- Email: [email protected]
Effect of dietary omega-3 fatty acids on high-density lipoprotein apolipoprotein AI kinetics in type II diabetes mellitus
Atherosclerosis 157 (2001) 131– 135 www.elsevier.com/locate/atherosclerosis
Effect of dietary omega-3 fatty acids on high-density lipoprotein apolipoprotein AI kinetics in type II diabetes mellitus R. Fre´nais a, K. Ouguerram a, C. Maugeais a, P. Mahot b, B. Charbonnel b, T. Magot a, M. Krempf a,b,* a
Centre de Recherche en Nutrition Humaine, Groupe Me´tabolisme, INSERM U539, Hoˆtel Dieu, Nantes, France b Clinique d’Endocrinologie, Maladies Me´taboliques et Nutrition, Hoˆtel Dieu, Nantes, France Received 25 April 2000; received in revised form 22 September 2000; accepted 19 October 2000
Abstract The effect of a dietary fish oil supplementation on metabolism of HDL was studied in type II diabetes mellitus. Endogenous labeling of HDL-apo AI was performed using a 14 h primed infusion of D3-leucine in five diabetic patients before and 2 months after treatment with maxEPA®. Isotopic enrichment curves were analyzed using a monoexponential function. After treatment, plasma cholesterol level remained unchanged (205.4 941.9 vs. 206.8 9 30.7 mg/dl, NS), whereas plasma triglycerides were decreased (155.4 967.9 vs. 202.6 932.2 mg/dl, P =0.06). Plasma apo AI was similar under maxEPA® (116.09 25.6 vs. 111.8 9 25.4 mg/dl, NS), and HDL-cholesterol and HDL-triglycerides were also not markedly changed (30.2 910.0 vs. 27.1 910 mg/dl, and 15.3 9 9.8 vs. 19.2 9 10.4 mg/dl, NS). HDL-apo AI fractional catabolic rate (FCR) and absolute production rate (APR) were significantly decreased after treatment with maxEPA® (0.2790.09 vs. 0.37 9 0.08 pool day, PB 0.05, and 12.1 92.8 vs. 16.1 9 3.3 mg/kg per day, PB0.05). These findings showed an effect of maxEPA® on kinetics of apolipoprotein AI in type II diabetes mellitus, probably linked to changes in plasma triglyceride level. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: n-3 polyunsaturated acids; HDL; apo AI; Kinetic; Type II diabetes mellitus
1. Introduction Regular fish consumption has been correlated with a reduced mortality from coronary artery disease [1]. This event has been ascribed to the beneficial effect of two major long chain polyunsaturated omega-3 fatty acid (v3-FA); eicosapentanoic acid (EPA, C20:5 v3), and docosahexaenoic acid (DHA, C22:6 v3). In patients with diabetes mellitus, plasma triglycerides were decreased and total cholesterol remain unchanged under v3-FA [2–6]. Furthermore, most intervention studies [2 – 4,7], but not all [5,6] showed no effect of v3-FA on the total HDL-cholesterol level. Discrepancies in lipid profile may be attributed to the characteristics of patients, the lack of control group, the amount of treatment administrated or the length of the study.
Furthermore, THE effect of v3-FA on lipoprotein metabolism was partly established, yielding a reduced hepatic synthesis of VLDL-apo B100 [8]. However, their effects on HDL metabolism are not documented. We already demonstrated that the decrease of plasma apo AI level in type II diabetes was due to the increase of HDL catabolic rate, closely linked to plasma triglyceride level and HDL composition [9]. Besides, HDL clearance is related to plasma triglyceride concentration [9–11]. In this study, we have therefore tested the hypothesis that the reduction of plasma triglyceride level, achieved by dietary fish oil supplementation, may play a direct role in the kinetic aspects of HDL metabolism in type II diabetic patients.
2. Subjects, materials and methods * Corresponding author. Tel.: + 33-240-083642; fax: +33-240083079. E-mail address: [email protected] (M. Krempf).
