Construction of Semisynthetic DNA–Protein Conjugates with Phi X174

Technical Note pubs.acs.org/bc

Construction of Semisynthetic DNA−Protein Conjugates with Phi X174 Gene-A* Protein Yasumasa Mashimo, Hitomi Maeda, Masayasu Mie, and Eiry Kobatake* Department of Biological Information, Graduate School of Bioscience and Biotechnology, Tokyo Institute of Technology, 4259, Nagatsuta, Midori-ku, Yokohama-shi, 226-8501, Japan S Supporting Information *

ABSTRACT: DNA−protein conjugates have frequently been used as versatile molecular tools for a variety of applications in biotechnology to harness synergistic effects of DNA and protein functions. With applications for DNA−protein conjugates growing, easy-to-use and economical methods for the synthesis of DNA−protein conjugates are required. In this study, we developed a method for site-specific labeling of single-stranded DNA (ssDNA) to a recombinant protein of interest (POI) through the Gene-A* protein (Gene-A*) from bacteriophage phi X174, without any chemical modifications of ssDNA. Gene-A* protein is an enzyme that site-selectively cleaves an oligodeoxyribonucleotide (ODN) containing a Gene-A* recognition sequence, at which point a tyrosine residue of Gene-A* is bonded to the 5′-phosphoryl group of the cleavage site via a stable phosphotyrosine linkage. Here, we constructed three kinds of recombinant proteins fused to Gene-A*: N-terminally Gene-A*-fused enhanced green fluorescent protein (EGFP), C-terminally Gene-A*-fused EGFP, and N-terminally Gene-A*-fused firefly luciferase (FLuc). The reaction yields of DNA−protein conjugation catalyzed by the Gene-A* moiety reached 80−90% in the three proteins, and kinetic study revealed that the reaction achieved a steady state after 10 min. Moreover, dot blot analyses were performed to evaluate the hybridization and aptamer-forming ability of ssDNA conjugated to the Gene-A* moiety of a recombinant Gene-A*-FLuc protein. This study demonstrated that a strategy using recombinant proteins fused to Gene-A* could offer a versatile, rapid, easy-to-use, and economical platform for producing DNA−protein conjugates.

■

INTRODUCTION In recent years, semisynthetic DNA−protein conjugates have been used as versatile molecular tools for a variety of applications in biotechnology and materials science.1,2 In particular, the utility of DNA−protein conjugates has emerged in ultrasensitive detection methods to quantify target biomolecules,3,4 and DNA-directed immobilization techniques for the fabrication of protein microarrays.5 Moreover, some potential applications of DNA−protein conjugates are being developed in cutting-edge biotechnologies, such as kinetic analysis of complementation of split-proteins,6,7 biomolecular delivery systems,8,9 and programmable extracellular matrices.10 Despite the potential applicability of DNA−protein conjugates, synthetic methods remain challenging. Total chemical synthesis of DNA−protein conjugates has been successful over the past years,11−13 although the incorporation of oligonucleotides into larger recombinant proteins, such as antibodies, is still poorly developed, and it is not possible, in many cases, to engineer a single chemically accessible cysteine into the protein of interest (POI). Furthermore, the methods require extensive purification of the conjugates to remove excessive protein and oligonucleotides after each coupling step.14 As alternatives, expressed protein ligation reactions15 and enzyme-mediated (e.g., sortase and transglutaminase)16−18 conjugation reactions © 2012 American Chemical Society

have been reported. They are advantageous because one can achieve a relatively high yield (50−75%) of the reaction product stoichiometrically conjugated to single-stranded DNA (ssDNA). Another approach to generate DNA−protein conjugates harnesses a tag-protein with specific self-labeling activity (e.g., methyltransferase and O6-alkylguanine-DNAalkyl-transferase),19,20 possessing the marked features of a high yield (∼85%) and few preparation steps. In these methods, however, it is indispensable to prepare DNA substrates modified with a specific functional moiety, which is cumbersome and not economical. In the present study, we developed an easy-to-use and economical method for site-selective conjugation of a recombinant POI with nonmodified ssDNA. This approach made use of phi X174 Gene-A* (Gene-A*) of coliphage phi X174 (Figure 1). The A-gene of phage phi X174 encodes two polypeptides:21 the Gene-A protein (55 kDa) and the Gene-A* protein (37 kDa), which results from internal translation of the gene-A message beginning at the 173rd codon. The Gene-A protein cleaves the double-stranded phi X174 DNA between Received: March 12, 2012 Revised: May 19, 2012 Published: May 23, 2012 1349

dx.doi.org/10.1021/bc300118m | Bioconjugate Chem. 2012, 23, 1349−1355

Bioconjugate Chemistry

Technical Note

Table 1. Sequences of Gene-A* Substrate Oligodeoxyribonucleotides (S-ODN) name

sequence

S-ODN31-FITC S-ODN-51

5′-TCG ACA ACT TGAa TAT TAA TAA CAC TAT AGA CFITC-3′ 5′-TCG ACA ACT TGA TAT TAA TAA CAC TAT AGA CCA CCG CCC CGA AGG GGA CGAb-3′ 5′-TCG ACA ACT TGA TAT TAA TAA CAC TAT AGA CGG TTG GTG TGG TTG Gc-3′

