Fingerprint quality detection of Solanum nigrum using high-performance liquid chromatography–evaporative light scattering detection

Abstract

Context: Solanum nigrum Linn. (Solanaceae), a traditional Chinese medicine (TCM), has been used for cancer therapy. It is urgent to develop a novel quality standard to validly detect its quality.

Objective: To control its quality, a novel, accurate, and valid fingerprint method was developed by high-performance liquid chromatography–evaporative light scattering detection (HPLC-ELSD) in the current case. We could evaluate the quality of different batches and assure the stability of herbs’ quality in subsequent research.

Materials and methods: The HPLC-ELSD fingerprints have been developed through analyzing 41 batches of raw herbs collected from different areas in different harvesting time.

Results: We have determined the optimum extraction and detection conditions in the process of establishing herb fingerprint. And, we could establish reference fingerprint to control such herb quality. Also, we could determine optimum collecting location and harvesting time according to the fingerprint.

Discussion and conclusion: It is the first time a new method has been established to control the quality of S. nigrum through HPLC-ELSD. We developed combining similarity evaluation to identify and distinguish raw materials efficiently from different sources. For S. nigrum the most influenced factor on herb quality was the collecting location, and the next was the harvesting time. So, in order to get the consistent raw materials, the collecting location and the harvesting time should be fixed.

Keywords::

• HPLC-ELSD
• Solanum nigrum
• fingerprint

Introduction

Nowadays, much scientific research in modern biomedicine has laid emphasis on traditional Chinese medicine (TCM), which aims at maintaining the dynamic balance of a whole body to achieve a harmony between human and nature. There has been a growing interest in TCM, which had long been used as a remedy and obtained largely from wild plants around the world over the past decade. Longkui, the whole herb of Solanum nigrum Linn. (Solanaceae), has long been used as a remedy to treat trachitis, cancer, hepatic damage, mastitis, and serodontalgia (Meng, 1954; Ikeda et al., 2003; Raju et al., 2003; Hebbar et al., 2004; Zakaria et al., 2009).

S. nigrum is widely used as one of the herbal ingredients in prescriptions of TCM to treat liver, mammary, uterine cervix, gastric, and other cancers (Sammon, 1998; Yen et al., 2001; Son et al., 2003). It also has attracted increasing interest in the process of searching anticancer Chinese medicines, since it is a prolific producer of secondary metabolites such as solasonine, solasnine, oleic acid, linolic acid, and saponins (Saijo et al., 1982; Tsuyoshi et al., 2000). And, some of these constituents are well-known for their pharmacological and biological activities (Li et al., 2010; Jimoh et al., 2010; Yang et al., 2010). This herb grows wild in many provinces of the China. Because of the complicated terrains and diverse climates in China, its secondary metabolites often vary highly depending on the various circumstances. Considering growing extensive applications of S. nigrum, it was necessary and urgent to establish a novel quality standard to validly control its quality before its anticancer research.

Both the Food and Drug Administration (FDA) (Anon., 2000) and European Medicines Agency (EMEA) (Anon., 2001) clearly admit that the appropriate fingerprint chromatogram could be applied to assess the consistency of botanical drugs. Thus, the researchers gradually used many kinds of fingerprint to control the quality of TCM in the process of research. Among the available methods, high-performance liquid chromatographic (HPLC) fingerprint has been used widely because of it is simple, perspicuous, and accurate, and is also recommended by the WHO as a strategy for identification and quality control of TCM (Chen et al., 2007; Chang et al., 2009; Jian et al., 2010).

Unfortunately, although many chemical and pharmacological studies on S. nigrum have been already reported systematically many years ago (Meng, 1954), no chromatographic fingerprints about it have been reported until now. In our present research, we focused on developing a simple, accurate, and valid chromatographic fingerprint to control the quality of S. nigrum through HPLC–evaporative light scattering detection (HPLC-ELSD). In comparison with HPLC-DAD, HPLC-ELSD was more sensitive, and more specific to some compounds, such as solasonine, solasnine, oleic acid, linolic acid, saponin, and so on. In the process of establishing this method, multiple batches of sample were examined to develop a mean global chromatogram and obtain a representative standard fingerprint. Also, the similarity of each chromatogram against the mean global chromatogram was calculated using the professional software Computer Aided Similarity Evaluation (CASE). In comparison with the representative standard fingerprint, we could evaluate the efficacy of each batch of the sample. Furthermore, the stability of herbs’ quality in subsequent research is assured despite the influence of complicated terrains and diverse climates.

Methods

Materials

Forty-one batches of raw herbs of S. nigrum collected from 15 provinces of China were investigated (Table 1), all of which were identified by Prof. Xiang Chen (Guizhou University, P. R. China). Voucher specimens and samples were stored at the Institute of Biology, Guizhou Academy of Sciences.