Five type II diabetic patients were recruited, including four non menopausal women at the beginning of
0021-9150/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S0021-9150(00)00723-1
132
R. Fre´ nais et al. / Atherosclerosis 157 (2001) 131–135
their menstrual cycle (Table 1). They were not taking any medication that could affect carbohydrate or lipid for at least one month before study, and were instructed by a dietician to eat a weight-maintenance diet (50% carbohydrates, 35% fat and 15% protein), for 1 month before the study. The experimental protocol was approved by the Ethical Committee of Nantes University Hospital, and informed consent was obtained before the study was started. Patients underwent a basal kinetic study the day before the beginning of the treatment. Then they received six capsules v3-FA per day with meals for 8 weeks (maxEPA®, Pierre Fabre Sante, Castres, France). Each capsule contained 1000 mg methyl ester fatty acids, providing 180 mg eicosapentaenoic acid (EPA, 20:5 v3), 120 mg docosahexaenoic acid (DHA, 22:6 v3) and 1.75 mg alpha tocopherol acetate. The second kinetic was carried out at the end of treatment period. The kinetic protocol was similar to that previously described [9]. Briefly, the endogenous labeling of apolipoprotein AI was performed by administration of a primed constant infusion of 10 mmol/kg per h L[5,5,5-2H3]-leucine (Cambridge Isotope Laboratories, Andover, MA, USA) for 14 h. All the subjects fasted overnight before the study, and remained fasting during the entire protocol. Venous blood samples were withdrawn in EDTA tubes at baseline and at regular times until the end of the infusion. VLDL were isolated by a sequential ultracentrifugation (Himac CP70, Hitachi). HDL were then isolated by a density gradient ultracentrifugation (Centrikon T 2060, Kontron Instruments). Plasma and HDL cholesterol and triglyceride levels were measured using enzymatic kits (Boehringer Mannheim GmbH, Germany). Plasma apo AI concentration was measured by immunonephelometry (Behring, Rueil Malmaison, France). The apo AI pool size was calculated by multiplying the mean plasma apo AI concentration by 0.0380.041, assuming a plasma volume of 3.8– 4.1% of body weight according to age, gender and body weight of each study subject. HDL-apo AI and VLDL-apo B100 were isolated by SDS-PAGE and were hydrolysed with HCl (Sigma, St Quentin Fallavier, France). Hydrolysates were purified by cation exchange chromatography. Leucine was derivatized and isotopic ratios were analyzed by electron-impact gas chromatography-mass spectrometry [9]. Calculations of apo AI kinetic parameters were based on the tracer-to-tracee mass ratio. We applied a one-compartment model to HDL-apo AI and VLDL-apo B100 enrichment curves (SAAM II, Resource Facility for Kinetic Analysis, SAAM Institute, Seattle, WA). The plateau of VLDL-apo B100 tracer-to-tracee ratio was used as precursor pool enrichment, assuming that apo B100 and the majority of apo AI are synthesized by the liver. We estimated k the
fractional catabolic rate (FCR, per day) and the absolute production rate (APR, mg/kg per day). APR was the product of FCR multiplied by apo AI pool size in HDL. We used the non-parametric Wilcoxon’s signed ranktest to determine significant differences before and after v3-FA treatment (two-tailed). Furthermore, maxEPA® being considered as an hypotriglyceridemic treatment, we reported a one-tailed P value for plasma triglyceride levels.
3. Results Treatment with maxEPA® did not lead to significant variations of body mass index (−1%), glycated haemoglobin level (− 4%), fasting blood glucose level (− 6%) or insulinemia (− 16%) (Table 1). Whereas v3-FA acids did not change total plasma cholesterol level (−1%), plasma triglyceride level was markedly decreased (− 23%, P= 0.06). The v3-FA treatment did not significantly change HDL composition, despite a trend to a higher HDL-cholesterol level (+11%), and a decreased HDL-triglycerides level (− 20%). Apo AI level was also similar (+ 4%) (Table 1). As previously reported [9], a plateau of tracer-totracee ratio was observed for VLDL-apo B100 but not for apo AI-HDL, meaning a slow synthetic rate for this apolipoprotein (data not shown). Mean HDL-apo AI fractional catabolic rate (FCR) and absolute production rate (APR) were significantly decreased after treatment with maxEPA® (0.279 0.09 vs. 0.379 0.08 pool per day, PB 0.05, and 12.192.8 vs. 16.193.3 mg/kg per day, PB 0.05) (Table 1).
