Anal. Chem. 1999, 71, 1786-1791
Use of a Mixed-Mode Packing and Voltage Tuning for Peptide Mixture Separation in Pressurized Capillary Electrochromatography with an Ion Trap Storage/Reflectron Time-of-Flight Mass Spectrometer Detector Peiqing Huang, Xiaoying Jin, Yajuan Chen, Jannavi R. Srinivasan, and David M. Lubman*
Department of Chemistry, The University of Michigan, Ann Arbor, Michigan 48109-1055
A mixed-mode (reversed-phase/anion-exchange) stationary phase has been used as the capillary column packing for investigation of the separation of peptide mixtures in pressurized capillary electrochromatography (pCEC). This stationary phase contains both octadecylsilanes and dialkylamines. The amine groups of the stationary phase determine the charge density on the surface of the packing and can produce a strong and constant electroosmotic flow (EOF) at low pH. A comparison was made in terms of the capability of separating tryptic digests between the mixed-mode phase and C18 reversed phase. In addition, the constant EOF enabled the tuning of the retention and the selectivity of the separation by adjusting the mobile phase pH from 2 to 5. Furthermore, the magnitude and the polarity of the electric voltage were demonstrated to greatly influence the elution profiles of the peptides in pCEC. An ion trap storage/reflectron time-of-flight mass spectrometer was used as an on-line detector in these experiments due to its ability to provide rapid and accurate mass detection of the sample components eluting from the separation column. Pressurized capillary electrochromatography (pCEC) is a novel analytical method in which both pressure and electric field are applied to a packed capillary to achieve separation of analytes. pCEC combines various aspects of capillary electrochromatography (CEC) and liquid chromatography (LC). In pCEC, an electroosmotic flow (EOF) caused by the voltage is superimposed on a pressure-induced hydrodynamic flow. Therefore, the separation efficiency is intermediate between that of pure CEC and LC. The major advantages of the method over pure CEC include the stability of the mobile-phase flow due to bubble suppression by the application of pressure, the increased speed of separation, and the enhanced selectivity for charged analytes. This method has been used to separate a wide range of compounds.1-5 (1) Eimer, T.; Unger, K. K.; Tsuda, T. Fresenius J. Anal. Chem. 1995, 352, 649-653. (2) Dekkers, S. E. G.; Tjaden, U. R.; Greef, J. J. Chromatogr., A 1995, 712, 201-209. (3) Behnke, B.; Bayer, E. J. Chromatogr., A 1994, 680, 93-98. (4) Kitagawa, S.; Tsuda, T. J. Microcolumn Sep. 1994, 6, 91-96.
1786 Analytical Chemistry, Vol. 71, No. 9, May 1, 1999
Most CEC applications have dealt with the separation of neutral compounds,6-8 and octadecylsilane (ODS) bonded silica has been the most commonly used stationary-phase material in CEC. Relatively high pH7,8 is usually used to generate strong electroosmotic flow to drive the analytes to the detector. Ionizable species, however, are very sensitive to the pH and generally require a different pH range for the separation. At relatively high pH, basic compounds such as most of the peptides undergo strong interaction with the negatively charged silanol groups. This interaction can cause serious peak tailing and thus poor column efficiency. Thus, peptide separations are usually performed at a low pH in order to obtain high efficiency. The low pH, however, suppresses the ionization of the silanol groups and greatly reduces the EOF for the ODS stationary phase. As a result, the elution time becomes much longer. Since the electroosmotic flow is closely associated with the charge density on the surface of the stationary phase, special stationary phases with different surface properties need to be investigated to provide the appropriate conditions for CEC in low-pH mobile phases. There have been several reports describing the use of different stationary phases for CEC. Smith and Evans9 reported the separation of basic compounds such as tricyclic antidepressants using a strong cation-exchange stationary phase. By using a strong cation exchanger instead of C18 silica, extremely high efficiency was obtained with very little evidence of peak tailing. Kitagawa et al.10 used anion-exchange resins to separate sulfite, sulfate, and thiosulfate ions. They also used cation-exchange resins to separate metal cations such as lithium, sodium, and potassium ions. Li and co-workers11 separated iodide and iodate ions by using anionexchange packing materials. These applications resolved the analytes by two separation modes, i.e., electrophoresis and ion exchange. There are other examples of CEC separations using (5) Wu, J.