Acute Hyperlipidemia Increases Oxidative Stress More in African Americans Than in White Americans

Abstract

Oxidative stress emerges as a potential factor in the pathogenesis of hypertension and a common signal transduction mechanism by which various risk factors mediate cardiovascular and renal disease. A greater level of oxidative stress could contribute to higher rates of hypertension and related cardiovascular and renal complications in African Americans than in white Americans. The objective of this study was to compare oxidative stress induced by a standardized episode of acute hyperlipidemia in the two groups. Fifteen African Americans (37 ± 1 years, 9 women/6 men, body mass index 30 ± 2 kg/m²) and 15 whites (38 ± 2 years, 9 women/6 men, body mass index 27 ± 1 kg/m²) were evaluated. Acute hyperlipidemia was induced by a 4-h long infusion of Intralipid and heparin. Blood samples were drawn at baseline and after 2 and 4 h of acute hyperlipidemia for nonesterified fatty acids, triglycerides, and F2-isoprostanes, a biomarker of oxidative stress. Plasma nonesterified fatty acids and triglycerides increased significantly and similarly in African Americans and whites after 2 and 4 h of the Intralipid and heparin infusion. Although baseline plasma F2-isoprostanes did not differ between groups, F2-isoprostanes increased more in African Americans than in whites at 2 h (12.2 ± 2.6 v 5.0 ± 2.9 pg/mL, P < .05) and 4 h (13.9 ± 3.1 v 3.0 ± 3.0 pg/mL, P < .05) of acute hyperlipidemia. The results indicate that acute hyperlipidemia increases F2-isoprostanes more in African Americans than in whites. The findings may have important implications for ethnic differences in the prevalence of hypertension and cardiovascular and renal disease. Am J Hypertens 2003;16: 331–336 @ 2003 American Journal of Hypertension, Ltd.

Intralipid, heparin, oxidative stress, F2-isoprostanes, African Americans

Hypertension and related cardiovascular and renal complications are more frequent and severe in African Americans than in whites.¹ Differences in access to health care, socioeconomic status, stress, physical activity, diet, and obesity, explain some ethnic differences.^1,2,3,4,5 However, these factors alone probably do not account for all ethnic differences in cardiovascular and renal risk and disease, which suggests that other biologic variables may contribute significantly.

Oxidative stress has been linked to cardiovascular risk factors and complications.^6,7 For example, hypertension induced by long-term, low-dose infusion of angiotensin II is associated with increased production of superoxide and attenuated by liposomal delivery of superoxide dismutase.⁸ Depletion of glutathione in normal rats induces severe hypertension, which is attenuated by the antioxidant vitamins C and E.⁹ Evidence also suggests that African Americans are more salt sensitive than whites.¹⁰ A high salt diet induces larger increases in circulating and renal markers of oxidative stress in Dahl salt-sensitive than in salt-resistant rats.^11,12 Oxidative stress may contribute not only to hypertension but also to insulin resistance, renal disease, endothelial dysfunction, vascular remodeling, and rupture of atherosclerotic plaque.^{6,7,13,14,15,16}

Collectively, the evidence suggests that ethnic differences in oxidative stress could contribute to variations in the prevalence and severity of cardiovascular and renal risk factors and disease. We have shown that acute hyperlipidemia increases F2-isoprostanes, a biomarker of oxidative stress in human subjects.¹⁷ In this study, we compared the responses of F2-isoprostanes in African Americans and whites to a standardized episode of acute hyperlipidemia in the clinical laboratory.

Methods

Subjects

Thirty volunteers, 21 to 54 years of age, were enrolled in this study. Eighteen of the subjects and some of their data were included in a previous report.¹⁷ The remaining 12 volunteers were selected from the baseline (usual) dietary phase of another study.¹⁸ Subjects for this report were selected from those two studies based on a match of African Americans and whites by age and gender, as well as 24-h urine Na⁺ values within ±20%. All volunteers signed a written, informed consent document approved by the Institutional Review Board. Each volunteer had history, physical, and laboratory examination to verify health status before beginning the study.

