超高效合相色谱 串联质谱法快速测定枸杞中甜菜碱含量 , 师彦平 (1. 中国科学院兰州化学物理研究所, 中国科学院西北特色植物资源化学重点实验室和

Transcription

1 2018 年 5 月 Vol.36 No.5 May 2018 Chinese Journal of Chromatography 417 ~ 424 Article DOI: / SP.J Fast determination of betaine in Lycii Fructus by ultra performance convergence chromatography tandem mass spectrometry ZHANG Xianfei 1,2, YANG Junli 1, CHEN Juan 1, SHI Yanping 1 (1. CAS Key Laboratory of Chemistry of Northwestern Plant Resources and Key Laboratory for Natural Medicine of Gansu Province, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences (CAS), Lanzhou , China; 2. University of Chinese Academy of Sciences, Beijing , China) Abstract: Over the past 2500 years, Lycii Fructus has been widely used as a functional food and tonic in Chi nese herbal medicine. It can nourish the liver and kidneys, moisten the lungs, and improve eyesight. In this study, a new rapid and sensitive method has been developed for the quantitative determination of betaine ( index compound) in Lycii Fructus by using ultra performance convergence chromatography tandem mass spectrometry (UPC 2 MS). The separation of betaine was successfully achieved on an ACQUITY UPC 2 BEH 2 EP column (150 mm 2 1 mm, 1 7 μm), with isocratic elution by a CO 2 methanol (80 20, v / v) solvent at a flow rate of 0 7 ml / min. The conditions used in the separation process were as follows: modifier, 0 1% (v / v) for mic acid in methanol; column temperature, 40 ; backpressure, Pa; injection volume, 1 μl; and retention time, 3 min. The MS system was equipped with an electrospray ionization ( ESI) ion source and oper ated in the selected ion recording ( SIR) and positive ion mode. Under the abovementioned conditions, the cali bration curve was obtained. The linear range of detection was μg / ml, with a correlation coefficient of , and the limit of detection ( LOD) was found to be μg / ml. The validity of the method was tested by analyses of precision, repeatability, stability, and accuracy (average recovery: 96 3%). Finally, the developed method was applied to analyze 11 batches of samples. The results indicated that this method was suitable for evaluating the quality of Lycii Fructus. Key words: ultra performance convergence chromatography tandem mass spectrometry (UPC 2 MS); betaine; Lycii Fructus; quality control CLC number: O658 Document code: A Article IC: (2018) 超高效合相色谱 串联质谱法快速测定枸杞中甜菜碱含量 张弦飞 1,2, 杨军丽 1, 陈娟 1 1, 师彦平 (1. 中国科学院兰州化学物理研究所, 中国科学院西北特色植物资源化学重点实验室和 甘肃省天然药物重点实验室, 兰州 ; 2. 中国科学院大学, 北京 ) 摘要 : 枸杞作为传统的中药材, 具有滋补肝肾 益精明目的功效, 亦可作为功能性食品被广泛食用 基于超高效合 相色谱 质谱 (UPC 2 MS), 建立了一种快速 灵敏地定量测定枸杞中甜菜碱 ( 指标化合物 ) 含量的新方法 采用 ACQUITY UPC 2 BEH 2 EP 色谱柱 (150 mm 2 1 mm, 1 7 μm), 以 0 7 ml / min 超临界 CO 2 甲醇 (80 20, v / v) 等 度洗脱, 成功分离了枸杞中的甜菜碱 在该分离过程中,0 1% (v / v) 甲酸作为改性剂 ; 背压为 Pa; 柱温为 40 ; 进样体积为 1 μl; 保留时间为 3 min 质谱检测工作于电喷雾 ( ESI) 正离子模式和选择离子监测 ( SIR) 模 式 在上述条件下得到线性回归方程 其相关系数为 , 检测范围为 0 5 ~ 50 0 μg / ml, 检出限为 Received date: Corresponding author. Tel: ; E mail: chenjuan@ licp. cas. cn ( CHEN Juan), shiyp@ licp. cas. cn ( SHI Yan ping). Foundation item: National Natural Science Foundation of China (Nos , ); the Top Priority Program of One Three Five Strategic Planning of Lanzhou Institute of Chemical Physics of Chinese Academy of Sciences.

