Integration of normal phase liquid chromatography with supercritical fluid chromatography for analysis of fruiting bodies of Ganoderma lucidum

Transcription

1 J. Sep. Sci. 2010, 33, Liang Gao 1,2 Jie Zhang 1 Weibing Zhang 1,3 Yichu Shan 1 Zhen Liang 1 Lihua Zhang 1 Yushu Huo 1 Yukui Zhang 1 1 Key Laboratory of Separation Science for Analytical Chemistry, National Chromatographic Research and Analysis Center, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian, 2 Graduate School of the Chinese Academy of Sciences, Beijing, 3 East China University of Science and Technology, Shanghai, Received June 22, 2010 Revised September 12, 2010 Accepted September 12, 2010 Short Communication Integration of normal phase liquid chromatography with supercritical fluid chromatography for analysis of fruiting bodies of Ganoderma lucidum In this study, a comprehensive 2-D chromatography was constructed, consisting of normal phase LC (NPLC) with a CN column as the first dimension, and supercritical fluid chromatography (SFC), with a Merck Chromolith Flash C 18 column as the second dimension, which were connected by a 10-port, dual-position valve controlled automatically by a self-designed software. Such platform was applied into the analysis of the fruiting bodies of Ganoderma lucidum, a traditional Chinese medicine, and within 2 h analysis, the obtained peak capacity of the 2-D-NPLC-SFC system was about 350, obviously higher than that of each dimension. These results demonstrate that 2-D-NPLC- SFC is not only of good orthogonality, but also of high throughput for the analysis of complex samples. Keywords: Comprehensive 2-D chromatography / Normal phase LC / Supercritical fluid chromatography / Traditional Chinese medicines DOI /jssc Introduction Complex samples require analytical methods characterized by an extremely high resolving power to provide thorough analysis of the sample components. Due to the restriction in separation capability of one-dimensional separation [1], much attention has been paid on multi-dimensional systems [2, 3]. Giddings has shown theoretically that the peak capacity could be greatly enhanced by coupling columns with different (orthogonal) retention mechanisms [4]. Until now, complex samples in research fields of proteomics, metabolomics, polymers, traditional Chinese medicines (TCMs), etc., have been analyzed with multi-dimensional systems, such as 2-D-IEC-RPLC [5 8], 2-D-SEC-RPLC [9, 10], 2-D-normal phase LC (NPLC)-RPLC [11], 2-D-HILIC-RPLC [12 14], and 2-D-RPLC-RPLC [15 17]. Recently, supercritical fluid chromatography (SFC) can be an alternative to HPLC because of its superior selectivity and speed. Most of the advantages of SFC compared to Correspondence: Professor Lihua Zhang, Key Laboratory of Separation Science for Analytical Chemistry, National Chromatographic Research & Analysis Center, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian , lihuazhang@dicp.ac.cn Fax: Abbreviations: SFC, supercritical fluid chromatography; TCM, traditional Chinese medicine HPLC arise from the lowered intramolecular energy of interaction of supercritical fluids [18], resulting in lower viscosity, higher diffusivity, and lower column pressure drops compared to HPLC [19]. Since diffusion coefficients are an order of magnitude higher in supercritical fluids than in liquids, compared to HPLC, the number of theoretical plates generated per unit time in SFC increases approximately threefold [20]. Van Deemter plots illustrate that the range of optimal linear velocity can be increased up to fivefold with SFC compared to HPLC [21]. In addition, supercritical fluid viscosities are lower than those of liquids, facilitating the use of longer columns in SFC while maintaining optimal conditions for high-speed separations [22]. SFC has been coupled with a number of separation techniques, such as SFC-SFC [23], SFC-GC [24], and SFE- SFC [25]. However, few studies have been reported about coupling SFC to HPLC, which is possibly due to the