4. Discussion We have studied the effect of reduction of plasma triglyceride level, achieved by a dietary fish oil supplementation, on the kinetic aspects of HDL metabolism in type II diabetes. After 2 months of treatment, patients showed a marked reduction of plasma triglyceride concentration, whereas, plasma apoA1 and HDLcholesterol remained unchanged, as already reported [2–5]. Both the clearance and production rates of HDL-apo AI decreased significantly after treatment in all the patients. Although in the range of inclusion criteria (type II diabetes mellitus, plasma TG level 1.5–5 g/l and HbA1c 512%), the small-sized study population was to some extent heterogeneous and this could constitute a limitation of the study. Patients also showed heterogeneity of response to the treatment regarding HDL components, probably explaining the lack of significant difference with treatment. We have applied a one-com-
Table 1 Clinical characteristics, plasma and HDL cholesterol and triglycerides levels, and HDL-apo AI kinetic parameters in type II diabetic patients before (base) and after (post) dietary fish oil supplementationa
c1 c2 c3 c4 c5
Age (years)
Base Post Base Post Base Post Base Post Base Post Mean base S.D. Mean post S.D.
37 54 55 36 65 49 (13)
BMI (kg/m2)
HbA1c (%)
Insulin (mU/l)
Glucose (mmol/l)
Plasma CH (mg/dl)
Plasma TG (mg/dl)
HDL-CH (mg/dl)
HDL-TG (mg/dl)
Apo AI (mg/dl)
FCR (pool per day)
APR (mg/kg per day)
32.2 30.1 31.8 32.4 26.6 27.0 33.3 32.9 29.0 28.6 30.6 (2.7) 30.2 (2.5)
6.1 5.2 5.2 5.6 5.1 5.9 8.7 9.1 11.1 8.6 7.2 (2.6) 6.9 (1.8)
16.4 6.5 ND ND 26.9 22.0 12.0 15.1 17.8 17.9 18.3 (6.3) 15.4 (6.6)
8.7 6.1 7.1 7.6 7.7 7.1 13.0 13.3 11.3 10.5 9.5 (2.5) 8.9 (3.0)
212 159 204 214 251 219 165 171
214 72 186 177 193 144 168 128 202 256 202.6 (32.2) 155.4b (67.9)
34.7 27.8 15.8 19.1 25.3 45.2 20.1 24.6 39.7 34.2 27.1 (10.0) 30.2 (10.0)
11.7 8.4 33.4 31.4 17.6 17.6 7.6 8.6 25.5 10.4 19.2 (10.4) 15.3 (9.8)
94 80 104 99 152 142 120 128 89 131 111.8 (25.4) 116.0 (25.6)
0.48 0.40 0.31 0.30 0.34 0.28 0.29 0.19 0.43 0.19 0.37 0.08 0.27c 0.09
17.2 12.5 12.8 11.7 21.3 16.6 14.0 9.9 15.2 9.7 16.1 3.3 12.1c 2.8
264 206.8 (30.7) 205.4 (41.9)
R. Fre´ nais et al. / Atherosclerosis 157 (2001) 131–135
Subject
a
CH, cholesterol; TG, triglyceride; FCR, fractional catabolic rate; APR, absolute production rate. P =0.06. c PB0.05, post versus base. b
133
134
R. Fre´ nais et al. / Atherosclerosis 157 (2001) 131–135
partment to our enrichment data, which appears the most relevant under our study conditions (constant infusion) compared with munch complex model, and allows the study of apo AI overall catabolism. The lack of change in plasma apo AI level after fish oil supplementation, already reported [2,7], was in agreement with both decrease of production and catabolic rates of apoA1-HDL. The benefic effect of MaxEPA® is validated since the production of HDL-apoA1 was similar to control value (12.0 9 4.2 mg/kg per day) we have previously reported [9]. However, the decrease of HDL-apo AI clearance rate, although significant, was not strong enough to reach normal values (0.2190.06 pool per day). This could be an explanation of the low plasma apo AI level observed at the end of the study. But these values at the basal state were in agreement with previous studies on HDL kinetics, underlying the accuracy of our method. A decrease in both HDL-apo AI FCR and APR was observed in every study subjects, indicating that v3 fatty acids had an impact on HDL metabolic. HDL production was adapted to the increase of catabolism before and after fish oil supplementation. This is however not the case for all patients with glucose impaired tolerance (IGT) or type II diabetes mellitus [9,12], who showed no change in HDL absolute production rate. This would suggest that HDL production could depend on several genetic and environmental factors, which may finally control plasma apo AI concentration. We already reported that the FCR of HDL increased in type II diabetes, and related to plasma and HDLtriglycerides levels [9]. The present study was therefore designed to evaluate whether plasma triglycerides directly — or secondarily via HDL-triglyceride level — could act upon HDL clearance rate. A reduction of plasma triglyceride level, achieved by means of v3-FA, may induce a change in HDL composition, and particularly a decrease of HDL-triglyceride level. Several studies have indicated that HDL catabolism was related to HDL-triglyceride concentration or HL activity [13,14]. Thus, a reduction may be associated with a decrease of HDL clearance. However, this probably does not occur in this study since maxEPA® did not induce a clear-cut reduction of HDL-triglycerides, which was not consequently significantly correlated with HDL-apo AI catabolism (data not shown). This lack of change in HDL composition is in agreement with the unaffected HL activity and the increased LPL activity [7], as well as the decrease of VLDL production [8] after fish oil supplementation. In contrast, a lowered plasma triglyceride level, achieved by means of v3-FA, induces a decrease of HDL clearance rate. Therefore, these results suggest that plasma triglyceride level per se may play a role in the control of HDL metabolism. Nevertheless, definite conclusions are precluded by the small number of patients studied.
