Original Paper Ann Nutr Metab 1992;36:336-342
Department of Dietetics, University of Gent. Belgium
Keyw ords
Protein-sparing modified fast Weight reduction program Obesity Serum lipid fatty acid compositions
Effects of Weight Loss on the Fatty Acid Composition of Serum Lipids in Obese Women
Abstract The effect of a weight reduction regimen, consisting of a pro tein-sparing modified fast and an exercise program, on the fatty acid composition of serum phospholipids and cholesterol esters of obese women, is described. In phospholipids, this treatment did not induce any significant change of the different fatty acid families (total saturated, monounsaturated, to9, co 7, co6 and trans-fatty acids), except for total co3 fatty acids which increased. Within families, individual fatty acids change however. The changes are compatible with increased A 5 and A6 desaturase activity. In cholesteryl esters, signifi cant changes occurred which are suggestive of an increase in serum of the fraction of cholesteryl esters of intracellular ori gin. The changes in fatty acid compositions may not be benefi cial with respect to atherosclerosis.
Introduction
In order to reduce weight, the energy bal ance has to be negative. This can be achieved either by reducing energy intake, increasing energy output or a combination of both. En ergy input can be reduced to zero (‘total star vation’) but this results in a negative protein balance with a considerable loss of muscular tissue and a considerable risk of cardiac ar rhythmia and occasionally sudden death.
Received: October 16.1991 Accepted: July 20. 1992
With very low calory diets, of about 400600 keal/day, 60-65% of which as protein (a protein-sparing modified fast or PSMF), about 200-250 g of adipose tissue is lost per day without affecting significantly the nitro gen balance [ 1]. Several authors have reported on the effects of very low energy diets on the fatty acid composition of adipose tissue [2, 3] and serum lipids [3, 4], In a previous study from this laboratory, the effects of a PSMF and physical exercise program on plasma lip-
Dr. A Chrislophe laboratory for Dietetics Clinic Building 12. Floor 6 (6K12) University Hospital. Dc Pintelaan 185 B-9000 Gent (Belgium)
© 1992 S. Kargcr AG. Basel 0250-6807/92/ 0366-0336$2.75/0
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
A. Christophe A. Vermeiden
Material and Methods Patients All patients participating in this study (n = 22) were women (mean age ± standard deviation: 36.8 ± 18.5 years: range 15-66 years) with moderate to severe obe sity (mean body mass index ± standard deviation: 37.2 ± 7 .1 : range 27.1-53.8). All were in good health and none was on medication. Women on birth control pills had stopped hormonal contraception at least 3 months before the start of the study. The patients underwent a weight reduction pro gram for 4 weeks in hospital. The program consisted of a PSMF and an exercise program. The diet provided 400 keal/dav (63 energy % protein. 25 energy % carbo hydrate. 12 energy % fat). As it has been reported that varying the fatty acid composition of energy-restricted diets containing a much higher percentage of fat than used in this study did not affect fatty acid composi tions of serum lipids [4], the fatty acid composition of the diet was not determined. In order to promote dietary compliance, the patients were motivated to adhere to the diet by a psychologist before and on sev eral occasions during the treatment. Acetonuria was determined daily. Positive results were considered as indicating compliance with the diet. The exercise pro gram consisted of a daily session of swimming (20 min) and gymnastics (30 min). Average weight loss was 10.4 ± 2.3 kg. Laboratory Methods Blood was taken before and immediately at the end of the treatment between 8 and 10 a.m. after an over
night fast. Scrum was deep-frozen until analyzed. Preand posttreatment samples were analyzed simulta neously. Lipids were extracted [6] and the phospholip ids and cholesteryl esters isolated by thin layer chro matography [7], Their fatty acids were converted into methyl esters [8] whose weight percent composition was determined after separation on a 25 m X 250 pm X 0.2 pm df Silar IOC column (initial temperature 150°C; I min isothermal: programed to 200 "C at I °C/min). installed in a Varian Model 3500 gas chro matograph equipped with a (lame ionization detector (275 °C) and a glass splitter (250 °C). Split ratio was 1/15. Peak identification was done by spiking with authentic standards (Alltech). Peak integration and calculation was performed electronically with a Varian Model 4290 integrator. Statistical A nalysis Results are given as a percentage ± standard devia tion of the total fatty acids in a lipid class. Signifi cances between pre- and posttreatment values were calculated by paired Student's t test.
Results
Fatty acids that changed significantly in serum phospholipids or cholesteryl esters are shown in table 1. In serum phospholipids, the sums of the percentages of all saturated fatty acids, m-monoenes, /ram-monoencs, 0)9. 0)7, branched chain and odd chain fatty acids were not affected significantly neither as indi vidual trans-, branched chain or odd chain fatty acids [results not shown]. The sums of the percentages of co6 fatty acids and of 20:3co6 + 20:4co6 (fatty acids are designated by number of carbon atoms: number of double bonds followed by designation of the bio chemical series) were not affected signifi cantly but 20:3 co 6 was decreased (65% of the pretreatment value; p < 0.01). whereas its A 5 desaturation product 20:4co6 (116%; p < 0.01) and the ratio of 20:4co6 to 20:3w6 (210%; p < 0.001) were increased significant ly. Total co3 fatty acids were increased (114%; p < 0.01) but in this fatly acid family, 20:5co3 337
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
ids and lipoproteins in obese women were reported [5], In brief, treatment resulted in a highly significant decrease of about 20% of plasma phospholipids, total cholesterol and LDL-cholestcrol as well as in a significant decrease in HDL-cholesterol which was en tirely due to reduced HDLi-cholesterol and not to HDLi-cholesterol. The HDL-cholestcrol to LDL-cholesterol ratio decreased slightly. In view of these important changes in concentration, we considered it of interest to determine the effects of the treatment on the fatty acid composition of serum phospholip ids and cholesteryl esters.
