Original Papers

Effect of Processing on the Traditional Chinese Herbal Medicine Flos Lonicerae: An NMR-based Chemometric Approach

Authors

Jianping Zhao 1, Mei Wang 1, Bharathi Avula 1, Lingyun Zhong 4, Zhonghua Song 5, Qiongming Xu 6, Shunxiang Li 7, Ikhlas A. Khan 1, 2, 3

Affiliations

The affiliations are listed at the end of the article

Key words " Lonicera japonica l " Caprifoliaceae l " processing l " TCM l " NMR l " PCA l " chemometric l

Abstract

received revised accepted

June 29, 2014 March 31, 2015 April 19, 2015

Bibliography DOI http://dx.doi.org/ 10.1055/s-0035-1546081 Published online June 3, 2015 Planta Med 2015; 81: 754–764 © Georg Thieme Verlag KG Stuttgart · New York · ISSN 0032‑0943 Correspondence Prof. Dr. Ikhlas A. Khan National Center for Natural Products Research University of Mississippi University, MS 38677 USA Phone: + 1 66 29 15 78 21 Fax: + 1 66 29 15 79 89 [email protected]

Zhao J et al. Effect of Processing …

!

The processing of medicinal materials, known as Pao Zhi in traditional Chinese medicine, is a unique part of traditional Chinese medicine and has been widely used for the preparation of Chinese materia medica. It is believed that processing can alter the properties and functions of remedies, increase medical potency, and reduce toxicity and side effects. Both processed and unprocessed Flos Lonicerae (flowers of Lonicera japonica) are important drug ingredients in traditional Chinese medicine. To gain insights on the effect of processing factors (heating temperature and duration) on the change of chemical composition, nuclear magnetic resonance combined with chemometric analysis was applied to investigate the processing of F. Lonicerae. Nuclear magnetic resonance spectral data were analyzed by means of a heat map and principal components analysis. The results indicated that the composition changed significantly, particularly when processing at the higher temperature (210 °C). Five chemical components, viz. 3,4-dicaffeoylquinic acid, 4,5-dicaffeoylquinic acid, 3,5-dicaffeoylquinic acid, chlorogenic acid, and myo-inositol, whose concentration changed significantly during the processing, were isolated and identified. The patterns for the concentration change observed from nuclear magnetic resonance analysis during the processing were confirmed and quantitatively deter-

Introduction !

Lonicera japonica Thunb. (family Caprifoliaceae) is a perennial plant with showy, fragrant flowers, native to East Asia. The flower, known as “Jin Yin Hua” (JYH) or “Ren Dong Hua” in TCM, is one of the most commonly used ingredients in the herbal formulas of traditional remedies. “Shen Nong Ben Cao Jin” (Divine Farmerʼs Materia Medica), a

Planta Med 2015; 81: 754–764

mined by ultrahigh-performance liquid chromatography analysis. The study demonstrated that a nuclear magnetic resonance-based chemometric approach could be a promising tool for investigation of the processing of herbal medicines in traditional Chinese medicine.

Abbreviations !

APCI:

atmospheric pressure chemical ionization DAD: diode array detector DCQA: dicaffeoylquinic acid HRESI‑MS: high-resolution electrospray ionization mass spectroscopy HSQC: heteronuclear single quantum coherence JYH: Jin Yin Hua (flowers of Lonicera japonica) MS: mass spectroscopy NMR: nuclear magnetic resonance PCA: principal component analysis PC1: first principal component PC2: second principal component SARS: severe acute respiratory syndrome TCM: traditional Chinese medicine UHPLC: ultrahigh-performance liquid chromatography 2D NMR: two-dimensional NMR

famous classical book of Chinese materia medica written more than 3000 years ago, recorded this plant and the usage of its flowers for the treatment of diseases. The cultivation history of L. japonica can be traced back to more than 1500 years [1]. Currently, the plant is widely cultivated in China to meet the needs of the TCM pharmaceutical industry [2]. In TCM theory, JYH is classified with a “cold” nature [3], and the “cold” desig-

Downloaded by: Karolinska Institutet. Copyrighted material.

754

Original Papers

to support its use. Accumulating evidence has demonstrated that most changes of chemical compositions in Chinese materia medica occurred during processing, and thus produced different curative effects [17]. It is important to investigate and understand the role or mechanism of Pao Zhi in TCM, because the safety, efficacy, and quality of TCM prescriptions are closely related to the chemical content that may be altered when the materia medica ingredients are processed. In addition, due to the lack of the rigorous regulations for most of the traditional processing methods which are currently used in the TCM pharmaceutical industry, standardization of the processing procedures is urgently needed for modernization of classical TCM. Aiming at those ultimate objectives, the present study was performed using an NMR-based metabolomic approach to investigate the effects of different processing temperatures and heating durations on the change of chemical composition in JYH. The intention of the study was to gain insights into the relationship of the change of chemical composition to the processing conditions, and to provide information that could be valuable for establishing standard procedures for JYH processing.

Results and Discussion !

