Nature Exposed to Our Method of Questioning: Reliability of Natural Product Measurements

Natural product research encompasses a spectrum of biomedical and chemical investigations ranging from new compound discovery to pharmacokinetic and other clinical studies. Common to such investigations is the need to demonstrate integrity and reproducibility of data. Data quality efforts range from assuring that biomass is properly identified and herbarium specimens cataloged to determining that quantitative chemical measurements are accurate and precise, and that the methods are transferrable to other laboratories. Modern research on botanicals may include discovery of bioactive phytochemicals, including investigations of synergistic effects of complex mixtures in the botanical matrix. In the phytomedicine field, botanicals and their contained mixtures are considered the active pharmaceutical ingredient (API), and natural product scientists are increasingly called upon to supplement their molecular discovery work by assisting in the development of analytical tools for assessing complex products. Unlike single-chemical APIs, botanicals are variable because their composition depends on genotypic and phenotypic variation, geographical origin, weather exposure, harvesting practices, and processing. Inherent variability in raw materials can result in inconsistent research materials and commercial products that are under-potent, over-potent, and/or contaminated. Natural product chemists have routinely developed quantitative methods for phytochemicals of interest as part their overall investigations. Publication of such methods occurs at the discretion of individual investigators, but when published, the methods described often serve as starting points for methods used by researchers, regulatory, and quality control scientists. Ideally, published methods should be accurate, precise, and reproducible. Accordingly, this review discusses the principles of assuring data integrity and provides an overview of approaches to demonstrating method performance.