A Novel Approach of Protein Immobilization for Protein Chips Using an

Dec 28, 2005 - protein (EGFP) and an EGFP-stathmin fusion protein, both of which contained a Cys-tag. We also included an oligo-histidine tag to allow...
0 downloads 0 Views 585KB Size
Download PDF
A Novel Approach of Protein Immobilization for Protein Chips Using an Oligo-Cysteine Tag Teruhisa Ichihara,† Junko K. Akada,‡ Shuichi Kamei,† Saemi Ohshiro,‡ Daisuke Sato,‡ Masanori Fujimoto,‡ Yasuhiro Kuramitsu,‡ and Kazuyuki Nakamura*,‡ Technical Research Laboratory, Toyo Kohan Company, Ltd., Higashi-toyoi, 1296, Kudamatsu, Yamaguchi 733-8611, Japan, and Department of Biochemistry and Functional Proteomics, Yamaguchi University Graduate School of Medicine, 1-1-1, Minami-kogushi, Ube, Yamaguchi 755-8505, Japan Received December 28, 2005

Protein chip technology is essential for high-throughput functional proteomics. We developed a novel protein tag consisting of five tandem cysteine repeats (Cys-tag) at termini of proteins. The Cys-tag was designed to allow covalent attachment of proteins to the surface of a maleimide-modified, diamondlike, carbon-coated silicon substrate. As model proteins, we created an enhanced green fluorescent protein (EGFP) and an EGFP-stathmin fusion protein, both of which contained a Cys-tag. We also included an oligo-histidine tag to allow its purification by the use of Ni beads, and we expressed the protein in Escherichia coli. The purified Cys-tagged EGFP could be captured on the maleimide-coated substrate efficiently so that 50 pg of the fusion protein was detected by fluorescence, and as little as 5 pg was immunodetected by combination with enhanced chemiluminescence. This highly sensitive immunodetection may be due to the strong covalent binding of the Cys-tag to the substrate combined with efficient exposure of the protein to the surrounding solution. Thus, the Cys-tag should be useful for developing a novel protein printing method for protein chips that requires very low amounts of protein and can be used for high-performance analysis of protein-ligand interactions. Keywords: cysteine tag • thiol group • maleimide group • DLC • maleimide-coated substrate • orientation • protein chip

Introduction Since completion of the human genome project,1,2 biological and medical science has focused on the transcriptome, and more recently, on the proteome.3,4 Following the development of DNA microarray chip technology for analysis of the transcriptome, protein chip technology for proteomic analysis has also advanced and been applied for the simultaneous identification, quantification, and functional analysis of proteins.5 In early versions of protein chips, the chip surface was treated with an aldehyde-containing silane reagent for Schiff’s base linkages to amines, lysine, or NH2-termini in proteins.6 Nickel-coated slides have also been used to immobilize yeast proteome-wide recombinant proteins through 6× histidine fusion tags (His-tags).7 Recently, nitrocellulose-coated glass slides have been employed for investigating protein-protein interactions,8,9 finding specific antibodies from serum with antigen array,10,11 and finding protein antigens with antibody arrays.12-14 Bait proteins on chips must be firmly immobilized so that they remain attached after washing and incubation * To whom correspondence should be addressed at Department of Biochemistry and Functional Proteomics, Yamaguchi University Graduate School of Medicine, 1-1-1, Minami-kogushi, Ube Yamaguchi 755-8505, Japan. Tel/fax, +81 836 22 2212; e-mail, [email protected]. † Toyo Kohan Company, Ltd. ‡ Yamaguchi University Graduate School of Medicine.

2144

Journal of Proteome Research 2006, 5, 2144-2151

Published on Web 07/29/2006

steps. The immobilization must also allow interacting molecules to access and bind to the protein efficiently. In these regards, protein chips are not yet as advanced as DNA microarray chips, and further modifications are needed to develop them into high-quality tools for proteomic analysis. Here, we describe a novel method for immobilizing protein using a fusion tag of five tandem cysteines (Cys-tag; Figure 1). The thiol group of cysteine has been utilized for several protein analyses. For example, cysteine residues in any proteins could be labeled by maleimide group-linked fluorescent dyes,15 applied for antibody labeling or protein staining before or after electrophoresis in gel. Short tetracysteine motifs expressed inside of a recombinant protein could trap a fluorophore later for the labeling of only the protein in vivo16,17 and in vitro.17 For chip application, N-terminally cysteine-containing peptides were immobilized on thioester slides,18 which were applied to a study of kinase activity on peptide arrays.19 This covalent binding property of cysteine is useful for the chip application of larger recombinant proteins, too, via a fusion Cys-tag. As a model protein, we generated a Cys-tag fusion of enhanced green fluorescent protein (EGFP),20 which was used as a marker of gene expression and protein targeting in intact cells and organisms because of its strong and stable green fluorescence. We were able to immobilize and detect the purified Cys-tagged EGFP at levels as low as 5 pg (0.2 fmol). We also immobilized 10.1021/pr0504889 CCC: $33.50

 2006 American Chemical Society

research articles

Protein Immobilization Using an Oligo-Cysteine Tag

Figure 1. A novel approach to protein immobilization using a Cys-tag. (A) Scheme for reaction between a maleimide group and a thiol group of a cysteine in a protein. (B) The three steps for immobilization of a Cys-tagged protein on the maleimide-coated substrate. Step 1, manufacturing of a substrate for a protein chip with maleimide groups on its surface; step 2, preparation of a model Cys-tagged fusion protein (EGFP); step 3, printing of the Cys-tagged protein on the maleimide-coated substrate to produce the protein chip. Protein immobilization occurs by any cysteine residue in Cys-tag or inside of a target-protein theoretically. C, cysteine.

