Effects of immediate‐release niacin and dietary fatty acids on acute insulin and lipid status in individuals with metabolic syndrome

For Peer Review 1 Effects of immediate-release niacin and dietary fatty acids 1 on acute insulin and lipid status in individuals with 2 metabolic syndrome 3 4 Sergio Montserrat-de la Paz 1 , Sergio Lopez 1 , Beatriz Bermudez 2 , Juan M. 5 Guerrero 3 , Rocio Abia 1 and Francisco J.G. Muriana 1, * 6 7 1 Laboratory of Cellular and Molecular Nutrition, Instituto de la Grasa (CSIC), Seville, Spain; 8 2 Department of Cell Biology, Faculty of Biology, University of Seville, Seville, Spain; and 9 3 Department of Clinical Biochemistry, University Hospital Virgen del Rocio, 10 IBiS/CSIC/University of Seville, Seville, Spain 11 12 *Address correspondence to FJG Muriana, Laboratory of Cellular and Molecular Nutrition, 13 Instituto de la Grasa, CSIC, Ctra. de Utrera km. 1, Edificio 46, Campus Universitario Pablo 14 de Olavide, 41013 Seville, Spain. Telephone number: 95 4611550. Fax number: 95 4616790. 15 Email: [email protected] 16 17 Dedicated to the memory of Professor Jose Villar Ortiz who died on March 28th, 2016, and is 18 a celebration of his work and contribution to the care and well being of people with MetS and 19 any kind of person. He always hosted and guided our studies on olive oil and the postprandial 20 metabolism of dietary fats. He has been and is an inspiration. 21 22 23 Page 1 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 This is the peer reviewed version of the following article: Journal of the Science of Food and Agriculture 98 (6): 2194-2200 (2018), which has been published in final form at https:// doi.org/10.1002/jsfa.8704. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-Archiving For Peer Review 2 ABSTRACT 24 Background: The nature of dietary fats profoundly affects postprandial hypertriglyceridemia 25 and glucose homeostasis. Niacin is a potent lowering-lipid agent. However, limited data exist 26 on postprandial triglycerides and glycemic control following co-administration of high-fat 27 meals with a single-dose of niacin in subjects with metabolic syndrome (MetS). Thus, the 28 study aim was to explore whether a fat challenge containing predominantly saturated (SFAs), 29 monounsaturated (MUFAs) or MUFAs plus omega-3 long-chain polyunsaturated (LCPUFAs) 30 fatty acids together with a single-dose of immediate-release niacin have a relevant role on 31 postprandial insulin and lipid status in subjects with MetS. 32 Results: In a randomized crossover within-subject design, 16 subjects with MetS were given 33 a single-dose of immediate-release niacin (2 g) and ~15 calorie per kilogram of body weight 34 meals containing either SFAs, MUFAs, MUFAs plus omega-3 LCPUFAs or no fat. At 35 baseline and hourly over 6 h, plasma glucose, insulin, C-peptide, triglycerides, free fatty acids 36 (FFAs), total cholesterol, HDL-, and LDL-cholesterol were assessed. Co-administered with 37 niacin, high-fat meals significantly increased the postprandial concentrations of glucose, 38 insulin, C-peptide, triglycerides, FFAs, and postprandial indices of β-cell function. However, 39 postprandial indices of insulin sensitivity were significantly decreased. These effects were 40 significantly attenuated with MUFAs or MUFAs plus omega-3 LCPUFAs when compared 41 with SFAs. 42 Conclusion: In the setting of niacin co-administration and compared to dietary SFAs, 43 MUFAs limit the postprandial insulin, triglyceride, and FFA excursions and improve the 44 postprandial glucose homeostasis in the MetS. 45 46 Keywords: niacin, dietary fatty acids, MUFAs, SFAs, postprandial, metabolic syndrome 47 48 Page 2 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 3 INTRODUCTION 49 The metabolic syndrome (MetS) is a constellation of interrelated and additive risk factors for 50 cardiovascular disease (CVD) that comprises dysglycemia, dyslipidemia, hypertension, and 51 central obesity. 