Comparison of Two Different Astragali Radix by a 1H NMR-Based

Subscriber access provided by UNIV OF NEBRASKA - LINCOLN

Article

Comparison of Two Different Astragali Radix by H NMR based Metabolomic Approach 1

Ai-Ping Li, Zhen-Yu Li, Hai-Feng Sun, Ke Li, Xue-Mei Qin, and Guan-Hua Du J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/pr501167u • Publication Date (Web): 06 Apr 2015 Downloaded from http://pubs.acs.org on April 11, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Proteome Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Comparison of Two Different Astragali Radix by 1

H NMR based Metabolomic Approach

Ai-Ping Li,a,b,1 Zhen-Yu Lia*,1, Hai-Feng Sun a,b, Ke Lia, Xue-Mei Qina*, Guan-Hua Dua,c a

Modern Research Center for Traditional Chinese Medicine of Shanxi University, No. 92, Wucheng Road, Taiyuan 030006, Shanxi, People’s Republic of China

b

College of Chemistry and Chemical Engineering of Shanxi University, No. 92, Wucheng Road 92, Taiyuan 030006, Shanxi, People’s Republic of China

c

Institute of Materia Medica, Chinese Academy of Medical Sciences, Beijing 100050, People’s Republic of China

* To whom correspondence should be addressed. Tel (Fax): +86-351-7018379. E-mail: [email protected] (Xue-Mei Qin); [email protected] (Zhen-Yu Li) 1

These two authors contributed equally to this study.

1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT: Astragali Radix (AR) is a commonly used herbal drug in Traditional Chinese Medicine, and widely used for the treatment of diabetes, cardiovascular diseases, nephropathy and neuropathy. The main source of AR is the dried root of Astragalus membranaceus var. mongholicus (Bge.) Hsiao in China, and both cultivated and wild ARs are used in the clinic. Systematic comparison of cultivated AR and wild AR should be performed to ensure the efficacy and safety in clinic. In this study, the chemical compositions of two different ARs, collected in Shanxi (wild) and Gansu (cultivated) Provinces, were compared by NMR based metabolic fingerprint coupled with multivariate analysis. Then SX-AR and GS-AR induced metabolic changes of endogenous metabolites in mice were also compared. The results showed that SX-AR and GS-AR differed significantly not only in the primary metabolites, but also the secondary metabolites. However, alterations of endogenous metabolites in serum, lung, liver and spleen were relatively small between them. This study provided a novel and valuable method for consistency and diversity evaluation of herbal drugs, and further studies should be conducted on the difference of polysaccharide as well as biological effect between the two kinds of ARs.

KEYWORDS: Astragali Radix, metabolomics, NMR, chemical differences, metabolic changes

2


Page 2 of 32

Page 3 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION

Astragali Radix (AR), also known as Huangqi in China, is derived from the dried root of Astragalus membranaceus (Fisch.) Bge. or A. membranaceus var. mongholicus (Bge.) Hsiao, which belongs to the Leguminosae family. 1 It is a traditional folk medicine that has been used for many therapeutic purposes in Asia, and pharmacological studies have indicated that AR has immunostimulant, hepatoprotective, tonic, diuretic, antidiabetic expectorant, analgesic, and sedative properties. 2, 3 Nowadays, the dried roots of A. membranaceus var. mongholicus (Bge.) Hsiao are the main sources of AR in the commercial market of China, and there exist two types of growth pattern (cultivated and wild). The wild A. membranaceus var. mongholicus (Bge.) Hsiao is mainly distributed in droughty mountainous areas, especially in the north of Shanxi Province, and grows for more than five years before harvest. While the cultivated AR is often cultivated in wet and flat soil, especially in Gansu Province, and it only grows for two years. 4 Usually, the wild AR has the longer and thicker roots due to the longer growing years. In the market, AR is graded by the root length, diameter, and physical appearance: the longer and thicker the roots are, the higher the quality. The chemical composition of the herbal drugs is always influenced by the weather, geographic location, soil conditions, and cultivation patterns, thus the wild and cultivated ARs differed not only in the appearance, but also in chemical compositions. In recent years, the main source of AR changed from the wild to the cultivated due to the increasing demands. However, both of them meet the quality standard of Chinese Pharmacopeia, 1 and they are not distinguished in the clinical use. To ensure the efficacy and safety in clinic, systematic comparison should be conducted to disclose their similarities and differences. Metabolic fingerprint approach has been widely used as a state of art technique in medicinal plant research. Compared with gas chromatography-mass-spectrometer (GC-MS) and liquid chromatography-mass spectrometry (LC-MS), nuclear magnetic resonance (NMR) spectroscopy served as a rapid, non-destructive and high-throughput method requires minimal sample preparation.

