A Proteomics Approach for the Identification of DNA Binding Activities Observed in the Electrophoretic Mobility Shift Assay

Abstract

Transcription factors lie at the center of gene regulation, and their identification is crucial to the understanding of transcription and gene expression. Traditionally, the isolation and identification of transcription factors has been a long and laborious task. We present here a novel method for the identification of DNA-binding proteins seen in electrophoretic mobility shift assay (EMSA) using the power of two-dimensional electrophoresis coupled with mass spectrometry. By coupling SDS-PAGE and isoelectric focusing to EMSA, the molecular mass and pI of a protein complex seen in EMSA were estimated. Candidate proteins were then identified on a two-dimensional array at the predetermined pI and molecular mass coordinates and identified by mass spectrometry. We show here the successful isolation of a functionally relevant transcription factor and validate the identity through EMSA supershift analysis. Transcription factors lie at the center of gene regulation, and their identification is crucial to the understanding of transcription and gene expression. Traditionally, the isolation and identification of transcription factors has been a long and laborious task. We present here a novel method for the identification of DNA-binding proteins seen in electrophoretic mobility shift assay (EMSA) using the power of two-dimensional electrophoresis coupled with mass spectrometry. By coupling SDS-PAGE and isoelectric focusing to EMSA, the molecular mass and pI of a protein complex seen in EMSA were estimated. Candidate proteins were then identified on a two-dimensional array at the predetermined pI and molecular mass coordinates and identified by mass spectrometry. We show here the successful isolation of a functionally relevant transcription factor and validate the identity through EMSA supershift analysis. The isolation and identification of sequence-specific DNA-binding proteins is not trivial. The first step involves the mapping of promoter regions that interact with potential transcription factors. DNase I footprinting and electrophoretic mobility shift assays (EMSA) 1The abbreviations used are: EMSA, electrophoretic mobility shift assay; 2DE, two-dimensional electrophoresis; IEF, isoelectric focusing; IPG, immobilized pH gradient; MALDI-TOF, matrix-assisted laser desorption ionization time-of-flight; MS, mass spectrometry; MM, molecular mass; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid. are generally employed for the rapid characterization of such regions including the definition of cis-acting elements (1.Dent C.L. Latchman D.S. Latchman D.S. Transcription Factors: A Practical Approach. Oxford University Press, Oxford1993: 1-26Google Scholar). If the defined DNA elements conform to consensus transcription factor binding sites, and if specific antibodies are available, then EMSA supershifts can be employed to possibly identify the interacting proteins. However, if appropriate supershift antibodies are not available, then EMSA is of limited value in the identification of novel DNA-binding proteins. The traditional route to the identification of cognate trans-acting factors, the biochemical isolation and identification of DNA-binding proteins, is usually a long and labor-intensive process. Purification of transcription factors often involves four or five different chromatographic steps, including ion exchange, gel filtration, and nonspecific and sequence-specific DNA affinity columns (2.Gadgil H. Jurando L.A. Jarrett H.W. DNA affinity chromatography of transcription factors.Anal. Biochem. 2001; 290: 147-178Google Scholar). The major impediment to the rapid identification of a transcription factor of interest is the fact that they are generally present in low concentrations, usually less than 0.1% of the total nuclear protein. Additionally they often bind with moderate affinity (3.Ren L. Chen C. Sternberg A.S. Tethered bandshift assay and affinity purification of a new DNA-binding protein.Biotechniques. 1994; 16: 852-855Google Scholar). Recent advents in proteomics and mass spectrometry have created unprecedented power in protein identification. For example, proteins have recently been analyzed directly by matrix-assisted laser desorption ionization time-of-flight mass spectrometry utilizing DNA probes harboring specific sequence motifs (4.Nordhoff E. Krogsdam A.