Fifteen African Americans (37 ± 1 years, 9 women/6 men, body mass index 30 ± 2 kg/m², 7 hypertensives and 8 normotensives) and 15 whites (38 ± 2 years, 9 women/6 men, body mass index 27 ± 1 kg/m², 10 hypertensives and 5 normotensives) participated in this study. Normotensive subjects were defined by blood pressure (BP) readings on three screening visits consistently less than 130/85 mm Hg. Hypertensive subjects were defined by the presence of BP off medications between 130 and 159 mm Hg systolic or 85 and 99 mm Hg diastolic on the three qualifying visits. Hypertensive subjects discontinued medications at least 2 weeks before participating in the clinical protocol. Hypertensive subjects received equipment for home BP monitoring and were trained in its use. They were instructed to measure their BP twice daily and inform the study doctor/principal investigator if their home BP was equal to or greater than 160/100 mm Hg.

Hemodynamic measurements

Blood pressure and heart rate

Blood pressure at the qualifying visits was measured using a mercury sphygmomanometer. Systolic BP was defined by the first and diastolic by the fifth Korotkoff sound. The qualifying BP were measured in the seated position after 5 min of rest. Three BP readings were obtained at 2-min intervals, and the mean of the last two readings was used. Heart rate was measured between the first and second and the second and third BP measurements, and the values were averaged.

Hemodynamic data

Arterial compliance, stroke volume, cardiac ejection time, large and small vessel distensibility, systemic vascular resistance, and total vascular impedance were measured or calculated by measurement of BP and analysis of the radial artery pulse waveform using the H.D.I./PulseWaveTM CR-2000 (Hypertension Diagnostics, Eagan, MN).¹⁸

Biochemical measurements

Blood samples were obtained on the qualifying visit for fasting glucose and lipid profile, electrolytes, and measures of renal and hepatic function. Blood samples on the study day were obtained for triglycerides, nonesterified fatty acids (NEFA), and F2-isoprostane measurements as previously described.^17,19,20

Plasma NEFA

Blood for NEFA was drawn into chilled tubes containing EDTA and paraoxon to inhibit lipolysis in vitro.²¹ Plasma for NEFA was stored at −70°C until analysis; plasma NEFA concentrations were measured using the ³Ni method.¹⁹

Plasma F2-Isoprostanes

Blood for plasma F2-isoprostanes was placed into chilled tubes containing EDTA and stored at −70°C after separation. Plasma F2-isoprostanes were measured by gas chromatography/negative-ion chemical ionization mass spectrometry.^20,22

Homeostatic model assessment of insulin resistance

Fasting plasma insulin levels were measured on frozen samples (−70°C) by radioimmunoassay. Homeostasis model assessment for insulin resistance (HOMAir) was calculated as described by Matthews et al.²³

Protocol

On the study day, subjects closed the 24-h urine for sodium and creatinine at 7:00 AM then reported to the Clinical Physiology Laboratory in the outpatient General Clinical Research Center at 8:00 AM after an overnight fast. With the volunteer supine, an intravenous catheter was placed in one arm for blood sampling and in the contralateral arm for infusion of Intralipid and heparin. Baseline BP and hemodynamic (HDI PulseWave) measurements were obtained in the supine position at 5-min intervals during a 30-min baseline period. At the end of the baseline period, blood samples were drawn for NEFA, triglycerides, and F2-isoprostanes. Intralipid 20% (Baxter Healthcare Corp. Glendale, CA) was infused at 0.8 mL/m²/min for 4 h. An initial heparin bolus of 1000 U was followed by 200 U/h infusion during the 4-h period to activate endothelial lipoprotein lipase and accelerate hydrolysis of fatty acids from the glycerol backbone of the triglycerides. Blood samples were obtained at 2 and 4 h of Intralipid and heparin infusion for triglycerides, NEFA, and F2-isoprostanes.

Statistical analysis

For nominal, discrete variables (eg, gender), the χ² test was used. Continuous numeric variables such as age, body mass index, BP, and the biochemical and metabolic data were analyzed using the Student paired and unpaired t test as appropriate. Changes in BP and other hemodynamic and metabolic variables from baseline during the infusion of Intralipid and heparin were assessed using the General Linear Model for repeated measures. All statistical analyses were performed with the SPSS/PC 10.0 statistical software package (SPSS, Chicago, IL). A P value less than .05 was accepted as statistically significant.