2 418 色谱第 36 卷 μg / ml 随后, 通过对精度 重复性 稳定性和加标回收率 ( 平均值 96 3%) 的分析, 验证了该方法的有效性 最后, 将所建立的方法应用于 11 批样品的分析 结果表明, 该方法可较好地用于枸杞的质量控制与分析评价 关键词 : 超高效合相色谱 串联质谱 ; 甜菜碱 ; 枸杞 ; 质量控制 Lycium barbarum L., which belongs to the So lanaceae family, is widely distributed in the wild in most regions of China and is mainly cultivated in Northwestern China, including Ningxia, Gan su, Qinghai, Inner Mongolia, and Xinjiang [1,2]. Lycii fructus is widely used as a functional health food [3] or as a valuable tonic in Chinese herbal medicine for nourishing the liver and kidneys, moistening the lungs, and improving eyesight [ 4, 5]. Nowadays, Lycii fructus is marketed as a diet ary supplement and functional tea in many coun tries in Asia, Europe, and North America [ 6]. Betaine is one of the major functional components in Lycii fructus and is a zwitterionic quaternary ammonium compound. It has many beneficial effects on the human body, including anti athero sclerosis, anti osteoporosis, anti tumor effects; further, it lowers the blood pressure, heals peptic ulcers, and protects the liver against chemical in jury [7]. Both the Chinese Pharmacopoeia [ 8] and the Korean Pharmacopoeia [ 9] stipulate betaine as the index compound of Lycii Fructus. In the Chi nese Pharmacopoeia, the method for determina tion of betaine in Lycii Fructus is stipulated as thin layer chromatography ( TLC) scanning. The precision of TLC is not sufficient, and it is typi cally used for qualitative analysis. Furthermore, tailing often occurs during chromatography. In re cent years, some quantitative analysis methods have been developed. Shin et al. [10] determined betaine in Lycium chinense fruits by liquid chro matography electrospray ionization mass spec trometry ( LC ESI MS) and observed high base line noise. Huang et al. [11] analyzed the betaine content in Lycii Fructus by high performance liq uid chromatography (HPLC) using a Luna 5u SCX 100A column and an evaporative light scattering detector (ELSD), but the target peak of betaine was not well separated from the other peaks. Zhao et al. [12] used an HILIC column and ELSD detector to analyze betaine in Lycii Fructus. The obtained peak shape of betaine was acceptable, but the retention time was slightly long, about 20 min. Besides Lycii Fructus, the determination of betaine in other matrices such as plasma [ 13], feed premixes [14], and human urine [15] was also reported, and good results were obtained. In 2012, Waters Company launched ultra per formance convergence chromatography ( UPC 2 ), which integrated the advantages of both supercrit ical fluid chromatography ( SFC) and ultrahigh performance liquid chromatography ( UPLC). The mobile phase contained supercritical carbon diox ide, which is an inert fluid and a small amount of organic solvents [ 16]. In the supercritical state, the polarity of carbon dioxide is similar to that of hexane, so a small amount of modifier is com monly added to increase the separation efficiency [17]. UPC 2 has many advantages, such as short analysis time, simple pretreatment, high sensitivi ty, good reproducibility, minimal consumption of organic reagents, and environmental friendliness. Nowadays, UPC 2 is widely used in medicine [18-21], food [22,23], biology [24,25], chemistry, and other fields [26-28]. Until now, there are only a few notable reports on the determination of betaine in Lycii Fructus, but there is no detailed research based on UPC 2. The main purpose of this work was to establish a new quantitative analysis method based on UPC 2 MS. 1 Experimental 1.1 Reagents and materials In 2016, 11 batches of dried fruits of L. barba rum were collected from different regions of Ningxia and Gansu provinces in China and

