Assessment of cytotoxicity exerted by leaf extracts from plants of the genus Rhododendron towards epidermal keratinocytes and intestine epithelial cells

Abstract

Background: Rhododendron leaf extracts were previously found to exert antimicrobial activities against a range of Gram-positive bacteria. In this study, we investigated which of the extracts with these antimicrobial properties would be best suited for further exploitation. Specifically, the project aims to identify biologically active compounds that affect bacterial but not mammalian cells when applied in medical treatments such as lotions for ectopic application onto skin, or as orally administered drugs.

Methods: Different concentrations of DMSO-dissolved remnants of crude methanol Rhododendron leaf extracts were incubated for 24 h with cultured epidermal keratinocytes (human HaCaT cell line) and epithelial cells of the intestinal mucosa (rat IEC6 cell line) and tested for their cytotoxic potential. In particular, the cytotoxic potencies of the compounds contained in antimicrobial Rhododendron leaf extracts were assessed by quantifying their effects on (i) plasma membrane integrity, (ii) cell viability and proliferation rates, (iii) cellular metabolism, (iv) cytoskeletal architecture, and (v) determining initiation of cell death pathways by morphological and biochemical means.

Results: Extracts of almost all Rhododendron species, when applied at 500 μg/mL, were potent in negatively affecting both keratinocytes and intestine epithelial cells, except material from R. hippophaeoides var. hippophaeoides. Extracts of R. minus and R. racemosum were non-toxic towards both mammalian cell types when used at 50 μg/mL, which was equivalent to their minimal inhibitory concentration against bacteria. At this concentration, leaf extracts from three other highly potent antimicrobial Rhododendron species proved non-cytotoxic against one or the other mammalian cell type: Extracts of R. ferrugineum were non-toxic towards IEC6 cells, and extracts of R. rubiginosum as well as R. concinnum did not affect HaCaT cells. In general, keratinocytes proved more resistant than intestine epithelial cells against the treatment with compounds contained in Rhododendron leaf extracts.

Conclusions: We conclude that leaf extracts from highly potent antimicrobial R. minus and R. racemosum are safe to use at 50 μg/mL in 24-h incubations with HaCaT keratinocytes and IEC6 intestine epithelial cells in monolayer cultures. Extracts from R. rubiginosum as well as R. concinnum or R. ferrugineum are applicable to either keratinocytes or intestinal epithelial cells, respectively. Beyond the scope of the current study, further experiments are required to identify the specific compounds contained in those Rhododendron leaf extracts that exert antimicrobial activity while being non-cytotoxic when applied onto human skin or gastrointestinal tract mucosa. Thus, this study supports the notion that detailed phytochemical profiling and compound identification is needed for characterization of the leaf extracts from specific Rhododendron species in order to exploit their components as supplementary agents in antimicrobial phyto-medical treatments.

Fig. 1

Effects of Rhododendron leaf extracts on IEC6 (a) and HaCaT (b) cells incubated with three different concentrations (5, 50, and 500 μg/mL) of leaf extracts as indicated. Cell viability and proliferation was analyzed by the MTT assay upon incubation at 37 °C for 24 h. The percentage of MTT reduction for each extract concentration was normalized to that of the 0.5 % DMSO solvent. Values are given as mean ± standard deviations from three independent experiments, each performed in triplicates. Statistical evaluation was performed by one way ANOVA-analysis; levels of significance are indicated as *for p < 0.05

Fig. 2

Effects of Rhododendron leaf extracts on the cell numbers of IEC6 (a) and HaCaT (b) cells after 24 h incubation at 37 °C with three different concentrations (5, 50, and 500 μg/mL) of leaf extracts as indicated. The total number of cells as determined by Draq5™ staining reflects the effects of leaf extracts on cell viability and adhesion since only monolayer-associated cells were stained and counted in this assay. Values are given as mean ± standard deviations from three independent experiments, each performed in triplicates. Statistical evaluation was performed by one way ANOVA-analysis; levels of significance are indicated as *for p < 0.05

Fig. 3

Effects of Rhododendron leaf extracts on the mitochondrial membrane potential of IEC6 (a) and HaCaT (b) after 24 h incubation at 37 °C with three different concentrations (5, 50, and 500 μg/mL) of leaf extract as indicated. The intensity of MitoTracker® Red CMXRos signal reflects the accumulation of the dye within the mitochondrial matrix, which depends on an intact inner mitochondrial membrane potential, and thus on the metabolic activity of the cells. Values are given as mean ± standard deviations from three independent experiments, each performed in triplicates. Statistical evaluation was performed by one way ANOVA-analysis; levels of significance are indicated as *for p < 0.05

