From quality markers to data mining and intelligence assessment: A smart quality-evaluation strategy for traditional Chinese medicine based on quality markers

Phytomedicine xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Phytomedicine journal homepage: www.elsevier.com/locate/phymed

From quality markers to data mining and intelligence assessment: A smart quality-evaluation strategy for traditional Chinese medicine based on quality markers ⁎

Gang Baia, , Tiejun Zhangb, Yuanyuan Houa, Guoyu Dinga, Min Jianga, Guoan Luoc a

State Key Laboratory of Medicinal Chemical Biology and College of Pharmacy, Tianjin Key Laboratory of Molecular Drug Research, Nankai University, Tianjin 300350, People's Republic of China b Department of Traditional Chinese Medicine, State Key Laboratory of Drug Delivery and Pharmacokinetics, Tianjin Institute of Pharmaceutical Research, Tianjin 300193, People's Republic of China c Analysis Center, Tsinghua University, Room 139, Building of Life Science, Beijing 100084, People's Republic of China

A R T I C L E I N F O

A B S T R A C T

Keywords: Q-marker Artificial intelligence UPLC/Q-TOF NIRS TCM

Background: The quality of traditional Chinese medicine (TCM) forms the foundation of its clinical efficacy. The standardization of TCM is the most important task of TCM modernization. In recent years, there has been great progress in the quality control of TCM. However, there are still many issues related to the current quality standards, and it is difficult to objectively evaluate and effectively control the quality of TCM. Purpose: To face these challenge, we summarized the current quality marker (Q-marker) research based on its characteristics and benefits, and proposed a reasonable and intelligentized quality evaluation strategy for the development and application of Q-markers. Methods: Ultra-performance liquid chromatography-quadrupole/time-of-flight with partial least squares-discriminant analysis was suggested to screen the chemical markers from Chinese medicinal materials (CMM), and a bioactive-guided evaluation method was used to select the Q-markers. Near-infrared spectroscopy (NIRS), based on the distinctive wavenumber zones or points from the Q-markers, was developed for its determination. Then, artificial intelligence algorithms were used to clarify the complex relationship between the Q-markers and their integral functions. Internet and mobile communication technology helped us to perform remote analysis and determine the information feedback of test samples. Chapters: The quality control research, evaluation, standard establishment and quality control of TCM must be based on the systematic analysis of Q-markers to study and describe the material basis of TCM efficacy, define the chemical markers in the plant body, and understand the process of herb drug acquisition, change and transmission laws affecting metabolism and exposure. Based on the advantages of chemometrics, new sensor technologies, including infrared spectroscopy, hyperspectral imaging, chemical imaging, electronic nose and electronic tongue, have become increasingly important in the quality evaluation of CMM. Inspired by the concept of Q-marker, the quantitation can be achieved with the help of artificial intelligence, and these subtle differences can be discovered, allowing the quantitative analysis by NIRS and providing a quick and easy detection method for CMM quality evaluations. Conclusion: The concept of Q-markers focused on unique CMM differences, dynamic changes and their transmission and traceability to establish an overall quality control and traceability system. Based on the basic attributes, an integration model and artificial intelligence research path was proposed, with the hope of providing new ideas and perspectives for the TCM quality management.

Abbreviations: TCM, traditional Chinese medicine; Q-marker, quality marker; CMM, Chinese medicinal materials; NIRS, near-infrared spectroscopy; APIs, active pharmaceutical ingredients; 3,5-diCQA, 3,5-dicaffeoylquinicacid; PCA, principal component analysis; MA, Medicinal attributes; ME, medicinal efficacy; β2AR, β2-adrenergic receptor; GPCR, G proteincoupled receptor; UPLC/Q-TOF, ultra-performance liquid chromatography-quadrupole/time-of-flight; PLS, partial least squares ⁎ Corresponding author. E-mail addresses: [email protected], [email protected] (G. Bai). https://doi.org/10.1016/j.phymed.2018.01.017 Received 11 September 2017; Received in revised form 20 December 2017; Accepted 20 January 2018 0944-7113/ © 2018 Elsevier GmbH. All rights reserved.

Please cite this article as: Bai, G., Phytomedicine (2018), https://doi.org/10.1016/j.phymed.2018.01.017

Phytomedicine xxx (xxxx) xxx–xxx

G. Bai et al.

Fig. 1. The Q-marker identification of Ephedra Herbaand Ephedra Radix was based on the traditional applicationof herbalism.

Introduction

of CMM are extremely complex. Therefore, we should understand, evaluate and control TCM quality beginning with the formation, change and transfer of all chemical components at the chemomics level. Families with similar kinship have similar chemical compositions and, therefore, similar efficacies. The geography, kinship, source pathways and other factors related to medicinal plants should be combined with traditional pharmacology to analyze their chemical compositions and to predict the function of CMM, to provide useful clues when searching for new drugs and to serve as an important foundation for identifying quality evaluation indicators.

The quality of traditional Chinese medicine (TCM) forms the foundation of its clinical efficacy. The standardization of TCM is the most important task of TCM modernization. In recent years, there has been great progress in the quality control of TCM. However, there are still many issues related to the current quality standards, and it is difficult to objectively evaluate and effectively control the quality of TCM. To solve this problem, Academician Liu has recently proposed the novel concept of the TCM quality marker (Q-marker) (Liu et al., 2016a). Focusing on the core concept of a Q-marker, Chinese scholars have been conducting quality research of TCM (Xiong and Peng, 2016; Sun et al., 2016), and they have gradually formed a new quality research model (Jiang and Wang, 2016; Zhang et al., 2017). In the past several decades, rapid quality control analysis has focused on holistic characterizations and determinations of the primary ingredients in Chinese medicinal materials (CMM). Recently, near infrared spectroscopy (NIRS) has attracted growing interest in TCM qualitative and quantitative control (Zhang and Su, 2014). However, the safety and activity test results related may not be sufficient. In addition, the complexity of multi-compounds and multi-target effects, as well as unclear standardizations, impedes the pace of modernization and internationalization of TCM (Han et al., 2011). To face these challenge, in this review, we summarized the current Q-marker research, based on its characteristics and benefits, we proposed a reasonable and intelligentized quality evaluation strategy for the development and application of Q-markers.