Cys-tagged and EGFP-fused stathmin on the chip as the second protein model, and confirmed it immunologically and directly by LC-MS/MS. This protein-tag and the related printing method described here allow the covalent binding of proteins to the surface of protein chips with high performance.

Experimental Procedures Manufacturing Process of a Maleimide-Coated Substrate. To manufacture a maleimide-coated substrate (Figure 2A), we deposited a diamond-like carbon (DLC) film on a silicon substrate using the ion deposition method.21 First, the substrate was placed in a reactor evacuated to under 8 × 10-3 Pa and then cleaned by plasma using hydrogen as the feed gas. The flow rate of hydrogen gas was set at 40 standard cc/min (sccm), and the radio frequency (RF) power of self-bias for the substrate was 100 W. After the treatment, methane and hydrogen gas were introduced into the reactor to deposit the DLC film on the substrate at ambient temperature. The flow rates of methane and hydrogen were 47.5 and 2.5 sccm, respectively. The working pressures in the reactor and the RF power for the substrate were maintained at 3 Pa and 200 W, respectively. Next, the surface of the DLC film was aminated by plasma using ammonium as feed gas at a flow rate of 18 sccm and a working pressure of 3 Pa. To introduce maleimide groups to the aminated surface of the DLC film, the substrate was immersed in phosphate-buffered saline (PBS) containing 0.1 M N-(6maleimidocaproyloxy) sulfosuccinimide, sodium salt (SulfoEMCS; Dojin, Kumamoto, Japan)22 at 25 °C for 1 h. The substrate was then rinsed with distilled water and ethanol. Finally, the substrate was dried under vacuum. The surface of the substrate was certificated by X-ray photoelectron spectros-

Figure 2. Process of manufacturing a maleimide-coated substrate and its certification. (A) Process of manufacturing a maleimidecoated substrate. (B, C) X-ray photoelectron spectroscopy (XPS) charts of a substrate before and after introduction of maleimide groups on its surface. (B) Wide scan chart. Range, 0-1100 eV; pass energy, 178.95 eV; step, 1 eV; time/step, 60 eV. (C) Narrow scan chart around C1s. Range, 280-300 eV; pass energy, 71.55 eV; step, 0.2 eV; time/step, 50 eV. Table 1. Sequence of Oligo DNA for the Test of Maleimide Chip Substrate no.

sequence

size (mer)

modification

1 2 3

ACTGGCCGTCGTTTTACAACGT ACTGGCCGTCGTTTTACAACGT ACGTTGTAAAACGACGGCCAGT

22 22 22

5′-amino 5′-thiol 5′-Cy5

copy (XPS; Physical Electronics, Chanhassen, MN). The prepared substrates were cut in size to 3 mm × 3 mm squares, vacuumed, sealed, and stored in the dark at room temperature until further uses. Evaluation of the Maleimide-Coated Substrate Using Oligo DNA. The maleimide-coated substrate was evaluated using modified oligo DNA (Table 1). Two picomoles of 5′-aminomodified oligo DNA and 5′-thiol-modified oligo DNA (nos. 1 and 2 in Table 1) were printed on the substrate in PBS, after which the substrate was incubated for 1 h at 80 °C. Next, the substrate was washed at room temperature for 15 min with 2× SSC (0.3 M NaCl/30 mM sodium citrate)/0.2% sodium dodecyl sulfate (SDS), for 5 min at 95 °C with same solution, and once for 10 min with distilled water at room temperature. Journal of Proteome Research • Vol. 5, No. 9, 2006 2145

research articles

Ichihara et al.

Table 2. PCR Primers Used to Construct Plasmids primer name

sequence

size (bp)

restriction enzyme

PstpET14F NBXBKpET Nco5xCys NdeEGFPF XhoEGFPR Kpn6xHis Kpn5xCys XhoStaF KpnStaR

TTGCCATTGCTGCAGGCATCGTG GCGGCAGCCATATGAGATCTCGAGGATCCGGTACCGCTGCTAACAAAGCCCGAAAG TATACCATGGGCAGCAGCTGTTGTTGTTGTTGTAGCAGCGGCCTGGTGCCGCG GCGGCAGCCATATGGTCGCCACCATGGTGAGCAAG TCGAAGCTTGAGCTCGAGATCTGA ATCCGGTACCAGCAGCCATCATCATCATCATCATTAAGAAGCTGAGTTGGCTGCTGCC ATCCGGTACCAGCAGCTGTTGTTGTTGTTGTTAAGAAGCTGAGTTGGCTGCTGCC TCTTCCTCGAGCCATGGCTTCTTCTGATATCCAGGT GAACGGTACCGTCAGCTTCAGTCTCGTCAGCA

