Effect of soy-based breakfast cereal on blood lipids and oxidized low-density lipoprotein

Effect of Soy-Based Breakfast Cereal on Blood Lipids and Oxidized Low-Density Lipoprotein David J.A. Jenkins, Cyril W.C. Kendall, Edward Vidgen, Vladimir Vuksan, Chung-Ja Jackson, Livia S.A. Augustin, Brenda Lee, Marcella Garsetti, Sanjiv Agarwal, A. Venket Rao, Gloria B. Cagampang, and Victor Fulgoni III Consumption of soy protein may reduce the risk of cardiovascular disease both through reduction in serum lipids and by the antioxidant properties of protein-associated soy isoflavones. However, the effect that processing required for the manufacture of breakfast cereals may have on the lipid lowering and antioxidant activities of soy has not been studied. We have therefore assessed the health benefits of soy incorporation into breakfast cereals. Twenty-five hyperlipidemic men and women took soy (providing 36 g/d soy protein and 168 mg/d isoflavones) and control breakfast cereals, each for 3 weeks in a randomized crossover study with a 2-week washout period between treatments. Fasting blood samples were obtained pretreatment and at weeks 2 and 3 of each treatment. No significant difference was seen in serum lipids between treatments at week 3 apart from a 3.8% ⴞ 1.5% higher apolipoprotein A-1 level on control versus soy (P ⴝ .021). However, oxidized low-density lipoprotein (LDL) was reduced on the test compared with the control both as total dienes in LDL and as the ratio of conjugated dienes to cholesterol in the LDL fraction by 9.2% ⴞ 4.3% (P ⴝ .042) and 8.7% ⴞ 4.2% (P ⴝ .050), respectively. High isoflavone intakes in soy breakfast cereals may decrease the risk of cardiovascular disease by reducing oxidized LDL, while having no significant effect on the absolute concentration of LDL cholesterol. Copyright r 2000 by W.B. Saunders Company

S

OY FOODS ARE recognized to contain potential antioxidants as isoflavones, and a number of studies have shown the antioxidant properties of the soy protein-associated isoflavones both in vivo and in vitro.1-3 This property of soy is of particular interest at a time when the US Food and Drug Administration (FDA) have accepted health claims for soy protein in relation to cardiovascular disease reduction through cholesterol lowering.4 The antioxidant properties of soy therefore provide an additional mechanism by which consumption of soy foods may decrease the risk of coronary heart disease. Breakfast cereals may be a convenient vehicle for soy consumption especially for those unfamiliar with soy foods. Nevertheless, the dry heat required in the manufacture of breakfast cereals may modify soy protein structure and so influence its physiologic effects. We have therefore assessed the effect of soy-based breakfast cereal on serum lipids and oxidized low-density lipoprotein (LDL). Our aim was to determine whether consumption of a moderate amount of soy protein and associated isoflavones in breakfasts cereal form can favorably alter these markers of increased cardiovascular disease risk.5-7

From the Clinical Nutrition and Risk Factor Modification Center and Department of Nutritional Sciences, St. Michael’s Hospital, University of Toronto, Toronto, Ontario; Laboratory Services Division, University of Guelph, Guelph, Ontario, Canada; McKee Foods Corporation, Collegedale, TN; Nutrition Impact, Battle Creek, MI; and the Division of Human Nutrition, Department of Food Science and Technology and Microbiology, University of Milan, Milan, Italy. Submitted February 7, 2000; accepted April 9, 2000. Supported by the University-Industry Research Partnership Program of The Natural Sciences and Engineering Research Council of Canada and The Kellogg Company, Toronto, Ontario, Canada. Address reprint requests to David J.A. Jenkins, MD, Clinical Nutrition and Risk Factor Modification Center, St. Michael’s Hospital, 61 Queen St East, Toronto, Ontario, Canada, M5C 2T2. Copyright r 2000 by W.B. Saunders Company 0026-0495/00/4911-0011$10.00/0 doi:10.1053/meta.2000.17703 1496