Traditional pharmacological properties of TCM Based on thousands of years of application experience, basic knowledge regarding the variety, origin, efficacy, and quality of TCM has been established. Herbalism evidence plays an important role in the quality evaluation of CMM. Hence, it is essential to understand the safety, efficacy, and quality of the traditional CMM pharmacology. For instance, plants of the Ephedra species are widely used in TCM. Based on traditional applications, Ephedra Herba is used to cure a cold fever by inducing sweating, whereas Ephedra Radix is used to treat hyperhidrosis. Although they come from the same plant, Ephedra sinica Stapf, the two herb drugs play different clinical application roles (Persky et al., 2004). Ephedra Herba contains ephedrine alkaloids, which drive the physiological changes in sweating, heart rate and blood pressure (Ma et al., 2007). In a recent report, the opposite effects on sweating that Ephedra Herba and Ephedra Radix have can be attributed to ephedrine and mahuannin B, an effective antihydrotic agent that inhibits the production of cAMP through suppression of adenylate cyclase activity (Fig. 1) (Wang et al., 2017). The characteristic components and active pharmaceutical ingredients (APIs) of TCM form the foundation for the screening quality markers. The history of and accumulated experience with traditional medicine has determined the basic varieties and habitual use of CMM. Hence, these experiences form the basis of quality research development. In addition, CMM quality also depends on its origin, as well as the ecological factors associated with the origin. Due to differences in their growth and ecological environments, the same species have corresponding variations and differences, resulting in quality differences. The origin of high-quality herbs is known as authentic “Daodi”, and high-quality herbs are known as “Daodi” herbs. Flos Chrysanthemi, a common herbal medicine, has five major cultivars (Boju, Chuju, Gongju, Hangbaiju and Huaiju) from different sources, but only three types (Chuju, Gongju and Hangbaiju) are

The importance of TCM characteristics and benefits in quality evaluations TCM differs from other chemical drugs that are composed of single chemical materials. TCM is derived from biological organisms, and it primarily exists in the form of clinical application compounds. In particular, preparing proprietary TCM is a complex process. CMM quality is based on characterizations of the effectiveness of the materials, which determines differences in the safety and efficacy of TCM. Knowledge of TCM efficacy and the basis of its clinical applications are derived from TCM theory and thousands of years of clinical experience. Therefore, quality evaluations of TCM should be based on the basic theory and clinical function of TCM. As a complex chemical substance system, the chemical composition of TCM is an important determining factor for quality evaluation and control. The formation, change and expression of the biological effects 2

Phytomedicine xxx (xxxx) xxx–xxx

G. Bai et al.

administration, glycyrrhizin is hydrolyzed to glycyrrhetic acid by human intestinal bacteria (Akao et al., 1991). Co-treatment with glycyrrhizin and β2-adrenergic receptor (β2AR) agonists yields synergistic anti-asthmatic effects, and glycyrrhizin enhances β2AR signaling by increasing cAMP accumulation (Bai et al., 2008). Hence, we speculate that in a G protein-coupled receptor (GPCR), such as β2ARs, as significant fraction of their signaling partners is embedded in lipid raft micro-domains. Glycyrrhetic acid decrease the level of cholesterol in lipid rafts, changes the fluidity of the cell membrane, and leads to the release of raft-embedded Gαs, as well as increased interaction with β2ARs (Shi et al., 2012). For the synergetic mechanism, GPCR/Gαs mediated signal transduction enhancement was regarded as the cause of synergism. As the Q-marker ingredient, glycyrrhizin contributes to both MA and ME in TCM prescription.

included in the Chinese Pharmacopoeia 2015. The content limits of chlorogenic acid (CA), luteolin and 3,5-dicaffeoylquinicacid(3,5diCQA) are set for quality management. However, the APIs of these cultivars have not been system characterized with respect to their traditional bioactivity. To develop a method for the standardization and quality control of Flos Chrysanthemi based on its effectiveness, ultraperformance liquid chromatography-quadrupole/time-of-flight, principal component analysis (PCA) and anti-inflammatory APIs were screened through an artificial neural network, combined with a NF-κB luciferase reporter gene assay system. In the results, nineteen chemical markers were confirmed to contribute to the cluster. Among them, only eleven ingredients were found to reveal potential anti-inflammatory activity, and only CA, luteolin-7-O-glucoside, 3,5-diCQA and luteolin were finally selected as the Q-markers for the quality control of Flos Chrysanthemi (Han et al., 2015). Moreover, this integrated approach may prove to be a powerful method for the rapid establishment and application of quality control procedures based on Q-markers (Ding et al., 2016).

Compatibility of TCM prescription is a basic and effective method of clinical usage The effectiveness of TCM is a comprehensive expression of the biological effects of APIs in herb drugs. It is also a comprehensive manifestation of the overall attributes of the chemical components as a whole compound medicine. A single ingredient cannot replace the ME and MA of the original medicine, and the whole effect is not a simple sum of the components. Currently, TCM therapies are becoming increasingly important in treating complex diseases because they can act on multiple targets in the disease network (Morphy and Rankovic, 2009; Pujol et al., 2010). Therefore, the varying quality of TCM is not only reflected in the chemical composition but also, in most cases, in the amounts of the API and their relative proportions. Qingfei Xiaoyan Wan, a classical formula-derived TCM for prevalent chronic lung disease, was used in our present work. A mutually enhanced bioactivity guided ultra-performance liquid chromatographyquadrupole/time-of-flight (UPLC/Q-TOF) characterization method was proposed, coupled with a dual-luciferase reporter assay for β2ARagonist cofactor screening (Hou et al., 2014). Ephedrine, as the principal bronchodilator, was indicated as the β2AR agonist, and four lignin compounds (arctiin, arctigenin, descurainoside and descurainolide B) that showed synergistic bronchodilation effects with ephedrine were also confirmed. However, these lignin compounds did not show any β2AR agonistic effects alone. For this reason, arctigenin bound to PDK1 and reduced the PKB/Akt induced phosphorylation of PDE4D and cAMP accumulation. Hence, the inhibition of PDK1 demonstrated a synergistic function with β2AR agonists (Fang et al., 2015). The pharmacological treatments summarized in the Global Initiative for Chronic Obstructive Lung Disease guidelines for managing stable chronic obstructive pulmonary disease only include β2AR agonists (Vestbo et al., 2013). However, in this case of Qingfei Xiaoyan Wan, the lignin compounds were also monitored as Q-markers to improve the quality of TCM.