23 56 53 35 23 58 55 36 32

PstI NdeI, BglII, XhoI, BamHI, KpnI NcoI NdeI XhoI KpnI KpnI XhoI KpnI

Cy5-labeled complementary oligo DNA (1 nmol/mL; no. 3 in Table 1) was hybridized to the immobilized oligo DNA in 5× SSC/0.5% SDS for 1 h at 50 °C. The substrate was then washed with 2× SSC/0.2% SDS, followed by 2× SSC. Fluorescence on the substrate was scanned using an Image Analyzer FLA-8000 (Fujifilm, Tokyo, Japan) with a 532-nm filter. Construction of His- and Cys-Tagged EGFP and Stathmin Plasmids. The pET-14b vector (Merck Novagen, Madison, WI), which was an N-terminal His-tag expression plasmid, was modified to construct the Cys-tag expression vectors. First, new cloning sites (NdeI, BglII, XhoI, BamHI, and KpnI) were introduced into pET-14b using a DNA fragment generated by PCR amplification of pET-14b with primers NBXBKpET and PstpET14 (Table 2), followed by digestion with NdeI and PstI, thus, generating the new 6×His plasmid. The 5× Cys-encoding sequence was amplified from the 6×His plasmid by PCR using primers Nco5×Cys and PstpET14F and digested with NcoI and PstI. The 5× Cys-encoding fragment was then inserted in place of the corresponding 6× His-encoding sequence in the 6×His plasmid, generating the 5×Cys plasmid. The EGFP-encoding sequence was next PCR-amplified from pEGFP-C1 (BD Bioscience Clonetech, Palo Alto, CA) using primers NdeEGFPF and XhoEGFPR and then cloned into the 6×His and 5×Cys plasmids using the NdeI and XhoI sites, consisting of the amino-terminal (N-terminal) tag-fused 6×His-EGFP and 5×Cys-EGFP plasmids, respectively. To introduce carboxyl-terminal (C-terminal) tags, the 6×His- or 5×Cys-encoding sequence was amplified by PCR using primers Kpn6×His and PstpET14F or Kpn5×Cys and PstpET14F, respectively, digested with KpnI and PstI, and then inserted downstream of the EGFP gene in the 5×Cys-EGFP or 6×His-EGFP plasmids, respectively. Thus, 6×His-EGFP-5×Cys and 5×Cys-EGFP-6×His plasmids were constructed. As the second model protein, PCR-amplified human stathmin genes (a gift from Dr. A. Sobel) using primers XhoStaF and KpnStaR were inserted downstream of EGFP using XhoI and KpnI sites. All plasmid cloning was carried out in Escherichia coli JM109 cells using standard methods. All plasmid-inserted, PCRamplified DNA fragments were verified by DNA sequencing. Expression and Purification of Proteins. E. coli BL21 (DE3) cells (Novagen) were transformed with plasmid vectors encoding 6×His, 6×His-EGFP, 5×Cys-EGFP, 5×Cys-EGFP-6×His (Figure 4) and 6×His-EGFP-5×Cys (Figures 4 and 8A), or 6×His-EGFP-stathmin-5×Cys (Figure 8A). The cells were cultured overnight at 37 °C on a Luria-Bertani (LB) agar plate containing 50 µg/mL ampicillin. Single colonies were suspended in 30 mL of LB medium containing 50 µg/mL ampicillin and 0.4 mM isopropyl-β-thiogalactopyranoside (IPTG; Sigma, St. Louis, MO) and cultured for 24 h at 23 or 28 °C with shaking at 160 rpm. The cultured cells were centrifuged, and the pellet was washed twice with PBS (pH 7.4) and then lysed by incubation for 20 min at room temperature in 5 mL/g cell pellet 2146