Heating temperature and duration are two of the most important factors in JYH processing. These two parameters directly affect the quality of the processed products, hence the efficacy of the final pharmaceutical products. However, in the traditional practice of JYH processing, there are no accurate specifications for these parameters. Instead, two rough temperature levels, “Wen Huo” and “Wu Huo”, are used to indicate the temperature of the heat source for JYH processing. Meanwhile, the heating duration is mainly determined by observing the color change of the JYH material being heated, or by the personal experience of various processing operators. In the present study, JYH samples processed at two temperatures (120 °C and 210 °C) with different heating durations were inves" Table 1. The tigated. The experimental conditions are shown in l change of chemical composition during JYH processing was first evaluated by means of NMR. NMR spectroscopy is one of the most powerful chemical analytical tools and is widely used for characterization and structure determination of natural products. NMR spectra can provide qualitative and quantitative information regarding the components of a mixture such as a plant " Fig. 1 shows typical 1H NMR spectra of methaextract [18, 19]. l nolic extracts from the unprocessed and two processed samples. In comparison with the spectrum of the unprocessed JYH sample, the changes of signal intensity in the spectra of processed JYH samples can be observed throughout the entire chemical shift range, particularly for the one processed under the higher tem" Fig. 1 C). To facilitate information mining, a perature (210 °C) (l " Fig. 2) was generated from the NMR dataset to illusheat map (l trate the changes of intensity (relative to that of the unprocessed) for the signals in the spectral range from 0.7 to 8.0 ppm. Both the decreasing (green) and the increasing (red) signal intensities were observed for all of the stages of processing, suggesting that the chemical reactions and/or conversions of the compounds occurred during the processing procedure. Significant changes of chemical composition occurred when JYH was processed at the higher temperature (210 °C) for only 14 min, which were much more significant than those observed when the sample was processed at 120 °C for 80 min. These results suggest that tempera-

Zhao J et al. Effect of Processing …

Planta Med 2015; 81: 754–764

Downloaded by: Karolinska Institutet. Copyrighted material.

nation specifically refers to its antitoxin, antipyretic, and anti-inflammatory properties. JYH is notable for being used to address what are so-called excess heat conditions such as fevers, skin rashes, ulcers, and sore throat. In 2003, JYH was used as one of the most effective TCM ingredients to treat SARS coronavirus [1, 4]. Modern pharmacological studies have shown that JYH also possesses various biological activities including antibacterial [5], antiviral [6, 7], antioxidant [8], anti-inflammatory [9], antinociceptive [10], and immune enhancement [6]. As an important ingredient, JYH is used in many patented TCM prescriptions, including the well-known ones such as “Yin Qiao Jie Du Pian”, “Ying Huang Kou Fu Ye”, “Xiao Yin Pian”, and “Shuang Huang Lian Shuan”, all of which are listed in the Chinese Pharmacopoeia and are widely used in clinical practice. More than 500 prescriptions traditionally used in China to treat various diseases contain JYH [1]. Two forms of JYH materia medica, viz. unprocessed (natural and dry flowers, called “Sheng Yao” in TCM) and processed (stir-fried flowers), are used in TCM clinical practice for the treatment of different symptoms [11]. Traditionally, the processed form is obtained by means of stir-frying from the unprocessed raw material. Depending upon the treatment degree, the processed JYH can be further classified into two types, called “Chao Yao” (fried drug) and “Tan Yao” (charred drug). In accordance with the traditional technique for JYH processing, Chao Yao is produced by stir-frying the flowers above a moderate fire (“Wen Huo”) to make them become yellow in color; Tan Yao is produced by stir-frying above a very hot fire (“Wu Huo”) to make them become dark in color. In TCM, it is believed that the different processed forms of JYH materia medica possess different therapeutic functions and properties. For instance, the unprocessed JYH is usually used to dispel the “heat-evil” and treat diseases by acting on the “Upper Jiao” (Jiao is a term found in TCM theory, referring to the organs or functional body parts that maintain the health of human bodies), while Chao Yao and Tan Yao are mainly used to treat diseases by acting on the “Middle Jiao” and “Lower Jiao”, respectively. Many studies have been conducted to understand the role of JYH processing in its therapeutic function and clinical applications [12]. Investigations of the different effects of unprocessed and processed JYH on blood coagulation in vivo and the difference in their antibacterial activities in vitro were conducted by Nan et al. [13]. The results showed that the blood coagulation time was significantly shorter for the mouse group treated with the processed JYH than the group treated with the unprocessed JYH, and the extract of the processed JYH had a stronger antibacterial effect than the unprocessed JYH. The anti-inflammatory effects of the decoctions of the unprocessed and the processed JYH were studied using the models of xylene-induced mouse ear edema and albumin-induced rat paw edema [14]. It was found that the anti-inflammatory activity of the unprocessed JYH was stronger than the processed form. Concentration differences for some targeted components, such as chlorogenic acid, tannin, inorganic elements, and polysaccharides, were observed in different forms of JYH [15, 16]. Many herbs used in TCM are subjected to specific treatments (such as frying, boiling, steaming, fermenting, etc.) before they are used in clinical practices, and the method used to process Chinese medicinal materials has the specific term “Pao Zhi”. It is a specific item in the Chinese Pharmacopeia and a unique characteristic of TCM practice. In 2006, Pao Zhi was designated as one of the national intangible culture heritages of China. Although it has been practiced in China for centuries, Pao Zhi is still one of the esoteric parts of TCM and additional scientific evidence is needed

755

Original Papers

Table 1 Processing conditions for the investigated samples. Temperature Duration (min.) Sample code Temperature Duration (min.) Sample code

120 °C 5 A-5 210 °C 2 B-2

10 A-10

15 A-15

20 A-20

30 A-30

40 A-40

50 A-50

60 A-60

80 A-80

3 B-3

4 B-4

5 B-5

7 B-7

9 B-9

11 B-11

14 B-14

Unprocessed N-0

Fig. 1 1H NMR spectra of JYH extracts. A Unprocessed; B processed at 120 °C for 30 min; C processed at 210 °C for 11 min.

