Anal. Chem. 2004, 76, 6967-6972
DNA Directed Protein Immobilization on Mixed ssDNA/Oligo(ethylene glycol) Self-Assembled Monolayers for Sensitive Biosensors Christina Boozer,† Jon Ladd,† Shengfu Chen,† Qiuming Yu,† Jiri Homola,‡ and Shaoyi Jiang*,†
Department of Chemical Engineering, University of Washington, Seattle, Washington 98195, and Institute of Radio Engineering and Electronics, Academy of Sciences of the Czech Republic, Prague, Czech Republic
A stable and versatile biosensor surface is prepared by site-directed immobilization of protein-DNA conjugates onto a mixed self-assembled monolayer (SAM) composed of ssDNA thiols and oligo(ethylene glycol) (OEG) terminated thiols. The protein conjugates consist of an antibody chemically linked to a ssDNA target with a sequence complementary to the surface-bound ssDNA probes and are immobilized on the surface via sequence-specific hybridization. Compared to standard antibody immobilization techniques, this approach offers many advantages. The exceptional specificity of DNA hybridization combined with the diversity of potential sequences makes this platform perfect for multichannel sensors. Once a surface is patterned with the appropriate probe sequences, sequence-specific hybridization will sort out the target conjugates and direct them to the appropriate spots on the surface. In addition, the DNA SAMs are very stable and well suited to recycling by dehybdridization of the conjugates from the surface-bound probes. In this work, we demonstrate the specificity, sensitivity, and convenience of using protein-DNA conjugates to convert a DNA/ OEG SAM surface into a biosensor surface and apply this platform to the detection of human chorionic gonadotropin using surface plasmon resonance. Protein arrays are an attractive platform for applications such as drug screening, biosensors, and medical diagnostics. Due to challenges with fabrication, storage, and protein stability, they have not realized their full potential.1-3 In contrast, DNA chips are widely used for DNA sequencing, gene mapping, and molecular diagnostics.4-7 Fabrication of DNA arrays is routine, and unlike protein arrays, DNA arrays are stable and robust. In an attempt to exploit the strengths of DNA technology, self-assembly techniques based on DNA hybridization have been used in recent * To whom correspondence should be addressed. E-mail: sjiang@ u.washington.edu. † University of Washington. ‡ Academy of Sciences of the Czech Republic. (1) Huang, R. P. Front. Biosci. 2003, 8, D559-D576. (2) Cutler, P. Proteomics 2003, 3, 3-18. (3) Kodadek, T. Chem. Biol. 2001, 8, 105-115. (4) Southern, E.; Mir, K.; Shchepinov, M. Nat. Genet. 1999, 21, 5-9. (5) Wang, J. Nucleic Acids Res. 2000, 28, 3011-3016. (6) Service, R. F. Science 1998, 282, 396-+. (7) Ramsay, G. Nat. Biotechnol. 1998, 16, 40-44. 10.1021/ac048908l CCC: $27.50 Published on Web 10/22/2004
© 2004 American Chemical Society
years to spatially assemble an assortment of molecular components, such as nanoparticles, proteins, and polypeptides.8-11 While the feasibility of using DNA directed immobilization to prepare protein surfaces has been studied, the specificity and sensitivity of protein surfaces created by this method have not been investigated. The focus of this work is to address these issues by developing and characterizing a novel DNA probe surface, specially designed for DNA directed immobilization of proteins. In this paper, we introduce a novel single-stranded DNA (ssDNA) probe surface, which, when used in conjunction with a simple DNA-antibody conjugate, produces a biosensor surface with superior sensitivity and protein resistance. The ssDNA probe surface is a mixed self-assembled monolayer (SAM) of ssDNA and oligo(ethylene glycol) (OEG) terminated thiols. The protein conjugate consists of an antibody chemically linked to a ssDNA target with a sequence complementary to the surface-bound ssDNA probe. As shown in Figure 1, DNA hybridization will result in immobilization of the antibody on the surface. Previously, other forms of protein conjugates have been used for the assembly proteins on DNA surfaces. One group has prepared DNA-steptavidin conjugates that they have used to immobilize biotinylated antibodies on a DNA surface.8,12-14 In this work, the biotinylated antibodies are incubated with the DNAstreptavidin conjugates and then assembled via DNA directed immobilization. Although the end result is the conversion of a DNA array into a protein array, the process requires an extra incubation step, as well as the biotinylation of antibodies. In contrast, our conjugates directly convert a DNA surface into a protein surface in one simple step. Another group is using mRNA-protein fusions,15,16 which consist of a polypeptide covalently linked to its corresponding mRNA, to assemble proteins (8) Niemeyer, C. M. Trends Biotechnol. 2002, 20, 395-401. (9) Mirkin, C. A.; Letsinger, R. L.; Mucic, R. C.; Storhoff, J. J. Nature 1996, 382, 607-609. (10) Weng, S.; Gu, K.; Hammond, P. W.; Lohse, P.; Rise, C.; Wagner, R. W.; Wright, M. C.; Kuimelis, R. G. Proteomics 2002, 2, 48-57. (11) Ladd, J.; Boozer, C. L.; Chen, S.; Yu, Q.; Homola, J.; Jiang, S. Langmuir 2004, 20, 8090-8095. (12) Niemeyer, C. M.; Boldt, L.; Ceyhan, B.; Blohm, D. Anal. Biochem. 1999, 268, 54-63. (13) Niemeyer, C. M.; Burger, W.; Hoedemakers, R. M. J. Bioconjugate Chem. 1998, 9, 168-175. (14) Niemeyer, C. M.; Sano, T.; Smith, C. L.; Cantor, C. R. Nucleic Acids Res. 1994, 22, 5530-5539. (15) Roberts, R. W.; Szostak, J. W. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 1229712302.
