Aristolochia species: Toxicological risk assessment using a metabolomic approach

Aristolochia species are toxic plants used as herbal medicines worldwide [1]. Their use has become a global public health concern [2] since they are known to cause aristolochic acid nephropathy, a devastating disease associated with kidney failure and kidney cancer [3]. The aims of this project are to assess the health risks associated with the use of different Aristolochia species and to elucidate the active principle behind their nephrotoxic effects. 44 medicinally used Aristolochia spp. were extracted with 70% methanol. The extracts were analyzed using LC-MS and NMR fingerprinting. The fingerprints were compared using principal component analysis and several aristolochic acid analogues were identified and quantified. The cytotoxicity and genotoxicity of 30 Aristolochia extracts was measured in human kidney (HK-2) cells. IC 50 values were determined using the sulforhodamine B (SRB) assay [4]. Genotoxicity, apoptosis, cytotoxicity and cell cycle effects were measured using the flow cytometry-based micronucleus test (MNT). The content and nature of aristolochic acid analogues (AAAs) in Aristolochia species varies greatly. AA I and AA II are the most common AAAs, but aristolactam I, AA IV, AA C and AA D are widespread as well. In general, Aristolochia leaves contain less AAAs than seeds, roots and flowers. Extracts containing high amounts of AA I showed only moderate genotoxicity and cytotoxicity in HK-2 cells. On the other hand, micronuclei formation was also observed in extracts, which only contain aristolactams (e.g. A. guentheri ). Therefore aristolactam I and related compounds need to be considered as potential nephrotoxic agents in Aristolochia spp. The pharmacological effects of these agents require further research. References: [1] Heinrich, M., et al (2009)J Ethnopharmacol, 125, 108 – 144. [2] Chen, C. H., et al. (2012) Proc Natl Acad Sci U S A, 109, 8241 – 8246. [3] Nortier J.L., et al. (2000) N Engl J Med, 342, 1686 – 1692. [4] P. Houghton, et al. (2007) Methods, 42, 377 – 387.