Table 1. Weight % ± SD of fatty acids and of fatty acid families that change significantly in serum phospholipids or in scrum cholesteryl esters Fatty acid
Lipid class phospholipids
cholesteryl esters before
after
0.14 ±0.05 11.35 ± 1.42
0 .2 1 ± 0 . 1 2 ** 12.56 ± 1.53*
2.73 ± 0.91
2.13 ±0.69**
52.55 ±4.79 0.79±0.20 8.08 ±2.64
47.17 ± 5.69** 0.50 ±0.16*** 10.80 ±2.75***
0.56 ± 0.17
0.76 ±0.23***
before
after
28.60+1.89 12.37 ± 1.18 1.27 ± 0.61
30.48 ±2.73* 1 1 .0 0 ± 1.2 ** 0.85 ±0.32***
16:1 to 7 24:1 d)9
2.87±0.84
3.70 ±1.04*
18:2m6 20:3(o 6 20:4(o 6
2.68 ±0.69 12.29 ±2.72
1.74±0.53** 13.87 ±2.52**
20:5(0 3 22:5 0)3 2 2 :6 m3
1.12 ± 0.38 0.91 ±0.31 3.87 ±0.90
SAT MONO
42.67 ±1.64 12.35 ±2.08
42.74 ±2.72 12.88 ± 1.31
13.04 ±1.52 19.56 ±2.74
14.53 ±1.6* 19.03 ± 1.75
Sum 0 ) 6 Sum 0) 3
32.04 ±2.44 5.56 ± 1.05
31.56 ±2.08 6.22 ±1.13**
61.63 ±3.90 1.18 ±0.32
58.64 ±4.67* 1.38 ±0.43*
15:0 16:0 18:0 24:0
0.74 ±0.29** 1.08 ±0.36* 4.40±0.9I***
* p < 0.05: ** p < 0.01; *** p < 0.001. SAT = Sum of saturated fatty acids: MONO = sum of monoenes.
338
(134%: p < 0.001) was increased significant ly. 22:6co 3 (136%: p < 0.001) and the sum of all co3 fatty acids were increased (118%; p < 0.05).
Discussion
Methodology The fatty acid composition of phospholip ids and cholesteryl esters does not change appreciably in a time frame of a few days and is considered to reflect the dietary intake of the preceding weeks [9]. Therefore, only a sin gle sample was taken at the beginning and at the end of the treatment. The standard devia
Christophe/Vermculcn
Effect of Weight Reduction on Fatty Acid Compositions
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
was decreased (66%; p < 0.01) and 22:5 co3 (119%: p < 0.05) and 22:6w3 (114%; p < 0.001) were increased significantly. The 22:6(o3 to 22:5co3 ratio was not affected sig nificantly but the ratio of the sum of desaturase-elongase products of 20:5co3 (i.e. 22:5co3 + 22:6co3) to 20:5(o3 (114%; p < 0.001) was increased significantly. In serum cholesteryl esters, the sum of all saturated fatty acids were increased (111%; p < 0.05) at the end of the weight reduction period. Total monoenes were not affected sig nificantly but 16:10)7 (78%; p < 0.001) was decreased. Total to 6 fatty acids (90%; p < 0.05), 18:2co6 (90%: p < 0.01) and 20:3o6 (63%: p < 0.001) were decreased, but 20:4o)6
Effect o f Treatment on the Fatty Acid Composition o f Serum Phospholipids In man, the fatty acid composition of se rum phospholipids is influenced by many fac tors such as age [ 10], sex [ 10], disease [11.12], medication [ 13] and diet [14], Of all these fac tors, diet was the only variable in this study. In accordance with our results, increases in 16:0 and 20:4to 6 and decreases in 18:0 and 20:3co6 were also found in serum phospholip ids after 4 weeks of very low calory diets [4]. In the latter study, to 3 fatty acids were not reported however. Increases in 16:0 and 20:4(o6 were also observed after 6 weeks of weight reduction of grossly obese patients [3]. Significant increases of 20:4co6 have been de scribed in plasma total lipids after total star vation [15]. An increase of 22:6co 3 and of the ratio of 20:4co6 to 18:2to6 ( 188%: p < 0.05) as observed after PSMF has also been found in patients with anorexia nervosa, in whom energy intake was reduced, compared to con trols [ 16], After PSMF. 18:0 was reduced and the 16:0/18:0 ratio increased (120%: p < 0.01) in serum phospholipids. Similar changes were found in surgical patients un dergoing a protein-sparing therapy by paren teral administration of amino acids, with en ergy provision below their energy needs [17], In neither condition, nor in this study, were there signs of essential fatty acid deficiency [16,17], co 3, o)6 and tram-fatty acids cannot be formed in the body. The fact that they were not decreased after treatment suggests that
enough of these fatty acids were mobilized and reached a pool available for the acvlating enzymes. In human depot fat co6 and to3 fatty acids have been found [2] as well as /ram-fatty acids when these were present in the pre viously consumed diet [18]. During very low calory dieting, there is a considerable mobili zation of fat from adipose tissue and most of the energy is provided for by fatty acid oxida tion. However, fatty acid mobilization is con siderably higher than oxidation [19], the ex cess of fatty acids being used for the synthesis of lipids, part of which appear in serum. The suggestion that differential utilization of mobilized fatty acids for oxidation and incorporation might explain the effects of en ergy restriction on serum lipid fatty acid com position [4] was considered. Based on pub lished data on adipose tissue weight and fattyacid composition before and after low calory dieting [2], changes in weight of individual fatty acids in adipose tissue, induced by such treatment, were calculated. Such calculations show that the decrease in weight of 18:3co3 is substantial. In contrast, there is no decrease in weight of longer chain co3 fatty acids in adi pose tissue. Thus, net mobilization of these long chain to 3 fatty acids does not occur and cannot explain the increase of 22:5co3 and 22:6co3 in serum phospholipids. The latter could be explained by partial conversion of the mobilized 18:3co3 in longer chain co3 fatty acids, involving A5 and A6 desatura tion. Desaturation of fatty acids is under nutri tional and hormonal control [19, 20], A 5 and A6 desaturase activity increases in rat liver upon removal of cholesterol from the diet [21 ] whereas cholesterol supplementation de presses these activities [22], In rat plasma, 18:2 co6 increases and 20:4 co6 decreases upon cholesterol supplementation in the diet [23], If reverse changes occur upon removal of cho lesterol from the diet and the same relation
339
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
tions of the different fatly acids reflect subject variation. They are usually somewhat smaller in the posttreatment group than in the pre treatment group (table 1). This is probably due to the fact that the pretreatmcnl diet was not controlled whereas the composition of the diet during treatment was the same for all patients.