S-ODNTBA a

The underline shows the Gene-A* cleavage site. bThe italic characters show the hybridization site. cThe bold characters show the trombin binding aptamer sequence.

recognition sequence,25 and all other oligonucleotide sequences were purchased from Greiner Bio-One. Reagents for a bioluminescent assay were obtained from PicaGene kit (TOYO B-Net). All other chemicals and proteins were of analytical grade. Plasmid Construction. The phi X174 Gene-A* protein gene fragment was amplified by the polymerase chain reaction (PCR) from ϕX174 DNA (Takara Bio), and inserted into the pBSII(SK+) (Cosmo Bio) plasmid. The expression plasmid pET28b-Gene-A* with the N-terminal His-tag sequence was constructed from the pBSII(SK+) plasmid encoding Gene-A* (pBS-A*). It turns out that the Y303H mutation is necessary for overexpression of the Gene-A* protein in E. coli.24 Therefore, Y303H mutation was introduced into pET28bGene-A* as described in the instruction of QuikChange II SiteDirected Mutagenesis Kits (Agilent Technologies). In all of the following plasmid construction procedures, the plasmids encoding the Gene-A* gene with the Y303H mutation were used. Enhanced green fluorescent protein (EGFP) and firefly luciferase (FLuc) gene fragments were amplified by PCR from pEGFP-N3 (BD Biosciences) and pBS-FLuc encoding a FLuc gene, which was constructed previously, respectively. The plasmid pET28b-NHis-EGFP-A* was generated by inserting the EGFP gene fragment into the N-terminal of the Gene-A* gene of pET28b-A*. The plasmid pET28b-NHis-A*-EGFP and pET28b-NHis-A*-FLuc were constructed by inserting EGFP and FLuc gene fragments into the C-terminal of Gene-A* gene of pET28b-A*. Protein Expression and Purification. E. coli BL21(DE3) transformed with each of the three pET28b-NHis-(A*-EGFP, EGFP-A*, A*-FLuc) plasmids was grown in 50 mL of Luria broth supplemented with 0.5% glucose at 37 °C to an apparent optical density of 0.3 at 600 nm. Expression of the Gene-A* fusion protein was induced by the addition of isopropyl-β-Dthiogalactopyranoside (IPTG) to 0.5 mM. After the incubation at 19 °C for 16 h, cells were pelleted, resuspended in 5 mL/ gpellet of sucrose buffer (50 mM HEPES, 20% sucrose, 1 mM EDTA pH 7.9) before repelleting by centrifugation at 7000g for 30 min at 4 °C. The supernatant was discarded and the pellet resuspended in 5 mL/gpellet of 5 mM MgCl2 and incubated on ice for 10 min. Cells were pelleted by centrifugation at 4500g for 20 min, the supernatant was discarded, and the pellet was resuspended in 1.5 mL/gpellet lysis buffer (50 mM sodium phosphate, 300 mM NaCl, 10 mM MgCl2, 10% glycerol, 10% sucrose, 10 mM β-mercaptoethanol, 50 mM L-arginine hydrochloride, pH 8.0). Cell lysis was carried out by adding BugBuster (Merck Chemicals), incubation for 30 min at 4 °C, and then sonication for 5 min in 5 s pulses. The lysates were centrifuged at 15 000 rpm for 15 min. All steps were performed on ice or at 4 °C. Then, the collected supernatants were applied

Figure 1. Schematic representation of strategy for conjugation of a recombinant protein of interest (POI) with ssDNA through the fused Gene-A* moiety (A*). The recombinant protein cleaves the recognition sequence, and a phosphodiester bond is formed between the 5′-phosphoryl group of the cleavage site and the tyrosine residue of the Gene-A* moiety. An underline in the substrate oligodeoxyribonucleotide (ODN) shows the Gene-A* recognition sequence. The GpA in the recognition sequence denotes the cleavage site.

