Development of Discrimination Method for the Three Varieties of Panax vietnamensis using Leaf Phytometabolites = 잎 식물대사산물을 이용한 Panax vietnamensis 3품종 감별법 개발

다국어 초록 (Multilingual Abstract)

Panax vietnamensis, indigenous to Vietnam and southern China, is classified into three subspecies: Panax vietnamensis Ha et Grushv (PVV), Panax vietnamensis var. fuscidiscus (PVF), and Panax var. lanbianesis (PVL). They are renowned for its rich medicinal components and high research value, which significantly enhance its economic importance. The high cost of Vietnamese ginseng makes it vulnerable to adulteration, posing safety risks. Developing rapid and accurate identification methods is crucial for quality control. However, the similarities among the three types of Vietnamese ginseng in various plant parts, such as roots, stems, leaves, and flowers, make visual differentiation challenging when these varieties are interplanted. A method to distinguish these varieties in their intact form is absent, which poses a possible risk of misclassification. Meanwhile, there has no research undertaken to discriminate three different varieties of Panax vietnamensis, or to discover phytochemicals as markers other than ginsenoside derivatives. Here, I aimed to devise a plant metabolite-based discrimination algorithm for the three varieties, without causing significant damage to individual plants. A multivariate analysis on mass spectral data of PVV, PVF, and PVL revealed that a peak at m/z 426, which was subsequently identified as an indole alkaloid glycoside, was exclusive to PVF and therefore clearly distinguished PVF from PVV and PVL. Additionally, global metabolic profiling by using liquid chromatography—mass spectrometry was conducted to elucidate the discrimination markers between PVV and PVL, OPLS-DA analysis can clearly distinguish PVV and PVL, with lysophospholipids being more abundant in PVL, while hydroxy fatty acids are significantly higher in PVV. Therefore, lysophospholipids and hydroxy fatty acids were selected as potential discrimination markers. The performance of these markers was validated by cross-validation using machine learning algorithm. As a result, I have devised a standard protocol for discriminating intact samples of the three variants of Panax vietnamensis through comprehensive profiling of their phytochemical attributes employing various analytical modalities.

국문 초록 (Abstract)

베트남산 삼백수는 베트남과 중국 남부 원산지로, Panax vietnamensis Ha et Grushv (PVV), Panax vietnamensis var. fuscidiscus (PVF), Panax var. lanbianesis (PVL)의 3개 하종으로 분류된다. 이들은 풍부한 약용 성분과 높은 연구 가치로 유명하며, 이는 경제적 중요성을 크게 향상시킨다. 베트남 삼백수의 고가로 인해 이를 대체하는 일이 있어 안전 문제가 발생할 수 있다. 빠르고 정확한 식별 방법 개발은 품질 통제에 중요하다. 그러나 이들 사이의 유사성으로 인해 근본적으로 시각적 구분이 어렵다. 또한, 인도의 방법이 없어 잘못된 분류의 위험이 있다. 한편, Panax vietnamensis의 3가지 다른 종을 구분하는 연구나 진세노 사이드 유도체 외의 식물 대사 물질을 발견한 연구가 없었다. 따라서 본 연구에서는 개별 식물에 심각한 손상을 입히지 않고도 세 가지 다양한 종류의 Panax vietnamensis의 식별 알고리즘을 개발하기 위해 식물 대사 물질을 기반으로 한 구별 알고리즘을 설계하였다. PVV, PVF 및 PVL의 질량 스펙트럼 데이터에 대한 다변량 분석 결과, 이들 중 PVF에만 발생하는 m/z 426의 피크가 PVF를 PVV 및 PVL과 명확하게 구별할 수 있음을 확인하였다. 또한, 액체 크로마토그래피-질량 분석을 사용한 전역 대사 프로파일링을 통해 PVV와 PVL 간의 구별 표지자를 규명하였다. OPLS-DA 분석 결과, 라이소인지질은 PVL에서 보다 풍부하며, PVV에서는 수소화 지방산이 상당히 높았다. 따라서 라이소인지질과 수소화 지방산을 잠재적인 구별 표지자로 선택하였다. 이러한 표지자들의 성능은 기계 학습 알고리즘을 사용한 교차 검증을 통해 검증되었다. 결과적으로, 본 연구는 다양한 분석 기법을 활용하여 세 가지 다양한 Panax vietnamensis 종의 식물 대사 물질 속성을 종합적으로 프로파일링하여 인텍스트 샘플의 구별 표준 프로토콜을 개발하였다.

목차 (Table of Contents)

• Table of Content
• Abstract i
• 1. Introduction 1
• 1.1 Ginseng Overview 1
• 1.1.1 History of ginseng 1
• 1.1.2 Morphology of ginseng 1
• 1.1.3 Pharmacological components of ginseng 3
• 1.2 Panax vietnamensis 6
• 1.2.1 Introduction to Panax vietnamensis 6
• 1.2.2 Classification of Panax vietnamensis 7
• 1.2.3 Current Methods of Distinguishing Subspecies 7
• 1.3 Ginseng discrimination method 9
• 1.3.1 DNA analytical methods 9
• 1.3.2 Methods for metabolomic analysis 9
• 1.3.3 Other Methods 10
• 1.4 Confusion matrix of machine learning 11
• 1.4.1 Introduction of machine learning 11
• 1.4.2 Receiver operating characteristic curve 13
• 2. Experimental methods and materials 14
• 2.1 Scheme of the research 14
• 2.2 Chemicals and Reagents 16
• 2.3 Experimental methods 16
• 2.3.1 Sample Acquisition and Extraction 16
• 2.3.2 Gene Analysis by Using KAPS Markers 17
• 2.3.3 Flow Injection Analysis Coupled to Mass Spectrometry (FIA-MS): Untargeted Metabolomics 18
• 2.3.4 Thin Layer Chromatography 18
• 2.3.5 PVF key compound confirmation 19
• 2.3.6 Global Profiling of Metabolites: Large-scale Targeted Metabolomics 19
• 2.3.7 Statistics Analysis ２０
• 2.3.8 Differentiation between PVV and PVL by machine learning algorithm 20
• 3. Results 21
• 3.1 Morphological characteristics have little power in the discrimination of three varieties of Panax vietnamensis 21
• 3.2 Panax vietnamensis is classified by Pvdm12, but not the three varieties 22
• 3.3 PLS-DA on mass spectrum and TLC visualization reveal a PVF-specific compound 24
• 3.4 MS and NMR of PVF-specific compound 27
• 3.5 Potential differentiating markers between PVV and PVL 35
• 3.6 An intact sample discrimination algorithm 41
• 4. Discussion 42
• 5. Conclusion 44
• 6. Abbreviations Used 45
• 7. Bibliography 46
• 8. Abstract in Korean 53
• 9. Acknowledgements 54