Discovery and ADME profiling of CNS-active natural products

Gamma aminobutyric acid type A (GABAA) receptors are the major inhibitory neurotransmitter receptors in the Central Nervous System (CNS). Several clinically important drugs used to treat anxiety, insomnia and epilepsy act via an allosteric modulation of postsynaptic GABAA receptors. The currently used drugs are associated with serious side-effects, mainly due to a lack of receptor subtype selectivity. This raises the medical need for discovery of novel types of GABAA receptor modulators. We previously identified a series of allosteric GABAA receptor agonists with the aid of HPLC-based activity profiling, whereby activity was tracked with an electrophysiological assay in Xenopus oocytes expressing GABAA receptors of desired subunit composition. In an expansion of our investigations, aiming at acceleration of the activity profiling, an in-house assay was established using larval zebrafish locomotor activity model. In that, larval convulsions were provoked by the pro-convulsant GABAA receptor antagonist pentylenetetrazol (PTZ), and GABAA receptor agonistic extracts and compounds were identified through a decrease in larval locomotion. The assay was validated with the aid of known GABAergic compounds that had previously shown activity in the Xenopus oocyte assay. Assay validation was approached with respect to parameters relevant for the quality of results, including PTZ concentration, number of larvae, concentration of test samples, duration of incubation with test solutions and tracking of larval locomotion, as well as data visualization protocol. The validated protocol was subsequently translated into an HPLC-based activity profiling protocol using ethyl acetate extracts of Valeriana officinalis and Magnolia officinalis. The zebrafish larvae locomotor assay was later employed for activity profiling of South African medicinal plants traditionally used for the treatment of epilepsy and other neurological disorders. An initial screening of medicinal plants in Xenopus oocytes patch clamp assay revealed GABAergic activity of a dichloromethane extract from leaves of Searsia pyroides. The extract significantly lowered PTZ-provoked locomotion in zebrafish larvae when tested at 4 μg/mL. HPLC-based profiling followed by targeted isolation of phytochemicals in the active timewindow, revealed identification of three anacardic acid derivatives (1-3). Also, three structurally related compounds (4-6) were purified from inactive areas. After assessment of the isolated compounds with both Xenopus oocyte and larval zebrafish models, GABAA receptor modulation activity of the extract was successfully correlated with the phytochemicals in the active time-window. Additionally, in the zebrafish larval assay a series of GABAA receptor agonistic natural products, previously identified by the Xenopus oocyte assay, were tested. Lowering of locomotor activity was found for these compounds with exception of sanggenone C. Physicochemical and biochemical properties (PSA, cLogP, number of H-donor and acceptor sites, and number of rotatable bonds) of tested compounds were calculated in silico. A lack of permeability across the blood-brain-barrier (BBB) was concluded for sanggenone C, inferring its lack of activity in vivo, which also remarked the exclusivity of the zebrafish larvae assay for discovery of BBB permeable natural products. The last part of our research focused on the membrane permeability of kaempferol (KMF), a sedative flavonoid which targets GABAA receptors, and its major intestinal metabolite, 4-hydroxyphenylacetic acid (4-HPAA). In previous studies, KMF induced sedative effects in mice only after oral administration (i.o.) but not after intraperitoneal application (i.p.). However, 4- HPAA, the biotransformation product of KMF by intestinal microflora, induced behavioral changes after i.p. injection. To further explore the relation between KMF and 4-HPAA bioactivities and their route of administration, their ability to cross biological barriers has been examined. Intestinal barrier permeability studies were performed with Caco-2 cells, and bloodbrain barrier transport studies were done with an immortalized mono-culture human model and a primary triple-co-culture rat model. UHPLC–MS/MS quantification methods for bioanalysis of KMF and 4-HPAA in the corresponding transport media were developed and validated according to international guidelines. The fundamental validation parameters included accuracy, precision, specificity, selectivity, sensitivity, repeatability, reproducibility, short-term and long-term stabilities. Data obtained with all barrier models were indicative of high intestinal and BBB permeation of KMF, and no permeation for 4-HPAA when compared with the fluorescent barrier integrity markers. In all experiments efflux ratios were below two, indicating that no active transport processes were involved for both compounds. Within a calcein-AM uptake assay in porcine brain capillary endothelial cells, kaempferol and 4-HPAA were found neither Pglycoprotein substrates nor P-glycoprotein inhibitors. Our in vitro data supported the previous described in vivo CNS effects of KMF while the role of 4-HPAA needs to be further studied.