Development of a Single Chain Antibody Using a Phage Display Cloning Method for the Detection of 2,4-Dinitrotoluene
Jung-Hyun Na, Man-Seok Joo, Won-Kyu Lee, Hyunbo Shim,† Si-Hyung Lim,‡ Sang Taek Jung, and Yeon Gyu Yu*
Department of Chemistry, Kookmin University, Seoul 136-702, Korea. *E-mail: ygyu@kookmin.ac.kr
†Department of Life Science, Ewha Womans University, Seoul 120-750, Korea
‡School of Mechanical Engineering, Kookmin University, Seoul 136-702, Korea Received October 31, 2012, Accepted November 15, 2012
Single-chain variable fragments of antibodies (scFv) specific to 2,4-dinitrotoluene (DNT) were isolated from a phage library displaying synthetic human scFv fragments with 6 diversified complementary determining regions (CDRs). A DNT derivative that contained an extended amine group was synthesized and conjugated to the NHS-group that was linked to magnetic beads. Phages specific to the immobilized DNT derivatives were isolated from the library after 4 rounds of sequential binding and elution processes. The displayed scFv fragments from the isolated phages showed consensus CDR sequences. One DNT-specific scFv was expressed in E. coli and purified using Ni-affinity chromatography. The purified DNT-specific scFv binds specifically to the immobilized DNT-derivative with KD value of 6.0× 10−7 M. The scFv and DNT interaction was not disrupted by the addition of 4-nitrotoluene or benzoic acid. These data demonstrate that the screened scFv from the phage displayed library could be used for selective and sensitive detection of explosives such as TNT.
Key Words : DNT, Sensor, Single chain antibody, Biopanning, Phage display
Introduction
Effective detection of 2,4,6-trinitrotoluene (TNT) is requir- ed for public safety and health because it is the most widely used explosive that produces toxic effect for humans.1,2 Currently, gas chromatography coupled with a mass spectro- meter, HPLC, or other analytical device is used for nitro- aromatic explosive compound detection.3,4 However, these methods involve time-consuming extraction steps and require a laboratory facility, which limits rapid on-site detection. For the development of an explosives detection system, 2,4- dinitrotoluene (DNT) has been considered a target molecule for recognition elements because it is a major impurity of military grade TNT as well as has higher vapor pressure and stability compared to TNT.5,6
Various explosives detection methods have been develop- ed using DNT or TNT-specific biopolymers, such as pep- tides or antibodies. Short peptides specific to DNT have been isolated from a phage display screen using immobiliz- ed DNT as bait, and these peptides showed specific binding to DNT in solution as well as in the gas phase.7,8 Polyclonal antibodies raised against DNT-coupled KLH as the antigen could be used for DNT detection in solution by ELISA or surface plasmon resonance methods.9,10 These DNT-specific polyclonal antibodies showed selectivity against other compounds containing nitro groups, and had dissociation constants of 0.12-3.7× 10−7 M,10 which are slightly lower than the KD values of DNT-specific peptides.8 The higher affinity of antibody to peptide is expected because a 3- dimensional pocket structure consisting of at least 6 CDRs would provide more specificity and affinity than the binding moieties of peptides, which consist of linearly located amino
acids. However, polyclonal antibodies are assumed to have some limitations in sensing material due to their lower stabilities compared to synthetic polymer or peptides. In addition, using genetic manipulation to enhance the sensi- tivity and selectivity of polyclonal antibodies is restricted due to the complication of obtaining the cDNAs of specific polyclonal antibodies.
The antigen recognition domain of an antibody consists of diverse variable fragments of heavy and light chains that could be constructed as a single chain polypeptide (scFv).11 The scFv represents the antigen-binding domain of the antibody and consists of 6 variable heavy and light chain CDRs. The 6 CDRs of scFv have been constructed as random sequences, and the resulting diversified scFvs were fused at the N-terminus of the pIII protein of M13 phages resulting in scFv-displaying library with a diversity of 7.6× 109.12 This library has been successfully used to identify scFvs that specifically bind to target proteins or chemical compounds using a biopanning method. Because scFv is much smaller than immunoglobulin and lacks a flexible loop, scFv is considered much more stable than an intact antibody. Furthermore, the selectivity and affinity of the selected scFv to target molecules can be significantly en- hanced by genetic engineering using the cDNA representing the scFv.13
In this study, we have isolated M13 phages from the M13 scFv-displaying library using a DNT derivative immobilized on magnetic beads. The cDNA representing the scFv region was cloned, expressed in E. coli, and purified. The binding properties of the purified scFv to DNT were determined. The identified scFv or its genetic variants could be used as sensing material for TNT-based explosives.
