Quality Assessment and Ecotype Distinction for Panax quinquefolius L . from China and Canada by 1 H NMR and Chemometrics

Article history: Received on: 09/11/2016 Accepted on: 07/01/2017 Available online: 30/05/2017 Panax quinquefolius L. is one of the most widely consumed and cultivated herbal medicines around the world. China and Canada are the two major producing countries. However, research on the ecotype distinction of P. quinquefolius from Canada has never been reported, and the quality evaluation of P. quinquefolius between China and Canada using nuclear magnetic resonance ( 1 H-NMR) is limited. Here, we investigated the 1 H NMR key signals and ecological factors of P. quinquefolius samples, and the data were further analyzed by principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA). The key signals were identified as sugar and methyl, which distinguished all the samples into two different ecotypes. This is the first report of ecotype division for P. quinquefolium worldwide. The results showed quality variation of P. quinquefolium from different geographic areas, implying the ecological adaptation and biodiversity. Our findings also demonstrate the critical need for improving quality and quality standardization, appropriate ecological regionalization and promoting industrialized development of P. quinquefolium.


INTRODUCTION
Panax quinquefolius L. is one of the most important herbs in the world, and is native to North America in USA and Canada; it is now widely cultured in many parts all over the world, including China (Christensen et al., 2006).The history of P. quinquefolius, in China, was first recorded in Ben Cao Gang Mu in the Ming Dynasty (Li, 2004).Presently, the cultivation of the herb has been extended on a large scale in Jilin and Shandong provinces.Canada, China and USA are the three major producing areas.China has gradually become the major production, consumption and export country instead of traditional import country.
It was reported that the contents and composition of perceived pharmacological properties varied significantly among populations (Li et al., 1996;Tang et al., 2016;Qi et al., 2016).The pharmacological effects of ginseng roots has been attributed primarily to ginsenosides, a triterpenoid saponin glycoside.
Ginsenosides were recorded as indicator compound in the Chinese Pharmacopoeia and the United States Pharmacopeia.The grading and pricing of P. quinquefolius are primarily determined by the ginsenosides of harvested roots (Lim et al., 2005).The ecotype is a population concept put forward by Turesson in 1921, andOdum improved it (Liu et al., 2004;Odum, 1996).The ecotype of P. quinquefolius from China has been investigated in our lab, which showed that two chemoecotypes of P. quinquefolius in China, ginsenosides Rb1-Re from outside Great Wall and Rg2-Rd from inside Great Wall with distinct climatic characteristics (Wang and Huang, 2015;Wang et al., 2015;Huang et al., 2013).To date, some studies have been reported for quality assessment of P. quinquefolius (Ludwiczuk et al., 2005;Xu et al., 2011;Chan et al., 2000;Sun et al., 2012;Zhao et al., 2015;Yang et al., 2012), but the ecotype of P. quinquefolius from major productive country Canada has never been conducted.The current article investigates the differences in quality and ecotype of P. quinquefolius from Canada and China by 1 H-NMR spectroscopy coupled with chemometrics, which includes PCA and PLS-DA.This paper aims to establish an effective and rapid metabolomics method for geographical origin traceability of P. quinquefolius in the world.

Sample collection
The samples of P. quinquefolius were collected from 6 locations, which include China and Canada populations (Table 1).Three independent plants were collected from each location.Professor Huang Linfang identified the botanical specimens, and the voucher specimens were deposited in the Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences, Beijing, China.The typical root and aerial part ofP.quinquefolius is shown in Fig. 1.

Preparation of samples
CD 3 OD, DMSO-d6 and C 5 D 5 N were tested to dissolve the lyophilized samples; CD 3 OD was found the best solvent.30mg of each samplewas extracted with 1 ml of CD 3 OD.The extractions were vortexed vigorously for 30s, and then were sonicated for 40 min at room temperature.NMR samples were prepared by shaking manually, sonicated the insoluble material at 2000 g for a further 5 min, and adjusted to pH 7.0 ± 0.003 using deuterated base and/or acid.Finally, it wastransferring the supernatant of the extracts into 5 mm NMR tubes for NMR analysis (Shin et al., 2007).

H NMR analysis
One-dimensional 1 H NMR spectra were measured at a temperature of 300Kon a 600.13MHzBrukerAvance spectrometer (BrukerAnalytische GmbH, Rheinstetten, Germany) equipped with a broad-band-observe (BBO) probe.A zgcppr pulse sequence wasapplied to suppress the residual water signal.A total of 128transients were collected in 32 K data points with relaxation delay of 2s.Aspectral width of 9615.4Hz and an acquisition time per scan of 1.70s were used.
Prior to Fourier transformation, an exponential line broadening function of 0.30 Hz was applied to the free induction decay.The chemical shifts for all samples were referenced to TMS (tetramethylsilane) at 0.00 ppm.

Ecological factors
The data of ecological factors were collected from the NASA Atmospheric Science Data Center, which covered the whole growth process of P. quinquefolius from the year 2011 to 2015.The average values of ecological factors listed in Table 2, including annual air temperature, relative humidity, annual precipitation, annual average solar isolation, average air temperature in January, and minimum air temperature in January, average air temperature in July, and maximum air temperature in July, at 6 locations.

