Versatile C-Terminal Specific Biotinylation of Proteins Using Both a

Jul 31, 2014 - A versatile puromycin-linker using cnvK for high-throughput in vitro selection by cDNA display. Yuki Mochizuki , Takeru Suzuki , Kenzo ...
0 downloads 0 Views 2MB Size
Download PDF
Technical Note pubs.acs.org/ac

Versatile C‑Terminal Specific Biotinylation of Proteins Using Both a Puromycin-Linker and a Cell-Free Translation System for Studying High-Throughput Protein−Molecule Interactions Naoto Nemoto,*,†,‡ Takayuki Fukushima,† Shigefumi Kumachi,† Miho Suzuki,† Koichi Nishigaki,† and Tai Kubo§ †

Graduate School of Science and Engineering, Saitama University, 255 Shimo-Okubo, Sakura-ku, Saitama 338-8570, Japan CREST, Japan Science and Technology Agency (JST), 5 Sanbancho, Chiyoda-ku, Tokyo 102-0075, Japan § Molecular Profiling Research Center for Drug Discovery, National Institute of Advanced Industrial Science and Technology (AIST), 2-4-7 Aomi, Koto-ku, Tokyo 135-0064, Japan ‡

ABSTRACT: Immobilization of a protein in a functionally active form and correct orientation for high-throughput analysis is crucial for surface-based protein−molecular interaction studies and should aid progress in associated nanotechnologies. Here, we present a general method for controlled and oriented immobilization of proteins by a puromycin-linker for cDNA display technology. The utility and potential of this method was demonstrated by examining the interaction between the B domain of protein A and immunoglobulin G (IgG) by surface plasmon resonance. This study revealed that the mRNA fragment of the mRNA−protein fusion (i.e., mRNA display) interferes with the interaction between the protein (B domain) and its target molecule (IgG). This results in a reduction of the apparent affinity by ∼10-fold. This method is expected to find wide appeal in the fields of surface-based studies of protein−protein interactions, drug screening, and single molecule analysis that require only a small amount of protein sample.

I

puromycin-linker.3,4 As the puromycin-linker links an mRNA and its coding protein covalently, protein modification with the puromycin-linker results in a protein modified at the Cterminus via the puromycin-linker with a label molecule, and thus, without direct chemical modification perturbing the function of the protein. In a previous study, the strategy was applied to a pull-down assay to examine protein−protein interactions rapidly and semiquantitatively using a puromycinlinker whose biotin attached to a streptavidin-magnetic bead.5 However, in the method presented herein the biotin molecule was used for immobilization of the protein fusion onto the sensor chip. Thus, a poly dA fragment was introduced into the puromycin-linker to purify a labeled protein under moderate conditions without reducing the protein function (Figure 1). The protein biotinylation using this novel puromycin-linker was applied to surface plasmon resonance (SPR) analysis, and the effect of the mRNA fragment toward the mRNA−protein fusion interaction with its target ligand was explored.

n the past decade, high-throughput screening of protein− protein interactions has been required for proteomics and directed evolution (especially antibody technology), and this method is particularly important to studies linked to various genome projects and drug discovery efforts. For this type of high-throughput process, apparatus and methods that operate with small sample amounts require rapid sample preparations of proteins from DNAs coding each gene simultaneously. The cell-free translation system is highly suitable for this purpose because protein synthesis occurs in a very short period when compared with cell-based production systems. Thus, efficient labeling and purification of the synthesized protein in a cell-free translation system is very important because it enables the preparation of a labeled sample required by particular highthroughput analysis apparatuses. The labeling procedure must also be rapid and cost-effective. In this study, an mRNA/cDNA display method for protein biotinylation at the C-terminus that specifically uses the cell-free translation system has been developed. The mRNA display involved the use of technology that links mRNA (genotype) and its coding protein (phenotype) via a puromycin-linker using a cell-free translation system for directed evolution.1,2 Recently, we have developed cDNA display to improve the stability and the efficient productivity of the mRNA display process using a novel © XXXX American Chemical Society

