ORIGINAL COMMUNICATION

European Journal of Clinical Nutrition (2005) 59, 508–517

& 2005 Nature Publishing Group All rights reserved 0954-3007/05 $30.00 www.nature.com/ejcn

ORIGINAL COMMUNICATION Effect of CLA supplementation on immune function in young healthy volunteers H-J Song1,2, I Grant2, D Rotondo3, I Mohede4, N Sattar5, SD Heys6 and KWJ Wahle1,2* 1

Rowett Research Institute, Aberdeen, UK; 2The Robert Gordon University, Aberdeen, UK; 3Strathclyde University, Glasgow, UK; Loders-Croklaan,Wormerveer, The Netherlands; 5Pathological Biochemistry, Glasgow University, Glasgow, UK; and 6Surgical & Nutritional Oncology, Medical School, Aberdeen University, UK 4

Objectives: This study investigated the effect of dietary CLA supplementation (3g/day; 50:50 mix of the two major isomers) on the immune system and plasma lipids and glucose of healthy human (male and female) volunteers. Design: Double-blind, randomized, reference-controlled study. Subject and intervention: A total of 28 healthy male and female participants aged 25–50 y received either high oleic sunflower oil (reference) or 50% CLA 9–11 and 50% CLA 10–12 CLA isomers (50:50 CLA-triglyceride form). The treatments were given as supplements in soft-gel capsules providing a total 3 g (6 500 mg capsules) per day in treatment groups for 12 weeks. A 12-week washout period followed the intervention period. Results: Levels of plasma IgA and IgM were increased (Po0.05 and 0.01 respectively), while plasma IgE levels were decreased (Po0.05). CLA supplementation also decreased the levels of the proinflammatory cytokines, TNF-a and IL-1b (Po0.05), but increased the levels of the anti-inflammatory cytokine, IL-10 (Po0.05). Another aspect of immune function, delayed type hypersensitivity (DTH) response, was decreased during and after CLA supplementation (Po0.05). However, plasma glucose, lipids, lymphocyte phenotypic results were not affected significantly by CLA. Conclusion: This is the first study to show that CLA, a fatty acid naturally found in dairy and meat products, can beneficially affect immune function in healthy human volunteers. Sponsorship: This study was supported by Loders-Croklaan, The Netherlands and SEERAD (Scottish Executive Environmental Rural and Agriculture Department).

European Journal of Clinical Nutrition (2005) 59, 508–517. doi:10.1038/sj.ejcn.1602102 Published online 12 January 2005 Keywords: conjugated linoleic acid (CLA); immune function; fatty acids; human

Introduction Conjugated linoleic acid (CLA) is found naturally in many animal products, especially those from ruminant sources where it is synthesized by rumen bacteria from linoleic acid or in ruminant tissues, particularly mammary tissue, by D9 desaturation of trans-vaccenic acid (t-11, 18:1) (Griinari et al, 2000). CLA can also be synthesized in tissues from nonruminants and is found in nonruminant meat sources (Chin

*Correspondence: KWJ Wahle, The Robert Gordon University, St Andrew Street, Aberdeen, AB25 1HG, UK. E-mail: [email protected] Guarantor: KWJ Wahle. Contributors: KJWJ, DR, IM, SDH designed the study supervised experiment work. H-JS, IG conducted experiment work. NS analysed CRP results. H-JS wrote the manuscript with input from all authors. Received 24 February 2004; revised 16 November 2004; accepted 18 November 2004; published online 12 January 2005

et al, 1992) partly through ingestion of feed containing CLA. Increased dairy fat consumption has been shown to be associated with increased CLA levels in human adipose tissue and human milk (Jiang et al, 1999). The term CLA refers to a mixture of positional and geometric isomers of octadecadienoic (cis,cis 9,12–18:2, linoleic) acid, with conjugated double bonds in either the cis or trans configuration located at positions D7,9-, D8,10-, D9,11-, D10,12-, and D11,13- of the carbon chain. The c9, t11 isomer along with the t10, c12 isomer are the most common CLA isomers with the former predominating (cis-9,trans-11 is 70–80% of total CLA (Parodi, 1994). CLA has been implicated in eliciting a variety of potential human health benefits through studies with animal models of disease, including positive effects on cancer, body composition, diabetes and cardiovascular disease (Ip et al, 1994; Visonneau et al, 1997; Cesano et al, 1998). In addition,

CLA and immune functions H-J Song et al

509 animal studies have suggested that CLAs have beneficial effects on certain aspects of immune function (Turek et al, 1998; Hayek et al, 1999; Yang et al, 2000). Most of the animal studies were performed with a mixture of CLA 9–11 and CLA 10–12 in roughly equal amounts. These mixtures have been found to decrease TNF-a and IL-6 production in rats (Turek et al, 1998). Also, splenic levels of immunoglobulin A (IgA), IgG, and IgM increased, while those of IgE decreased significantly in rats fed a 1.0% CLA diet. These results supported the view that CLA can mitigate the food-induced allergic reaction. Despite their importance in animal studies, there is a lack of data demonstrating a direct beneficial effect of CLA on immune function in human subjects. The present double-blind, randomized, reference-controlled trial was designed to determine whether supplementation with a mixture of CLA isomers similar to those found to affect immune function in animals also has similar effects in humans. Healthy men and women aged 25–50 y were therefore supplemented for 12 weeks with 3 g of 50:50 mixture of CLA 9–11 and CLA 10–12 or 3 g of high oleic sunflower oil (reference oil). Effects on immune function were measured, including delayed-type hypersensitivity (DTH) responses, plasma immunoglobulin levels and LPSinduced cytokine production in isolated immune cells. Plasma lipid profiles were also determined.