H-NMR spectra profiling
To obtain the best separation for all the integrated signals in 1 H NMR spectra, NMR solvents were optimized in this study.Based on the dissolvability of the main pharmacological components, three solvent systems involving CD 3 OD, DMSO-d6 and C 5 D 5 N were investigated.CD 3 OD showed better separation for the signals of the analyzed ginsenosides, and it was the preferred 1 H NMR solvent finally.In Fig. 2I, examples of NMR spectra of P. quinquefolius are shown.Differences can be observed from the spectra, PCA and PLS-DA was utilized to further analyze the differences in the spectra of the samples.

Chemometric analysis
PCA was performed on the pretreated NMR spectra of all the studied P. quinquefolius samples.Fig. 3A presents the differences between samples from China and Canada.The samples from Jilin in China and the samples from Canada clustered into one group, while samples from Shandong in China were in a distinct group, which is similar result with our previous research (Huang et al., 2013).In order to further analyze the spectra of different samples, 16 main peaks were selected based on their intensity data.
These data were then imported into Excel to generate a curve graph (Fig. 3B).The figure shows the differences in content of the main compounds in various samples.It is interesting to note that the peaks in the range of 3.0 ppm to 5.5 ppm (sugar region) and those in the range of 0.5 ppm to 2.0 ppm (aliphatic region) show an opposite trend (Shin et al., 2007;Lee et al., 2009;Jocham et al., 2007).The peaks in the range of 0.5 ppm to 2.0 ppm and 3.0 ppm to 5.5 ppm are pointing to opposite directions, meaning that if higher 0.5 ppm to 2.0 ppm peaks are present, then lower 3.0 ppm to 5.5 ppm peaks are present, and vice versa.
This result shows that location may influence the quality of P. quinquefolius significantly.In United States Pharmacopeia USP35, the content of ginsenosides acceptance criteria is no less than 4.0% of total ginsenosides (Rg 1, Re, Rb 1 , Rc, Rb 2 and Rd) on the dried basis, while in Chinese Pharmacopoeia 2015, the acceptance criteria is no less than 2.0% of total ginsenosides (Rg 1, Re and Rb 1 ) on the dried basis.The content requirement of ginsenosides is different in the two Pharmacopeias, which illustrate the importance of quality control.

Ecological factors analysis and its effect on phytochemical composition
Principal component analysis was used to analyze the differences in climate in China (Shandong and Jilin) and Canada.The PLA-DA scores plot shows that ecological factors of Jilin in China and those in Canada clustered into one group, and those in Shandong, China clustered into another group (Fig. 3C).The PCA loadings plot indicated that the annual air temperature and maximum air temperature in July made a great influence on the discrimination of the 2 types of climate (Fig. 3D).Shandong has a relatively higher annual air temperature and relatively lower maximum air temperature in July,which of these above all verified our previous results.
The result was also validated by our previous experiments.In 2008, we reported that the northeast of China including Jilin is the most suitable area for the growth of P. quinquefolius, because the climate is very similar to North America (Chen et al., 2008).And in 2011, our data further showed the temperature was the most important environmental factors affecting the content of ginsenosides of P. quinquefolius roots (Dong, 2011).Jochum.GM, et al, demonstrated that plants grown, at high temperatures, had less root biomass and greater concentrations of storage root ginsenosides (49%), than plants grown at low temperatures (Jocham et al., 2007).In the present study, Shandong samples (grown at high environmental temperature) contain higher signal levels in the range of 0.5 ppm to 2.0 ppm, and the selected peaks, 1.59, 1.53, 1.27, 1.19, 0.91 0.82 ppm, were mainly correlated to the methyl group of ginsenosides that is referred to in the supplementary information (Fig. 2II).The samples from Jilin and Canada (grown at lower environmental temperature) contain relatively higher levels of signals in the sugar region.We performed discriminatory NMRbased chemical profile studies on P. quinquefolius roots from China and Canada.Samples from Shandong, China contain higher content of ginsenosides and relatively lower saccharides than those from Jilin in China and those from Canada.And chemical discrimination was in accordance with their ecological discrimination, and temperature factors were concluded as the most influential ecological variable.All the above results are consistent with our previous work.Consequently, we assume that the ecotype of Canadian P. quinquefolius was outside Great Wall due to the similar latitude to the Northeast of China.

CONCLUSION
The P. quinquefolius are observed with big differences in quality and ecotype between China and Canada, which indicate that the P. quinquefolius from Canada is the ecotype of outside Great Wall.This is the first time studied the ecotype distinction of P. quinquefolius around the world, and the present approach is important and reliable to control the quality and distinguish ecotype.

Fig. 2 :
Fig. 2: I Representative 1 H NMR spectra of P. quinquefoliusfrom China and Canada.II NMR spectra of ginsenoside Rb1, Rc, Rd and Re.

Fig. 3 :
Fig. 3: A The PLS-DA analysis of different P. quinquefolius samples (R 2 X [1]=0.400836R 2 X [2]=0.173504);B Intensities of selected peaks from 1H NMR profiles of Various P. quinquefolius; C The PLS-DA score plots of ecological factors of sampling location (R 2 X [1]=0.570915R 2 X [2]=0.299042);D The PLS-DA loading plot of ecological factors of sampling location.X1 Annual air temperature; X2 Relative humidity; X3 Annual precipitation; X4 Annual average solar insolation; X5 Average air temperature in January; X6 Minimum air temperature in January; X7Average air temperature in July; X8 Maximum air temperature in July.

Table 1 :
P. quinquefolius samples collected from different locations

Table 2 :
The ecological factors of sampling locations.