Anal. Chem. 2010, 82, 1580–1583
Letters to Analytical Chemistry Capillary Isoelectric Focusing with an Open Tubular Immobilized pH Gradient Chun Yang,* Shuangshou Wang, Chunyan Chang, Yang Wang, and Xiaoya Hu* College of Chemistry and Chemical Engineering, Yangzhou University, Yangzhou 225002, China Capillary isoelectric focusing (CIEF) was first performed in an open tubular mode. It depended on an immobilized pH gradients from the polymerization of several solutions, which contained acrylamide, N,N′-methylenebisacrylamide, allylamine, and acrylic acid. Different ratios of the basic and acidic monomers lead to solutions of various pHs. When injected into a capillary in the sequence of their pHs, the solutions polymerized and resulted into polymers with a series of isoelectric points (pIs). These polymers formed a covalently attached hydrophilic layer in the capillary, acting as both column bonding and carrier ampholytes. When five polymers were orderly lined up in a capillary, an immobilized pH gradient was established. This open tubular column was subjected to CIEF, showing excellent performance properties for protein focusing. Capillary isoelectric focusing separates ampholytic compounds, such as peptides and proteins, according to their pIs. It plays an important role in bioanlysis, especially in the era of postgenomics.1 Great methodological success has been made in the past few years. CIEF enables simultaneous separation and concentration of the analytes. Besides applications in the separation or purification of intact proteins and peptides, it shows a charming perspective in building two- or multidimensional methodologies for proteomic research.2-6 Powerful technical platforms have been made when coupling CIEF with mass spectrometry (MS). Such a platform can supply information of both pIs and molecular weights of target proteins, which greatly facilitates proteome profiling and biomarker discovery.7-14 * To whom correspondence should be addressed. E-mail:
[email protected]. (1) Silvertand, L. H. H.; Torao`o, J. S.; van Bennekom, W. P.; de Jong, G. J. J. Chromatogr., A 2008, 1204, 157–170. (2) Yang, C.; Liu, H.; Yang, Q.; Zhang, L.; Zhang, W.; Zhang, Y. Anal. Chem. 2003, 75, 215–218. (3) Kang, D.; Moon, M. H. Anal. Chem. 2006, 78, 5789–5798. (4) Tragas, C.; Pawliszyn, J. Electrophoresis 2000, 21, 227–237. (5) Chen, J.; Lee, C. S.; Shen, Y.; Smith, R. D.; Baehrecke, E. H. Electrophoresis 2002, 23, 3143–3148. (6) Herr, A. E.; Molho, J. I.; Drouvalakis, K. A.; Mikkelsen, J. C.; Utz, P. J.; Santiago, J. G.; Kenny, T. W. Anal. Chem. 2003, 75, 1180–1187. (7) Zhang, M.; Rassi, Z. E. J. Proteome Res. 2006, 5, 2001–2008. (8) Guo, T.; Rudnick, P. A.; Wang, W.; Lee, C. S.; Devoe, D. L.; Balgley, B. M. J. Proteome Res. 2006, 5, 1469–1478. (9) Zhou, F.; Johnston, M. V. Anal. Chem. 2004, 76, 2734–2740.
1580
Analytical Chemistry, Vol. 82, No. 5, March 1, 2010
Nowadays, MS ionization methods for proteins, i.e., electrospray and matrix-assisted laser desorption, always suffer from matrix interferences from the carrier ampholytes (CAs) when online coupled with CIEF. In fact, CIEF could be carried out without CAs by the electrolysis of water or other methods.15 Pawliszyn et al. recently reported that a pH gradient slightly different from that built by current commercial CAs.16 It completely got rid of the usage of CAs and would greatly improve the sensitivity of MS detection. Peptides and proteins are amphoteric compounds themselves. They can be protonated or deprotonated and migrate in an electric field with a base as the catholyte and an acid as the anolyte. However, without a special means to stabilize the formed pH gradient, peptides or proteins could not be separated effectively. They tended to stack in a limited length of the capillary during the focusing.17 Another