Downloaded by: Karolinska Institutet. Copyrighted material.

756

Fig. 2 Heat map for visualizing the changes of 1HNMR signal intensity for JYH samples processed at 120 °C (A series) and 210 °C (B series). The changes relative to the unprocessed (N-0) sample are represented using a red-to-

Zhao J et al. Effect of Processing …

Planta Med 2015; 81: 754–764

green scaled color map. Red represents an increase and green a decrease. (Color figure available online only.)

Original Papers

757

Table 2 Chemical shifts list (ppm) of signals whose intensities changed significantly during the processing. Decrease

Processing temperature Chemical shift (ppm)

120 °C 1.47 2.63–2.69 2.95–2.97

3.67–3.69 3.75 3.89–3.91 4.01–4.05 4.49 4.59 4.63 4.67–4.69 5.23–5.27 5.31 5.45–5.57 5.69–5.77 6.29 6.39 6.43 6.79 6.97 7.07–7.09 7.33–7.55 7.41–7.45 7.57

a

Increase 210 °C 1.47 2.63–2.69 2.95–2.97 3.43–3.45 3.63–3.65 3.67–3.69 3.73–3.85 3.89–3.91 4.01–4.05 4.09–4.11 4.49a 4.59 4.63 4.67–4.69 5.23–5.27 5.31 5.39–5.41 5.45–5.57 5.61–5.65 5.69–5.83 6.27–6.31 6.39 6.43 6.77–6.79 6.93–6.97 7.05–7.09 7.33–7.55 7.41–7.45 7.55–7.63

120 °C 0.75–0.77

2.03–2.09 2.19–2.23 2.41–2.43 2.55–2.57 2.77–2.85 2.89

4.13–4.15 4.19–4.43 4.51–4.57 4.73–4.77 5.33–5.35 5.89 6.01–6.25 6.35–6.37 6.45–6.69 6.73–6.75 6.87–6.91 6.99–7.03 7.15–7.25 7.53 7.75–7.95

210 °C 0.73–0.77 0.93–0.99 1.17–1.25 1.29–1.37 1.61–1.67 1.79–1.87 1.94 2.03–2.09 2.19–2.23 2.41–2.43 2.47–2.49 2.55–2.59 2.77–2.85 2.89b 3.55–3.57 3.99 4.07 4.13–4.15 4.19–4.47 4.51–4.57 4.71 4.73–4.79 5.33–5.35 5.89–5.93 6.01–6.25 6.35–6.37 6.45–6.69 6.59b 6.73–6.75b 6.87–6.91b 6.99–7.03b 7.15–7.25 7.53b 7.75–7.95

The signal intensity decreases first, then increases along the course of processing; b the signal intensity increases first, then decreases along the course of processing

ture is a vital factor in JYH processing. The chemical shifts of the signals whose intensities changed significantly during processing " Table 2. It was found that the intensities of many are listed in l signals within the sugar region (3.0–5.0 ppm) decreased and many within the fatty/amino acid region (0.7–3.0 ppm) increased, while those within the aromatic region (6.0–8.0 ppm) displayed both decreasing and increasing intensity. To date, more than 90 chemical constituents have been isolated and identified from JYH flowers [1]. The identified compounds represent different chemical categories including organic acids, flavonoids, terpenoids, saponins, and iridoids. The complexity of the chemical composition makes it challenging to assign the signals in the NMR spectra to their corresponding chemical components, especially for those generated during the processing. Nevertheless, with the aid of multivariate analysis, it was possible to obtain additional information on the overall chemical composition changes during the procedure of JYH processing from the NMR data without assignment of the individual resonance frequencies. In the present study, a PCA was performed for the data set derived from the 1H NMR spectra of the JYH samples. PCA is an unsupervised pattern recognition method that requires no prior knowledge of the data set and acts to reduce a large number of original variables to a smaller number of principal compo-

" Fig. 3 shows the scores scatter plot of PC1 versus PC2 nents. l for the PCA model. The first two principal components accounted for 91.4 % of the total variance within the spectral data set. The mean coordinates expressed in the PCA plot are produced for each group of samples processed under the same heating temperature and duration, and the connection of the mean coordinates for a certain heating temperature in the order of processing generates a developmental trajectory. Two distinct develop" Fig. 3) were illustrated in the mental trajectories (T-1 and T-2, l plot. As the nearness/farness between the samples groups in the PCA scores plot is related to the difference of their chemical compositions, T-1 and T-2 trajectories may summarize the changes of JYH composition throughout the processing procedure for the two different heating temperatures (120 °C and 210 °C, respec" Fig. 3, for the case that JYH was protively). As illustrated in l cessed at 120 °C, the trajectory (T-1) reflected that the chemical composition changed immediately after JYH was heated for 5 min, and the change slowed down after heating for 10 min. The sample groups processed for 15, 20, 30, and 40 min were closely located in the plot, indicating that the composition did not change dramatically at those processing stages, or the change might reach a plateau when the processing duration was in the range from 15 min to 40 min. Significant changes occurred con-

Zhao J et al. Effect of Processing …

Planta Med 2015; 81: 754–764

Downloaded by: Karolinska Institutet. Copyrighted material.

