Evaluation of environmental factors affecting the quality of Codonopsis pilosula based on chromatographic fingerprint and MaxEnt model

Abstract

The quality control of Traditional Chinese medicine (TCM) is rather challenging owing to the diversity and complexity of chemical components, which fluctuate with geographic origin and growth environment. In this study, a comprehensive strategy based on ultra-high performance chromatographic (UPLC) fingerprint technique and MaxEnt model was proposed to evaluate the effects of environmental factors on the quality of Codonopsis pilosula. An UPLC analytical method was developed and validated by precision, repeatability and stability, and then applied to analyze 91 sets of samples collected from different sites of Dingxi district, Gansu province, China (one of the geo-authentic producing areas of Codonopsis pilosula). With a similarity evaluation software, a standard UPLC fingerprint was obtained and 23 common peaks were found out. The MaxEnt model was established based on the principle of maximum entropy and its accuracy was evaluated by the receiver operating characteristic (ROC) curve and the area under the curve (AUC). Based on the UPLC fingerprint and MaxEnt model, the potential habitat suitability distribution of Codonopsis pilosula in Dingxi district was obtained, and the correlation model between ecological factors and chemical components was established. With the spatial analysis function of ArcGIS, the quality zoning map of Codonopsis pilosula was further drawn integrating the above distribution map of potential habitat suitability and the correlation model. The research results could provide a reference for the selection of planting area and the production regionalization of an herbal medicine.