Received: December 19, 2013 Accepted: July 31, 2014

A

dx.doi.org/10.1021/ac501601g | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Technical Note

same molar concentration under annealing conditions (heating to 90 °C for 2 min followed by incubation at 70 °C for 1 min and then cooling to 4 °C) in ligation buffer (50 mM Tris-HCl pH 7.5, containing 10 mM MgCl2, 10 mM DTT, and 1 mM ATP). Following the addition of T4 RNA ligase (6 U/pmol mRNA, Takara, Japan) and T4 polynucleotide kinase (2 U/ pmol mRNA, Takara), the ligation reactions were performed at 25 °C for 30 min. The ligation products were analyzed using 8 M urea denaturing 8% PAGE. The ligated products were visualized by FITC fluorescence using the fluoroimager. The final products were purified with a RNA purification column kit (RNeasy MinElute Cleanup Kit, Qiagen, Hilden, Germany). Cell-Free Translation. The mRNA-linker conjugates (3−5 pmol) were added to 50 μL of the cell-free translation system (Promega), and the mixture was incubated at 30 °C for 30 min. To increase the yield of mRNA−protein fusions, the posttranslation fusion reaction was incubated at 37 °C for an additional 60 min in the presence of a high concentration of salts (KCl and MgCl2 at final concentrations of 800 and 80 mM, respectively). The products (mRNA-linker-protein) were resolved by 8 M urea containing 6% SDS−PAGE and detected using FITC fluorescence with the fluoroimager. Purification of the mRNA-Protein Fusions by Oligo-dT Beads. Oligo-dT magnetic beads [Dynabeads Oligo (dT), Life Technologies] were washed twice with the washing solution (10 mM Tris-HCl pH 7.5, 150 mM LiCl, and 1 mM EDTA) according to the manufacturer’s instructions. The washed oligodT beads (1.0 mg) and binding buffer (20 mM Tris-HCl, pH 7.5, 1 M LiCl, 2 mM EDTA) were added to the mixture containing the mRNA−protein fusions (50−100 pmol) and incubated for 60 min at room temperature using the thermo block rotator (NISSIN, Japan). After the incubation, the mixture was kept on ice for 5 min. The washing buffer was added and mixed to the mixture in equal volume. The microcentrifuge tube containing the mixture was attached to a magnetic stand and left on for 3 min. After the supernatant in the tube was removed, the tube was taken off the stand. After the oligo-dT beads were washed once with the binding buffer, 100−150 μL of DEPC-treated water was added and mixed. After the mixture was incubated at room temperature for 5 min, the supernatant containing the mRNA−protein fusions was collected from the tube attached to the magnetic stand and stored in a clean microcentrifuge tube. His-Tag Purification of the mRNA-Protein Fusions by Ni-NTA Beads. The 20 μL of Ni-NTA beads (His Mag Sepharose Ni, GE Healthcare Bio-Sciences, Piscataway, NJ) were washed twice with the His-tag-washing buffer (20 mM sodium phosphate, pH 7.4, 500 mM NaCl, 5 mM imidazole, 0.05% Tween-20). The washed Ni-NTA beads and 5× His-tagbinding buffer (100 mM sodium phosphate, pH 7.4, 2 M NaCl, 20 mM imidazole, 0.25% Tween-20) were added to the mixture containing the mRNA−protein fusions (6 pmol) and incubated for 60 min at room temperature using the thermo block rotator (NISSIN, Japan). After the Ni-NTA beads were washed twice with the His-tag-washing buffer, the elution buffer (20 mM sodium phosphate, pH 7.4, 500 mM NaCl, 250 mM imidazole, 0.05% Tween-20) was added and mixed. After the mixture was incubated for 20 min at room temperature, the supernatant containing the mRNA−protein fusions was collected from the tube attached to the magnetic stand and stored in a clean microcentrifuge tube. Immobilization of the mRNA-Protein Fusions to the Sensor Chip. All SPR experiments were carried out on a

Figure 1. Schematic diagram of the C-terminal specific biotinylation and immobilization of proteins using the puromycin-linker. The puromycin-linker containing both a covalently attached oligo-dA and biotin is ligated with an mRNA. After the ligated mRNA is translated by the cell-free translation system, the protein-puromycin-mRNAlinker conjugate is purified with oligo-dT magnetic beads. The purified biotinylated protein conjugate with or without mRNA degradation is immobilized onto the streptavidin-coated surface.