-T.; Huang, P. Q.; Li, M. X.; Lubman, D. M. Anal. Chem. 1997, 69, 2908-2913. (6) Yan, C.; Dadoo, R.; Zare, R. N. Anal. Chem. 1996, 68, 2726-2730. (7) Colon, L. A.; Guo, Y.; Fermier, A. Anal. Chem. 1997, 69, 461A-467A. (8) Huber, C. G.; Choudhary, G.; Horvath, C. Anal. Chem. 1997, 69, 44294436. (9) Smith, N. W.; Evans, M. B. Chromatographia 1995, 41, 197-203. (10) Kitagawa, S.; Tsuji, A.; Watanabe, H.; Nakashima, M.; Tsuda, T. J. Microcolumn Sep. 1997, 9, 347-356. (11) Li, D.; Knobel, H. H.; Remcho, V. T. J. Chromatogr., B 1997, 695, 169174. 10.1021/ac980813u CCC: $18.00
© 1999 American Chemical Society Published on Web 03/19/1999
ion-exchange materials.12,13 All of these examples involve the separation of compounds with low molecular weight. These ionexchange stationary phases are not appropriate for separation of peptides since there are no long carbon chains to interact with the hydrophobic part of peptides and to retain the peptides. A mixed-mode phase was employed by Dittman and Rozing14 to separate polyaromatic hydrocarbons over a wide range of pH (48). The stationary phase has both C18 alkyl chains and strong cation-exchange groups, i.e., propanesulfonic acid. The sulfonic acid carries a negative charge even at low pH values. Therefore such phases can maintain a stable EOF over a broad pH range. In this work, a mixed-mode (reversed-phase/anion-exchange) stationary phase was used to separate peptide mixtures in pressurized capillary electrochromatography. C18 chains were used to interact with the peptides hydrophobically and anion-exchange sites, i.e., secondary amine groups, determine the charge density on the surface of the packing particles and are responsible for generating EOF. There are two advantages of using the mixedmode stationary phase. First, it can generate strong EOF at low pH, so that it is appropriate for separations that require acidic conditions. Second, a constant EOF can be achieved with a pH between 2 and 5. Therefore, the selectivity can be tuned over this pH range to optimize the separations, which are pH sensitive. Also, the mixed-mode phase demonstrated sufficient resolving capability in analyzing protein digest products. In addition, in pCEC, the retention of the peptides is dependent on their partitioning between the stationary and mobile phases and their electrophoretic migration. The latter is caused by the application of the voltage. The amplitude and polarity of the voltage controls the extent and direction of the electrophoretic migration. It is shown in this work that a wide range of selectivity for peptides can be obtained by changing the magnitude and polarity of the voltage. EXPERIMENTAL SECTION Materials. Trifluoroacetic acid (TFA), acetic acid, ammonium hydroxide, acetonitrile, methanol, ammonium bicarbonate, standard peptides, and horse heart myoglobin were obtained from Sigma (St. Louis, MO). TPCK-treated trypsin was purchased from Promega (Madison, WI). These materials were used without further purification. The water was deionized prior to use by a Milli-Q water purification system (Millipore Corp., Bedford, MA). Standard peptides were dissolved in deionized water to a concentration of 2 × 10-5 M. For protein digestion, 50 µg of protein was incubated for 20-24 h at 37 °C with a protein-to-enzyme ratio of 25:1 (w/w) in 50 mM ammonium bicarbonate solution at pH 8.2. The digested materials were then vacuum-dried to remove the salt and reconstituted in water to a concentration of 5 × 10-6 M original protein. Column Preparation. Fused-silica capillaries were obtained from Polymicro Technology (Phoenix, AZ) with 150-µm i.d. and 360-µm o.d. The mixed-mode (C18 reversed-phase/anion-exchange) stationary phase was obtained from Alltech (Deerfield, IL). The material consists of a high-purity, 100-Å pore size, 5-µm spherical silica substrate bonded with a single ligand containing both reversed-phase (C18) and dialkylamine in a fixed 1:1 ratio. (12) Watanabe, H.; Kitagawa, S.; Nakashima, M.; Tsuda, T. Chromatographia 1994, 15, 220. (13) Wei, W.; Luo, G. A.; Yan, C. Am. Lab. 1998, 30, 20C-20E. (14) Dittman, M. M.; Rozing, G. P. J. Microcolumn Sep. 1997, 9, 399-408.