1 The dyslipidemic state of MetS is characterized by raised triglycerides, low 52 HDL-cholesterol, and abnormal elevation in postprandial, apolipoprotein B48 (apoB48)-53 containing triglyceride-rich lipoproteins (TRLs), mainly composed of dietary fatty acids. 54 Subjects with MetS often have impaired glucose metabolism resulting from defects in insulin 55 secretion, insulin action, or both. 56 57 Niacin (also commonly known as nicotinic acid or vitamin B3) is a water-soluble vitamin 58 effective for lowering triglycerides and raising HDL-cholesterol. 2 Data from recent studies 59 indicated that niacin has long-term benefits in lowering postprandial secretion of apoB48 3 and 60 TRLs 4 in statin-treated patients with type 2 diabetes. Less evident are the long-term benefits 61 of niacin on postprandial glucose homeostasis in dyslipidemic patients with or without type 2 62 diabetes. 5 Of interest, the extended-release form of niacin has been shown to acutely suppress 63 postprandial hypertriglyceridemia in healthy subjects. 6 However, the role of immediate-64 release niacin dosed with a high-fat meal on postprandial insulin secretion or action and 65 hypertriglyceridemia, particularly in subjects with MetS, remains largely unknown. 66 67 We have previously shown that the nature of the dietary fats in the meal influences on 68 postprandial triglycerides and control of insulin secretion and sensitivity in subjects with 69 normal 7 and high 8 fasting triglycerides. Our studies provided evidence that subjects had 70 decreased postprandial β-cell function and became less insulin resistant postprandially as the 71 proportion of MUFAs compared with SFAs in dietary fats increased. These effects were 72 specifically associated to the content of MUFA oleic acid and SFA palmitic acid in the meal. 9 73 Page 3 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 4 In addition, omega-3 long-chain PUFAs (LCPUFAs) have long-term benefits in lowering 74 triglycerides, 10 notably when combined with extended-release niacin, 2 in individuals with 75 MetS. Omega-3 LCPUFAs may also attenuate postprandial abnormalities in lipid metabolism 76 associated with MetS. 11 However, recent trials have not produced evidence for additional 77 clinical benefits of omega-3 PUFAs on glucose homeostasis in either the fasting or 78 postprandial period. 10,12 79 80 Here, we explored the impact of immediate-release niacin co-ingested with a meal rich in 81 SFAs, MUFAs or MUFAs plus omega-3 LCPUFAs on postprandial insulinemic and lipemic 82 responses as well as on postprandial parameters and indices of β-cell function and insulin 83 sensitivity in subjects with MetS. 84 85 MATERIALS AND METHODS 86 Participants and design 87 Inclusion criteria for MetS consisted of at least 3 of the following components: waist 88 circumference >102 cm, fasting plasma HDL-cholesterol ≤ 1.03 mmol/l, fasting plasma 89 triglycerides ≥1.7 mmol/l, systolic blood pressure (SBP) ≥130 mmHg or diastolic blood 90 pressure (DBP) ≥85 mmHg, and fasting plasma glucose ≥5.6 mmol/l. 