method to avoid CAs is to perform CIEF in a monolithic immobilized pH gradient (M-IPG).18 Commercial CAs with amino groups reacted with the aldehyde-functionalized polymers in a monolithic column to build an immobilized gradient with a pH span similar to the CAs. Additionally methods could also be developed to make narrow19 and hydrophilic20 M-IPGs. Actually the pH scope and resolving power of such a gradient is dependent on the used commercial CAs. Besides focusing, some other mechanism, such as hydrophobic interaction between the proteins and the organic polymer media, exists during the analytes moving. It occasionally induces protein precipitation and makes it difficult to relate pIs with the migration time. (10) Fujita, M.; Hattori, W.; Sano, T.; Baba, M.; Someya, H.; Miyazaki, K.; Kamijo, K.; Takahashi, K.; Kawaura, H. J. Chromatogr., A 2006, 1111, 200–205. (11) Haselberg, R.; De Jong, G. J.; Somsen, G. W. J. Chromatogr., A 2007, 1159, 81–109. (12) Tang, Q.; Harrata, A. K.; Lee, C. S. Anal. Chem. 1997, 69, 3177–3182. (13) Jensen, P. K.; Pasˇa-Tolic´, L.; Anderson, G. A.; Horner, J. A.; Lipton, M. S.; Bruce, J. E.; Smith, R. D. Anal. Chem. 1999, 71, 2076–2084. (14) Wang, Y.; Rudnick, P. A.; Evans, E. L.; Li, J.; Zhuang, Z.; Devoe, D. L.; Lee, C. S.; Balgley, B. M. Anal. Chem. 2005, 77, 6549–6556. (15) Huang, T.; Wu, X.-Z.; Pawliszyn, J. Anal. Chem. 2000, 72, 4758–4761. (16) Zhan, Y.; Lemma, T.; Musteata, M. F.; Pawliszyn, J. J. Chromatogr., A 2009, 1216, 2928–2933. (17) Storms, H. F.; Van Der Heijden, R.; Tjaden, U. R.; Van Der Greef, J. Electrophoresis 2004, 25, 3461–3467. (18) Yang, C.; Zhu, G.; Zhang, L.; Zhang, W.; Zhang, Y. Electrophoresis 2004, 25, 1729–1734. (19) Zhu, G.; Yang, C.; Zhang, L.; Zhang, W.; Zhang, Y. Talanta 2006, 70, 2–6. (20) Zhu, G.; Yuan, H.; Zhao, P.; Zhang, L.; Liang, Z.; Zhang, W.; Zhang, Y. Electrophoresis 2006, 27, 3578–3583. 10.1021/ac902223y 2010 American Chemical Society Published on Web 02/11/2010
Table 1. Composition of the Monomer Solutions solution no. reagents (g)
1
2
3
allylamine acrylic acid acrylamide N,N′-methylenebisacrylamide APS pHa
0.2070 1.3043 0.1652 0.0518 0.2095 3.37
0.2180 1.3410 0.1645 0.0523 0.2014 5.56
0.0627 0.0890 0.1565 0.0512 0.2052 7.56
a
4
5
0.9552 0.9797 0.2484 0.2358 0.1719 0.1624 0.0524 0.0546 0.2063 0.2018 9.56 10.41
The pH values were measured before the addition of APS.
Compared with their packed partners, open tubular stationary phases retain analytes by just a thin film and exhibit a significant performance advantage. Because of the very small thickness of the layered stationary phases, open tubular columns endow fast mass transfer and quite low mechanical impedance. This enables great resolving powers and high separation speeds. Because of these favored features, open tubes have been widely applied in gas chromatography (GC),21 high-performance liquid chromatogrphy (HPLC),22 and capillary electrochromatography (CEC).23 As well as utilized in the analysis of small molecules, open tubular columns found extensive applications in the rapid analysis of peptides,24,25 proteins,26-30 biofilms,31 and even a whole cell.32 Here we report a new method to achieve an immobilized pH gradient in the form of open tubular style. Four monomers, allylamine, acrylamide, N,N′-methylenebisacrylamide, and acrylic acid were used to make a series of solutions with different pH values, from 3.37 to 10.41. In situ polymerization of the solutions in capillaries produced polymeric ampholytes with different pIs. The polymers not only modified the capillaries but also acted as focusing media. A rapid, CAs-free CIEF is available in this open tubular immobilized pH gradient. At the same time, nonspecific adsorption and protein precipitation are limited because of the hydrophilicity and open tubular feature of the polymers.