incompatibility of solvents and differences in operational pressure. Currently, TCMs have attracted considerable attention even in European and North American countries [26]. Ganoderma lucidum is one of the most famous medicinal funguses that have been widely used for thousands of years, which is composed of triterpenoids, polysaccharides, proteins, and other active small molecules [27]. Due to the complexity of TCMs, multi-dimensional separation techniques are required to improve the peak capacity [28 30]. In this article, a comprehensive 2-D-NPLC-SFC system is described for the separation of components in the fruiting bodies of G. lucidum. The resultant peak capacity was

2 3818 L. Gao et al. obtained in a short analysis time, which showed the high resolution and high throughput of the technique in the separation of complex mixtures. 2 Materials and methods 2.1 Chemicals and materials Hexane and methonal were purchased from Kemel (Tianjin, China). Isopropanol was bought from Shenlian (Shenyang, China). The mobile phase of supercritical CO 2 was ordered from Guangming Gas factory (Dalian, China). Syringe filters (0.45 mm 25 mm) were ordered from Millipore (Bedford, MA, USA). 2.2 Sample preparation The fruiting bodies of G. lucidum were dried, ground into powder, and stored at 41C prior to use. One gram of the powder was mixed with 10 ml methanol and ultrasonicated for 1 h at room temperature. After centrifugation at 4300 rpm for 0.5 h, the supernatant was filtered by a syringe filter and then freeze dried. The residues were redissolved in 1 ml methanol for NPLC or SFC analysis, and stored at 41C prior to use. 2.3 One-dimensional separation with NPLC and SFC NPLC separation was performed on an HPLC system (Jasco International, Kyoto, Japan), consisting of a PU-2089 quaternary gradient pump, an AS-2055 intelligent sampler and a Hypersil-CN column (5 mm, 120 Å, mm i.d.). The binary mobile phase was composed of hexane and isopropanol, with a gradient started from 0% isopropanol, programmed to 40% isopropanol over 30 min. The flow rate was set at 1 ml/min, and the injection volume was 20 ml. The effluents were detected at 254 nm by a UV detector. Without special statement, all the ratios represented volume ratios. The SFC system was composed of a PU-1580-CO 2 pump, a CO-2060 plus column thermostat, a BP back pressure regulator, a UV-1575 intelligent UV/VIS detector (Jasco International), and a Hypersil-C 18 (5 mm, 100 Å, mm i.d.). The apparent flow rate of CO 2 was set at 1 ml/min according to the data shown on the CO 2 pump, and the injection volume was 20 ml. The column was maintained at 401C, and the outlet pressure was set at 15 MPa to maintain the supercritical state of CO 2 throughout the column. The eluants were detected at 254 nm by the UV detector containing a high-pressure resistant cell D system Figure 1 shows the systematic diagram for the comprehensive 2-D-NPLC-SFC system. The key component of the system was a 10-port dual-position valve that enabled the continuous and alternate sampling the eluates from the primary column into the secondary one through two equivalent 200-mL sampling loops every 1 min electronically controlled by a self-designed software. For NPLC separation, the gradient of mobile phase was prolonged from 0 to 30% isopropanol over 120 min. The flow rate was set at 0.2 ml/min, and the injection volume was 100 ml. For SFC, a Merck Chromolith Flash C 18 column (Merck KGaA, Darmstadt, Germany) ( mm i.d.) was used instead of the Hypersil-C 18 column. The apparent flow rate of CO 2 was set at 2 ml/min. The column was maintained at 351C, and the outlet pressure was set at 10 MPa. 2.5 Data processing J. Sep. Sci. 2010, 33, The data obtained from UV detectors were exported into Origin 7.5, and converted to a matrix with rows corresponding to 1 min duration and data columns covering all successive second-dimensional separation. Then the matrix was exported Figure 1. Systematic diagram of comprehensive 2-D-NPLC-SFC system.

