|
UK researchers are confident that in silico screening of Chinese medicines will enable faster drug discovery. The team has also revealed several potential leads from an analysis of such materials used in traditional Chinese medicine that could one day be useful in treating HIV/AIDS, cancer, Alzheimer's Disease, arthritis and various other conditions. Some of the more outlandish claims of herbal medicine do not bear fruit, but in Traditional Chinese medicine there are many health claims for the likes of Ginkgo biloba that might bear closer scrutiny. Pharmaceutical chemist David Barlow and colleagues Peter Hylands and Thomas Ehrman at King's College London hoped to apply the multiple decision tree technique known as Random Forest to unearth the root of the activity of natural products found in TCM by mining a large database of its ingredients with known medical indications and constituents. Barlow says that his PhD student Ehrman came up with the idea of screening the herbal databases to help them catalogue the active chemical constituents in TCM and to marry that information to the medical conditions for which various herbal remedies are indicated by practitioners. The team built a database containing 8264 compounds from the 240 most common Chinese herbs and used a second database of 2597 known active phytochemicals and related natural products as a training set in the largest study of its kind so far. Ehrman then carried out a search using Random Forest to identify any substances within the database having potential for treating a range of human diseases. The drug targets in question included cyclic adenosine 3'-5'-monophosphate phosphodiesterases, protein kinase A, cyclooxygenases, lipoxygenases, aldose reductase, and three HIV targets-integrase, protease, and reverse transcriptase. Additionally, the team investigated compound able to inhibit the expression of inducible nitric oxide synthase and/or nitric oxide production in vivo. "Random Forest was found to perform well, even on highly unbalanced data characteristic of ligand-based screening where the compounds to be screened are far more numerous than the number of active compounds used in training," the researchers say. Perhaps unsurprisingly to practitioners some 62% of the herbs screened contained 'candidate' drug compounds with the potential for use in treating a single disease. Moreover, more than half of the compounds would have medical activity against at least two diseases. The researchers then carried out a literature search to uncover evidence of known TCM active ingredient and found 83 herb-target predictions that matched their results. Ehrman cites several prominent examples. Ginseng and rhubarb, for instance, could have anti-inflammatory activity, while ginkgo and mint may represent useful leads in treating diabetes. Pomegranate and clove harbour natural products, including lignans, carotenoids and phenolics, that could lead to antiviral compounds for fighting HIV. "What this work shows is that the relatively new science of molecular informatics may have much to offer in furthering our understanding of how herbs work," Ehrman says, "and in developing new, and perhaps safer, medicines based on nature's own resources." The researchers point out that there is still a relatively limited amount of data on phytochemicals available in a format that might be screened using this approach. This "places appreciable constraints on informatics-based research in this area," the team says, although efforts are being made to address this problem. Ligand-receptor docking experiments and further theoretical approaches to understanding the interaction of herbal extracts with putative targets are now needed before next-generation pharmaceuticals with fewer side-effects will be possible. Related links:
|
Ginseng, the root of... |