The C18 stationary phase (5-µm, 300-Å pore size) was purchased from Vydac (Hesperia, CA). The columns were packed using the method described elsewhere.15 pCEC/MS. A Varian Star 9012 solvent delivery pump (Varian Associates, Inc., Walnut Creek, CA) provided supplementary pressure and delivered solvent gradient at a flow rate of 200 µL/ min. A split valve was used to split the solvent flow before the injection valve. The inlet end of the column was connected to a Valco six-port injection valve. A positive or negative high voltage was applied to the column end fitting at the outlet end of the column. The high voltage was delivered from a 0-30-kV power supply (model CZE 1000 R, Spellman High Voltage Electronics Corp., Plainview, NY). The injector was set to ground potential to protect the pump from any damage from the high voltage. The outlet end of the column was connected to an electrospray ionization (ESI) source via a 40-µm-i.d. fused-silica capillary. Solvent A consists of a certain concentration of TFA or acetic acid-ammonium acetate in aqueous solution. Solvent B consists of the equal amount of the electrolytes in acetonitrile. For all the standard peptide separations, the column was equilibrated with 100% A. After injection, 25% B was used to elute the sample from the column. For horse heart myoglobin digest analysis, a gradient of 0-100% B in 40 min with 0.025% TFA was employed. An ion trap storage/reflectron time-of-flight mass spectrometer as described in previous work16 was used as the on-line detector for pCEC separations. The analyte ions produced by an electrospray source were introduced into a differentially pumped interface. The ions were transported via a set of Einzel lens into the quadrupole ion trap and were stored in the ion trap under a preset radio frequency voltage (1250 Vpp) for 500 ms. The ions were then ejected by a dc pulse on the end cap of the ion trap into the reflectron time-of-flight mass spectrometer for detection and mass analysis. The ions were detected by a 40-mm triple-microchannel plate detector (model C-2501, R. M. Jordan Co.). The mass signals were collected using a 250-MHz transient recorder (model 9846, Precision Instruments Inc., Knoxville, TN) embedded in a Pentium 66-MHz PC-compatible computer (model P5-66, Gateway 2000, North Sioux City, SD).17 All the measurements in this work were performed at a sampling rate of 1 s/acquisition. Mass calibration was conducted using several standard peptides, based on the equation (m/z)1/2 ) aT + b, where T is the flight time. RESULTS AND DISCUSSION Typically, ODS stationary phases are used in CEC separations. EOF results from the ionization of the residual silanol groups on the surface of the stationary phase, where the ionization of the silanol groups increases with pH. To date most CEC applications have focused on the analysis of various neutral compounds. Separation of neutral compounds is not affected by the pH so that a mobile phase with pH of 7-8 is generally used to ensure a strong and reproducible EOF. For the analysis of ionic compounds, however, pH plays a very important role in achieving fast and efficient separations. Some separations require low pH values to (15) Huang, P. Q.; Wu, J.-T.; Lubman, D. M. Anal. Chem. 1998, 70, 3003-3008. (16) Wu, J.-T.; Huang, P. Q.; Li, M. X.; Qian, M. G.; Lubman, D. M. Anal. Chem. 1997, 69, 320-326. (17) Qian, M. G.; Wu, J.-T.; Parus, S.; Lubman, D. M. Rapid Commun. Mass Spectrom. 1996, 10, 1209-1214.