Figure 1. Schematic presentation of the preparation of an open tubular immobilized pH gradient.
monosodium phosphate, and urea were all analytical reagents from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Instrumentation. CIEF was carried out on a P/ACE MDQ system (Beckman Coulter, Fullerton, CA) equipped with a UV detector. The capillary (100 µm i.d., 360 µm o.d.) was purchased from Reafine Chromatographic Device Co., Ltd. (Yongnian, Hebei, China). Microscopic analysis was carried out on an Hitachi-S4800 (Kanagawa, Japan) at 15.0 kV. Preparation of the Open Tubular pH Gradients. A long capillary (∼3 m) was treated and functionalized with γ-MAPS as described previously.18 All monomers were dissolved in 20 mL of deionized water (Table 1). The solutions were ultrasonically degassed before the pHs measured. After the addition of APS and degassed again, three capillaries (32 cm) were filled with one of the solutions, 1, 3, and 5, respectively (Table 1). Another capillary was filled with all the five solutions in a turn of their pH values, each occupying an equal length (∼4.5 cm) in the capillary. A length of 10 cm remained without filling any solution. It was kept for the fabrication of a detection window. With both ends sealed with rubber septa, all the capillaries were put in a oven at 60 °C for 24 h. Sample Preparation and CIEF. Fresh egg white was diluted with an 8-time volume of phosphate buffer solution (20 mmol/L, pH 5.6, with urea in the concentration of 10 mmol/L). The sample solutions was filtrated and kept at 4 °C before use. Before electrophoresis, each capillary was rinsed with the sample buffer for 1 min. Sample injection was performed by a
EXPERIMENTAL SECTION Chemicals. Allylamine, acrylic acid, acrylamide, 3-methylpropyleneacylpropylmethoxysilane (γ-MAPS, 98%), and N,N′-methylenebisacrylamide were purchased from Haopeng Chemical Plant (Jinan, Shandong, China). Ammonium peroxydisulfate (APS), sodium hydroxide, sodium borate, disodium hydrogen phosphate, (21) Rykowska, I.; Wasiak, W. J. Chromatogr., A 2009, 1216, 1713–1722. (22) Hulthe, G.; Petersson, M. A.; Fogelqvist, E. Anal. Chem. 1999, 71, 2915– 2921. (23) Guihen, E.; Glennon, J. D. J. Chromatogr., A 2004, 1044, 67–81. (24) Wu, J.-T.; Huang, P.; Li, M. X.; Qian, M. G.; Lubman, D. M. Anal. Chem. 1997, 69, 320–326. (25) Yang, Y.; Boysen, R. I.; Matyska, M. T.; Pesek, J. J.; Hearn, M. T. W. Anal. Chem. 2007, 79, 4942–4949. (26) Wang, X.; Cheng, C.; Wang, S.; Zhao, M.; Dasgupta, P. K.; Liu, S. Anal. Chem. 2009, 81, 7428–7435. (27) Luo, Q.; Yue, G.; Valaskovic, G. A.; Gu, Y.; Wu, S.-L.; Karger, B. L. Anal. Chem. 2007, 79, 6174–6181. (28) Bakry, R.; Gjerde, D.; Bonn, G. K. J. Proteome Res. 2006, 5, 1321–1331. (29) Liu, H.; Stupak, J.; Zheng, J.; Keller, B. O.; Brix, B. J.; Fliegel, L.; Li, L. Anal. Chem. 2004, 76, 4223–4232. ¨o (30) Kuldvee, R.; D’ulivo, L.; Yohannes, G.; Lindenburg, P. W.; Laine, M.; O ¨rni, K.; Kovanen, P.; Riekkola, M.-L. Anal. Chem. 2006, 78, 2665–2671. (31) Chen, J.; Fallarero, A.; Ma¨a¨tta¨nen, A.; Sandberg, M.; Peltonen, J.; Vuorela, P. M.; Riekkola, M.-L. Anal. Chem. 2008, 80, 5103–5109. (32) Wang, K.; Marshall, M. K.; Garza, G.; Pappas, D. Anal. Chem. 2008, 80, 2118–2124.
Figure 2. SEM of the open tubular immobilized pH gradient. Analytical Chemistry, Vol. 82, No. 5, March 1, 2010
1581
Figure 3. CIEF of egg white proteins in capillaries bonded with polymers from a single solution: (a) pH 3.37, (b) pH 7.56, and (c) pH 10.41. Focusing voltage, 30 kV; detection wavelength, 280 nm; capillary length, 30.6 cm total, 20.5 cm effective; anolyte, NaOH (10 mmol/L); catholyte, H3PO4 (20 mmol/L).