Then, maxEPA® induces changes in HDL catabolism and this could be added to the beneficial cardiovascular effect of this treatment. These current kinetic data, even observed on a small-sized study group, may provide new insights about the mechanisms of action of fish oil, which remain open to further investigations.
Acknowledgements We thank I. Grit, C. Le Valegant, and P. Mauge`re for their excellent technical assistance and D. Darmaun for his review and advice. This study was supported by grants from the department of Clinical Research of Nantes Hospital and from Pierre Fabre Sante Laboratories.
References [1] Kromhout D, Bosscheiter EB, de Lezenne Coulander C. The inverse relation between fish consumption and 20-year mortality from coronary heart disease. New Engl J Med 1985;312:1205 –9. [2] Schectman G, Kaul S, Kissebah H. Effect of fish oil concentrate on lipoprotein composition in NIDDM. Diabetes 1988;37:1567 – 73. [3] Goh YK, Jumpsen MA, Ryan EA, Clandinin MT. Effects of w-3 fatty acids on plasma lipids, cholesterol and lipoproteins fatty acid content in NIDDM patients. Diabetologia 1997;40:45 –52. [4] Morgan WA, Raskin P, Rosenstock J. A comparison of fish oil or corn oil supplements in hyperlipidemic subjects with NIDDM. Diabetes Care 1995;18:83 – 6. [5] Dunstan DW, Mori TA, Puddey IB, Burke V, Morton AR, Stanton KG. The independent and combined effects of aerobic exercise and dietary fish intake on serum lipids and glycemic control in NIDDM: a randomized controlled study. Diabetes Care 1998;20:913 – 21. [6] Holub BJ, Bakker DJ, Skeaff CM. Alterations in molecular species of cholesterol esters formed via lecithin-cholesterol acyltransferase in human subjects consuming fish oil. Atherosclerosis 1987;66:11 – 8. [7] Kasim SE, Stern B, Khilnani S, McLin P, Baciorowski S, Jen KLC. Effects of w 3 fish oil on lipid metabolism, glycemic control, and blood pressure in type II diabetic patients. J Clin Endocrinol Metab 1988;67:1 – 5. [8] Fisher WR, Zech LA, Stacpoole PW. Apolipoprotein B metabolism in hypertriglyceridemic diabetic patients administered either a fish oil- or vegetable oil-enriched diet. J Lipid Res 1998;39:388 – 401. [9] Fre´ nais R, Ouguerram K, Maugeais C, et al. High density lipoprotein apolipoprotein AI kinetics in NIDDM: a stable isotope study. Diabetologia 1997;40:578 – 83. [10] Brinton EA, Eisenberg S, Breslow JL. Increased apo AI and apo AII fractional catabolic rate in patients with low high density lipoprotein-cholesterol levels with or without hypertriglyceridemia. J Clin Invest 1991;87:536 – 44. [11] Nikkila¨ EA, Taskinen MR, Sane T. Plasma high density lipoprotein concentration and subfraction in relation to triglyceride metabolism. Am Heart J 1987;113:543 – 8. [12] Pietzch J, Julius U, Nitzsche S, Hanefeld M. In vivo evidence for increased apolipoprotein A-I catabolism in subjects with impaired glucose tolerance. Diabetes 1998;47:1928 – 34.
R. Fre´ nais et al. / Atherosclerosis 157 (2001) 131–135 [13] Saku K, Gartside PS, Hynd BA, Mendoza SG, Kashyap ML. Apolipoprotein AI and AII metabolism in patients with primary high density lipoprotein deficiency associated with familial hypertriglyceridemia. Metabolism 1985;34:754 –64.
.
