| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH COMMUNICATION |
1 From the Clinical Nutrition Research Unit, Department of Public Health and Caring Sciences/Geriatrics, Uppsala University, Uppsala, Sweden
2 Supported by the Swedish Medical Research Council, the Dr P Håkansson Foundation, and the Swedish Society of Medicine. The fatty acid preparations were supplied by Natural Lipids Ltd AS, Hovebygda, Norway. 3 Reprints not available. Address correspondence to U Risérus, Clinical Nutrition Research Unit, Department of Public Health and Caring Sciences/Geriatrics, Box 609, 75125 Uppsala, Sweden. E-mail: ulf.riserus{at}pubcare.uu.se.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: We investigated the effects of c9,t11 CLA supplementation on insulin sensitivity, body composition, and lipid peroxidation in a group at high risk for cardiovascular disease.
Design: In a randomized, double-blind, placebo-controlled study, 25 abdominally obese men received 3 g c9,t11 CLA/d or placebo (olive oil). Before and after 3 mo of supplementation, we assessed insulin sensitivity (hyperinsulinemic euglycemic clamp), lipid metabolism, body composition, and urinary 8-iso-prostaglandin F2
(a major F2-isoprostane) and 15-keto-dihydro-prostaglandin F2, markers of in vivo oxidative stress and inflammation, respectively.
Results: All subjects completed the study. Compared with placebo, c9,t11 CLA decreased insulin sensitivity by 15% (P < 0.05) and increased 8-iso-prostaglandin F2 and 15-keto-dihydro-prostaglandin F2 excretion by 50% (P < 0.01) and 15% (P < 0.05), respectively. The decreased insulin sensitivity was independent of changes in serum lipids, glycemia, body mass index, and body fat but was abolished after adjustment for changes in 8-iso-prostaglandin F2 concentrations. There were no differences between groups in body composition.
Conclusions: A CLA preparation containing the purified c9,t11 CLA isomer increased insulin resistance and lipid peroxidation compared with placebo in obese men. Because c9,t11 CLA occurs in commercial supplements as well as in the diet, the present results should be confirmed in larger studies that also include women.
Key Words: trans Fatty acids • fatty acids • conjugated linoleic acid • dairy fat • diet • insulin resistance • abdominal obesity • inflammation • oxidative stress • lipid peroxidation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
90% of dietary CLA is c9,t11 CLA and <10% is t10,c12 CLA) (2). In addition, CLA is found in weight-loss products sold over the counter as isomer mixtures, usually containing c9,t11 CLA and t10,c12 CLA in equal proportions. Furthermore, c9,t11 CLA seems to be endogenously formed in humans from trans vaccenic acid (4) found in dairy fat. Metabolic effects of CLA in animals are divergent, which is partly explained by isomer-specific properties of CLA as well as differences in phenotypes studied (1). There is little information available about isomer-specific effects on glucose and lipid metabolism in humans because CLA mixtures were used in most studies (5-8). Our recent studies showed that t10,c12 CLA caused marked insulin resistance, oxidative stress, and dyslipidemia in obese men (9, 10). Because c9,t11 CLA is the major dietary CLA isomer (2), it is relevant to investigate the metabolic effects of c9,t11 CLA in obese men. Because of the prooxidative (10-12) and HDL-lowering effects (9) of CLA mixtures, a study that uses purified c9,t11 CLA is needed to gain more information about this particular CLA isomer in humans.