13 As shown in Table 1, 91 the participants' average body mass index (BMI), waist circumference, waist-to-hip ratio, and 92 blood pressure were above the standard values, while average HDL-cholesterol was below the 93 standard value, reflecting characteristics of individuals with MetS. None of the participants 94 had impaired renal, thyroid or liver function; none had cardiovascular disease or 95 gastroparesis; none had anemia or pulmonary, psychiatric, immunological or neoplastic 96 diseases; none used tobacco, consumed special diets or took medication known to alter gastric 97 emptying, insulin secretion or insulin action. Ethics approval was obtained from the Human 98 Page 4 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 5 Clinical Research and Ethics Committee of the University Hospital Virgen del Rocio (UHVR, 99 Seville), and the study complied with the current revision of the Declaration of Helsinki. All 100 subjects gave written informed consent. This study was registered at the ClinicalTrials.gov 101 registry and the clinical trial registration number is NCT02061267. 102 103 The study was designed as a within-subject crossover in which the participants attended the 104 Clinic Experimental Research Unit for Vascular Risk at the UHVR on 4 separate occasions. 105 Fasting blood samples (t = 0) were taken at 0800 after a 12-h overnight fast. The test meals in 106 combination with a single 2 g dose of immediate-release niacin (Twinlab, American Fork, 107 UT) were given in random order with an interval of ~1 wk between meals. The high-fat meals 108 consisted in an emulsion prepared according to a method previously described, 14 with water, 109 sucrose (30 g/m 2 of body surface area), and fat (50 g/m 2 of body surface area). Dietary fats 110 were cow's milk cream, refined olive oil or refined olive oil plus a dose of omega-3 111 LCPUFAs, which consisted of 920 mg of EPA and 760 mg of DHA in the form of ethyl esters. 112 The fatty acid compositions of dietary fats are described in Table S1 in the Supporting 113 Information. The participants also consumed a test meal prepared as indicated above, but not 114 including fat, as a control meal. After the ingestion of the meals within 10 min, blood samples 115 were collected each 60 min in tubes containing EDTA for the measurement of glucose, 116 insulin, C-peptides, triglycerides, FFAs, total cholesterol, HDL-cholesterol, and LDL-117 cholesterol over 360 min. In this study, each participant served as his own control. Subject 118 flow through the protocol is shown in Figure S1 in the Supporting Information. 119 120 Biochemical analyses 121 Glucose was immediately measured using the glucose/oxidase method (Glucose GOD-PAP; 122 Biolabo, Madrid, Spain). Insulin and C-peptide were measured by using ELISA (Diagnostic 123 Page 5 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 6 System Laboratories, Webster, TX). Total cholesterol and triglycerides were determined by 124 enzymatic methods (CHOD-PAP and GPO-PAP, respectively; Roche Diagnostics, Basel, 125 Switzerland). HDL-cholesterol was determined after precipitation with phosphotungstic acid. 126 LDL-cholesterol was measured using an Advia 2400 Clinical Chemistry System (Siemens 127 Healthcare Diagnostics, Erlangen, Germany). FFAs were measured with an ACS-ACOD 128 assay (Wako Chemicals GmbH, Neuss, Germany). Glycated hemoglobin (HbA 1c ) was 129 measured according to the Standard Operating Procedure of the IFCC Reference, with an 130 automated HPLC analyzer (Bio-Rad, Milan, Italy). 