The multiple properties of CMM The quality of CMM can also be reflected in its biological variation on multiple levels. Differences in the interspecific chemical composition of the polygene lead to differences in clinical efficacy. CMM usually has multiple effects, and different effects correspond to different chemical compositions. Therefore, for different biological effects, the corresponding efficacies of the APIs should be selected as the evaluation index. For example, many species of Bulbus Fritillariae are used as BeiMu in China, but their clinical applications are not standardized. To clarify the differences in and homologies of anti-muscarinic and anti-inflammatory effects, an integrated strategy combining Q-marker screening and function evaluation was established. The BeiMu extract from SongBei and QingBei showed the best anti-muscarinic effects, and several steroidal alkaloids, such as solanidine, contributed to these effects. However, the extract from ZheBei was superior in reducing the inflammatory response with the other components: puqiedine, zhebeiresinol, 2-monopalmitin, N-demethylpuqietinone, and isoverticine. Using these potential Q-markers, the new cluster method was able to illustrate their different functions, which was helpful for BeiMu quality control (Fig. 2) (Zhou et al., 2017). Characteristics of CMM biological effects Dual characteristics of the traditional features: medicinal attributes and medicinal efficacy Medicinal attributes (MA), known as “YaoXing”, and medicinal efficacy (ME), known as “YaoXiao”, are objective descriptions of CMM efficacy from different perspectives. MA usually refers to the macroscopic effect of an herbal medicine, and the ME is a specific function of expression. The two definitions have a complex interrelationship. Therefore, the characterizations of MA and ME are used to describe the pharmacological substance, principle and compatibility of TCM, to further guide their use in clinical practice. Many problems existed in the previous research model of ME and API, which neglected the holistic characteristics of CMM and lacked a characterization of API associated with MA. Therefore, the research strategy of systems biology and network pharmacology is more aligned with a holistic view, and it should be connected to the characteristics of Q-marker research (Bai et al., 2016b). This approach is important to inherit and carry forward valuable experience, to restore and illustrate the theory, to highlight the characteristics of TCM synergetic theory, and to guide clinical practice. Liquorice, a guide drug, “ShiYao”, has been widely used in TCM prescriptions (Wang et al., 2013). As a major API, the glycyrrhizinin demonstrates various nutraceutical and functional activities (Kao et al., 2014). Pharmacological studies have revealed that after oral

Establishment of quality research model based on Q-marker From chemical marker to Q-marker Q-markers are the core indicators associated with efficacy and safety in the quality evaluation of TCM. To understand the chemical components of TCM, the source and unique characteristics of the chemical composition were analyzed, and representative and specific chemical compositions were selected to further focus on establishing quality control directed by Q-markers. The quality control research, evaluation, standard establishment and quality control of TCM must be based on the systematic analysis of Q-markers to study and describe the material basis of TCM efficacy, define the chemical markers in the plant body, and understand the process of herb drug acquisition, change and transmission laws affecting metabolism and exposure. In addition, the Q-markers form the basis of the study of the effectiveness and 3

Phytomedicine xxx (xxxx) xxx–xxx

G. Bai et al.

Fig. 2. The cluster method was used to illustrate the different biological properties of BeiMu with anti-muscarinic or anti-inflammatory different Q-markers.

pharmaceutical evaluation and TCM quality control to solve related problems (Liu et al., 2017; Yoshida et al., 2014). At present, fingerprinting is widely accepted for the management of TCM quality using chromatography techniques. Chemometric techniques provide a good approach for mining more useful chemical information from chromatographic fingerprinting including peak alignment information features and baseline correction (Liu et al., 2016b). And the most common chemometrics tools in metabonomics are consisted with multivariate statistical analyses including principal components analysis (PCA) and partial least squares-discriminant analysis (PLS-DA) (Lee et al., 2017). The advantages of infrared spectroscopy for the quality control of TCM are also based on the holistic spectroscopic fingerprinting of all ingredients in the tested TCM sample. The principle for the identification of TCM is based on the differences in specificity of the secondary metabolites and the specificity of the herb medicines. Mid-infrared and 2D-infrared correlation spectroscopy have been applied to identify and detect some chemical compounds in CMM samples, and try to monitor the production process as good manufacturing practice in the TCM products industry (Sun et al., 2010; Liu et al., 2010). The principle of NIRS for the identification of TCM is also based on differences in the specificities of secondary metabolites and the specificities of herbal medicines. NIRS can comprehensively reflect the global chemical information and this holistic approach can be used for the identification and cluster analysis of CMM (Wang et al., 2014; Zhu et al., 2014; Lucio-Gutiérrez et al., 2013) and to distinguish processed products or adulterants (Xin et al., 2012a; Zhang et al., 2016; Nie et al., 2013). However, inspired by the concept of Q-markers, multiple API quantitation can be achieved with the help of chemometrics, and these subtle differences can be discovered, allowing the quantitative analysis of CMM and providing a quick and easy detection method for CMM quality evaluations (Bai et al., 2016a).