SUBJECTS AND METHODS Twenty-five hyperlipidemic subjects (15 men, 10 postmenopausal women) aged 60 ⫾ 2 years (range, 42 to 79) with a body mass index of 25.1 ⫾ 0.5 kg/m2 (range, 20.7 to 31.5 kg/m2 ) completed two 3-week ad libitum diets separated by at least a 2-week washout period in a randomized crossover design. During the washout period, subjects returned to their pretreatment self-selected National Cholesterol Education Program (NCEP) Step 2 diets. All subjects had elevated serum LDL cholesterol concentrations (⬎4.1 mmol/L)8 and triglyceride levels below 4.0 mmol/L at recruitment. None had clinical or biochemical evidence of diabetes, liver, or renal disease. One woman was taking an hepatic hydroxymethyl glutaryl coenzyme A (HMG CoA) reductase inhibitor, lovastatin, 40 mg/d (Mevacor; Merck Frost Canada, Kirkland, Quebec, Canada); 1 woman was on hormone replacement therapy, 17␤-estradiol, 50 µg/d (Estroderm Patch; Ciba-Geigy Canada, Mississauga, Ontario, Canada); 2 women were taking L-thyroxine (0.15 mg/d); 1 man and 1 woman were taking ␤-blocking agents; and 5 men and 1 woman were taking vitamin E (400 to 800 mg/d). Dosage levels of all medications were held constant for both study periods. Subjects were also asked to keep their level of physical activity constant. Blood samples were obtained after 12- to 14-hour overnight fasts before starting the study and at weeks 2 and 3 of each phase. Body weight was measured at the start and at weekly intervals on both phases, and blood pressure was measured at the beginning and end of each phase. Twenty-four–hour urine collections were obtained on an outpatient basis at the end of both phases. The Ethics Committee of the University of Toronto approved the study. Informed consent was obtained from all subjects.

Diets Subjects had been instructed on a NCEP Step 2 diet (⬍30% energy as total fat, ⬍7% saturated fat, and ⬍200 mg/d dietary cholesterol)4 at least 1 month before starting the study. Subjects were provided with boxes of the breakfast cereal supplement at the start of each treatment period together with electronic scales on which to weigh daily portions. On the control diet, they were also given a measured amount of soy oil to be weighed and consumed daily with food. Containers were returned at the end of each treatment period to be weighed and compliance assessed. The 2 supplements were identical in total fat (9.6 g/d) and daily energy content (test, 376 kcal/d; and control 378 kcal/d) (Table 1). The control supplement was a commercial breakfast cereal product. The soy supplement was produced from soybeans, which had undergone infraMetabolism, Vol 49, No 11 (November), 2000: pp 1496-1500

SOY CEREAL AND OXIDIZED LDL

red irradiation (Infranization, Briess Industries, Chilton, WI), was defatted with hexane (Food Protein Research and Development Center, Texas A & M University, College Station, TX), and the resulting defatted flour used for cereal production. Approximately 70% of the cereal was soy flour. The other major ingredients were malted sugars and wheat and corn flours. The lower fat content of the control supplement was increased to that of the test by addition of 7.9 g soy oil. The test supplement contributed 36 g/d soy protein compared with the control supplement, which provided 8 g/d wheat protein. The difference in calories was made up with carbohydrate (test, 37 g/d; control, 65 g/d). The total isoflavonoid content of the soy supplement was 168 mg/100 g, the daily dose of the soy supplement (diedzein, 880 µg/g; glycitein, 110 µg/g; genistein, 690 µg/g; total isoflavone, 168 mg/100 g of cereal measured as the aglycone).