Effect o f Treatment on the Fatty Acid Composition o f Serum Cholesteryl Esters In the normal-fed healthy individual, most of the serum cholesteryl esters are formed in plasma under the action of lecithin-choles terol acyltransferase (LCAT) [26], The com position of the fatty acids present in the sn-2 position of lecithin and the acyl specificity of the enzyme are the main determinants of the composition of the fatty acids of the so formed cholesteryl esters. Cholesteryl esters are also formed intracellularly. These choles teryl esters of intracellular origin are poorer in 18:2o)6 and richer in 16:0 and 18:0 than those of LCAT origin. They contribute in different conditions to a different extent to serum cho-
340
lesteryl esters [26-28]. Starvation reduces LCAT activity substantially [29] and. indeed, the fraction of serum cholesterol that is esterificd drops during energy restriction [24], Our observation that 16:0 and 18:0 were increased and that 18:2 to 6 was decreased at the end of the treatment period in serum cholesteryl es ters suggests that a larger fraction of this lipid class may be of intracellular origin in this con dition. The changes in 20:3(o6, 20:4o)6 and 22:6(03 in serum cholesteryl esters are in the same direction as the change of these fatty acids in serum phospholipids and may be due to changes in availability of these fatty acids in liver or in plasma precursor. Clinical Relevance Total o)3 fatty acids were increased in serum phospholipids and cholesteryl esters at the end of the weight reduction program. Fish oil feeding also results in an increase of total 0)3 fatty acids in these lipid fractions and reduces cardiovascular risk factors and dis ease [30], Different long chain to 3 fatty acids have different lipoproteinemic responses [31] and probably reduce the risk by different mechanisms and to a different extent [32], The consumption of tuna fish which is rich in 22:6(0 3 reduces vulnerability to cardiac ar rhythmias in an experimental model [33]. Other beneficial effects of 22:6co3 have been described [34], At the end of the weight reduction program. 22:6to3 was increased in both phospholipids and cholesteryl esters. Some cardiovascular risk factors are probably related to the ratio of 20:4oo6 to 20:5o)3 [35. 36], This ratio is decreased in serum phos pholipids by fish oil feeding but increased (12.86 ± 7.93 vs. 21.89 ± 10.02: p < 0.005) at the end of the treatment. Thus, as to ath erosclerosis, some of the changes in fatty acid composition of serum phospholipids ob served after PSMF may be beneficial (in crease in 22 :6 (03) whereas others (reduced
Christophe/Vermeulen
Effect of Weight Reduction on Fatty Acid Compositions
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
ship holds in man as in rats, these findings arc in line with ours (nonsignificant decrease in 18:2co6 and significant increase in 20:4to6). Cholesterol is mobilized from adipose tissue during starvation [24] but the ratio of choles terol to fat is low in adipose tissue, even in massive obesity, compared to that of a West ern-type diet [25], The ratio of the sum of desaturase-elongase products to their precur sor in serum phospholipids has been pro posed as an estimation of desaturase activity [ 12 ], In the (o6 fatty acid cascade, the ratio of the sum of linoleic acid metabolites beyond the A 6 desaturase step (18:3 oo6 + 20:3cu6 + 20:4ü)6) to their precursor (linoleic acid) in creases 9%, though significance is not reached, whereas in the 0)3 series it increases 14% (p < 0.01). The ratio of 20:4 0)6 to its A 5 desaturase precursor (20:3a)6) increases 85% in the 0)6 series (p < 0.001) whereas the sum of20:5co3 + 22:5o)3 + 22:6co3 increases 14% (p < 0.01 ) in the to3 series (for this series, the A 5 desaturase precursor is not incorporated in phospholipids). Competitive interactions between 0)6 and 0)3 fatty acids are known to exist, making that the estimation of desatu rase activity is different for these fatty acids.
20:5co3 and increased 20:4co6 to 20:5co3 ra tio) may not be. The decrease of 18:2co6 in serum cholesteryl esters may have important consequences, as myocardial infarction is strongly associated with lower levels of this fatty acid in this lipid fraction [37, 38], This may be a factor contrib
uting to early atherosclerosis in adults after repeated weight loss [39] even though several other risk factors change favorably [5], How long the changes in serum lipid fatty acid composition, induced by treatment, persist after resuming an unrestricted feeding regi men was not determined however.
References 9
10
11
12
13
14
15
Glatz FC. Soffers AEMF, Katan MB: Fatty acid composition of se rum cholesteryl esters and ery thro cyte membranes as indicators of linoleic acid intake in man. Am J Clin Nutr 1989:49:269-276. Holman RT. Smythe L. Johnson S: Effect of sex and age on fatty acid composition of human serum lipids. Am J Clin Nutr 1979:32:23902399. Holman RT. Johnson S: Changes in essential fatty acid profile in serum phospholipids in human disease. Prog Lipid Res 1981:20:67-73. Holman RT. Johnson SB: Essential fatty acid deficiencies in man: in Perkins EG. Visek WJ. (eds): Di etary Fats and Health. Champaign. American Oil Chemists' Society, 1983. pp 247-260. Christophe A. Vermeulen A: Changes in polyunsaturated fatty acids in the major serum lipid classes during simvastatin treat ment: in Malmendier C fed): Ab stract Book. 5e Colloque interna tional de la fondation de recherche sur l'athérosclérose. Brussels. 1990. p 26. Lopes SM. T rimbo SL. Mascioli EA. Blackburn GL: Human plasma fatty acid variations and how they are related to dietary intake. Am J Clin Nutr 1991:53:628-637. Korf II. Tsagikian TA. Meshcheriakova VA. Levachev MM, Arutiunova MB: Changes in fatty acid composition of the plasma and erythrocytes in osteochondrosis. V oprM edK him 1989:35:81-84.
