Background: In a recent pilot study, the intake of elderberry juice resulted in a significant decrease in serum cholesterol concentrations and an increase in low-density lipoprotein (LDL) stability. This study was designed to verify the preliminary results.
Objective: We investigated the impact of elderberry juice on cholesterol and triglyceride concentrations as well as antioxidant status in a cohort of young volunteers.
Design: Study A: The randomized, placebo-controlled trial for studying the effect of anthocyanes on lipid and antioxidant status, 34 subjects took capsules with 400 mg spray-dried powder containing 10% anthocyanes t.i.d. equivalent to 5 ml elderberry juice for 2 weeks. A subgroup of 14 subjects continued for an additional week to test for resistance to oxidation of LDL. Study B: To investigate the short-term effects on serum lipid concentrations, six subjects took a single dose of 50 ml of elderberry juice (equivalent to 10 capsules) along with a high-fat breakfast.
Results: In the placebo-controlled study, there was only a small, statistically not significant change in cholesterol concentrations in the elderberry group (from 199 to 190 mg/dl) compared to the placebo group (from 192 to 196 mg/dl). The resistance to copper-induced oxidation of LDL did not change within 3 weeks. In the single-dose experiment increases in postprandial triglyceride concentrations were not significantly different when the six subjects were investigated with and without elderberry juice.
Conclusions: Elderberry spray-dried extract at a low dose exerts a minor effect on serum lipids and antioxidative capacity. Higher, but nutritionally relevant doses might significantly reduce postprandial serum lipids.
Sponsorship: This study was supported by the Austrian Industrial Research Fund (FFF, Project Nr. 802230) and the Steirische Beerenfrost GmbH.
Atherogenesis is, among multiple risk factors, causally related to elevated serum lipid levels (Castelli et al, 1986). Additionally, a relation to the oxidation of low-density lipoprotein (LDL) in the arterial wall is assumed (Halliwell, 1993; Esterbauer et al, 1997). Such a risk estimation might be further influenced by the presence of postprandial hyperlipidemia (Ebenbichler et al, 1995). Dietary products with a lipid-lowering and/or antioxidative capacity may be important in the reduction of risk of coronary heart disease (Hollman et al, 1996; Piironen et al, 2000). Flavonoids, which have been shown to exert antioxidant actions, have been extensively studied. In the Zutphen elderly study, the intake of flavonoids was inversely associated with mortality from coronary heart disease (Hertog et al, 1993) and in the ‘seven countries study’, a cross-cultural correlation study, mortality from coronary heart disease was inversely associated with the average intake of flavonoids (Hertog et al, 1995). The recognition of the ‘French paradox’ led to the suggestion that some components of red wine protect against coronary heart disease and again focused attention on dietary anthocyanes (Renaud & Lorgeril, 1992). Previous studies suggested a beneficial effect of anthocyanes on lipid peroxidation (Satué-Garcia et al, 1997). The in vitro experiments of Abuja et al (1998) showed that anthocyanes of elderberry have a high antioxidative potential and that the action of the antioxidant substances from elderberry juice is similar to that of vitamin C. Additionally, a pilot study with six subjects showed a cholesterol-lowering effect of a spray-dried elderberry and an increase in the resistance to oxidation of LDL (Murkovic et al, unpublished data).
The aim of the present study was to investigate the effects of anthocyanes on serum lipid concentrations and serum and LDL-antioxidative properties as determinants of the risk of developing cardiovascular disease. The effects on serum lipids, antioxidants, and LDL resistance to oxidation were tested in a randomized, double-blind placebo-controlled 3 weeks trial and the effects of a single oral loading dose on postprandial lipids were tested in a high-fat meal tolerance test.