■

EXPERIMENTAL SECTION Chemicals and Reagents. The modified oligonucleotides, “Psorelen fragment (PF)” and “Poly-dA fragment (PAF)” were obtained from Geneworld (Tokyo, Japan). The PF had the following composition: 5′-(Ps)TACGACGATCTCGAACGAACCACCCCCCCCGCCGCCCCCC GTCC(F)(Z)(Z)(Z)CC(P)-3′, where (Ps) = 5′psoralen-C6, (F) = fluorescein-dT, (Z) = spacer phosphoramidite 18 (Glen Research, Sterling, VA), and (P) = puromycin CPG. The PAF had the following composition: 5′-CCCGTGGTTCGTTCGAGATCGTCGTAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA(B)-3′, where (B) = 3′biotin TEG. All other regents were of analytical grade. Vectors, Oligonucleotides, and Library Construction. The B domain of protein A was obtained from the pEZZ 18 protein A gene fusion vector (GE Healthcare, Piscataway, NJ). The forward primer incorporated a T7 promoter, the tobacco mosaic virus “omega” 50-untranslated region, a Kozak sequence, and an ATG start codon. The reverse primer harbored a hexa-histidine tag, a spacer sequence (GGGGGAGGCAGC), and a complementary sequence (AGGACGGGGGGCGGGGAAA) for the purification-linker DNA at the 3′terminus to enable ligation between the mRNA and the purification-linker DNA. In the case of the FLAG-tag (which fused the helix IV of Pou-specific DNA-binding domain of Oct1) and mouse FAS ligand, the template was generated by replacing the B domain region with the FLAG-tag and the Fas ligand. Synthesis of the Purification-Linker DNA. Cross-linking of the psoralen reaction with ultraviolet (UV) irradiation is a modification of the method of Sinden and Hagerman.6 Briefly, a total of 0.5 nmol of PF and 2 nmol of PAF were added to 50 μL of 25 mM Tris-HCl buffer (pH 6.8) with 100 mM NaCl, and the mixture was annealed from 90 to 25 °C gradually for 1 h with a thermal cycler and irradiated with ultraviolet rays (main peak λ = 365 nm) for 15 min. The product was analyzed by 8 M urea denaturing 8% polyacrylamide gel electrophoresis (PAGE) and detected using FITC fluorescence using a fluoroimager (Pharos Fx, BioRad). Ligation of mRNA to the Purification-Linker DNA. The 3′-ends of mRNA molecules were hybridized to the complementary strands of the purification-linker DNAs at the B

dx.doi.org/10.1021/ac501601g | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Technical Note

Biacore J system (GE Healthcare) at 25 °C. To remove nonspecific absorbed streptavidin on the sensor chip, the washing buffer (1 M NaCl in 50 mM NaOH) was loaded into the Fc1 and Fc2 channels of a Sensor Chip SA for 1 min three times. The mRNA-protein fusions (28 pmol) in 100 mM NaCl (200 μL) were injected into the Fc1 channel and loaded with the running buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 3 mM EDTA, and 0.005% (w/v) surfactant P20) for 12 min at a low flow rate to immobilize the mRNA−protein fusions onto the sensor chip. Molecular Interactions Analysis. Analytes at appropriate ranges of concentrations (2.5, 5, 10, and 20 μg/mL) in PBS buffer were injected at a flow rate of 5 μL/min into Fc1 and Fc2 channels to measure the affinity. Data from 80 to 250 s of association and 250 s of dissociation were collected. The sensorgram of the assay channel (Fc2) was double-subtracted by the buffer control and the Fc1 channel (control), and then overlaid for kinetic fitting to obtain the binding on (ka) and off (kd) rates and affinity (KD = kd/ka). The kinetic fitting was carried out with the Biacore J evaluation software using a 1:1 Langmuir binding model (A + B = AB).7 All experiments were repeated three times, and the average KD with standard deviation (SD) is shown.

