created: 11th August 1998, last updated: 24th May 1999© 1998 ABRF

Abstract

A high concentration of substrate protein facilitates efficient in-gel protease digestion of samples prior to HPLC separation and peptide micro sequencing. To facilitate quantitative recovery and concentration of proteins separated on large SDS-PAGE gels, a method of concentrating proteins using funnel tube (FT) gels was developed. This method allows the use of very large sample sizes (up to 5 ml of sample buffer and gel slices per funnel tube gel) and yielded quantitative recovery when 5 µg of a protein standard was concentrated. Funnel tube gel protein concentration was developed during the purification of several components of the yeast Exocyst complex. The yeast Exocyst complex was immunoprecipitated and its components were separated by sequential rounds of SDS-PAGE followed by funnel tube gel protein concentration. The individual protein components were concentrated a total of 676-fold into 71 cubic mm gel slices. The resulting in-gel protein concentrations were at least 0.2 µg/cubic mm. The FT gel protein concentration methodology should be useful whenever a small amount of protein must be concentrated from a very large gel or sample volume prior to protease digestion and HPLC or other characterization techniques.

Introduction

A significant problem in proteolytic digestion of SDS-PAGE purified proteins is obtaining a high in-gel protein concentration. A high in-gel protein concentration greatly facilitates complete proteolytic digestion (1). The slab funnel gel method uses SDS-PAGE to concentrate protein in a gel slice avoiding the need to use other protein concentrating methods (2). However, this technique is limited in capacity and the gels have a tendency to leak the applied sample from the gel sandwich due to inefficient acrylamide polymerization at the gel-spacer interface. The funnel tube (FT) gel method for concentrating proteins, described below, does not leak and efficiently concentrates protein from very large sample volumes (up to 5 ml of gel slices and buffer) into a small gel volume containing high in-gel protein concentration.

This method was developed during the purification of a multiprotein complex, the Exocyst, which is required for exocytosis in Saccharomyces cerevisiae (3, 4). Prior to this study, the Exocyst was known to contain at least eight polypeptides including Sec6p, Sec8p and Sec15p (4). The Exocyst complex proteins were isolated by immunoprecipitation and separated by SDS-PAGE. The unknown Exocyst components were then concentrated by funnel tube gel protein concentration. The high recovery of protein allowed for a second round of separation and concentration which completely removed contaminating IgG from the proteins of interest. Sufficient quantities of the unknown Exocyst proteins were recovered for efficient in-gel proteolytic digests, HPLC, mass spectrometry and peptide micro sequencing.

Materials and Methods

A. Construction and Design of the Funnel Tubes and the Electrophoretic Apparatus

Each FT was constructed by a scientific glassblower from a 3/8 inch (10 mm outside diameter) heavy-walled glass tube (inside diameter is about 5.5 mm) to which a 25 mm thin-walled glass tube was affixed. The top 25 mm tube was cut to about 2 cm in height to form a large bowl-shaped funnel above the 3/8 inch tube. The bottom 3/8 inch tube was then cut to about 19 cm in length. In addition, a Maria (a bulge) was placed in the neck of the tube 4.5 cm from the top so that the FT would rest on this bulge when placed in the electrophoresis tank (Figure 1).

Figure 1. Schematic diagram of a funnel tube gel. Sample buffer or a mixture of gel slices and sample buffer are loaded into the top of the funnel tube gel. Proteins concentrate at the stacking gel/running gel interface and enter the running gel just behind the lagging edge of the sample buffer dye.

The electrophoresis tank used in this study was improvised from a 25 year old Bio-Rad tube gel apparatus (Hercules, CA, http://www.bio-rad.com/index.html) capable of handling 11 mm outside diameter tubes, but most university shops should be able to fashion a workable unit. Also, a Bio-Rad Model 175 tube cell could be used, but with smaller capacity tubes (8 mm outside diameter).

