Are Supplements Safe? Effects of Gallic and Ferulic Acids on In Vitro Cell Models

Abstract

Polyphenols display health-promoting properties linked to their biological activities. They are initially absorbed in the small intestine, then they are largely metabolized in the colon, whereupon they are able to exert systemic effects. The health-promoting properties of polyphenols have led to the development of food supplements, which are also largely consumed by healthy people, even if data on their safety are still yet lacking. In the present paper, the content of gallic acid and ferulic acid was analyzed in two supplements, and shown to be higher than the relative contents found in fruit and flour. To evaluate the effects of these phenolic compounds on epithelial intestinal tissue, gallic and ferulic acids were added to a new in vitro model of the intestinal wall at different concentrations. The effects on viability, proliferation and migration of these compounds were respectively tested on three different cell lines (Caco2, L929 and U937), as well as on a tridimensional intestinal model, composed of a mucosal layer and a submucosa with fibroblasts and monocytes. Results indicated that gallic and ferulic acids can exert toxic effects on in vitro cell models at high concentrations, suggesting that an excessive and uncontrolled consumption of polyphenols may induce negative effects on the intestinal wall.

Keywords: ferulic acid; gallic acid; intestinal wall models; supplements.

Figure 1

Gallic and ferulic acid effects on L929 mouse fibroblast proliferation and viability. L929 fibroblasts were treated with gallic and ferulic acids at 2.5, 5, 10, 20 and 40 mg/L, respectively. The MTT (3-(4,5-Dimethylthiazol-2-Yl)-2,5-Diphenyltetrazolium Bromide) assay (a,b) and viability assays (c,d) were performed 24 h after treatment. For caspase 3/7 analysis, L929 cells were treated with gallic and ferulic acid at the concentration of 20 µg/mL or diluent as the control. After 24 h, cells were stained with CellEvent™ Caspase-3/7 Green Detection Reagent (e). Data are expressed as a percentage of the control. Statistical analysis was performed and, between brackets, different letters indicate mean values that are significantly different at p < 0.05.

Figure 2

Wound-healing assay using L929 with gallic and ferulic acid. (a) number of fibroblasts migrated after treatment with gallic and (b) ferulic acids at 2.5, 5, 10, 20 and 40 mg/L. (c) Significant pictures of the migrated cells. Data were expressed as percentage compared to the control. Statistical analysis was performed, and between brackets, different letters indicate mean values that are significantly different at p < 0.05.

Figure 3

Gallic and ferulic acid effects on Caco2 intestinal cell proliferation and viability. Caco2 cells were treated with gallic and ferulic acids (2.5, 5, 10, 20 and 40 mg/L), respectively. The MTT (a,b) and viability assays (c,d) were performed 24 h after treatment. For caspase 3/7 analysis, Caco2 cells were treated with gallic and ferulic acids at either 20 mg/L (e) or 40 mg/L (f) and diluent as the control. After 24 h, cells were stained with CellEvent™ Caspase-3/7 Green Detection Reagent. Data are expressed as a percentage of the control. Statistical analysis was performed, and between brackets, different letters indicate mean values that are significantly different at p < 0.05.

Figure 4

Gallic and ferulic acid effects on U937 monocytes cell proliferation and viability. U937 cells were treated with gallic and ferulic acids (2.5, 5, 10, 20 and 40 mg/L) respectively. The MTT (a,b) and viability assays (c,d) were performed 24 h after treatment. For caspase 3/7 analysis, U937 cells were treated with gallic and ferulic acids, respectively, at 20 mg/L or diluent as the control. After 24 h, cells were stained with CellEvent™ Caspase-3/7 Green Detection Reagent (e). U937 cell migration was measured as indicated in the Materials and Methods section, either with or without Caco2 cells. The mean number ± SD of migrated cells were counted and data are expressed as percentage of the control (f,g). Statistical analysis was performed, and between brackets, different letters indicate mean values that are significantly different at p < 0.05.

Figure 5

Scatter plot of the 10 investigated treatments, namely (a) gallic and ferulic acids, at doses of 2.5, 5, 10, 20 and 40 mg/L, respectively, which were defined by the first two canonical functions of linear discriminant analysis and (b) projection of variables (as reported in the material and methods section) on the first two dimensions.

Figure 6

Intestinal equivalents obtained by seeding Caco2 cells on dermal equivalents that were paraffin-embedded. (a) Sections were stained with hematoxylin and eosin (H&E), thiazine and occludin markers; Fast Red was used as a chromogen. (b) The thickness of the epithelial layer was measured by image pixel count using ImageJ 2.0.0-rc-69/1.521 (c) Stained areas were evaluated by image pixel count using ImageJ 2.0.0-rc-69/1.521. Experiments were conducted in triplicate from different samples. Statistical analysis was performed, and between brackets, different letters indicate mean values that are significantly different at p < 0.05.