Reference group (n=14) / CLA group (n=14) Supplementation

Washout period

Blood sampling 0 wk (base line) 6 wk

12 wk

24 wk

Figure 1 Protocol for clinic days. Subjects were requested to fast overnight and attended at baseline and every 6 weeks during supplementations and then 12 weeks after stop supplementation for blood sampling.

were provided in plastic bags each containing a daily supply. Nonconsumed capsules were counted to monitor compliance. A 12-week washout period where dietary supplementation with CLA had ceased was also monitored.

Blood sampling Fasting blood samples were obtained from each volunteer by a trained phlebotomist on 0, 6, 12 and 12 weeks after supplementation (washout period) by venepuncture and collected into vacutainers containing lithium heparin for plasma and the isolation of plasma and peripheral blood mononuclear cells (PBMC). Plasma was stored at 801C until analysed.

Subjects and methods Subjects and study design All protocols were approved by the Joint Ethical Committee of Grampian Health Board and The University of Aberdeen. Male and female volunteers were recruited through an advertisement on the Rowett Research Institute network Intranet message board. Once a potential volunteer expressed an interest they were screened as to their suitability and the chosen volunteers were asked for written consent to confirm their willingness to take part in the study. Exclusion criteria included known allergies, clinical disorders, medication and smoking. Subjects were asked to refrain from taking any sort of medication while on the study (eg aspirin, paracetamol) or in the event that they felt they could not avoid their use they were requested to make a note of the date and dosage. In total, 28 subjects, aged 25–50 y, were randomly allocated to one of the two treatment groups. This study was carried out double-blinded in that neither the investigators nor the volunteers knew which group was the CLA and which was the reference. The design of the study is outlined in Figure 1. Subjects were asked to keep their normal dietary habits and physical activity. They received one of the following treatments: (1) triglyceride form of mixture of 50% CLA 9–11 and 50% CLA 10–12 fatty acids (CLA 50:50, Safflorint, Loders Croklaan, Netherlands); (2) high oleic sunflower oil fatty acids (reference oil, Loders Croklaan, Netherlands). Supplements were given as soft-gel capsules (6 capsules of 500 mg/day) for 12 weeks; capsules

DTH response Tuberculin infection induces the T-cell response, which is cell-mediated (CMI). Firstly, 100 ml of whole blood and 900 ml of RPMI-1640 media were placed into a 12-round well plate (NUNC, Hereford, UK). Then, 0, 1, 10 or 100 mg/ml of tuberculin purified protein derivatives (PPD, Veterinary Laboratories Agency, Surrey, UK) was added to each well. The plate was then placed in an incubator for 7 days at 371C. After 7 days, diluted plasma was carefully taken and used for the detection of IFN-g, the end point marker for PPD treatment, using a commercially available enzyme-linked immunosorbent assay (ELISA) kit (Biosourse, Nivelles, Belgium).

Immunoglobulin analysis For the analysis of IgA, IgG and IgM a Konelab selective chemistry analyser (Labmedics Ltd, Manchester, UK) was used. A standard was used with the spectral calibrator (Konelab, Labmedics Ltd., Manchester, UK) in PBS. A volume of 115 ml of buffer (IgA, IgG and IgM kits) was mixed with 5 ml of plasma samples and the mix was incubated at 371C for 200 s. An amount of 30 mg of porcine anti-IgA, anti-IgG or anti-IgM were added and incubated at 371C for 420 s to permit the immune reaction and the formation of the immune-complex. The immuno-precipitation was evaluated at OD 340 nm in a plate reader and the concentration of European Journal of Clinical Nutrition


510 immunoglobulins was expressed in g/l. A quality control was carried out each time with a special standard protein (Konelab, Labmedics Ltd, Manchester, UK). IgE was analysed using an ELISA kit (Bethyl, Montgomery, USA).

Isolation of PBMC PBMC were isolated on density gradients (histopaque-1077, Sigma Dorset, UK), washed twice in PBS and resuspended in RPMI-1640 medium supplemented with 10% autologous plasma and penicillin/streptomycin.

LPS-induced cytokine production by PBMC Replicate samples of PBMC were stimulated with 10 mg/ml LPS at 371C in 5% CO2 for 6 h (for analysis of TNF-a) or 24 h (for analysis of IL-1b and IL-10). Cells and medium were collected and the cells were then isolated by centrifugation at 400 g for 10 min. Supernatants were stored at 801C until further analysis. Cytokines were analysed in duplicate using commercially available ELISA kits (Biosourse, Nivelles, Belgium) according to the manufacturers’ instructions.

Plasma lipid and glucose analysis Plasma lipid profiles (cholesterol, HDL-cholesterol, LDLcholesterol, triacylglycerols (TAG) and free fatty acids (FFA)) were determined using a Konelab Autoanalyser (Labmedics Ltd, Manchester, UK).

C-reactive protein determinations Plasma C-reactive protein was measured in the West of Scotland Coronary Prevention Study (WOSCOPS). Details of the CRP assay are given in Packard et al (2000).