nucleotides G-4305 and A-4306 at the replication origin of the viral strand and binds covalently to the 5′-phosphoryl group of A-4306 via a stable phosphotyrosine link. In the presence of E. coli rep DNA helicase and DNA polymerase III, elongation of of phi X174 DNA is initiated at the 3′-OH of G-4305 generated by Gene-A. Gene-A* also has the property of cleaving and binding covalently to ssDNA as well as Gene-A.22,23 Remarkably, Gene-A* can link to ssDNA in vitro in the presence of only divalent metal ions (e.g., Mg2+) and does not require other proteins or biomolecules.24 Specifically, Gene-A* can couple itself to ssDNA containing the conserved nucleotide sequence present in the replication origin of phi X174 DNA, without any chemical modification.24,25 In this paper, we describe the synthesis of novel DNA−protein conjugates using three recombinant proteins with the autolabeling activity: Nterminally Gene-A*-fused enhanced green fluorescent protein (EGFP), C-terminal Gene-A*-fused EGFP, and N-terminally Gene-A*-fused firefly luciferase (FLuc). The recombinant proteins cleave the recognition sequence, and a phosphodiester bond is formed between the 5′-phosphoryl group of the cleavage site and the tyrosine residue of the Gene-A* moiety. We demonstrated that the robust autolabeling activity of GeneA* could be maintained with any proteins. Moreover, the Gene-A* moiety did not interfere with the functions of the fusion partner protein or conjugated ssDNA.

■

EXPERIMENTAL PROCEDURES Materials. All substrate oligodeoxyribonucleotide (S-ODN) sequences (Table 1), containing the conserved Gene-A* 1350

dx.doi.org/10.1021/bc300118m | Bioconjugate Chem. 2012, 23, 1349−1355

Bioconjugate Chemistry

Technical Note

another protein has the capacity to form a complex with a substrate oligodeoxyribonucleotide (ODN), sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) analysis of the conjugates of the Gene-A* fusion protein with the substrate ODN in a cell lysate was performed. A cell lysate of an E. coli BL21(DE3) expressing the N-terminally thioredoxinHis6-tagged Gene-A* (Trx-Gene-A*) protein was mixed with the 3′-FITC-labeled substrate ODN (S-ODN-31-FITC). The substrate ODN sequence is the 31 nucleotide derived from the replication origin of phi X174 DNA, including the conserved Gene-A* recognition sequence (CAA CTT GATA). The ODN-Trx-Gene-A* conjugates in the cell lysate were electrophoretically separated, imaged with FITC labeled via the conjugated ODN, and verified as a single band (Figure S1A in the Supporting Information). It has been reported that both of two tyrosine residues, Tyr-343 and Tyr-347, of the Gene-A* protein are directly involved in Gene-A*-mediated cleavage and joining.27 Therefore, we constructed Y343F, Y347F, and Y343F/Y347F Gene-A* mutants fused to Trx to examine the stoichiometrically DNA binding activity of wild-type (WT) Gene-A* fusions by SDS-PAGE (Figure S1B in the Supporting Information). It was found that the fusion proteins with WT, Y343F, and Y347F Gene-A* could cleave the single-stranded substrate ODN and join the cleavage site to itself stoichiometrically, while the Y343F/Y347F mutant could not. Since no differences in the binding abilities to the substrate ODN were observed among the proteins, we used Gene-A* WT for the following experiments. To ensure whether Gene-A* could form a complex with DNA without affecting the DNA binding activity and the activity of the accompanied fused protein, plasmids coding Nterminally EGFP-fused Gene-A* (N-EA*) and C-terminally EGFP-fused Gene-A* (A*E-C) proteins were constructed. The proteins with His6-tag were overexpressed in E. coli BL21(DE3) and purified in a single step using a TALON metal affinity gel. First, the fluorescence of N-EA* and A*E-C proteins was confirmed by spectrofluorometric analysis (FP-6500, Jasco) (data not shown). The expected sizes from their amino acid sequences of the N-EA* and A*E-C proteins were 67.9 kDa and 67.2 kDa, respectively. The estimated molecular weights of the proteins from SDS-PAGE were a little smaller than the expected sizes. The bands of both N-EA* and A*E-C were shifted to higher molecular weights, after the reaction with SODN-31-FITC (Figure 2A,B). Thus, it was shown as with the case of Trx that the Gene-A* moiety of Gene-A* and EGFP fusion proteins had the DNA binding activity. The singleshifted bands of FITC-labeled substrate ODN after the reaction with N-EA* and A*E-C were confirmed by fluorescence imaging. Those results indicated that each product contained a single molecular species, with only slight impurities of the unconjugated protein (SI Figure S2). Subsequently, we examined the yield of the DNA−protein conjugation reaction. On the basis of the protein bands in SDS-PAGE analysis, both the reaction yields of N-EA* and A*E-C were estimated to be approximately 80−90% when the conjugation reaction was conducted with a 2-fold molar excess of S-ODN-31-FITC to fusion proteins (Figure 2B). The yield of present method is comparable to that of previously reported conjugation methods advantageous in yield.16,20 We speculate as one reason of the high yields of DNA−N-EA* and DNA−A*E-C conjugates that the reaction efficiency is increased because the substrate ODN includes the Gene-A* interacting sequence (CAC TAT AGA C) as well as the recognition sequence.25