mechanism of action, and they also provide a complete and clear approach for the transmission and traceability of TCM quality. Chemical markers can profile the distinguishing information for identification, but a satisfactory discrimination associated with the bioactive function was not observed. In the published literature, we described a paradigm that involved a set of integrated strategies to improve the chemical markers for Q-markers in TCM quality management (Ding et al., 2017). The Q-markers are demonstrated as the API and can be used to clarify the relationship between the components and their MA and ME in TCM. Artificial intelligence techniques for TCM quality control Artificial intelligence techniques have demonstrated potential in most drug R&D fields (Ramesh et al., 2004; Song et al., 2014). With the rapid development of data mining, cheminformatics, computational biology, network pharmacology and systems biology techniques, great progress has been made recently in the artificial intelligence study of TCM mechanism analysis and clinical medication rule of TCM prescriptions (Gu and Pei, 2017; Zhang et al., 2016). Confronted with a huge volume of TCM data, artificial intelligence can be used not only to effectively explore these resources using the techniques of knowledge discovery in historically accumulated and recently obtained databases but also be applied to pattern recognition for TCM quality assessments (Feng et al., 2006; Zhao and Li, 2002). Based on the advantages of chemometrics and big data analysis, new sensor technologies, including infrared spectroscopy, hyperspectral imaging, chemical imaging, electronic nose and electronic tongue, have become increasingly important in the quality evaluation of CMM (Miao et al., 2017; Sandasi et al., 2014). In addition, artificial intelligence sense technology (AIST), including electronic tongue, nose, eye, ear and skin, can simulate the function of real human sense organ. Those ideas are advantage and unique that can get the favor of scientists and have been introduced in 4

Phytomedicine xxx (xxxx) xxx–xxx

G. Bai et al.

Table 1 The application characteristic of NIRS in TCM quality control. Test plant

State of sample

Quantitative compounds

Algorithms

App.

Ref.

Lonicerae japonicae flos

Powder

PLS

Q

Li et al. (2012)

Lonicerae japonicae flos Lonicerae japonicae flos Lonicerae japonicae flos Notoginseng radix et rhizoma

Liquid Liquid Liquid Liquid

iPLS PLS PLS UVE-PLS

M M M M

Wu et al. (2012) Wu et al. (2013) Xiong et al. (2012) Jiang and Qu (2015)

Citri reticulatae pericarpium

Liquid

5-OeCA; CA; 3,4-diCQA; 3,5-diCQA; 4,5-diCQA; caffeic acid CA CA;5-OeCA;3,5-diCQA;3,4-diCQA;4,5-diCQA CA notoginsenoside R1, ginsenoside Rg1, ginsenoside Re, ginsenoside Rb1, ginsenoside Rd hesperidin; nobiletin

M

Zhou et al., 2016

Coptidis rhizoma Scutellariae radix

Liquid Powder

siPLS/biPLS, bagging-PLS, boosting-PLS PLS PLS/PCA

M Q

Dai et al. (2016) Navarro Escamilla et al. (2013)

Angelicae sinensis radix Acorus calamus L. & Acorus tatarinowii Schott

Powder Powder

GA-MLR PCA/RF PLS

Q Q

Li et al. (2014) Ying et al. (2014)

Ginkgo folium Curcumae longae rhizoma Paridis rhizoma

Powder Powder Powder

PLS/DA PLS/DA PLS PLS/DA

Q Q Q

Liu et al. 2015 Kasemsumran et al. 2014 Li et al. (2016)

Pueraria lobata Atractylodis macrocephalae rhizoma Fritillaria

Liquid Powder

α-asarone Flavonol glycoside; moisture; extract contents curcumin polyphyllinI; polyphyllinII; polyphyllinVI; polyphyllin VII puerarin; daidzin; daidzein; total isoflavonoid atractylenolide I; tractylenolide III

biPLS; SPA PLS

M Q

Wang et al. (2015) Shao et al. (2014)

Powder

total alkaloids

PLS Factorization

Q

Meng et al. (2015)

berberine baicalin; total baicalein ferulic acid β-asarone;

M: Monitor of extraction process; Q: Quality control of the original medicinal materials or preparations.

Fig. 3. UPLC/Q-TOF integrated with NIRS was proposed to set-up an intelligentized systematic quality assessment of CMM based on Q-markers.

The progress and trend of TCM quality control by NIRS

extract solution, and the approach was found to be encouraging and reliable for real-time monitoring in the precipitation process (Li et al., 2012), multivariate analysis (Wu et al., 2012), and batch-to-batch reproducibility (Wu et al., 2013). As an excellent process analytical technology tool, NIRS has also been applied in on-line monitoring process of Panax Notoginseng (Xiong et al., 2012), Pericarpium Citri Reticulatae (Jiang and Qu, 2015), Coptis Rhizome (Zhou et al., 2016) and Puerarialobate (Dai et al., 2016) in the extraction and manufacturing processes. Based on the distinctive wavenumber section of API, partial least squares (PLS) regression models were developed for NIRS quantitative analysis of botanical raw materials. Thus, some APIs have been successfully quantified, including baicalin and total baicalein in

Due to its unique advantages, NIRS technology has been developed as a simple and reliable method to quickly evaluate the quality of CMM. Table 1 summarizes the progress and trends of modern TCM quality evaluation using NIRS to monitor the extraction process, API determination, and taxonomic identification of CMM preparations or sections. And most common chemometrics algorithms used in NIRS are based on PLS. CA and its derivatives, as the most important APIs in Lonicera japonica, were often used as examples to investigate the feasibility of NIRS evaluation systems and to monitor the extraction process and content variation. NIRS could be collected by fiber optic probes in the 5

Phytomedicine xxx (xxxx) xxx–xxx

G. Bai et al.

Fig. 4. The schematic diagram of an intelligent quality management system for Chinese medicinal materials.

marker content as the input variable and employing quantitative activity relationship network simulation prediction analysis, the CMM efficacy can be evaluated, providing an intelligent evaluation method for TCM grade identification and quality management (Ding et al., 2016; Ding et al., 2017). Hence, the established Q-marker-coupled NIRS pattern is a convenient, reliable and smart method of quantitative analysis. In addition, internet and mobile communication technology can help us to perform remote analysis and receive information feedback for test samples. Now, NIRS technique has been successfully applied in clinical trials, rehabilitation therapy, drug discovery and remote access field (Schachner et al., 2008; Safaie et al., 2013). Finally, convenient and reliable quality assessment patterns can be established using artificial intelligence for holistic potency evaluations of CMM or botanical products (Fig. 4).