Intensity change

Original Papers

Fig. 3 PCA scores plot of PC1 versus PC2 for unprocessed (N-0) and processed samples [A series (green) – processed at 120 °C; B series (blue) – processed at 210 °C]. The two solid lines (red line for T-1 and black line for T-2)

tinuously after processing for the longer heating durations (50, 60, and 80 min). For the case of the JYH samples processed at 210 °C, the trajectory (T-2) shows overlaps with T-1 at its short time stages where the processing was carried out for 2 and 3 min. The closeness of the locations between the sample group B-3 (processed at 210 °C for 3 min) and the sample group A-80 (processed at 120 °C for 80 min) indicates that the JYH chemical composition changed more quickly when processed at the higher temperature. The trajectories clearly illustrate the significance of temperature in JYH processing. The trajectory T-2 diverges to some degree along with the PC1 axis at the stages where the processing occurred for 3–7 min. At longer processing times (> 7 min), T-2 deviates mainly along the PC2 axis. The overall geometry form of trajectory T-2 is U-shaped. The samples (B-14) that were processed for 14 min at 210 °C appear far from the group of unprocessed samples, implying a significant change of chemical composition occurred. B-14 samples could be classified as Tan Yao based on their dark appearance. " Fig. 4 A, B shows the PCA loading plots for PC1 and PC2, respecl " Fig. 3 has a larger span along tively. Since the trajectory T-2 in l the PC1 axis in comparison with the trajectory T-1, the PC1 loadings can be used to measure the importance of each variable

Zhao J et al. Effect of Processing …

Planta Med 2015; 81: 754–764

that connect the mean of PC scores for each group of samples represent the developmental trajectories for 120 °C and 210 °C processing, respectively. (Color figure available online only.)

(NMR peaks) in accounting for the variability (change of composition) when JYH was processed under the higher temperature (210 °C). The trajectory T-1 and the short time stages of T-2 mainly involved changes along the PC2 axis, therefore the PC2 loadings can reflect the important variables when JYH was processed at the lower temperature (120 °C) or in the early stages of the processing undertaken at the higher temperature (210 °C) for short durations. The positive peaks in the loading plots correspond to the NMR signals of the compounds whose concentration levels decreased during the processing and vice versa. The " Taloading data showed consistency with the results given in l ble 2. A close examination of those positive peaks within the " Fig. 4 A, combining with the HSQC 3.2–4.1 ppm region in l 2D‑NMR analyses, revealed that these peaks were mostly attributable to the protons attached to oxygenated carbons. Furthermore, the negative peaks within 0.7–2.8 ppm were mostly attributed to methylene protons, and both the positive and negative peaks within 6.2–7.6 ppm corresponded to olefinic protons. The results indicated that some constituents were converted to other compounds, and chemical reactions occurred when JYH materials were processed.

Downloaded by: Karolinska Institutet. Copyrighted material.

758

759

Downloaded by: Karolinska Institutet. Copyrighted material.

Original Papers

Fig. 4

PCA loading plots of PC1 (A) PC2 (B).

Zhao J et al. Effect of Processing …

Planta Med 2015; 81: 754–764

Original Papers

Fig. 5

In order to verify these NMR signals whose intensities changed significantly during the processing and to identify their corresponding compounds, column chromatographic isolations with the guidance of NMR analysis were conducted. Five compounds were isolated and characterized in the present study, namely 3,4-dicaffeoylquinic acid (1), 4,5-dicaffeoylquinic acid (2), 3,5-dicaffeoylquinic acid (3), chlorogenic acid (4), and myo-inositol (5). Of those compounds, 1 and 2 were isolated from the JYH material processed under 120 °C for 30 min, 3 and 4 from the unprocessed JYH material, and 5 from the material processed under 210 °C for 11 min. The structures of these five compounds are given in " Fig. 5. Their identities were determined by NMR and HRMS l analyses. Compounds 1–5 have been identified from JYH in previous studies [20–22]. Chlorogenic acid and DCQAs were considered to be the important active compounds responsible for various biological activities of JYH [1]. Inositol was found in many plants and has been considered a member of the vitamin B complex. The five isolated compounds were used to further explore the effect of processing temperature and duration on the change of " Figs. 6 and 7 illuschemical composition by UHPLC analyses. l trate the results from the UHPLC analyses for processing at 120 °C and 210 °C, respectively. Interestingly, compounds 1 and 2 showed similar pattern for the concentration changes. In the case of processing at 120 °C, the concentrations for both compounds increased as the heating duration increased. The concentrations of compounds 1 and 2 increased from 0.81 ± 0.03 (mean ± standard deviation) and 4.0 ± 0.2 mg/g to 1.4 ± 0.1 and 10.3 ± 0.4 mg/ g, respectively, after being processed for 80 min. About twofold increases were observed for the two compounds. In the case of processing at 210 °C, the concentrations of compounds 1 and 2 increased initially, and then decreased when the heating time was longer than 4 min. Compounds 3 and 4 showed a similarity in the pattern of concentration changes, but the changes were different from those of compounds 1 and 2. The concentrations of compounds 3 and 4 declined slightly during processing from 0 to 80 min when processed at 120 °C. When processed at 210 °C, their concentrations dropped significantly after JYH was heated for 2 min. The concentrations of compounds 3 and 4 in the unprocessed JYH material were 17 ± 2 and 45 ± 4 mg/g, respectively. They changed to 11.0 ± 0.8 and 45 ± 2 mg/g after heating for 80 min at 120 °C, but dropped to 1.3 ± 0.1 and 21 ± 1 mg/g after heating for 11 min at 210 °C, respectively. Unlike the other four compounds, the concentration of compound 5 decreased slightly from 53 ± 2 mg/g to 45 ± 2 mg when processed for 80 min at 120 °C; but when processed at 210 °C, the content went down first to as low as 43 ± 3, then increased to 89 ± 5 mg/g after heating for 11 min.