135
[14] Brinton EA, Eisenberg S, Breslow JL. Human HDL cholesterol levels are determined by apo AI fractional catabolic rate, which correlates inversely with estimates of HDL particle size. Arterioscler Thromb 1994;14:707 – 20.
Effect of dietary omega-3 fatty acids on high-density lipoprotein apolipoprotein AI kinetics in type II diabetes mellitus R. Fre´nais a, K. Ouguerram a, C. Maugeais a, P. Mahot b, B. Charbonnel b, T. Magot a, M. Krempf a,b,* a
Centre de Recherche en Nutrition Humaine, Groupe Me´tabolisme, INSERM U539, Hoˆtel Dieu, Nantes, France b Clinique d’Endocrinologie, Maladies Me´taboliques et Nutrition, Hoˆtel Dieu, Nantes, France Received 25 April 2000; received in revised form 22 September 2000; accepted 19 October 2000
Abstract The effect of a dietary fish oil supplementation on metabolism of HDL was studied in type II diabetes mellitus. Endogenous labeling of HDL-apo AI was performed using a 14 h primed infusion of D3-leucine in five diabetic patients before and 2 months after treatment with maxEPA®. Isotopic enrichment curves were analyzed using a monoexponential function. After treatment, plasma cholesterol level remained unchanged (205.4 941.9 vs. 206.8 9 30.7 mg/dl, NS), whereas plasma triglycerides were decreased (155.4 967.9 vs. 202.6 932.2 mg/dl, P =0.06). Plasma apo AI was similar under maxEPA® (116.09 25.6 vs. 111.8 9 25.4 mg/dl, NS), and HDL-cholesterol and HDL-triglycerides were also not markedly changed (30.2 910.0 vs. 27.1 910 mg/dl, and 15.3 9 9.8 vs. 19.2 9 10.4 mg/dl, NS). HDL-apo AI fractional catabolic rate (FCR) and absolute production rate (APR) were significantly decreased after treatment with maxEPA® (0.2790.09 vs. 0.37 9 0.08 pool day, PB 0.05, and 12.1 92.8 vs. 16.1 9 3.3 mg/kg per day, PB0.05). These findings showed an effect of maxEPA® on kinetics of apolipoprotein AI in type II diabetes mellitus, probably linked to changes in plasma triglyceride level. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: n-3 polyunsaturated acids; HDL; apo AI; Kinetic; Type II diabetes mellitus
1. Introduction Regular fish consumption has been correlated with a reduced mortality from coronary artery disease [1]. This event has been ascribed to the beneficial effect of two major long chain polyunsaturated omega-3 fatty acid (v3-FA); eicosapentanoic acid (EPA, C20:5 v3), and docosahexaenoic acid (DHA, C22:6 v3). In patients with diabetes mellitus, plasma triglycerides were decreased and total cholesterol remain unchanged under v3-FA [2–6]. Furthermore, most intervention studies [2 – 4,7], but not all [5,6] showed no effect of v3-FA on the total HDL-cholesterol level. Discrepancies in lipid profile may be attributed to the characteristics of patients, the lack of control group, the amount of treatment administrated or the length of the study.
Furthermore, THE effect of v3-FA on lipoprotein metabolism was partly established, yielding a reduced hepatic synthesis of VLDL-apo B100 [8]. However, their effects on HDL metabolism are not documented. We already demonstrated that the decrease of plasma apo AI level in type II diabetes was due to the increase of HDL catabolic rate, closely linked to plasma triglyceride level and HDL composition [9]. Besides, HDL clearance is related to plasma triglyceride concentration [9–11]. In this study, we have therefore tested the hypothesis that the reduction of plasma triglyceride level, achieved by dietary fish oil supplementation, may play a direct role in the kinetic aspects of HDL metabolism in type II diabetic patients.
2. Subjects, materials and methods * Corresponding author. Tel.: + 33-240-083642; fax: +33-240083079. E-mail address: [email protected] (M. Krempf).