16 Lagan SM. Farrell P: Vitamin E. vitamin A and essential fatty acid status o f patients hospitalized for anorexia nervosa. Am J Clin Nutr 1985:41:1054-1060. 17 Steginck LD. Freeman JB. Wispc J. Connor WE: Absence of the bio chemical symptoms of essential fatty acid deficiency in surgical pa tients undergoing protein sparing therapy. Am J Clin Nutr 1977:30: 388-393. 18 Schrock CG. Connor WE: Incorpo ration of the dietary trans fatty acid ( 0 8 :1 ) in the serum lipids, the se rum lipoproteins and adipose tissue. Am J Clin Nutr 1975:28:10201027. 19 Olson RE. Vester JW: Nutrition Endocrine interrelationship in the control of fat transport in man. Physiol Rev 1960:40:677-733. 20 Brenner RR: Nutritional and hor monal factors influencing desaturation of essential fatty acids. Prog Lipid Res 1981:20:41-47. 21 Lcikin AI. Brenner RR: In vivo cho lesterol removal from liver microsomes induces changes in fatty acid desaturase activities. Biochim BiophysActa 1988:963:311-319. 22 Leikin AI. Brenner RR: Cholesterolinduced microsomal changes modu late desaturase activities. Biochim Biophys Acta 1987:922:294-303. 23 Garg ML. Snoswcll AM. Sabine JR: Influence of dietary cholesterol on desatinase enzymes of rat liver microsomes. Prog Lipid Res 1986:25: 639-644.
341
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
1 Jequier E: Situation métabolique des obèses au cours de cure d'amai grissement. Cah Nutr Diét 1980:15: 111-114. 2 Phinnev SD. Tang AB. Johnson SB. Holman RT: Reduced adipose 18:3 to 3 with weight loss by very low calory dieting. l ipids 1990:25:7988 06.’ 3 Rossner S. Walldius G. Bjorvell H: Fatty acid composition of serum lip ids and adipose tissue in severe obe sity before and after six weeks of weight loss. Int J Obcs 1989:13:603612. 4 Wilson JH. Rietveld T. Van den Berg JW. Jansen H. Swart GR. Lam berts FW: The effect of very low energy diets on the fatty acid com position of serum lipids. Int J Obes I989:l3(suppl 2):5l-60. 5 Vermeulen A: Effects of a short term (4 weeks) protein-sparing modified fast on plasma lipids and lipopro teins in obese women. Ann Nutr Metab 1990:34:133-142. 6 Folch-Pi J. Lees M, Sloane-Stanly GH: A simple method for the isola tion and purification of total lipids from animal tissues. J Biol Chem 1957:226:497-509. 7 Christophe A. Matthys F: New method for the determination of the fatty acid pattern of serum lipid classes. Clin Chim Acta 1967:16: 39-43. 8 Muskiet FAJ. Van Doormaal JJ, Martini 1A. Wolthers BG. Van dcr Silk W: Capillary gas chromato graphic profiling of total long-chain fatty acids and cholesterol in biolog ical materials. J Chromatogr 1983: 278:231-244.
24 Kudchodkar BJ. Sohdi HS, Mason DT, Borhani NO: Effects of acute caloric restriction on cholesterol me tabolism in man. Am J Clin Nutr 1977:30:1135-1146. 25 Angel A. Bray GA: Synthesis of fatty acids and cholesterol by liver, adi pose tissue and intestinal mucosa from obese and control patients. Eur J Clin Invest 1979:9:355-362. 26 Subbaiah PV. Banerji B. Gregg RE. Bagdade JD: Molecular species of cholesteryl esters formed in abctalipoproteincmia: Effect of apoprotein B-containing lipoproteins. J l ipid Res 1990:21:927-932. 27 Bisgaicr CL, Siebenkas MV. Brown ML, Inazu A. Koizumi J. Mabushi H. Tall AL: Familial cholesteryl es ter transfer protein deficiency is as sociated with triglyceride-rich low density lipoproteins containing cholestcryl esters of probable intracellu lar origin. J Lipid Res 1991:32:2133. 28 Norum KR. Gjone E: Familial plasma lecithin cholesterol acyltransferase deficiency. Biochemical study of a new inborn error of me tabolism. Scand J Clin Lab Invest 1967:20:231-243.
29 Wallentin L, Sköldstam L: Lipopro teins and cholesterol esterification rate in plasma during a 10-day mod ified fast in man. Am J Clin Nutr 1980:33:1925-1931. 30 Kromhout B, Bosschieter EB, Coulandcr CL: The inverse relationship between fish consumption and 20year mortality from coronary heart disease. N Engl J Med 1985:312: 1205-1209. 31 Childs MT. King IB. Knopp RH: Divergent lipoproteincmic re sponses to fish oils with various ra tios of eicosapcntaenoic acid and docosahexaenoic acid. Am J Clin Nutr 1990:52:632-639. 32 Leaf A, Weber PC: Cardiovascular effects of n-3 fatty acids. N Engl J Med 1988:318:549-557. 33 McLennen P L Abeywardena MY. Charnock JS: Reversal of the arrhythmogenic effects of long term saturated fatty acid intake by di etary n-3 and n-6 polyunsaturated fatty acids. Am J Clin Nutr 1990:51: 53-58.
342
Christophe/ Vermeulen
34 Yamazaki K, Hamazaki T. Yano S. Funada T. Ibuki F: Changes in fatty acids in rat blood and organs after infusion of docosahexaenoic acid ethyl ester. Am J Clin Nutr 1991:53: 620-627. 35 Kannel WB. Wolf PA. Castelli WP, D'Agostino RB: Fibrinogen and risk of cardiovascular disease. The Framingham study. JAMA 1987; 258:1183-1186. 36 Anonymous: Fish oil and the devel opment of athcroslcrosis. Nutr Rev 1987:45:90-92. 37 Kingsbury KJ, Morgan DM. Stovold R. Brett CG: Polyunsaturated fatty acids and myocardinal infarc tion. Lancet 1969;ii: 1325—1329. 38 Kingsbury KJ. Brett C. Stolved R. Chapman A. Anderson J. Morgan DM: Abnomal fatty acid composi tion and human atherosclerosis. Postgrad Med J 1974:50:425-440. 39 Hamm PG. Schekelle RB. Stamler J: Large fluctuations in body weight during young adulthood and 25 years risk o f coronary death in men. Am J Epidemiol 1989:129:312—
Effect of Weight Reduction on Fatty Acid Compositions
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
318.
