobilisation of IMCL during exercise in obese non-diabetic adults; 2) the effect of an exercise training intervention on IMCL storage and mobilisation in a subject with MD. All obese subjects underwent a 4 month lifestyle intervention with weekly meetings with a dietitian and an exercise physiologist. Of the 92 subjects, 73 completed the intervention. They showed significant decreases in body weight (8%), fat mass (14%) and total cholesterol (5%). The exercise prescription of 1500 kcal.week-1 resulted in variable compliance with the prescription (1224 ± 1085 kcal.week-1) measured by heart rate monitor. Those who did most exercise and also those who had less weekly variability in their exercise, had greater reductions in body weight and fat mass. The total activity energy expenditure measured by accelerometry did not change post-intervention but there was a reduction in low intensity activity and an increase in moderate and high intensity activity. A submaximal treadmill test and resting metabolic rate (RMR) using indirect calorimetry was measured before and after the intervention to investigate factors regulating FAO and energy expenditure. Subjects showed increases in FAO without change in energy expenditure for the same walking speed post intervention, but the volume of exercise completed during the intervention was not associated with these changes. To investigate body fat distribution in obesity, VAT, IHL and soleus muscle IMCL was measured in a sub-group of 18 males by magnetic resonance imaging (MRI) and spectroscopy (MRS) along with measurement of maximal aerobic capacity. Fitness increased significantly with significant decreases in VAT (29%) and IHL (54%), without significant change in IMCL. Subjects who had the greatest decrease in VAT were those who exercised for longer durations during the intervention. IHL was the only measure of excess lipid that correlated with IR. The measurement of IMCL before and after 1-hour of cycle ergometer exercise showed no significant mobilisation of IMCL either at baseline or after the lifestyle intervention. The intensity of the acute exercise was adjusted to correspond to each individual’s maximal fatty acid oxidation (MFAO) which increased by over 60% post intervention. In the subject with MD, an 8 week exercise training intervention without dietary intervention increased IMCL stores by 27%, but there was no marked change in IMCL with acute exercise at both time points. The findings of this thesis demonstrate that a clinically relevant and achievable lifestyle intervention incorporating weight loss through diet and objectively measured exercise can achieve improvements in blood lipid profile, body composition and FAO. The differential effects of the intervention on the various fat depots and their associations to metabolic markers suggest that individualised strategies may be required dependent upon body fat distribution. The non detection of mobilisation of IMCL by MRS suggests that these lipids may not be present as a substrate source in this population but rather an ectopic lipid depot related to increased energy consumption in diet. The relatively low capacity for FAO in both the obese and MD subjects may have affected the results. This thesis discusses implications for clinical practice, discusses novel findings related to the energy metabolism in obesity and MD and informs clinical and basic science about important future directions.