In the present study, we hypothesized that c9,t11 CLA could improve glucose metabolism and serum lipid profile on the basis of recent data. First, results in obese mice suggest that c9,t11 CLA, in contrast to t10,c12 CLA, reduced serum lipids without impairing insulin action (13). Second, in a recent 8-wk controlled study in healthy normolipidemic subjects there was a significant decrease in plasma VLDL cholesterol after c9,t11 CLA supplementation (14). To date, that study provides the only data on the effect of purified c9,t11 CLA in humans. Because that study included only lean healthy subjects, it would be interesting to investigate the effects of c9,t11 CLA on lipid and glucose metabolism in a high-risk group of abdominally obese men. Thus, in a randomized, double-blind, controlled trial, we investigated the effects of c9,t11 CLA on insulin sensitivity and lipid peroxidation in vivo by using direct methods for determining these 2 variables, euglycemic hyperinsulinemic clamp (15) and F2-isoprostane production (16, 17), respectively.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Protocol and study design
A randomized, placebo-controlled, double-blind study design with 2 parallel groups was used. Twenty-five men were randomly assigned to 1 of 2 groups receiving 3 g/d (6 capsules/d; 1% of total energy intake) as c9,t11 CLA (83% triacylglycerol) or placebo (olive oil; 83% triacylglycerol). The isomer composition of the CLA preparation was assessed with the use of high-resolution gas chromatography of fatty acid ethyl esters (sodium ethylate method) as described in detail by Saebo (18). The isomer content of the CLA preparation was 83.3% c9,t11 CLA; 7.3% t10,c12 CLA; 5.5% c9-18:1; 0.46% c9,c11 CLA; 0.2% c10,c12 CLA; and 1.4% t9,t11+t10,t12 CLA, which provided 2.5 g c9,t11 CLA/d and 0.2 g t10,c12 CLA/d. The major fatty acid content of placebo was 82.7% c9-18:1; 5.6% c9-18:2,c12; 3.6% 16:0; and 3.3% 18:0. The preparations did not contain any added antioxidants. Capsules were kept cold and in the dark in sealed boxes before the study. During the study, all participants were instructed to keep the boxes in the refrigerator to avoid any possible oxidation. All preparations (identical in appearance) were prepared by Natural Lipids Ltd (Hovebygda, Norway), which also generated randomization numbers and blinding.
Primary outcome measures were insulin sensitivity, serum lipoproteins, and free radical and cyclooxygenase-dependent lipid peroxidation in vivo (urinary 8-iso-prostaglandin F2 and 15-keto-dihydro-prostaglandin F2, respectively). Secondary outcome measures were anthropometric variables. All variables were determined after 12-h fast, before and after 3-mo supplementation. Subjects were asked to restrain from smoking, physical activity, and alcohol within 48 h preceding clinical investigations. Subjects were encouraged to maintain their usual diet and physical activity habits during the study.
Nonenzymatic lipid peroxidation
Concentrations of free 8-iso-prostaglandin F2 (a major F2-isoprostane) were analyzed in morning urine without any extraction, with the use of a highly specific and sensitive radioimmunoassay as previously described (19). 8-iso-Prostaglandin F2 concentrations were adjusted for creatinine values measured with a commercial kit (IL Test; Monarch Instrument, Amherst, NH).
Enzymatic lipid peroxidation
Urinary samples were analyzed for 15-keto-dihydro-prostaglandin F2, a major metabolite of prostaglandin F2 and proinflammatory marker, without any extraction, by using a radioimmunoassay as described previously (20).
Body composition
Sagittal abdominal diameter was measured at the L4–5 level in the supine position, and waist girth was measured as previously described in detail (7). Anthropometric variables were measured by a single investigator. Bioelectrical impedance analysis was measured with the use of a multifrequency analyzer (Xitron Technologies Inc, San Diego). From the estimation of body water, body fat content was calculated, with the assumption that fat-free mass contains 73.2% water (21). Lean body mass was calculated with the formula provided by the manufacturer.
Hyperinsulinemic clamp
Hyperinsulinemic euglycemic clamp was used to determine insulin sensitivity in vivo according to DeFronzo et al (15), slightly modified as previously described (9). Blood glucose was assayed with the use of a HemoCue Analyzer (HemoCue AB, Ängelholm, Sweden) and an enzymatic method. The glucose infusion rate (M; mg · kg body wt–1 · min–1) was calculated during the last hour of the clamp (15). Insulin sensitivity (M/I) was calculated by dividing M by the mean plasma insulin concentration (in mU/L) during the last 60 min of the clamp multiplied by 100 (I) (15).
Biochemical analyses
Venous blood was drawn into evacuated tubes, coagulated, centrifuged (2500 x g, 10 min, room temperature), and then frozen at –20 °C. Serum samples were stored at –70 °C. Serum insulin was measured with the use of an enzyme-linked immunosorbent assay kit (Mercodia AB, Uppsala, Sweden) in a Bio-Rad Coda automated electroimmunoassay analyzer (Bio-Rad Laboratories AB, Hercules, CA). Lipoproteins were isolated from fresh serum by a combination of preparative ultracentrifugation (22) and precipitation with a sodium phosphotungstate and magnesium chloride solution (23). Serum lipoproteins were assayed by using enzymatic techniques with a Monarch 2000 centrifugal analyzer (Instrumentation Laboratories, Lexington, MA).