Zhao J et al. Effect of Processing …

Planta Med 2015; 81: 754–764

Structures of the isolated compounds.

All of the compounds did not show significant changes in concentration within the processing time window from 15 to 40 min " Fig. 6). Interestingly, as when JYH was processed at 120 °C (l mentioned above, based on the observation of the T-1 trajectory in the PCA scores plot, the JYH composition did not change significantly within the same processing time window (from 15 to 40 min). The results confirmed that the PCA trajectory could be used to reflect the chemical composition change situation. To the best of our knowledge, the present study is the first to report that the contents of compounds 1 and 2 in JYH could be increased through heat processing. The decrease of content could be explained as the result of decomposition or degradation of the compounds when heat was applied. But as for the twofold content increases of compounds 1 and 2 when processed under certain conditions, the mechanism remains unclear. The loss of water content (< 5 % in unprocessed material) in JYH material after being heated is not sufficient to account for such significant concentration increases. Chemical transformation, release of free molecules from complex molecules, or other unknown chemical reactions might be involved to cause the content increase. As for compound 5, we speculated that the increasing amount might come from the degradation of phytate (inositol hexakisphosphate) to inositol. In conclusion, an NMR methodological approach was used in the present study to explore the changes of chemical composition in association with the effects of heating temperatures and times for JYH processing. In principle, any components containing hydrogen can be detected by 1H NMR. This advantage provides unbiased detection compared to other techniques. NMR can be applied to compounds of all organic chemical classes, and allows direct comparisons of the quantities of the compounds present in the sample [23]. The heat map derived from the NMR spectral data was used to pinpoint the intensity changes (increase or decrease) of the signals that changed significantly in the different stages of processing. The PCA of the NMR data set could provide a holistical view for the evaluation of the chemical composition change during JYH processing, and could facilitate the determination of the signals that changed significantly in intensity during the processing. The PCA developmental trajectory could be potentially applicable in other Pao Zhi studies for TCM to summarize the effects of processing temperature and duration, the two most significant factors associated with the quality of processed products. In TCM, JYH is traditionally processed under two different heating conditions (“Wen Huo” and “Wu Huo”), and two different forms of processed products (“Chao Yao” and “Tan Yao”) are used in the practice. The current study demonstrated that the heating temperature is a crucial factor in JYH processing. When processed at the higher temperature (210 °C), the chemical composition in

Downloaded by: Karolinska Institutet. Copyrighted material.

760

Original Papers

761

Downloaded by: Karolinska Institutet. Copyrighted material.

Fig. 6 Content changes of compounds 1–5 in JYH processed at 120 °C with different durations. A 3,4dicaffeoylquinic acid (1); B 4,5-dicaffeoylquinic acid (2); C 3,5-dicaffeoylquinic acid (3); D chlorogenic acid (4); E myo-inositol (5).

JYH changed significantly after heating only for a short period of time. When processed at 120 °C, the change of composition was not significant when the processing time varied from 15 to 40 min. According to the TCM theory, it is believed that Pao Zhi can alter the properties and functions of herbs, increase medical potency, and reduce toxicity and side effects. In the present study, five compounds were isolated by using column chromatography and their structures were identified with the aid of spectroscopic analysis. The changes of their content related to the different processing conditions were investigated by a UHPLC method. Different patterns in content change for the five compounds were observed along the course of processing. The evidence on the changes of content for the five compounds might reflect, to some extent, the change of chemical composition in JYH under different processing conditions. It is certain that except for those five compounds, some other compounds might change in quantity during the processing based on NMR analyses. In addition, some new compounds might be generated as a result of processing. Further studies are needed to investigate the changes of chemical

composition caused by JYH processing to the biological activity variations, and to gain insights on the mechanism of JYH processing.

Materials and Methods !

Plant material The dry flowers of L. japonica Thunb. (JYH) (Batch number: MM#5144SF) were supplied by Kalyx. The authentication of materials was verified by Dr. Vijayasankar Raman at the University of Mississippi. The processed JYH samples were prepared according to the procedures described in the Chinese Pharmacopeia [3]. A CBR-101 coffee roaster equipped with temperature control and time programming functions was used for the processing. The sample information and associated processing conditions are " Table 1. Triplicate experiments were performed presented in l for each processing condition. Voucher specimens of all the samples (#5208) were deposited at the National Center for Natural Products Research (NCNPR), University of Mississippi, Mississip-

Zhao J et al. Effect of Processing …

Planta Med 2015; 81: 754–764

Original Papers

Fig. 7 Content changes of compounds 1–5 in JYH processed at 210 °C with different processing durations. A 3,4-dicaffeoylquinic acid (1); B 4,5-dicaffeoylquinic acid (2); C 3,5-dicaffeoylquinic acid (3); D chlorogenic acid (4); E myo-inositol (5).

Downloaded by: Karolinska Institutet. Copyrighted material.

762

pi, USA. The samples were vacuum-dried and pulverized into a fine powder using an IKA grinder to be analyzed.