Lymphocyte subset analysis A measure of 100 ml of the whole blood samples collected in heparin vacutainer tubes was added to each of seven polystyrene round-bottom tubes of 5 ml (12 75 mm style, nonpyrogenic, sterile, Falcon-Becton Dickinson, Plymouth, UK) and the tubes were left on ice. The first tube was the negative control and no antibody was added. A volume of 10 ml of anti-CD3/anti-CD4 (to distinguish T lymphocytes as CD3 þ and T helper lymphocytes as CD3 þ CD4 þ , Serotec, Oxon, UK), anti-CD3/anti-CD8 fluorochrome-labelled antibodies (to distinguish cytotoxic T lymphocytes as CD3 þ CD8 þ , Serotec, Oxon, UK), anti-CD3/anti-CD16 (to distinguish natural killer cells as CD3-CD16 þ , Serotec, Oxon, UK), anti-CD3/anti-CD62L (to distinguish L-selectin expressing T lymphocytes, Serotec, Oxon, UK), anti-CD14/ anti-CD54 (to distinguish monocytes as CD14 þ and to determine the expression of ICAM-1 on monocytes, Serotec, Oxon, UK) and anti-CD14/anti-CD62L (to distinguish Lselectin expressing moncytes, Serotec, Oxon, UK) were European Journal of Clinical Nutrition

respectively added into each tube and then vortexed. After incubation for 30 min in the dark, the samples were lysed by adding 2 ml of diluted FACS lysing solution (1:10) (Becton Dickinson, Plymouth, UK) for 10 min while on ice and in the dark. Tubes were centrifuged at 200 g for 7 min at 41C to permit the removal of the supernatant. Stained leukocytes were fixed with 200 ml of diluted CellFIX (1:10) (Becton Dickinson, Plymouth, UK) and then vortexed. Fixed leukocytes were analysed with a Becton Dickinson FACS Calibur flow cytometer (Becton Dickinson, Oxford,UK). Fluorescence data were collected on 10 000 cells. Lymphocytes and monocytes were identified by forward and side scatter properties. Data were analysed using Cellquest software (Becton Dickinson, Oxford, UK).

Statistics The change from week 0 to weeks 6, 12 and after the 12-week washout period within-treatment groups were tested for significance with a paired t-test. However, differences between the two treatment groups were analysed by ANOVA test and analysis of covariance. The statistical analyses were performed using the GenStat6 software (Bioss, Rowett Research Institute, UK).

Results Sex differences Although the current study had both female and male volunteers in each group, none of the parameters measured showed any sex-specific differences. Hence, the results are presented only as two supplemented groups; reference oil vs CLA, and each volunteer was his or her own control.

Baseline characteristics Randomization of volunteers within the groups was successful as age, height, weight, BMI, total cholesterol levels and gender number did not differ between groups (Table 1). No adverse events were observed in the two treatment groups during the course of the study.

Table 1

Subject baseline characteristics

Sex F M Age (y) Height (cm) Weight (kg) BMI (kg/m2) Total cholesterol (mmol/l)

Reference

CLA

10 4

10 4

30.977.14 169.9710.47 70.05711.13 24.2373.69 5.070.7

31.876.88 168.579.55 69.27717.4 24.373.8 4.970.6


6000

4500

3000

1500

0 0

b

IFN- gamma level in PPD stimulated blood samples (%)

Effect of CLA consumption on ex vivo DTH response Specific cellular recall responses were assessed as DTH response to tuberculin antigen. Various concentrations of tuberculin purified protein derivative (PPD) (0, 1, 10 and 100 mg/ml) were added to whole blood samples in order to determine the optimum concentration for the production of IFN-g (Figure 2a). Both 10 and 100 mg/ml of PPD were found to release IFN-g maximally; however, 10 mg/ml of PPD was used throughout these studies to analyse IFN-g levels in whole blood samples. The effect of lipid supplementation on IFN-g in plasma samples after stimulation of blood with 10 mg/ml of PPD for 7 days is shown in Figure 2b. As the graph shows, both reference oil and CLA supplementation decreased IFN-g after 6, 12 weeks and this effect was still evident in the washout period. However, only the CLA supplementation elicited a significant decrease in IFN-g after 6 weeks supplementation, which was still evident after 12 weeks of washout period (22, 35 and 35% respectively, for 12 weeks supplementation and washout, Po0.05).

a

Concentration of IFN-gamma (pg/ml)

511 Effect of CLA and reference-oil supplementation on plasma lipid profiles and glucose Plasma total cholesterol was not affected by CLA supplementation but the reference oil elicited a significant but small decrease at 12 weeks (10%, Po0.05). (Table 2). HDLcholesterol was decreased slightly but significantly (Po0.05) at 12 weeks by the CLA supplement but not the reference oil, whereas LDL-cholesterol was not altered by either supplement. Plasma FFA levels were reduced by both oil supplements but only significantly (Po0.05) by the CLA (Table 2). Plasma glucose concentrations were reduced slightly but nonsignificantly at 6 weeks supplementation by both oils and returned to normal or control values at 12 weeks and after the washout period (Table 2).

140

1ug/ml 10ug/ml 100ug/ml Concentration of tuberculin PPD

REFERENCE CLA

120 *

*

100 80 60 40 20 0

0

6 weeks 12 weeks Treatment time (wk)

/ /

wash out

Figure 2 Production of IFN-g in tuberculin PPD-stimulated whole blood ex vivo was measured by ELISA (a). Effect of reference (filled bars) and CLA (open bars) supplementation on IFN-g levels in healthy young volunteers (n ¼ 14 each group (b). IFN-g levels at the baseline (0 week) are shown as 100%, the levels during and after supplementation are expressed as a percentage of these levels. The washout period was 12 weeks after the supplementation period. Significant difference from presupplementation levels is indicated by (*) Po0.05 (t-test).

Table 2 Effect of reference and CLA supplementation on plasma lipids and glucose levels in healthy young volunteers (n ¼ 14 each group) 6 weeks

s.d.

12 weeks

s.d.

Washout

s.d.