Department of Dietetics, University of Gent. Belgium
Keyw ords
Protein-sparing modified fast Weight reduction program Obesity Serum lipid fatty acid compositions
Effects of Weight Loss on the Fatty Acid Composition of Serum Lipids in Obese Women
Abstract The effect of a weight reduction regimen, consisting of a pro tein-sparing modified fast and an exercise program, on the fatty acid composition of serum phospholipids and cholesterol esters of obese women, is described. In phospholipids, this treatment did not induce any significant change of the different fatty acid families (total saturated, monounsaturated, to9, co 7, co6 and trans-fatty acids), except for total co3 fatty acids which increased. Within families, individual fatty acids change however. The changes are compatible with increased A 5 and A6 desaturase activity. In cholesteryl esters, signifi cant changes occurred which are suggestive of an increase in serum of the fraction of cholesteryl esters of intracellular ori gin. The changes in fatty acid compositions may not be benefi cial with respect to atherosclerosis.
Introduction
In order to reduce weight, the energy bal ance has to be negative. This can be achieved either by reducing energy intake, increasing energy output or a combination of both. En ergy input can be reduced to zero (‘total star vation’) but this results in a negative protein balance with a considerable loss of muscular tissue and a considerable risk of cardiac ar rhythmia and occasionally sudden death.
Received: October 16.1991 Accepted: July 20. 1992
With very low calory diets, of about 400600 keal/day, 60-65% of which as protein (a protein-sparing modified fast or PSMF), about 200-250 g of adipose tissue is lost per day without affecting significantly the nitro gen balance [ 1]. Several authors have reported on the effects of very low energy diets on the fatty acid composition of adipose tissue [2, 3] and serum lipids [3, 4], In a previous study from this laboratory, the effects of a PSMF and physical exercise program on plasma lip-
Dr. A Chrislophe laboratory for Dietetics Clinic Building 12. Floor 6 (6K12) University Hospital. Dc Pintelaan 185 B-9000 Gent (Belgium)
© 1992 S. Kargcr AG. Basel 0250-6807/92/ 0366-0336$2.75/0
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
A. Christophe A. Vermeiden
Material and Methods Patients All patients participating in this study (n = 22) were women (mean age ± standard deviation: 36.8 ± 18.5 years: range 15-66 years) with moderate to severe obe sity (mean body mass index ± standard deviation: 37.2 ± 7 .1 : range 27.1-53.8). All were in good health and none was on medication. Women on birth control pills had stopped hormonal contraception at least 3 months before the start of the study. The patients underwent a weight reduction pro gram for 4 weeks in hospital. The program consisted of a PSMF and an exercise program. The diet provided 400 keal/dav (63 energy % protein. 25 energy % carbo hydrate. 12 energy % fat). As it has been reported that varying the fatty acid composition of energy-restricted diets containing a much higher percentage of fat than used in this study did not affect fatty acid composi tions of serum lipids [4], the fatty acid composition of the diet was not determined. In order to promote dietary compliance, the patients were motivated to adhere to the diet by a psychologist before and on sev eral occasions during the treatment. Acetonuria was determined daily. Positive results were considered as indicating compliance with the diet. The exercise pro gram consisted of a daily session of swimming (20 min) and gymnastics (30 min). Average weight loss was 10.4 ± 2.3 kg. Laboratory Methods Blood was taken before and immediately at the end of the treatment between 8 and 10 a.m. after an over
night fast. Scrum was deep-frozen until analyzed. Preand posttreatment samples were analyzed simulta neously. Lipids were extracted [6] and the phospholip ids and cholesteryl esters isolated by thin layer chro matography [7], Their fatty acids were converted into methyl esters [8] whose weight percent composition was determined after separation on a 25 m X 250 pm X 0.2 pm df Silar IOC column (initial temperature 150°C; I min isothermal: programed to 200 "C at I °C/min). installed in a Varian Model 3500 gas chro matograph equipped with a (lame ionization detector (275 °C) and a glass splitter (250 °C). Split ratio was 1/15. Peak identification was done by spiking with authentic standards (Alltech). Peak integration and calculation was performed electronically with a Varian Model 4290 integrator. Statistical A nalysis Results are given as a percentage ± standard devia tion of the total fatty acids in a lipid class. Signifi cances between pre- and posttreatment values were calculated by paired Student's t test.
Results
Fatty acids that changed significantly in serum phospholipids or cholesteryl esters are shown in table 1. In serum phospholipids, the sums of the percentages of all saturated fatty acids, m-monoenes, /ram-monoencs, 0)9. 0)7, branched chain and odd chain fatty acids were not affected significantly neither as indi vidual trans-, branched chain or odd chain fatty acids [results not shown]. The sums of the percentages of co6 fatty acids and of 20:3co6 + 20:4co6 (fatty acids are designated by number of carbon atoms: number of double bonds followed by designation of the bio chemical series) were not affected signifi cantly but 20:3 co 6 was decreased (65% of the pretreatment value; p < 0.01). whereas its A 5 desaturation product 20:4co6 (116%; p < 0.01) and the ratio of 20:4co6 to 20:3w6 (210%; p < 0.001) were increased significant ly. Total co3 fatty acids were increased (114%; p < 0.01) but in this fatly acid family, 20:5co3 337
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
ids and lipoproteins in obese women were reported [5], In brief, treatment resulted in a highly significant decrease of about 20% of plasma phospholipids, total cholesterol and LDL-cholestcrol as well as in a significant decrease in HDL-cholesterol which was en tirely due to reduced HDLi-cholesterol and not to HDLi-cholesterol. The HDL-cholestcrol to LDL-cholesterol ratio decreased slightly. In view of these important changes in concentration, we considered it of interest to determine the effects of the treatment on the fatty acid composition of serum phospholip ids and cholesteryl esters.