Table 1.  The samples of Solanum nigrum: geographical origins, harvesting time, and similarity.

Sample no.	Geographical origins	Harvesting time	Similarity
1	Taian, Shandong	August 2006	0.954
2	Dezhou, Shandong	August 2006	0.853
3	Changchun, Jilin	August 2006	0.942
4	Haihekou, Jilin	August 2006	0.962
5	Siping, Jilin	August 2006	0.974
6	Shijiazhuang, Hebei	August 2006	0.964
7	Tangshan, Hebei	August 2006	0.882
8	Langfang, Hebei	August 2006	0.911
9	Tongliao, Inner Mongolia	August 2006	0.983
10	Chifeng, Inner Mongolia	August 2006	0.863
11	Lixian, Tientsin	August 2006	0.979
12	Huangshan, Anhui	August 2006	0.847
13	Hegang, Heilongjiang	September 2006	0.362
14	Chaoyang, Liaoning	August 2006	0.933
15	Jinzhou, Liaoning	July 2006	0.971
16	Shenyang, Liaoning	August 2006	0.942
17	Dalian, Liaoning	August 2006	0.557
18	Dongling, Liaoning	July 2006	0.954
19	Haicheng, Liaoning	August 2006	0.965
20	Chenzhou, Hunan	August 2006	0.83
21	Shunyi, Beijing	August 2006	0.985
22	Wulumuqi, Sinkiang	August 2006	0.801
23	Haikou, Hainan	August 2006	0.587
24	Shanghai	August 2006	0.406
25	Xinyang, Henan	September 2006	0.909
26	Pingdingshan, Henan	August 2006	0.984
27	Zhengzhou, Henan	August 2006	0.927
28	Shenzhen, Guangdong	August 2006	0.363
29	Hangzhou, Zhejiang	July 2006	0.518
30	Guilin, Guangxi	July 2006	0.283
31	Yuncheng, Shanxi	August 2006	0.961
32	Xian, Shanxi	August 2006	0.941
33	Xianyang, Shanxi	August 2006	0.978
34	Shenyang, Liaoning	June 20, 2006	0.414
35	Shenyang, Liaoning	July 5, 2006	0.537
36	Shenyang, Liaoning	July 20, 2006	0.953
37	Shenyang, Liaoning	August 5, 2006	0.978
38	Shenyang, Liaoning	August 19, 2006	0.945
39	Shenyang, Liaoning	September 4, 2006	0.935
40	Shenyang, Liaoning	September 19, 2006	0.943
41	Shenyang, Liaoning	October 5, 2006	0.876

Twelve reference compounds, uttroside B (a), solanigroside J (b), 5α-pregn-16-en-3β-ol-20-one lycotetraoside (c), linolic acid (d), palmitic acid (e), oleic acid (f), solasonine (g), solamargine (h), β₂-solamargine (i), nigrumnin I (j), degalactotigonin (k) and tigogenin 3-O-β-[d-glucopyranosyl-(1→2)-O-[β-d-glucopyranosyl-(1→3)]-O-β-d-glucopyranosyl-(1→4)-O-β-d-galactopyranoside (l), were isolated previously from the dried samples using repeated silica gel, Sephadex LH-20, and Rp-18 silica gel column chromatography, and their structures were elucidated by comparison of their spectral data (UV, IR, MS, ¹H NMR, ¹³C NMR, HMBC, and COSY). The purity of these chemicals was verified to be higher than 98% by normalization of the peak areas as detected by HPLC with ELSD and showed to be very stable in methanol solution.

Instrumentation and reagents

The chromatographic analysis was carried out on a Shimadzu LC-10ADvp series HPLC systems (Shimadzu Corporation, Japan) with a vacuum degasser, binary pump, auto injector, thermostated column compartment, and a Sedex 75 ELSD (Seder Corporation, France). The column was Phenomenex RP column (C₁₈, 5 µm, 250 × 4.6 mm). An SK8200H ultrasonic cleaner (KUDOS Company, China) was used for extraction. Isolation was performed by LC-8A prep. Methanol, acetonitrile, and water were of HPLC grade. Ethanol was of analytical grade.

Reference compounds

The standards of 12 compounds (Figure 1) were isolated, purified, and identified using ¹H NMR, ¹³C NMR, and MS in our laboratory. These reference compounds were dissolved in methanol and then injected into HPLC after filtration with filter membrane (0.45 µm).

Figure 1.  Structures of analytes used in the study.

Sample preparation

Powder of each batch of dried herbs (2.0 g) was extracted, respectively, with 20.0 mL ethanol:water (6:4, v/v) solution in an ultrasonic water bath for 20 min. The procedure was repeated two times. The extracted solution was blended and filtered with analytical filter papers. The filtered solution was evaporated at 50°C by vacuum. The dried extract was dissolved in 25.0 mL ethanol:water (6:4, v/v) and then filtered through 0.45 µm filter membrane.