B. Funnel Tube Gel Electrophoresis

The constructed funnel tubes were used to concentrate protein from sample buffer/gel slice mixtures. First, 5 cm tall running and stacking gels were sequentially poured into the funnel tubes and allowed to polymerize (1.4 ml of unpolymerized gel mixture each). After the gels polymerized, the FT was glued into the electrophoresis tank using RTV Multi-Purpose Sealant Model 732 (Dow Corning, Midland MI). Up to 5 ml of protein contained in sample buffer or a sample buffer/gel slice mixture is then loaded into the top of the funnel tube gel. For a FT gel with the dimensions described above, the voltage during the electrophoresis run was limited to 150 V and the current was limited to 7 mamps per tube to avoid heating. The concentrated protein elutes from the gel slices and stacks at the interface of the stacking and running gels. The protein concentrates at the stacking gel/running gel interface and enters the running gel just behind the dye front. When the trailing edge of dye was about 4 or 5 mm into the top of the running gel, the electrophoresis was stopped. Following an electrophoresis run, each FT was merely pulled hard to remove from the tank, although sometimes the sealant had to be partially cut away. The running gel can then be easily extruded from the FT by placing the bottom 10 cm of the tube under running hot water until the glass is quite warm to the touch. The gel can then be pushed out of the bottom of the FT with a 5 mm diameter glass rod. The top 4 or 5 mm of the running gel that contains the concentrated protein was then excised.

C. Concentrating Protein Standards

Funnel tube gels were used to concentrate 5 and 15 µg samples of beta-galactosidase, and the efficiency of protein recovery was quantitated. The protein samples were first boiled in 6 ml of sample buffer (3% SDS, 10% glycerol, 2% 2-mercaptoethanol, 2% bromophenol blue, 0.125 M Tris-HCl, pH 6.8) and applied separately to single well 16 cm x 14 cm x 1.5 mm, 7% acrylamide SDS-PAGE slab gels. After "separation", each gel was stained with Coomassie Blue for 10 min (0.125% Coomassie Blue, 50% MeOH, 10% acetic acid) and destained (one wash with 50% MeOH + 10% acetic acid followed by several washes with 5% MeOH + 7% acetic acid). The beta-galactosidase protein band was excised. The gel slices were then incubated at room temperature with several changes of water until the pH was between 5 and 6. They were then rocked with an equal volume of 2X sample buffer for at least 2 hours and then frozen overnight. The sample buffer/gel slice mixture was thawed and placed on a rocking platform for another two hours prior to loading the FT gel. Following completion of the FT gel protein concentration, the resulting gel slice was fixed with three changes of solution containing 50% MeOH and 10% acetic acid. Protein recovery was quantitated by amino acid composition analysis performed by the W. M. Keck Laboratory, Yale University.

D. Concentrating Exocyst Complex Components

The Exocyst complex proteins were purified by immunoprecipitation and sequential rounds of slab gel/FT gel SDS-PAGE electrophoresis. The Exocyst complex proteins were isolated from a lysate containing 6.4 g total protein by immunoprecipitation with 3.6 ml of ascites (anti-c-myc, monoclonal antibody 9E10) as previously described (3, 4). The immunoprecipitate was boiled in 48 ml of sample buffer to liberate the immunoisolated proteins. One quarter of the total immunoextract (12 ml) was multi-loaded into a one well 16 cm x 14 cm x 1.5 mm, 7% acrylamide SDS-PAGE slab gel. Four gels were used for the initial electrophoresis. The slab gels were stained and destained as above. Eight Exocyst complex polypeptides specifically co-immunoprecipitate (Figure 2) as previously shown (3, 4).