Scutellariae Radix powder (Navarro Escamilla et al., 2013), ferulic acid in Radix Angelicae Sinensis samples (Li et al., 2014), β-asarone and αasarone in dried rhizome part of Acorus Calamus L. and Acorus Tatarinowii Schott (Ying et al., 2014), flavonol glycoside in ginkgo leaves (Liu et al., 2015), curcumin in turmeric herbal medicine powder (Kasemsumran et al., 2014), several polyphyllin compounds in Rhizomaparidis (Li et al., 2016), major bioactive isoflavonoid compounds in Puerarialobate (Wang et al., 2015), two atractylenolides in Rhizoma Atractylodis Macrocephalae (Shao et al., 2014), and total alkaloids from Fritillaria samples (Meng et al., 2015). A smart quality-evaluation systems for TCM quality management The use of UPLC/Q-TOF, along with multivariate statistical analysis, is an efficient approach to choose potential chemical markers between similar CMM (Fig. 3). However, it is not suitable as a routine analysis method because of its high maintenance cost and time requirements. NIRS as a substitute method as some advantages over other analytical techniques in that it is fast, easily used, low cost, and it records the spectra of solid and liquid samples with simple pretreatment in a short time and allows for the quantification of multiple components (Li et al., 2015; Shi et al., 2016). The use of NIRS in TCM quality control has usually been limited by its blindness to qualitative or quantitative analysis and because the analytical marker is easily ignored. Here, the use of UPLC/Q-TOF integrated with NIRS has been proposed as a systematic quality assessment of Chinese Eaglewood, a resinous heartwood from the Aquilariasinensis (Lour.) Gilg, with agarotetrol and its derivatives as Q-markers for CE quality assessment (Ding et al., 2015). Therefore, based on the key distinctive wave number Q-marker points, the NIRS method was proposed as being more suitable because of its fast determinations. In addition, the complex relationship between the Q-markers and the overall ME can be further predicted by machine learning, artificial neural networks, genetic algorithm-support vector regression and other algorithms (Ma et al., 2016; Xin et al., 2012b). In detail, using the Q-

Quality by design and full process supervision mode The transmission and traceability of Q-markers are reflected in every process of the TCM production. The study of quality control methods and determination of quality standards should focus on the whole process of TCM formation, using a two-way concept of transmission and traceability for whole-process quality control. All preparation processes, such as the collection, pre-treatment, extraction, purification, condensation and drying of CMM, produce complex changes in chemical composition, and different process routes and process parameters generate different results. Therefore, according to the Quality by design concept (Pramod et al., 2016), "good" quality control and quality standards should reflect the overall quality control and quality traceability based on Q-markers. Conclusions The concept of Q-markers as the primary evaluation is based on APIs, focusing on unique CMM differences, dynamic changes and quality and their transmission and traceability to establish an overall 6

Phytomedicine xxx (xxxx) xxx–xxx

G. Bai et al.

quality control and traceability system. The exploration of a TCM quality research model is a long-term, gradually improved system engineering process. The Q-marker is the core concept of quality evaluation, as well as an open concept, requiring continuous exploration to gradually enrich and improve its research ideas and methods, ultimately forming a scientific quality evaluation and quality control system to improve TCM quality standards. Based on the basic attributes and clinical features of TCM, this paper proposes an integration model and research path for TCM Q-marker research, with the hope of providing new ideas and perspectives for the quality research, overall quality management and intelligent control of TCM.

