Screening of herbal extracts for TLR2- and TLR4-dependent anti-inflammatory effects

Herbal extracts represent an ample source of natural compounds, with potential to be used in improving human health. There is a growing interest in using natural extracts as possible new treatment strategies for inflammatory diseases. We therefore aimed at identifying herbal extracts that affect inflammatory signaling pathways through toll-like receptors (TLRs), TLR2 and TLR4. Ninety-nine ethanolic extracts were screened in THP-1 monocytes and HeLa-TLR4 transfected reporter cells for their effects on stimulated TLR2 and TLR4 signaling pathways. The 28 identified anti-inflammatory extracts were tested in comparative assays of stimulated HEK-TLR2 and HEK-TLR4 transfected reporter cells to differentiate between direct TLR4 antagonistic effects and interference with downstream signaling cascades. Furthermore, the ten most effective anti-inflammatory extracts were tested on their ability to inhibit nuclear factor-κB (NF-κB) translocation in HeLa-TLR4 transfected reporter cell lines and for their ability to repolarize M1-type macrophages. Ethanolic extracts which showed the highest anti-inflammatory potential, up to a complete inhibition of pro-inflammatory cytokine production were Castanea sativa leaves, Cinchona pubescens bark, Cinnamomum verum bark, Salix alba bark, Rheum palmatum root, Alchemilla vulgaris plant, Humulus lupulus cones, Vaccinium myrtillus berries, Curcuma longa root and Arctostaphylos uva-ursi leaves. Moreover, all tested extracts mitigated not only TLR4, but also TLR2 signaling pathways. Seven of them additionally inhibited translocation of NF-κB into the nucleus. Two of the extracts showed impact on repolarization of pro-inflammatory M1-type to anti-inflammatory M2-type macrophages. Several promising anti-inflammatory herbal extracts were identified in this study, including extracts with previously unknown influence on key TLR signaling pathways and macrophage repolarization, serving as a basis for novel lead compound identification.