Journal of Proteome Research • Vol. 5, No. 9, 2006

of Protein Extraction Reagent (Novagen) supplemented with 25 U/mL Benzonase (Novagen) and 5 kU/mL rLysozyme (Novagen). After the centrifugation of 10 000g, 10 min at 4 °C, the supernatant was passed through a 0.22-µm filter (Millipore, Billerica, MA), affinity-purified using a Ni-chelate column (GE Healthcare Bio-Science, Piscataway, NJ), eluted with 250 mM imidazole, and desalted with a Microcon YM-10 (Millipore). Protein Immobilizing on a Maleimide-Coated Substrate. Single- or double-tagged EGFP or EGFP-stathmin in PBS/PEG (PBS with 50% polyethylene glycol 300 [Kishida Chemical, Osaka, Japan]) was printed on the substrate as 0.5-µL spots of 1 ng/mL to 100 µg/mL solution or 8 µL spots of 1 mg/mL solution all over a 5 × 5 mm substrate and then incubated in a humid chamber at room temperature for 1 h to overnight (Tipping methods). Alternatively, the substrate was immersed in solutions of 1-10 mg/mL EGFP in PBS at room temperature for 1 h to overnight (Dipping methods). The spotted or immersed substrates were washed sequentially with TBS/ Tween-20 (25 mM Tris, 137 mM NaCl, 2.68 mM KCl, and 0.05% Tween-20, pH 7.4), PBS, and distilled water. The protein chips were overlaid with a cover slip to prevent drying. Detection of EGFP and Stathmin on the Substrate. EGFP fluorescence on the substrate was detected with a FLA-8000 fluorescence scanner (Fujifilm) using a 415-nm filter. EGFP on the substrate was also immunologically detected as follows. The EGFP-bound substrate was treated with blocking reagent (Roche, Basel, Switzerland, or 5% albumin/TBS/0.5% Tween20) for 1 h at room temperature and then incubated with antiEGFP mouse monoclonal or anti-stathmin rabbit polyclonal antibody (Sigma). After washing twice with PBS, it was incubated with horseradish peroxidase-conjugated or Alexa fluora 555-conjugated anti- mouse IgG antibody (Sigma, and Molecular Probes, Eugene, OR) for 1 h, then washed twice with PBS or TBS with 0.5% Tween-20. Chemiluminescence or chemifluorescence detection was performed using SuperSignal West Femto Maximum Sensitivity Substrate (Pierce, Rockford, IL) or ECL-Plus (GE healthcare Bio-Science), respectively. On-Chip Digestion and Mass Spectrometry. The protein samples on chips were washed once in H2O and reduced twice in 50% ACN, 50 mM ammonium bicarbonate, and 5 M DTT for 10 min, then dehydrated in 100% ACN twice for 30 min. The samples were digested in 10 µg/mL trypsin (Promega, Madison, WI), 30% ACN, 50 mM ammonium bicarbonate, and 5 M DTT at 30 °C overnight, lyophilized overnight, then dissolved in 0.1% formic acid. The peptide samples were analyzed by an Agilent 1100 LC/MSD Trap XCT (Agilent Technologies, Palo Alto, CA). Protein identification was performed in the Spectrum Mill MS Proteomics Workbench against the Swiss-Prot protein database search.

Protein Immobilization Using an Oligo-Cysteine Tag

research articles

Figure 4. Structures of the four kinds of tagged EGFP fusion proteins examined in this study: (A) 6×His-EGFP, (B) 5×CysEGFP, (C) 5×Cys-EGFP-6×His, and (D) 6×His-EGFP-5×Cys. H, histidine; C, cysteine.

Figure 3. Scanning image of (A) 5′-thiol- or (B) 5′-amino-modified oligo DNA immobilized on the maleimide-coated substrate after hybridization of Cy5-labeled complementary oligo DNA. The fluorescence was scanned using a 532-nm filter, and the fluorescence intensity is shown with a 32-color scale.

Results Principle and Design of the New Protein Chip. Because cysteine has a thiol group, it specifically covalently reacts with maleimide groups23 (Figure 1A). We designed a novel protein immobilization method that took advantage of this reaction (Figure 1B). The design of the current studies were consisted with three steps, which are to (1) manufacture a substrate for a protein chip with maleimide groups on its surface, (2) prepare a model Cys-tagged fusion protein (EGFP) for expression in E. coli, and (3) print the Cys-tagged protein on the maleimidecoated substrate. Introduction of Maleimide Groups on a Substrate. The new protein chip substrate was manufactured by introduction of maleimide groups on the surface of a DLC-coated silicon substrate (Figure 2A). The surface of the new substrate was examined by XPS (Figure 2B,C). The wide-scan chart (Figure 2B) revealed an increase in the oxygen and nitrogen peaks following the chemical modification. In addition, the narrowscan chart around C1s (Figure 2C) showed a new weak peak around 290 eV for the modified substrate (arrow in Figure 2C). The new peak, along with the increased oxygen and nitrogen peaks, indicates the specific introduction of carbonyl carbons from maleimide groups. This analysis confirmed that maleimide groups were introduced on the DLC-coated silicon substrate. Evaluation of the Maleimide-Coated Substrate Using Oligo DNA. Prior to examining protein immobilization to the modified substrate, we evaluated its specificity for sulfhydryl groups using amino- and thiol-modified oligo DNA (Table 1, nos. 1 and 2). The substrate was incubated with oligo DNA, followed by Cy5-labeled complementary oligo DNA (Table 1, no. 3). We observed strong hybridization signals for the thiol- but not the amino-modified oligo DNA (Figure 3, spots A and B, respectively). This indicated that the modified substrate specifically immobilizes molecules containing thiol groups. Expression and Purification of Various Tagged EGFP Proteins. To test immobilization of Cys-tagged proteins on the modified substrate, we prepared the following EGFP fusion proteins containing various combinations of the Cys-tag and a conventional His-tag (Figure 4): 6×His-EGFP, which has an N-terminal His-tag (as a control); 5×Cys-EGFP, which has an

Figure 5. Expression, isolation, and purification of the tagged EGFP proteins. (A) E. coli BL 21 culture at 28 °C after overnight induction of recombinant proteins, which are 6×His only (no. 1), 6×His-EGFP (no. 2), 6×His-EGFP-5×Cys (no. 3), 5×Cys-EGFP (no. 4), and 5×Cys-EGFP-6×His (no. 5). (B) Proteins were separated by 12.5% SDS-PAGE and stained with Coomassie Brilliant Blue. Lanes 1-5, supernatants after centrifugation of crude extracts; lanes 6-11, Ni-affinity-purified samples. The proteins are as follows: 6×His (lane 1), 6×His-EGFP (lanes 2, 6, and 7), 6×HisEGFP-5×Cys (lanes 3, 8, and 9), 5×Cys-EGFP (lane 4), and 5×CysEGFP-6×His (lanes 5, 10, and 11). Lanes 7, 9, and 11 are 10-fold dilutions of lanes 6, 8, and 10, respectively.