Results

African American and white volunteers had comparable values, respectively, for age (37 ± 1 v 38 ± 1, P = not significant [NS]), gender (9 men:6 women both groups), and body mass index (30 ± 2 v 27 ± 1, P = NS). Twenty-four-hour urine Na⁺ was also comparable in African Americans and whites (157 ± 11 v 167 ± 11 mmol/d, P = NS). Baseline values for BP, heart rate, total cholesterol, LDL-cholesterol, plasma insulin, NEFA, and insulin resistance as assessed by the HOMAir index were statistically similar in the two groups (Table 1). At baseline, African American subjects had higher plasma glucose and marginally higher HDL-cholesterol as well as lower plasma triglycerides and very low-density lipoprotein-cholesterol compared to whites.

Table 1

Baseline descriptive characteristics of the African American and white volunteers participating in this study

Variables	African Americans (n = 15)	White (n = 15)	P
SBP (mm Hg)	121 ± 3	124 ± 4	.57
DBP (mm Hg)	85 ± 2	84 ± 3	.98
Heart rate (beats/min)	73 ± 2	72 ± 2	.99
Total cholesterol (mg/dL)	196 ± 7	192 ± 12	.75
LDL-cholesterol (mg/dL)	119 ± 8	109 ± 11	.49
HDL-cholesterol (mg/dL)	59 ± 4	50 ± 4	.08
VLDL-cholesterol (mg/dL)	18 ± 2	33 ± 4	.003
Triglycerides (mg/dL)	91 ± 9	166 ± 20	.003
Plasma NEFA (μM/L)	540 ± 58	587 ± 46	.51
Glucose (mg/dL)	92 ± 2	84 ± 2	.02
Insulin (IU/mL)	10.2 ± 1.2	8.3 ± 1.3	.28
HOMAir	42 ± 5	31 ± 5	.15

SBP = systolic blood pressure; DBP = diastolic blood pressure; VLDL = very low-density lipoprotein; NEFA = nonesterified fatty acids; HOMAir = homeostatis model assessment for insulin resistance.

Plasma NEFA and triglycerides increased significantly and similarly in whites and African Americans (Fig. 1). Plasma F2-isoprostanes increased significantly in both groups during the Intralipid and heparin infusion. However, the increase of plasma F2-isoprostanes was significantly greater in African Americans than in whites after 2 and 4 h of acute hyperlipidemia (Fig. 2).

Changes from baseline of plasma triglyclerides (top) and nonesterified fatty acids (NEFA, lower) in response to the infusion of Intralipid and heparin are depicted separately for African Americans (solid circles) and whites (open circles). As shown, plasma triglycerides and NEFA increased significantly and similarly in both groups. Asterisk denotes significant change (P < .05) from baseline within group.

Figure 1.

Changes from baseline in plasma F2-isoprostanes in response to acute hyperlipidemia (Intralipid and heparin) are shown separately for African Americans (solid circles) and whites (open circles). F2-isoprostanes increased significantly in both groups during the infusion, but the increase was significantly greater in African Americans than whites (#P < .05).

Figure 2.

African Americans and whites showed an increase (P < .05) in systolic and diastolic BP as well as heart rate (not shown) after 4 h of the Intralipid and heparin infusion (Fig. 3). Diastolic BP increased more (P < .05) during acute hyperlipidemia in whites than in African Americans. Large and small artery elasticity indices decreased significantly and similarly in both groups during acute hyperlipidemia. Cardiac ejection time and stroke volume decreased and systemic vascular resistance increased significantly (P < .05) in whites but not in African Americans during acute hyperlipidemia, although the differences between groups were not significant.

Changes of SBP and DBP (top) as well as large and small artery elasticity indices (SAEI, LAEI) (lower) during acute hyperlipidemia are depicted for African Americans (solid circles) and whites (open circles). As indicated, SBP and DBP increased, whereas large and small artery elasticity declined during the infusion of Intralipid and heparin. SBP = systolic blood pressure; DBP = diastolic blood pressure.

Figure 3.

Discussion

The principal new observation in this report is that a standardized acute hyperlipidemic stress increases plasma F2-isoprostanes more in African Americans than in whites. The findings confirm our previous report that acute hyperlipidemia induced by an infusion of Intralipid and heparin increases plasma F2-isoprostanes, a stable biomarker of oxidative stress.^17,21 Despite an identical infusion rate of Intralipid and heparin and a comparable increase of fatty acids and triglycerides (Fig. 1), the increase in plasma F2-isoprostanes (Fig. 2) during acute hyperlipidemia is greater in African Americans than in whites.