Five type II diabetic patients were recruited, including four non menopausal women at the beginning of
0021-9150/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S0021-9150(00)00723-1
132
R. Fre´ nais et al. / Atherosclerosis 157 (2001) 131–135
their menstrual cycle (Table 1). They were not taking any medication that could affect carbohydrate or lipid for at least one month before study, and were instructed by a dietician to eat a weight-maintenance diet (50% carbohydrates, 35% fat and 15% protein), for 1 month before the study. The experimental protocol was approved by the Ethical Committee of Nantes University Hospital, and informed consent was obtained before the study was started. Patients underwent a basal kinetic study the day before the beginning of the treatment. Then they received six capsules v3-FA per day with meals for 8 weeks (maxEPA®, Pierre Fabre Sante, Castres, France). Each capsule contained 1000 mg methyl ester fatty acids, providing 180 mg eicosapentaenoic acid (EPA, 20:5 v3), 120 mg docosahexaenoic acid (DHA, 22:6 v3) and 1.75 mg alpha tocopherol acetate. The second kinetic was carried out at the end of treatment period. The kinetic protocol was similar to that previously described [9]. Briefly, the endogenous labeling of apolipoprotein AI was performed by administration of a primed constant infusion of 10 mmol/kg per h L[5,5,5-2H3]-leucine (Cambridge Isotope Laboratories, Andover, MA, USA) for 14 h. All the subjects fasted overnight before the study, and remained fasting during the entire protocol. Venous blood samples were withdrawn in EDTA tubes at baseline and at regular times until the end of the infusion. VLDL were isolated by a sequential ultracentrifugation (Himac CP70, Hitachi). HDL were then isolated by a density gradient ultracentrifugation (Centrikon T 2060, Kontron Instruments). Plasma and HDL cholesterol and triglyceride levels were measured using enzymatic kits (Boehringer Mannheim GmbH, Germany). Plasma apo AI concentration was measured by immunonephelometry (Behring, Rueil Malmaison, France). The apo AI pool size was calculated by multiplying the mean plasma apo AI concentration by 0.0380.041, assuming a plasma volume of 3.8– 4.1% of body weight according to age, gender and body weight of each study subject. HDL-apo AI and VLDL-apo B100 were isolated by SDS-PAGE and were hydrolysed with HCl (Sigma, St Quentin Fallavier, France). Hydrolysates were purified by cation exchange chromatography. Leucine was derivatized and isotopic ratios were analyzed by electron-impact gas chromatography-mass spectrometry [9]. Calculations of apo AI kinetic parameters were based on the tracer-to-tracee mass ratio. We applied a one-compartment model to HDL-apo AI and VLDL-apo B100 enrichment curves (SAAM II, Resource Facility for Kinetic Analysis, SAAM Institute, Seattle, WA). The plateau of VLDL-apo B100 tracer-to-tracee ratio was used as precursor pool enrichment, assuming that apo B100 and the majority of apo AI are synthesized by the liver. We estimated k the
fractional catabolic rate (FCR, per day) and the absolute production rate (APR, mg/kg per day). APR was the product of FCR multiplied by apo AI pool size in HDL. We used the non-parametric Wilcoxon’s signed ranktest to determine significant differences before and after v3-FA treatment (two-tailed). Furthermore, maxEPA® being considered as an hypotriglyceridemic treatment, we reported a one-tailed P value for plasma triglyceride levels.
3. Results Treatment with maxEPA® did not lead to significant variations of body mass index (−1%), glycated haemoglobin level (− 4%), fasting blood glucose level (− 6%) or insulinemia (− 16%) (Table 1). Whereas v3-FA acids did not change total plasma cholesterol level (−1%), plasma triglyceride level was markedly decreased (− 23%, P= 0.06). The v3-FA treatment did not significantly change HDL composition, despite a trend to a higher HDL-cholesterol level (+11%), and a decreased HDL-triglycerides level (− 20%). Apo AI level was also similar (+ 4%) (Table 1). As previously reported [9], a plateau of tracer-totracee ratio was observed for VLDL-apo B100 but not for apo AI-HDL, meaning a slow synthetic rate for this apolipoprotein (data not shown). Mean HDL-apo AI fractional catabolic rate (FCR) and absolute production rate (APR) were significantly decreased after treatment with maxEPA® (0.279 0.09 vs. 0.379 0.08 pool per day, PB 0.05, and 12.192.8 vs. 16.193.3 mg/kg per day, PB 0.05) (Table 1).
4. Discussion We have studied the effect of reduction of plasma triglyceride level, achieved by a dietary fish oil supplementation, on the kinetic aspects of HDL metabolism in type II diabetes. After 2 months of treatment, patients showed a marked reduction of plasma triglyceride concentration, whereas, plasma apoA1 and HDLcholesterol remained unchanged, as already reported [2–5]. Both the clearance and production rates of HDL-apo AI decreased significantly after treatment in all the patients. Although in the range of inclusion criteria (type II diabetes mellitus, plasma TG level 1.5–5 g/l and HbA1c 512%), the small-sized study population was to some extent heterogeneous and this could constitute a limitation of the study. Patients also showed heterogeneity of response to the treatment regarding HDL components, probably explaining the lack of significant difference with treatment. We have applied a one-com-
Table 1 Clinical characteristics, plasma and HDL cholesterol and triglycerides levels, and HDL-apo AI kinetic parameters in type II diabetic patients before (base) and after (post) dietary fish oil supplementationa
c1 c2 c3 c4 c5
Age (years)
Base Post Base Post Base Post Base Post Base Post Mean base S.D. Mean post S.D.
