A New Isoelectric Focusing Gel for Two-Dimensional Electrophoresis Constructed in Microporous Hollow Fiber Membranes Jon Klein,†,‡,§,| George Harding,† and Elias Klein*,† Department of Medicine and Department of Biochemistry and Molecular Biology, and the Core Proteomics Laboratory, University of Louisville School of Medicine, Louisville, Kentucky 40202, and Veterans Administration Medical Center, Louisville, Kentucky Received November 7, 2001
We describe the preparation of IEF tube gels inside a nonwetting microporous plastic tubing. The gel in the tube need not be extruded after the first dimension separation. Instead, the porous structure of the tubes is made wettable, and the proteins are electrophoresed “through-the-wall” into the second dimension PAGE gel. Commercial ampholytes and reagents are suitable for the procedure. A useful pI range of 4.5-9.5 can be obtained when pI 3-10 ampholyte mixtures are used. Because of the high surface area of the porous material, precautions must be exercised to reduce oxygen inhibition during polymerization and dehydration of the gel during storage and use. A sheath device is described that satisfies these requirements. The plastic tubes can be disposed of by incineration and pose no biohazard. Keywords: microporous tube • IEF • 2-D PAGE
Introduction The scientific potential of high-resolution separation and identification methods for proteins has been amply documented in the current literature.1-3 A number of techniques are being examined to accelerate throughput of analysis to characterize post-translational modifications and to identify global protein expression. A principal analytical method at this time is the two-dimensional electrophoretic separation of proteins first reported independently by Klose and by O’Farrell.4,5 It provides high separation factors based on both charge and mass. The stained protein gel patterns resulting from the twodimensional separations are amenable to analysis by optical scanning devices that translate each protein “spot” into a coordinate. Replications of the separation can be used to provide a measure of coordinate precision so that samples representing differing cell conditions can be compared with some measure of reliability to identify newly expressed or repressed proteins. When proteins of interest are so identified, they can be excised from the gel, using robots driven to the proper coordinates.1,6 The excised spot can be digested with trypsin to produce a peptide fragmentation pattern that is then identified by MALDI analysis or other analytical techniques.7-9 Despite the ability of optical recognition programs to average coordinates, the reliability of the procedure depends on the ability of the electrofocusing process to produce reproducible separations on the 2-D gel plate. The search algorithms used * To whom correspondence should be addressed. E-mail: jon.klein@ louisville.edu. † Department of Medicine, University of Louisville School of Medicine. ‡ Department of Biochemistry and Molecular Biology, University of Louisville School of Medicine. § Core Proteomics Laboratory, University of Louisville School of Medicine. | Veterans Administration Medical Center. 10.1021/pr0155073 CCC: $22.00
2002 American Chemical Society
in peptide mass fingerprinting rely not only on the fragmentation pattern from the MALDI analyses but also on knowledge of the pI and mass range data obtained from the 2-D gel. The more reliable the coordinates on the latter are, the better the chances for identification. When quantitation of differential protein expression is desired, the reproducibility of spot location becomes even more important. Several methodologic limitations have been recognized in the use of mobile ampholyte-based IEF. First, mobile ampholyte pH gradients degrade in a time-dependent manner during electrophoresis. However, Gombocz et al. have demonstrated that carrier ampholyte drift can be attenuated by altering and supplementing the composition of the ampholyte mixture.10 In this paper, we deal with one of the problems leading to loss of coordinate reproducibility. In the 2-D procedure, the first step is based on the separation of the protein components on the basis of their charge. Firstdimension isoelectric focusing is traditionally performed using either immobilized pH gradients (IPG) or mobile ampholyte gels poured in glass tubes. While mobile pH gradient gels have significant potential in protein analysis, they are used less often than IPG gels due to several technical limitations. One significant disadvantage of mobile ampholyte gels is the distortion and breakage of the gel when it is extruded from the glass tube. Isoelectric focusing in mobile ampholyte gels separates proteins in low solids poly(acrylamide) gels that have been cross-linked with a small amount of ethylene bis-acrylamide contained within glass capillaries of less than 2 mm diameter. The low concentration of monomer and cross-linker results in fragile gels with low tensile moduli. Generally, high-resolution gels are on the order of 20 cm long, which causes additional handling problems. At the end of the electrolysis period, the gel must be extruded mechanically for transfer to the top layer Journal of Proteome Research 2002, 1, 41-45
41
Published on Web 01/12/2002
research articles of the second dimension gel to be used for sizing on the basis of mass. The relative positions of all proteins having the same pI must be reproducible since it is one of the input variables of eventual identification routines. However, as we show experimentally later in this paper, the force needed to extrude the focused gel from its glass capillary housing is sufficient to cause compression in the axial direction. The extruded gel is then placed into a conditioning bath with reducing and (sometimes) alkylating agents before transfer to the top layer of the second dimension gel. Depending on the methodology used, tensile forces sufficient to elongate the gel are applied in such manual transfers. Ideally, one would like to eliminate both the compressive force needed for extrusion as well as the tensile forces needed to transfer the fragile gel to its final location. To avoid these manipulations, we describe here a method for preparing the IEF gel inside a microporous, nonwetting hollow fiber. The gel is formed as usual but need not be extruded for transfer to the second dimension gel. Rather, the wettability of the microporous structure is changed and the proteins are transported through the hollow fiber wall on to the second dimension gel by the same electric field gradient that is used to size the proteins in the SDS gel.