Figure 2. Characteristic protein pattern of an Exocyst complex immunoprecipitate. Shown are the Coomassie Blue stained polypeptides that co-immunoprecipitate with c-myc-tagged Sec8p with anti-c-myc antibody. This is an image of one of the four Coomassie Blue stained gels from the first SDS-PAGE protein separation. The gel is heavily overloaded with protein, in particular IgG. Bands B (c-myc-Sec8p), C (Sec15p), and G (Sec6p) had been previously identified (4). Bands A, D, E, F, and H were previously shown to co-immunoprecipitate with c-myc-Sec8p, but their identities were unknown prior to their purification and peptide sequencing (3). Click on the image to see a higher resolution version. (Warning: the larger version of this figure is 95K and may take some time to download!)

 

The protein bands corresponding to positions A, D, E, F, and H were excised from the gels, neutralized with water and equilibrated with an equal volume of fresh 2X sample buffer. The proteins contained in each band were then concentrated with FT gels as above.

The first round of SDS-PAGE slab gel separation failed to cleanly separate the eight Exocyst complex polypeptides from each other due to the extreme overloading of the initial slab gel. The proteins concentrated from the first slab gel (Bands A, D, E, F, and H) actually contain a mixture of Exocyst components and IgG (Figure 3). A second round of SDS-PAGE slab gel separation was necessary to resolve the individual components. The gel plugs from the first FT concentration gel were chopped into 1 mm cubes and equilibrated with an equal volume of fresh 2X sample buffer for 2 hours. The gel cubes and sample buffer were then loaded into a second 16 cm x 14 cm x 1.5 mm, 7% acrylamide SDS-PAGE slab gel (10-well comb) and electrophoresed. The second separating gel was stained and destained as above. The second separating gel gives a clean separation of all the concentrated proteins. One lane each of the second separating gel for bands D and E is shown in Figure 3.

Figure 3. Second SDS-PAGE gel separation. The bands corresponding to positions D and E from the first slab gel (Figure 2) were concentrated with FT gels and reseparated on a second slab gel. Shown is the Coomassie Blue stained second separating gel. Note that for both bands D and E, the material concentrated from the first slab gel was contaminated with other Exocyst complex proteins and with IgG. The first separating gel was heavily overloaded which results in proteins from a given band smearing and contaminating adjacent protein bands. The second separating gel cleanly resolves the mixture of proteins.

Gel slices from the second separating gel corresponding to a given unknown were then processed exactly as above and concentrated in a final FT gel. The resulting 5.5 mm by 3 mm gel slices were fixed overnight with at least 3 changes of solution containing 50% MeOH and 10% acetic acid. The gel slices containing the polypeptides of bands A, D, E, F, and H were then given to the W. M. Keck Laboratory for amino acid analysis, trypsin and/or lys-C in-gel digestion, HPLC separation, mass spectrometry and micro sequencing of the proteolytic peptides.

Results

We used FT gels to concentrate known amounts of beta-galactosidase from SDS-PAGE slab gel slices to test the efficiency of the funnel tube gel protein concentration method. Gel slices from the SDS-PAGE slab gels containing either 5 or 15 µg of beta-galactosidase were concentrated with funnel tube gels into 71 cubic mm gel plugs. We recovered an average of 100% (n=3) of the 5 µg protein standard and 65% (n=2) of the 15 µg of protein standard. Recovery was less efficient with larger (15 µg) than with smaller (5 µg) protein samples possibly due to overloading of the tube gel resulting in poor stacking. For the 5 µg and 15 µg protein standards, this yielded protein concentrations of 0.07 µg/cubic mm and 0.14 µg/cubic mm of gel volume, respectively. For both, the protein was concentrated about 70-fold (from 5,000 cubic mm to 71 cubic mm).