linkers that were used in in vitro selection of a peptide aptamer.3 The synthesis of the puromycin-linker has been shown to be very laborious and high-performance liquid chromatography (HPLC) purification has been indispensable for the separation of the final product from other byproducts.3,4 In our method presented herein, there is no purification step following the synthesis of the final linker after UV irradiation of this linker. This will enable researchers who are not familiar with cDNA display to apply this method easily to various kinds of molecular interactions. Rapid Purification of Protein Synthesized with a CellFree Translation System. The cell-free translated products (mRNA−protein fusion) were purified by an oligo-dT column. Each band of the mRNA−protein fusion and the linker mRNA could be identified before and after purification on the gel by FITC imaging (Figure 3a, left). However, these bands could not be identified when the gel was stained with SYPRO Ruby because the amount of product was very low in comparison with other endogenous proteins in the rabbit reticulocyte lysate (Figure 3a, right). After the purification, the endogenous proteins derived from the rabbit reticulocyte were essentially removed, whereas the synthesized proteins remained. This purity is very important for high-sensitivity molecular interaction analysis [e.g., SPR and Quartz Crystal Microbalance (QCM)] to decrease nonspecific absorption. By way of comparison, the mRNA−protein fusions were purified by a Ni-NTA column and analyzed before and after purification on a gel by FITC imaging and an oligo-dT column (Figure 3a, left). Although the band representing the mRNA−protein fusion was easily identified before purification, no bands appeared on the gel after purification. Similarly, any bands without the endogenous proteins derived from the rabbit reticulocyte were detected on the gel by SYPRO Ruby staining (Figure 3a, right). These results show that the puromycin-linker with an oligo-dA region is very useful for the purification of the mRNA−protein fusion from a cell-free translation system. In a previous study, we developed a puromycin-linker containing a biotin molecule to perform a pull-down method using a cell-free translation system.5 Although this puromycinlinker is capable of immobilizing a cell-free synthesized protein with a streptavidin-coated surface, many other endogenous molecules without mRNA−protein fusions could easily clog the microfluid channel in the sensor chip of the SPR unit because endogenous contaminants could not be removed from the cellfree translation system efficiently. By the novel puromycinlinker used in this study, the final efficiency of the mRNA− protein fusion synthesis from the initial amount of input mRNA was ∼20%; although the mRNA−protein fusion efficiency was around 30% before oligo-dT purification (Figure 3b). Regardless of whether it is the mRNA-linker only or the mRNA−protein fusion, the purification efficiency was ∼60% because it depends on the hybridization efficiency between the oligo-dT and oligo-dA region of the linker. Although the amount of purified mRNA-protein fusion is very low, as confirmed by the absence of a band on the gel following SYPRO staining, the amount is still sufficient to analyze molecular interactions by SPR. In general, proteins and peptides syntheses are time-consuming and laborious, regardless of whether they are performed by cell-based expression (e.g., protein expression in E. coli and purification) or chemical synthesis (e.g., peptide solid-phase synthesis). In the previous study, a combination of His-tag purification and a His-tag chip were applied for high-throughput preparation of a biochip using

■

RESULTS AND DISCUSSION Synthesis of Purification-Linker and its Ligation Performance. After the psoralen fragment and the poly dA fragment were hybridized at a ratio of 1:4, the fragments were cross-linked by the psoralen reaction with ultraviolet (UV) irradiation (Figure 2a). The cross-linking efficiency was around 80% (Figure 2b). As an example, the synthesized linker was ligated with the mRNA of the B domain of protein A (BDA) and T4 RNA ligase. The efficiency was estimated to be ∼80% by fluorescence intensity analysis of the band on the gel. This efficiency is almost the same as previously reported puromycin-

Figure 2. Schematic depiction and synthesis of the puromycin-linker. (a) Specific construction of the puromycin-linker. After the psoralenfragment and the poly dA-fragment were hybridized, the two DNA fragments are covalently linked by UV irradiation; PEG (polyethylene glycol). (b) Linkage efficiency of the two DNA fragments by UV radiation. The linkage products were analyzed by 8 M urea PAGE. C

dx.doi.org/10.1021/ac501601g | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Technical Note