-M. Jorgensen H.F. Kallipolitis B.H. Clark B.F.C. Roepstorff P. Kristiansen K. Rapid identification of DNA-binding proteins by mass spectrometry.Nat. Biotechnol. 1999; 17: 884-888Google Scholar). In this paper, we have developed a powerful method for the identification of DNA-binding proteins seen in EMSA. Utilizing the power of two-dimensional electrophoresis (2DE) and mass spectrometry (MS), we have established a novel technique to isolate transcription factors. More importantly, our method obviates the need for laborious and extensive purification of the protein of interest. In this paper, the methodology required and the successful isolation of a functionally relevant transcription factor have been described using our novel proteomics approach. We were interested in the identity of an EMSA complex that bound to a CCAT repeat sequence. This repeat forms part of a functionally important microsatellite repressor sequence within the CD30 promoter (5.Croager E. Gout A.M. Abraham L.J. Involvement of Sp1 and microsatellite repressor sequences in the transcriptional control of the human CD30 gene.Am. J. Pathol. 2000; 156: 1723-1731Google Scholar). Traditional methods such as sequence-specific DNA affinity chromatography, coupled with chromatographic purification of nuclear proteins, proved unsuccessful because of the high abundance and affinity of nonspecific nuclear proteins. Instead, by estimating the pI and molecular mass (MM) of the protein by coupling SDS-PAGE or isoelectric focusing (IEF) with EMSA, it was possible to identify candidate protein spots on a two-dimensional array of nuclear proteins. These candidates were characterized further by excision from a two-dimensional gel at the predetermined pI and MM. Proteins were then eluted, renatured, and tested for original activity in EMSA, and candidate spots were subsequently analyzed by mass spectrometry, and their identity was determined. Finally, we confirmed the identity of the protein isolated via our novel method using EMSA supershift analysis. An overview of the method is shown in Fig. 1. Jurkat (T cell) cell lines were obtained from ATCC and maintained in RPMI 1640 medium with 2 mm l-glutamine, 100 μg/ml each of streptomycin and penicillin, and 10% fetal bovine serum at 37 °C with 5% CO2. Cells were harvested at a density of 8 × 105 cells/ml, and nuclear extracts were prepared essentially as described previously (6.Li Y. Ross J. Scheppler J.A. Franza B.R. An in vitro transcriptional analysis of early responses of the human immunodeficiency virus type I long terminal repeat to different transcriptional activators.Mol. Cell. Biol. 1991; 11: 1883-1893Google Scholar). Nuclear extracts were then subjected to an ammonium sulfate cut (0.33 g/ml extract) and pelleted by centrifugation as described previously (7.Marshak D.R. Kadonaga J.T. Burgess R.R. Knuth M.W. Brennan Jr., W.A. Lin S. Strategies for Protein Purification and Characterization: A Laboratory Course Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1996Google Scholar). The resulting extract was then dissolved in TM buffer (50 mm Tris-Cl, pH 7.9, 100 mm KCl, 12.5 mm MgCl2, 1 mm EDTA, 10% glycerol (v/v)) and loaded onto a HiPrep® Sephacryl S-300 high resolution column (Amersham Biosciences). Fractions were snap-frozen in liquid N2 and stored at −80 °C. Determination of protein concentration of the fractions was performed using the Bio-Rad protein assay kit (Bio-Rad) according to the manufacturer’s protocol. MM determination of unknown proteins by SDS-PAGE was performed as described previously (8.Ossipow V. Laemmli U. Schibler U. A simple method to renature DNA-binding proteins separated by SDS-polyacrylamide electrophoresis.Nucleic Acids Res. 1993; 21: 6040-6041Google Scholar). Briefly, concentrated crude Jurkat nuclear extract (60 μg) was denatured in standard SDS loading buffer for 5 min at 95 °C. Proteins were electrophoresed at 240 V on an 8% SDS-polyacrylamide gel (9.Laemmli U.K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4.Nature. 