Analytical Chemistry, Vol. 76, No. 23, December 1, 2004 6967
Figure 1. Site-directed immobilization of DNA-protein conjugates.
on DNA microarrays.10,17 The mRNA-protein fusions are prepared in vitro and are assembled on the DNA surface via hybridization of the mRNA with surface-bound DNA. Although elegant, this approach is currently limited to small, unstructured proteins and requires the user to identify an appropriate capture probe sequence from the mRNA of each protein fusion. As a result, the DNA arrays can be utilized only for assembling the corresponding mRNA-protein fusion. In contrast, our platform provides more versatility, because any protein can be conjugated to any DNA sequence, effectively making the DNA probe surface universal. While many ssDNA probe surfaces have been studied extensively,18 they are mainly designed for use with DNA, not protein. As a result, these ssDNA surfaces are often not protein resistant. If DNA-antibody conjugates are to be used to convert a DNA chip into a viable biosensor, the surface must be protein resistant to (1) ensure proper placement of the antibody conjugates on the sensor surface and (2) eliminate false alarms due to nonspecific binding of nontarget analytes. In this work, mixed DNA/OEG SAMs are used because they are a simple method for preparing a custom-tailored ssDNA surface with a protein-resistant background. Previously, mixed ssDNA SAMs have been prepared with mercaptohexanol (MCH) as a second thiol component.18,19 Although mixed SAMs of ssDNA and MCH have been shown to resist nonspecific binding of noncomplementary strands, MCH is not sufficiently protein resistant. Therefore, these DNA/MCH SAMs are appropriate for applications such as DNA sequencing or gene mapping yet are a not suitable for biosensor development. Modification of surfaces by OEG is a common and effective means for rendering surfaces protein resistant;20 hence, OEG was chosen as a diluent thiol. The resulting surface is a mixed ssDNA/OEG SAM that selectively hybridizes with complementary ssDNA yet resists nonspecific binding of both proteins and other DNA strands. (16) Nemoto, N.; MiyamotoSato, E.; Husimi, Y.; Yanagawa, H. FEBS Lett. 1997, 414, 405-408. (17) Xu, L. H.; Aha, P.; Gu, K.; Kuimelis, R. G.; Kurz, M.; Lam, T.; Lim, A. C.; Liu, H. X.; Lohse, P. A.; Sun, L.; Weng, S.; Wagner, R. W.; Lipovsek, D. Chem. Biol. 2002, 9, 933-942. (18) Herne, T. M.; Tarlov, M. J. J. Am. Chem. Soc. 1997, 119, 8916-8920. (19) Petrovykh, D. Y.; Kimura-Suda, H.; Whitman, L. J.; Tarlov, M. J. J. Am. Chem. Soc. 2003, 125, 5219-5226. (20) Prime, K. L.; Whitesides, G. M. J. Am. Chem. Soc. 1993, 115, 10714-10721.