Reference (%) Cholesterol HDL-C LDL-C Triglyceride FFA Glucose

103.11 100.86 103.45 104.65 78.86 97.39

9.35 11.71 15.61 48.47 25.94 4.7

94.33w 94.22 100.51 157.83ww 73.99 99.69

9.53 10.13 14.03 53.11 18.57 8.3

96.31 95.91 95.84 142.65 76.04 99.6

18.50 22.24 18.64 61.26 20.06 8.14

CLA (%) Cholesterol HDL-C LDL-C Triglyceride FFA Glucose

100.12 100.20 101.67 126.87 69.54ww 96.91

6.41 11.27 15.36 61.13 15.77 13.02

99.48 92.83w 106.72 137.17ww 79.96 97.83

9.87 10.00 22.23 57.40 24.24 12.05

96.29 93.86 96.32 125.8 69.18 98.77

14.04 20.21 18.66 50.52 15.67 10.06

Values during and after supplementation are expressed as a percentage of the base line (0 week, 100%). The washout period was 12 weeks after the supplementation period. Significant difference from presupplementation levels (0 week) is denoted by (w) Po0.05, (ww) Po0.01 (t-test). Error bars are7s.d.

European Journal of Clinical Nutrition


512 release of IL-1b ex vivo in response to LPS over the 12-week study period for the groups supplemented with CLA or reference-oil is shown in Figure 7. The CLA group decreased

130

** *

120 ** 110

100

90

80

Effect of CLA consumption on ex vivo PBMC function To assess the effect of CLA consumption on PBMC function, cells were stimulated ex vivo with LPS, a well-known activator of human monocytes. Decreases in the release of TNF-a in response to LPS were observed following dietary supplementation with CLA. This decrease was statistically significant (10%, Po0.05) after 12 weeks of CLA supplementation, when compared to baseline levels (Figure 6). Unlike the CLA group, the reference-oil group showed a trend to increased TNF-a levels after 6 and 12 weeks supplementation. However, this was not statistically significant between groups. The

**

REFERENCE CLA

Plasma IgM levels (%)

Effect of CLA consumption on Ig levels IgA levels were increased by 10% after 12 weeks supplementation; this increase was continued after cessation of supplementation and still evident in the washout period. In contrast, the reference group decreased IgA levels after 6 weeks but these increased again after 12 weeks supplementation (Figure 3). Plasma IgM levels were also increased in both the CLA and control group after supplementation. However, only the CLA group was significantly increased (by 8 and 20%, at the 6-week and washout period, respectively, Po0.01) (Figure 4). The reference-oil group, like the CLA group, also exhibited a significant increase in IgM but only after 12 weeks of supplementation (10%, Po0.05). However, CLA had the opposite effect on IgE levels to that observed with IgA or IgM and elicited a 10% decrease in plasma IgE levels at both 12 weeks supplementation and after the washout period. IgE levels were also significantly increased at 6 weeks and then decreased after 12 weeks of supplementation and after the 12-week washout period in the referenceoil group (Figure 5).

0

/ / 6 weeks 12 weeks wash out Supplementation time (wk)

Figure 4 Effect of reference (filled bars) and CLA (open bars) supplementation on plasma IgM levels in healthy young volunteers (n ¼ 14 each group). IgM levels at the baseline (0 week) are shown as 100%, the levels during and after supplementation are expressed as a percentage of these levels. The washout period was 12 weeks after the supplementation period. Significant differences from presupplementation values are denoted by (*) at Po0.05 and (**) at Po0.01 (t-test). Error bars are 7s.d.

*

140

REFERENCE CLA

Plasma IgA levels (%)

130 REFERENCE

120

*

CLA

*

110

* **

100

120

110

100

*

90

90 80

Plasma IgE levels (%)

130

0

12 weeks / / wash out 6 weeks Supplementation time (wk)

Figure 3 Effect of reference (filled bars) and CLA (open bars) supplementation on plasma IgA levels in healthy young volunteers (n ¼ 14 each group). IgA levels at the baseline (0 week) are shown as 100%, the levels during and after supplementation are expressed as a percentage of these levels. The washout period was 12 weeks after the supplementation period. Significant difference from presupplementation levels is denoted by (*) Po0.05, (**) Po0.01 (t-test). Error bars are 7s.d.


80

0

/ / 6 weeks 12 weeks Treatment time (wk)

wash out

Figure 5 Effect of reference (filled bars) and CLA (open bars) supplementation on plasma IgE levels in healthy young volunteers (n ¼ 14 each group). IgE levels at the baseline (0 week) are shown as 100 %, the levels during and after supplementation are expressed as a percentage of these levels. The washout period was 12 weeks after the supplementation period. Significant difference between pre- and post-supplementation values are indicated by (*) at Po0.05 (t-test). Error bars are 7s.d.


513 220

REFERENCE CLA

120 110

200 IL-10 release in LPS stimulated PBMCs (%)

TNF- alpharelease in LPS stimulated PBMCs (%)

130

*

100 90 80 70

0

6 weeks 12 weeks / / wash out Treatment time (wk)

Figure 6 Effect of reference (filled bars) and CLA (open bars) supplementation on TNF-a levels in healthy young volunteers (n ¼ 14 each group). Production of TNF-a in LPS-stimulated PBMCs ex vivo was measured by ELISA. TNF-a levels at the baseline (0 week) are shown as 100%, the levels during and after supplementation are expressed as a percentage of these levels. The washout period was 12 weeks after the supplementation period. Significant difference from presupplementation values is indicated by (*) at Po0.05 (t-test). Error bars are 7s.d.