131 132 Calculations 133 Fasting β-cell function was estimated by 2 methods: 1) the homeostasis model assessment of 134 insulin secretion (HOMA-B), by using the formula HOMA-B = I 0 × 3.33/(G 0 – 3.5); and 2) 135 the basal disposition index (DI 0 ), which gives an adjusted measure by insulin sensitivity 136 according to the HOMA-IR, by using the formula DI 0 = HOMA-B × (1/HOMA-IR). 15 137 138 Fasting insulin resistance and its reciprocal (fasting insulin sensitivity) were estimated by 4 139 methods: 1) HOMA-IR = I 0 × G 0 divided by 22.5; 2) the revised-quantitative insulin 140 sensitivity check index [rQUICKI = 1/(log I 0 + log G 0 + log FFA 0 )]; 3) the basal insulin 141 sensitivity index (ISI 0 ) for glycemia {ISI(G) 0 = 2/[([I 0 × G 0 ) + 1]}; and 4) the basal ISI 0 for 142 blood FFAs {ISI(FFA) 0 = 2/[(I 0 × FFA 0 ) + 1]}. I 0 , G 0 , and FFA 0 refer to fasting 143 concentrations (t = 0) of insulin, glucose, and FFAs, respectively. 144 145 Postprandial β-cell function was estimated by 4 methods: 1) the insulinogenic index (IGI), 146 which is a surrogate measure of first-phase insulin secretion and was calculated by using the 147 difference between the postprandial insulin peak (t = 60 min) and the fasting insulin in 148 Page 6 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 7 relation to the difference in glucose (IGI = ∆I 0–60 /∆G 0–60 ); 2) the ratio of the IGI to the 149 HOMA-IR, which gives an adjusted measure of β-cell function that accounts for variations in 150 insulin sensitivity; 3) the ratio of the insulin to glucose areas under the curve 151 (AUC INS /AUC GLU ), which significantly correlates with glucose sensitivity and early-phase 152 insulin secretion, calculated by using the trapezoidal method from 0 to 120 min; and 4) the 153 total disposition index from 0 to 120 min (DI 120 ) by using the product of AUC INS /AUC GLU 154 with the Matsuda insulin sensitivity index, 16 which incorporates both the hepatic and muscle 155 components of insulin resistance and correlates well with the euglycemic insulin clamp. 156 157 Postprandial insulin sensitivity was estimated by 4 methods: 1) the insulin sensitivity index 158 IS 0-∞ ; 2) the postprandial insulin sensitivity index (ISI 0-∞ ) for glycemia [ISI(G) 0-∞ ]; and 3) the 159 ISI 0-∞ for blood FFAs [ISI(FFA) 0-∞ ]. The infinity symbol represents the values at any 160 postprandial time. The details of the equations and additional references using this 161 multisampling protocol were previously described. 7 162 163 Statistical analyses 164 The summary data (the fasting and postprandial response) were analyzed by using one-factor 165 repeated-measures ANOVA. The incremental AUC (iAUC) was calculated by the trapezoidal 166 rule. The postprandial time courses after the test meals were analyzed by using 2-factor 167 repeated-measures ANOVA, and Bonferroni correction was applied for the post hoc detection 168 of significant pairwise differences. When indicated, hypothesis testing for differences 169 between groups was performed by the one-sample t test for independent samples. The data 170 were analyzed by using STATVIEW for WINDOWS (SAS Institute, Cary, NC). P < 0.05 was 171 considered significant. 