to TALON Metal Affinity Resins (Takara Bio), and the fusion proteins were purified by metal-ion affinity chromatography using the His-tag. The fusion proteins were concentrated by the use of Amicon Ultra-15 10K (Merck), and dialyzed against the storage buffer (50% glycerol, 1 M NaCl, 50 mM Tris-HCl, 1 mM EDTA, 5 mM β-mercaptoethanol, pH 7.5), and stored at −20 °C. All fusion proteins were analyzed by SDS-PAGE (8% acrylamide gels). Each concentration of the purified proteins was determined with the bicinchoninic acid protein assay reagent kit (Pierce). Synthesis of DNA−Protein Conjugates. The appropriate concentration of each Gene-A* fusion protein was mixed with a 2× concentration of S-ODN in the reaction buffer (150 mM NaCl, 50 mM Tris-HCl, pH 7.5, 5 mM dithiothreitol (DTT), and 5 mM MgCl2), and then, the DNA cleavage and joining reaction was carried out at 30 °C for 30 min, resulting in the tyrosyl residue (Tyr-343 or Tyr-347) of Gene-A* being condensed in a phosphodiester bond formation. To remove unreacted S-ODNs, the fusion proteins reacted with S-ODNs were dialyzed against the storage buffer using 20 000 MWCO Slide-A-Lyzer (Thermo Scientific), and stored at −20 °C until use. Dot Blot Analysis. Dot blot analysis was performed on a synthetic complementary oligodeoxyribonucleotide (C-ODN) and a human α-thrombin. The C-ODN and α-thrombin were dissolved in formaldehyde/SSC buffer (1.8 M NaCl, 0.1 M sodium phosphate, 0.01 M EDTA, 6.15 M formaldehyde, pH 7.7) and selection buffer (20 mM Tris-acetate, 140 mM NaCl, 50 mM KCl, 10 mM CaCl2, pH 7.4),26 respectively. Before spotting, the C-ODN in the buffer was heated at 95 °C for 3 min, and then chilled on ice for at least 3 min. Then, the CODN and α-thrombin were spotted onto the Hybond-C nitrocellulose membranes (Amersham Life Science) prewetted in 10× standard saline citrate (SSC) and selection buffer, respectively. Subsequently, the membranes were dried, and the membranes spotted with the C-ODN were baked at 80 °C for 2 h. The sample spotted membranes were incubated in prebinding buffer for C-ODN (5× sodium chloride/sodium potassium/EDTA (SSPE), 500 μg/mL sonicated salmon sperm DNA, 0.2× Blocking One (Nacalai Tesque), 5 mM DTT, 0.05% (v/v) Tween 20, pH 7.7), and for α-thrombin (1× selection buffer, 500 μg/mL sonicated salmon sperm DNA, 0.2× Blocking One (Nacalai Tesque), 0.05% (v/v) Tween 20, pH 7.4) at 37 °C for 3 h. Then, DNA−protein conjugates were added to the prebinding solution, and incubated at 37 °C for 3 h. Following the binding assay, the filters were washed by incubating themselves in the wash buffer (C-ODN: 2 × SSPE, 0.05% Tween 20, 1 mM DTT, pH 7.7; α-thrombin: 1 × selection buffer, 0.05% Tween 20, 1 mM DTT, pH 7.4) for 15 min, repeatedly, and then the wash buffer was replaced with the second wash buffer (C-ODN: 1 × SSPE, 0.05% Tween 20, 1 mM DTT, pH 7.7; α-thrombin: 1 × selection buffer, 0.05% Tween 20, 1 mM DTT, pH 7.4), and washed for 15 min. After two washes with the second wash buffer, the membranes were immersed in a 1 mL PicaGene solution as a substrate for luciferase and monitored for a fixed amount of time by LumiCube (Liponics) to measure the luminescence of each spot.

■

RESULTS AND DISCUSSION Characterization of Gene-A* Fused to POI. Gene-A* from coliphage phi X174 has not been analyzed for its activity as a fusion protein. To determine whether Gene-A* fused to 1351

dx.doi.org/10.1021/bc300118m | Bioconjugate Chem. 2012, 23, 1349−1355

Bioconjugate Chemistry

Technical Note

Figure 2. Gel electrophoretic analysis of the ODN−EGFP conjugates. (A,B) 10% SDS-PAGE analysis of 2.5 μM N-EA* (A) and A*E-C (B) in the reactions with different concentrations of single-stranded substrate ODN: lane M, SDS Broadrange Marker (Bio-Rad); lane 1, no substrate ODN; lane 2, 0.5 μM S-ODN-31-FITC; lane 3, 1 μM SODN-31-FITC; lane 4, 2 μM S-ODN-31-FITC; lane 5, 5 μM S-ODN31-FITC; lane 6, 10 μM S-ODN-31-FITC; lane 7, 20 μM S-ODN-31FITC. (C) The fractions of the ODN−EGFP conjugates were quantified using ImageJ software. The yield was calculated as a ratio by value of a band intensity of the ODN−protein conjugates and a total band intensity of the proteins reacted and nonreacted with S-ODN.