HPLC procedures

A Phenomenex C₁₈ column (5 µm, 250 × 4.6 mm) was used for the procedures. The mobile phase consisted of 0.3% triethylamine in acetonitrile (A) and 0.3% triethylamine water:methanol (80:20, v/v) (B), performing a gradient program of 10–15% (A) in 10 min, 15–30% (A) in 15 min, 30–34% (A) in 30 min, 34–50% (A) in 52 min, 50–55% (A) in 55 min, and 55% (A) in 80 min. The flow rate was 1.0 mL/min and the column temperature was maintained at 40°C. About the ELSD, its drift tube temperature was 40°C and the nebulizer nitrogen gas flow rate was controlled at 3.5 bars.

Data analysis

The correlation coefficients and the similarities of entire chromatographic profiles among tested samples and the simulative mean chromatogram were calculated through a professional software named “Similarity Evaluation System for Chromatographic Fingerprint of Traditional Chinese Medicine” (version 2004 A), which was recommended by State Pharmacopoeia Committee of the People’s Republic of China.

Results

Optimization of extraction conditions

Considering a variety of compounds with relatively high polarity in S. nigrum, ethanol was the preferred choice of extraction solvent to effectively obtain good resolution in the present work Multiple related extraction conditions were designed and evaluated, which involved in the following experimental factors and corresponding levels: ethanol concentration (50, 60, or 70%, v/v), extraction method (ultrasonication, reflux, or Soxhlet). Then, the experimental factors of 60% ethanol and ultrasonication were confirmed to be more accurate than others. The orthogonal experiment involved three factors: (A) extraction time (10, 15, or 20 min), (B) ultrasonication times (1, 2, or 3 times), and (C) solvent volume (20, 16, or 12 mL). The orthogonal designs L9 (3⁴) were studied in this article. Comparing the sum of all characteristic peaks’ numbers and areas in each chromatogram of different factors, we could quantify the overall extraction of an herb, and use the recovery as a criterion to select optimal condition. Then, the optimal condition for extraction of S. nigrum was selected and presented in detail in Sample preparation.

Optimization of HPLC conditions

To develop a fingerprint for S. nigrum, an optimized strategy for HPLC conditions was performed. The chromatographic analysis was carried out on a Shimadzu LC-10 AD vp series HPLC system (Shimadzu Corporation, Japan). The different HPLC parameters including mobile phase (acetonitrile–water, acetonitrile–triethylamine–water, or acetonitrile–triethylamine–methanol–water) (Table 2), category of column (Phenomenex RP C₁₈ column 250 × 4.6 mm I.D. 5 µm, Ultimate^™ C₁₈ column 250 × 4.6 mm I.D. 5 µm, or SHIM-PACK vp-ODS C₁₈ column 250 mm × 4.6 mm I.D. 5 µm), and column temperature (30°C or 40°C) were all examined and compared. Finally, an optimized HPLC condition was determined by comparing comprehensively their solution, base line, elution time, and number of characteristic peaks.

Table 2.  The tried mobile phases in optimization of HPLC conditions.

Systems	Gradients
A: Acetonitrile	40% A in 90 min
B: Water
A: Acetonitrile	20–50% A in 60 min
B: Water
A: Acetonitrile (0.3‰ triethylamine)	(1) 20–30% A in 10 min
B: Water	(2) 30–36% A in 28 min
	(3) 36–50% A in 52 min
	(4) 50% A in 80 min
A: Acetonitrile (0.3‰ triethylamine)	(1) 10–15% A in 10 min
B: Water:methanol (8:2) (0.3‰ triethylamine)	(2) 15–30% A in 15 min
	(3) 30–34% A in 30 min
	(4) 34–50% A in 52 min
	(5) 50–55% A in 55 min
	(6) 55% A in 80 min

The main parameters needed to be controlled in the process of ELSD response optimization were the flow rate of nebulizer gas (pressure) and drift tube temperature (Cardenas et al., 1999; Lee et al., 2010; Liu et al., 2010). Under the fixed chromatographic conditions, two parameters were evaluated by the injection of linolic acid and degalactotigonin, which were the tested fatty acids and saponins for optimizing ELSD conditions, at different detector temperatures from 40°C to 80°C and the pressure from 2.0 to 3.5 bars, respectively. In this study, the drift tube temperature of 40°C and gas pressure of 3.0 bars were employed to determine the analytes via comparing peak area values. These optimized parameters resulted in a complete solvent evaporation and produced negligible base line noise in experiment. The optimal HPLC condition was shown in HPLC procedures.