Figure 3. Synthesis and purification of the mRNA-protein fusion with biotin at the C-terminus. (a) Analysis image of the mRNA−protein fusion after translation with or without purification combined with Oligo-dT beads or Ni-NTA beads by PAGE. In the case of Ni-NTA beads, the puromycin-linker lacking of Oligo dA region (i.e., a short linker) was applied. The red arrows indicate each protein-mRNA fusion. (b) Efficiency of the mRNA−protein fusion synthesis and purification efficiency of the mRNA−protein fusion by the oligo-dT column.

a cell-free translation system. The protein array termed “protein in situ array” (PISA) is rapidly generated in one step directly from PCR-generated DNA fragments by a cell-free protein expression and in situ immobilization onto Ni-NTA-coated wells.8 However, this method has not been applied to protein immobilization on a Ni-NTA surface for SPR analysis because of the limitation of the affinity of the His-tag and doublehexahistidine tag.9 Although the his-tag immobilization strategy for SPR using some types of His-tag derivatives were studied intensively,10 these His-tag fused proteins were not prepared with a cell-free translation system suitable for a high-throughput procedure. In this method, the biotin−streptavidin interaction as a high-affinity tag was used for protein immobilization on the chip surface instead of a His-tag. Thus, because only a small amount of protein material is needed for SPR, rapid preparation of the puromycin-linker attached to biotin using a cell-free translation system in several hours can be achieved and is a suitable protein preparation approach for use in SPR studies. In this method, although the linker-mRNA fusions without protein remained at a similar abundance to the mRNA− protein fusions in the purified sample, it is clear that the effect against the sensorgram analysis of SPR is almost negligible, as shown in Figure 4. This shows that the linker-mRNA fusions might not interfere with the interactions between the mRNA− protein fusions and the target molecule during the SPR measurement. Effect of the mRNA Fragment on Protein-Molecular Interactions. In this mRNA display method, biotinylation of a protein involved linking an mRNA molecule with its coding protein via a puromycin-linker that contained biotin. In general, an mRNA molecule is larger than its coding protein in

molecular weight. Furthermore, as mRNA is a polymer, the mRNA may interfere with the molecular interaction between its coding protein and the target molecule. To examine this effect, the mRNA display molecule (mRNA−protein fusion) was directly immobilized onto the surface of a streptavidin-coated SPR sensor chip. In addition, a RNase-treated mRNA display molecule (protein-linker only) was also immobilized on a streptavidin-coated chip. The mRNA was encoded by the B domain of protein A (BDA), which interacts with immunoglobulin G (IgG) (Figure 4a). Binding curves were obtained with a BIAcore streptavidin sensor chip. Figure 4b shows the binding and dissociation curves of IgG at different concentrations. As a result of the SPR analysis, we found that the response of the sensorgrams on the mRNA display molecule was less than that observed on the RNase-treated mRNA display molecule. As results of the above data analysis, the determined dissociation constants (Kd) between immunoglobulin G (IgG) and the B domain of protein A, which was displayed by either the mRNA display molecule or the RNase-treated mRNA display molecule, were 190 and 17 nM, respectively (Figure 4b). In the case of the mRNA display, the Kd value was larger than the value reported previously. On the other hand, in the case of the RNase-treated mRNA display, the Kd value was consistent with the previous study. This was the first example showing that the mRNA fragment part of the mRNA display interfered with the interaction between the protein part and its target molecule. An mRNA can move flexibly in the liquid phase. Consequently, the flexible motion of the mRNA fragment of the mRNA display may reduce the likelihood that the target molecule interacts with its coded protein on the mRNA display. This is most likely because the mRNA fragment is located near its coding protein D

dx.doi.org/10.1021/ac501601g | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Technical Note