Statistics
Values are means ± SDs. Shapiro-Wilk W tests were used to test distributions. Variables with skewed distributions [all measures except cholesterol and free fatty acid (FFA)] were logarithmically transformed. All measures were normally distributed after logarithmic transformation. The mean changes within groups from baseline to 12 wk were assessed with the use of the paired t test. Analysis of covariance (ANCOVA) was used to compare changes between groups when baseline values and change (
) from baseline values for each outcome measure were taken into account. It was calculated that 13 subjects per group would be needed to detect a mean difference between groups in insulin sensitivity index (M/I) of 1 unit with a power of 0.80 at a significance level of 0.05 with the use of the unpaired t test.
Pearson’s correlation coefficient was determined from pairwise correlations. Partial correlation analysis for values was assessed with the use of both baseline values and values as covariates. A two-tailed P value < 0.05 was regarded as significant. For statistical analyses, the JMP software package (version 3.2) was used (SAS Institute Inc, Cary, NC).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The age of the subjects was 56 ± 6.0 y in the placebo group and 54 ± 5.5 y in the c9,t11 CLA group. All 25 men completed the study. Complete data from the euglycemic clamp test were available from 23 men: 1 subject in each treatment group did not complete the clamp test at follow-up. All other metabolic measures were obtained in these 2 subjects at baseline and at follow-up. Compliance (capsule count) was 91% (n = 25) with no differences between groups. No side effects were reported, and no adverse effects on liver enzymes occurred (data not shown).
|
).
Glucose metabolism
Insulin sensitivity (M/I) was significantly lower after c9,t11 CLA than after placebo (–15%; P < 0.05; Table 1). This effect remained significant after control for baseline values of insulin sensitivity (ANCOVA). The mean insulin-mediated glucose uptake (M) was also lower after c9,t11 CLA than after placebo (P < 0.05; data not shown). No significant differences were found between the groups in fasting blood glucose or serum insulin concentrations. The significant impairment of insulin sensitivity was abolished when 8-iso-prostaglandin F2 and 15-keto-dihydro-prostaglandin F2 were adjusted for (ANCOVA) but remained significant after the addition of age, BMI, body fat, waist girth, FFA, triacylglycerols, or blood glucose as covariates.
Lipid peroxidation and inflammation
Concentrations of 8-iso-prostaglandin F2 and 15-keto-dihydro-prostaglandin F2 were higher [50% (P < 0.01) and 15% (P < 0.05), respectively] after c9,t11 CLA than after placebo. These measures remained unchanged in the placebo group (Table 1). The significant increase in 8-iso-prostaglandin F2 remained significant after adjustment for all other variables, including -insulin sensitivity (ANCOVA).
Body composition
No significant differences were found between the groups in weight, BMI, total body fat, lean body mass, sagittal diameter, or waist girth after adjustments (ANCOVA) for baseline values (Table 1). Within the c9,t11 CLA group (paired t test), there was a significant increase in body weight and BMI from baseline to 12 wk (Table 1).
Correlation analysis (whole-group analysis)
The changes in insulin sensitivity were inversely correlated with 8-iso-prostaglandin F2 (r = –0.43, P < 0.05, n = 23), VLDL cholesterol (r = –0.63, P < 0.01), and total body fat (r = –0.46, P < 0.05). The changes in 8-iso-prostaglandin F2 concentrations were not correlated with changes in other measures, except for insulin sensitivity (r and P values presented above) and 15-keto-dihydro-prostaglandin F2 (r = 0.40, P < 0.05).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Interestingly, there was a significant difference between groups in insulin sensitivity but not in serum lipids or glucose concentrations. In contrast, t10,c12 CLA was previously shown to aggravate both dyslipidemia and glycemia (9). This finding is accordant with isomer-specific effects in mice and hamsters, whereby metabolic disorders occurred after t10,c12 CLA but not after c9,t11 CLA (13, 26, 27).
