Improved Preventive Effects of Combined Bioactive Compounds Present in Different Blueberry Varieties as Compared to Single Phytochemicals

Abstract

Blueberries contain many different phytochemicals which might be responsible for their disease preventive properties. In a previously conducted human dietary intervention study, we showed that a 4-week intervention with blueberry⁻apple juice protected the participants against oxidative stress and modulated expression of genes involved in different genetic pathways contributing to the antioxidant response. The present study investigates the effect of different blueberry varieties (Elliot, Draper, Bluecrop, and Aurora, and the blueberry⁻apple juice from our previous human dietary intervention study), and four different single compounds (vitamin C, peonidin, cyanidin, and quercetin) on antioxidant capacity and gene expression changes in colonic cells in vitro, and compares the outcome with the earlier in vivo findings. The results demonstrate that all blueberry varieties as well as the blueberry⁻apple juice were more effective in reducing oxidative stress as compared to the single compounds (e.g., DNA strand break reduction: EC₅₀: Elliot 8.3 mg/mL, Aurora and Draper 11.9 mg/mL, blueberry⁻apple juice 12.3 mg/mL, and Bluecrop 12.7 mg/mL; single compounds). In addition, the gene expression profiles (consisting of 18 selected genes from the in vivo study) induced by the blueberry varieties were more similar to the profile of the human intervention study (range 44⁻78%). The blueberry variety Elliot showed the strongest and most similar effects, almost 80% of gene expression modulations were similar compared to the in vivo results. From the single compounds (range 17⁻44%), quercetin induced the most comparable gene expression changes, i.e., 44%. This approach could be useful in agriculture for identifying crop varieties containing combinations of phytochemicals which show optimal preventive capacities.

Keywords: (combinations of) phytochemicals; antioxidant capacity; blueberry extracts; disease risk reduction; gene expression profiles.

Figure 1

Level of DNA damage in Caco-2 cells as measured by the alkaline comet assay, expressed as average median tail moments. Error bars indicate standard deviations. Caco-2 cells were pre-incubated for 2 h with different concentrations of the extract of blueberry–apple juice or extracts of four different blueberry varieties (a), or single compounds (b) and subsequently exposed to 100 µM tert-butylhydroperoxide (TBH) for 1 h. Next, DNA strand breaks were measured using the alkaline comet assay. Pre-incubation for 2 h with medium, solvent control (0.5% end concentration of 70% methanol/0.1% formic acid), the highest concentration of the different extracts (i.e., 7 mg/mL), or 100 µM of single compounds did not induce any DNA damage. A dose-dependent decrease in DNA damage was observed for all blueberry extracts and single compounds tested (ANOVA, P < 0.01). In order to investigate which of the blueberry extracts and single compounds possessed the highest chemopreventive properties, linear log regression was applied. From the log linear regression equation, the EC₅₀ was estimated which is shown in the legend.

Figure 2

Radical formation in Caco-2 cells as measured by ESR spectroscopy. Results are expressed as percentage of solvent control levels. AUC: area under the curve of radical specific signals. Error bars indicate standard deviations. Caco-2 cells were pre-incubated for 2 h with different concentrations of the extract of blueberry–apple juice or extracts of four different blueberry varieties (a), or single compounds (b) and subsequently exposed to 150 µM tert-butylhydroperoxide (TBH) for 30 min. Pre-incubation for 2 h with medium, solvent control (0.5% end concentration of 70% methanol/0.1% formic acid), the maximal concentration of the different extracts (i.e., 7 mg/mL), or 100 µM of single compounds did not induce significant levels of radical formation. ** P < 0.01; * P < 0.05, significantly different from Caco-2 cells exposed to solvent control for 2 h and challenged with 150 µM TBH for 30 min.

Figure 2

Radical formation in Caco-2 cells as measured by ESR spectroscopy. Results are expressed as percentage of solvent control levels. AUC: area under the curve of radical specific signals. Error bars indicate standard deviations. Caco-2 cells were pre-incubated for 2 h with different concentrations of the extract of blueberry–apple juice or extracts of four different blueberry varieties (a), or single compounds (b) and subsequently exposed to 150 µM tert-butylhydroperoxide (TBH) for 30 min. Pre-incubation for 2 h with medium, solvent control (0.5% end concentration of 70% methanol/0.1% formic acid), the maximal concentration of the different extracts (i.e., 7 mg/mL), or 100 µM of single compounds did not induce significant levels of radical formation. ** P < 0.01; * P < 0.05, significantly different from Caco-2 cells exposed to solvent control for 2 h and challenged with 150 µM TBH for 30 min.

Figure 3

Hierarchical clustering dendrogram of different blueberry extracts (Elliot, Draper, Bluecrop, and Aurora), the extract of the blueberry–apple juice (BBAJ) from the human dietary intervention, and of four single compounds (quercetin, vitamin C, peondin, and cyanidin) based on the effect on the expression of 18 genes selected from a previously conducted human dietary interventions study. In order to compare these gene expression profiles with the profile generated in the human dietary intervention study, the gene expression profile of the human dietary intervention study is included as well (HDIS). Hierarchical clustering analysis was carried out using http://heatmapper.ca. As clustering method complete linkage was used, and distance between the clusters was computed based on Spearman rank correlation. The different conditions are displayed on the y-axis and the genes are shown on the x-axis.