N-terminal Cys-tag; 5×Cys-EGFP-6×His, which has an Nterminal Cys-tag and a C-terminal His-tag; and 6×His-EGFP5×Cys, which has an N-terminal His-tag and a C-terminal Cystag. These four proteins along with a no-EGFP His-tag control (6×His vector) were expressed in E. coli strain BL21 and induced overnight at 28 °C. Fusion proteins 6×His-EGFP and 6×His-EGFP-5×Cys expressed well in the bacterial cultures as shown by green color (Figure 5A, nos. 2 and 3, respectively). Although 5×Cys-EGFP and 5×Cys-EGFP-6×His did not express well (nos. 4 and 5 in Figure 5A), all three His-tagged EGFP proteins could be effectively purified and concentrated by Nicolumn chromatography (lanes 6, 8, and 10 in Figure 5B). We used purified 5×Cys-EGFP-6×His, 6×His-EGFP-5×Cys, and 6×His-EGFP and unpurified 5×Cys-EGFP for further analyses. Journal of Proteome Research • Vol. 5, No. 9, 2006 2147

research articles

Ichihara et al.

Figure 6. Immobilization of EGFP on a maleimide-coated substrate. (A) Tipping method. Various EGFP fusion proteins were spotted on the substrate, and unbound protein was removed by washing. (B) Dipping method. Maleimide-coated substrates were immersed in EGFP solutions for 1 h and then washed. In both panels A and B, the following proteins were analyzed: 1, purified 6×His-EGFP; 2, purified 6×His-EGFP-5×Cys; 3, crude 5×CysEGFP; 4, purified 5×Cys-EGFP-6×His; 5, no protein (negative control). The fluorescence was scanned for 30 s using a 415-nm filter. The fluorescence intensity is shown with a 32-color scale.

Protein Immobilization on the Maleimide-Coated Substrate. We next examined two methods for assessing the binding of the tagged EGFP proteins to the maleimide-modified substrate. In the first, which we call the tipping method (Figure 6A), protein solutions are spotted on the substrate and incubated for 1 h at room temperature. In the second, the dipping method (Figure 6B), individual substrates are dipped into separate protein solutions. On the basis of fluorescence measurements, two kinds of purified 6×His-EGFP-5×Cys and 5×Cys-EGFP-6×His remained strongly attached to the substrate after washing (spots 2 and 4 in Figure 6). On the other hand, no fluorescence was detected for purified 6×His-EGFP (no. 1) or unpurified 5×Cys-EGFP (no. 3). These results indicated that both the N- and C-terminal Cys-tag allows immobilization of the protein on the maleimide-coated surface and that purification of Cys-tagged protein is essential for its immobilization. Efficiency of Cys-Tagged EGFP Immobilization. We next quantitatively evaluated the efficiency of protein immobilization on the maleimide-coated surface. We spotted a series of dilutions of purified 6×His-EGFP-5×Cys (1.5 pg to 1.25 ng) on a maleimide-coated substrate (Figure 7). After the substrate was washed, it was possible to detect as little as 50 pg of protein by GFP fluorescence (Figure 7B). This indicated that the threedimensional conformation of EGFP was essentially intact after immobilization and washing. To determine the lower limit of protein needed for immobilization, we measured bound protein by immunodetection with an anti-GFP antibody followed by enhanced chemiluminescence (Figure 7C). GFP amount spotted on the chip (Figure 7A) and fluorescence intensity on the chip after immobilization and washing (Figure 7B) have a strong correlation (R2 ) 0.9948, Figure 7D) in 50 pg to 1.25 ng amount. By this method, Cys-tagged protein was immobilized 2148

Journal of Proteome Research • Vol. 5, No. 9, 2006

Figure 7. Immunological detection of the amount of EGFP bound to the maleimide-coated substrate. (A) GFP fluorescent image of a protein chip spotted with a series of 6×His-EGFP-5×Cys dilutions. (B) GFP fluorescent image of (A) after washing with PBS followed by TBS/Tween-20. (C) Chemiluminescent image of (B) after immunodetection. The amount of protein spotted was indicated on theleft side for (A-C). (D) GFP fluorescent intensity (FL intensity) on chips after washing. Each FL intensity of immobilized protein spots after washing (B) were detected and plotted on the graph. The correlation (R2) is statistically significant.

on the chip in a dose-dependent manner, and we found that 5 pg of protein could bind on the chip even when the GFP fluorescence was undetectable, confirming that the immunodetection was more sensitive than the measurement of GFP fluorescence. Cys-Tagged EGFP-Fused Stathmin Immobilization and Detection on Chip. As the second model of recombinant

Protein Immobilization Using an Oligo-Cysteine Tag

research articles

Figure 9. Identification of stathmin on chip by MS/MS. (A) LCMS spectra of tryptic digests on EGFP-stathmin chip. The precursor ion m/z is 695.8 marked by a red diamond. The spectrum of the precursor shows a doubly charged peptide ion. (B) LC-MS/MS spectrum of the precursor ion with m/z 695.8. The original position is indicating as a red diamond. (C) The MS/MS spectrum same as (B) but processed with a Spectrum Mill Workbench. The spectrum is identified as tryptic peptide ASGQAFELILSPR from stathmin.