Figure 3. Reaction kinetics of ODN−Gene-A*-EGFP conjugation. (A,C) N-EA* proteins (2.5 μM) (A) and A*E-C proteins (2.5 μM) (C) were mixed with S-ODN-31-FITC (5 μM) for the times indicated at different temperatures. The proteins were separated by SDS-PAGE and visualized by Coomassie blue staining. The retarded bands correspond to the ODN−Gene-A*-EGFP conjugates. (B,D) The fractions of conjugates of N-EA* (B) and A*E-C (D) with S-ODN were quantified using ImageJ software. The yield was calculated as a ratio by value of a band intensity of the ODN−protein conjugate and a total band intensity of the proteins reacted and nonreacted with SODN.

The reaction kinetics were investigated with preformed DNA−A*E-C and DNA−N-EA* conjugates. The yield of conjugates was assessed by SDS-PAGE followed by Coomassie blue staining (Figure 3A,C). The upper band, appearing over time at various temperatures, corresponds to the conjugated DNA−protein complex. The DNA−protein conjugation reactions mediated by the Gene-A* moiety reached a plateau after 10 min at 30 and 37 °C (Figure 3B,D). These results indicated that there was little differences between DNA binding kinetics of the Gene-A* moiety of recombinant fusion proteins and natural Gene-A*.23 Given that there are few methods which could complete the reaction for synthesis of DNA− protein conjugates within 10 min, the present strategy allowed the extremely rapid reaction. Moreover, the yields of DNA− A*E-C and DNA−EA* conjugates were similar at 4, 30, and 37 °C at each time point. Thus, these data showed that Gene-A* could be fused to both the N-terminus and the C-terminus of a POI without significant reduction in DNA binding activity of Gene-A* or decline in the fused protein function. In the next stage, we investigated the versatility of the GeneA* fusion strategy and the function of ssDNA conjugated to the Gene-A* fusion protein. It has been reported that there are some problems in processing of the genetically engineered firefly luciferase (FLuc) fusion protein when the C-terminus of FLuc is fused to other proteins.28 Therefore, we constructed the plasmids coding for a C-terminally FLuc-fused Gene-A* (A*L) protein with His6-tag, and a protein was overexpressed and purified in the same procedures as in the preparation of NEA* and A*E-C. To evaluate the activity of the A*L protein, three μM A*L protein and different concentrations of S-ODN31-FITC were mixed, separated by SDS-PAGE and visualized by Coomassie blue staining and fluorescence (SI Figure S3). The expected size of the A*L protein was 100.2 kDa, and the estimated molecular weight from SDS-PAGE was a little smaller than the expected one. The bands of the A*L proteins

mixed with various concentrations of S-ODN-31-FITC were shifted in a manner similar to that of both N-EA* and A*E-C proteins (SI Figure S3A), and the single-shifted band of SODN-31-FITC was observed in samples mixed with the A*L protein (SI Figure S3B). The reaction efficiency of the GeneA* moiety of the A*L protein was similar to that of the N-EA* and A*E-C proteins (SI Figure S3C). Thus, the Gene-A* protein fusion strategy for fabricating DNA−protein conjugates could be applied to the FLuc protein. One reason for the versatility of our strategy would be that the Gene-A* moiety is not susceptible to steric hindrance by the fused protein of interest because the Gene-A* protein has the relatively large molecular weight, unlike a small reactive group (e.g., cysteine residue).13 Dot Blot Analysis of DNA−Protein Conjugates Immobilized by Hybridization. To evaluate the function of the fused ssDNA, we fabricated the A*L protein conjugated to the S-ODN-51 substrate oligodeoxyribonucleotide. The conjugation reaction was conducted with a 2-fold molar excess of S-ODN-51 to the A*L protein, and unreacted oligonucleotides were removed by dialysis. The presence of some unconjugated fusion protein did not affect the efficiency of binding to the membrane, because that process was exclusively dependent on the presence of the conjugated oligonucleotide. First, we investigated whether the ODN conjugated to the A*L protein could hybridize to a complementary oligonucleotide sequence. Thus, ODN−A*L conjugates were immobilized on a nitrocellulose membrane through hybridization with a complementary oligodeoxyribonucleotide (C-ODN: 5′-AAA ACC 1352

dx.doi.org/10.1021/bc300118m | Bioconjugate Chem. 2012, 23, 1349−1355

Bioconjugate Chemistry

Technical Note

ATT TTT CGT CCC CTT CGG GGC GGT GGT CTA TAG TGT TAT TAA TAT-3′). The bioluminescent signals were observed in membranes spotted with different amounts of C-ODN and a noncomplementary oligodeoxyribonucleotide (NC-ODN: 5′-TTT ATC GTG GAT CGT GAT AAG TCT TTA ATC-3′) after hybridization with the ODN−A*L conjugates (Figure 4). The bioluminescence from the ODN−