3 第 5 期 ZHANG Xianfei, et al: Fast determination of betaine in Lycii Fructus by ultra performance convergence chromatography tandem mass spectrometry 419 identified by Prof. Lin Jin, Gansu University of Traditional Chinese Medicine, Gansu, China. The betaine standard ( purity > 98 0%) was purchased from the National Institutes for Food and Drug Control ( Beijing, China). Chromatographic grade methanol was obtained from Merck Co. Ltd. ( Darmstadt, Germany), and all other chemicals of analytical grade were provided by Tianjin Chemical Reagent Co. ( Tianjin, China ). Water was purified using an OKP TS purification system ( Exceed AC 16, Shanghai Laikie Instrument Co. Ltd., China). 1.2 Chromatographic system UPC 2 analysis was performed on an ACQUITY UPC 2 system ( Waters, Milford, MA, USA ), which was equipped with a binary solvent delivery pump, a fixed loop autosampler, a column ther mostat, and an automated backpressure regulator (ABPR). The separation of betaine was success fully achieved on an ACQUITY UPC 2 BEH 2 EP column (150 mm 2 1 mm, 1 7 μm) by isocratic elution with a CO 2 methanol (80 20, v / v) mix ture at a flow rate of 0 7 ml / min. The following were the conditions for this separation process: modifier, 0 1% formic acid in methanol; column temperature, 40 ; injection volume, 1 μl; and ABPR, Pa. The system was coupled to an MS apparatus e quipped with an ESI source, and the quantifica tion of betaine was performed by MS analysis in selected ion recording ( SIR) mode and positive ion mode. The MS analysis conditions were opti mized as follows: source temperature, 120 ; desolvation temperature, 350 ; capillary volt age, 2 3 kv; cone voltage, 20 V; cone gas flow rate, 150 L / h; and desolation gas (nitrogen) flow rate, 600 L / h. Data acquisition and system control were performed using a MassLynx 4 1 worksta tion (Waters, USA). 1.3 Preparation of standard solutions 12 0 mg of betaine was accurately weighed and dissolved in methanol in a 25 ml volumetric flask to prepare the stock solution. In order to obtain the calibration curve, the stock solution was fur ther diluted to 0 5, 1 0, 2 0, 5 0, 10 0, 20 0, and 50 0 μg / ml with methanol, filtered through a 0 22 μm millipore membrane, and stored at 4 before use. 1.4 Preparation of sample solutions 50 0 mg of the accurately weighed sample from each batch of Lycii Fructus was air dried, crushed into powder, and extracted with 50 ml methanol in an ultrasonicator for 1 h. The extracts were separated in a centrifuge machine ( TDL 5 A, Shanghai Anting Scientific Instrument Factory, China) at r / min for 10 min. The obtained supernatants were concentrated to 5 ml using a vacuum rotavapor ( EYELA N 1001, EYELA Co. Ltd., China). The resulting sample solutions were filtered through a 0 22 μm millipore membrane filter and injected into the UPC 2 MS system for analysis. The betaine contents were calculated ac cording to the calibration curve, and the yields were the mass ratios of the extracted betaine to the weighed medicinal herbs. 1.5 Method validation The limit of detection (LOD) and limit of quan tification ( LOQ) were defined by the signal to noise ratios of 3 and 10, respectively. Precisions within one day ( intraday) and within three con secutive days ( interday) were analyzed by repli cate injections of the same sample solution ( n = 6). Stability was tested at ~ 2 h intervals for three days. Repeatability was evaluated by the relative standard deviation of six different sample solu tions, which were prepared in parallel with the same batch of sample S 01. Accuracy was investi gated by spiking sample S 01 with three concen tration levels of betaine standards, and the ex traction and analysis were repeated three times for each level. 2 Results and discussion 2.1 Optimization of chromatographic condi tions To obtain good peak shapes and shorten the