Materials and Methods
Materials. NHS-activated magnetic beads were obtained from Bioclone, Inc. (San Diego, CA). The M13 phage library displaying scFv with randomized amino acid sequences at 6 CDR regions was prepared by the previously described method.12 We synthesized 4-(2,4-dinitrophenyl)butan-1- amine from benzyl bromide by the previously described method.8 The Q-Sense His-Tag capturing QCM-D sensor (QSZ340) was purchased from Biolin Scientic AB (Sweden).
All other reagents used were of reagent grade.
scFv-Displayed M13 Library Biopanning Using DNT Derivative on Magnetic Beads. Magnetic beads (1 mg/mL) linked with the NHS-group were conjugated with (1 4-(2,4- dinitrophenyl)butan-1-amine (1 mM) by incubating for 2 h at 25oC in reaction buffer (10 mM HEPES, pH 7.9). The conjugation reaction was terminated by incubating for 1 h in 100 mM glycine. The covalently modified 4-(2,4-dinitro- phenyl)butan-1-amine magnetic beads were recovered using a magnet and were washed with buffer A (50 mM Tris-HCl, 150 mM NaCl, pH 7.4). Approximately 0.1 mg of conju- gated beads were incubated for 1 h at room temperature with 500 μL of M13 scFv phage display library (1.5 × 1013 pfu/
mL) in buffer A, and then washed 5 times with buffer A. The bound phages were eluted after incubating for 10 min with 0.1 mL of triethylamine and neutralized by adding 0.1 mL of 1 M Tris-HCl (pH 7.4). The eluted phages were applied again to the 4-(2,4-dinitrophenyl)butan-1-amine conjugated magnetic beads for a second round of biopanning, and the bound phages were eluted as described above. The 3rd and 4th biopanning rounds were performed similarly, except buffer A contained 0.2% Tween 20. To determine the titer of the eluted phages, 1 μL of eluted phage solution was mixed with 1 mL of mid-log phase ER2537 E. coli cells for 30 min at 37oC, and then plated on LB plates containing carbeni- cillin. The phage DNAs from ER2537 colonies harboring specific M13 phages were purified using a DNA isolation kit (Qiagen, USA). The sequences of the isolated phage DNAs were determined, and the displayed peptide sequences were
deduced from the DNA sequence.
scFv Expression and Purification. The scFv protein from the selected M13 phage was prepared as described previously.12 Briefly, a single ER2537 colony harboring a scFv clone in the pComb3X vector was inoculated and grown at 37 °C in 1 L of LB carbenicillin media, until the absorbance at 600 nm was 0.5. scFv protein expression was induced by adding 1 mM IPTG, and the cultures were allow- ed to grow overnight at 30 °C. After harvesting the cells by centrifugation, the cell pellet was resuspended in 10 mL of ice-cold TES buffer (50 mM Tris, 1 mM EDTA, 20% sucrose, pH 8.0). Cold 0.2 × TES buffer (15 mL) was subsequently added. After incubation for 30 min, the suspension was centrifuged and the supernatant containing the periplasmic fraction was loaded onto Ni-NTA agarose columns (Novagen).
The scFv protein was eluted from the column using 200 mM imidazole in equilibration buffer (50 mM Tris-HCl, pH 7.4).