Analyses Serum stored at ⫺70°C was analyzed in a single batch according to the Lipid Research Clinics protocol9 for total cholesterol, triglyceride, and high-density lipoprotein (HDL) cholesterol, after dextran sulfatemagnesium chloride precipitation using an automated clinical chemistry analyzer (CH1000; Technicon, Tarrytown, NY).10 Previous studies showed that the average between-run coefficient of variation (CV) for these analyses were as follows: total cholesterol, 1.5% (range, 0.8% to 3.2%); HDL cholesterol, 3.2% (range, 1.6% to 5.3%); and triglyceride, 3.0% (range, 1.9% to 5.0%).11 LDL cholesterol concentrations were calculated.12 Serum apolipoprotein A-1 and B were measured by end-point nephelometry (Behring Werke, Marburg, Germany).13 Within run CV for apolipoprotein A-1 was 3.4% (range, 3.0% to 3.5%) and for apolipoprotein B, the CV was 2.7% (range, 1.8% to 2.9%).11 For direct assessment of LDL oxidation, LDL particles were isolated by precipitation with buffered heparin at their isoelectric point (pH 5.05).14 The LDL precipitate was centrifuged at 1,000 ⫻ g and resuspended in saline. LDL cholesterol was estimated enzymatically15 on an aliquot of the saline resuspension using a commercial cholesterol assay kit (Sigma Chemical, St Louis, MO). On a further aliquot, LDL oxidation was estimated as conjugated dienes in LDL fatty acids. Lipids from the resuspended LDL were extracted with chloroform:methanol (2:1), dried under nitrogen, dissolved in cyclohexane, and analyzed spectophotometrically at 234 nm using a molar extinction coefficient of 29,500 mol⫺1 · L · cm⫺1 for conjugated dienes.16 Oxidized LDL was expressed as total LDL conjugated dienes (µmol/L serum) and as the ratio of conjugated dienes (µmol) per mmol LDL cholesterol.16 The CV for this assay on 6 replicates was 2.5% for conjugated dienes. The isoflavone content of the test and control cereals was measured by high pressure liquid chromatography (HPLC)17,18 using a 600E multisolvent delivery system with a photodiode array detector monitoring at 200 to 350 nm (Waters, Marlborough, MA) and a YMC-pack ODS-AM 303 column (5 µm, 250 mm by 4.6 mm internal diameter) (YMC, Wilmington, NC) equipped with an AM direct connect C18 guard column. Biochanin A was used as an internal standard with recovery values ranging from 80% to 100%. Urinary isoflavones were measured by HPLC after acid hydrolysis.18 Chromatographs were obtained from the 3-dimensional array using a photodiode array detector at 258 nm to allow assessment of common regions of relatively high absorbence for daidzein, genistein, and the added recovery standard, flavone.

Statistical Analysis The results are expressed as means ⫾ SE. Weight change was expressed as kilogram per month. The percentage difference between the end-point (week 3) values for both diets was assessed by Student’s t test (2-tailed) for paired data and 2-sample t test for comparison of treatment effects between subgroups. The absolute difference between treatments was assessed using the General Linear Model procedure and SAS software (PROC GLM/SAS)19 with the end of treatment value as

1497

Table 1. Daily Contribution and Composition of Cereals as Analyzed

Daily supplement Energy (kcal) Protein Fat SFA MUFA PUFA Available carbohydrate Total dietary fiber

Control (g/d)

Soy (g/d)

Control (% of energy)

Soy (% of energy)

96 378* 8.0* 9.6* 1.0* 5.1* 3.0*

100 376 36.0 9.6 1.5 2.2 5.4

8.5 22.9 2.4 12.1 7.2

38.3 23.0 3.5 5.3 13.0

65.0* 8.0*

37.0 11.0

68.8 21.2†

39.4 29.3†

Abbreviations: SFA, saturated fatty acids; MUFA, monounsaturated fatty acids; PUFA, polyunsaturated fatty acids. *Includes 7.9 g/d soy oil. †Values expressed as g/1000 kcal.

the response variable and the following main effects: diet, diet by sex, sex by sequence, a random term representing subject nested within the sex by sequence interaction, and the baseline value as a covariate where measured. RESULTS

Of the 25 subjects, 11 received the test cereal first. The diets were well accepted and compliance was good, with 99.5% of the test and 99.4% of the control breakfast cereals being consumed, and 100% of the soy oil provided was consumed on the control diet. Significant differences were seen between diets in the percentage of energy derived from protein and carbohydrate due to replacement of protein by carbohydrate in the control breakfast cereal (Tables 1 and 2). There were no significant changes in body weight across either treatment, and Table 2. Calculated Dietary Intakes (Mean ⴞ SE) for Weeks 2 and 3 of Treatment Periods (n ⴝ 25)