We next applied the technique of sequential SDS-PAGE separating gel/FT concentrating gel electrophoresis to the purification of the uncharacterized components of the yeast Exocyst complex (Figures 2 and 3). We estimate that 640 µg of protein corresponding to bands D and E are present in the 6.4 g of starting lysate used for their purification (each Exocyst protein is about 0.01% of total cellular protein, ref. 3). In the large scale immunoprecipitation used for these purifications, approximately 5% of the total c-myc-Sec8p was immunoprecipitated (data not shown and ref. 3). Since the Exocyst complex proteins are present in a 1:1 weight ratio to Sec8p (3), the sample buffer loaded onto the first separating gels contained about 32 µg of the band D and E proteins. We recovered 20 µg of band D protein (62%) and 9.7 µg of band E protein (30%) after two SDS-PAGE separating gel and FT concentrating gel electrophoresis cycles. The yield of purified protein from the starting material (640 µg) was 3.1% for band D and 1.5% for band E with most of the protein being lost by the inefficient immunoprecipitation. The concentration of protein in the final gel slices was 0.42 and 0.20 µg/cubic mm, respectively. The results for the other Exocyst components were comparable to that obtained for bands D and E (data not shown).

We were able to obtain peptide sequence from bands D, E, F, and H and determine the identity of band A by mass pattern matching (3). Comparison of the peptide sequences to a translation of the Saccharomyces cerevisiae genome database, and our own cloning and sequencing results, identified D as Sec5p, E as Sec10p, F as a C-terminal breakdown product of Sec3p and H as a novel protein of 70 kDa we termed Exo70p (3). Band A was identified as full-length Sec3p by mass pattern matching (3).

Discussion

We have described a method of concentrating protein using FT gels and applied this method to the purification of the unknown protein components of a multiprotein complex required for exocytosis in yeast. This method has several advantages: 1) very large starting sample volume, 2) FT gels do not leak, 3) high resulting in-gel protein concentrations, 4) quantitative protein recovery from dilute samples. The quantitative recovery of the 5 µg beta-galactosidase protein standard suggests that this method could be applied with equal success to samples containing much lower protein amounts.

The high efficiency of protein recovery allowed the use of multiple cycles of SDS-PAGE protein separation and FT gel protein concentration. This combination was a powerful aid in obtaining highly concentrated, pure Exocyst component proteins free of contaminating IgG. We were able to concentrate the purified Exocyst proteins a total of 676-fold with the FT gel method. The minimum in-gel protein concentration obtained (0.20 µg/cubic mm) was well above the desired minimum for efficient proteolytic digestion (0.03 µg/cubic mm) (1). The concentrated protein was suitable for in-gel protease digestion, HPLC separation of the resulting peptides, and peptide micro sequencing.

Acknowledgments

We would like to thank Ralph Stevens and Alan Brown for making the funnel tubes. We would also like to thank all the members of the W. M Keck Sequencing Facility at Yale for their help and patience, in particular, Kathryn Stone, Myron Crawford, Edward Papacoda, Mary Lo Presti and Kenneth Williams. Finally, we would like to thank James Jamieson for the use and gift of the Bio-Rad tube gel tank. This research was supported by NIH grant GM35370, by USUHS grant RO71EK and by a grant from the Edward Mallinckrodt, Jr. Foundation. The opinions and assertions contained herein are private ones of the authors and are not to be construed as official or reflecting the views of the Department of Defense or the Uniformed Services University of the Health Sciences.

References

1. Williams, K. R. and K. L. Stone. In gel digestion of SDS PAGE-separated proteins: Observations from internal sequencing of 25 proteins. In John W. Crabb (Ed.) Techniques in Protein Chemistry VI, Academic Press, Inc., San Diego, p. 143-152 (1995).

2. Lombard-Platet, G. and P. Jalinot. Funnel-well SDS-PAGE: A rapid technique for obtaining sufficient quantities of low-abundance proteins for internal sequence analysis. Biotechniques 15: 669-672 (1993).

3. TerBush, D. R., T. M. Maurice, D. Roth, and P. Novick. The Exocyst is a multiprotein complex required for exocytosis in Saccharomyces cerevisiae. EMBO J 15: 6483-6494 (1996).

4. TerBush, D. R., and P. Novick. Sec6, Sec8, and Sec15 are components of a multisubunit complex which localizes to small bud tips in Saccharomyces cerevisiae. J. Cell Biol. 130: 299-312 (1995).