Experimental Section Materials. A standard stock solution containing 30% acrylamide and 0.4% ethylene bis-acrylamide, as well as ampholytes (pH 3-10), was obtained from Genomic Solutions, Inc. (Ann Arbor, MI). Dilutions of the stock solution were made volumetrically to produce the gels after polymerization with either (NH4)2S2O3 or Na2S2O3. Urea, dithiothreitol (DTT), and sodium dodecyl sulfate (SDS) were obtained from Sigma Chemical Co. (St. Louis, MO). Microporous polypropylene hollow fibers, Accurel, type S6/2 (Membrana Gmbh, Wuppertal, Germany) were used as the housing for the gels. The fibers had a maximum pore size (determined by bubble point) of 0.64 µm, an inside diameter (i.d.) of 1800 µ,m and a wall thickness of 450 µm. Procedures. Preparation of Gels. The hollow fibers were cleaned of residual extrusion chemicals by washing with acetonitrile and drying. Fiber lengths needed for transfer to the second dimension were selected and encased in a thin gasimpermeable plastic tubing so that only a very small annulus remained between the hollow fiber and its plastic sheath. Heated inert gas was then blown into the fiber lumens and allowed to escape through the fiber walls and to atmosphere at the end of the dense plastic sheaths. This purging was necessary to remove adsorbed oxygen from the large internal surface of the microporous structure. The sheath was left in place throughout polymerization and during isoelectric focusing. To prepare the gel within each fiber, a monomer solution with 8 M urea, 5% acrylamide, and 0.2% bis-acrylamide was sparged with argon. Persulfate catalyst was used to catalyze the polymerization. Aliquots of monomer solution were catalyzed in 2.0 mL batches and held at 10 °C prior to use to increase pot life. The solution was quickly introduced into the fiber lumen in an upward flow until the fiber overflowed. The open end of the fiber was closed with a stretched Parafilm, and the modules were stored in a vertical position until polymerization was complete. When ready for use, the end fitting through which the monomer solution had been introduced was cut from the fiber and the end covered with nonwoven nylon. The upper end was 42
Journal of Proteome Research • Vol. 1, No. 1, 2002
Klein et al.
Figure 1. pH distributions along the length of tube gel in microporous fiber after 5500 V h electrolysis.
cut smoothly to produce a flat interface on the gel; it was then inserted into a 1.5 cm segment of Pharmed tubing that served as a sample reservoir/adapter and wrapped with stretched Parafilm to seal the reservoir. The gel-containing fibers were fitted into a standard electrophoretic apparatus (Hoefer, Inc.). The small reservoir at the top of the fiber was filled with a protein protective solution consisting of ampholyte, Triton X-100, urea, and DTT; the electrolyte reservoirs were filled with 10 mN diethylamine and phosphoric acid and the gels prefocused for 30-60 min at 100 V potential. Protein samples, Biorad IEF standards, or a laboratory mixture were added by underlayering the protein protective solution and a program of stepped increases in the voltage potential run for 7050 V h at a maximum voltage of 2000 V and a maximum current of 110 µA. IEF was performed at 4 °C. The focused gels were removed immediately and incubated for 30 min with a solution containing 112 mM Tris/acetate, 5% SDS, 0.01% bromophenol blue, and 50 mM DTT for application to the second dimension. The equilibrated gels were applied immediately to SDS gels and electrophoresed or were frozen at -85° C for application at a later time. SDS, acting as a surfactant, wetted the porous walls of the fiber and allowed the solution to reach the gel inside the fiber. Extrusion Force Measurements. To determine the force necessary to extrude 5% mobile ampholyte gels from commercially available glass capillaries, we utilized an 0.9 mm i.d. commercial IEF capillary (Genomic Solutions, Ann Arbor, MI). The capillary was connected to a saline filled 5 mL syringe via a T junction. The third opening of the T junction was connected to a digital manometer (Renal Systems, Minneapolis, MN) calibrated in mmHg. A nonwetting gas-permeable membrane was placed between the T connector and the manometer. Measurement of pI Gradient. Gels that had undergone focusing without protein samples were snap frozen in liquid nitrogen, dissected into 1 cm slices, and equilibrated with 1.0 mL of CO2-free distilled water. The pH of each sample in the distilled water was measured using a small diameter pH probe.