4 420 色谱第 36 卷 retention time of betaine, the optimal chromato graphic conditions were investigated. For UPC 2, the mobile phase, column type, injection volume ( μl), column temperature (35-55 ), flow rate ( ml / min), and backpressure ( Pa) were analyzed. For MS, the ionization mode, capillary voltage ( kv), cone voltage (10-50 V), desolvation tem perature ( ), and desolvation gas ( ni trogen) flow rate ( L / h ) were examined. A suitable chromatographic column plays an important role in the acceptable separation of components. As shown in Fig. 1a, three columns, Waters ACQUITY UPC 2 BEH 2 EP (150 mm 2 1 mm, 1 7 μm), Waters ACQUITY UPC 2 BEH C18 (100 mm 3 0 mm, 1 7 μm), and Waters ACQU ITY UPC 2 CSH Fluoro Phenyl (150 mm 2 1 mm, 1 7 μm), were screened. The Waters ACQUITY UPC 2 BEH 2 EP column is filled using hybrid sili ca with 2 ethylpyridine bonding. The hydrophilic interactions in the silica column are conducive for the elution of alkaloids [ 29 ]. As indicated in Fig. 1a, this column was proved to be the best since it possessed a basic character at its surface, and was chosen for subsequent analyses. Different modifiers ( methanol, methanol: isopropanol, and methanol: acetonitrile) and additives ( formic acid, ammonium formate, and ammonium ace tate) were screened, and 0 1% ( v / v ) formic acid in methanol was found to be the best. Chan ging the injection volume or the cone voltage hardly influenced the retention time but signifi cantly changed the peak shape. As shown in Figs. 1b and c, the peak height and peak width increased with an increase in the injection volume or cone voltage, and the peak shape worsened. Given that larger areas are required for quantita tive analysis, the injection volume and cone voltage were selected as 1 μl and 20 V, respectively. Density is the main factor contributing to reten tion and selectivity, and it depends on three pa rameters: mobile phase composition, column temperature, and pressure [ 30,31]. Temperature is known to affect the distribution constant, and thereby, the retention and separation factors of the analytes. The density of CO 2 decreases with an increase in temperature, so the elution capaci ty decreases and the retention time increases [32]. As presented in Fig. 1d, the retention time increased slightly when the temperature was in creased to 55 ; further, the peak became wider as the temperature increased. This was because temperature variations would modify the chemical Fig. 1 Effects of (a) column type, (b) injection volume, (c) cone voltage, and (d) column temperature on the peak shape of betaine

5 第 5 期 ZHANG Xianfei, et al: Fast determination of betaine in Lycii Fructus by ultra performance convergence chromatography tandem mass spectrometry 421 characteristics of the analytes and the stationary phase, leading to a change in selectivity. In this work, the column temperature was set at 40. The backpressure hardly impacted the peak shape but had a notable influence on the retention time in this work. A higher backpressure would increase the density of and solvation power of the mobile phase solvation power, thereby shortening the retention time [ 33]. As seen in Fig. 2a, the retention time decreased with an increase in back pressure, so the backpressure was selected as Pa. Under operation at high temperature and low backpressure, the fluid will behave like a gas, thus causing problems with repeatability and sensitivity [ 34 ]. Hence, intermediate values of temperature and backpressure may be more suit able. In Ref. [34], 40 and Pa were suggested as the temperature and backpressure, respectively, which were consistent with the re sults of our investigation. In this study, the flow rate, too, had no clear influence on the peak shape. However, as seen from Fig. 2b, the retention time would decrease with an increase in the flow rate. It is well known that the inlet pressure increases with increasing flow rate, and this leads to an increased density of the mobile phase in the column. Thus, the re tention time decreases [33]. Given that increasing the flow rate by 0 1 ml / min would result in an increase of about Pa in the system pres sure and very high pressure would hinder system maintenance, the flow rate was selected as 0 7 ml / min. Other parameters such as desolvation gas flow rate and desolvation temperature were also investigated and demonstrated to have minor effects on the sensitivity. Fig. 2 Effects of (a) backpressure and (b) flow rate on the retention time of betaine The detection of betaine was performed with the ESI source and in SIR mode; the obtained UPC 2 MS total ion chromatograms are presented in Fig. 3. As shown in Fig. 3a, the retention times of the standard and the sample solutions in posi tive ion mode agreed well with one another, and Fig. 3 UPC 2 MS total ion chromatograms obtained in (a) positive ion mode and (b) negative ion mode