5, 6

In addition, NMR is a very useful technique for structure elucidation due to

various two-dimensional NMR measurement without the further fractionation, which makes NMR an ideal choice for chemical analysis of medicinal plants, such as Ephedra species, 3


7

Tussilago


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

farfara L,

8

tobacco leaf,

9

Astragalus Roots

10

and Rehmanniae Radix

Page 4 of 32

11

etc. Metabolomic

technique can be also used to study holistic response of organism to drugs or other stimulus. All drugs are xenobioics to animals, and metabolomic approaches have been proven powerful in understanding the xenobiotic induced metabolomic changes in animal biofluids and tissues, such as gallic acids, 12 chamomile tea, 13 arginine, 14 and dietary flavonoids. 15 In this study, the NMR based metabolic fingerprint coupled with multivariate analysis, was applied to compare the chemical difference between the two types of ARs growing in Shanxi (SX-AR) and Gansu (GS-AR) Province. Changes of endogenous metabolites in vivo induced by SX-AR and GS-AR were also compared for quality evaluation.

MATERIALS AND METHODS

Chemicals and Plant materials Analytical grade acetonitrile, petroleum ether, ethyl acetate (EtOAc), n-butanol (n-BuOH), Na2HPO4 and KH2PO4 were from Beijing Chemical Works (Beijing, China). D2O was purchased from Norell (Landisville, Pennsylvania, USA). Sodium 3-trimethlysilyl [2, 2, 3, 3-d4] propionate (TSP) was purchased from Cambridge Isotope Laboratories Inc (Andover, MA, USA). Phosphate buffer was prepared by dissolving KH2PO4 and Na2HPO4 in water (0.1 M, pH 7.4) containing 0.01% TSP and 10% D2O. Formononetin, ononin, calycosin, calycosin-7-O-β-D-glucoside, 1,2-dihydroxy-3,4-dimethoxyisoflavan-7-O-β-D-glucoside, 9,10-dimethoxypterocarpan-3-O-β-D-glucoside, astragaloside I, astragaloside II were obtained from Shanghai forever-Biotech Co., Ltd. (Shanghai, China) AR samples were collected at the same time of year from the same planting area from Hunyuan County, Shanxi Province (SX-AR-0001-0008) and Minxian County, Gansu Province (GS-AR-0001-0008) of China, respectively, and authenticated as the roots of Astragalus membranaceus (Fisch.) Bge. var. mongholicus (Bge.) Hsiao by Prof. Xue-Mei Qin, and the voucher specimens were deposited in the herbarium of Modern Research Center for Traditional Chinese Medicine of Shanxi University. AR aqueous crude extract preparation for chemical analysis and biological evaluation 4


Page 5 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


According to the traditional use, 100 g of AR sample was extracted with 1000 mL of water by refluxing twice (2 h each). The combined solution was filtered, concentrated under reduced pressure and subsequently freeze-dried. The yield ratios of SX-AR and GS-AR samples are 30.86% and 33.32%, respectively. Chemical analysis of AR aqueous crude extract by NMR based profiling Sample Preparation for NMR analysis.