Energy (kcal/d) Total protein (g/d) (%) Animal protein (g/d) (%) Vegetable protein (g/d) (%) Animal to vegetable protein ratio Available carbohydrate (g/d) (%) Total dietary fiber (g/d) (g/1,000 kcal) Total fat (g/d) (%) SFA (g/d) (%) MUFA (g/d) (%) PUFA (g/d) (%) Dietary cholesterol (mg/d) (mg/1,000 kcal) Alcohol (g/d) (%)

Control

Soy

P

1,975 ⫾ 112 77 ⫾ 5 15.7 ⫾ 0.4 43 ⫾ 4 8.5 ⫾ 0.5 35 ⫾ 2 7.2 ⫾ 0.3

1,856 ⫾ 90 97 ⫾ 3 21.3 ⫾ 0.5 37 ⫾ 3 8.0 ⫾ 0.5 60 ⫾ 2 13.3 ⫾ 0.5

.149 ⬍.001 ⬍.001 .061 .226 ⬍.001 ⬍.001

1.3 ⫾ 0.1 273 ⫾ 15 55.5 ⫾ 1.0 28 ⫾ 2 14.6 ⫾ 0.8 58 ⫾ 4 26.3 ⫾ 0.8 15 ⫾ 1 6.8 ⫾ 0.3 23 ⫾ 2 10.5 ⫾ 0.4 14 ⫾ 1 6.4 ⫾ 0.3 159 ⫾ 14 79 ⫾ 4 8⫾2 2.5 ⫾ 0.8

0.6 ⫾ 0.0 234 ⫾ 14 50.1 ⫾ 1.2 29 ⫾ 2 16.2 ⫾ 0.9 55 ⫾ 3 26.5 ⫾ 1.0 15 ⫾ 1 7.0 ⫾ 0.4 19 ⫾ 1 9.2 ⫾ 0.4 16 ⫾ 1 8.1 ⫾ 0.4 145 ⫾ 16 76 ⫾ 7 6⫾1 2.2 ⫾ 0.4

⬍.001 .004 ⬍.001 .399 .012 .310 .818 .717 .601 .003 .001 .033 ⬍.001 .355 .676 .272 .541

1498

JENKINS ET AL

final body weights were similar (Table 3). On the soy diet, the 24-hr urinary total isoflavone output was 80.2 mg/d. No isoflavones were detected in the urine on the control phase. Serum Lipids No significant treatment differences were seen in baseline values between serum lipids, lipoproteins, or their ratios Table 3. The only difference between treatments at week 3 was a 3.8% ⫾ 1.5% (P ⫽ .021) higher apolipoprotein A-1 on the control compared with the test period. This effect was confirmed by a borderline significance between absolute test and control treatments using the General Linear Model procedure (1.56 ⫾ 0.04 g/L v 1.49 ⫾ 0.05 g/L, respectively, P ⫽ .066). Exclusion of the 3 subjects taking lipid-lowering medications further reduced the borderline significance in apolipoprotein A-1 (P ⫽ .200). There were no significant absolute or percentage treatment differences in the apolipoprotein B:A-1 ratio, LDL cholesterol, or the LDL:HDL cholesterol ratio. No interaction between drugs and diet was seen using the General Linear Model procedure.

the ratio of conjugated dienes to LDL cholesterol in the LDL fraction (P ⫽ .032). Three subjects (1 vitamin E taker and 2 nontakers) had baseline values for LDL conjugated dienes, which were more than 2 standard deviations from the mean. When these subjects were excluded, a significant difference in the ratio of conjugated diene to LDL cholesterol was noted between subjects taking vitamin E and those who did not take vitamin E supplements. The mean of the 2 baseline and the 3-week control values of the vitamin E takers (n ⫽ 5) was significantly below the mean value of those not on vitamin E (P ⫽ .036). Furthermore, assessment of these 2 groups separately indicated that only in the subjects not taking vitamin E (n ⫽ 17) was a decrease seen in LDL conjugated dienes on the soy cereal phase. The change on soy in conjugated dienes in the LDL fraction was ⫺11.7% ⫾ 5.0%, P ⫽ .034 for vitamin E nontakers versus 2.6% ⫾ 12.5%, P ⫽ .847 for vitamin E takers. The respective figures for the ratio of conjugated dienes to LDL cholesterol were ⫺12.9% ⫾ 4.7%, P ⫽ .015 versus 4.7% ⫾ 12.2%, P ⫽ .719. The significance of the soy effect was confirmed using the General Linear Model Procedure. Individual data points are shown in Fig 1.