6 422 色谱第 36 卷 good peak shapes were observed. These results indicated that this mode was suitable for betaine detection. As presented in Fig. 3b, UPC 2 MS anal ysis in negative ion mode was also attempted, but this did not give acceptable results. The mecha nism for the fragmentation pathways of betaine in the mass spectrum is shown in Fig. 4. In the posi tive ion mode, the mass to charge ratios ( m / z) were detected based on the formation of [ M + Na] + ions. Fig. 4 Mass scan spectrum in positive ion mode and fragmentation pathways of betaine 2.2 Optimization of sample preparation con ditions Using a mono factorial experimental design, four sample preparation conditions ( solvent, ex traction time, sample amount, and extraction method) were optimized. Methanol and ethanol at six volume percentages ( 50%, 60%, 70%, 80%, 90%, and 100%) were investigated in sample as says using an ultrasonicator, and 100% methanol was proved to be the best. Five extraction times ( 30, 45, 60, 75, and 90 min) were examined with 100% methanol as the solvent, and the ex traction was saturated after 60 min. Samples of six different amounts (30, 40, 50, 60, 70, and 80 mg) were extracted with 50 ml methanol for 60 min, and the extraction efficiency for 50 mg was proved to be the best. Finally, two extraction methods ( reflux and ultrasonication ) were at tempted, and the extraction efficiency with the ultrasonicator was found to be slightly better. 2.3 Method validation The calibration curve was established by linear regression analysis, and the calibration range was μg / ml. The regression equation was expressed as y = x , where y is the peak area and x is the concentra tion ( μg / ml) of betaine. The correlation coeffi cient ( r 2 ) of implied that the linear re gression of the calibration curve was good within the calibration range; the low LOD ( μg / ml) and LOQ ( μg / ml ) meant that overload and pollution of the chromatographic columns could be avoided. In Ref. [12], betaine in Lycii Fructus was analyzed using HPLC ELSD, and the LOD was 2 31 μg / ml. In Ref. [15], be taine was determinated using UPLC MS / MS, and the LODs for plasma and urine were 0 03 and 3 00 μg / ml, respectively. In the present work, the LOD was μg / ml, which was much low er than the values obtained with HPLC and UPLC. With the present method, the relative standard deviations ( RSDs) of the intraday and interday (three days) precisions were found to be 0 9% and 1 1% ( n = 6), respectively, which indicated high precision of the instrument. The stability of the sample was reasonable over the tested period (RSD, 1 6% ), and the repeatability of the meth od was acceptable ( RSD, 1 4% ). The accuracy was validated by the recovery ( R), which was calculated using the following formula: R = [( found amount - original amount ) / spiked amount] 100%. As shown in Table 1, the average

7 第 5 期 ZHANG Xianfei, et al: Fast determination of betaine in Lycii Fructus by ultra performance convergence chromatography tandem mass spectrometry 423 recovery of 96 3% with RSD of 0 5% demonstra ted the high accuracy of the method. Table 1 Analyte Study of recovery for validation of accuracy Spiked level / μg Recovery / % Average / % RSD / % Betaine ± ± ±0.56 Recovery: mean±standard deviation (SD); n = Sample analysis The developed UPC 2 MS method was applied to analyze the betaine contents in 11 batches of Lycii Fructus samples, which were collected from elev en different counties of Ningxia and Gansu prov inces in China. In Table 2, the obtained betaine contents are listed. It could be seen that the beta ine contents in the Lycii Fructus samples from dif ferent counties were slightly different. The betaine content of sample S 01 was the best ( 1 39 μg / mg), while that of S 09 was the worst (1 20 μg / mg). This was because the difference in the growing conditions, such as growing regions, growing years, and harvest seasons, would result in variations in quality [2]. Table 2 Betaine contents of 11 batches of Lycii Fructus samples Sample No. County Content / (μg / mg) S 01 Zhongning 1.39±0.08 S 02 Guyuan 1.36±0.21 S 03 Yinchuan 1.31±0.09 S 04 Zhongwei 1.36±0.14 S 05 Pingluo 1.32±0.19 S 06 Shizuishan 1.33±0.22 S 07 Tongxin 1.35±0.31 S 08 Jingyuan 1.24±0.17 S 09 Jiuquan 1.20±0.21 S 10 Jinta 1.29±0.18 S 11 Minqin 1.31±0.26 Content: mean±sd; n = 3. 