IL-1beta release in LPS stimulated PBMCs (%)

140

REFERENCE CLA

130 120

100 90 80 0

6 weeks 12 weeks / / Treatment time (wk)

180

*

* * **

160 140 120 100 80 60

0

/ / 6 weeks 12 weeks wash out Supplementaion time (wk)

Figure 8 Effect of reference (filled bars) and CLA (open bars) supplementation on IL-10 levels in healthy young volunteers (n ¼ 14 each group). Production of IL-10 in LPS-stimulated PBMCs ex vivo was measured by ELISA. IL-10 levels at the base line (0 week) are shown as 100%, the levels during and after supplementation are expressed as a percentage of these levels. The washout period was 12 weeks after the supplementation period. Significant differences from presupplementation levels are denoted by (*) at Po0.05 and (**) at Po0.01(t-test). Error bars are 7s.d.

respectively) (Figure 8). In the CLA group, the IL-10 levels were increased at 6 and 12 weeks supplementation by 20 and 30%, respectively but only the 12-week values were significant (Po0.01). Significant increases in the production of IL-10 ex vivo on stimulation of PBMCs with LPS were also seen in the reference-oil group following supplementation for 6 and 12 weeks (25, and 40%, both Po0.05, respectively).

*

110

70

REFERENCE CLA

wash out

Figure 7 Effect of reference (filled bars) and CLA (open bars) supplementation on IL-1b levels in healthy young volunteers (n ¼ 14 each group). Production of IL-1b in LPS-stimulated PBMCs ex vivo was measured by ELISA. IL-1b levels at the baseline (0 week) are shown as 100%, the levels during and after supplementation are expressed as a percentage of these levels. The washout period was 12 weeks after the supplementation period. Significant difference from presupplementation levels is denoted by (*) Po0.05 (t-test). Error bars are 7s.d.

the release of IL-1b after 6 weeks by 10% (Po0.05). However, this decrease was not evident after 12 weeks of supplementation and/or the washout period. The reference-oil supplemented group tended to increase the IL-1b levels during the supplementation period at both 6 and 12 weeks (10 and 15%, respectively) but not significantly. The IL-1b levels were slightly decreased after the washout period, but again not significantly. Both CLA and reference-oil groups gradually and significantly increased the levels of the anti-inflammatory cytokine, IL-10 (Po0.01 and 0.05

Subset analysis results The proportions of natural killer cells were not different among the treatment groups at baseline and were not significantly affected by the treatments (Table 3). The proportions of T helper lymphocytes were not changed in the control group. However, the CLA group showed nonsignificant increases (by approximately 10%) during and after supplementation (Table 3). Again, the referenceoil group did not change the proportions of T lymphocytes as cytotoxic cells but the CLA group gradually, but nonsignificantly, decreased these cells at 6 and 12 weeks and after the washout period (2, 6 and 12% respectively). The proportion of monocytes expressing ICAM-1 was increased nonsignificantly by approximately 10% after supplementation in the reference-oil group but was not affected by CLA (Table 3). Both the reference-oil and CLA group increased the proportion of lymphocytes expressing Lselectin; however, the increased L-selectin expression was more evident in the control group but still not significant (Table 3). European Journal of Clinical Nutrition


514 Table 3 PBMC subsets in the treatment groups Helper T cellsa Treatment groups

Cytotoxic T cellsb

NK cellsc

ICAM-1d

L-selectine

Time (weeks)

Mean

s.d.

Mean

s.d.

Mean

s.d.

Mean

s.d.

Mean

s.d.

0 6 12 Washout

100 102.7 105.4 98.2

9.0 10.1 12.0

100 96.0 101.2 96.1

18.7 20.5 14.2

100 99.9 97.7 99.0

1.5 2.0 0.5

100 114.5 110.29 110.02

18.0 14.4 16.7

100 123.9 121.3 104.9

18.9 15.5 18.1

0 6 12 Washout

100 107.2 112.6 109.8

7.7 6.4 5.9

100 98.4 94.05 88.44

6.3 5.6 6.7

100 101.2 100.6 99.1

1.57 1.99 1.24

100 106.9 96.73 106.4

15.6 16.4 14.5

100 119.3 111.8 97.4

18.4 13.4 14.6

Control

CLA

a

Defined as % of CD3 þ lymphocytes staining positive for CD4. Defined as % of CD3 þ lymphocytes staining positive for CD8. c Defined as % of lymphocytes staining negative for CD3 and positive for CD16. d Defined as % of monocytes staining negative for CD14 and positive for CD54. e Defined as % of lymphocytes staining negative for CD3 and positive for CD62. b

Discussion The effects of dietary supplementation with CLA or high oleic sunflower oil (the reference oil) on plasma lipid concentrations were small (Table 2). Despite CLA supplementation having no effect on cholesterol levels, high oleic sunflower oil supplementation significantly decreased total cholesterol levels. These findings support previous observations that this reference oil reduces cholesterol levels. CLA significantly increased TAG levels while significantly decreasing FFA levels by 30% at 6 weeks (Po0.01) in the present study. This is possibly due to the effect of CLA on general lipid metabolism but particularly its documented inhibitory effect on the activity of lipoprotein lipase (LPL) (Xu et al, 2003) which would be expected to decrease plasma FFA levels. Previous studies reported that CLA, at 3.9 g/day for 63 days (Benito et al, 2001) and at 4.2 g/day for 12 weeks (Smedman and Vessby, 2001), did not elicit any differences on plasma lipid concentrations. In another human trial (Noone et al, 2002) showed that CLA supplementation (3 g/ day, 80:20 of CLA 9–11 and 10–12 isomeric blend) significantly reduced VLDL-cholesterol concentrations (Noone et al, 2002). This result may indicate that the CLA 9–11 isomer may be the more potent isomer in terms of having a beneficial effect on plasma lipids. This hypothesis could be addressed by supplementation studies using purified individual isomers of CLA. DTH reactions are cell-mediated immunity (CMI) responses that, depending on the antigen involved, mediate beneficial (resistance to viruses, bacteria, fungi and tumours) or harmful (allergic dermatitis, autoimmunity) aspects of immune function. CMI is related to T cells that bind to the surface of other immune cells that display specific antigens (antigen-presenting cells) and then these antigens initiate the release of various cytokines, in particular IFN-g. This process is more rapid and effective than other types of antigen-induced responses. Tuberculin infection induces these rapid T-cell responses. Intradermal injection of tuberEuropean Journal of Clinical Nutrition