Oxidized LDL There was evidence of reduced oxidized LDL on the test compared with the control at the end of treatment (Table 3). Total conjugated dienes in the LDL fraction as a marker of oxidized LDL cholesterol were significantly reduced on the test compared with the control (9.2% ⫾ 4.3%, P ⫽ .042), and the ratio of conjugated dienes to cholesterol in the LDL fraction was also reduced (8.7 ⫾ 4.2, P ⫽ .050). The respective absolute treatment differences were confirmed using the General Linear Model for both total conjugated dienes in LDL (P ⫽ .021) and

Associations No significant associations were seen between the treatment differences in serum lipids or measures of oxidative stress and the urinary isoflavones either singly or combined either before or after correction for 24-hour urinary creatinine. Differences were seen between treatments in the proportion of dietary mono- and polyunsaturated fatty acids. The control diet contained more monounsaturated fatty acids and the test diet more polyunsaturated fatty acids. However, these dietary

Table 3. Mean (ⴞSE) Body Weight, Serum, and Blood Pressure Data on Control and Soy Treatment Periods (n ⴝ 25) Control Cereal Baseline (week 0)

Body weight (kg) Cholesterol Total-C (mmol/L) LDL-C (mmol/L)† HDL-C (mmol/L) Triglyceride (mmol/L) Apolipoproteins Apo A-1 (g/L) Apo B (g/L) Ratios Total-C:HDL-C LDL-C:HDL-C† Apo B:Apo A-1 Oxidized LDL LDL conjugated dienes LDL conjugated dienes: LDL-C Blood pressure (mm Hg) Systolic Diastolic

Soy Cereal

Treatment (week 2)

Treatment (week 3)

69.9 ⫾ 2.3

69.6 ⫾ 2.3

69.8 ⫾ 2.3

6.92 ⫾ 0.29 4.68 ⫾ 0.25 1.14 ⫾ 0.06 2.41 ⫾ 0.36

6.86 ⫾ 0.27 4.67 ⫾ 0.28 1.19 ⫾ 0.06 2.21 ⫾ 0.22

1.52 ⫾ 0.05 1.80 ⫾ 0.09

Baseline (week 0)

Treatment (week 2)

Treatment (week 3)