Figure 4. Dot blot analysis of DNA-directed immobilization of SODN-51−A*L conjugates. (A) The bioluminescent image of the spots of various concentrations of C-ODN (0.064, 0.32, 1.6, 8, 40, and 200 pmol) and NC-ODN (200 pmol) visualized by the bioluminescence of immobilized S-ODN-51−A*L conjugates. (B) The bioluminescent intensity of the respective spots was quantified using ImageJ software. Error bars represent the mean ± s.d., n = 3.

Figure 5. Dot blot analysis of the S-ODN-TBA−A*L conjugates specifically binding to thrombin. The respective bioluminescent intensities of the S-ODN-TBA−A*L conjugates immobilized on spots of indicated amounts of thrombin and lysozyme (A) and 1 μg of thrombin, BSA, lysozyme, and myoglobin (B) on a nitrocellulose membrane were quantified using ImageJ software. The insets in A and B show the bioluminescent image of spots of the respective proteins visualized by the bioluminescence of immobilized S-ODN-TBA−A*L conjugates. Error bars represent the mean ± s.d., n = 3.

A*L conjugates immobilized on the spots was directly proportional to the amount of C-ODN, not NC-ODN. Thus, the results demonstrated that ssDNA conjugated to the GeneA* moiety could hybridize selectively to the complementary sequence. Dot Blot Analysis of TBA−Protein Conjugates binding to Thrombin. We then investigated whether a ssDNA aptamer conjugated to A*L could bind targets depending upon the folding and three-dimensional structure. In this study, a thorombin binding aptamer (TBA) was used as a model. The A*L protein conjugated to a S-ODN-TBA substrate oligodeoxyribonucleotide, containing a Gene-A* recognition sequence and a TBA sequence, was obtained in the same manner as the S-ODN-51−A*L conjugates. The functional analysis was conducted by using a dot blot method with a nitrocellulose membrane spotted with thrombin and other proteins (Figure 5). The different amounts (μg per spot) of thrombin or lysozyme were spotted on a nitrocellulose membrane, and the membrane was incubated in the binding buffer containing 50 nM TBA−A*L conjugates at 37 °C for three hours. After washout of nonbinding conjugates, the bioluminescence on the respective spots was visualized (Figure 5A). We observed that the bioluminescence signal increased significantly over 0.125−1 μg of thrombin. Furthermore, using a membrane spotted with 1 μg of thrombin, BSA, lysozyme, or myoglobin, thrombinspecific binding of TBA−A*L conjugates was examined. The strong bioluminescence signal was detected specifically from the spot of thrombin (Figure 5B). Thus, the results showed that the TBA conjugated to the Gene-A* moiety maintained its aptamer forming ability. Although 1 μg of thrombin corresponds approximately to 27 pmol, the signal intensity

was much weaker than that of conjugates immobilized through hybridization. A possible explanation for this outcome is that the aptamer structure is susceptible to the incubation conditions and buffer composition. Alternatively, the instability of the TBA might have increased due to the addition of 20 nucleotides and large protein moiety to its 5′-end.29 To improve the performance of the assay, optimization of assay conditions should be pursued. Parameters should include incubation temperature and time, the concentration of TBA− protein conjugates, the pore size and affinity of the membrane, and buffer composition. Moreover, utilization of chemical ligation methods, or the fusion protein strategy with other DNA binding proteins that could covalently bond themselves to 3′-terminus of substrate ODN,30 instead of the Gene-A* protein, could be beneficial.

■

CONCLUSION We have described an easy-to-use and economical method for conjugating recombinant proteins site-specifically to the 5′-end of ssDNA, using the Gene-A* protein. The method consists of the breakage and joining of DNA strands via stable transesterification reactions, without any chemical modification of the substrate ODN. Once the recombinant protein fused to either N-terminus or C-terminus of Gene-A* is constructed using genetic engineering methods, the present method should 1353