culin (the tuberculin skin test) is used worldwide to determine whether an individual has an immunological reactivity to mycobacterial antigens. While the tuberculin skin test is a useful aid in identifying tuberculosis infection, it has a number of drawbacks. These include the need for a return visit to the clinic or laboratory to allow for dermal readings, problems in interpretation due to cross-reactivity with other mycobacterial species, the booster effect, falsenegative results because of intercurrent immunosuppression, as well as the variability inherent in its subjective application and reading. For these reasons, the whole blood samples were incubated with TB antigen and the IFN-g released in response to the antigen was determined in the current study. This represents a more direct and less subjective determination of the DTH response. The 30% decrease in the classic CMI cytokine (IFN-g) response elicited by CLA supplementation at 12 weeks. Additionally, this decreasing effect was continued after the washout period (35%, Po0.05). This result indicates an important regulatory role for this fatty acid in cell-mediated immunity. However, the analyses of CLA isomer concentrations in plasma and white blood cells during the washout period may identify whether the effects observed during this period are directly related to residual CLA in future. This is, to our knowledge, the first report of such a regulatory role of CLA in human volunteers and contrasts with the observations of Kelley et al (2000, 2001). There are three possible mechanisms by which these fatty acids can regulate CMI. Firstly, by altering the EFA concentrations in the phospholipids of the plasma membranes of lymphocytes they could cause alterations in the cell’s immunological reactivity via changes in membrane fluidity or through the activity of membrane bound enzyme systems. Secondly, effects of these fatty acids on CMI may be mediated by their known inhibition of prostaglandin (PG) production, particularly of the E-type PG which can influence lymphocyte activity via modulation of cyclic AMP and cyclic GMP activity (Mertin, 1981). Thirdly, CLA


515 have been reported to directly affect gene expression of inflammatory cytokines like IL-1b and TNF-a and this may also be true for IFN-g (Cook et al, 2003). The precise mechanisms remain to be elucidated. Sugano et al (1998) showed that, in rats, dietary CLA enhanced immunoglobulin (Ig) production in immunocompetent organs and increased plasma IgA, IgG and IgM concentrations while those of IgE decreased significantly. This is the first known report of a change in IgE with CLA ingestion. Yamasaki et al (2000) observed that CLA enhanced overall Ig production in rat spleen but did not affect serum IgG levels. These observations are in general agreement with our present findings in human volunteers which showed that plasma IgA was significantly increased (Po0.05) but IgE was decreased (Po0.05) during and after CLA supplementation suggesting a decreased sensitivity to allergen actions. CLA also increased IgM levels throughout this human volunteer study (Po0.05). These increases in Ig levels would be expected to enhance the volunteers’ capacity to combat infection, in particular the increases in IgM since this immunoglobulin is the vanguard of this defence against infection. The cytokine results in the present study demonstrate the ability of CLA to modulate the release of inflammatory IL-1b, TNF-a and anti-inflammatory IL-10 in LPS-stimulated PBMC ex vivo when given as a dietary supplement. The present observations in human volunteers support the findings in animals where a suppression of inflammatory TNF-a production and a reduction in immune-induced cachexia was observed following feeding with CLA (Akashoshi et al, 2002; Cook et al, 2003). Published investigations of the effect of dietary CLA on cytokine production in human subjects is surprisingly low. However, similar findings of a suppression of IL-1b and TNF-a production following fish oil supplementation have been reported (Cooper et al, 1993; Endres et al, 1994; Meydani, 1996). This indicates that specific fatty acids can significantly regulate cytokine production in man. The observed increase in production of the anti-inflammatory cytokine, IL-10 by 20% following 12 weeks of dietary supplementation of CLA (Po0.01) has not been shown previously and supports the suggestion that CLA can attenuate inflammatory responses in man. Interestingly, a similar significant increase in IL-10 release was observed in the reference group (by 25% at 12 weeks, Po0.05). This is possibly an effect of the high levels of cis-monounsaturated fatty acids, which the reference group consumed throughout the study. Monounsaturated fatty acids reduce the oxidative stress potential in vivo and favour maintenance of tissue and plasma antioxidant levels compared to certain PUFA (Collette et al, 2003). Surprisingly, the plasma concentrations of highly sensitive C-reactive protein, regarded as an indicator of general oxidative stress and a biomarker for CVD risk (Sattar et al, 2003), were not affected by CLA ingestion in these volunteers (results not shown) despite the reduction in TNF-a elicited by ingestion of this fatty acid. In contrast to these findings, Riserus et al (2002) reported a significant