172 173 Page 7 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 8 RESULTS 174 Basal conditions of participants 175 There were no significant differences between fasting values for plasma glucose, insulin, C-176 peptide, triglycerides, and FFAs at the beginning of each of the four intervention periods 177 (Table S2 in the Supporting Information). Neither were differences in fasting plasma total 178 cholesterol, HDL-cholesterol, and LDL-cholesterol (data not shown). Likewise, we found no 179 significant differences in the basal values for HOMA-B, DI 0 , HOMA-IR, rQUICKI, and the 180 basal Belfiore indices for glycemia and blood FFAs (Table S3 in the Supporting Information). 181 These data indicated that all of the participants had similar basal β-cell function and insulin 182 sensitivity prior to the ingestion of the high-fat meals and the pharmacological dose of 183 immediate-release niacin. 184 185 Effect of co-ingestion of immediate-release niacin with meals on plasma lipids, glucose, 186 insulin, and C-peptide 187 The plasma concentrations of postprandial glucose, insulin, C-peptide, triglycerides, FFAs, 188 total cholesterol, HDL-cholesterol, and LDL-cholesterol after the meal with no fat and the 189 meals enriched in SFAs, MUFAs or MUFAs+omega-3 LCPUFAs plus niacin are shown in 190 Figures 1A-1H. As expected, no changes in plasma triglycerides were observed when fat was 191 not included in the meal. The plasma total cholesterol, HDL-cholesterol, and LDL-cholesterol 192 were also not affected during the postprandial period. The high-fat meals markedly increased 193 (P < 0.05) plasma triglycerides to a peak at 60 min (3.02 ± 0.26 mmol/l with SFAs, 3.05 ± 194 0.18 mmol/l with MUFAs, and 2.87 ± 0.23 mmol/l with MUFAs+omega-3 LCPUFAs vs. 195 1.79 ± 0.16 mmol/l with no fat), and the iAUC values (0-6 h) for triglycerides also increased 196 (+544% with SFAs, +428% with MUFAs, and +408% with MUFAs+omega-3 LCPUFAs vs. 197 no fat, 100%) (Table S2 in the Supporting Information). All of the meals co-administered 198 Page 8 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 9 with niacin induced a decrease in plasma FFAs at 240 min (174 ± 32 µmol/l with SFAs, 173 ± 199 25 µmol/l with MUFAs, and 159 ± 64 µmol/l with MUFAs+omega-3 LCPUFAs vs. 138 ± 22 200 µmol/l with no fat, P < 0.05). Thereafter, plasma FFAs rebounded but not reaching basal 201 values. The high-fat meals increased (P < 0.05) the iAUC values (0-6 h) for FFAs (+120% 202 with SFAs, +111% with MUFAs, and +110% with MUFAs+omega-3 LCPUFAs vs. no fat, 203 100%) (Table S2 in the Supporting Information). The postprandial glucose response was 204 similar after ingestion of any of the meals plus niacin, increasing from a basal concentration 205 to a peak at 60 min (7.94 ± 1.39 mmol/l with SFAs, 8.51 ± 1.14 mmol/l with MUFAs, and 206 8.15 ± 1.01 mmol/l with MUFAs+omega-3 LCPUFAs vs. 8.11 ± 0.86 mmol/l with no fat, P > 207 0.05) and returning to basal values between 120 and 180 min. However, the high-fat meals 208 markedly increased (P < 0.05) plasma insulin to a peak at 60 min (452 ± 44 pmol/l with SFAs, 209 353 ± 15 pmol/l with MUFAs, and 352 ± 23 pmol/l with MUFAs+omega-3 LCPUFAs vs. 210 301 ± 26 pmol/l with no fat), and the iAUC values (0-6 h) for insulin also increased (+812% 211 with SFAs, +474% with MUFAs, and +493% with MUFAs+omega-3 LCPUFAs vs. no fat, 212 100%) (Table S2 in the Supporting Information). These effects of fat-enriched meals on 213 insulin were close to those observed on C-peptide, with a marked increase (P < 0.05) to a 214 peak at 60 min (2608 ± 229 pmol/l with SFAs, 2219 ± 120 pmol/l with MUFAs, and 2235 ± 215 155 pmol/l with MUFAs+omega-3 LCPUFAs vs. 2029 ± 136 pmol/l with no fat), and the 216 iAUC values (0-6 h) for C-peptide also increased (+442% with SFAs, +271% with MUFAs, 217 and +253% with MUFAs+omega-3 LCPUFAs vs. no fat, 100%) (Table S2 in the Supporting 218 Information). 