Chemicals and reagents All chemical reagents used in this study were of analytical grade. Deuterated methanol (CD3OD, 99.8 % D), water (D2O, 99.9 %), chloroform (CDCl3, 99.8 %), dimethyl sulfoxide (DMSO-d6, 99.9 %), and acetone (CD3 COCD3, 99.9 %) were purchased from Cambridge Isotope Laboratories, and sodium 3-trimethylsilyl [2,2,3,3-2H4] propanoate (TSP) was obtained from Sigma-Aldrich, Inc. The reference standards, 3,4-dicaffeoylquinic acid, 3,5-dicaffeoylquinic acid, and 4,5-dicaffeoylquinic acid, which were used for the UHPLC analysis, were purchase from ChromDex, Inc. Myo-insitol and chlorogenic acid were obtained from Sigma-Aldrich, Inc. HPLC grade acetonitrile was also purchased from Sigma-Aldrich.

Zhao J et al. Effect of Processing …

Planta Med 2015; 81: 754–764

Sample preparation for nuclear magnetic resonance analysis JYH extracts (from unprocessed and processed ones) were obtained by using different solvent systems, i.e., MeOH, MeOH/H2O (1 : 1), MeOH/CHCl3 (1 : 1), DMSO, and acetone, and were first analyzed by NMR. Among the 1H NMR spectra of all the extracts, the spectra of the methanolic extracts presented the richest information regarding the number and intensity of resonant signals as well as the intensity change of resonant signals when comparing the spectrum of the unprocessed with that of the processed. Thus, methanol was used as the extractant for the NMR study. A powdered sample (200 mg) was vortexed with 0.6 mL dueterated methanol containing 0.03 % (w/v) TSP, and sonicated at 25 °C for 12 min. The mixture was centrifuged at 13 400 rpm for 9 min, and the supernatant was transferred to a 3-mm NMR tube for NMR measurement.

Nuclear magnetic resonance measurements

Isolation and identification of compounds 1–5

NMR experiments were performed at 25 °C on an Agilent DD2– 500 NMR spectrometer equipped with a oneNMR probe and Agilent Vnmrj 3.2 software. The 1H NMR spectra were measured at 499.79 MHz, and the H2O signal was suppressed using a presaturation pulse program. For each sample, 64 transients were collected into 64 K data points. The spectra were recorded with the following parameters: spectral width = 6010 Hz, relaxation delay = 5 s, acquisition time = 5.453 s, pulse width = 7.20 µs (90°). Free induction delays (FIDs) were Fourier transformed with 0.3 Hz line broadening. The spectra were manually phased and baseline corrected using Mnova software (v. 5.2, Mestrelab Research S. L.). The 1H NMR spectra were calibrated to the signal of TSP at chemical shift δ = 0.00 ppm. The 13C spectra were measured at 125.67 MHz. The parameters for the 13C spectra were as follows: spectral width = 30 488 Hz, relaxation delay = 2 s, pulse width = 3.50 µs (45°). Chemical shifts reported were calibrated to the CD3OD solvent signal at δ 49.15 ppm. The 2D NMR spectra, including 1H-1H correlation spectroscopy (COSY), 1H-13C heteronuclear single quantum coherence (HSQC), and 1H-13C heteronuclear multiple bonds coherence (HMBC) spectra, were recorded for the isolated compounds. The measurements of 2D spectra were performed using the standard pulse sequences in the Vnmrj software. The optimized coupling constants for HSQC and HMBC were 145 and 8 Hz, respectively.

The isolations were guided by NMR monitoring on the targeted signals, which changed significantly during the processing. A total of 10.0 g of the JYH sample (A-30) processed at 120 °C for 30 min was ground to homogeneneity and extracted with methanol (15 mL) three times, yielding 3.26 g of extract. The extract was partitioned with ethyl acetate and water (1 : 1) three times, and the collected organic phase yielded 0.53 g of residue. The residue was loaded onto a Sephadex LH-20 column eluting with chloroform/methanol (5 : 1) to produce 12 fractions. Fractions 4 and 5 were determined to contain the targeted compounds, which showed the characteristic NMR signals at around 6.4 ppm. Purifications were conducted on SNAP C-18 columns with a Biotage Isolera chromatographic system (Biotage LLC) for these two fractions. Compound 1 (12.3 mg) was obtained from fraction 4, and confirmed to be 3,4-dicaffeoylquinic acid by analysis of the spectral data (1D and 2D‑NMR and HRESI‑MS) and comparison with the published spectral data [24]. Compound 2 (2.5 mg) was obtained from fraction 5, and identified to be 4,5-dicaffeoylquinic acid by the spectral data analysis and comparison with the literature data [20]. Compounds 3 and 4, identified as 3,5-dicaffeoylquinic acid and chlorogenic acid, respectively, were isolated from the unprocessed JYH material in a previous study [22]. Compound 5 was obtained from the processed JYH sample B-11 (processed at 210 °C for 11 min). Significant intensity changes were observed for signals in the 3.50–3.70 ppm range in the 1 H‑NMR spectrum of sample B-11, thus the isolation was guided to target these characteristic signals. A total 3.0 g of the B-11 powder was extracted with methanol/water (8 : 2) three times and yielded 1.51 g of residue after drying. The residue was subjected to chromatographic separations by using a Biotage Isolera System. The separation on a Flash M + 40 silica gel column with an elution solvent system of ethyl acetate/chloroform/methanol/ water (15 : 5 : 8 : 2) offered a fraction containing the targeted compound 5. The fraction was further purified on a new Flash M + 40 column eluted with chloroform/methanol/water (6.5 : 3.5 : 1) to give a fraction containing a high concentration of compound 5. After crystallization in methanol, the white crystals of compound 5 (3.9 mg) were obtained. By analysis of the spectral data (1D and 2D‑NMR and HRESI‑MS) and comparison with the literature data [25], compound 5 was identified as myo-inosital.