Subjects and methods
Randomized placebo-controlled trial
A pilot study on six subjects, which had been performed prior to this study showed a reduction of plasma cholesterol concentrations after 2 weeks and an increase in the resistance to oxidation of LDL after 3 weeks. Data from that pilot study were used for power calculation (n=34) and choice of time points (2 weeks for cholesterol; 3 weeks LDL resistance to oxidation) of the present study. During the 2-week study period, 34 volunteers took a continental breakfast (coffee, rolls, butter, jam) at home. For lunch and dinner, the subjects received a diet standardized for total energy content and percent distribution of carbohydrates, fat, and protein without further restrictions or recommendations, prepared in a local restaurant, containing typical local food to reach a final caloric intake of 33 kcal/kg bodyweight/day (45% carbohydrate, 35% fat, 20% protein).
At baseline, subjects were randomized to receive 400 mg t.i.d of either spray-dried elderberry juice (containing 10% anthocyanes) in gelatinous capsules (Iprona, Lana, Italy) or placebo. In addition to 100 mg anthocyanes, the spray-dried elderberry powder contained per g: 51 mg glucose, 45 mg fructose, 57 mg citric acid, 6.2 mg potassium, 1.6 mg calcium, 1.7 mg phosphate, 0.9 mg magnesium, amino acids, and maltodextrin as the matrix. Distribution of anthocyanes was 42% cyanidine-3-glucoside, 42% cyanide-3-sambubioside, 12% cyanidine-3-sambubioside-5-glucoside, 4% cyanidine-3,5-diglucoside.
Compliance was checked by pill-count. The subjects were asked to report any side effects of the treatment. Prior to the study and both after 2 weeks and 3 weeks plasma samples were obtained for determination of antioxidants, lipids, and clinical routine parameters.
A subgroup of unselected 14 subjects (seven males, seven females) volunteered to continue the study for an additional week. From these subjects, additional plasma (50 ml) was obtained for LDL preparation and further analysis after 3 weeks after an overnight fast between 8.00 and 9.00.
Short-term effects of a single oral loading dose of anthocyanes in the postprandial phase
After. a 4-week wash-out period, a subgroup of six subjects volunteered for two meal tolerance tests (97 g fat, 56 g carbohydrate, 34 g protein) on consecutive days, randomly assigned to either with or without a single dose of the spray-dried elderberry juice (4000 mg, equivalent to 50 ml of elderberry juice) ingested simultaneously with the meal in a crossover design. It has been shown previously that after uptake of this amount the anthocyanes became detectable in plasma (Murkovic et al, 2000b). Fasting blood samples were drawn prior to the test meal after an overnight fast between 8.00 and 9.00 and after 1, 2, 3, 4, and 6 h.
In all, 34 healthy volunteers (20 males, 14 females) not regularly taking antioxidant supplements from the Erzherzog-Johann University of Graz, Austria, took part in a short-term nutrition intervention study. The protocol was approved by the Ethics Committee of the Karl-Franzens-University Graz, Austria and informed consent was obtained prior to the study from all the subjects. Anthropometric data (BMI, waist and hip measurements, waist-to-hip ratios) were determined by routine methods, and data on smoking habits and alcohol consumption (units per day) were obtained (Table 1).
HPLC analysis of anthocyanes in serum
Serum was injected directly (500 μl) onto an HPLC system using an on-line clean-up and enrichment precolumn with a column switching device. The analytical method is described in detail in Murkovic et al (2000a,2000b).
Serum for routine laboratory analyses and serum lipid determination (total cholesterol, triglycerides, and HDL-cholesterol) was obtained as previously described (Toplak et al, 1994). Cholesterol concentrations were determined using the test kit from Roche Diagnostics (Mannheim, Germany), LDL-cholesterol was calculated using the Friedewald formula.
Plasma vitamin C, E, and carotenoid concentrations were determined by HPLC as described by Esteve et al (1997) and Vuilleumier et al (1983). EDTA-plasma samples for the LDL resistance to oxidation measured by the induction (lag) time of copper-induced LDL oxidation were obtained by immediate centrifugation of the EDTA-blood at 4°C as previously described (Abuja et al, 1998).