37 54 55 36 65 49 (13)
BMI (kg/m2)
HbA1c (%)
Insulin (mU/l)
Glucose (mmol/l)
Plasma CH (mg/dl)
Plasma TG (mg/dl)
HDL-CH (mg/dl)
HDL-TG (mg/dl)
Apo AI (mg/dl)
FCR (pool per day)
APR (mg/kg per day)
32.2 30.1 31.8 32.4 26.6 27.0 33.3 32.9 29.0 28.6 30.6 (2.7) 30.2 (2.5)
6.1 5.2 5.2 5.6 5.1 5.9 8.7 9.1 11.1 8.6 7.2 (2.6) 6.9 (1.8)
16.4 6.5 ND ND 26.9 22.0 12.0 15.1 17.8 17.9 18.3 (6.3) 15.4 (6.6)
8.7 6.1 7.1 7.6 7.7 7.1 13.0 13.3 11.3 10.5 9.5 (2.5) 8.9 (3.0)
212 159 204 214 251 219 165 171
214 72 186 177 193 144 168 128 202 256 202.6 (32.2) 155.4b (67.9)
34.7 27.8 15.8 19.1 25.3 45.2 20.1 24.6 39.7 34.2 27.1 (10.0) 30.2 (10.0)
11.7 8.4 33.4 31.4 17.6 17.6 7.6 8.6 25.5 10.4 19.2 (10.4) 15.3 (9.8)
94 80 104 99 152 142 120 128 89 131 111.8 (25.4) 116.0 (25.6)
0.48 0.40 0.31 0.30 0.34 0.28 0.29 0.19 0.43 0.19 0.37 0.08 0.27c 0.09
17.2 12.5 12.8 11.7 21.3 16.6 14.0 9.9 15.2 9.7 16.1 3.3 12.1c 2.8
264 206.8 (30.7) 205.4 (41.9)
R. Fre´ nais et al. / Atherosclerosis 157 (2001) 131–135
Subject
a
CH, cholesterol; TG, triglyceride; FCR, fractional catabolic rate; APR, absolute production rate. P =0.06. c PB0.05, post versus base. b
133
134
R. Fre´ nais et al. / Atherosclerosis 157 (2001) 131–135
partment to our enrichment data, which appears the most relevant under our study conditions (constant infusion) compared with munch complex model, and allows the study of apo AI overall catabolism. The lack of change in plasma apo AI level after fish oil supplementation, already reported [2,7], was in agreement with both decrease of production and catabolic rates of apoA1-HDL. The benefic effect of MaxEPA® is validated since the production of HDL-apoA1 was similar to control value (12.0 9 4.2 mg/kg per day) we have previously reported [9]. However, the decrease of HDL-apo AI clearance rate, although significant, was not strong enough to reach normal values (0.2190.06 pool per day). This could be an explanation of the low plasma apo AI level observed at the end of the study. But these values at the basal state were in agreement with previous studies on HDL kinetics, underlying the accuracy of our method. A decrease in both HDL-apo AI FCR and APR was observed in every study subjects, indicating that v3 fatty acids had an impact on HDL metabolic. HDL production was adapted to the increase of catabolism before and after fish oil supplementation. This is however not the case for all patients with glucose impaired tolerance (IGT) or type II diabetes mellitus [9,12], who showed no change in HDL absolute production rate. This would suggest that HDL production could depend on several genetic and environmental factors, which may finally control plasma apo AI concentration. We already reported that the FCR of HDL increased in type II diabetes, and related to plasma and HDLtriglycerides levels [9]. The present study was therefore designed to evaluate whether plasma triglycerides directly — or secondarily via HDL-triglyceride level — could act upon HDL clearance rate. A reduction of plasma triglyceride level, achieved by means of v3-FA, may induce a change in HDL composition, and particularly a decrease of HDL-triglyceride level. Several studies have indicated that HDL catabolism was related to HDL-triglyceride concentration or HL activity [13,14]. Thus, a reduction may be associated with a decrease of HDL clearance. However, this probably does not occur in this study since maxEPA® did not induce a clear-cut reduction of HDL-triglycerides, which was not consequently significantly correlated with HDL-apo AI catabolism (data not shown). This lack of change in HDL composition is in agreement with the unaffected HL activity and the increased LPL activity [7], as well as the decrease of VLDL production [8] after fish oil supplementation. In contrast, a lowered plasma triglyceride level, achieved by means of v3-FA, induces a decrease of HDL clearance rate. Therefore, these results suggest that plasma triglyceride level per se may play a role in the control of HDL metabolism. Nevertheless, definite conclusions are precluded by the small number of patients studied.