Thanapase, W., 2014. Evaluation of fourier transform-near infrared spectroscopic measurements for the quantification of curcumin in turmeric herbal medicines. J. Near Infrared Spectrosc. 22, 113–120. Lee, K.M., Jeon, J.Y., Lee, B.J., Lee, H., Choi, H.K., 2017. Application of metabolomics to quality control of natural product derived medicines. Biomol. Ther. (Seoul) 25, 559–568. Li, B., Wang, C., Xi, L., Wei, Y., Duan, H., Wu, X., 2014. Qualitative and quantitative analysis of Angelica sinensis using near infrared spectroscopy and chemometrics. Anal. Methods 6, 9691–9697. Li, W.L., Wang, Y.F., Qu, H.B., 2012. Near infrared spectroscopy as a tool for the rapid analysis of the Honeysuckle extracts. Vib. Spectrosc. 62, 159–164. Li, Y., Cai, W., Weng, X., Li, Q., Wang, Y., Chen, Y., Zhang, W., Yang, Q., Guo, Y., Zhu, X., Wang, H., 2015. Lonicerae Japonicae Flos and Lonicerae Flos: a systematic pharmacology review. Evid. Based Complement. Alternat. Med. 2015, 905063. Li, Y., Liu, B., Geng, S., Kim, S., Jin, Y., Liu, X., Luan, L., Wu, Y., Chen, Y., 2016. An approach combining real-time release testing with near-infrared spectroscopy to improve quality control efficiency of Rhizoma paridis. Spectrochim. Acta A Mol. Biomol. Spectrosc. 157, 186–191. Liu, C., Chen, S., Xiao, X., Zhang, T., Hou, W., Liao, M., 2016a. A new concept on quality marker of Chinese material medica: quality control for Chinese medicinal products. Chin. Tradit. Herb. Drugs 47, 1443–1457. Liu, R., Chen, P., Li, X., Wu, Z., Gao, X., Chen, X., Zhang, L., Wang, Q., Li, Z., 2017. Artificial intelligence sense technology: new technology in pharmaceutical sciences. Chin. J. Pharm. Anal. 37, 559–567. Liu, S., Liang, Y.Z., Liu, H.T., 2016b. Chemometrics applied to quality control and metabolomics for traditional Chinese medicines. J. Chromatogr. B 1015-1016, 82–91. Liu, X., Tao, L., Du, W., Jin, Y., Wu, Y., 2015. Quality control of Ginkgo biloba leaves by real time release testing in combination with near infrared spectroscopy. J. Near Infrared Spectrosc. 23, 381–389. Liu, Y., Zhang, G.J., Sun, S.Q., Noda, I., 2010. Study on similar traditional Chinese medicines cornu Cervi pantotrichum, cornu Cervi and cornu Cervi degelatinatum by FT-IR and 2D-IR correlation spectroscopy. J. Pharm. Biomed. Anal. 52, 631–635. Lucio-Gutiérrez, J.R., Coello, J., Maspoch, S., 2013. Expeditious identification and semiquantification of Panax ginseng using near infrared spectral fingerprints and multivariate analysis. Anal. Methods 5, 857–865. Ma, L., Zhang, Z., Zhao, X., Zhang, S., Lu, H., 2016. The rapid determination of total polyphenols content and antioxidant activity in Dendrobium officinale using nearinfrared spectroscopy. Anal. Methods 8, 4584–4589. Ma, G., Bavadekar, S.A., Davis, Y.M., Lalchandani, S.G., Nagmani, R., Schaneberg, B.T., Khan, I.A., Feller, D.R., 2007. Pharmacological effects of ephedrine alkaloids on human α1- and α2-adrenergic receptor subtypes. J. Pharmacol. Exp. Ther. 322, 214–221. Meng, Y., Wang, S., Cai, R., Jiang, B., Zhao, W., 2015. Discrimination and content analysis of Fritillaria using near infrared spectroscopy. J. Anal. Methods Chem., 752162. Miao, X., Cui, Q., Wu, H., Qiao, Y., Zheng, Y., Wu, Z., 2017. New sensor technologies in quality evaluation of Chinese material medica: 2010-2015. Acta Pharm. Sin. B 7, 137–145. Morphy, R., Rankovic, Z., 2009. Designing multiple ligands-medicinal chemistry strategies and challenges. Curr. Pharm. Des. 15, 587–600. Navarro Escamilla, M., Rodenas Sanz, F., Li, H., Schönbichler, S.A., Yang, B., Bonn, G.K., Huck, C.W., 2013. Rapid determination of baicalin and total baicalein content in Scutellariae radix by ATR-IR and NIR spectroscopy. Talanta 114, 304–310. Nie, P., Wu, D., Sun, D.W., Cao, F., Bao, Y., He, Y., 2013. Potential of visible and near infrared spectroscopy and pattern recognition for rapid quantification of notoginseng powder with adulterants. Sensors 13, 13820–13834. Persky, A.M., Berry, N.S., Pollack, G.M., Brouwer, K.L., 2004. Modelling the cardiovascular effects of ephedrine. Br. J. Clin. Pharmacol. 57, 552–562. Pramod, K., Tahir, M.A., Charoo, N.A., Ansari, S.H., Ali, J., 2016. Pharmaceutical product development: a quality by design approach. Int . J. Pharm. Investig. 6, 129–138. Pujol, A., Mosca, R., Farre´s, J., Aloy, P., 2010. Unveiling the role of network and systems biology in drug discovery. Trends Pharmacol. Sci. 31, 115–123. Ramesh, A.N., Kambhampati, C., Monson, J.R., Drew, P.J., 2004. Artificial intelligence in medicine. Ann. R. Coll. Surg. Engl. 86, 334–338. Safaie, J., Grebe, R., Abrishami Moghaddam, H., Wallois, F., 2013. Toward a fully integrated wireless wearable EEG-NIRS bimodal acquisition system. J. Neural Eng. 10, 637. Sandasi, M., Vermaak, I., Chen, W., Viljoen, A.M., 2014. Skullcap and Germander: preventing potential toxicity through the application of hyper spectral imaging and multivariate image analysis as a novel quality control method. Planta Med 80, 1329–1339. Schachner, T., Bonaros, N., Bonatti, J., Kolbitsch, C., 2008. Near infrared spectroscopy for controlling the quality of distal leg perfusion in remote access cardiopulmonary bypass. Eur. J. Cardiothorac. Surg. 34, 1253–1254. Shao, Q.S., Zhang, A.L., Ye, W.W., Guo, H.P., Hu, R.H., 2014. Fast determination of two atractylenolides in Rhizoma Atractylodis Macrocephalae by fourier transform near-infrared spectroscopy with partial least squares. Spectrochim. ActaA Mol. Biomol. Spectrosc. 120, 499–504. Shi, Q., Hou, Y., Hou, J., Pan, P., Liu, Z., Jiang, M., Gao, J., Bai, G., 2012. Glycyrrhetic acid synergistically enhances β2-adrenergic receptor-Gs signaling by changing the location of Gαs in lipid rafts. PLoS One 7, e44921. Shi, Z., Liu, Z., Liu, C., Wu, M., Sun, H., Ma, X., Zang, Y., Wang, J., Zhao, Y., Xiao, X., 2016. Spectrum-effect relationships between chemical fingerprints and antibacterial effects of Lonicerae Japonicae Flos and Lonicerae Flos base on UPLC and microcalorimetry. Front. Pharmacol. 7, 12. Song, J., Xie, J., Li, C., Lu, J.H., Meng, Q.F., Yang, Z., Lee, R.J., Wang, D., Teng, L.S., 2014. Near infrared spectroscopic (NIRS) analysis of drug-loading rate and particle