Results Establishment of a pH Gradient in Hollow Fiber IEF Gels. To determine if hollow fiber gels maintained a pH gradient, we measured pH in segments of the gels using a small diameter pH probe. Shown in Figure 1 is the essentially linear relationship between pH and location of a gel slice measured from the end of a gel focused within a microporous fiber. Although the ampholyte mixture was quoted as having a range of pH units from 3 to 10, the measured values of the extremities of the gel yielded values of 8.25-8.75 on the cathodic end and
Isoelectric Focusing Gel for Two-Dimensional Electrophoresis
research articles
Figure 2. Tube gels extruded from microporous hollow fibers after 5500 V h electrolysis: (a) BioRad standards, unstained; (b) protein mixture stained with Coomassie Blue.
Figure 3. Second dimension gels from tube gels separated in microporous hollow fibers: (a) tube gel extruded before placement on 2 nd dimension gel; (b) proteins transported through the HF shell after IEF separation, but not extruded.
4.25-4.6 at the anodic end. Considering the difficulty of measuring the pH of a small sample that is absorbing atmospheric gases, the reproducibility is quite good. However, the data indicate very basic and very acidic samples may not become adequately separated with the available ampholyte mixture. IEF in Hollow Fiber Membranes. We then determined if hollow fiber IEF gels could consistently separate a group of proteins with a wide pI range. BioRad IEF standards underwent isoelectric focusing using the protocol described in the Experimental Section. After focusing was complete, the hollow fiber was removed and the naturally colored protein standards were visualized. Figure 2a shows an extruded gel that has been used to separate a BioRad IEF mixture inside a microporous hollow fiber. The unstained bands are well separated, tightly focused, and positioned correctly relative to their known pI. We then
performed IEF in gels cast in hollow fibers using a protein mixture produced in our laboratory composed of insulin, Rand β-lactoglobulin, ovalbumin, bovine serum albumin, and polyclonal human IgG. Figure 2b shows an extruded gel that has been used to separate this protein mixture. Several replicates of the same sample are shown and demonstrate the complexity of the protein mixture and the good resolution of bands. Direct Transfer of Proteins through the Hollow Fiber Wall into the Second Dimension Gel. The next series of experiments were performed to determine if proteins resolved by IEF in hollow fiber membranes could transfer by electrophoresis into second dimension sizing gels. IEF was performed using a simple protein mixture composed of insulin, R- and β-lactoglobulin, ovalbumin, bovine serum albumin, and polyclonal human IgG. Shown in Figure 3 is a comparison of second Journal of Proteome Research • Vol. 1, No. 1, 2002 43
research articles
Klein et al. needed to begin the extrusion procedure was 1224 ( 140 mmHg, or nearly 2 atm of applied pressure. On the basis of the standard error of measured pressures, it is likely that the pressure applied to different gels is variable and results in variability of the protein pattern in the second dimension.
