Ethanolic Echinacea purpurea Extracts Contain a Mixture of Cytokine-Suppressive and Cytokine-Inducing Compounds, Including Some That Originate from Endophytic Bacteria

Abstract

Echinacea preparations, which are used for the prevention and treatment of upper respiratory infections, account for 10% of the dietary supplement market in the U.S., with sales totaling more than $100 million annually. In an attempt to shed light on Echinacea's mechanism of action, we evaluated the effects of a 75% ethanolic root extract of Echinacea purpurea, prepared in accord with industry methods, on cytokine and chemokine production from RAW 264.7 macrophage-like cells. We found that the extract displayed dual activities; the extract could itself stimulate production of the cytokine TNF-α, and also suppress production of TNF-α in response to stimulation with exogenous LPS. Liquid:liquid partitioning followed by normal-phase flash chromatography resulted in separation of the stimulatory and inhibitory activities into different fractions, confirming the complex nature of this extract. We also studied the role of alkylamides in the suppressive activity of this E. purpurea extract. Our fractionation method concentrated the alkylamides into a single fraction, which suppressed production of TNF-α, CCL3, and CCL5; however fractions that did not contain detectable alkylamides also displayed similar suppressive effects. Alkylamides, therefore, likely contribute to the suppressive activity of the extract but are not solely responsible for that activity. From the fractions without detectable alkylamides, we purified xanthienopyran, a compound not previously known to be a constituent of the Echinacea genus. Xanthienopyran suppressed production of TNF-α suggesting that it may contribute to the suppressive activity of the crude ethanolic extract. Finally, we show that ethanolic extracts prepared from E. purpurea plants grown under sterile conditions and from sterilized seeds, do not contain LPS and do not stimulate macrophage production of TNF-α, supporting the hypothesis that the macrophage-stimulating activity in E. purpurea extracts can originate from endophytic bacteria. Together, our findings indicate that ethanolic E. purpurea extracts contain multiple constituents that differentially regulate cytokine production by macrophages.

Fig 1. Partitioning scheme for the Echinacea purpurea extract.

Dried E. purpurea roots (1.9 kg) were extracted in 75:25 ethanol:water, consistent with standard practices in the dietary supplements industry. The ethanolic extract was evaporated by rotary evaporation and subjected to two stages of liquid:liquid partitioning with hexane:methanol and chloroform. The resulting chloroform layer was fractioned into thirteen pooled fractions using normal phase flash chromatography. These fractions were then profiled for chemical composition and subjected to biological assays. Marc, residue after extraction; EE, ethanol extract; HL, hexane layer, ML, methane layer; AL, aqueous layer; CL, chloroform layer.

Fig 2. Influence of 75% ethanol extract and liquid:liquid partitions from Echinacea purpurea extract on TNF-α production by RAW 264.7 cells.

Extract and partitions were tested at concentrations of 50 μg/mL (A and C) and 100 μg/mL (B and D) expressed as mass of extract per assay well volume. Cells were unstimulated in A and B and stimulated with 100 ng/mL LPS in C and D. Supernatants were harvested after 16–18 hr. and levels of TNF-α measured by ELISA. M, media; L, LPS; EE, ethanol extract; HL, hexane layer, ML, methane layer; WL, water layer; CL, chloroform layer. Values shown are means +/- SD from a single representative experiment. Statistical analysis was performed using the Student’s T test, *p<0.05, **p<0.01, ***p<0.001.

Fig 3. Alkylamides from Echinacea purpurea.

Chromatogram (A) of the crude chloroform layer (CL) from the E. purpurea extract showing the presence of a series of alkylamides (3,4,5,8,9,10,11A,11B,14,15) which were identified according to molecular weight and retention time. The relative amounts of these alkylamides in the silica gel column pooled fractions (1–13) are indicated by the peak areas shown in panel B. The numbering system used for these alkylamides is consistent with that used in a previous report [18]. * indicates compounds detected, but not identified. The structures for seven alkylamides selected to represent the alkylamide content are shown in panel C.

Fig 4. The effects of Echinacea purpurea extract fractions on production of TNF-α.

Fractions from the flash chromatography separation of the E. purpurea extract chloroform layer (CL) were tested for their effects on the production of TNF-α from RAW 264.7 cells in the absence (A) or presence (B) of 10 ng/mL LPS. Treatments were for 16–18 h and levels of TNF-α in supernatants were quantified by ELISA. Each fraction was tested at 50 and 100 μg/mL. All data shown are means +/- SEM from three independent experiments. Statistical analysis was performed using the Student’s T test, *p<0.05, **p<0.01. M, media; L, LPS; CL, chloroform layer.

Fig 5. Cytotoxicity of Echinacea purpurea extract fractions and chloroform layer.

Fractions from the silica gel column and the chloroform layer were tested for cytotoxicity towards RAW 264.7 cells. Treatments were for 24 h and cytotoxicity was determined using the LDH release assay. Each fraction was tested at 50 and 100 μg/mL. Data shown are means +/- SEM from three independent experiments. Statistical analysis was performed using the Student’s T test, *p<0.05, **p<0.01, ***p<.001. M, media; CL, chloroform layer; Ly, lysis buffer.

Fig 6. Structure and activity of xanthienopyran.

The structure of xanthienopyran was elucidated by comparison of NMR and MS data with published values (A). The influence of xanthienopyran on TNF-α secretion by RAW 264.7 macrophage-like cells alone (B) or in the presence of 10 ng/mL of LPS (C) was evaluated. The influence of dodeca-2E,4E-dienoic acid isobutylamide (15) on TNF-α secretion by RAW 264.7 macrophage-like cells in the presence of 10 ng/mL of LPS (C) was evaluated as comparison. Treatments were for 16–18 hr. and levels of TNF-α in supernatants were quantified by ELISA. Data shown are means +/- SEM from three independent experiments. Statistical analysis was performed using the Student’s T test, *p<0.05.

Fig 7. Echinacea purpurea fraction effects on CCL3 and CCL5.

Fractions from the silica gel column were tested for their effects on the production of CCL3 (MIP1-α) and CCL5 (RANTES) from RAW 264.7 cells in the absence (A and B) or presence (C and D) of 10 ng/mL LPS. Treatments were for 16–18 hr. and chemokine levels in supernatants were quantified by ELISA. Each fraction was tested at 50 μg/mL. Data shown are means +/- SEM or SD from three (CCL3) and two (RANTES) independent experiments. Statistical analysis was performed using the Student’s T test, *p<0.05. M, media; CL, chloroform layer.

Fig 8. Echinacea purpurea plants grown from sterilized seeds.

E. purpurea seeds were sterilized following removal of the epidermis and grown under sterile conditions for six weeks. The negative control was prepared using the same reaction vessels and solvents as for the individual plants, but without plant material. The seed used to grow plant 49 was subjected to surface sterilization but with epidermis intact (see S1 Table). Plants were harvested and extracts tested for LPS content (A) and ability to induce production of TNF-α from RAW 264.7 cells after 16–18 h incubation (B). Data show the means ± SD of duplicate experiments measuring TNF-α secretion in panel A, and the mean ± SEM of triplicate measurements of LPS content in panel B. Statistical analysis was performed using the Student’s T test, ***p < 0.001.