69.7 ⫾ 2.3

69.8 ⫾ 2.3

69.8 ⫾ 2.3

.351

6.79 ⫾ 0.27 4.69 ⫾ 0.27 1.19 ⫾ 0.05 1.99 ⫾ 0.20

6.99 ⫾ 0.25 4.84 ⫾ 0.25 1.20 ⫾ 0.06 2.12 ⫾ 0.21

6.86 ⫾ 0.27 4.73 ⫾ 0.26 1.20 ⫾ 0.06 2.04 ⫾ 0.20

6.67 ⫾ 0.23 4.65 ⫾ 0.24 1.17 ⫾ 0.06 1.87 ⫾ 0.17

.272 .238 .808 .533

1.56 ⫾ 0.05 1.77 ⫾ 0.08

1.56 ⫾ 0.04 1.74 ⫾ 0.07

1.56 ⫾ 0.05 1.77 ⫾ 0.07

1.54 ⫾ 0.05 1.75 ⫾ 0.07

1.49 ⫾ 0.05 1.69 ⫾ 0.07

.066 .215

6.51 ⫾ 0.48 4.35 ⫾ 0.30 1.21 ⫾ 0.07

6.12 ⫾ 0.37 4.15 ⫾ 0.30 1.17 ⫾ 0.07

5.97 ⫾ 0.36 4.11 ⫾ 0.29 1.14 ⫾ 0.06

6.15 ⫾ 0.36 4.26 ⫾ 0.30 1.17 ⫾ 0.06

6.02 ⫾ 0.37 4.14 ⫾ 0.29 1.17 ⫾ 0.06

6.02 ⫾ 0.37 4.20 ⫾ 0.31 1.16 ⫾ 0.07

.572 .410 .168

74 ⫾ 7

67 ⫾ 9

68 ⫾ 4

68 ⫾ 2

65 ⫾ 8

59 ⫾ 4

.021

18 ⫾ 1

15 ⫾ 1

15 ⫾ 1

14 ⫾ 1

14 ⫾ 1

13 ⫾ 1

.032

128 ⫾ 4 84 ⫾ 4

127 ⫾ 3 79 ⫾ 2

128 ⫾ 3 81 ⫾ 2

130 ⫾ 4 81 ⫾ 2

129 ⫾ 4 81 ⫾ 2

127 ⫾ 3 80 ⫾ 2

.276 .452

P*

NOTE. To convert cholesterol and triglyceride values to mg/dL, multiply by 38.67 and 88.57, respectively. To convert apolipoprotein A-1 and B values to mg/dL, multiply by 100. *P represents the significance of the treatment difference between week 3 values for soy and control cereals using the General Linear Model Procedure. †For LDL-C and LDL-C:HDL-C n ⫽ 24. LDL-C could not be calculated for 1 subject at all 4 time points due to high triglyceride concentrations (⬎4 nmol/L).

SOY CEREAL AND OXIDIZED LDL

1499

Fig 1. Conjugated dienes in the LDL function at weeks 0 and 3 on control and soy cereals in vitamin E nontakers (n ⴝ 17, 䊊) and takers (n ⴝ 5, 䊉).

fatty acid treatment differences were not related to corresponding treatment differences in serum lipids. DISCUSSION

The present study indicated a beneficial effect of soy in reducing indices of LDL oxidation. Positive effects of the soy cereal were not observed on the serum lipid profile in this study possibly related to the effect on soy protein of the infrared irradiation and dry heat used in cereal production. The essential features of soy isoflavones, on the other hand, are more resistant to destruction by heat, as indicated by the high soy cereal isoflavone content, which may explain the maintenance of the soy antioxidant effect. This study supports previous reports that soy isoflavone consumption may protect LDL cholesterol from oxidative damage.1-3 Oxidized LDL is considered to be more readily taken up by the macrophages of the scavenger system in the arterial wall and so may contribute to plaque formation.6,7 Consumption of antioxidant flavonoids in tea, fruit, and vegetables, lycopene in tomato products, and vitamin E taken as a supplement have all been associated with a reduced risk of coronary heart disease in some,20 although not all reports.21 Of the antioxidants tested, vitamin E and lycopene have been shown to have powerful antioxidant properties, reducing LDL oxidation and oxidative damage to plasma proteins.5 In this respect, it is of interest that subjects with established cardiovascular disease fed soy as part of their intervention showed angiographic improvement in coronary arterial diameter compared with control subjects.22 We used direct measurement of conjugated dienes in serum as our marker of oxidative damage rather than indirectly by assessing the resistance of LDL to oxidation ex vivo after exposure to an oxidative stress. Conjugated dienes are lipid peroxidation products formed by the oxidation of unsaturated fatty acids and thus appear to be good indicators of oxidative stress.23 Moreover, conjugated dienes have been used previously to study the role of oxidized LDL in the development of atherosclerosis.5 Although it is not possible to define the ideal level of oxidized LDL cholesterol that must be achieved to reduce cardiovascular disease risk, antioxidant drugs (such as probucol) appear to prevent experimental arteriosclerosis6 despite reducing HDL cholesterol levels. In the present study, vitamin E takers tended to have lower