Discussion Mobile ampholyte gels cast in glass tubes were the original method of isoelectric focusing used in two-dimensional gel electrophoresis.4,5 However, the use of mobile ampholyte gels cast in glass tubes has declined as a result of several disadvantages including cathodic drift, poor reproducibility of transfer to the second dimension, loss of work product to broken gels and limited shelf life. Many of these limitations do not apply to isoelectric focusing using immobilized pH gradient (IPG) technology. However, IPG techniques do have some limitations of protein uptake and entry into the second dimension caused by poor hydration and electroendosmotic effect, respectively.11 By constructing mobile ampholyte gels cast in microporous hollow fiber membranes, we attempted to produce an isoelectric focusing gel that retains the theoretical advantages of mobile ampholyte gels in protein transfer and that has the extended shelf life, durability, and convenience of IPG isoelectric focusing. We first demonstrated that IEF gels could be polymerized within microporous hollow fiber membranes. The physical property that permits the retention of the first dimension gel within the porous hollow fiber during the polymerization stage is based on the liquid pressure needed to penetrate a microporous structure as expressed by the LaPlace equation:
Figure 4. Rat fibroblast proteins separated by IEF focusing in the porous fiber and transferred across the fiber wall for the SDS separation.
dimension gels resulting from IEF using gels extruded from glass tubes and those cast in hollow fiber membranes with passage of the proteins through the fiber wall. A similar general spot pattern was seen with both methods, but in general, spot margins were sharper with the hollow fiber technique. We then performed IEF of a more complex protein mixture in a hollow fiber gel. Shown in Figure 4 is a two-dimensional gel performed using hollow fiber IEF gels to resolve mouse fibroblast proteins. Protein spot margins were well defined, and patterns were consistent with those using standard techniques. Shelf Life of Hollow Fiber IEF Gels. We postulated that the nonwetted microporous membrane could prolong the shelf life of mobile ampholyte gels by preventing evaporation. To determine the role of hollow fiber membranes in prolonging utility, we cast IEF gels in hollow fibers and used them to perform IEF after 10 months of storage. Shown in Figure 5 are four 10-month-old mobile ampholyte IEF gels performed using BioRad IEF naturally colored standards. The protein bands are sharp and comparable to that observed with freshly cast gels as seen in Figure 2a. These data demonstrate that hollow fiber IEF gels have an extended functional life. Extrusion Pressure. Because hollow fiber IEF gels do not require manual extrusion, we wished to determine what extrusion forces are avoided that might lead to distortion of the protein pattern entering the second dimension gel. To determine the force needed to remove mobile IEF gels from their glass tubes, we measured the pressure required to begin extrusion of five replicate gels. The average pressure that was 44
Journal of Proteome Research • Vol. 1, No. 1, 2002
∆Pcr )
2γ cos Θcr rp
In this equation, ∆Pcr is the pressure difference needed for a liquid to displace a gas in the pore when the effective interfacial tension between the two is given by γ, and the average pore radius of a nonwetted capillary is rp. The bubble point, i.e., the pressure at which the first capillary is emptied of gas, reflects the largest pore present. For the poly(propylene) fibers used in this study, this is reported to be 13.8 psi with isopropyl alcohol as the wetting liquid; for water, the bubble point pressure would be well in excess of 100 psi. Thus, there is no risk that the aqueous polymerization mixture would penetrate into the pores facing the lumen side. This was confirmed by the fact that an integral cylindrical gel of low acrylamide solids could be extruded from the hollow fibers. Since the fiber’s lumen surface is not as smooth as the conventional glass tubings used, there was some concern whether adherence of the gel at the gel/polymer interface would interfere with the electrofocusing. This apparently was not a problem. However, the porous structure introduced other unforeseen problems. Specifically, the microporous matrix of the fiber wall provides excellent gas transfer rates between the interior and exterior surfaces. The same fiber types are used as oxygen-exchange devices in clinical oxygenators. The internal surface of the pore structure approaches that of chromatographic media and thus provide a large area for gas adsorption. It was necessary prior to the polymerization step to exchange adsorbed gas for an inert gas in order to avoid inhibition of catalysis. After polymerization, the same microporous structure acted as a vapor transfer medium so that it was found beneficial to maintain the impermable gas sleeve, or sheath, on the fibers.
research articles
Isoelectric Focusing Gel for Two-Dimensional Electrophoresis
Figure 5. Separation of Bio Rad standards in plastic tubing after 10 months of gel storage before the IEF run.