Figure 4. Effect of mRNA on the mRNA-protein fusion against molecular interactions. (a) Schematic diagram of the interaction between analyte (IgG) and the B domain of protein A (BDA) with or without its coding mRNA. (b) Baseline-corrected sensorgrams of BDA against IgG. The sensorgrams of BDA-mRNA (left) and BDA only with RNase-treatment (right) are presented for the various concentrations of IgG examined. The affinities between BDA and IgG from the analysis of the sensorgrams are indicated in the frames. (c) Schematic diagram of the interaction between Fas ligand and anti-Fas mAb (left). Baseline-corrected sensorgrams of the Fas ligand against anti-Fas mAb (right). The Fas ligand was biotinylated and immobilized onto the sensor chip without mRNA. The anti-Fas mAb (analyte) was injected at various concentrations. The affinity from the analysis of the sensorgram is provided within the frame.

as the Fas ligand/anti-Fas mAb interaction, as well as standard ligand−Ab interactions, can be measured by this method. With the rapid progress of genome projects, many approaches that analyze protein−molecular interactions have been required. Proteomics tools have improved with the advent of nanotechnology with reduction in assay times and required sample volumes representing two major benefits over conventional methods.12 However, the conventional protein and peptide sample preparation techniques have not necessarily been convenient and inexpensive for the emerging proteomics tools that use a small amount of sample. Furthermore, protein modification methods (e.g., biotinylation) generally require complicated and cumbersome chemical methods. Although cell-free translation systems are useful for the production of

and therefore inhibits the interaction with the target molecule. To confirm this method, an analysis of the interaction between the Fas ligand and anti-Fas ligand monoclonal antibody (antiFas mAb) was performed according to the above RNase-treated method. The determined Kd value was 170 nM. As expected, the dissociation constant was consistent with a previous study.5 Additionally, as a positive control, analysis of the interaction between a FLAG peptide and an anti-FLAG monoclonal antibody was performed according to the above RNase-treated method. The determined Kd value was 38 nM (data not shown), which was similar to that determined in a previous study (Kd = 15 nM).11 These experiments showed that weak (∼micromolar) molecular interactions between proteins, such E

dx.doi.org/10.1021/ac501601g | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Technical Note

immobilization of proteins at the C-terminus through the puromycin−biotin spacer in a high-throughput manner. The present method is anticipated to find wide applications in the fields of surface-based studies of protein−protein interactions (e.g., the 2D-SPR imaging system), single molecule analysis, biochip production, and drug screening.

small amounts of proteins, it is not generally suitable for preparation of labeled proteins because of its poor yield and the difficulty of purifying the labeled protein.13 In this paper, the puromycin-linker labeling method overcomes these challenges because the biotinylated puromycin-linker conjugated protein can be purified from the cell-free translation system efficiently by an oligo-dT column purification step, as well as using the benefits of a poly-A-tailed mRNA, as mentioned in the previous section. Thus, in this SPR measurement, the time for protein− molecular interaction analysis was dramatically shorter when compared with the available conventional sample preparation methods. Although SPR has historically been limited by its throughput, the simultaneous analysis of many thousands of interactions has been possible by protein array technologies and nanotechnologies.14,15 Recently, the two-dimensional surface plasmon resonance (2D-SPR) imaging system16 has been developed. Thus, the development of an easy-to-use protein biotinylation process for immobilization of proteins onto chip surfaces is important. In particular, protein biotinylation using a cell-free translation system is suitable for rapid preparation with a small amount of protein. For this purpose, a position-specific incorporation method with a biotinylated non-natural amino acid using a cell-free translation system has been developed.17 However, the biotin moiety incorporated into the synthesized protein can interfere with molecular interactions because the biotin exists on the protein surface. On the other hand, there can be negligible interference between a protein synthesized by our method and the target protein because the biotin is linked to the protein via a spacer. In addition, special reagents (e.g., biotinylated non-natural amino acid) and the redesign of the DNA template are not required in our method. Thus, our method is easy to use and cost-effective when compared with previous methods. Recently, the quartz crystal microbalance (QCM) technology that requires only a small amount of sample for measurement has also been improved by the introduction of a multichannel, thereby enabling high-throughput analysis.18 Thus, we believe that the combined use of a biotinylated protein preparation method with a cell-free translation system such as this method will contribute to high-throughput molecular interaction analyses using microarray technologies and nanofabricated devices, as well as SPR and QCM. Adaptation of this method to various types of protein should be systematically assessed. Relatively low molecular weight proteins (up to ∼200 amino acid residues) should be easily analyzed by this method; however, for proteins with higher molecular weight or with a high content of disulfide bridges the cell-free translation system may need to be optimized.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +81-48-858-3531. Fax: +81-48-858-3531. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank Prof. Wang Qingyue (Saitama University) and Ms. Suzuko Kobayashi (AIST) for his/her technical assistance with SPR measurements (Biacore J). This research was supported by JST, CREST.