Figure 8. Immobilization of EGFP-Stathmin on a maleimidecoated substrate. (A) Recombinant proteins 6×His-EGFP-Stathmin-5×Cys (lower panel) were prepared in E. coli containing the corresponding plasmid vector, as a fusion protein modified from 6×His-EGFP-5×Cys (upper panel). H, histidine; C, cysteine. (B) Expression, isolation, and purification of the tagged EGFPStathmin. E. coli BL 21 cultures at 28 °C after overnight induction of recombinant proteins 6×His-EGFP-5×Cys or 6×His-EGFPStathmin-5×Cys were extracted and purified by Ni-beads affinity purification. The proteins before (ext) and after (beads) affinity purification were separated in 5-20% SDS-PAGE and stained with Coomassie Brilliant Blue R-250. Eluted solution of purified recombinant proteins, 6×His-EGFP-5×Cys(*) and 6×His-EGFPStathmin-5×Cys(**), were used for protein immobilization on chips. (C) GFP fluorescence and immunodetection of stathmin on chips. EGFP-Stathmin no. 3 chip and EGFP no. 4 chip were treated with the anti-stathmin mouse IgG as the first antibody. Then Alexa Fluor 555-conjugated anti-mouse IgG was used as the secondary antibody on the chip no. 3, no. 4 and also no. 2 (no first antibody) as the controls. After washing, GFP fluorescence of the chips was detected by Cy2 setting, and stathmin was immunodetected by Cy3 setting via Alexa Fluor 555 fluorescence of the secondary antibody. No.1 chip is the original maleimide substrate without protein immobilization as fluorescence control.

protein for protein chip, we tested stathmin, a human cytoplasmic protein that had double tags and EGFP fusion (6×HisEGFP-Stathmin-5×Cys, Figure 8A). Recombinant double-tagged EGFP-Stathmin was expressed and purified in the same manner as double-tagged EGFP (Figure 8B) and then eluted, and both protein solutions were spotted by the tipping method all over the maleimide substrate overnight and washed. GFP fluorescence of EGFP-stathmin chips was almost equivalent to that of EGFP chip under image scanner under Cy2 setting (Figure 8C, upper panel). To confirm whether stathmin is undoubtedly immobilized on the chip, the chips were immunodetected by stathmin antibody followed by Alexa Fluor 555-labeled second antibody and analyzed under Cy3 setting scanner to detect Alexa Fluora 555 fluorescence (Figure 8C, upper panel). The Alexa 555 fluorescence was detected on the EGFP-stathmin chip (no. 3 chip in Figure 8C) treated with both of anti-stathmin antibody and secondary antibody, but not on the EGFP chip (no. 4 chip) with same treatment, demonstrating that the no. 3 chip contains stathmin. There was almost no detection of the secondary antibody nonspecific background (no. 2 chip) and the original substrate background (no. 1 chip). Each one of the EGFP chips and EGFP-stathmin chips in the same preparations was trypsinized, then the chip substrate was taken out, and the peptide solutions were applied by LC-MS/MS analysis (Figure 9, Table 3). Four peptide sequences annotated as stathmin were detected from the EGFP-stathmin chip but not others. One peptide sequence annotated as GFP was detected from the EGFP-stathmin chip and the EGFP chip. Taken together from the protein fluorescence detection, immunological detection, and direct MS detection, it was shown that the Cys-tagged purified proteins were immobilized on maleimide-coated chips. Journal of Proteome Research • Vol. 5, No. 9, 2006 2149

research articles

Ichihara et al.

Table 3. The Result of LC-MS/MS for the EGFP-Chip, EGFP-Stathmin Chip, and Control Chip protein on chip

protein peptide % AA detection (no.) coverage score

EGFP

GFP Stathmin EGFP-Stathmin GFP Stathmin

1 0 1 5

4 0 4 31

PBS control

0 0

0 0

a

GFP Stathmin

15.43 0 16.94 69.92

0 0

sequence

SAMPEGYVQER SAMPEGYVQER AIEENNNFSK ASGQAFELILSPRa DLSLEEIQK RASGQAFELILSPR SKESVPEFPLSPPK -

Sequence identified by the spectrum data shown in Figure 9.