Table 1. Weight % ± SD of fatty acids and of fatty acid families that change significantly in serum phospholipids or in scrum cholesteryl esters Fatty acid
Lipid class phospholipids
cholesteryl esters before
after
0.14 ±0.05 11.35 ± 1.42
0 .2 1 ± 0 . 1 2 ** 12.56 ± 1.53*
2.73 ± 0.91
2.13 ±0.69**
52.55 ±4.79 0.79±0.20 8.08 ±2.64
47.17 ± 5.69** 0.50 ±0.16*** 10.80 ±2.75***
0.56 ± 0.17
0.76 ±0.23***
before
after
28.60+1.89 12.37 ± 1.18 1.27 ± 0.61
30.48 ±2.73* 1 1 .0 0 ± 1.2 ** 0.85 ±0.32***
16:1 to 7 24:1 d)9
2.87±0.84
3.70 ±1.04*
18:2m6 20:3(o 6 20:4(o 6
2.68 ±0.69 12.29 ±2.72
1.74±0.53** 13.87 ±2.52**
20:5(0 3 22:5 0)3 2 2 :6 m3
1.12 ± 0.38 0.91 ±0.31 3.87 ±0.90
SAT MONO
42.67 ±1.64 12.35 ±2.08
42.74 ±2.72 12.88 ± 1.31
13.04 ±1.52 19.56 ±2.74
14.53 ±1.6* 19.03 ± 1.75
Sum 0 ) 6 Sum 0) 3
32.04 ±2.44 5.56 ± 1.05
31.56 ±2.08 6.22 ±1.13**
61.63 ±3.90 1.18 ±0.32
58.64 ±4.67* 1.38 ±0.43*
15:0 16:0 18:0 24:0
0.74 ±0.29** 1.08 ±0.36* 4.40±0.9I***
* p < 0.05: ** p < 0.01; *** p < 0.001. SAT = Sum of saturated fatty acids: MONO = sum of monoenes.
338
(134%: p < 0.001) was increased significant ly. 22:6co 3 (136%: p < 0.001) and the sum of all co3 fatty acids were increased (118%; p < 0.05).
Discussion
Methodology The fatty acid composition of phospholip ids and cholesteryl esters does not change appreciably in a time frame of a few days and is considered to reflect the dietary intake of the preceding weeks [9]. Therefore, only a sin gle sample was taken at the beginning and at the end of the treatment. The standard devia
Christophe/Vermculcn
Effect of Weight Reduction on Fatty Acid Compositions
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
was decreased (66%; p < 0.01) and 22:5 co3 (119%: p < 0.05) and 22:6w3 (114%; p < 0.001) were increased significantly. The 22:6(o3 to 22:5co3 ratio was not affected sig nificantly but the ratio of the sum of desaturase-elongase products of 20:5co3 (i.e. 22:5co3 + 22:6co3) to 20:5(o3 (114%; p < 0.001) was increased significantly. In serum cholesteryl esters, the sum of all saturated fatty acids were increased (111%; p < 0.05) at the end of the weight reduction period. Total monoenes were not affected sig nificantly but 16:10)7 (78%; p < 0.001) was decreased. Total to 6 fatty acids (90%; p < 0.05), 18:2co6 (90%: p < 0.01) and 20:3o6 (63%: p < 0.001) were decreased, but 20:4o)6
Effect o f Treatment on the Fatty Acid Composition o f Serum Phospholipids In man, the fatty acid composition of se rum phospholipids is influenced by many fac tors such as age [ 10], sex [ 10], disease [11.12], medication [ 13] and diet [14], Of all these fac tors, diet was the only variable in this study. In accordance with our results, increases in 16:0 and 20:4to 6 and decreases in 18:0 and 20:3co6 were also found in serum phospholip ids after 4 weeks of very low calory diets [4]. In the latter study, to 3 fatty acids were not reported however. Increases in 16:0 and 20:4(o6 were also observed after 6 weeks of weight reduction of grossly obese patients [3]. Significant increases of 20:4co6 have been de scribed in plasma total lipids after total star vation [15]. An increase of 22:6co 3 and of the ratio of 20:4co6 to 18:2to6 ( 188%: p < 0.05) as observed after PSMF has also been found in patients with anorexia nervosa, in whom energy intake was reduced, compared to con trols [ 16], After PSMF. 18:0 was reduced and the 16:0/18:0 ratio increased (120%: p < 0.01) in serum phospholipids. Similar changes were found in surgical patients un dergoing a protein-sparing therapy by paren teral administration of amino acids, with en ergy provision below their energy needs [17], In neither condition, nor in this study, were there signs of essential fatty acid deficiency [16,17], co 3, o)6 and tram-fatty acids cannot be formed in the body. The fact that they were not decreased after treatment suggests that
enough of these fatty acids were mobilized and reached a pool available for the acvlating enzymes. In human depot fat co6 and to3 fatty acids have been found [2] as well as /ram-fatty acids when these were present in the pre viously consumed diet [18]. During very low calory dieting, there is a considerable mobili zation of fat from adipose tissue and most of the energy is provided for by fatty acid oxida tion. However, fatty acid mobilization is con siderably higher than oxidation [19], the ex cess of fatty acids being used for the synthesis of lipids, part of which appear in serum. The suggestion that differential utilization of mobilized fatty acids for oxidation and incorporation might explain the effects of en ergy restriction on serum lipid fatty acid com position [4] was considered. Based on pub lished data on adipose tissue weight and fattyacid composition before and after low calory dieting [2], changes in weight of individual fatty acids in adipose tissue, induced by such treatment, were calculated. Such calculations show that the decrease in weight of 18:3co3 is substantial. In contrast, there is no decrease in weight of longer chain co3 fatty acids in adi pose tissue. Thus, net mobilization of these long chain to 3 fatty acids does not occur and cannot explain the increase of 22:5co3 and 22:6co3 in serum phospholipids. The latter could be explained by partial conversion of the mobilized 18:3co3 in longer chain co3 fatty acids, involving A5 and A6 desatura tion. Desaturation of fatty acids is under nutri tional and hormonal control [19, 20], A 5 and A6 desaturase activity increases in rat liver upon removal of cholesterol from the diet [21 ] whereas cholesterol supplementation de presses these activities [22], In rat plasma, 18:2 co6 increases and 20:4 co6 decreases upon cholesterol supplementation in the diet [23], If reverse changes occur upon removal of cho lesterol from the diet and the same relation
339
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
tions of the different fatly acids reflect subject variation. They are usually somewhat smaller in the posttreatment group than in the pre treatment group (table 1). This is probably due to the fact that the pretreatmcnl diet was not controlled whereas the composition of the diet during treatment was the same for all patients.