Two different sample preparation procedures were

used for comprehensive chemical characterization of AR aqueous crude extract. In the first procedure (M1), the freeze-dried powder (about 30 mg) was redissolved in KH2PO4 buffer in D2O (adjusted to pH 6.0 by 1 N NaOD) containing 0.05% TSP. Then the sample was centrifuged for 10 min at 13 000 rpm. The supernatants (600 μl) of all the samples were transferred into 5 mm NMR tube for NMR analysis. In the second procedure (M2), 4 g freeze-dried powder was suspended in 100 mL of water, and partitioned by equal volume of petroleum ether, EtOAc, and n-BuOH in sequence. The soluble fractions in petroleum ether (M2P), EtOAc (M2E), n-BuOH (M2B), and the residual water (M2W) fractions were transferred separately into round bottomed flasks and taken to dryness with a rotary vacuum evaporator. The M2P and M2E were redissolved in CDCl3 and CD3OD, respectively, whereas the M2B and M2W were reconstituted in KH2PO4 buffer in D2O. All the samples were transferred into 5 mm NMR tubes for NMR analysis. NMR analysis. 1H NMR spectra was recorded at 25 ◦C on a Bruker 600-MHz AVANCE III NMR spectrometer (600.13 MHz proton frequency, Bruker BioSpin, Germany). Each 1H NMR spectrum for M1 and M2E was acquired using zg30 pulse sequence, consisted of 32 scans requiring 2.654 s acquisition time with the following parameters: spectral width = 12 345.7 Hz, spectral size = 65 536 points, pulse width (PW) = 30°, and relaxation delay (RD) = 1.0 s. For other fractions, zg30 pulse sequence was used for M2P, and noesygppr1d pulse sequence was used for M2B and M2W. NMR data preprocessing. The 1H NMR spectra were manually phase- and baseline-corrected, and calibrated to TSP at 0.00 ppm for M1, and methanol at 3.30 ppm for M2E using MestReNova (version 8.0.1, Mestrelab Research, Santiago de Compostella, Spain). For M1, the spectrum was divided into integrated regions of equal width (0.04 ppm) corresponding to the region of 0.78-9.22. The region of  4.66-5.06 was excluded from the analysis because of the residual signals of HOD. For M2E, the 1H NMR spectral region from  0.16-9.76 was segmented into regions with widths of 0.04 ppm, and the two regions of  4.68-5.40 and  3.28-3.36 were 5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

excluded from the analysis due to the residual signal of HOD and CHD2OD. Biological comparison of SX-AR and GS-AR in vivo Animal protocol and AR administration. The freeze dried extract was suspended in distilled water to 1.0 g/mL, which is equivalent to the crude drug. According to the extraction yields, the final concentration for SX-AR and GS-AR were 0.31 g/mL and 0.33 g/mL, respectively. For the low dose group, the drug concentration was diluted to 0.25 g (equivalent to the crude drug)/ml. The animal study was performed according to the International Rules considering Animal Experiments and the Internationally Accepted Ethical Principles for Laboratory Animal Use and Care. Four-week-old, male Kunming (KM) mice, weighing 18-22 g, were purchased from Experimental Animal Center of Academy of Military Medical Sciences, the Chinese people's liberation army (PLA) [SCXK (Military) 2012-0004]. All animals were housed at room temperature (20-25 ◦C) and constant humidity (40-70%) under a 12 h light/dark cycle in SPF (Specific Pathogen Free) grade laboratory conditions. Animals were allowed to have access to food and water ad libitum, except for fasting periods before experiments. After a one-week adaptive feeding period, KM mice were randomly divided into 5 groups (n = 8) for a 7-day treatment intervention: control group (NS); GS-AR low-dose group (GS-AR-L); GS-AR high-dose group (GS-AR-H); SX-AR low-dose group (SX-AR-L); SX-AR high-dose group (SX-AR-H). Aqueous extracts of AR were administrated into mice via gavage for the low-dose (5g/kg, equivalent to the crude drug) and high-dose (20g/kg, equivalent to the crude drug) groups, the control group was given the same volume of distilled water. 0.2 mL/10 g body weight was administered to the mouse once a day. Sacrifice and sample collection. Blood (0.5-0.8 mL) was withdrawn from retro-orbital plexus of mice and collected into eppendorf tubes on 8th day after fasting for 12 h. Then blood was separated using refrigerated centrifugation at 13 000 rpm for 10 min to afford the serum. The serum samples, liver, spleen and lung tissues were snap-frozen in liquid nitrogen and stored at −80 °C for further analysis. Samples preparation. Each serum sample (300 μl) was mixed with 300 μl of D2O, centrifuged (13 000 rpm, 4 °C, 10 min), and 550 μl of supernatant was transferred into 5 mm NMR tubes for NMR analysis. Liver (about 200 mg), spleen (about 100 mg) and lung (about 90 mg) were respectively extracted with 1 mL of acetonitrile/water (1:1) using an ultrasonic cell crusher 6