Quartz Crystal Microbalance with Dissipation (QCM- D) analyses. The specific binding of DNT to isolated scFv was examined using QCM-D. Purified His-tag labeled scFv (1.5 mg/mL, 10 mM MES, pH 6.5) was immobilized onto the His-Tag capturing QCM-D sensor (QSZ340) using a 50 µL/min flow rate. After washing the unbound scFv using 10 mM MES (pH 6.5), 1 mM DNT in the same buffer was applied at a ow rate of 30 μL/min, and the resonance frequency change was monitored at 25 °C using a Q-Sense E4 system (Biolin Scientic AB, Västra Frölunda, Sweden).
Affinity Determination. The affinity of purified scFv to DNT was analyzed by measuring the amount of bound scFv compared to immobilized biotinylated DNT derivative. NHS- biotin (1 mM in 100 mM HEPES, pH 7.5; Pierce, USA) was reacted with 4-(2,4-dinitrophenyl)butan-1-amine (5.0 mM) for 3 h at room temperature. The reaction was terminated by adding 20 mM glycine to inactivate any residual NHS-bio- tin. The reaction product was added to streptavidin-coated 96-well plates. After incubation for 1 h, the plate was wash- ed and incubated with different concentrations of scFv solution. After washing, the wells were incubated for 1 h at room temperature with the HRP-labeled His-tag antibody.
Figure 1. Schematic representation of the immobilization of 4-(2,4-dinitrophenyl)butan-1-amine on the NHS-coated magnetic beads and the biopanning selection process of scFv-displayed M13 phages.
The color development reaction was performed using TMB solution (Sigma, USA), and the absorbance of each well was measured at 450 nm using a TRIAD microplate reader (Dynex Technologies, VA, USA).
Results and Discussion
DNT Derivative Immobilization on Magnetic Beads and Biopanning of Phage Display Library. 4-(2,4-Dinitro- phenyl)butan-1-amine (Fig. 1), a derivative of DNT, was prepared as previously described.8 The amine group of the DNT derivative was covalently attached to the NHS group linked to magnetic beads (Fig. 1). The DNT-attached mag- netic beads were used for biopanning of scFv-displayed M13 phage library. The biopanning process consisted of binding, washing, and elution steps that were repeated for 4 rounds. During this process, eluted phage amplification was omitted to minimize uneven distribution of the eluted phages due to different growth rates of phages.14 The phage titer before and after each round of biopanning decreased from 1.0 × 1013 cfu/ml to 0.2 × 109 cfu/mL (Fig. 2(a)). The ratio of input phages to eluted phages increased gradually after each round and was saturated after the 3rd round of biopanning (Fig. 2(b)), indicating that the fraction of DNT-specific phages was enriched during the biopanning process. After the 4th round of biopanning, 8 clones of the M13 phages were selected, and their DNA was sequenced. Among them, 3 clones (DNT-scFv-2), and 2 clones (DNT-scFv-3) showed identical sequences. Rest of them (DNT-scFv-1, DNT-scFv- 4 and DNT-scFv-5) showed unique sequences (Table 1). The presence of degenerated sequences indicated that these phages displaying DNT-specific scFvs were selectively isolated.
scFv Protein Expression and Purification from the Identified Phage Clones. To prepare the scFv proteins from the identified M13 phage clones, we examined the expre- ssion level of scFvs. The 5 different scFv clones (DNT-scFv- 1-5) were expressed in E. coli, and their expression was measured by western blot analyses using the His-tag anti- body that recognizes the 6-histidine sequence at the C- terminus of the scFv protein. Among the 5 clones, only 2 clones expressed substantial levels of scFv (Fig. 3(a)). The poor expression of the 3 clones is likely due to the early termination at the amber codons that occur 1-2 times in the reading frame. The 2 scFC clones (DNT-scFv-2 and DNT- scFv-3) that showed high expression levels were further purified using IMAC columns. As shown in Figure 3(b),
proteins efficiently purified to > 90% homogeneity. When the purified proteins were analyzed by size-exclusion chromatography, they appeared to be monomers (data not shown).
Characterization of Purified scFvs Binding to DNT. To examine the sensitivity of the purified scFvs to DNT, we determined the binding affinity of the purified scFv proteins to DNT-biotin immobilized on streptavidin-coated plates.