This prevented loss of water, and shrinkage, from within the fiber lumen. Addition of humectants would reduce the vapor pressure resulting in less water loss. The sheath was removed immediately prior to conditioning the gels in SDS and DTT containing equilibration buffers for transfer to the second dimension gel. The SDS lowered the interfacial tension between the poly(propylene) and the aqueous treating solution sufficiently to allow complete wetting of the pore structure. The wetting of the pore structure was a prerequisite for the proteins to be able to migrate through the wall onto the second dimension gel. With polymerization problems solved, we then determined if hollow fiber IEF gels could maintain an adequate pH gradient, focus complex protein mixtures, have extended storage life, and most importantly, permit the transfer of proteins directly through the hollow fiber wall into the second dimension. Direct measurement of the pH gradient from extruded gels demonstrated a linearity of pH that remained constant after prolonged storage at 4 °C and that permitted excellent IEF of complex protein mixtures. The equilibration of the hollow fiber gel with a standard solution containing SDS resulted in the opening of the membrane pores and excellent transfer of proteins into the second dimension. These experiments demonstrate the feasibility of using microporous hollow fiber membranes to construct an improved isoelectric focusing gel. The improved gels have the advantages of IPG techniques in terms of storage life and durability, without being subject to protein retention in the first dimension resulting from severe electroendosmotic effects. We speculate that this improved mobile ampholyte gel may have only minimal cathodic drift as a result of not being cast in a glass tube. Some cathodic drift in tube gels results from the effect of the acidic glass surface on acrylamide and ampholytes. Experiments are currently in progress to determine if microporous hollow fiber gels have less cathodic drift. In summary, we have produced an IEF gel for the first dimension that is cast in a hollow fiber made of a microporous membrane. These gels have linear pH gradients and are capable of IEF of complex protein mixtures. Proteins focused in hollow fiber gels can then pass through the membrane wall when subjected to an electrophoretic field, thus eliminating the
physical forces associated with extrusion of the gel. Finally, IEF gels cast in hollow fibers can be stored up to 10 months, thereby permitting mass manufacturing and lowering costs.
Acknowledgment. This work was supported, in part, by the Jewish Hospital Foundation, Louisville. KY (E.K.). References (1) Humphery-Smith, I. Proteomics: from small genes to highthroughput robotics. J. Protein Chem. 1998, 17, 524-525. (2) Proteome Research: New Frontiers in Functional Genomics (Principles and Practice); Springer-Verlag: Berlin, Heidelberg, 1997. (3) Wilkins, M. R.; Sanchez, J. C.; Gooley, A. A.; Appel, R. D.; Humphery-Smith, I.; Hochstrasser, D. F.; Williams, K. L. Progress with proteome projects: why all proteins expressed by a genome should be identified and how to do it. Biotechnol. Genet. Eng. Rev. 1969, 13, 19-50. (4) Klose, J. Protein mapping by combined isoelectric focusing and electrophoresis of mouse tissues. A novel approach to testing for induced point mutations in mammals. Humangenetik. 1975, 26, 231-243. (5) O’Farrell, P. H. High-resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 1975, 250, 4007-4021. (6) Lopez, M. F. Better approaches to finding the needle in a haystack: optimizing proteome analysis through automation. Electrophoresis 2000, 21, 1082-1093. (7) Shevchenko, A.; Jensen, O. N.; Podtelejnikov, A. V.; Sagliocco, F.; Wilm, M.; Vorm, O.; Mortensen, P.; Boucherie, H.; Mann, M. Linking genome and proteome by mass spectrometry: large-scale identification of yeast proteins from two-dimensional gels. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 14440-14445. (8) Gooley, A. A.; Ou, K.; Russell, J.; Wilkins, M. R.; Sanchez, J. C.; Hochstrasser, D. F.; Williams, K. L. A role for Edman degradation in proteome studies. Electrophoresis 1997, 18, 1068-1072. (9) Yan, J. X.; Wilkins, M. R.; K, Ou; Gooley, A. A.; Williams, K. L.; Sanchez, J. C.; Golaz, O.; Pasquali, C.; Hochstrasser, D. F. Largescale amino acid analysis for proteome studies. J. Chromatogr. A 1996, 736, 291-302. (10) Gombocz, E.; Cortez, E. Carrier ampholytes rehabilitated: gel isoelectric focusing on pH gradients visualized in real-time by automated fluorescence scanning in the HPGE-1000 apparatus. Electrophoresis 1999, 20, 1365-1372. (11) Gorg, A.; Obermaier, C.; Boguth, G.; Harder, A.; Scheibe, B.; Wildgruber, R.; Weiss, W. The current state of two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 2000, 21, 1037-1053.
PR0155073
Journal of Proteome Research • Vol. 1, No. 1, 2002 45