3 J. Sep. Sci. 2010, 33, Other Techniques 3819 into Fortner Transform version 3.4 software (Fortner, Savoy, IL, USA), and converted to a 2-D contour plot. 3 Results and discussion 3.1 Choice of separation modes Most components in the fruiting bodies of G. lucidum are small, and consist of low-/non-polarity molecules [31]. However, for one thing, the fatty acid esters and triterpenoids in the fruiting bodies of G. lucidum have strong retention on a C 18 column and could be eluted hardly, for another, high concentration of water as mobile phase could cause deterioration in chromatographic performance of SFC. Therefore, in our study, NPLC with a CN column and SFC with a C 18 column were chosen for the separation of the methanol extracts of the fruiting bodies of G. lucidum, respectively. Although the baseline rise of NPLC was caused by the continuously changed UV adsorption of the mobile phase under gradient elution in NPLC, as shown in Fig. 2, in a comprehensive 2-D system, all eluants from the first dimension are transferred to the second dimension for the further separation, and no peaks are missed. Furthermore, the noise of the baseline was rather low. Therefore, only few peaks were hidden in the NPLC baseline. Compared to NPLC, different peak patterns and higher separation efficiency could be obtained by SFC, as shown in Figs. 2 and 3. Generally in comprehensive 2-D system, the seconddimensional column should be performed at high speed to meet the rate of fractionation from the first-dimensional separation. Thus, due to the inherence advantages of superior selectivity, high throughput and high speed, SFC was chosen as the second dimension. Also, due to the high permeability and excellent mass-transfer properties of the skeleton [30], a monolithic silica C 18 column was chosen to replace the packed Hypersil-C 18 column in 2-D-NPLC-SFC to fasten the analysis speed. Furthermore, since the usually used mobile phase in NPLC, hexane and isopropanol could be dissolved in supercritical CO 2, the compatibility of mobile phases is better than 2-D-NPLC-RPLC. Therefore, 2-D-NPLC-SFC might be a good system to achieve the high resolution, high speed, and high peak capacity separation of complex samples. 3.2 Construction of 2-D-NPLC-SFC Although the pressure drop in SFC was lower than that in HPLC, owing to the low viscosity of supercritical fluids, the back pressure of the second dimension was still higher than that in the first dimension, to maintain the supercritical state of CO 2 (at least 7.3 MPa). Therefore, as shown in Fig. 1, a back pressure regulator was introduced in the waste pathway of the 10-port valve, to balance the system pressures between the first- and the second-dimensional separation. Analysis duration for the second dimension separation was an important factor for a comprehensive 2-D system. Although each eluant from NPLC could be completely transferred to the second-dimensional separation, within a short cycle of SFC, the injected fraction could hardly be well separated. However, with an increased SFC analysis cycle, the flow rate in NPLC would be too low to obtain enough sampling times of each peak, resulting in extremely long total analysis time. In this study, the cycle duration for SFC was optimized as 1 min. Consequently, the flow rate at 0.2 ml/min was chosen in NPLC, and 200-mL sampling loops were used for sampling to SFC, enabling the eluants from the primary column to be injected into the second one. After optimization, the apparent flow rate in SFC was set as 2 ml/min, to obtain high-speed and high-resolution separation. Figure 2. Chromatogram of methanol extracts from the fruiting bodies of G. lucidum analyzed by NPLC. Conditions were as described in Section 2.3. Figure 3. Chromatogram of methanol extracts from the fruiting bodies of G. lucidum analyzed by SFC. Conditions were as described in Section 2.3.