The cyclooxygenase-mediated lipid peroxidation (15-keto-dihydro-prostaglandin F2), which reflects a proinflammatory response (20), also increased after c9,t11 CLA. The changes were related to free radical–mediated lipid peroxidation (8-iso-prostaglandin F2).
On the basis of previous data (10), we assume that the very small t10,c12 CLA content (200 mg) in the present c9,t11 CLA preparation does not explain the increased lipid peroxidation, although a small effect of t10,c12 CLA cannot be excluded. Our assumption is supported by evidence that shows increased lipid peroxidation (8-iso-prostaglandin F2) after dietary supplementation with trans vaccenic acid, a precursor of c9,t11 CLA in the human body (4). Furthermore, it is not likely that the small amount of t10,c12 CLA in the c9,t11 CLA preparation solely mediated the current lowering of insulin sensitivity, considering that CLA mixtures containing 1 g t10,c12 CLA do not significantly alter insulin sensitivity (9). However, a possible antagonistic effect of t10,c12 CLA and c9,t11 CLA present in CLA mixtures would complicate interpretations because effects of a single isomer might be different from those of the 2 isomers combined. In addition, when comparing the amount of t10,c12 CLA in the t10,c12 CLA concentrate and CLA isomer mixture, respectively, the previous study (10) indicated a possible dose-response effect of t10,c12 CLA on lipid peroxidation, which was not apparent for insulin sensitivity (9). Thus, before we can conclude that c9,t11 CLA alone increases lipid peroxidation, future studies that use entirely pure c9,t11 CLA preparations are needed to clarify the isomer-specific effects of CLA.
No significant effect of c9,t11 CLA on VLDL-cholesterol concentrations was observed, which stands in contrast to the results of a recent controlled study in lean and healthy subjects that reported decreased VLDL cholesterol after c9,t11 CLA (14). In that study, a slightly less purified c9,t11 CLA preparation (3 g/d; 80:20 CLA isomer blend of c9,t11 and t10,c12) was used. However, in agreement with the results by Noone et al (14), there were no effects on BMI or on lipid, glucose, or insulin concentrations compared with placebo. In contrast with our data, BMI did not increase within the c9,t11 CLA group in the study by Noone et al (14). Because our study included obese men and the duration of the intervention period and CLA preparations differed slightly, caution should be taken when comparing these 2 studies. Unfortunately, insulin sensitivity and lipid peroxidation were not assessed by Noone et al (14).
Our results suggest that c9,t11 CLA has no antiobesity effects and are in accord with evidence in mice (28), which suggests that t10,c12 CLA is the antiadipogenic isomer (29). It is unlikely that insulin resistance was mediated by the modest increase in mean body weight (30) that occurred in both groups, because change in BMI and body fat was adjusted for. Nor did adjustments for insulin sensitivity and age at baseline or changes in body composition affect the significant effect of c9,t11 CLA on insulin sensitivity when compared with placebo.
A limitation of this study could be the restricted sample size. The power calculation suggested that the present sample size was large enough to detect significant differences between groups in insulin sensitivity. However, before firm conclusions can be drawn about the effect of c9,t11 CLA on body composition and metabolic variables, the present results should be confirmed by conducting large studies that include both obese men and women.
Another limitation was that we did not assess dietary intake or physical activity during the study. Thus, the possibility that changes in dietary habits affected insulin sensitivity cannot be completely excluded, although the randomized design should have clearly lowered the risk of such bias, and all subjects were instructed to maintain their usual lifestyle habits. From a nutritional viewpoint, the CLA preparation used in the present study (83% c9,t11 CLA, 7% t10,c12 CLA) mirrors dietary CLA isomer distribution quite well (2). However, note that the absolute dose of c9,t11 CLA was 
8 times as high as the average dietary CLA intake in a similar population (7), which indicates that it is unlikely that possible variations in dietary CLA intake (ie, from high amounts of dairy fat) significantly influenced the present results.
The clinical relevance of the present results is uncertain, but the 15% reduction in insulin sensitivity, which corresponds to 5–10 kg of weight gain as estimated from dose-response clamp studies (26), and the 50% increase in lipid peroxidation are pronounced effects that might be proatherogenic in obese subjects having high, long-term CLA intakes from dairy foods (2), hydrogenated vegetable oils (3), or dietary supplements. However, with regard to the metabolic syndrome, epidemiologic data suggest no hazardous effects of consuming dairy fat. If anything, it seems as if a high intake of dairy products can protect against developing the metabolic syndrome (31, 32). The present study suggests that such associations could not be explained by c9,t11 CLA present in dairy fat.