Discussion In the current studies, we designed and developed a new technology for protein chips (Figure 1). We first created a maleimide-modified silicon substrate as a solid support for covalently attaching thiol-containing biomolecules. XPS measurements of the modified substrate surface confirmed that maleimide groups could be introduced on the surface of the silicon substrate. Using oligo DNAs, we demonstrated thiolspecific binding to the maleimide-coated substrate. Finally, we showed that EGFP with a Cys-tag could be bound to the modified substrate. Purification of target protein was essential for its binding to the substrate; unpurified 5×Cys-EGFP could not bind. We suspect that cysteine residues of various proteins in the crude E. coli extract prevent 5×Cys-EGFP from binding to the substrate. Thus, the combination of a 6×His-tag for purification along with a 5×Cys-tag for immobilization is the most efficient for this system. The specific locations of the two tags in the recombinant protein were not critical, but they may be important for expression in E. coli. The most efficient combination for the production, purification, and immobilization of the recombinant protein was an N-terminal His-tag and a Cterminal Cys-tag. Finally, we found that this protein immobilization system is highly efficient. Fluorescence measurement could detect a signal using 50 pg (0.5 µL spot of 100 pg/µL) of EGFP, indicating that the protein’s three-dimensional conformation was essentially intact after immobilization. Moreover, an immunological method using enhanced chemiluminescence could detect a signal when as little as 5 pg (0.2 fmol, 0.5 µL spot of 10 pg/µL) of EGFP protein was used. This is efficient when it is compared with other protein immobilization systems; for example, coupling on nitrocellulose-coated slides was done with 1 µg/µL of protein solution,8 0.1-12.8 ng/µL of solution,13 and 10-50 ng/spot;12 or coupling on BSA-N-hydroxysuccinimide slides was via amino groups of 0.1-0.5 µg/mL of protein solution,6 although binding on Ni-coated slides with 6×Histagged 10-950 fg of protein/spot7 is more efficient. N-terminal cysteine-containing peptides (usually CGG- and under 10 amino acids) were printed covalently using 3-5 mg/mL peptide solution on thioester slide.19 In our method, target proteins are covalently immobilized the same as this peptide array, but to maleimide groups via a long linker, so we suspect that the protein is easy to react, attached in a uniform orientation, and exposed-well to the surrounding solution. This is also suggested in the Ni-coated slides/His-tagged protein system.7 This property allows efficient binding of the primary and secondary antibodies as well as the substrates for enhanced chemilumi2150

Journal of Proteome Research • Vol. 5, No. 9, 2006

nescence. The main benefit of the maleimide/Cys-tag covalent binding over the Ni/His-tag affinity binding24 is stability of immobilization, theoretically, which should allow stringent washes and give lower background signals. Furthermore, chelating agents, which interfere with affinity of His-tagged proteins to Ni-coated slides, may not interfere with the immobilization of Cys-tagged proteins to the maleimide substrate. Nitrocellulose-coated glass slides have become popular as a substrate for protein immobilization.10-14 Proteins bind tightly to nitrocellulose via hydrophobic interactions.25,26 Some highly charged proteins, however, may be difficult to bind to nitrocellulose, and detergent prevents nitrocellulose from tight binding of proteins.27 Also, nitrocellulose is expected to bind proteins in random orientations via internal hydrophobic residues. This may explain why nanogram quantitiessa relative high amounts of protein are utilized for nitrocellulose-coated chips. Cystagged proteins also have a possibility to be immobilized via its internal cysteine residues (Figure 1B, the third case in step 3), but it may happen in relatively low frequency because the terminal five cysteine residues will dominate to react with maleimide residues on the substrate. Cys-tag may have a role to retain the three-dimensional structure of the proteins fused with it, which will be required for the highly sensitive detection of biological functions on a chip, including protein-ligand interactions and enzymatic activity. In conclusion, we successfully developed a novel protein chip system, consisting of a maleimide-coated substrate and a fusion protein containing both Cys- and His-tags. This will be useful for the development of new protein chip applications. We are continuing to improve this system, including the number of cysteine residues for optimal function of the Cys-tag.

Acknowledgment. We thank Dr. Michifumi Tanga at Toyo Kohan Co., Ltd., and Dr. Susumu Tsunasawa at Shimadzu Co., Ltd., for valuable advice and discussions and Dr. Andre Sobel for sharing a stathmin plasmid. Plasmid construction works were carried out at the Gene Research Center of Yamaguchi University. This work was supported by grants-inaid for science research 17590648 (to J.K.A.), 17659166 (to Y.K.), and 16659142 (to K.N.) from the Ministry of Education, Science, Sports, and Culture of Japan. References (1) International Human Genome Sequencing Consortium. Initial sequencing and analysis of the human genome. Nature 2001, 409, 860-921. (2) Venter, J. C.; Adams, M. D.; Myers, E. W.; Li, P. W.; Mural, R. J.; Sutton, G. G.; Smith, H. O.; Yandell, M.; Evans, C. A.; Evanc, C. A.; et. al. The sequence of the human genome. Science 2001, 291, 1304-1351. (3) de Hoog, C. L.; Mann, M. Proteomics. Annu. Rev. Genomics Hum. Genet. 2004, 5, 267-293. (4) Resing, K. A.; Ahn, N. G. Proteomic strategies for protein identification. FEBS Lett. 2005, 579, 885-889. (5) Stoll, D.; Templin, M. F.; Bachmann, J.; Joos, T. O. Protein microarrays: Applications and future challenges. Curr. Opin. Drug Discovery Dev. 2005, 8, 239-252. (6) MacBeath, G.; Schreiber, S. L. Printing proteins as microarrays for high-throughput function determination. Science 2000, 289, 1790-1763. (7) Zhu, H.; Bilgin, M.; Bangham, R.; Hall, D.; Casamayor, A.; Bertone, P.; Lan, N.; Jasen, R.; Bidlingmaier, S.; Houfek, T.; Mitchell, T.; Miller, P.; Dean, R. A.; Gerstein, M.; Snyder, M. Global analysis of protein activities using proteome chips. Science 2001, 293, 2101-2105. (8) Espejo, A.; Cote, J.; Bednarek, A.; Richard, S.; Bedford, T. A protein-domain microarray identifies novel protein-protein interaction. Biochem. J. 2002, 367, 697-702.