3 Conclusions In the present study, a novel UPC 2 MS method operating in positive ion mode was developed to analyze the betaine contents of Lycii Fructus sam ples. The results showed that the separation of betaine was successful on an ACQUITY UPC 2 BEH 2 EP column, and the UPC 2 MS total ion chromatograms showed good baseline resolution and peak shapes within a short time (3 min). The calibration curve was established, and the method validation indicated that the precision, repeat ability, stability, and accuracy of this method were excellent. In particular, the LOD was much lower than that in previous reports. The devel oped UPC 2 MS method was further applied to an alyze the quality of eleven batches of Lycii Fruc tus samples. The betaine contents in samples from different areas were slightly different. The present method is thus expected to be very useful for evaluating the quality of Lycii Fructus. References: [1] Wu D T, Cheong K L, Deng Y, et al. Carbohyd Polym, 2015, 134: 12 [2] Lu W, Jiang Q, Shi H, et al. J Agric Food Chem, 2014, 62: 9073 [3] Seeram N P. J Agric Food Chem, 2008, 56: 627 [4] Potterat O. Planta Med, 2010, 76: 7 [5] Amagase H, Farnsworth N R. Food Res Int, 2011, 44: 1702 [6] Duan H, Chen Y, Chen G. J Chromatogr A, 2010, 1217: 4511 [7] Craig S A S. Am J Clin Nutr, 2004, 80: 539 [8] The Pharmacopoeia Commission of the People s Republic of China. Pharmacopoeia of People s Republic of China, Part 1. Beijing: China Medical Science Press, 2015: 249 [9] Korea Food and Drug Administration. Korean Pharmacopoe ia IX, Part II. Seoul: Shinil Books, 2007: 913 [10] Shin Y G, Cho K H, Kim J M, et al. J Chromatogr A, 1999, 857: 331 [11] Huang H, Chen X S, Liao Q. Chinese Journal of Experi mental Traditional Medical Formulae, 2010, 16: 59 [12] Zhao B T, Jeong S Y, Hwangbo K, et al. Arch Pharm Res, 2013, 36: 1231 [13] Holm P I, Ueland P M, Kvalheim G, et al. Clin Chem, 2003, 49: 286 [14] Suo D, Li L, Zhang S, et al. Anal Methods, 2013, 5: 59 [15] Ocque A J, Stubbs J R, Nolin T D. J Pharm Biomed Anal, 2015, 109: 128 [16] Berger T A. J Chromatogr A, 2015, 1421: 171 [17] Zhu L L, Zhao Y, Xu Y W, et al. J Pharm Biomed Anal, 2016, 120: 72 [18] Breitenbach S, Rowe W F, McCord B, et al. J Chromatogr A, 2016, 1440: 201 [19] Hicks M B, Regalado E L, Tan F, et al. J Pharm Biomed Anal, 2016, 117: 316 [20] Ganipisetty V N R, Ravi B, Reddy C R, et al. Anal Meth ods, 2015, 7: 1092 [21] Berger T A, Berger B K. Chromatographia, 2013, 76: 1631 [22] Lin C, Xie X, Fan N, et al. Chinese Journal of

8 424 色谱第 36 卷 Chromatography, 2015, 33: 397 [23] Said A B, Guinot C, Ruiz J C, et al. J Supercrit Fluid, 2016, 110: 22 [24] Perrenoud A G G, Guillarme D, Boccard J, et al. J Chrom atogr A, 2016, 1450: 101 [25] Hegstad S, Havnen H, Helland A, et al. J Chromatogr B, 2017, 1061 / 1062: 103 [26] Lesellier E, Mith D, Dubrulle I. J Chromatogr A, 2015, 1423: 158 [27] Dai X, Wei B, Wang X, et al. Chinese Journal of Chroma tography, 2015, 33: 1059 [28] Li W, Li X, Li G, et al. Chinese Journal of Chromatogra phy, 2016, 34: 795 [29] West C, Lemasson E, Bertin S, et al. J Chromatogr A, 2016, 1440: 212 [30] West C, Bouet A, Routier S, et al. J Chromatogr A, 2012, 1269: 325 [31] Åsberg D, Enmark M, Samuelsson J, et al. J Chromatogr A, 2014, 1374: 254 [32] Li K, Fu Q, Xin H, et al. Analyst, 2014, 139: 3577 [33] Jumaah F, Larsson S, Essén S, et al. J Chromatogr A, 2016, 1440: 191 [34] Desfontaine V, Guillarme D, Francotte E, et al. J Pharm Biomed Anal, 2015, 113: 56