Conflict of interest None of the authors have any conflicts of interest to declare. Acknowledgments This project was financially supported by the National Natural Science Foundation of China (81430095, 81673616) and International Cooperation and Exchange of the National Natural Science Foundation of China (81761168039). References Akao, T., Akao, T., Hattori, M., Kanaoka, M., Yamamoto, K., Namba, T., Kobashi, K., 1991. Hydrolysis of glycyrrhizin to 18 beta-glycyrrhetyl monoglucuronide by lysosomal beta-D-glucuronidase of animal livers. Biochem. Pharmacol. 41, 1025–1029. Bai, G., Ding, G., Hou, Y., Jiang, M., Gao, J., 2016a. Rapid identification and evaluation of Chinese material medica quality by near infrared spectroscopy. ZhongguoZhongYaoZaZhi 41, 3501–3505. Bai, G., Hou, Y., Jiang, M., Gao, J., 2016b. Integrated systems biology and chemical biology approach to exploring mechanisms of traditional Chinese medicines. Chin. Herb. Med. 8, 99–106. Bai, G., Yang, Y., Shi, Q., Liu, Z., Zhang, Q., Zhu, Y.Y., 2008. Identification of higenamine in Radix Aconiti Lateralis Preparata as a beta2-adrenergic receptor agonist. Acta. Pharmacol. Sin. 29, 1187–1194. Dai, X., Song, H., Liu, W., Yao, S., Wang, G., 2016. On-line UV-NIR spectroscopy as a process analytical technology (PAT) tool for on-line and real-time monitoring of the extraction process of Coptis Rhizome. RSC Adv. 6, 10078–10085. Ding, G., Hou, Y., Han, Y., Zhang, J., Liu, A., Liu, Z., Peng, J., Jiang, M., Bai, G., 2016. A rapid score evaluation system based on near infrared spectroscopy for quality control of Flos Chrysanthemi. J. Pharm. Biomed. Anal. 131, 391–399. Ding, G., Nie, Y., Hou, Y., Liu, Z., Liu, A., Peng, J., Jiang, M., Bai, G., 2015. An integrated strategy of marker ingredients searching and near infrared spectroscopy rapid evaluation for the quality control of Chinese eaglewood. J. Pharm. Biomed. Anal. 114, 462–470. Ding, G., Wang, Y., Liu, A., Hou, Y., Zhang, T., Bai, G., Liu, C., 2017. From chemical markers to quality markers: an integrated approach of UPLC/Q-TOF, NIRS and chemometrics for the quality assessment of honeysuckle buds. RSC Adv. 7, 22034–22044. Fang, R., Cui, Q., Sun, J., Duan, X., Ma, X., Wang, W., Cheng, B., Liu, Y., Hou, Y., Bai, G., 2015. PDK1/Akt/PDE4D axis identified as a target for asthma remedy synergistic with β2AR agonists by a natural agent arctigenin. Allergy 70, 1622–1632. Feng, Y., Wu, Z., Zhou, X., Zhou, Z., Fan, W., 2006. Knowledge discovery in traditional Chinese medicine: state of the art and perspectives. Artif. Intell. Med. 38, 219–236. Gu, S., Pei, J., 2017. Chinese herbal medicine meets biological networks of complex diseases: a computational perspective. Evid. Based Complement Altern. Med. 2017, 7198645. Han, X., Liu, X., Wang, L., Shi, N., Yu, W., Wang, Y., Lv, A., 2011. Standardization strategy on modernization and internationalization of traditional Chinese medicine. World Sci. Technol./Mod. Tradit. Chin. Med. Mater. Med. 3, 737–742. Han, Y., Zhou, M., Wang, L., Ying, X., Peng, J., Jiang, M., Bai, G., Luo, G., 2015. Comparative evaluation of different cultivars of Flos Chrysanthemi by an anti-inflammatory based NF-kB reporter gene assay coupled to UPLC-Q/TOF MS with PCA and ANN. J. Ethnopharmacol. 174, 387–395. Hou, Y., Cheng, B., Zhou, M., Fang, R., Jiang, M., Hou, W., Bai, G., 2014. Searching for synergistic bronchodilators and novel therapeutic regimens for chronic lung diseases from a traditional Chinese medicine, Qingfei Xiaoyan Wan. PLoS One 9, e113104. Jiang, C., Qu, H., 2015. A comparative study of using in-line near-infrared spectra, ultraviolet spectra and fused spectra to monitor Panaxnoto ginseng adsorption process. J. Pharm. Biomed. Anal. 102, 78–84. Jiang, Z., Wang, Y., 2016. A pattern of hierarchical progression for quality standard of Chinese material medica based on “herbal origin-material basis-quality markersquality control method”. Chin. Tradit. Herb. Drugs 47, 4127–4133. Kao, T.C., Wu, C.H., Yen, G.C., 2014. Bioactivity and potential health benefits of licorice. J. Agric. Food Chem. 62, 542–553. Kasemsumran, S., Apiwatanapiwat, W., Suttiwijitpukdee, N., Vaithanomsat, P.,