These results could have important implications for human nutrition and the food industry. At the present time, attempts in several countries are being made to enrich the amount of c9,t11 CLA in foods by altering cattle feeding (33). Future epidemiologic studies together with large long-term trials could help answer the important question of whether relatively small, but apparently potent, amounts of dietary CLA are safe with regard to cardiovascular disease risk.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Raff, T. Tholstrup, S. Toubro, J. M. Bruun, P. Lund, E. M. Straarup, R. Christensen, M. B. Sandberg, and S. Mandrup Conjugated Linoleic Acids Reduce Body Fat in Healthy Postmenopausal Women J. Nutr., July 1, 2009; 139(7): 1347 - 1352. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Tholstrup, M. Raff, E. M. Straarup, P. Lund, S. Basu, and J. M. Bruun An Oil Mixture with Trans-10, Cis-12 Conjugated Linoleic Acid Increases Markers of Inflammation and in Vivo Lipid Peroxidation Compared with Cis-9, Trans-11 Conjugated Linoleic Acid in Postmenopausal Women J. Nutr., August 1, 2008; 138(8): 1445 - 1451. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Raff, T. Tholstrup, S. Basu, P. Nonboe, M. T. Sorensen, and E. M. Straarup A Diet Rich in Conjugated Linoleic Acid and Butter Increases Lipid Peroxidation but Does Not Affect Atherosclerotic, Inflammatory, or Diabetic Risk Markers in Healthy Young Men J. Nutr., March 1, 2008; 138(3): 509 - 514. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. L. Neuhouser, M. J. Barnett, A. R. Kristal, C. B. Ambrosone, I. King, M. Thornquist, and G. Goodman (n-6) PUFA Increase and Dairy Foods Decrease Prostate Cancer Risk in Heavy Smokers J. Nutr., July 1, 2007; 137(7): 1821 - 1827. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. D Whigham, A. C Watras, and D. A Schoeller Efficacy of conjugated linoleic acid for reducing fat mass: a meta-analysis in humans Am. J. Clinical Nutrition, May 1, 2007; 85(5): 1203 - 1211. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. E. Steck, A. M. Chalecki, P. Miller, J. Conway, G. L. Austin, J. W. Hardin, C. D. Albright, and P. Thuillier Conjugated Linoleic Acid Supplementation for Twelve Weeks Increases Lean Body Mass in Obese Humans J. Nutr., May 1, 2007; 137(5): 1188 - 1193. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Kuhnt, A. Wagner, J. Kraft, S. Basu, and G. Jahreis Dietary supplementation with 11trans- and 12trans-18:1 and oxidative stress in humans. Am. J. Clinical Nutrition, November 1, 2006; 84(5): 981 - 988. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Larsen, S. Toubro, O. Gudmundsen, and A. Astrup Conjugated linoleic acid supplementation for 1 y does not prevent weight or body fat regain Am. J. Clinical Nutrition, March 1, 2006; 83(3): 606 - 612. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. S.W. Taylor, S. R.P. Williams, R. Rhys, P. James, and M. P. Frenneaux Conjugated Linoleic Acid Impairs Endothelial Function Arterioscler. Thromb. Vasc. Biol., February 1, 2006; 26(2): 307 - 312. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Desroches, P Y. Chouinard, I. Galibois, L. Corneau, J. Delisle, B. Lamarche, P. Couture, and N. Bergeron Lack of effect of dietary conjugated linoleic acids naturally incorporated into butter on the lipid profile and body composition of overweight and obese men Am. J. Clinical Nutrition, August 1, 2005; 82(2): 309 - 319. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Valeille, J. Ferezou, G. Amsler, A. Quignard-Boulange, M. Parquet, D. Gripois, V. Dorovska-Taran, and J.-C. Martin A cis-9,trans-11-conjugated linoleic acid-rich oil reduces the outcome of atherogenic process in hyperlipidemic hamster Am J Physiol Heart Circ Physiol, August 1, 2005; 289(2): H652 - H659. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