■

REFERENCES

(1) Nemoto, N.; Miyamoto-Sato, E.; Husimi, Y.; Yanagawa, H. FEBS Lett. 1997, 414, 405−408. (2) Roberts, R. W.; Szostak, J. W. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 12297−12302. (3) Yamaguchi, J.; Naimuddin, M.; Biyani, M.; Sasaki, T.; Machida, M.; Kubo, T.; Funatsu, T.; Husimi, Y.; Nemoto, N. Nucleic Acids Res. 2009, 37, e108. (4) Mochizuki, Y.; Biyani, M.; Tsuji-Ueno, S.; Suzuki, M.; Nishigaki, K.; Husimi, Y.; Nemoto, N. ACS Comb. Sci. 2011, 13, 478−485. (5) Mochizuki, Y.; Kohno, F.; Nishigaki, K.; Nemoto, N. Anal. Biochem. 2013, 434, 93−95. (6) Sinden, R. R.; Hagerman, P. J. Biochemistry 1984, 23, 6299−6303. (7) Khan, S. H.; Farkas, K.; Kumar, R.; Ling, J. Anal. Biochem. 2012, 421, 385−390. (8) He, M.; Taussig, M. J. Nucleic Acids Res. 2001, 29, e73. (9) Khan, F.; He, M.; Taussig, M. J. Anal. Chem. 2006, 78, 3072− 3079. (10) Fischer, M.; Leech, A. P.; Hubbard, R. E. Anal. Chem. 2011, 83, 1800−1807. (11) Wegner, G. J.; Lee, H. J.; Corn, R. M. Anal. Chem. 2002, 74, 5161−5168. (12) Gonzalez-Gonzalez, M.; Jara-Acevedo, R.; Matarraz, S.; JaraAcevedo, M.; Paradinas, S.; Sayagües, J. M.; Orfao, A.; Fuentes, M. Eur. J. Pharm. Sci. 2012, 45, 499−506. (13) Lue, R. Y.; Chen, G. Y.; Hu, Y.; Zhu, Q.; Yao, S. Q. J. Am. Chem. Soc. 2004, 126, 1055−1062. (14) Boozer, C.; Kim, G.; Cong, S.; Guan, H.; Londergan, T. Curr. Opin. Biotechnol. 2006, 17, 400−405. (15) Campbell, C. T.; Kim, G. Biomaterials 2007, 28, 2380−2392. (16) Pyo, H. B.; Shin, Y. B.; Kim, M. G.; Yoon, H. C. Langmuir 2005, 21, 166−171. (17) Watanabe, T.; Muranaka, N.; Iijima, I.; Hohsaka, T. Biochem. Biophys. Res. Commun. 2007, 361, 794−799. (18) Ogi, H.; Nagai, H.; Fukunishi, Y.; Yanagida, T.; Hirao, M.; Nishiyama, M. Anal. Chem. 2010, 82, 3957−3962.

■

CONCLUSIONS We have demonstrated a general method for protein biotinylation at the C-terminus for high-throughput use via linkage of a spacer containing both puromycin and biotin molecules during cell-free translation synthesis. Kinetic analysis of ligand binding by SPR revealed that the presence of the mRNA anchored with its coding protein through the puromycin-linker gives rise to a ∼10-fold lower binding affinity than the value measured with only the coding protein. This showed for the first time that the mRNA part of an mRNA− protein fusion construct can interfere with the interaction between its coding protein and the target molecule. Our results clearly underline the efficacy of the controlled and oriented F

dx.doi.org/10.1021/ac501601g | Anal. Chem. XXXX, XXX, XXX−XXX