conjugated dienes in their LDL fraction than those who were not taking vitamin E, indicating an antioxidant effect of vitamin E in reducing conjugated diene formation. Furthermore, in vitamin E takers, soy failed to reduce the level of conjugated dienes further. This result differs from an earlier study of ours where, in a similar group of patients, the vitamin E takers responded similarly to the nonusers in showing decreases in conjugated dienes on soy.3 However, in that study, the apparent decrease in conjugated dienes on soy in vitamin E takers may have been accentuated by an unexplained increase in conjugated dienes over the control period with which the soy period was compared.3 The majority of studies in which soy protein foods have been consumed have shown that LDL concentrations are reduced.24,25 Both the amino acid composition of soy proteins and their isoflavone content have been suggested as responsible for the cholesterol lowering action of soy.24-30 Not all studies have reported an isoflavone effect on serum cholesterol,31,32 but other benefits have been found including improved arterial compliance.31 The reason for the failure of soy protein to lower serum cholesterol by comparison with the control in this study is not clear, but could relate to 3 factors. First the control cereal in combination with the soy oil appeared to have a beneficial effect on apolipoprotein A-1. It therefore appears not to have been an entirely neutral control and so may have made it more difficult to show a lipid lowering effect of soy. In this respect, it is of interest that soy consumption resulted in a small reduction in total cholesterol and apolipoprotein B across the test phase. Secondly, the processing of the soy protein may have reduced its cholesterol lowering potential. At present, there is no literature on this issue. Lastly, the vegetable protein content of the control supplement reduced the total difference in vegetable protein between supplements to 28 g. This resulted in an animal to total vegetable protein (soy plus other vegetable proteins) ratio of 1.3 on the control diet versus 0.6 on the test diet. In previous studies in which we have noted benefits of soy on serum lipids, our differences between control versus test diets in animal to total vegetable protein ratios have been greater (3.5 v 0.1 and 2.1 v 0.4).3,33 We conclude that consumption of soy in breakfast cereal reduces the concentration of oxidized LDL cholesterol, most

1500

JENKINS ET AL

likely as a result of the increased intake of isoflavones associated with soy protein. The value of soy incorporation into breakfast cereals therefore appears to be as a vehicle for delivering isoflavones. High isoflavone intake per se was not associated with reduced LDL cholesterol for which the action of unmodified soy proteins may also be required. In the present

study, the dry heat used in cereal manufacture may have modified the action of the soy protein. ACKNOWLEDGMENT The authors wish to thank Yu-Min Li and George Koumbridis who provided excellent technical assistance.

REFERENCES 1. Tikkanen MJ, Wahala K, Ojala S, et al: Effect of soybean phytoestrogen intake on low density lipoprotein oxidation resistance. Proc Natl Acad Sci 95:3106-3110, 1998 2. Wei H, Bowen R, Cai Q, et al: Antioxidant and antipromotional effects of the soybean isoflavone genistein. Proc Soc Exp Biol Med 208:124-130, 1995 3. Jenkins DJ, Kendall CW, Garsetti M, et al: Effect of soy protein foods on low density lipoprotein oxidation and ex vivo sex hormone receptor activity—a controlled crossover trial. Metabolism 49:537-543, 2000 4. United States Food and Drug Administration: FDA Final Rule for: Food Labeling: Health Claims: Soy Protein and Coronary Heart Disease. Federal Register 64(Oct 26):57699-57733, 1999 5. Esterbauer H, Gebicki J, Puhl H, et al: The role of lipid peroxidation and antioxidants in oxidative modification of LDL. Free Radic Biol Med 13:341-390, 1992 6. Steinberg D, Parthasarathy S, Carew TE, et al: Beyond cholesterol. Modifications of low-density lipoprotein that increase its atherogenicity. N Engl J Med 320:915-924, 1989 7. Jialal I, Grundy SM: Effect of dietary supplementation with alpha-tocopherol on the oxidative modification of low density lipoprotein. J Lipid Res 33:899-906, 1992 8. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults: Summary of the Second Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel II). JAMA 269:3015-3023, 1993 9. Lipid Research Clinics Program: Manual of Laboratory Operations. Lipid and Lipoprotein Analysis (revised 1982). Washington, DC, US Government Printing Office, US Department of Health and Human Services Publication no. (NIH) 75-678, 1982 10. Warnick GR, Benderson J, Albers JJ: Dextran sulfate-Mg2⫹ precipitation procedure for quantification of high-density-lipoprotein cholesterol. Clin Chem 28:1379-1388, 1982 11. Jenkins DJ, Wolever TM, Vidgen E, et al: Effect of psyllium in hypercholesterolemia at two monounsaturated fatty acid intakes. Am J Clin Nutr 65:1524-1533, 1997 12. Friedewald WT, Levy RI, Fredrickson DS: Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 18499-18502, 1972 13. Fink PC, Romer M, Hasckel R, et al: Measurement of proteins with the Behring Nephelometer. A multicentre evaluation. J Clin Chem Clin Biochem 27:261-276, 1989 14. Wieland H, Seidel D: A simple specific method for precipitation of low density lipoproteins. J Lipid Res 24:904-909, 1983 15. Allain CC, Poon LS, Chan CS, et al: Enzymatic determination of total serum cholesterol. Clin Chem 20:470-475, 1974 16. Agarwal S, Rao AV: Tomato lycopene and low density lipoprotein oxidation: A human dietary intervention study. Lipids 33:981-984, 1998 17. Wang H-J, Murphy P: Isoflavone composition of American and