research articles

Protein Immobilization Using an Oligo-Cysteine Tag (9) Yeast ProtoArray PPI (Invitrogen), http://www.invitrogen.com. (10) Hueber, W.; Utz, P. J.; Steinman, L.; Robinson, W. H. Autoantibody profiling for the study and treatment of autoimmune disease. Arthritis Res. 2002, 4, 290-295. (11) Davies, D. H.; Liang, X.; Hernandez, J. E.; Randall, A.; Hirst, S.; Mu, Y.; Romero, K. M.; Nguyen, T. T.; Kalantari-Dehaghi, M.; Crotty, S.; Baldi, P.; Villarreal, L. P. Profiling the human immune response to infection by using proteome microarrays: Highthroughput vaccine and diagnostic antigen discovery. Proc. Nat. Acad. Sci. U.S.A. 2005, 102, 547-552. (12) Knezevic, V.; Leetahnakul, C.; Bichsel, V. E.; Worth, J. M.; Prabha, V. V.; Gutkind, J. S.; Liotta, L. A.; Munson, P. J.; Petricoin, E. F., III; Krizman, D. B. Proteimic profiling of the canner microenvironment by antibody arrays. Proteomics 2001, 1, 1271-1278. (13) Michaud, G.; Salcius, M.; Zhou, F.; Bargham; R.; Bonin, J.; Guo, H.; Snyder, M.; Predki, P. F.; Schwetzer, B. I. Analysing antibody specificity with whole proteome microarrays. Nat. Biotechnol. 2003, 21, 1509-1512. (14) Panorama Ab Microarray-Cell Signaling Kit (SIGMA), http:// www.sigma-aldrich.com. (15) CyDye Maleimide Reactive Dye (GE helthcare), http:// www.gehealthcare.com. (16) Griffin, B. A.; Adams, S. R.; Tsien, R. Y. Specific covalent labeling of recombinant protein molecules inside live cells. Science 1998, 281, 269-272. (17) Adams, S. R.; Campbell, R. E.; Gross, L. A.; Nartin, B. R.; Walkup, G. K.; Yao, Y.; Llopis, J.; Tsien, R. Y. New biarsenical ligands and tetracystein motifs for protein labeling in vitro and in vivo: Synthesis and biological applications. J. Am. Chem. Soc. 2002, 124, 6063-6076. (18) Lesaichirre, M.-L.; Uttamchandani, M.; Chen, G. Y. J.; Yao, S. Q. Developing site-specific immobilization strategies of peptides in microarray. Bioorg. Med. Chem. Lett. 2002, 12, 2079-2083.

(19) Uttamchandani, M.; Chen, G. Y. J.; Lesaichirre, M.-L.; Yao, S. Site∧specific peptide immobilization strategies for the rapid detection of kinase activity on microarrays. Methods Mol. Biol. 2004, 264, 191-204. (20) Tsien, R. Y. The green fluorescent protein. Annu. Rev. Biochem. 1998, 67, 509-544. (21) Takahashi, K.; Tanga, M.; Takai, O.; Okamura, H. Diamond Relat. Mater. 2003, 12, 572-576. (22) Staros, J. V. N-Hydroxysulfosuccinimide active esters: Bis (Nhydrosulfosuccinimide) esters of two diarboxylic acid are hydrophilic, membrane-impermeant, protein cross-linkers. Biochemistry 1982, 21, 3950-3955. (23) Shimada, K.; Mitamura, K. Derivatization of thil-containing compounds. J. Chromatogr., B 1994, 659, 227-241. (24) Terpe, K. Overview of tag protein fusions: from molecular and biochemical fundamentals to commercial systems. Appl. Microbiol. Biotechnol. 2003, 60, 523-533. (25) Farrah, S. R.; Shah, D. O.; Ingram, O. Effects of chaotropic and antichaotropic agents on elution of poliovirus adsorbed on membrane filters. Proc. Natl. Acd. Sci. U.S.A. 1981, 78, 12291232. (26) Gershoni, J.; Palade, G. E. Electrophoretic transfer of proteins from sodium dodecyl sulfate- polyacrylamide gels to positively charged membrane filter. Anal. Biochem. 1982, 124, 396-405. (27) Nakamura, K.; Tanaka, T.; Kuwahara, A.; Takeo, K. Microassay for proteins on nitrocellose filter using protein dye-staining procedure. Anal. Biochem. 1985, 148, 311-319.

PR0504889

Journal of Proteome Research • Vol. 5, No. 9, 2006 2151