219 220 Effect of co-ingestion of immediate-release niacin with meals on β ββ β-cell function and 221 insulin sensitivity 222 Page 9 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 16 14. Bermudez B, Pedroche JJ, Varela LM et al, inventors. Emulsified composition of 371 saturated fat in water, its preparation and its use for assessing tolerance to triglycerides. 372 Patent PCT/ES2014/070427, WO/2014/191597 (2014). 373 15. Cobelli C, Toffolo GM, Man CD et al. Assessment of β-cell function in humans, 374 simultaneously with insulin sensitivity and hepatic extraction, from intravenous and oral 375 glucose tests. Am J Physiol Endocrinol Metab 293:1-15 (2007). 376 16. Retnakaran R, Shen S, Hanley AJ, Vuksan V, Hamilton JK and Zinman B. Hyperbolic 377 relationship between insulin secretion and sensitivity on oral glucose tolerance test. 378 Obesity 16:1901-1907 (2008). 379 17. Pillarisetti S. Potential drug combinations to reduce cardiovascular disease burden in 380 diabetes. Trends Pharmacol Sci 37:207-219 (2016). 381 18. Pedrini MT, Niederwanger A, Kranebitter M et al. Postprandial lipaemia induces an 382 acute decrease of insulin sensitivity in healthy men independently of plasma NEFA levels. 383 Diabetologia 49:1612-1618 (2006). 384 19. Rijkelijkhuizen JM, Girman CJ, Mari A et al. Classical and model-based estimates of 385 beta-cell function during a mixed meal vs. an OGTT in a population-based cohort. 386 Diabetes Res Clin Pract 83:280-288 (2009). 387 20. Khoury DE, Hwalla N, Frochot V, Lacorte JM, Chabert M and Kalopissis AD. 388 Postprandial metabolic and hormonal responses of obese dyslipidemic subjects with 389 metabolic syndrome to test meals, rich in carbohydrate, fat or protein. Atherosclerosis 390 210:307-313 (2010). 391 21. Monnier L, Colette C, Dunseath GJ and Owens DR. The loss of postprandial glycemic 392 control precedes stepwise deterioration of fasting with worsening diabetes. Diabetes Care 393 30:263-269 (2007). 394 Page 16 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 17 22. Lopez S, Bermudez B, Abia R and Muriana FJ. The influence of major dietary fatty acids 395 on insulin secretion and action. Curr Opin Lipidol 21:15-20 (2010). 396 23. Rasmussen O, Lauszus FF, Christiansen C, Thomsen C and Hermansen K. Differential 397 effects of saturated and monounsaturated fat on blood glucose and insulin responses in 398 subjects with non-insulin-dependent diabetes mellitus. Am J Clin Nutr 63:249-253 (1996). 399 24. Blond E, Rieusset J, Alligier M. et al. Nicotinic acid effects on insulin sensitivity and 400 hepatic lipid metabolism: an in vivo to in vitro study. Horm Metab Res 46:390-396 401 (2014). 402 25. Koh Y, Bidstrup H and Nichols DL. Niacin increased glucose, insulin, and C-peptide 403 levels in sedentary nondiabetic postmenopausal women. Int J Womens Health 6:913-920 404 (2014). 405 26. Malin SK, Finnegan S, Fealy CE et al. β-Cell dysfunction is associated with metabolic 406 syndrome severity in adults. Metab Syndr Relat Disord 12:79-85 (2014). 407 27. Nakamura K, Miyoshi T, Yunoki K and Ito H. Postprandial hyperlipidemia as a potential 408 residual risk factor. J Cardiol 67:335-339 (2016). 409 28. Robertson MD, Jackson KG, Fielding BA, Williams CM and Frayn KN. Acute effects of 410 meal fatty acid composition on insulin sensitivity in healthy post-menopausal women. Br 411 J Nutr 88:635-640 (2002). 