Japanese soybeans in Iowa: Effects of variety, crop year, and location. J Agric Food Chem 42:1674-1677, 1994 18. Frank AA, Custer LJ, Cerna CM, et al: Rapid HPLC analysis of dietary phytoestrogens from legumes and from human urine. Proc Soc Exp Med 208:18-26, 1995 19. SAS Institute: SAS/STAT User’s Guide (ed 6.12). Cary, NC, SAS Institute, 1997 20. Rimm EB, Stampfer MJ, Ascherio A, et al: Vitamin E consumption and the risk of coronary heart disease in men. N Engl J Med 328:1450-1456, 1993 21. The Heart Outcomes Prevention Study. Vitamin E supplementation and cardiovascular events in high-risk patients. N Engl J Med 342:154-160, 2000 22. Ornish D, Scherwitz LW, Billings JH, et al: Intensive lifestyle changes for reversal of coronary heart disease. JAMA 280:2001-2007, 1998 23. Vasankari T, Kujala U, Heinonen O, et al: Measurement of serum lipid peroxidation during exercise using three different methods: Diene conjugation, thiobarbituric acid reactive material and fluorescent chromolipids. Clin Chim Acta 234:63-69, 1995 24. Kritchevsky D: Dietary protein, cholesterol and atherosclerosis: A review of the early history. J Nutr 125:589s-593s, 1995 (suppl) 25. Anderson JW, Johnstone BM, Cook-Newell ME: Meta-analysis of the effects of soy protein intake on serum lipids. N Engl J Med 333:276-282, 1995 26. Crouse JR III, Morgan T, Terry JG, et al: A randomized trial comparing the effect of casein with that of soy protein containing varying amounts of isoflavones on plasma concentrations of lipids and lipoproteins. Arch Intern Med 159:2070-2076, 1999 27. Carroll KK: Review of clinical studies on cholesterol-lowering response to soy protein. J Am Diet Assoc 91:820-827, 1991 28. Baum JA, Teng H, Erdman JW Jr, et al: Long-term intake of soy protein improves blood lipid profiles and increases mononuclear cell low-density-lipoprotein receptor messenger RNA in hypercholesterolemic, postmenopausal women. Am J Clin Nutr 68:545-551, 1998 29. Anthony MS, Clarkson TB, Bullock BC, et al: Soy protein versus soy phytoestrogens in the prevention of diet-induced coronary artery atherosclerosis of male cynomolgus monkeys. Arterioscler Thromb Vasc Biol 17:2524-2531, 1997 30. Potter SM: Soy protein and serum lipids. Curr Opin Lipidol 7:260-264, 1996 31. Nestel PJ, Yamashita T, Sasahara T, et al: Soy isoflavones improve systemic arterial compliance but not plasma lipids in menopausal and perimenopausal women. Arterioscler Thromb Vasc Biol 17:3392-3398, 1997 32. Hodgson JM, Puddey IB, Beilin LJ, et al: Supplementation with isoflavonoid phytoestrogens does not alter serum lipid concentrations: A randomized controlled trial in humans. J Nutr 128:728-732, 1998 33. Jenkins DJ, Kendall CW, Vidgen E, et al: The effect on serum lipids and oxidized LDL of supplementing self-selected low-fat diets with soluble fiber, soy and vegetable protein foods. Metabolism 49:67-72, 2000