HPLC fingerprints of S. nigrum

The reference fingerprint should be representative of the authentic S. nigrum in order to control herb quality. For this purpose, 41 batches of raw materials of S. nigrum from various locations, including nine-tenths of cultivation areas were analyzed by CASE to generate the representative standard fingerprint (Figure 2). The chromatogram for each sample was somewhat different from the representative fingerprint in peak area and/or numbers, which indicated different quality. About 16 peaks were shown in the standard chromatogram; 12 characteristic peaks among which were identified by comparing t_R with the reference standards (Figure 3).

Figure 2.  The representative fingerprint of Solanum nigrum [column: Phenomenex C18 column (5 µm, 250 × 4.6 mm); detector: ELSD].

Figure 3.  Comparison between representative fingerprints (A) and mixture of standards (B) [column: Phenomenex C18 column (5 µm, 250 × 4.6 mm); detector: ELSD]. a: Uttroside B, b: solanigroside J, c: 5α-pregn-16-en-3β-ol-20-one lycotetraoside, d: linolic acid, e: palmitic acid, f: oleic acid, g: solasonine, h: solamargine, i: β₂-solamargine, j: nigrumnin I, k: degalactotigonin; l: igogenin 3-O-β-[d-glucopyranosyl-(1→2)-O-[β-d-glucopyranosyl-(1→3)]-O-β-d-glucopyranosyl-(1→4)-O-β-d-galactopyranoside.

Discussion

The effects on herbs of various collecting locations

Forty-one batches of samples were extracted and the extracted solutions were injected into HPLC system. The obtained chromatograms were compared via the software presented in Data analysis. Based on the similarity values of all samples (Table 1), it was interesting that all samples could be divided into three groups: Group A, Group B, and Group C. The similarities of Groups A, B, and C were 0.2–0.5, 0.5–0.85, and 0.85–1.0, respectively. Group A consisted of four batches of samples collected from Guangxi, Guangdong, Shanghai, and Heilongjiang, and their representative fingerprint was shown in Figure 4A. Group B consisted of six batches of samples from Hainan, Dalian of Liaoning, Sinkiang, Hunan, and Anhui, and their representative fingerprint was shown in Figure 4B. Group C consisted of 23 batches of samples from other locations and their representative fingerprint was shown in Figure 4C. It was observed that the similarities of the herbs were closely relative among the samples from near locations, and highly different from those of other locations, suggesting the secondary metabolites of this species vary greatly among the locations. Moreover, even from the same collecting location, the similarity of herbs was much variably depended on the harvesting time; eight batches of samples were analyzed as described in the following section.

Figure 4.  Representative chromatographic fingerprints of Solanum nigrum. (A) Group A; (B) Group B; and (C) Group C [column: Phenomenex C18 column (5 µm, 250 × 4.6 mm); detector: ELSD].

The effects on herbs of harvesting time

Eight batches of samples were harvested every 15 days in June, July, August, September, and October in 2006, at the same location (Shenyang, Liaoning Province, China). The harvesting details and similarities of the chromatograms were shown in Table 1 and Figure 5. Interestingly, the similarities were distributed symmetrically. It was also clearly showed the growth cycle. The regular changes of secondary metabolites provided the guidelines for harvesting time. Based on Figure 5, the best harvesting time was from the last 15 days of July to the last 15 days of October.

Figure 5.  The similarities of Solanum nigrum in different harvesting time (each 15 days). 1: June 20, 2006; 2: July 5, 2006; 3: July 20, 2006; 4: August 5, 2006; 5: August 19, 2006; 6: September 4, 2006; 7: September 19, 2006; 8: October 5, 2006.

Conclusions

Above all, an unbiased, valid, and rapid chromatographic method and a new fingerprint analysis method were developed. Forty-one batches of herbs of S. nigrum were identified and distinguished. According to their similarities, those herbs were assorted to three groups. The taxonomy based on similarities had a fair consistency with the authentic chromatographic profiles (Figure 4).

Comparing the quantification of a few of markers or pharmacologically active constituents, the chromatographic fingerprint has more predominance in showing the authenticity of one kind of herb. For example, if only one peak and/or two peaks were quantified, Group B and Group C would not be distinguished distinctly. At present work, 12 characteristic peaks in chromatogram of S. nigrum were confirmed by comparison of reference compounds based on their retention time. Our result revealed that the chromatographic fingerprint we developed combining similarity evaluation could efficiently identify and distinguish raw herbs of S. nigrum from different sources.

The most important findings were that the most relevant factor on secondary metabolites of S. nigrum was the collecting location and the harvesting time. To get the consistent raw materials of S. nigrum, the collecting location should be fixed and then the harvesting time.

Acknowledgements

This work was supported by the Ministry of Science and Technology of Guizhou Province, China (G [2009]4013 and J [2009]2290, NY [2009]3021 and TZJF-2009-45). We would like to thank the Research Institute of Tsinghua University in Shenzhen for supporting the present work.

Declaration of interest

The authors explicitly declare that this article has no conflict of interest.