Then, maxEPA® induces changes in HDL catabolism and this could be added to the beneficial cardiovascular effect of this treatment. These current kinetic data, even observed on a small-sized study group, may provide new insights about the mechanisms of action of fish oil, which remain open to further investigations.
Acknowledgements We thank I. Grit, C. Le Valegant, and P. Mauge`re for their excellent technical assistance and D. Darmaun for his review and advice. This study was supported by grants from the department of Clinical Research of Nantes Hospital and from Pierre Fabre Sante Laboratories.
References [1] Kromhout D, Bosscheiter EB, de Lezenne Coulander C. The inverse relation between fish consumption and 20-year mortality from coronary heart disease. New Engl J Med 1985;312:1205 –9. [2] Schectman G, Kaul S, Kissebah H. Effect of fish oil concentrate on lipoprotein composition in NIDDM. Diabetes 1988;37:1567 – 73. [3] Goh YK, Jumpsen MA, Ryan EA, Clandinin MT. Effects of w-3 fatty acids on plasma lipids, cholesterol and lipoproteins fatty acid content in NIDDM patients. Diabetologia 1997;40:45 –52. [4] Morgan WA, Raskin P, Rosenstock J. A comparison of fish oil or corn oil supplements in hyperlipidemic subjects with NIDDM. Diabetes Care 1995;18:83 – 6. [5] Dunstan DW, Mori TA, Puddey IB, Burke V, Morton AR, Stanton KG. The independent and combined effects of aerobic exercise and dietary fish intake on serum lipids and glycemic control in NIDDM: a randomized controlled study. Diabetes Care 1998;20:913 – 21. [6] Holub BJ, Bakker DJ, Skeaff CM. Alterations in molecular species of cholesterol esters formed via lecithin-cholesterol acyltransferase in human subjects consuming fish oil. Atherosclerosis 1987;66:11 – 8. [7] Kasim SE, Stern B, Khilnani S, McLin P, Baciorowski S, Jen KLC. Effects of w 3 fish oil on lipid metabolism, glycemic control, and blood pressure in type II diabetic patients. J Clin Endocrinol Metab 1988;67:1 – 5. [8] Fisher WR, Zech LA, Stacpoole PW. Apolipoprotein B metabolism in hypertriglyceridemic diabetic patients administered either a fish oil- or vegetable oil-enriched diet. J Lipid Res 1998;39:388 – 401. [9] Fre´ nais R, Ouguerram K, Maugeais C, et al. High density lipoprotein apolipoprotein AI kinetics in NIDDM: a stable isotope study. Diabetologia 1997;40:578 – 83. [10] Brinton EA, Eisenberg S, Breslow JL. Increased apo AI and apo AII fractional catabolic rate in patients with low high density lipoprotein-cholesterol levels with or without hypertriglyceridemia. J Clin Invest 1991;87:536 – 44. [11] Nikkila¨ EA, Taskinen MR, Sane T. Plasma high density lipoprotein concentration and subfraction in relation to triglyceride metabolism. Am Heart J 1987;113:543 – 8. [12] Pietzch J, Julius U, Nitzsche S, Hanefeld M. In vivo evidence for increased apolipoprotein A-I catabolism in subjects with impaired glucose tolerance. Diabetes 1998;47:1928 – 34.
R. Fre´ nais et al. / Atherosclerosis 157 (2001) 131–135 [13] Saku K, Gartside PS, Hynd BA, Mendoza SG, Kashyap ML. Apolipoprotein AI and AII metabolism in patients with primary high density lipoprotein deficiency associated with familial hypertriglyceridemia. Metabolism 1985;34:754 –64.
.
135
[14] Brinton EA, Eisenberg S, Breslow JL. Human HDL cholesterol levels are determined by apo AI fractional catabolic rate, which correlates inversely with estimates of HDL particle size. Arterioscler Thromb 1994;14:707 – 20.