Page 295 - 2025中醫藥與天然藥物聯合學術研討會-中醫藥與天然藥物的挑戰Ｘ機遇與未來大會手冊

CM-28

Establishing identifiable characteristic fingerprints of mulberry leaves:

Integrating chemical composition and bioactivity through machine learning

3,4
5,6
1,2
Che Chun Lin, San Yuan Wang, Kowit Yu Chong, Vinh Tuyen T Le, Yi Ying
1,7
8
Lin, Lih Geeng Chen, Chia Jung Lee, 1,2,10 Ching Chiung Wang * ,1,2,3,10
9

1 Ph.D. Program in Clinical Drug Development of Herbal Medicine, College of Pharmacy,
Taipei Medical University, Taipei, Taiwan
2 Graduate Institute of Pharmacognosy, College of Pharmacy, Taipei Medical University,
Taipei, Taiwan
3 Master Program in Clinical Genomics and Proteomics, College of Pharmacy, Taipei Medical
University, Taipei, Taiwan
4 School of Pharmacy, College of Pharmacy, Taipei Medical University, Taipei, Taiwan
5 Department of Medical Biotechnology and Laboratory Science, College of Medicine, Chang
Gung University, Taoyuan, 333, Taiwan
6 Hyperbaric Oxygen Medical Research Laboratory, Bone and Joint Research Center, Linkou
Chang Gung Memorial Hospital, Taoyuan, 333, Taiwan
7 Department of Pharmacognosy - Traditional Pharmacy - Pharmaceutical Botany, College of
Pharmacy, Can Tho University of Medicine and Pharmacy, Can Tho, 941, Vietnam
8 Graduate Institute of Pharmacognosy, Taipei Medical University, Taipei, Taiwan.
9 Department of Microbiology, Immunology and Biopharmaceuticals, College of Life
Sciences, National Chiayi University, 300 University Road, Chiayi, 600, Taiwan
10 Traditional Herbal Medicine Research Center, Taipei Medical University Hospital, Taipei,
Taiwan
* E-mail: crystal@tmu.edu.tw

Abstract
Mulberry leaves (Morus alba L.) are widely consumed as herbal tea in Asia. They are
classified as traditional Chinese medicines because of their capacity to dispel wind heat, clear
the lungs, and moisten dryness. The Taiwan Herbal Pharmacopeia designates rutin as a
reference marker, but a single compound cannot fully reflect efficacy. This study integrated
HPLC fingerprinting, multivariate analyses, and artificial neural networks (ANN) to establish
a comprehensive quality evaluation. Analysis of 27 samples revealed two chemical groups, with
neochlorogenic acid, cryptochlorogenic acid, chlorogenic acid, rutin, isoquercitrin, and
astragalin identified as identification markers. A relevant content marker system was
established using 0.1% rutin as the classification threshold. A HPLC identification fingerprint
with 17 relative peaks was developed using ANN analysis. Qualified mulberry leaf samples
were used to evaluate their in vivo effects in a bleomycin-induced pulmonary fibrosis mouse
model. Mulberry leaf extract inhibited collagen accumulation and improved pulmonary fibrosis.
Further correlation analysis indicated that rutin inhibited MMP-13 expression, whereas
cryptochlorogenic acid inhibited MMP-13 and PAI-1 expression. These findings suggest that
rutin and cryptochlorogenic acid are key markers for the evaluation of mulberry leaves,
providing a basis for a multicomponent analytical method and quality assessment framework.

Keywords: Morus alba L.; Machine Learning; Identification Markers; Key Markers