1970; 227: 680-685Google Scholar), and the lane containing nuclear extract was sliced uniformly into molecular mass intervals. Gel slices were crushed into 1.5 volumes of renaturation buffer (3% Triton X-100, 20 mm Hepes, 100 mm NaCl, 5 mg/ml bovine serum albumin, 3 mm ZnCl2, 3 mm MgCl2, 2 mm dithiothreitol, 0.1 mm phenylmethylsulfonyl fluoride, 0.1 mm benzamidine-HCl) and incubated overnight at 4 °C. The polyacrylamide was pelleted by centrifugation, and the supernatant was then assayed for DNA binding activity in EMSA. Additionally, molecular mass standards were used to determine the molecular mass intervals of the excised gel slices. S-300 fractionated nuclear extract (180 μg) was resuspended in rehydration solution (8 m urea, 4% CHAPS, 0.5% IPG buffer, 2 mm tributyl phosphine) using ULTRAFREE centrifugal filters (Millipore). Four successive concentration and reconstitution cycles in rehydration solution ensured both buffer exchange and removal of salts for IEF. The samples (250 μl) were then loaded onto IPG Drystrips (13 cm; pH 3–10 linear or 4–7 linear) (Amersham Biosciences) and loaded onto an IPGphor (Amersham Biosciences) electrofocusing unit before commencing the following protocol: 12 h rehydration; 100 V for 100 V-h, 250 V for 250 V-h, 500 V for 1000 V-h, 1000 V for 2000 V-h, 8000 V for 60,000 V-h. Following IEF, strips were rocked in equilibration solution containing no urea (50 mm Tris, 30% glycerol (v/v), 2% SDS (w/v), 2 mm tributyl phosphine) for 15 min. IPG strips were sliced uniformly into pI intervals, the gel slices were crushed into 2 volumes of renaturation buffer, and the fractions were incubated overnight at 4 °C. The polyacrylamide was pelleted by centrifugation, and the supernatant was assayed for DNA binding activity using EMSA. Nuclear proteins were assayed for binding activity using an oligonucleotide containing 12 copies of a 4-bp repeat, 5′-G(C/C/A/T)12-3′, labeled with 32P[dCTP] using Klenow fragment. Nuclear proteins (2 μg) or pI/MM fractions (12.5 μl) were incubated for 10 min on ice with 1 μg of poly(dI·dC) in binding buffer consisting of 4% Ficoll, 20 mm HEPES, pH 7.9, 1 mm EDTA, 1 mm dithiothreitol, 50 mm KCl. Proteins were then incubated with the 32P-labeled oligonucleotide for 30 min on ice prior to being loaded onto a 6% polyacrylamide gel containing 0.25× Tris-taurine/EDTA. Gels were electrophoresed at 150 V for 3 h, dried, and exposed to x-ray film at −80 °C. Where indicated 500 ng of anti-YY1 (Yin Yang 1) antibody (Santa Cruz Biotechnology, Inc.) was incubated with extract for 10 min on ice prior to probe addition. Following IEF, strips were rocked in equilibration solution (6 m urea, 50 mm Tris, 30% glycerol (v/v), 2% SDS (w/v), 2 mm tributyl phosphine) for 30 min. Strips were rinsed briefly in SDS running buffer (25 mm Tris, 192 mm glycine, 0.1% SDS (w/v)) prior to being sealed into the top of a 10% SDS-polyacrylamide gel with 0.5% agarose in SDS running buffer. Electrophoresis was performed at 20 mA until the dye front reached the anodic end of the gels. Gels were subsequently stained using the silver staining kit (Amersham Biosciences) as per the manufacturer’s instructions. Alternatively, colloidal Coomassie Blue G250 was used if MS was being performed. Spots of interest were excised from 2DE gels, placed in microtiter plates and subjected to MALDI-TOF MS (Australian Proteome Analysis Facility). Samples were subjected to a 16-h tryptic digest at 37 °C. Peptides were extracted from the gel using a 50% (v/v) acetonitrile, 1% (v/v) trifluoroacetic acid solution. A 1-μl aliquot was spotted onto a sample plate with 1 μl of matrix (α-cyano-4-hydroxycinnamic acid, 8 mg/ml in 40% acetonitrile (v/v), 1% trifluoroacetic acid), and MALDI-TOF analysis was performed on a Micromass Tofspec time-of-flight mass spectrometer. The peptide masses obtained from MALDI-TOF spectra were analyzed using NCBI databases and utilizing the MS-FIT database tool located at the ProteinProspector website (10.Clauser K.R. Baker P.R. Burlingame A.L. Role of accurate mass measurement (+/− 10 ppm) in protein identification strategies employing MS or MS/MS and database searching.Anal. Chem. 1999; 71: 2871-2882Google Scholar). Monoisotopic peaks were searched against human proteins, 1–100 kDa, with a maximum of one missed cleavage, unmodified cysteines, and with a mass tolerance of 100 ppm. Binding of crude Jurkat nuclear extract to an oligonucleotide containing 12 CCAT repeat sequences resulted in the formation of one major and a number of minor DNA protein complexes designated A, B, D, and E (Fig. 2, lane NE). We