Page 6 of 32

Page 7 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(Ningbo Scientz biotechnology co., Ltd, China). The resultant supernatants were centrifuged (13 000 rpm, 4 °C, 10 min), lyophilized in nitrogen flow and redissolved into 600 μl phosphate buffer (0.1 M, KH2PO4/Na2HPO4, pH 7.4) containing 0.01% TSP and 10% D2O. After final centrifugation (13 000 rpm, 4 °C, 10 min), each supernatant (550 μl) was transferred into a 5 mm NMR tube for NMR analysis. NMR analysis of serum and tissues extract. 1H NMR spectra was acquired at 298 K on a Bruker 600-MHz AVANCE III NMR spectrometer equipped with a Bruker 5 mm double resonance BBI probe, using cpmgpr1d pulse sequence for serum samples, and noesygppr1d pulse sequence for tissue extracts, respectively. For assignment purposes, databases such as Chenomx NMR suite software (chenomx Inc. Edmonton, AB, Canada, evaluated), Human Metabolome Database (HMDB) (http://www.hmdb.ca/spectra/nmr_one_d/search), and Biological Magnetic Resonance Data Bank (BMRB) (http://www.bmrb.wisc.edu/) were used, and the assignment was also based on the reported literature data. 12, 16, 17 NMR data preprocessing. All spectra were manually phase- and baseline-corrected using MestReNova software. TSP with a chemical shift at δ 0.00 was used as a spectral reference for all tissue extracts, whereas the NMR spectra of the serum were referenced to the creatine at δ 3.04. The regions containing the resonance from residual water (δ 4.68-5.16) were excluded. Then, all spectra were segmented at 0.04 ppm intervals across δ 0.72-7.88 for serum sample and δ 0.80-9.00 for tissue extract samples. Normalization to a total sum of all integrals for serum and to the tissue weights for tissue extracts were conducted before multivariate analysis. Statistical analysis All preprocessed NMR data were imported into Simca-P 13.0 software (Umetrics, Sweden) for multivariate data analysis. Principal component analysis (PCA) was carried out on the mean centered data to obtain an overview and find possible outliers. Partial least squares discriminant analysis (PLS-DA) applies PLS to discriminate between groups of samples that are defined as separate response variables.

18

The model qualities were assessed with the total explained

variables (R2X values) and the model predictability (Q2 values) followed with rigorous permutation tests (number: 200). The orthogonal projection to latent structure with discriminant analysis (OPLS-DA)

19

was carried out using the NMR data as X-matrix and class information as

Y-matrix with pareto scaling and cross-validation. Pareto scaling, which divides each variable by 7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 32

the square root of its standard deviation, has the advantage that the loadings are less distorted and the dominating influence of high variance variables is reduced. 20 Compared to PLS-DA, the main benefit in interpretation using OPLS-DA lies in the ability of OPLS-DA to separate predictive from non-predictive (orthogonal) variation, and the validity of the model is certified by CV-ANOVA method.

20

The corresponding S-plot was employed to identify differential

metabolites contributing to the separation between classes. For quantitative data analysis of changes of endogenous metabolites by the treatment of SX-AR and GS-AR, relative amount of metabolites were evaluated based on the integrated regions (buckets) from the least overlapping NMR signals of metabolites. These semi-quantitative data were expressed in the form of mean ± standard error of the mean (S.E.M.) and were also subjected to classical one-way ANOVA analysis using SPSS 16.0 software to investigate the differences between the SX-AR and GS-AR at the same dose. Differences among groups were considered to be statistically significant if p < 0.05.

RESULTS

Chemical analysis of AR aqueous crude extract by NMR based profiling Metabolites assignment. In general, metabolomic studies should be designed to detect as many metabolites as possible. Therefore, solvents capable of extracting comprehensive groups of metabolites should be chosen during the process of sample preparation.

