Liu Yitong, Lawless Michael, Li Miao, Fairman Kiara, Embry Michelle R, Mitchell Constance A
Division of Toxicology, Office of Applied Research and Safety Assessment, Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, Laurel, Maryland, USA.
Simulations Plus, Lancaster, California, USA.
J Appl Toxicol. 2024 Aug;44(8):1236-1245. doi: 10.1002/jat.4617. Epub 2024 Apr 24.
Botanicals contain complex mixtures of chemicals most of which lack pharmacokinetic data in humans. Since physicochemical and pharmacokinetic properties dictate the in vivo exposure of botanical constituents, these parameters greatly impact the pharmacological and toxicological effects of botanicals in consumer products. This study sought to use computational (i.e., in silico) models, including quantitative structure-activity relationships (QSAR) and physiologically based pharmacokinetic (PBPK) modeling, to predict properties of botanical constituents. One hundred and three major constituents (e.g., withanolides, mitragynine, and yohimbine) in 13 botanicals (e.g., ashwagandha, kratom, and yohimbe) were investigated. The predicted properties included biopharmaceutical classification system (BCS) classes based on aqueous solubility and permeability, oral absorption, liver microsomal clearance, oral bioavailability, and others. Over half of these constituents fell into BCS classes I and II at dose levels no greater than 100 mg per day, indicating high permeability and absorption (%F > 75%) in the gastrointestinal tract. However, some constituents such as glycosides in ashwagandha and Asian ginseng showed low bioavailability after oral administration due to poor absorption (BCS classes III and IV, %F < 40%). These in silico results fill data gaps for botanical constituents and could guide future safety studies. For example, the predicted human plasma concentrations may help select concentrations for in vitro toxicity testing. Additionally, the in silico data could be used in tiered or batteries of assays to assess the safety of botanical products. For example, highly absorbed botanical constituents indicate potential high exposure in the body, which could lead to toxic effects.
植物药含有复杂的化学混合物，其中大多数在人体中缺乏药代动力学数据。由于物理化学和药代动力学特性决定了植物成分的体内暴露情况，这些参数对消费品中植物药的药理和毒理作用有很大影响。本研究试图使用计算（即计算机模拟）模型，包括定量构效关系（QSAR）和基于生理的药代动力学（PBPK）模型，来预测植物成分的特性。对13种植物药（如南非醉茄、 kratom和育亨宾）中的103种主要成分（如睡茄内酯、帽柱木碱和育亨宾碱）进行了研究。预测的特性包括基于水溶性和渗透性的生物药剂学分类系统（BCS）类别、口服吸收、肝微粒体清除率、口服生物利用度等。在每日剂量不超过100毫克的情况下，超过一半的这些成分属于BCS I类和II类，表明在胃肠道中具有高渗透性和高吸收性（%F > 75%）。然而，一些成分，如南非醉茄和亚洲人参中的糖苷，口服后由于吸收不良而生物利用度较低（BCS III类和IV类，%F < 40%）。这些计算机模拟结果填补了植物成分的数据空白，并可为未来的安全性研究提供指导。例如，预测的人体血浆浓度可能有助于选择体外毒性试验的浓度。此外，计算机模拟数据可用于分层或系列试验，以评估植物药产品的安全性。例如，可以高吸收的植物成分表明在体内可能有高暴露，这可能导致毒性作用。