dx.doi.org/10.1021/bc300118m | Bioconjugate Chem. 2012, 23, 1349−1355

Bioconjugate Chemistry

Technical Note

(7) Takeda, S., Tsukiji, S., Ueda, H., and Nagamune, T. (2008) Covalent split protein fragment-DNA hybrids generated through Nterminus-specific modification of proteins by oligonucleotides. Org. Biomol. Chem. 6, 2187−2194. (8) Chen, C. H., Dellamaggiore, K. R., Ouellette, C. P., Sedano, C. D., Lizadjohry, M., Chernis, G. A., Gonzales, M., Baltasar, F. E., Fan, A. L., Myerowitz, R., and Neufeld, E. F. (2008) Aptamer-based endocytosis of a lysosomal enzyme. Proc. Natl. Acad. Sci. U.S.A. 105, 15908−15913. (9) Tong, G. J., Hsiao, S. C., Carrico, Z. M., and Francis, M. B. (2009) Viral capsid DNA aptamer conjugates as multivalent celltargeting vehicles. J. Am. Chem. Soc. 131, 11174−11178. (10) Aldaye, F. A., Senapedis, W. T., Silver, P. A., and Way, J. C. (2010) A structurally tunable DNA-based extracellular matrix. J. Am. Chem. Soc. 132, 14727−14729. (11) Niemeyer, C. M., Sano, T., Smith, C. L., and Cantor, C. R. (1994) Oligonucleotide-directed self-assembly of proteins: semisynthetic DNA−streptavidin hybrid molecules as connectors for the generation of macroscopic arrays and the construction of supramolecular bioconjugates. Nucleic Acids Res. 22, 5530−5539. (12) Kukolka, F., and Niemeyer, C. M. (2004) Synthesis of fluorescent oligonucleotide−EYFP conjugate: towards supramolecular construction of semisynthetic biomolecular antennae. Org. Biomol. Chem. 2, 2203−2206. (13) Lapiene, V., Kukolka, F., Kiko, K., Arndt, A., and Niemeyer, C. M. (2010) Conjugation of fluorescent proteins with DNA oligonucleotides. Bioconjugate Chem. 21, 921−927. (14) Niemeyer, C. M. (2010) Semisynthetic DNA-protein conjugates for biosensing and nanofabrication. Angew. Chem., Int. Ed. Engl. 49, 1200−1216. (15) Lovrinovic, M., Seidel, R., Wacker, R., Schroeder, H., Seitz, O., Engelhard, M., Goody, R. S., and Niemeyer, C. M. (2003) Synthesis of protein-nucleic acid conjugates by expressed protein ligation. Chem. Commun., 822−823. (16) Pritz, S., Wolf, Y., Kraetke, O., Klose, J., Bienert, M., and Beyermann, M. (2007) Synthesis of biologically active peptide nucleic acid-peptide conjugates by sortase-mediated ligation. J. Org. Chem. 72, 3909−3912. (17) Tominaga, J., Kemori, Y., Tanaka, Y., Maruyama, T., Kamiya, N., and Goto, M. (2007) An enzymatic method for site-specific labeling of recombinant proteins with oligonucleotides. Chem. Commun., 401− 403. (18) Kitaoka, M., Tsuruda, Y., Tanaka, Y., Goto, M., Mitsumori, M., Hayashi, K., Hiraishi, Y., Miyawaki, K., Noji, S., and Kamiya, N. (2011) Transglutaminase-mediated synthesis of a DNA-(enzyme)n probe for highly sensitive DNA detection. Chemistry 17, 5387−5392. (19) Smith, S. S., Niu, L., Baker, D. J., Wendel, J. A., Kane, S. E., and Joy, D. S. (1997) Nucleoprotein-based nanoscale assembly. Proc. Natl. Acad. Sci. U.S.A. 94, 2162−2167. (20) Jongsma, M. A., and Litjens, R. H. (2006) Self-assembling protein arrays on DNA chips by auto-labeling fusion proteins with a single DNA address. Proteomics 6, 2650−2655. (21) Linney, E., and Hayashi, M. (1973) Two proteins of Gene A of ϕX174. Nat. New Biol. 245, 6−8. (22) Langeveld, S. A., van Mansfeld, A. D., Baas, P. D., Jansz, H. S., van Arkel, G. A., and Weisbeek, P. J. (1978) Nucleotide sequence of the origin of replication in bacteriophage ϕX174 RF DNA. Nature 271, 417−420. (23) Sanhueza, S., and Eisenberg, S. (1984) Cleavage of singlestranded DNA by the ϕX174 A* protein: The A*-single-stranded DNA covalent linkage. Proc. Natl. Acad. Sci. U.S.A. 81, 4285−4289. (24) Hanai, R., and Wang, J. C. (1993) The mechanism of sequencespecific DNA cleavage and strand transfer by phi X174 gene A* protein. J. Biol. Chem. 268, 23830−23836. (25) Heidekamp, F., Baas, P. D., and Jansz, H. S. (1982) Nucleotide sequences at the ϕX gene A protein cleavage site in replicative form I DNAs of bacteriophages U3, G14, and α3. J. Virol. 42, 91−99.