Analytical Chemistry, Vol. 71, No. 9, May 1, 1999
1787
Figure 1. Values of EOF velocity with pHs. Conditions: voltage, 6 kV; pressure, 80 bar; mobile phase, 0.04% TFA (pH 2.4) or 4 mM acetic acid-ammonium acetate in 25% acetonitrile-75% water (pH 3.2, 4.3, 5.2 and 6.1, respectively); columns, 120 mm × 150 µm, (0) 5-µm mixed-mode (C18 anion-exchange) stationary phase, and (b) 5-µm C18 stationary phase.
Figure 3. pCEC analysis of horse heart myoglobin. Conditions: Mobile-phase conditions as described in the Experimental Section. voltage, 1700 V; pressure, 80 bar; columns, (a) 85 mm × 150 µm packed with C18 reversed-phase and (b) 85 mm × 150 µm packed with mixed-mode phase.
Figure 2. pCEC separation of three peptides. Conditions: column, 85 mm × 150 µm; 0.014% TFA in the mobile phase; mobile-phase conditions as described in the Experimental Section; pressure, 70 bar. Peaks: (1) angiotensin III; (2) angiotensin I; (3) leucine enkephalin.
obtain high efficiency, such as the separation of peptides and many other basic compounds. Low pHs are required to reduce the interaction between the analytes and the ionized silanols of the stationary phase. This interaction is usually responsible for the severe peak tailing. For the analysis of small acidic compounds by pure CEC, low pH is also preferred to suppress the ionization of the compounds.18 At high pH the compounds are fully ionized, (18) Euerby, M. R.; Gilligan, D.; Johnson, C. M.; Roulin, S. C. P.; Myers, P.; Bartle, K. D. J. Microcolumn Sep. 1997, 9, 373-387.
1788 Analytical Chemistry, Vol. 71, No. 9, May 1, 1999
resulting in a strong electrophoretic migration which can almost counterbalance the EOF so that a long time is required for the analytes to reach the detector. To generate strong EOF at low pH, a mixed-mode (reversedphase C18/anion-exchange) stationary phase has been used in this work. The C18 chain of the stationary phase functions to retain the peptides by chromatographic interaction and the amine groups produce a considerable EOF at low pHs. EOF results from an electrical double layer on the surface of the packing particles in contact with the buffer. The velocity of the EOF, ueo, is given by
ueo ) ξE/4πη
where and η are the dielectric constant and viscosity of the liquid, respectively, E is the electric field strength, and ξ is the potential across the diffuse double layer. Among these parameters, and η depend on the property of the mobile phase. ξ is the only factor that is affected by the nature of the stationary phase, where ξ is proportional to the charge density on the surface of the stationary phase. The pKa of the secondary amine group on the mixed-mode
Figure 4. Mass spectrum corresponding to the last peak in Figure 3a.
stationary phase is 10,20 while the pKa of the silanol group on the conventional ODS stationary phase is ∼4.5.4 Therefore the amine groups are fully protonated while the silanol groups are slightly ionized in acidic buffer conditions. As a result, higher charge density and a larger EOF would be produced for the mixed-mode phase. Figure 1 shows the electroosmotic flow velocity at different pHs for the mixed-mode and C18 reversed-phase packing. Thiourea was used as the dead time marker. The elution time of thiourea with the application of the pressure only was measured to be t01, and the elution time of thiourea with the application of both the pressure and voltage was measured to be t02. The electroosmotic flow velocity, ueo, was calculated by ueo ) L/t02 - L/t01, where L is the column length. In Figure 1, the EOF for the mixed-mode packing was constant over a range of pH 2-5 and dropped greatly above pH 5.2. The decrease of the EOF was due to the partial deprotonation of the amine groups and the increased ionization of the silanol groups. For the C18 packing, EOF increased with pH and was substantially lower than that for the mixed-mode packing at pH 4 on the ODS stationary phase due to the interaction between the peptides and the residual silanol groups.23 The peak shapes in Figure 5, however, were better than expected at pH 4.3 and 5.2. This is attributed to the amine groups on the stationary phase, which can repulse the positively charged peptides and prevent them from approaching the residual silanol groups. The amine groups of the stationary phase not only generate the EOF but also can weaken the deleterious interaction between the peptides and the silanol groups. Figures 6 and 7 show the effect of changing the voltage on the elution profile in the separation of a five-peptide mixture. The peptide elution speed varied to a different extent in response to the voltage, resulting in different elution patterns. All the peaks except peak 4 changed the elution speed consistently corresponding to the voltage. For peak 4 in Figure 7, the elution speed varied randomly without any particular correlation with the voltage. It is probably due to the fact that the application of the voltage influenced not only the electrophoretic migration of the peptides but also ion pair formation and the hydrophobic interaction between the stationary phase and the peptides, which is determined by the nature of the peptides. An important advantage of (23) Mant, C. T., Hodges, R. S., Eds. High-Performance Liquid Chromatography of Peptides and Proteins: Separation, Analysis, and Conformation; CRC Press: Boca Raton, 1991; pp 298-299.