Nuclear magnetic resonance data analysis After phase and baseline corrections, the 1H NMR spectra were bucketed with equal bins in the width of 0.02 ppm between 0.60 to 9.80 ppm. The spectral regions from δ 4.80 to 5.20 ppm and 3.00 to 3.40 ppm were excluded to eliminate the effects of water suppression and the residual methanol signal. Each bin was integrated and normalized to the TSP peak area. The binned data were converted to an ASCII format. The mean intensity of the bins between 0.70 and 8.0 ppm from the triplicate data for each sample was compiled into an 18 × 326 matrix and used to construct a heat map summarizing the changes in signal intensity profiles during the course of JYH processing. Specifically, the mean intensities for all 326 bins were subtracted from the mean intensities for the unprocessed sample (N-0). The results were then divided by the mean intensities for N-0 to give the percentage changes. The values were plotted using a red-to-green color map to indicate an increase or decrease in relative intensity changes, respectively. (Color figure available online only.) PCA was carried out to look for compositional similarities and explore the overall NMR signal intensity changes during the JYH processing. The bin data between 0.6 and 9.80 ppm from all 54 1 HNMR spectra were compiled into a 54 × 420 matrix and subjected to PCA analysis using SIMCA‑P+ software (v. 12.0, Umetrics). Pareto scaling was used. PCA is a descriptive method used to reduce a larger dimensional data set into a smaller number of dimensions that represent sources of successively maximized variance of data. The main advantage of this approach was that the NMR spectra were treated as fingerprints, and all the changes in signal intensity were taken into account, giving an overall view of the data. Furthermore, with the illustration of the developmental trajectory in the PCA scores plot, the status and change of chemical composition along the processing course could be visualized.

Ultrahigh-performance liquid chromatography-diode array detector-mass spectroscopy analysis The ground powder samples (50 mg) were sonicated in 5 mL methanol/water (1 : 1) for 20 min followed by centrifugation for 15 min. The supernatant was transferred to 20 mL volumetric flasks. This procedure was repeated four times and the respective supernatants were combined. The final volume was fixed to 20 mL and the solution was diluted to an appropriate concentration for UHPLC analysis. An UHPLC‑DAD‑MS method was developed for the simultaneous analysis of the five marker compounds, viz. 3,4-dicaffeoylquinic acid (1), 4,5-dicaffeoylquinic acid (2), 3,5-dicaffeoylquinic acid (3), chlorogenic acid (4), and myo-insitol (5). The UHPLC system consisted of an Agilent 1290 Infinity series LC with a DAD, binary pump, vacuum degasser, ALS autosampler, and thermostated column compartment. The LC instrument was coupled to an Agilent 6120 quadrupole mass spectrometer with a dual APCI and ESI interface. The column was a Waters Acquity UPLC BEH C18 (2.1 × 150 mm, 1.7 µm) column. The column temperature was 30 °C.

Zhao J et al. Effect of Processing …

Planta Med 2015; 81: 754–764

763

Downloaded by: Karolinska Institutet. Copyrighted material.

Original Papers

764

The flow rate was 0.25 mL/min. The eluent was 5 % acetonitrile and water (0.05 % formic acid) programmed to 40% acetonitrile in 10 min, and then 95% in 12 min and held for 3 min. The posttime run was 8 min to re-equilibrate the column. The concentrations of 3,4-dicaffeoylquinic acid, 3,5-dicaffeoylquinic acid, 4,5dicaffeoylquinic acid, and chlorogenic acid were measured by DAD at 328 nm. Myo-insitol was not UV active and the analysis was performed with ESI negative MS in a selected ion (SIM) mode at [M–H]− 235. The fragmentor voltage was 120 V, and the capillary voltage was 4000 V. The drying gas flow rate was 10.0 L/min, the nebulizer pressure was 30 psi, and the drying gas temperature was 300 °C. The calibration plots were linear over the range 5–350 µg/mL for myo-insitol, and 5–120 µg/mL for the rest of the marker compounds with correlation coefficients of > 0.99. The average values as well as the standard deviations are shown in " Figs. 6 and 7. l

Acknowledgements !

This research was supported in part by “Science Based Authentication of Dietary Supplements” funded by the Food and Drug Administration grant number 1U01FD004246-04, the United States Department of Agriculture, Agricultural Research Service, Specific Cooperative Agreement No. 58-6408-1-603-04. The authors thank Prof. Dr. Jon Parcher for helpful discussions, and Dr. Vijayasankar Raman for authenticating the plant material.

Conflict of Interest !

The authors declare no conflict of interest.