be applicable for the attachment of proteins to DNA in fewer steps than currently available methods. Furthermore, the yield of DNA−protein conjugates in the present method reached approximately 80−90% in all three proteins for only 10 min. For the functional assessment of conjugated ssDNA, we developed two kinds of biomolecular detection systems to evaluate the hybridization ability toward cDNA and the thrombin binding ability of TBA. It has been demonstrated that S-ODN-51 and S-ODN-TBA conjugated to recombinant protein can selectively hybridize to the complementary oligonucleotide sequence and bind to thrombin. Recently, notable advances have been made in DNA-based nanotechnology such as DNA origami, DNA architecture, and DNA-fueled molecular machines.31−34 In combination with the technology, DNA−protein conjugates could become a platform for the construction of nanometer-sized functional biomaterials. To accelerate the research, development and diffusion of the functionalized nanostructures, therefore, an easy-to-use and economical method to conjugate proteins site-selectively and stoichiometrically to DNA is necessary. It is hoped that the present method for the synthesis of DNA−protein conjugates using the Gene-A* protein will serve as a valuable tool for DNA-based nanotechnology, or offer important implications for development of better conjugation strategies.

■

ASSOCIATED CONTENT

S Supporting Information *

Confirmation of the stoichiometrically DNA binding activity of wild-type and mutant Gene-A* proteins, SDS-PAGE analysis by fluorescence imaging, characterization of the DNA binding activity of A*L. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel.: +81-45-924-5760. Fax: +81-45-924-5779. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS This work was financially supported in part by the JSPS Global Centers of Excellence (COE) Program. REFERENCES

(1) Saccà, B., and Niemeyer, C. M. (2011) Functionalization of DNA nanostructures with proteins. Chem. Soc. Rev. 40, 5910−5921. (2) Röglin, L., and Seitz, O. (2008) Controlling the activity of peptides and proteins with smart nucleic acid-protein hybrids. Org. Biomol. Chem. 6, 3881−3887. (3) Burbulis, I., Yamaguchi, K., Gordon, A., Carlson, R., and Brent, R. (2005) Using protein-DNA chimeras to detect and count small numbers of molecules. Nat. Methods 2, 31−37. (4) Morin, I., Dixon, N. E., and Schaeffer, P. M. (2010) Ultrasensitive detection of antibodies using a new Tus-Ter-lock immunoPCR system. Mol. Biosyst. 6, 1173−1175. (5) Tang, Y. C., Wan, G., Ng, J. K., Ajikumar, P. K., and Too, H. P. (2008) DNA-directed assembly of protein microarrays. Front. Biosci. 13, 5755−5771. (6) Demidov, V. V., Dokholyan, N. V., Witte-Hoffmann, C., Chalasani, P., Yiu, H. W., Ding, F., Yu, Y., Cantor, C. R., and Broude, N. E. (2006) Fast complementation of split fluorescent protein triggered by DNA hybridization. Proc. Natl. Acad. Sci. U.S.A. 103, 2052−2056. 1354

dx.doi.org/10.1021/bc300118m | Bioconjugate Chem. 2012, 23, 1349−1355

Bioconjugate Chemistry

Technical Note

(26) Bock, L. C., Griffin, L. C., Latham, J. A., Vermaas, E. H., and Toole, J. J. (1992) Selection of single-stranded DNA molecules that bind and inhibit human thrombin. Nature 355, 564−566. (27) van Mansfeld, A. D., van Teeffelen, H. A., Baas, P. D., and Jansz, H. S. (1986) Two juxtaposed tyrosyl-OH groups participate in phi X174 gene A protein catalysed cleavage and ligation of DNA. Nucleic Acids Res. 14, 4229−4238. (28) Venkatesh, B., Arifuzzaman, M., Mori, H., Suzuki, S., Taguchi, T., and Ohmiya, Y. (2005) Use of GFP tags to monitor localization of different luciferases in E. coli. Photochem. Photobiol. Sci. 4, 740−743. (29) Baldrich, E., Restrepo, A., and O’Sullivan, C. K. (2004) Aptasensor development: elucidation of critical parameters for optimal aptamer performance. Anal. Chem. 76, 7053−7063. (30) Sekiguchi, J., and Shuman, S. (1994) Requirements for noncovalent binding of vaccinia topoisomerase I to duplex DNA. Nucleic Acids Res. 22, 5360−5365. (31) Nangreave, J., Han, D., Liu, Y., and Yan, H. (2010) DNA origami: a history and current perspective. Curr. Opin. Chem. Biol. 14, 608−615. (32) Rajendran, A., Endo, M., and Sugiyama, H. (2012) Singlemolecule analysis using DNA origami. Angew. Chem., Int. Ed. Engl. 51, 874−890. (33) Lin, C., Liu, Y., and Yan, H. (2009) Designer DNA nanoarchitectures. Biochemistry 48, 1663−1674. (34) Zhang, D. Y., and Seelig, G. (2011) Dynamic DNA nanotechnology using strand-displacement reactions. Nat. Chem. 3, 103−113.

1355

dx.doi.org/10.1021/bc300118m | Bioconjugate Chem. 2012, 23, 1349−1355