21

In this study, two

methods were used for NMR analysis of AR aqueous crude extracts. Figure 1 showed the typical 1

H NMR spectra of AR water extracts obtained from M1 (A) and M2E (B). (The stack NMR

spectra of all ARs in M1 (A) and M2E (B) were supplied in the Figure S1). The resonances in these spectra were assigned to individual metabolites according to the literature data, 22 NMR database such as HMDB, and BMRB, and also based on comparisons with the chemical shifts of standard compounds. It was apparent that the M1 fraction is dominated by amino acids, carbohydrates, organic acids, and nucleotide derivatives (Figure 1A). Amino acids such as isoleucine (0.96, 1.03), leucine (0.98, 0.99), valine (1.01, 1.06), arginine (1.70,  1.90, ), alanine ( 1.48), threonine (1.33) and proline ( 2.08,  2.38), GABA ( 2.30,  3.00), 8


Page 9 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


glycine ( 3.57) and several organic acids such as succinic acid ( 2.45), acetic acid ( 1.93) were identified in the amino and organic acids regions. In the sugar region, sucrose ( 5.42, 4.22,  3.68,  3.82) were identified as the major sugars (Figure 1A). In addition, small amounts of raffinose ( 5.43) and monosaccharides, such as glucose, galactose and xylose were also identified. Moreover, formic acid (8.47), fumaric acid (6.53), phenylalanine ( 7.33,  7.41), tyrosine (6.85,  7.18), trigonelline (8.10, 8.17,  4.46, 9.14), adenine (8.21, 8.26), adenosine (8.21, 8.35, 6.03) and uridine (5.91, 5.92, 7.88) were detected in the aromatic region. Due to the high concentration of primary metabolites, such as amino acids and sugars, contained in the M1 extracts, almost no secondary metabolites could be detected. Thus, another sample preparation method (M2) was used to eliminate the strong primary metabolites signals. In M2, the aqueous crude AR extracts were suspended in water and partitioned sequentially by petroleumether, EtOAc, and n-BuOH, to give four fractions, M2P, M2E, M2B, and M2W. As shown in Figure S2, most of the sugars and other primary metabolites, such as amino and organic acids were enriched in the M2W and M2B fraction, while saturated and unsaturated fatty acids or their esters were concentrated in M2P. Thus, these fractions were not subjected to the further multivariate analysis. For the M2E fraction (Figure 1B), no sugar signals were observed, but the flavonoids and saponins were enriched, which are used as chemical markers for quality evaluation of AR. Saponins such as astragaloside I (0.55, 0.99, 1.11, 1.19, 1.23, 1.25, 2.00, 2.03) and astragaloside II (0.25, 0.57, 0.89, 0.97, 2.08),and flavonoids such as calycosin

(8.04),

calycosin-7-O-β-D-glucoside

(8.18),

1,2-dihydroxy-3,4-dimethoxyisoflavan-7-O-β-D-glucoside (2.97, 3.70, .45, 6.54, 6.62, 6.76, 6.98), formononetin (8.12), 9,10-dimethoxypterocarpan-3-O-β-D-glucoside (5.54, 6.54, 6.63, 6.79, 6.94, 7.43) were detected and further confirmed by comparison with standard compounds. The chemical shifts and coupling constants of all the identified metabolites were summarized in Table 1. Multivariate data analysis. Visual inspection of M1 fraction revealed that SX-AR contained more acetic acid, threonine, succinic acid but less α-glucose compared with GS-AR. To obtain the subtle metabolite variations, multivariate data analysis methods were applied to the NMR data. The scores plot from principal component analysis (PCA) (Figure 2A) showed that more than 9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 32

70.9% variables can be explained with two principal components, and a clear separation could be seen between SX-AR and GS-AR samples. Marked chemical differences were identified by OPLS-DA and the corresponding S-plot (Figure 2B) indicated that SX-AR contained more sucrose, threonine, acetic acid, asparagine, raffinose, betaine, succinic acid, and less α-glucose, aspartic acid, GABA, leucine, choline, alanine, proline than GS-AR samples. For the M2E, the PCA score plot also showed a clear separation between the SX-AR and GS-AR (Figure 2C). PC1 accounted for 75.2% of the total variance, while PC2 accounted for 17.0% of the total variance. It is also interesting to notice that the close intragroup sample clusters suggested good reproducibility in the sample preparation procedures and NMR measurements. In order to evaluate the validity of the model, permutation tests and CV-ANOVA (p

Comparison of Two Different Astragali Radix by a 1H NMR-Based

Recommend Documents