412 29. Jackson KG, Wolstencroft EJ, Bateman PA, Yaqoob P and Williams CM. Acute effects 413 of meal fatty acids on postprandial NEFA, glucose and apo E response: implications for 414 insulin sensitivity and lipoprotein regulation? Br J Nutr 93:693-700 (2005). 415 30. Xiao C, Giacca A, Carpentier A and Lewis GF. Differential effects of monounsaturated, 416 polyunsaturated and saturated fat ingestion on glucose-stimulated insulin secretion, 417 sensitivity and clearance in overweight and obese, non-diabetic humans. Diabetologia 418 49:1371-1379 (2006). 419 Page 17 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 18 31. Campos H, Khoo C and Sacks FM. Diurnal and acute patterns of postprandial 420 apolipoprotein B-48 in VLDL, IDL, and LDL from normolipidemic humans. 421 Atherosclerosis 181:345-351 (2005). 422 32. Lamon-Fava S, Diffenderfer MR, Barrett PH et al. Extended-release niacin alters the 423 metabolism of plasma apolipoprotein (Apo) A-I and ApoB-containing lipoproteins. 424 Arterioscler Thromb Vasc Biol 28:1672-1678 (2008). 425 33. Campbell PJ, Carlson MG, Hill JO and Nurjhan N. Regulation of free fatty acid 426 metabolism by insulin in humans: role of lipolysis and reesterification. Am J Physiol 427 263:E1063-E1069 (1992). 428 34. Carpentier AC, Frisch F, Cyr D et al. On the suppression of plasma nonesterified fatty 429 acids by insulin during enhanced intravascular lipolysis in humans. Am J Physiol 430 Endocrinol Metab 289:E849-E856 (2005). 431 35. Ali AH, Mundi M, Koutsari C, Bernlohr DA and Jensen MD. Adipose tissue free fatty 432 acid storage in vivo: effects of insulin versus niacin as a control for suppression of 433 lipolysis. Diabetes 64:2828-2835 (2015). 434 36. McQuaid SE, Humphreys SM, Hodson L, Fielding BA, Karpe F and Frayn KN. Femoral 435 adipose tissue may accumulate the fat that has been recycled as VLDL and nonesterified 436 fatty acids. Diabetes 59:2465-2473 (2010). 437 37. Jackson KG, Wolstencroft EJ, Bateman PA, Yaqoob P and Williams CM. Increased 438 enrichment of triacylglycerol-rich lipoproteins with apolipoproteins E and C-III 439 following saturated compared with unsaturated fatty acid-rich meals. Am J Clin Nutr 440 81:25-34 (2005). 441 442 443 444 Page 18 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 19 Table 1. Baseline clinical and biochemical characteristics of participants. Value Participants, n 16 Age (y) 38.5 ± 4.3 Examination Body mass index (kg/m 2 ) 31.8 ± 4.2 Waist circumference (cm) 124.5 ± 16.1 Waist-to-hip ratio 1.18 ± 0.12 Systolic blood pressure (mmHg) 134.1 ± 8.2 Diastolic blood pressure (mmHg) 84.9 ± 8.8 Investigations Total cholesterol (mmol/l) 5.2 ± 0.5 HDL-cholesterol (mmol/l) 1.01 ± 0.11 LDL-cholesterol (mmol/l) 3.49 ± 0.28 Triglycerides (mmol/l) 1.49 ± 0.16 Glucose (mmol/l) 5.34 ± 0.65 Insulin (pmol/l) 135 ± 18 C-peptide (pmol/l) 1192 ± 120 HbA 1c (%) a 5.2 ± 0.4 Data are mean ± SD. These values were not statistical different when compared with those obtained in each occasion before the ingestion of the dose of immediate-release niacin and the corresponding high-fat meal or the meal with no fat. a 33.0 ± 0.2 in mmol/mol. Page 19 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 20 Table 2. Postprandial indices of β-cell function (IGI, IGI/HOMA-IR, AUC INS /AUC GLU , and DI 120 ) and insulin sensitivity [IS 0-∞ , ISI(G) 0-∞ , and ISI(NEFA) 0-∞ ] after the test meals coadministered with a single-dose of immediate-release niacin. Meal with no fat SFA meal MUFA meal