Potential of South African medicinal plants as treatments for Alzheimer’s disease through Aβ42 protein reduction and identification of their active compounds

Abstract

Alzheimer’s disease is a slow and progressive neurodegenerative disease that destroys memory and other important mental functions. APP (Amyloid precursor protein) is an integral membrane protein expressed in higher levels in brain and its abnormal cleavage produces beta amyloid (Aβ42). Accumulation of Aβ42 in the brain produces amyloid plaques, the hallmark of the Alzheimer’s disease. Most of the research for the treatment of Alzheimer’s disease on medicinal plants in South Africa, focuses on acetyl cholinesterase inhibition, however this does not play a significant role against amyloid plaques. The goal of this study is to identify, characterize and develop new natural ingredients from the South African medicinal plants for the treatment of Alzheimer’s disease by targeting the reduction of Aβ42 resulting in decreased levels of beta amyloid plaques. A literature survey was done for the selection of plants by using the keywords memory loss, aging, epilepsy, depression, mental health, dementia, forgetfulness, Alzheimer’s disease, African medicinal plants, acetylcholinesterase. The plants were prioritized by a scoring system following the key criteria- plant part use, strength of the traditional use in relative to memory loss, plant toxicity, plant availability, published information against acetylcholinesterase inhibition. A total of 21 plant species were identified for collection, extraction and screening. Different plant parts (leaves, bark, stem, roots and fruits) of the selected plants were collected from the University of Pretoria gardens and extracted with DCM:MeOH (1:1) after drying and grinding. The 34 plant extracts were screened to determine their inhibitory properties for Aβ42 production in HeLa cells stably transfected with APPsw. We found that of 34 extracts, 11 extracts (32.3%) were found significantly decrease the Aβ42 level. Amongst them, the extracts of the leaves of Xysmalobium undulatum, leaves of Cussonia paniculata, leaves of Schotia brachypetala and stems of Heteromorha arborescens potently reduced the secreted level of Aβ42 by 77.3 ± 0.5%, 57.5 ± 1.3%, 44.8 ± 0.1% and 68.6 ± 0.2%, respectively at 50 µg/ml. These extracts were selected for further studies and UPLC- QTOF-MS was used to identify the active ingredients in the extracts. A total of five compounds were tentatively identified including four flavonoids, isoquercetin (25), myricetin-3-O-alpha-L-rhamnopyranoside (26), quercetin-3-O-rhamnoside (27), quercetin (28) from leaf extract of S. brachypetala. The quercetin derivatives were previously studied for their potential to treat Alzheimer’s disease through Aβ reduction in different bioassays, in all likelihood the compounds would be responsible for the Aβ42 reduction in S. bracypetala. Using ESI negative mode of UPLC-QTOF-MS, a total of nine compounds were tentatively identified as quinic acid (29), 3,5-dicaffeoylquinic acid (30), rutin (31), valeriananoid E (32), acuminoside (33), dictamnoside D (34), 2''-O-β-D-glucopyranosylsaikosaponin B2 (35), clinoposaponin C (36), spinasaponin C (37) from the leaf extract of C. paniculata. Rutin is one of the most researched compound for the Alzheimer’s disease in previous studies. 3,5-dicaffeoylquinic acid was previously reported for AChE inhibition. Rutin or 3,5-dicaffeoylquinic acid are in all likelihood the active ingredients in C. paniculata leaf extract. Using ESI negative mode, a total of four compounds were tentatively identified named cynarin (1,5-dicaffeoylquinic acid) (38) alternoside IX (39), alternoside I (40) and saikogenin B4 (41) the stem extract of H. arborescens. Cynarin was previously reported for the reduction of Aβ cytotoxicity while the other identified compounds, triterpenoid saponins were not examined for Alzheimer’s disease, but found to exhibit neuroprotective properties. These compounds may in all likelihood contribute as the active ingredients from H. arborescens leaf extract for the reduction of Aβ42. The DCM:MeOH leaf extract of X. undulatum was fractionated using silica gel column chromatography to produce 15 fractions which were screened for Aβ42 production. Of these two fractions, were found to be active as they significantly reduced Aβ42 levels. Five pure compounds were successfully isolated from active fractions using preparative HPLC. Using NMR and UPLC-QTOF-MS, the structure of these compounds were elucidated as acetylated glycosydated crotoxigenin (56), xysmalogenin-3, β-D-glucopyranoside (57), crotoxigenin-3-O-β-digitalopyranosyl-(1-4)-O-β-digitoxopyanoside (58), crotoxigenin 3-O-glucopyranoside (59) and desglucouzarin (60) isolated from the leaves of X. undulatum. The present study reports for the first time, the evaluation of the compounds for their potential to reduce Aβ42 levels. The four cardenolide glycosides, acetylated glycosydated crotoxogenin (56), xysmalogenin-3, β-D-glucopyranoside (57), crotoxigenin-3-O-β-digitalopyranosyl-(1-4)-O-β-digitoxopyanoside (58), desglucouzarin (60) were found to significantly decreased the Aβ42 levels at 10μM by 75.04 ± 0.85%, 74.66 ± 0.1%, 65.55 ± 0.19% and 69.15 ± 0.25%, respectively as compared to the negative control. These results revealed that all the pure compounds including the inactive compound (crotoxigenin 3-O-glucopyranoside (59) ) isolated from X. undulatum leaf extract were cardiac glycosides and contain the characteristic lactone ring which is known to be responsible for the cardiotoxicity. Since this was present in both the inactive and active cardenolides in all probability the lactone ring is the active site of the molecules contributing towards activity as such any analogues synthetically prepared lacking the lactone ring may still retain Aβ42 reduction while also reducing cardiotoxicity. Further the results of dose response studies of the 4 cardenolides indicated that crotoxigenin-3-O-β-digitalopyranosyl-(1-4)-O-β-digitoxopyanoside (58) was the most potent compound as it significantly reduced Aβ42 levels at the low concentration of 0.025 μM. These results revealed that this is the most active compound responsible for Aβ42 reduction in the plant. The bioassay-guided approach used for identifying the active ingredient in X. undulatum responsible for the reduction of Aβ42 resulted in identification of 4 cardenolides active ingredient, with one being the most active at nanomolar levels, which could be develop further into pharmaceutical drugs.