were interested in identifying the components of complex E that bound to this repeat sequence. Jurkat nuclear extracts were fractionated using Sephacryl S-300 gel filtration. Resulting fractions were analyzed for complex E activity by EMSA. Fractionation of the extract resulted in partial purification of complex E, as activity was detected mainly in fraction 39 (Fig. 2, lane F39). Concentrated crude nuclear extract was analyzed by SDS-PAGE, and fractions of discrete molecular mass intervals were excised and eluted from the gel. Proteins of each fraction were renatured and tested for complex E activity in EMSA. Complex E activity was detected in the 55–66-kDa fraction (Fig. 3, lane 55–66). This lane also contains two other complexes that are not seen in the crude nuclear extract binding profile. These complexes may represent multimers of complex E constituents and suggested that complex E is most likely a monomer or homomeric protein complex although it was possible that it consisted of heteromeric subunits that were within the same molecular mass range. Sephacryl 300 fractionation demonstrated that fraction 36 (Fig. 1, lane F36) contained significant activity of complexes D and E, and to preserve the complex E peak fractions this fraction was used to determine the pI of complex E. Peak complex E activity was detected in two intervals, pI 5.6–5.8 and 5.8–6.15 with lower activity detected in neighboring intervals (Fig. 4). This result suggested that the pI of complex E is ∼5.8. Complex D activity was also reconstituted, and peak activity was seen in the pI 5.1–5.35 interval with lower activity in neighboring intervals. Also another complex, which was not seen in the crude nuclear extract binding profile, was seen with peak activity in the pI 4.8–5.1 interval. This complex may represent a nonspecific DNA-binding protein that is normally out competed in the S300 fractions but is able to bind in the IEF-purified fraction. The results indicated that complex E is a 60-kDa protein with a denatured pI of ∼5.8. To confirm these characteristics, nuclear proteins of fraction 39, the peak E fraction, were analyzed by two-dimensional electrophoresis over an IPG of pH 4–7 and transferred to an SDS-polyacrylamide gel. The region of interest was excised (MM 52–65, pI 5.5–6.0) and was dissected into 20 quadrants each corresponding to discrete MM and pI intervals (Fig. 5A). The proteins from each gel slice were eluted with re-naturation buffer and assayed for complex E activity using EMSA. Peak complex E activity was detected in gel slices 10 and 11 corresponding to pI intervals of 5.65–5.75 and 5.75–5.85, respectively, and a MM of 57–59 kDa. Smaller complex E activity was detected in neighboring fractions 6, 7, 14, and 15, which represent the same pI intervals, but neighboring MM intervals 54.5–57 kDa and 59–62 kDa (Fig. 5A). These results confirmed our earlier estimates of MM (Fig. 3) and pI (Fig. 4) of complex E. Examination of the excised region on a silver-stained gel indicated four candidate protein spots that were located within the quadrants that contained complex E activity. Candidates were denoted E1, E2, E3, and E4 (Fig. 5A), and the identity of these proteins spots was elucidated using MALDI-TOF MS. Two-dimensional electrophoresis of fraction 39 was repeated, and the resulting 2DE gel was stained with colloidal Coomassie Blue because of its compatibility with mass spectrometry. Two candidate protein spots were excised (Fig. 5A, E2 and E4), one from each quadrant, digested with trypsin and analyzed by MALDI-TOF MS, and monoisotopic peaks were searched against the NCBI database. Protein spot E4 matched the transcriptional repressor protein YY1 (Table I). YY1 is a zinc finger transcription factor with a pI of 5.8 and an apparent molecular mass of 60–68 kDa (11.Austen M. Lüscher B. Lüscher-Firzlaff J.M. Characterization of the transcriptional repressor YY1.J. Biol. Chem. 1997; 272: 1709-1717Google Scholar, 12.Harihan N. Kelley D.E. Perry R.P. (δ, a transcription factor that binds to downstream elements in several polymerase II promoters, is a functionally versatile zinc finger protein.Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 9799-9803Google Scholar), properties similar to our estimated MM and pI of the protein within complex E. Protein spot E2 also matched some YY1 peptides, although because of keratin contamination it had a significantly lower MOWSE score (data not shown). Protein spot E3 was not analyzed by MALDI-TOF MS but most likely represents a differentially modified form of YY1. Indeed, YY1 does contain several phosphorylation sites that may represent different protein spots on a two-dimensional array (13.Becker K.G. Jedlicka P. Templeton N.S. Liotta L. Ozato K. Characterization of hUCRBP (YY1, NF-E1, delta): a transcription factor that binds the regulatory regions of many viral and cellular genes.Gene. 1994; 150: 259-266Google Scholar).Table IProtein spot E4 matches the transcriptional repressor YY1Identified proteinMOWSE score% Peptides matchedpI/MMNCBI transcription protein repressor protein in a new To the identity of complex E as YY1 and to validate our anti-YY1 antibodies were used in EMSA to determine they were able to a supershift of complex E. 39 was in the supershift as it forms complex E in EMSA (Fig. 6, The of anti-YY1 antibodies two complexes with a no antibody was (Fig. 6, lane that complex E does contain YY1. Transcription factors are of gene expression. The purification and identification of these factors is crucial to the understanding of their on In the the isolation and identification of DNA-binding proteins has been a long and laborious task. we present a method for the identification of DNA-binding proteins seen in EMSA that the power of 2DE, coupled with mass spectrometry. EMSA is a technique for the and characterization of DNA-binding proteins in If candidate DNA-binding proteins can be by binding then appropriate antibodies can be in EMSA, and activity of the transcription factor can be However, if no candidates can be EMSA is of limited in the identification of novel DNA-binding proteins. We were interested in the identity of a DNA-binding protein seen in EMSA that was able to bind to a CCAT repeat which was to be characterized previously using standard purification Recent in mass spectrometry likely identification of the transcription factor through liquid chromatography mass spectrometry to identify proteins isolated from both the SDS-PAGE gel and the gel. However, coupling 2DE gel electrophoresis with EMSA was to be as simple analysis can be is because of the fact that the protein complex of interest in EMSA can be directly from the same nuclear protein sample the need for extensive on a 2DE gel results in less complex results as the transcription factors are further into spots that can also be in EMSA. our method determination of properties and MM and of the protein complex, which in identification of the functionally relevant transcription We have a novel technique for the identification of DNA-binding proteins seen in as with method are several The determination of the pI and MM of the protein on the of the protein to a following the of SDS-PAGE and IEF. we a previously (8.Ossipow V. Laemmli U. Schibler U. A simple method to renature DNA-binding proteins separated by SDS-polyacrylamide electrophoresis.Nucleic Acids Res. 1993; 21: 6040-6041Google that denatured proteins into a renaturation buffer containing a Triton X-100, that is able to SDS in it from with The to following is to the identification although if of renaturation is activity can be detected in EMSA. the of SDS-PAGE and and subunits of a DNA binding the DNA-binding protein of interest bind DNA as a monomer or complexes contain subunits that similar MM and pI the subunits be separated pI and MM fractionation and not be for analysis by EMSA. Finally, the DNA-binding protein in 2DE and be by mass spectrometry. and proteins from resolution or are during Two-dimensional electrophoresis of proteins using immobilized pH Biochem. 2000; Scholar). the DNA-binding protein of interest be to with 2DE and contain to mass identification. This technique has of in identifying DNA-binding proteins seen in EMSA the need for extensive purification of the protein of interest. We have used this technique to identify other EMSA binding such as (data not shown). DNA-binding proteins, transcription factors, lie at the center of gene regulation, and the identification of unknown factors is crucial to the understanding of transcriptional Recent in mass spectrometry and proteomics have rapid and accurate for protein identification and the identification of many transcription factors the need for purification