Affiliations 1

2

3

4

5 6 7

National Center for Natural Products Research, University of Mississippi, University, Mississippi, USA Division of Pharmacognosy, Department of BioMolecular Science, School of Pharmacy, University of Mississippi, University, Mississippi, USA Department of Pharmacognosy, College of Pharmacy, King Saud University, Riyadh, Saudi Arabia School of Pharmacy, Jiangxi University of Traditional Chinese Medicine, Nanchang, China Chinese Pharmacopeia commission, Beijing, China College of Pharmaceutical Science, Soochow University, Suzhou, China School of Pharmacy, Hunan University of Traditional Chinese Medicine, Changsha, China

References 1 Shang X, Pan H, Li M, Miao X, Ding H. Lonicera japonica Thunb.: ethnopharmacology, phytochemistry and pharmacology of an important traditional Chinese medicine. J Ethnopharmacol 2011; 138: 1–21 2 Guo CW, Zhou CF. Analysis of the geoherbalism of Lonicerae japonica flos produced in Xinmi of Henan province. Chin J New Drugs 2012; 21: 2913–2916 3 Commission CP. The Pharmacopoeia of the Peopleʼs Republic of China 2010, Vol. I: Chinese Materia Medica, Vol. I. Beijing: Chemical Industry Press; 2010

Zhao J et al. Effect of Processing …

Planta Med 2015; 81: 754–764

4 Jiao SG. Research and comprehensive utilization of honeysuckle. Qi Lu Yao Shi (Qilu Pharmaceutical Affairs) 2009; 28: 487–489 5 Rhee KH, Lee KH. Antimicrobial effects of Lonicera japonica against gram positive and gram negative anaerobic bacteria. Nat Prod Sci 2011; 17: 23–25 6 Xin N, Li W, Li YJ, Ma XK, Fu ZP, Li Y. Study of antivirus, antibacteria and immune functions. J Med Plant Res 2011; 5: 5407–5412 7 Li T, Peng T. Traditional Chinese herbal medicine as a source of molecules with antiviral activity. Antiviral Res 2013; 97: 1–9 8 Liao Y, Dong S, Kiyama R, Cai P, Liu L, Shen H. Flos Lonicerae extracts and chlorogenic acid protect human umbilical vein endothelial cells from the toxic damage of perfluorooctane sulphonate. Inflammation 2013; 36: 767–779 9 Tae J, Han SW, Yoo JY, Kim JA, Kang OH, Baek OS, Lim JP, Kim DK, Kim YH, Bae KH, Lee YM. Anti-inflammatory effect of Lonicera japonica in proteinase-activated receptor 2-mediated paw edema. Clin Chim Acta 2003; 330: 165–171 10 Sulaiman MR, Zakaria ZA, Moin S, Somchit MN, Chai YS. Antinociceptive and anti-inflammatory effects of Lonicera japonica thunb. Res J Pharmacol 2008; 2: 38–42 11 Wen YX. Applications of differently processed medicinal materials from Flos Lonicerae. Lishizhen Med Mater Med Res 2004; 15: 825 12 Wang YZ, Zhang ZW. Review on the processing of Flos Lonicerae. J Shandong Univ Tradit Chin Med 2000; 24: 66–67 13 Nan YS, Sun ZG, Wei CZ. Effects of Flos Lonicerae processing on coagulation and antibacterial activity. Chin Tradit Patent Med 1989; 11: 17– 18 14 Huang YY, Huang M, Lu ZH. A comprehensive study on Flos Lonicerae processing. J Chin Med Mater 1994; 17: 25–27 15 Cong Y, Wang Y, Li M, Wang Z, Li Q. Preliminary study on the machanism of processing Flos Lonicerae japonica. J Henan Univ (Medical Science) 2009; 28: 242–245 16 Nan YS, Wei CZ, Sun ZG. Effect of processing on chemical components of Flos Lonicerae japonica. Chin Tradit Patent Med 1990; 12: 20–21 17 Zhao Z, Liang Z, Chan K, Lu G, Lai Mei Lee E, Chen H, Li L. A unique issue in the standardization of Chinese materia medica: processing. Planta Med 2010; 76: 1975 18 Zhao J, Avula B, Chan M, Clément C, Kreuzer M, Khan IA. Metabolomic differentiation of maca (Lepidium meyenii) accessions cultivated under different conditions using NMR and chemometric analysis. Planta Med 2012; 78: 90–101 19 Kim HK, Choi YH, Verpoorte R. NMR-based metabolomic analysis of plants. Nat Protoc 2010; 5: 536–549 20 Lee EJ, Kim JS, Kim HP, Lee JH, Kang SS. Phenolic constituents from the flower buds of Lonicera japonica and their 5-lipoxygenase inhibitory activities. Food Chem 2010; 120: 134–139 21 Tong S, Yan J, Guan YX. Preparative separation of isomeric caffeoylquinic acids from Flos Lonicerae by pH-zone-refining counter-current chromatography. J Chromatogr A 2008; 1212: 48–53 22 Rumalla CS, Avula B, Zhao J, Smillie TJ, Khan IA. Quantitative determination of phenolic acids in Lonicera japonica Thunb. Using high performance thin layer chromatography. J Liq Chrom Relat Tech 2011; 34: 38– 47 23 Holzgrabe U, Wawer I, Diehl B. NMR spectroscopy in pharmaceutical analysis. Oxford: Elsevier; 2011 24 Shi S, Huang K, Zhang Y, Zhao Y, Du Q. Purification and identification of antiviral components from Laggera pterodonta by high-speed countercurrent chromatography. J Chromatogr B Analyt Technol Biomed Life Sci 2007; 859: 119–124 25 Rebecca OPS, Boyce AN, Somasundram C. Isolation and identification of myo-inositol crystals from dragon fruit (Hylocereus polyrhizus). Molecules 2012; 17: 4583–4594

Downloaded by: Karolinska Institutet. Copyrighted material.

Original Papers

Effect of Processing on the Traditional Chinese Herbal Medicine Flos Lonicerae: An NMR-based Chemometric Approach.

The processing of medicinal materials, known as Pao Zhi in traditional Chinese medicine, is a unique part of traditional Chinese medicine and has been...
319KB Sizes 0 Downloads 11 Views