Fig. 4

Morphological changes of mitochondria in IEC6 and HaCaT cells after 24 h exposure to three different concentrations (5, 50 and 500 μg/mL) of Rhododendron leaf extracts. Confocal fluorescence images of IEC6 (a) and HaCaT cells (b) labeled with MitoTracker® Red CMXRos. Cells treated with 0.5 % DMSO were used as controls. Also depicted are cells incubated with extracts from R. hippophaeoides var. hippophaeoides (A), R. xanthostephanum (B), R. hirsutum (C) and R. racemosum (D). Bars represent 20 μm

Fig. 5

Effects of Rhododendron leaf extracts on the F-actin cytoskeleton of IEC6 and HaCaT cells upon incubation with three different concentrations (5, 50, and 500 μg/mL) of leaf extract for 24 h. The intensity of the phalloidin signal in IEC6 (a) and HaCaT cells (b) reflects F-actin presence, which maintains the cellular architecture. Values are given as mean ± standard deviations from three independent experiments, each performed in triplicates. Statistical evaluation was performed by one way ANOVA-analysis; levels of significance are indicated as *for p < 0.05

Fig. 6

Morphological changes of F-actin structures in IEC6 and HaCaT cells after 24 h exposure to three different concentrations (5, 50 and 500 μg/mL) of Rhododendron leaf extracts. Confocal fluorescence images of IEC6 (a) and HaCaT cells (b) labeled with phalloidin (green) and Draq5™ (blue). Cells treated with 0.5 % DMSO served as controls for cells treated with leaf extracts of R. hippophaeoides var. hippophaeoides (A), R. xanthostephanum (B), R. hirsutum (C) and R. racemosum (D). Bars represent 20 μm

Fig. 7

Plasma membrane integrity and cell death by apoptosis as induced by a 24 h-exposure of IEC6 cells and HaCaT keratinocytes to 500 μg/mL of specific Rhododendron leaf extracts. Merged micrographs taken with a confocal laser scanning microscope depict IEC6 (a; panels A-E) and HaCaT cells (b; panels F-J). Violet signals in merged images are due to overlapping red, PI-derived signals, in cells with ruptured plasma membranes, and blue, Draq5™ staining of nuclei in all cells. Pictures A and F are control cells treated with 0.5 % DMSO, while cells in all other panels were incubated with extracts from R. cinnabarinum (B and G), R. ferrugineum (C and H), R. minus (D and I) and R. hippophaeoides var. hippophaeoides (E and J). Bars represent 50 μm

Fig. 8

Detection of caspase-3 activity in IEC6 cells. Cells were treated with 500 μg/mL of leaf extracts from R. cinnabarinum (a) and R. ferrugineum (b) for the indicated time intervals. Reactions were carried out at room temperature and fluorescence was measured in a fluorescence microplate reader using 496 nm for excitation and emission was detected at 520 nm. Non-treated cells and cells treated with DMSO (0.5 %) were used as negative controls, while staurosporine (10 μg/mL) treatment was used as a positive control (apoptosis inducer). Values are given as mean ± standard deviations from three independent experiments, each performed in triplicates. Statistical evaluation was performed by one way ANOVA-analysis; levels of significance are indicated as *for p < 0.05

Fig. 9

Profile map summarizing the results of cell biological assessment assays, combining the effects of Rhododendron crude extracts against B. subtilis depicted by minimum inhibitory concentration. The twelve Rhododendron species were classified into four groups (i.e. 50, 100, 150, and 300 μg/mL) according to the MIC results (black vertical line). Three concentrations (5, 50, and 500 μg/mL) of Rhododendron crude extracts were applied to the two different cell lines for 24 h representing low, medium, and high antimicrobial activity, respectively. The grey shading represents the toxicity that Rhododendron crude extracts exerted on mammalian cells, i.e. non-toxic extracts are depicted in light grey, and cytotoxic extracts are shown by dark grey boxes. Panel (a) represents the results of IEC6 cells for the assays on cell viability and proliferation rates (A), plasma membrane integrity (B), cellular architecture (C), total cell numbers (D), and cellular metabolism (E). Panel (b) represents the corresponding results for HaCaT cells