The mechanisms are not identified in this study by which acute hyperlipidemia induces greater increases of F2-isoprostanes in African Americans than in whites. Ethnic differences in nutritionally derived antioxidants (ie, antioxidant capacity),²⁴ could contribute. The 1987 National Health Interview Survey showed that whites eat a more varied diet, African Americans eat more fried and high-fat food, and Hispanics consume lower fat diets than the other two groups.²⁵ Preenrollment nutritional patterns in the Dietary Approaches to Stop Hypertension (DASH) study indicated that African American women had the least healthful diets, which included more atherogenic foods and fewer nutrients favorable to BP.²⁶

Ethnic differences in the distribution of genetic polymorphisms could also contribute to differences in oxidative stress. For example, the renin-angiotensin-aldosterone system has been linked to oxidative stress and target organ complications,^7,13,27 and ethnic differences in angiotensinogen genes related to cardiovascular risk and outcomes have been described.²⁸

Systolic and diastolic BP increased significantly in both groups during acute hyperlipidemia (Fig. 3). However, diastolic BP increased more in whites than in African Americans during acute hyperlipidemia. The finding that acute hyperlipidemia increased oxidative stress more in African Americans than in whites, despite relatively comparable BP changes, raise the possibility that nonoxidative stress-sensitive pathways contribute to the acute hemodynamic responses to Intralipid and heparin. Values for large and small artery elasticity declined as previously reported during Intralipid and heparin infusion. This observation suggests that acute hyperlipidemia impairs arterial distensibility, possibly through adverse effects on endothelial function as discussed previously.¹⁹

In this study, African Americans had higher baseline plasma glucose and a tendency for a higher HOMAir index, which suggest a greater level of insulin resistance compared to whites. This notion is supported by previous reports indicating a greater level of insulin resistance in African Americans than in whites.^29,30

Insulin resistance is implicated in the pathogenesis of oxidative stress and atherosclerosis.^6,7,13,31 Oxidative stress emerges as a common signal transduction mechanism by which various cardiovascular risk factors induce cardiovascular and renal remodeling and related adverse clinical sequalae.^{7,13,14,15,32} African Americans as compared to whites are more insulin resistant^29,30 and have more target organ complications such as stroke and end-stage renal disease.¹ Insulin-resistant subjects have greater excursions of plasma triglycerides and fatty acids than more insulin-sensitive subjects after standardized mixed and high-fat meals.^33,34 Collectively, these observations raise the possibility that postprandial lipid excursions and oxidative stress may be greater among African Americans who are more often insulin resistant.

The finding that African Americans have a greater oxidative stress response to hyperlipidemia may have clinical relevance that is not immediately apparent. As noted previously, African Americans tend to consume more high-fat foods than other ethnic groups^5,26,30 and would be expected to have greater postprandial increases of triglycerides, especially in the presence of insulin resistance. Thus, fasting lipid measurements may not adequately reflect lipoprotein-related cardiovascular risk among insulin-resistant subjects consuming high-fat diets. Although plasma F2-isoprostanes under basal conditions were similar in the two ethnic groups, the responses of this biomarker of oxidative stress to a standardized lipid stressor in the laboratory were markedly different (Fig. 2). We speculate that greater lipid-induced oxidative stress, particularly in the postprandial period, could contribute to the ethnic differences in cardiovascular and renal risk and complications.

Although the previous comments focused on postprandial lipid responses, it may be useful to discuss the implications of fasting lipids, which are more commonly used clinically in predicting cardiovascular risk. In this study, African Americans had lower triglycerides and very low-density lipoprotein-cholesterol and a tendency to greater HDL-cholesterol and plasma glucose levels than whites, which is consistent with previous reports.^35,36 The current findings raise the possibility that the biologic effects of lipids on oxidative stress signaling pathways linked to cardiovascular remodeling and atherosclerotic complications are greater among African Americans than whites. Consequently, the relationship of fasting lipid levels to cardiovascular risk may not be identical across ethnic groups.

In conclusion, the results indicate that acute hyperlipidemia increases F2-isoprostanes more in African Americans than in whites. The observations raise the possibility that heightened lipid-induced oxidative stress responses, especially in the postprandial period, could contribute to ethnic differences in cardiovascular and renal risk and outcomes.

We thank Ms. Kim Edwards for assistance in preparing this manuscript, Ms. Jackie Nguyen for support with the clinical laboratory procedures and data management, and the entire General Clinical Research Center staff for their expert help with multiple components of the study.