1790 Analytical Chemistry, Vol. 71, No. 9, May 1, 1999
Figure 5. Effect of pH on the separation of four peptides. Conditions: column, 120 mm × 150 µm; 4 mM acetic acid-ammonium acetate in the mobile phase; mobile-phase conditions as described in Experimental Section; voltage, 4 kV; pressure, 80 bar. Peaks: (1) angiotensin I; (2) methionine enkephalin-Arg-Phe; (3) methionine enkephalin; (4) leucine enkephalin.
pCEC over pure CEC is that the polarity of the voltage can be reversed during different runs in order to change the selectivity of the sample components and to optimize the separation. This is shown in Figure 7 for the peptide mixture of Figure 6 except that negative voltage has been used and varied to tune the elution of peaks for peptide separation. In pure CEC, only one polarity of the voltage is possible to ensure that the combination of the EOF and the electrophoretic mobility can carry the analytes to the detector. CONCLUSIONS A mixed-mode (reversed-phase/anion-exchange) stationary phase was used in pCEC to resolve peptide mixtures. The C18 chains of the stationary phase provided the chromatographic interaction with the peptides, while the amine groups were capable of generating a stable EOF in acidic mobile phases. A tryptic digest resulting from horse heart myoglobin was successfully separated using the mixed-mode phase in pCEC. This phase provided similar
Figure 7. pCEC separation of five peptides. All the conditions except the voltage were the same as in Figure 6.
Figure 6. pCEC separation of five peptides. Conditions: column, 75 mm × 150 µm; 0.014% TFA in the mobile phase; mobile-phase conditions as described in the Experimental Section; pressure, 70 bar. Peaks: (1) bradykinin; (2) angiotensin I; (3) methionine enkephalin; (4) methionine enkephalin-Arg-Phe; (5) leucine enkephalin.
separation efficiency compared to a conventional C18 reversed phase. In addition, selectivity tuning was achieved within pH 2-5, which cannot be readily achieved by using a conventional ODS stationary phase. Varying the magnitude of the voltage led to significant variations in the retention and selectivity of the peptides. In addition, the polarity of the voltage could be changed, and the voltage with opposite polarity resulted in totally different elution profiles. Also, this type of stationary phase should be appropriate for separating other cationic species or basic compounds in CEC. For anionic species, if the pH is sufficiently low to suppress the
ionization of the species, the stationary phase is an appropriate choice; otherwise, the anion exchange between the amine groups and the analytes would act as an additional separation mechanism to complicate the separation. This mixed-mode stationary phase has the potential to extend the range of compounds that can be analyzed by CEC. ACKNOWLEDGMENT We acknowledge Jing-Tao Wu at DuPont Pharmaceuticals for his helpful discussions. We gratefully acknowledge support of this work by the National Institutes of Health under Grant 2-R01GM49500-4 and the National Science Foundation under Grant BIR9513878. Received for review July 22, 1998. Accepted January 29, 1999. AC980813U
Analytical Chemistry, Vol. 71, No. 9, May 1, 1999
1791