MUFA + omega-3 LCPUFA meal IGI 0-60 min a 60.4 ± 6.5 130.1 ± 15.9* 75.8 ± 10.2*† 85.2 ± 11.1*† IGI/HOMA-IR b 1.89 ± 0.19 4.46 ± 0.55* 2.52 ± 0.26*† 2.78 ± 0.37*† AUC INS /AUC GLU 0-120 min a 31.4 ± 2.8 46.2 ± 4.1* 34.7 ± 2.5*† 37.2 ± 3.9*† DI 120 0-120 min c 85.3 ± 6.1 106.9 ± 8.5* 92.0 ± 5.6*† 94.8 ± 7.9*† IS 0-∞ 0-360 min d 486 ± 37 138 ± 13* 225 ± 16*† 216 ± 20*† Postprandial - Belfiore index for glycemia [ISI(G) 0-∞ ] 0-360 min c 1.00 ± 0.08 0.68 ± 0.06* 0.80 ± 0.07*† 0.78 ± 0.08*† Postprandial-Belfiore index for blood NEFA [ISI(NEFA) 0-∞ ] 0-360 min c 1.00 ± 0.09 0.55 ± 0.05* 0.68 ± 0.06*† 0.69 ± 0.09*† Data are mean ± SD (n = 16). Statistical differences are based on repeated-measures ANOVA with a Bonferroni correction. *P < 0.05 vs. meal with no fat; †P < 0.05 vs. SFA meal; a pmol/mmol; b l/mmol; c no units; d min -1 × dl × kg -1 /µU/ml. Page 20 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 21 Figure 1. Postprandial graphs for (A) glucose, (B) insulin, (C) C-peptide, (D) triglycerides, (E) FFAs, (F) total cholesterol, (G) HDL-cholesterol, and (H) LDL-cholesterol after the ingestion of the dose of immediate-release niacin and the meal with no fat, control meal Minutes mmol/l Total cholesterol mmol/l HDL-cholesterol mmol/l LDL-cholesterol Minutes A B C D E F G H Glucose mmol/l µmol/l FFAs * * † † * * * * C-peptide pmol/l * * * † * † * * † * * pmol/l Insulin * † * * * * † * † * mmol/l * * † * * † * † * * † Triglycerides * * 0 0.5 1 1.5 2 2.5 3 3.5 0 60 120 180 240 300 360 * * † 0 500 1000 1500 2000 2500 3000 * 0 100 200 300 400 500 * 0 2 4 6 8 10 0 100 200 300 400 500 * † * 0 2 4 6 8 0 2 4 6 8 0 2 4 6 8 0 60 120 180 240 300 360 Page 21 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 22 (white circle), rich in SFAs (black circle), rich in MUFAs (white square) or rich in MUFAs+omega-3 LCPUFAs (black square). *P < 0.05 between the high-fat meals and the control meal; †P < 0.05 between the MUFA meals and the SFA meal. The time effect and meal × time interaction were significant (P < 0.05), except for glucose, total cholesterol, HDL-cholesterol, and LDL-cholesterol. Page 22 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 23 TOC Graphic Page 23 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 24 SUPPORTING INFORMATION • Fatty acid composition of dietary fats. • Fasting values on triglycerides, free fatty acids, glucose, insulin, and C-peptide. • Fasting values on β-cell function and insulin action. Page 24 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review 1 ONLINE SUPPORTING INFORMATION Supporting Table 1. Fatty acid composition of dietary fats. Cow's milk cream Olive oil Olive oil + omega-3 LCPUFAs Fatty acid g/100 g of fatty acid 4:0, butyric 0.83 ± 0.16 - - 6:0, caproic 0.25 ± 0.02 - - 8:0, caprylic 0.61 ± 0.07 - - 10:0, capric 2.47 ± 0.13 - - 12:0, lauric 3.09 ± 0.42 - - 14:0, myristic 10.9 ± 0.91 - - 16:0, palmitic 35.50 ± 0.82 20.44 ± 0.89 20.48 ± 0.64 16:1(ω-7), palmitoleic 3.60 ± 0.32 0.97 ± 0.17 0.82 ± 0.12 18:0, stearic 11.54 ± 0.75 5.70 ± 0.11 4.49 ± 0.36 18:1(ω-9), oleic 25.33 ± 0.71 61.90 ± 1.23 61.51 ± 0.97 18:2(ω-6), linoleic 4.27 ± 0.82 7.97 ± 0.65 8.04 ± 0.53 18:3(ω-3), α-linolenic 0.39 ± 0.05 1.04 ± 0.13 0.94 ± 0.03 20:5(ω-3), eicosapentaenoic - - 0.92 ± 0.09 22:6(ω-3), docosahexaenoic - - 0.72 ± 0.10 Others 0.96 ± 0.42 2.05 ± 1.08 2.01 ± 0.88 Data are mean ± SD (n = 3). Page 25 of 29 [email protected] Journal of the Science of Food and Agriculture 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60