4 3820 L. Gao et al. J. Sep. Sci. 2010, 33, In addition, to improve the peak capacity of 2-D-NPLC- SFC, other experimental parameters were also optimized, among which the elution strength of the mobile phase in SFC played a significant role. Due to the low polarity of the analytes, pure CO 2 was chosen as the mobile phase without adding any organic solvents, such as methanol or acetonitrile. 3.3 Evaluation of 2-D-NPLC-SFC The 2-D-NPLC-SFC system was further evaluated by the analysis of methanol extracts from the fruiting bodies of G. lucidum. Only 17 and 34 peaks could be observed by onedimensional NPLC and SFC separation, respectively, as shown in Figs. 2 and 3. However, with the constructed 2-D-NPLC-SFC system, a total number of 250 peaks were observed by counting the peaks from the 120 secondary SFC chromatograms with 1 min duration, which demonstrated the improved resolution of such 2-D chromatography. Figure 4B shows one of the typical SFC chromatograms, Figure 5. 2-D contour plot of methanol extracts from the fruiting bodies of G. lucidum obtained by 2-D-NPLC-SFC. Conditions were as described in Section 2.4. in which a total of four peaks were recognized from peak 12 transferred from the first dimension of NPLC (as shown in Fig. 4A). Figure 5 shows the 2-D contour plots of the components in the fruiting bodies of G. lucidum analyzed by 2-D-NPLC- SFC. The x-axis and y-axis represented the retention time of the primary and the secondary columns, respectively, and the x-axis represented the intensity of UV signals, indicating the different component concentrations in the sample. In this figure, the peaks shown within 2 s of the second dimension (above the line) were caused by the baseline fluctuation during sample transfer, and those appeared within min of the first dimension and s of the second dimension were possibly caused by the decreased solubility of high-concentration isopropanol in supercritical CO 2. Besides such peaks, other peaks were all from components in the sample. The theoretical peak capacity was a key evaluation criterion for a comprehensive 2-D system. For one-dimensional separation, the average peak widths for NPLC and SFC were about 1.2 and 0.4 min, resulting in the theoretical peak capacity of 25 (30/1.2 min) and 50 (20/0.4 min), respectively. For 2-D-NPLC-SFC, the average peak width was about 1.7 min in NPLC, resulting in the peak capacity of 70 (120/1.7 min). In SFC, the peak capacity of 5 (1/0.2 min) was obtained. Therefore, the peak capacity was improved to 350 within 2 h analysis, which was obviously increased compared to one-dimensional separation. Furthermore, compared to other 2-D-HPLC for TCM analysis [28 30], the operation time of such platform was rather short, enabling the high-throughput analysis of TCMs. 4 Concluding remarks Figure 4. The amplified one-dimensional NPLC results of peak 12 in Fig. 2 (A) and the SFC chromatogram of peak 12 (B). Conditions were as described in Section 2.4. A comprehensive 2-D chromatography was constructed by NPLC with a CN column and an SFC with a Merck Chromolith Flash C 18 column. Compared to other HPLC