LDL resistance to oxidation
LDL was isolated from plasma by single-step density gradient ultracentrifugation in a Beckman NVT65 rotor. Prior to use, LDL was stored in a sterile, evacuated glass vial (TechneVial, Mallinckrodt) at 4°C in the dark for up to 1 week. To remove EDTA, LDL was gel filtered on an EconoPac 10DG column (Bio-Rad, Richmond, CA, USA) equilibrated with PBS, for copper-mediated oxidation. For oxidation, 0.1 mM LDL was incubated at 37°C in PBS. To start oxidation, either 1.6 μM CuSO4 or 1 mM AAPH was added. Oxidation was monitored by measuring the absorption of the conjugated dienes at 234 nm at intervals of 4 min in a Beckman DU-640 spectrophotometer equipped with a Peltier-thermostated six-cell holder. Duration of the lag time, tlag was determined as the time coordinate of the intersection of tangents to the increase in CD during the lag and propagation phases. Formation of CD was calculated from their absorbance at 234 nm, using the molar absorption coefficient ɛ234=29 500 M/cm. Data were evaluated using MS Excel 5.0 and Stat Graphics 2.1 for Windows. The method has been described in detail by Ramos et al (1995) and Bergmann et al (1997).
Study A: Differences between baseline and 2 (antioxidants, lipids) and 3 weeks (LDL resistance to oxidation) were analyzed by paired t-test. Study B: The paired Wilcoxon test was used for analyzing differences in serum cholesterol, Apo-B, and serum triglycerides between the two meal tolerance tests with and without anthocyanes, and Bonferroni correction was applied in order to take care of multiple testing. Data are expressed as mean ±s.d. P<0.05 was considered as statistically significant. For statistical analysis, SigmaStat vers. 2.03 (SPSS Science Software GmbH, Erkrath, Germany) was used.
The clinical biochemistry data including liver and kidney function, blood glucose, and serum lipase concentrations of the study subjects were well within the normal range. There were no differences between the treatment and placebo groups. No changes occurred during the treatment period.
No side effects of anthocyane supplementation, neither in the randomized, placebo-controlled trial nor after the single oral loading dose, were reported by the volunteers. Routine clinical laboratory profiles remained unchanged both in the verum and the placebo group.
Randomized, placebo-controlled trial
All 34 subjects completed the 14-day period (17 in each group). The two groups did not differ in terms of anthropometric data, smoking habits, and alcohol consumption (Table 1). Most of the lipid values were in the normal range, as expected in healthy volunteers. The effects of Sambucus nigra (vs placebo) on serum total cholesterol, triglycerides, and HDL-cholesterol as well as LDL-cholesterol are shown in Table 2. There were no significant differences in the changes in serum lipids when the elderberry group was compared with the placebo group (Mann–Whitney rank sum test). There was no effect of S nigra on vitamins C, E as well as carotenoids (Table 3).
At baseline, plasma vitamin C concentrations of the verum group (24.0±5.0 μg/ml) and the placebo group (22.2±6.5 μg/ml) were comparable at baseline. During the 2-week study period, vitamin C concentrations dropped significantly both in the placebo and the elderberry group (P<0.01). In addition, no differences in vitamin A and E and β-carotene concentrations were observed between the two groups (Table 3).
Resistance to oxidation of LDL
In all, 14 subjects (seven of either group) continued the study for an additional week, that is, for a total of 3 weeks. Using the absorption at 234 nm for measuring the conjugated diene formation, the resistance to oxidation of LDL was determined by the lag time of copper-induced LDL oxidation. No significant differences between the verum (baseline, 56±1 vs 3 weeks, 55±11 min) and the placebo group (58±13 vs 56±9 min) were observed.
Short-term effects of a single oral loading dose of anthocyanes in the postprandial phase
Serum anthocyane concentrations
Anthocyanes could not be detected in the serum prior to the uptake of the spray-dried elderberry juice (Figure 1). During the postprandial phase, an increase in serum total anthocyane levels was observed. The mean ±s.d. of the concentration time profile of serum total anthocyanes is shown in Figure 2. In three subjects, peak serum concentrations of the anthocyanes were achieved after 3 h and in the other three subjects after 4 h. The concentrations reached 2.1±1.4 and 1.0±0.4 ng/ml, respectively. In two subjects, the peak concentrations of the anthocyanes were more than three times the peak concentration of the others. The biological half-life of anthocyanes seemed to exceed 6 h, given that more than 50% of the increase was still present after the 6 h study period.