Effect o f Treatment on the Fatty Acid Composition o f Serum Cholesteryl Esters In the normal-fed healthy individual, most of the serum cholesteryl esters are formed in plasma under the action of lecithin-choles terol acyltransferase (LCAT) [26], The com position of the fatty acids present in the sn-2 position of lecithin and the acyl specificity of the enzyme are the main determinants of the composition of the fatty acids of the so formed cholesteryl esters. Cholesteryl esters are also formed intracellularly. These choles teryl esters of intracellular origin are poorer in 18:2o)6 and richer in 16:0 and 18:0 than those of LCAT origin. They contribute in different conditions to a different extent to serum cho-
340
lesteryl esters [26-28]. Starvation reduces LCAT activity substantially [29] and. indeed, the fraction of serum cholesterol that is esterificd drops during energy restriction [24], Our observation that 16:0 and 18:0 were increased and that 18:2 to 6 was decreased at the end of the treatment period in serum cholesteryl es ters suggests that a larger fraction of this lipid class may be of intracellular origin in this con dition. The changes in 20:3(o6, 20:4o)6 and 22:6(03 in serum cholesteryl esters are in the same direction as the change of these fatty acids in serum phospholipids and may be due to changes in availability of these fatty acids in liver or in plasma precursor. Clinical Relevance Total o)3 fatty acids were increased in serum phospholipids and cholesteryl esters at the end of the weight reduction program. Fish oil feeding also results in an increase of total 0)3 fatty acids in these lipid fractions and reduces cardiovascular risk factors and dis ease [30], Different long chain to 3 fatty acids have different lipoproteinemic responses [31] and probably reduce the risk by different mechanisms and to a different extent [32], The consumption of tuna fish which is rich in 22:6(0 3 reduces vulnerability to cardiac ar rhythmias in an experimental model [33]. Other beneficial effects of 22:6co3 have been described [34], At the end of the weight reduction program. 22:6to3 was increased in both phospholipids and cholesteryl esters. Some cardiovascular risk factors are probably related to the ratio of 20:4oo6 to 20:5o)3 [35. 36], This ratio is decreased in serum phos pholipids by fish oil feeding but increased (12.86 ± 7.93 vs. 21.89 ± 10.02: p < 0.005) at the end of the treatment. Thus, as to ath erosclerosis, some of the changes in fatty acid composition of serum phospholipids ob served after PSMF may be beneficial (in crease in 22 :6 (03) whereas others (reduced
Christophe/Vermeulen
Effect of Weight Reduction on Fatty Acid Compositions
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
ship holds in man as in rats, these findings arc in line with ours (nonsignificant decrease in 18:2co6 and significant increase in 20:4to6). Cholesterol is mobilized from adipose tissue during starvation [24] but the ratio of choles terol to fat is low in adipose tissue, even in massive obesity, compared to that of a West ern-type diet [25], The ratio of the sum of desaturase-elongase products to their precur sor in serum phospholipids has been pro posed as an estimation of desaturase activity [ 12 ], In the (o6 fatty acid cascade, the ratio of the sum of linoleic acid metabolites beyond the A 6 desaturase step (18:3 oo6 + 20:3cu6 + 20:4ü)6) to their precursor (linoleic acid) in creases 9%, though significance is not reached, whereas in the 0)3 series it increases 14% (p < 0.01). The ratio of 20:4 0)6 to its A 5 desaturase precursor (20:3a)6) increases 85% in the 0)6 series (p < 0.001) whereas the sum of20:5co3 + 22:5o)3 + 22:6co3 increases 14% (p < 0.01 ) in the to3 series (for this series, the A 5 desaturase precursor is not incorporated in phospholipids). Competitive interactions between 0)6 and 0)3 fatty acids are known to exist, making that the estimation of desatu rase activity is different for these fatty acids.
20:5co3 and increased 20:4co6 to 20:5co3 ra tio) may not be. The decrease of 18:2co6 in serum cholesteryl esters may have important consequences, as myocardial infarction is strongly associated with lower levels of this fatty acid in this lipid fraction [37, 38], This may be a factor contrib
uting to early atherosclerosis in adults after repeated weight loss [39] even though several other risk factors change favorably [5], How long the changes in serum lipid fatty acid composition, induced by treatment, persist after resuming an unrestricted feeding regi men was not determined however.
References 9
10
11
12
13
14
15
Glatz FC. Soffers AEMF, Katan MB: Fatty acid composition of se rum cholesteryl esters and ery thro cyte membranes as indicators of linoleic acid intake in man. Am J Clin Nutr 1989:49:269-276. Holman RT. Smythe L. Johnson S: Effect of sex and age on fatty acid composition of human serum lipids. Am J Clin Nutr 1979:32:23902399. Holman RT. Johnson S: Changes in essential fatty acid profile in serum phospholipids in human disease. Prog Lipid Res 1981:20:67-73. Holman RT. Johnson SB: Essential fatty acid deficiencies in man: in Perkins EG. Visek WJ. (eds): Di etary Fats and Health. Champaign. American Oil Chemists' Society, 1983. pp 247-260. Christophe A. Vermeulen A: Changes in polyunsaturated fatty acids in the major serum lipid classes during simvastatin treat ment: in Malmendier C fed): Ab stract Book. 5e Colloque interna tional de la fondation de recherche sur l'athérosclérose. Brussels. 1990. p 26. Lopes SM. T rimbo SL. Mascioli EA. Blackburn GL: Human plasma fatty acid variations and how they are related to dietary intake. Am J Clin Nutr 1991:53:628-637. Korf II. Tsagikian TA. Meshcheriakova VA. Levachev MM, Arutiunova MB: Changes in fatty acid composition of the plasma and erythrocytes in osteochondrosis. V oprM edK him 1989:35:81-84.