5 J. Sep. Sci. 2010, 33, Other Techniques 3821 modes, the low viscosity and high diffusivity of the mobile phase in SFC allowed increased flow rates and decreased pressure drops. Furthermore, with a monolithic column in SFC, high-speed analysis could be achieved without the loss of resolution, enabling enough sampling frequency from NPLC to SFC. Therefore, for such 2-D-NPLC-SFC platform, high resolution, high peak capacity, and high-throughput analysis of complex samples could be achieved. Furthermore, since the eluants from SFC could be easily splitted into MS with a tee valve, further study on the hyphenation of 2-D-NPLC-SFC with MS is underway, which could provide not only improved peak capacity, but also much qualitative information of TCMs. The authors are grateful for the financial support from National Key Technology R. & D. Program (2008BAK41B02). The authors have declared no conflict of interest. 5 References [1] Giddings, J. C., J. Chromatogr. 1995, 703, [2] Giddings, J. C., Anal. Chem. 1984, 56, [3] Bushey, M. M., Jorgenson, J. W., Anal. Chem. 1990, 62, [4] Giddings, J. C., J. High Resolut. Chromatogr. Commun. 1987, 10, [5] Wang, F. J., Dong, J., Jiang, X. G., Ye, M. L., Zou, H. F., Anal. Chem. 2007, 79, [6] Wang, F. J., Dong, J., Ye, M. L., Wu, R. A., Zou, H. F., J. Chromatogr. A 2009, 1216, [7] Zhai, B., Villen, J., Beausoleil, S. A., Mintseris, J., Gygi, S. P., J. Proteome Res. 2008, 7, [8] Chen, R., Jiang, X. N., Sun, D. G., Han, G. H., Wang, F. J., Ye, M. L., Wang, L. M., Zou, H. F., J. Proteome Res. 2009, 8, [9] Yuan, H. M., Zhang, L. H., Zhang, W. B., Liang, Z., Zhang, Y. K., J. Chromatogr. A 2009, 1216, [10] Yuan, H. M., Zhou, Y., Zhang, L. H., Liang, Z., Zhang, Y. K., J. Chromatogr. A 1216, [11] Tian, H. Z., Xu, J., Guan, Y. F., J. Sep. Sci. 2008, 31, [12] Liu, Y. M., Xue, X. Y., Guo, Z. M., Xu, Q., Zhang, F. F., Liang, X. M., J. Chromatogr. A 2008, 1208, [13] Wang, Y., Wang, J. S., Yao, M., Zhao, X. J., Fritsche, J., Schmitt-Kopplin, P., Cai, Z. W., Wan, D. F., Lu, X., Yang, S. L., Gu, J. R., Häring, H. U., Schleicher, E. D., Lehmann, R., Xu, G. W., Anal. Chem. 2008, 80, [14] Louw, S., Pereira, A. S., Lynen, F., Hanna-Brown, M., Sandra, P., J. Chromatogr. A 2008, 1208, [15] Zhang, J., Jin, Y., Liu, Y. F., Xiao, Y. S., Feng, J. T., Xue, X. Y., Zhang, X. L., Liang, X. M., J. Sep. Sci. 2009, 32, [16] Song, C. X., Ye, M. L., Han, G. H., Jiang, X. N., Wang, F. J., Yu, Z. Y., Chen, R., Zou, H. F., Anal. Chem. 2010, 82, [17] Liu, Y. M., Guo, Z. M., Jin, Y., Xue, X. Y., Xu, Q., Zhang, F. F., Liang, X. M., J. Chromatogr. A 2008, 1206, [18] Berger, T. A., Packed Column SFC, The Royal Society of Chemistry, London 1995, pp [19] Berger, T. A., Packed Column Supercritical Fluid Chromatography, Course Manual, Lake Tahoe, NV [20] Wenclawiak, B., Analysis with Supercritical Fluids: Extraction and Chromatography, Springer-Verlag, Berlin [21] Berger, T. A., Wilson, W. H., Anal. Chem. 1993, 65, [22] Berger, T. A., Packed Column SFC, The Royal Society of Chemistry, London 1995, pp [23] Juvancz, Z., Payne, K. M., Markides, K. E., Lee, M. L., Anal. Chem. 1990, 62, [24] Venter, A., Rohwer, E. R., Anal. Chem. 2004, 76, [25] Engelhardt, H., Gross, A., J. High Resolut. Chromatogr. Commun. 1988, 11, [26] Lee, K. H., J. Nat. Prod. 2004, 67, [27] Lin, Z. B., Modern Research on Ganoderma lucidum (Chinese). Beijing medical university press, Beijing 2001, p. 5. [28] Zhang, X. M., Hu, H. L., Xu, S. Y., Yang, X. H., Zhang, J., J. Sep. Sci. 2001, 24, [29] Chen, X. G., Kong, L., Su, X. Y., Fu, H. J., Ni, J. Y., Zhao, R. H., Zou, H. F., J. Chromatogr. A 2004, 1040, [30] Zhang, J., Tao, D. Y., Duan, J. C., Liang, Z., Zhang, W. B., Zhang, L. H., Huo, Y. S., Zhang, Y. K., Anal. Bioanal. Chem. 2006, 386, [31] Editorial Board of China Herbal, State Administration of Traditional Chinese Medicine, China Herbal. Shanghai Scientific and Technical Publishers, Shanghai 1999.