7

Phytomedicine xxx (xxxx) xxx–xxx

G. Bai et al.

Xiong, H., Gong, X., Qu, H., 2012. Monitoring batch-to-batch reproducibility of liquidliquid extraction process using in-line near-infrared spectroscopy combined with multivariate analysis. J. Pharm. Biomed. Anal. 70, 178–187. Xiong, L., Peng, C., 2016. Study on Q-Marker of Leonurus japonicus and Penthorum chinense based on basic conditions of Q-Marker. Chin. Tradit. Herb. Drugs 47, 2212–2220. Ying, X., Pei, Y., Liu, M., Ding, G., Jiang, M., Liang, Q., Wang, Y., Bai, G., Luo, G., 2014. Discrimination and quantification analysis of Acoruscalamus L. and Acorustatarinowii Schott with near-infrared reflection spectroscopy. Anal. Methods 6, 4212–4218. Yoshida, M., Haraguchi, T., Uchida, T., 2014. Bitterness evaluation of acidic pharmaceutical substances (NSAIDs) using a taste sensor. Chem. Pharm. Bull. (Tokyo) 62, 1252–1258. Zhang, C., Su, J., 2014. Application of near infrared spectroscopy to the analysis and fast quality assessment of traditional Chinese medicinal products. Acta Pharm. Sin. B 4, 182–192. Zhang, H., Wu, T., Zhang, L., Zhang, P., 2016a. Development of a portable field imaging spectrometer: application for the identification of sun-dried and sulfur-fumigated Chinese herbals. Appl. Spectrosc. 70, 879–887. Zhang, T., Wang, J., Cheng, C., Xu, J., Gong, S., Han, Y., Zhang, H., Liu, C., 2017. Method of quality marker research and quality evaluation of Chinese material medica based on drug properties and effect characteristics. Chin. Tradit. Herb. Drugs 48, 1051–1060. Zhang, N.D., Han, T., Huang, B.K., Rahman, K., Jiang, Y.P., Xu, H.T., Qin, L.P., Xin, H.L., Zhang, Q.Y., Li, Y.M., 2016b. Traditional Chinese medicine formulas for the treatment of osteoporosis: implication for antiosteoporotic drug discovery. J. Ethnopharmacol. 189, 61–80. Zhao, Y.X., Li, M.L., 2002. Application of pattern recognition to quality assessment of the traditional Chinese medicine. Chin. J. Chin. Mater. Med. 27, 808–811. Zhou, M., Ma, X., Ding, G., Wang, Z., Liu, D., Tong, Y., Zhou, H., Gao, J., Hou, Y., Jiang, M., Bai, G., 2017. Comparison and evaluation of anti-muscarinic and anti-inflammatory effects of five Bulbusfritillariae species based on UPLC-Q/TOF integrated dual-luciferase reporter assay, PCA and ANN analysis. J. Chromatogr. B. 1041-1042, 60–69. Zhou, Z., Li, Y., Zhang, Q., Shi, X., Wu, Z., Qiao, Y., 2016. Comparison of ensemble strategies in online NIR for monitoring the extraction process of Pericarpium Citri Reticulatae based on different variable selections. Planta Med. 82, 154–162. Zhu, J., Fan, X., Cheng, Y., Agarwal, R., Moore, C.M., Chen, S.T., Tong, W., 2014. Chemometric analysis for identification of botanical raw materials for pharmaceutical use: a case study using Panaxnoto ginseng. PLoS One 9, e87462.

size of risperidone microspheres by improved chemometric model. Int. J. Pharm. 472, 296–303. Sun, R., Li, X., Wang, L., Xu, N., Lyu, L., 2016. A research on reasonable identification and scientific control of quality marker (Q-marker) based on efficacy-toxicity correlation. World Sci. Technol./Mod. Tradit. Chin. Med. Mater. Med. 18, 1224–1231. Sun, S., Chen, J., Zhou, Q., Lu, G., Chan, K., 2010. Application of mid-infrared spectroscopy in the quality control of traditional Chinese medicines. Planta Med 76, 1987–1996. Vestbo, J., Hurd, S.S., Agustı´, A.G., Jones, P.W., Vogelmeier, C., Anzueto, A., Barnes, P.J., Fabbri, L.M., Martinez, F.J., Nishimura, M., Stockley, R.A., Sin, D.D., RodriguezRoisin, R., 2013. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am. J. Respir. Crit. Care Med. 187, 347–365. Wang, P., Zhang, H., Yang, H., Nie, L., Zang, H., 2015. Rapid determination of major bioactive isoflavonoid compounds during the extraction process of kudzu (Puerarialobata) by near-infrared transmission spectroscopy. Spectrochim. Acta A Mol. Biomol. Spectrosc. 137, 1403–1408. Wang, X., Zhang, H., Chen, L., Shan, L., Fan, G., Gao, X., 2013. Liquorice, a unique "guide drug" of traditional Chinese medicine: a review of its role in drug interactions. J. Ethnopharmacol. 150, 781–790. Wang, Y., Mei, M., Ni, Y., Kokot, S., 2014. Combined NIR/MIR analysis: a novel method for the classification of complex substances such as Illiciumverum Hook. F. and its adulterants. Spectrochim. Acta A Mol. Biomol. Spectrosc. 130, 539–545. Wang, Z., Cui, Y., Ding, G., Zhou, M., Ma, X., Hou, Y., Jiang, M., Liu, D., Bai, G., 2017. Mahuannin B an adenylate cyclase inhibitor attenuates hyperhidrosis via suppressing β2-adrenoceptor/cAMP signaling pathway. Phytomedicine 30, 18–27. Wu, Z., Sui, C., Xu, B., Ai, L., Ma, Q., Shi, X., Qiao, Y., 2013. Multivariate detection limits of on-line NIR model for extraction process of chlorogenic acid from Lonicera japonica. J. Pharm. Biomed. Anal. 77, 16–20. Wu, Z., Xu, B., Du, M., Sui, C., Shi, X., Qiao, Y., 2012. Validation of a NIR quantification method for the determination of chlorogenic acid in Lonicera japonica solution in ethanol precipitation process. J. Pharm. Biomed. Anal. 62, 1–6. Xin, N., Gu, X.F., Wu, H., Hu, Y.Z., Yang, Z.L., 2012a. Discrimination of raw and processed Dipsacus asperoides by near infrared spectroscopy combined with least squaressupport vector machine and random forests. Spectrochim. Acta A Mol. Biomol. Spectrosc. 89, 18–24. Xin, N., Gu, X., Wu, H., Hu, Y., Yang, Z., 2012b. Application of genetic algorithm-support vector regression (GA-SVR) for quantitative analysis of herbal medicines. J. Chemometr. 26, 353–360.

8