The amount of bound scFvs on the DNT immobilized plate was measured using His tag and HRP-labeled anti-mouse antibodies. As shown in Figure 4(a), one of the purified scFv proteins, DNT-scFv-3, exhibited a concentration-dependent binding curve. The apparent dissociation constant of scFv to immobilized DNT was calculated as 6.0 × 10−7 M. The other scFv protein, DNT-scFv-2, failed to show specific binding to immobilized DNT-biotin (data not shown). Therefore, DNT- scFv-3 was further characterized.
The selectivity of DNT-scFv-3 against DNT over other Figure 2. Phage titer after each round of biopanning. (a) The titer of eluted M13 phages of each round of biopanning. (b) The eluted phage titer percentage compared to the input titer from each round of biopanning.
Table 1. Amino acid sequences of the CDRs of isolated scFv clones
Clone (frequency) H1 H2 H3 L1 L2 L3
DNT-scFv-1 (1) NySms GiSPDGGSI KGLTQIIESQFDY TgsssnigNnTvS DNNK AAwdDslSA
DNT-scFv-2 (2) DyYms ViSHDGSST KGALDQQVAAASSAYAMDV SgsssnigNnNvY ANSH GAwdDslSG
DNT-scFv-3 (3) DyAms AiSYDGGSK RDAVTPTSTTTSYYNAMDV TgsssnigSnSvT DDNQ AAwdAslSG
DNT-scFv-4 (1) GyYms GiYPDGGSI KGRVKHINPFCSSADAMDV TgsssnigNnSvY SNSQ ASwdDslSG
DNT-scFv-5 (1) DyYms GiYSDGSRK RGRKRYRRAFDY TgsssnigNnAvN SNSH GTwdDslNG
The amino acid sequences of the 6 CDRs (H1, H2, H3, L1, L2, and L3) deduced from the 8 clones are represented as single letters. The randomized amino acid positions are indicated in capital lette
organic molecules containing a benzene ring was also ex- amined. The amount of bound DNT-scFv-3 on the DNT- immobilized plate was measured in the presence of DNT, TNT, toluene, 2-nitrotoluene, or benzene. In the presence of
5 μM of DNT or TNT, DNT-scFv-3 binding was completely blocked, indicating that both DNT and TNT efficiently bind Figure 3. SDS-PAGE analyses of scFvs. (a) The scFv proteins from E. coli extracts expressing 5 unique scFv clones from the isolated M13 phages were analyzed by western blotting using a His-tag specific antibody. Lane 1, DNT-scFv-1; lane 2, DNT-scFv-2; lane 3, DNT-scFv-3;
lane 4, DNT-scFv-4; lane 5, DNT-scFv-5. (b) The proteins during DNT-scFv-3 expression and purification were analyzed by SDS-PAGE.
Marker (M), periplasmic extract before (−) and induction (+), periplasmic fractions (Per), and the elution fraction from NTA column (NTA).
Figure 4. Binding analyses of purified DNT-scFv-3 to DNT. (a) Different DNT-scFv-3 concentrations were incubated in 96-well plates immobilized with a DNT derivative [4-(2,4-dinitrophenyl)- butan-1-amine], and the amount of bound DNT-scFv-3 was determined using an HRP-His6-antibody. (b) DNT-scFv-3 (1 µM) was incubated with different concentrations of DNT ( ), TNT ( ), 4-nitrotoluene ( ), or benzoic acid ( ), and the amount of bound DNT-scFv-3 on the DNT-immobilized plate was determined using a HRP-His6-antibody.
●
■
Figure 5. QCM analyses of DNT binding to isolated DNT-scFv-3.
(A) The frequency change of QCM chip upon the binding of purified DNT-scFv-3. The time of DNT-scFv-3 addition and washing are indicated by arrows. (B) The frequency change of QCM chip immobilized with DNT-scFv-3 upon addition of DNT (1 mM). The time of 1 mM DNT addition and washing are indicated by arrows.
to DNT-scFv-3 (Fig. 4(b)). However, 4-nitrotoluene and benzoic acid failed to interfere with the binding of DNT- scFv-3 to DNT, indicating that DNT-scFv-3 shows higher selectivity over 4-nitrotoluene or benzoic acid.