Figure 4. CIEF of egg white proteins in the open tubular immobilized pH gradient. pI values are given in the brackets. The conditions are the same in Figure 3. (1) lysozyme (11.0), (2) avidin (10.0), (3) apoovotransferrin (6.67), (4) ovotransferrin (5.65), (5) ovoglobulin (5.16), (6) ovalbumin (4.71), and (7) ovomucoid (4.14).
pressure of 0.5 psi. After 13 min of focusing, the sample zones were mobilized mechanically by a pressure of 0.1 psi. RESULTS AND DISCUSSION The solutions listed in Table 1 contain nearly the same amount of neutral monomers and different ratios of basic/acidic monomers. The pH values were made to cover a range from 3.37 to 10.41, which is a little wider than the pH span of a commercial gradient, i.e., Ampholine 3.5-10.0. A solution from Table 1 probably produces a polymeric ampholyte with a pI different from its original pH. However, under almost the same polymerization conditions, it is feasible to make the pIs in line with the sequence of pHs of the solutions. For example, the solution pHs in Table 1 appear as pH1 < pH2 < pH3 < pH4 < pH5. Hence, the pIs of the polymers should hold the same order, pI1 < pI2 < pI3 < pI4 < pI5. On the basis of this hypothesis, it is possible to make an immobilized pH gradient by rational design, synthesis, and orderly array of these polymeric ampholytes in a capillary (Figure 1). Without the addition of N,N,N′,N′-tetramethylethylenediamine, the solutions were polymerized initiated by the radicals from the thermal decomposition of APS. This led to colorful liquids, always yellow or brown. Liquid polymers were superior to commonly formed gels, in making open tubular columns. The later usually result in blocked capillaries. Figure 2 is the scanning electron microscopic (SEM) image of the cross section of the capillary 1582
Analytical Chemistry, Vol. 82, No. 5, March 1, 2010
filled with five solutions. The attached polymer shows a macroporous structure. CIEF in such an open tubular immobilized pH gradient could avoid most nonspecific interactions between the analytes and the medium. Polymerizing a monomer solution does not produce just one polymer molecule. The product is often a mixture of compounds with analogous structures and close molecular weights. Thus, there is not a fixed pI figure but a pI distribution spanning a short scope. In other words, each solution in Table 1 would produce a polymer mixture and form a narrow pH gradient. Figure 3 exhibits CIEF of egg white proteins in three columns. In the column with polymeric ampholytes from an acidic solution (pH 3.37), most of the proteins cannot be focused in such a narrow gradient, which is a little above pH 3.0. Along the total column, the proteins are positively charged and migrate to the cathode. This is a process could be described as “pH gradient driving electrophoresis” (PDGE).33 According to the difference of their pIs, the proteins are divided into two groups in the electropherogram (Figure 3a). The peak at about 2 min presents a basic protein, i.e., lysozyme, in the sample. It differs greatly from the polymers and other proteins in pI, hence appears apart in the electropherogram. On the other hand, the neutral and acidic proteins are concentrated and passed the detection window during 15-18 min (Figure 3a). It is very similar in the case of the column filled with a nearly neutral solution (pH 7.56), except that the peak of lysozyme appeared a little later (∼3 min in Figure 3b). In the capillary originally filled with the basic solution (pH 10.41), lysozyme does not migrate in advance due to the pI approximation with the gradient. All components are condensed together and gave a single peak at about 6 min, lagging until 20 min (Figure 3c). It is obvious such a gradient from a single solution is too narrow to distinguish the proteins with various pIs. When all the solutions were subjected to in situ polymerization in a capillary, there were groups of polymers, of whom the pIs were distributed around five discrete values. As a result, a practical pH gradient was established. The same sample focused in this column gave separated peaks in the electropherogram (Figure 4), which is similar to the result of CIEF with commercial CAs. The small volume of an open tubular stationary phase might lead to a low sample load, which limited subsequent detection sensitivities in HPLC, especially CEC. In the case of the focusing (33) Yang, C.; Zhang, W.; Zhang, L.; Zhang, J.; Daun, J.; Zhang, Y. J. Sep. Sci. 2005, 28, 78–86.
in the open tubular IPG, the separation is not determined by the retaining of the analytes in the ampholyte layer. Proteins are protonated or deprotonated in the gradient, migrating at different speeds to an isoelectric point. If there are adequate field-driven H+ and OH- in the capillary, and the ionic concentration gradient can be stable at the presence of the layered polymeric ampholyte, the open tubular IPG is effective to focus a general amount of proteins compared with that dealt with in conventional focusing. As a conclusion, open tubular IPG is a novel separation column. It enables CIEF, free of CAs, alleviating protein precipitation and
without sacrificing the sample load. Further improvement of resolution, reproducibility, and robustness will make it a potent candidate in building multidimensional separation platforms for proteomic research. ACKNOWLEDGMENT This project was partly financially supported by SRF (MOE, China). Received for review October 1, 2009. Accepted November 15, 2009. AC902223Y
Analytical Chemistry, Vol. 82, No. 5, March 1, 2010
1583