In a subgroup of six subjects, a meal tolerance test was performed twice, one in the absence and the other one in the presence of 4 g of a single dose of spray- dried elderberry juice (equivalent to 50 ml of fresh juice). Serum cholesterol and triglyceride concentrations were measured in the postprandial phase. Figure 3 shows that in the placebo group, serum triglyceride concentrations increased postprandially with a maximum achieved at 3 h after the meal, while the increase in the elderberry group appeared to be smaller. However, paired Wilcoxon tests performed for the different time points showed a difference only at 4 h close to significance (P=0.05), which disappeared after Bonferroni correction for multiple testing.
Anthocyanes are known to have antioxidant properties (Rice-Evans, 1999). It is well known that elderberries are a rich source of anthocyanes. In vitro studies of elderberry (S. nigra) demonstrated a significant antioxidative effect as determined by LDL oxidation (Abuja et al, 1998; Heinonen et al, 1998). A similarity with the action of vitamin C, a water-soluble antioxidant, was suggested. Antioxidative effects of vitamins have been demonstrated not only in vitro, but also after oral supplementation (Dieber-Rotheneder et al, 1991). Serum lipid concentrations and ex vivo resistance against oxidation of LDL using a well-established model system (Esterbauer et al, 1997) are thought to correlate with cardiovascular risk. As a consequence, reduction of serum lipid levels and enhancement of resistance to oxidation of LDL would contribute to reduction of cardiovascular disease. The present study showed a small but statistically not significant reducing effect of 400 mg spray-dried elderberry extract which is equivalent to 5 ml of fresh elderberry juice) on total cholesterol (minus 7%), LDL-cholesterol (minus 3%) and HDL-cholesterol concentrations (minus 7%). Fasting serum triglyceride concentrations remained unchanged. While there was a significant drop in plasma vitamin C concentrations in both the verum and the placebo group that can be explained by the study diet, there was no effect of the elderberry extract on the antioxidant status and resistance to oxidation of LDL.
Acute postprandial effects
Triglyceride intolerance has been shown to be an independent risk factor for coronary artery disease (Ebenbichler et al, 1995). Therefore, improvement of the vascular prognosis might be achieved in particular through the reduction of serum lipids in the postprandial phase. In the present study the short-term effect of a higher, but nutritionally relevant dose (4 g spray-dried elderberry extract which is equivalent to 50 ml of fresh elderberry juice) on serum concentrations after a high-fat meal was studied. There was only a trend of a smaller increase in postprandial triglyceride concentrations after the six volunteers had taken the elderberry juice, which was not statistically significant. This could be because of the small sample size of this open study. A randomized controlled and larger study is needed to find out whether elderberry juice can lower postprandial lipid levels.
The results of this study showing trends but not significant effects of elderberry juice on LDL resistance to oxidation can be explained by the fact that relatively low serum concentrations are achieved with the doses used in this study (low nanomolar range), and that anthocyanes are water-soluble compounds that are not present in the isolated LDL particle used for ex vivo oxidation experiments. In contrast to low serum concentrations, the concentrations of anthocyanes present in the digestive tract are expected to be high, which could interfere with the intestinal absorption of lipids. In addition, anthocyanes might exert antioxidant properties within the intestinal tract (Natella et al, 2002, Ursini & Sevanian, 2002).
Therefore, long-term experiments using such high doses have to be evaluated for efficacy and safety. The absence of apparent signs of toxicity in our subjects suggests that elderberry extract can be safely used for these studies.
Future studies also should comprise a larger group of subjects and include additional markers of lipid and protein oxidation.
Elderberry spray-dried extract at a low dose exerts a minor effect on serum lipids and antioxidative capacity in the absence of adverse effects. Higher, but nutritionally relevant doses might significantly reduce postprandial serum lipids. Further studies seem to be worthwhile.
We thank Andrea Hillebrand for excellent technical assistance.