Direct Binding Measurements of DNT to Surface Immobilized DNT-scFv-3 Using QCM-D. To use DNT- scFv-3 as an explosives sensor recognition molecule, surface immobilized DNT-scFv-3 should bind to free DNT. To examine whether DNT can bind to the surface immobilized DNT-scFv-3, the frequency change of a quartz crystal re- sonator coated with DNT-scFv-3 was measured using a quartz crystal microbalance with dissipation (QCM-D) (Biolin Scientic AB, Västra Frölunda, Sweden). The 6-His sequence at the C-terminus of DNT-scFv-3 binds the NTA-group coated on the surface of the quartz crystal chip, and the reduced frequency due to attachment of DNT-scFv-3 was readily detected (Fig. 4(a)). When 1 mM DNT was applied to the immobilized chip with DNT-scFv-3, a frequency change (Δf) was observed, indicating that immobilized DNT-scFv-3 can bind free DNT molecules (Fig. 4(b)).
In conclusion, we have successfully isolated an scFv clone that could be used for the detection of DNT. It specifically binds with high affinity to the DNT-derivative immobilized on surface, and free DNT also bound the surface immobiliz- ed scFv. The isolated scFv clone (or its genetic variants) could be used as a recognition molecule for explosives sensor.
Acknowledgments. This work was supported by grants from the Korea Dual Use Technology Program (DUTP)
under the Agency for Defense Development of Korea. We also thank the 2011 research program of Kookmin University, Korea.
References
1. Styles, J. A.; Cross, M. F. Cancer Lett. 1983, 20, 103.
2. Levine, B. S.; Furedi, E. M.; Gordon, D. E.; Barkley, J. J.; Lish, P.
M. Fundam. Appl. Toxicol. 1990, 15, 373.
3. Halasz, A.; Groom, C.; Zhou, E.; Paquet, L.; Beaulieu, C.;
Deschamps, S.; Corriveau, A.; Thiboutot, S.; Ampleman, G.;
Dubois, C.; Hawari, J. J. Chromatogr. A 2002, 963, 411.
4. Borch, T.; Gerlach, R. J. Chromatogr. A 2004, 1022, 83.
5. Jenkins, T. F.; Leggett, D. C.; Miyares, P. H.; Walsh, M. E.;
Ranney, T. A.; Cragin, J. H.; George, V. Talanta 2001, 54, 501.
6. Naddo, T.; Che, Y.; Zhang, W.; Balakrishnan, K.; Yang, X.; Yen, M.; Zhao, J.; Moore, J. S.; Zang, L. J. Am. Chem. Soc. 2007, 129, 6978.
7. Jaworski, J. W.; Raorane, D.; Huh, J. H.; Majumdar, A.; Lee, S.- W. Langmuir 2008, 24, 4938.
8. Jang, H.-J.; Na, J. H.; Jin, B.-S.; Lee, W.-K.; Lee, W.-H.; Jung, H.
J.; Kim, S.-C.; Lim, S.-H.; Yu, Y. G. Bull. Korean Chem. Soc.
2010, 31, 3704.
9. Wilson, R.; Clavering, C.; Hutchinson, A. Analyst 2003, 128, 480.
10. Nagatomo, K.; Kawaguchi, T.; Miura, N.; Toko, K.; Matsumoto, K. Talanta 2009, 79, 1142.
11. Huston, J. S.; Mudgett-Hunter, M.; Tai, M. S.; McCartney, J. E.;
Warren, F.; Haber, E.; Oppermann, H. Methods Enzymol. 1991, 203, 46.
12. Yang, H. Y.; Kang, K. J.; Chung, J. E.; Shim, H. Mol. Cells 2009, 27, 225.
13. Kobayashi, N.; Oyama, H.; Kato, Y.; Goto, J.; Soderlind, E.;
Borrebaeck, C. A. Anal. Chem. 2010, 82, 1027.
14. Markland, W.; Ley, A. C.; Lee, S. W.; Ladner, R. C. Biochemistry 1996, 35, 8045.