16 Lagan SM. Farrell P: Vitamin E. vitamin A and essential fatty acid status o f patients hospitalized for anorexia nervosa. Am J Clin Nutr 1985:41:1054-1060. 17 Steginck LD. Freeman JB. Wispc J. Connor WE: Absence of the bio chemical symptoms of essential fatty acid deficiency in surgical pa tients undergoing protein sparing therapy. Am J Clin Nutr 1977:30: 388-393. 18 Schrock CG. Connor WE: Incorpo ration of the dietary trans fatty acid ( 0 8 :1 ) in the serum lipids, the se rum lipoproteins and adipose tissue. Am J Clin Nutr 1975:28:10201027. 19 Olson RE. Vester JW: Nutrition Endocrine interrelationship in the control of fat transport in man. Physiol Rev 1960:40:677-733. 20 Brenner RR: Nutritional and hor monal factors influencing desaturation of essential fatty acids. Prog Lipid Res 1981:20:41-47. 21 Lcikin AI. Brenner RR: In vivo cho lesterol removal from liver microsomes induces changes in fatty acid desaturase activities. Biochim BiophysActa 1988:963:311-319. 22 Leikin AI. Brenner RR: Cholesterolinduced microsomal changes modu late desaturase activities. Biochim Biophys Acta 1987:922:294-303. 23 Garg ML. Snoswcll AM. Sabine JR: Influence of dietary cholesterol on desatinase enzymes of rat liver microsomes. Prog Lipid Res 1986:25: 639-644.
341
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
1 Jequier E: Situation métabolique des obèses au cours de cure d'amai grissement. Cah Nutr Diét 1980:15: 111-114. 2 Phinnev SD. Tang AB. Johnson SB. Holman RT: Reduced adipose 18:3 to 3 with weight loss by very low calory dieting. l ipids 1990:25:7988 06.’ 3 Rossner S. Walldius G. Bjorvell H: Fatty acid composition of serum lip ids and adipose tissue in severe obe sity before and after six weeks of weight loss. Int J Obcs 1989:13:603612. 4 Wilson JH. Rietveld T. Van den Berg JW. Jansen H. Swart GR. Lam berts FW: The effect of very low energy diets on the fatty acid com position of serum lipids. Int J Obes I989:l3(suppl 2):5l-60. 5 Vermeulen A: Effects of a short term (4 weeks) protein-sparing modified fast on plasma lipids and lipopro teins in obese women. Ann Nutr Metab 1990:34:133-142. 6 Folch-Pi J. Lees M, Sloane-Stanly GH: A simple method for the isola tion and purification of total lipids from animal tissues. J Biol Chem 1957:226:497-509. 7 Christophe A. Matthys F: New method for the determination of the fatty acid pattern of serum lipid classes. Clin Chim Acta 1967:16: 39-43. 8 Muskiet FAJ. Van Doormaal JJ, Martini 1A. Wolthers BG. Van dcr Silk W: Capillary gas chromato graphic profiling of total long-chain fatty acids and cholesterol in biolog ical materials. J Chromatogr 1983: 278:231-244.
24 Kudchodkar BJ. Sohdi HS, Mason DT, Borhani NO: Effects of acute caloric restriction on cholesterol me tabolism in man. Am J Clin Nutr 1977:30:1135-1146. 25 Angel A. Bray GA: Synthesis of fatty acids and cholesterol by liver, adi pose tissue and intestinal mucosa from obese and control patients. Eur J Clin Invest 1979:9:355-362. 26 Subbaiah PV. Banerji B. Gregg RE. Bagdade JD: Molecular species of cholesteryl esters formed in abctalipoproteincmia: Effect of apoprotein B-containing lipoproteins. J l ipid Res 1990:21:927-932. 27 Bisgaicr CL, Siebenkas MV. Brown ML, Inazu A. Koizumi J. Mabushi H. Tall AL: Familial cholesteryl es ter transfer protein deficiency is as sociated with triglyceride-rich low density lipoproteins containing cholestcryl esters of probable intracellu lar origin. J Lipid Res 1991:32:2133. 28 Norum KR. Gjone E: Familial plasma lecithin cholesterol acyltransferase deficiency. Biochemical study of a new inborn error of me tabolism. Scand J Clin Lab Invest 1967:20:231-243.
29 Wallentin L, Sköldstam L: Lipopro teins and cholesterol esterification rate in plasma during a 10-day mod ified fast in man. Am J Clin Nutr 1980:33:1925-1931. 30 Kromhout B, Bosschieter EB, Coulandcr CL: The inverse relationship between fish consumption and 20year mortality from coronary heart disease. N Engl J Med 1985:312: 1205-1209. 31 Childs MT. King IB. Knopp RH: Divergent lipoproteincmic re sponses to fish oils with various ra tios of eicosapcntaenoic acid and docosahexaenoic acid. Am J Clin Nutr 1990:52:632-639. 32 Leaf A, Weber PC: Cardiovascular effects of n-3 fatty acids. N Engl J Med 1988:318:549-557. 33 McLennen P L Abeywardena MY. Charnock JS: Reversal of the arrhythmogenic effects of long term saturated fatty acid intake by di etary n-3 and n-6 polyunsaturated fatty acids. Am J Clin Nutr 1990:51: 53-58.
342
Christophe/ Vermeulen
34 Yamazaki K, Hamazaki T. Yano S. Funada T. Ibuki F: Changes in fatty acids in rat blood and organs after infusion of docosahexaenoic acid ethyl ester. Am J Clin Nutr 1991:53: 620-627. 35 Kannel WB. Wolf PA. Castelli WP, D'Agostino RB: Fibrinogen and risk of cardiovascular disease. The Framingham study. JAMA 1987; 258:1183-1186. 36 Anonymous: Fish oil and the devel opment of athcroslcrosis. Nutr Rev 1987:45:90-92. 37 Kingsbury KJ, Morgan DM. Stovold R. Brett CG: Polyunsaturated fatty acids and myocardinal infarc tion. Lancet 1969;ii: 1325—1329. 38 Kingsbury KJ. Brett C. Stolved R. Chapman A. Anderson J. Morgan DM: Abnomal fatty acid composi tion and human atherosclerosis. Postgrad Med J 1974:50:425-440. 39 Hamm PG. Schekelle RB. Stamler J: Large fluctuations in body weight during young adulthood and 25 years risk o f coronary death in men. Am J Epidemiol 1989:129:312—
Effect of Weight Reduction on Fatty Acid Compositions
Downloaded by: Lund University Libraries 130.235.66.10 - 11/20/2018 2:51:14 PM
318.