effect of both the CLA mix and the individual 10,12 isomer on plasma CRP in obese patients CLA supplementation in this study did not result in any significant changes in plasma glucose levels. Riserus et al (2002) showed that the 10,12 isomer of CLA resulted in adverse effects on insulin resistance using the euglycaemic clamp. In the present study, the lack of change in plasma glucose with CLA mix suggests that this treatment is not having any adverse effects on glucose homeostatsis in healthy volunteers, but we did not conduct euglycaemic clamp studies, which are a more sensitive indicator of insulin resistance (DeFronzo et al, 1979). Riserus et al (2002) did not report any adverse effects of CLA mix on insulin resistance in obese volunteers, which supports our findings with the mix in healthy volunteers. In the current study, there was a trend for CLA to increase the proportions of helper T lymphocytes and decreased cytotoxic T lymphocytes. However, CLA did not affect the proportion of NK cells. These results differ from those found by Cook et al (1999) in naı¨ve mice. However, they agree with a previous human subject study found that CLA had no significant effect on T lymphocytes, helper T lymphocytes and cytotoxic T lymphocytes in young healthy women (Kelley et al, 2000). Expression of adhesion molecules, ICAM-1 and L-selectin, were not affected either in the CLA or reference group (Table 3). This result may be due to the use of nonstimulated PBMC, in the present study. Preliminary work from our laboratory has shown that CLA can attenuate the cytokineinduced expression of adhesion molecules on smooth muscle cells in culture (Goua and Wahle, 2002). Similarly, Collie-Duguid and Wahle (1996) showed that EPA and DHA did not affect the expression of ICAM-1, VCAM-1 or Eselectin in resting human umbilical vascular endothelial cells but decreased their cytokine-induced expression. There are only a few studies in which the effects of CLA supplementation on human immune cell function were examined and the results were not positive (Kelley et al, 2000, 2001; Albers et al, 2003). Importantly, the current data show for the first time the beneficial effects of CLA on various aspects of immune function in healthy human subjects. There are a number of possible reasons for the reported negative effects of CLA on immune function in man. These include the different methods employed by these authors to assess immune function, different supplementation regimens and different types of subjects. For example, to assess the DTH response, Kelley et al (2001) and Albers et al (2003) used the multitest CMI kit that injected seven antigens to volunteers whereas we used the more direct whole blood treatment with PPD and determination of IFN-g. Moreover, most previous studies used the FFA forms of CLA 50:50 main isomers ratio (CLA 9–11: CLA 10–12), whereas the current study used the triacylglyceride form of the CLA isomers, which may affect rates of absorption from the intestinal tract. Measuring the levels of CLA in plasma from volunteers on both forms of CLA would confirm this. European Journal of Clinical Nutrition


516 In conclusion, the current study examined the impact of feeding CLA to healthy volunteers on specific markers of immune function. It is one of the few studies reporting the effects of CLA on immune function. Overall, our results suggest that supplementation of healthy human volunteers with CLA mix could improve their ability to respond more effectively to infectious agents and enhance their resistance to allergic agents. This is the first time that IgE levels have been measured in healthy human volunteers in response to CLA supplementation. Similarly, CLA supplementation reduced the level of the inflammatory cytokine IL-1b but was without effect on CRP. However, further studies with healthy volunteers and specific concentrations of individual isomers and mixtures of CLA are required to confirm the above effects and to identify the most effective individual isomer.

Acknowledgements We thank Loders-Croklaan, The Netherlands and SEERAD (Scottish Executive Environmental Rural and Agriculture Department) for their sponsorship. In addition we thank all volunteers for their participation.

References Akashoshi A, Goto Y, Murao K, Miyazaki T, Yamasaki M, Nonaka M, Yamada K & Sugano M (2002): Conjugated linoleic acid reduces body fats and cytokine levels of mice. Biosci. Biotechnol. Biochem. 66, 916–920. Albers R, wan der Wielen RPJ, Brink EJ, Hendriks HFJ, DorovskaTaran VN & Mohede ICM (2003): Effect of cis-9, trans-11 and trans10, cis-12 conjugated linoleic acid (CLA) isomers on immune function in heatlhy men. Eur. J. Clin. Nutr. 57, 595–603. Benito P, Nelson GJ, Kelley DS, Bartolini G, Schmidt PC & Simon V (2001): The effect of conjugated linoleic acid on plasma lipoproteins and tissue fatty acid composition in humans. Lipids 36, 229–236. Cesano A, Visonneau S, Scimeca JA, Kritchevsky D & Santoli D (1998): Opposite effects of linoleic acid and conjugated linoleic acid on human prostatic cancer in SCID mice. Anticancer Res. 18, 1429–1434. Chin SF, Liu W, Storkson JM, Ha YL & Pariza M (1992): Dietary sources of conjugated dienoic isomers of linoleic acid, a newly recognized class of anticarcinogens. J. Food Comp. Ann. 5, 185–197. Collette C, Percheron C, Pares-Herbute N, Michel F, Pham T-C, Brillant L, Descomps B & Monnier L (2003): Exchanging carbohydrates for monounsaturated fats in energy-restricted diets: effects on metabolic profile and other cardiovascular risk factors. Int. J. Obes. Relat. Metab. Disord. 27, 648–656. Collie-Duguid ESR & Wahle KWJ (1996): Inhibitory effect of fish oil n-3 polyunsaturated fatty acids on the expression of endothelial cell adhesion molecules. Biochem. Biophys. Res. Commun. 220, 969–974. Cook ME, DeVoney D, Drake B, Pariza MW, Whigham L & Yang M (1999): In Advances in Conjugated Linoleic Acid Research, Vol. 1, Chapter 17, Yurawecz MP, Mossoba MM, Kramer JKG, Pariza MW & Nelson GJ, eds, pp 226–237. Champaign, IL: AOCS Press. Cook ME, Butz D, Li G, Pariza M, Whigham L & Yang M (2003): Conjugated linoleic acid enhances immune responses but protects against the collateral damage of immune events. In Advances in Conjugated Linoleic Acid Research, Vol. 2, Sebedio J, Christie WW & Adolf R, eds, pp 283–291. Champaign, IL: AOCS Press.


Cooper AL, Gibbons L, Horan MA, Little RA & Rothwell NL (1993): Effect of dietary fish oil supplementation on fever and cytokine production in human volunteers. Clin. Nutr. 12, 321–328. DeFronzo RA, Tobin JD & Andres R (1979): Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am. J. Physiol. 237, E214–E223. Endres S, Eisenhut T & Sinha B (1994): n-3 polyunsaturated fatty acids in the regulation of human cytokine synthesis. Biochem. Soc. Trans. 23, 277–281. Goua M & Wahle KWJ (2002): Fatty acid modulation of adhesion molecule expression in human aortic smooth muscle cells. Chem. Phys. Lipid ICBL Abstract, p 118. Griinari JM, Coral BA, Lacy SH, Chouinard PY, Normela KV & Baumann DE (2000): Conjugated linoleic acid is synthesized endogenously lactating dairy cows by delta (9)-desaturase. J. Nutr. 130, 2285–2291. Hayek MG, Han SN, Wu D, Watkins BA, Meydani M, Dorsey JL, Smith DE & Meydani SN (1999): Dietary conjugated linoleic acid influences the immune response of young and old C57BL/6NCrlBR mice. J. Nutr. 129, 32–38. Ip C, Singh M, Thompson HJ & Scimeca JA (1994): Conjugated linoleic acid suppresses mammary carcinogenesis and proliferative activity of the mammary gland in the rat. Cancer Res. 54, 1212–1215. Jiang J, Wolk A & Vessby B (1999): Relation between the intake of milk fat and the occurrence of conjugated linoleic acid in human adipose tissue. Am. J. Clin. Nutr. 70, 21–27. Kelley DS, Taylor PC, Rudolph IL, Benito P, Nelson GL, Mackey BE & Erickson KL (2000): Dietary conjugated linoleic acid did not alter immune status in young healthy women. Lipids 35, 1055–1166. Kelley DS, Simon VA, Taylor PC, Rudolph IL, Benito P, Nelson GL, Mackey BE & Erickson KL (2001): Dietary supplementation with conjugated linoleic acid increased its concentration in human peripheral blood mononuclear cells, but did not alter their function. Lipids 36, 669–674. Mertin J (1981): Essential fatty acids and cell-mediated immunity. Prog. Lipid. Res. 20, 851–856. Meydani SN (1996): Effect of (n-3) polyunsaturated fatty acids on cytokine production and their biologic function. Nutrition 12, S8–S14. Noone EJ, Roche HM, Nugent AP & Gibney MJ (2002): The effect of dietary supplementation using isomeric blends of conjugated linoleic acid on lipid metabolism in healthy human subjects. Br. J. Nutr. 88, 243–251. Packard CJ, O’Reilly DS, Caslake MJ, McMahon AD, Ford I, Cooney J, Macphee CH, Suckling KE, Krishna M, Wilkinson FE, Rumley A & Lowe GD (2000): Lipoprotein-associated phospholipase A2 as an independent predictor of coronary heart disease. West of Scotland Coronary Prevention Study Group. N. Engl. J. Med. 343, 1148–1155. Parodi PW (1994): Conjugated linoleic acid: an anticarcinogenic fatty acid present in milk fat. Austr. J. Dairy Technol. 49, 93–97. Riserus U, Basu S, Jovinge S, Fredrikson GN, Arnlov J & Vessby B (2002): Supplementation with conjugated linoleic acid causes isomer-dependent oxidative stress and elevated C-reactive protein: a potential link to fatty acid-induced insulin resistance. Circulation 106, 1925–1929. Sattar N, Gaw A, Scherbakova O, Ford I, O’Reilly DS, Haffner SM, Isles C, Macfarlane PW, Packard CJ, Cobbe SM & Shepherd J (2003): Metabolic syndrome with and without C-reactive protein as a predictor of coronary heart disease and diabetes in the West of Scotland Coronary Prevention Study. Circulation 108, 414–419. Smedman A & Vessby B (2001): Conjugated linoleic acid supplementation in humans—The metabolic effects. Lipids 36, 773–781. Sugano M, Tsujita A, Yamasaki M, Noguchi M & Yamada K (1998): Conjugated linoleic acid modulates tissue levels of chemical mediators and immunoglobulins in rats. Lipids 33, 521–527. Turek JJ, Li Y, Schoenlein IA, Allen KGD & Watkins BA (1998): Modulation of macrophage cytokine production by conjugated linoeic acids is influenced by the dietary n-6:n-3 fatty acid ratio. J. Nutr. Biochem. 9, 258–266.


517 Visonneau S, Cesano A, Tepper SA, Scimeca JA, Santoli D & Kritchevsky D (1997): Conjugated linoleic acid suppresses the growth of human breast adenocarcinoma cells in SCID mice. Anticancer Res. 17, 969–973. Xu X, Storckson J, Kim S, Sugimoto K, Park Y & Pariza MW (2003): Short-term intake of conjugated linoleic acid inhibits lipoprotein lipase and glucose metabolism but does not enhance lipolysis in mouse adipose tissue. J. Nutr. 133, 663–667.

Yamasaki M, Kishihara K, Mansho K, Ogino Y, Kasai M, Sugano M, Tachibana H & Yamada K (2000): Dietary effect of conjugated linoleic acid increases immunoglobulin productivity of SpragueDawley rat spleen lymphocytes. Biosci. Biotechnol. Biochem. 64, 2159–2164. Yang M, Pariza MW & Cook ME (2000): Dietary conjugated linoleic acid protects against end stage disease of systemic lupus erythematosus in the NZB/W F1 mouse. Immunopharmacol. Immunotoxicol. 22, 433–449.