6968
Analytical Chemistry, Vol. 76, No. 23, December 1, 2004
Compared to standard antibody immobilization techniques, this approach offers many advantages. First, the diversity of available DNA sequences combined with the high specificity of base pairing makes this platform ideal for producing multichannel biosensors or protein arrays. If each different antibody or protein is conjugated to a different target sequence, and the surface is patterned with the appropriate complementary oligonucleotide probes, a mixture of the conjugates can be applied to the surface and sequence-specific hybridization will direct the target conjugates to the appropriate spots on the surface. In one simple step, a DNA array is converted into a protein array. Second, a ssDNA chip is much more stable than a standard protein chip, where the antibodies may become denatured and lose activity over time. By storing the protein conjugates in solution and not immobilizing them until the time of use, protein activity can be preserved. Third, the ssDNA probe surface acts as a universal surface, with an infinite number of potential applications. The desired end use of the chip dictates its functionalization, not the other way around. Whereas a sensor surface that has been prespotted with prescribed antibodies can only be used to detect the associated antigens, a ssDNA chip can be used to create a custom surface by immobilizing any desired protein conjugates. Last, DNA hybridization is reversible, allowing for the chips to be regenerated. A protein surface can be converted back into DNA surface by dehybridizing the DNA, thus removing the associated proteins from the surface. EXPERIMENTAL METHODS Materials. Two sets of oligonucleotides and antibodies were used to prepare two unique DNA-antibody conjugates. The DNA sequences and descriptions can be found in Table 1. These sequences were chosen because they have been shown previously to have high hybridization efficiency.14 Briefly, thiolated ssDNA A and B were used to prepare the ssDNA probe surfaces, while the corresponding thiolated complements, c-A and c-B, were cross-linked to anti-human chorionic gonadotropin (hCG) and antilysozyme, respectively. Most of the surface characterization and optimization was performed using the A probe surfaces and antihCG/c-A conjugates. The anti-lysozyme/c-B conjugate was used for control experiments to test cross reactivity.
Table 1. DNA Sequencesa
a
name
sequence14
description
A c-A B c-B
5′ TCC TGT GTG AAA TTG TTA TCC GCT 3′ 5′ AGC GGA TAA CAA TTT CAC ACA GGA 3′ 5′ GTA ATC ATG GTC ATA GCT GTT 3′ 5′ AAC AGC TAT GAC CAT GAT TAC 3′
probe sequence used to prepare DNA SAMs target sequence conjugated to anti-hCG probe sequence used to prepare DNA SAMs target sequence conjugated to anti-lysozyme
All sequences are thiolated at the 5′ end.
All DNA used in this work was purchased from Synthegen, and the OEG terminated thiols were custom synthesized at the University of Washington. Anti-hCG, monoclonal to β-hCG, was purchased from Scripps Laboratory, as was the hCG antigen and polyclonal goat anti-hCG. Anti-lysozyme was purchased from Research Diangnostics, Inc. BSA and lysozyme were purchased from Sigma. Synthesis of DNA-Antibody Conjugates. The DNAantibody conjugates were synthesized by chemically cross-linking a thiolated ssDNA (sequence c-A or c-B) with the respective antibody. The antibodies (5 mg/mL) were reacted with a 10-fold excess of sulfosuccinimidyl 4-(p-maleimidophenyl) butyrate (sulfoSMPB; Pierce) in PBS (100 mM phospate buffer, pH 7.4, 150 mM NaCl). After incubation for 30 min at room temperature, the derivatized antibodies were desalted by ultrafiltration (30 000 MW cutoff membrane; Millipore) and the buffer was changed to PBE (phosphate-buffered EDTA; 100 mM phosphate buffer, pH 6.8, 5 mM EDTA). Thiolated ssDNA was added to the antibodies at a 1:1 ratio, and the mixture was incubated at room temperature for 30 min. Unreacted ssDNA was removed by ultrafiltration (100 000 MW cutoff), and the purification was verified by nondenaturing PAGE. Surface Functionalization. Prior to surface functionalization, all sensor surfaces were rinsed with ethanol and water, blown dry with nitrogen, and cleaned by 20 min of UV/ozone. After UV treatment, the sensor surfaces were rinsed again with water and ethanol and dried under a stream of N2. Sensor surfaces were functionalized immediately following cleaning. Pure ssDNA or mixed ssDNA/OEG SAMs were formed by immersing clean Au sensor substate in a 1.0 M KH2PO4 buffer solution of ssDNA and OEG thiols. The ssDNA thiol concentration was held constant at 100 nM for all experiments, while the OEG thiol concentration ranged from 0 to 100 µM. Following overnight assembly, the samples were rinsed thoroughly with water and dried by nitrogen. Samples were used immediately. Surface Plasmon Resonance (SPR) Sensor Surfaces and Instrumentation. SPR sensor substrates were prepared by coating clean BK-7 glass substrates with a 2-nm adhesion layer of chromium followed by a 50-nm layer of Au by electron beam evaporation. The SPR sensor setup has been described previously.21 Briefly, it is a home-built instrument based on the Kretschmann configuration of the attenuated total reflection method. The glass side of the gold-coated substrate is index matched to the prism coupler while the functionalized surface is mechanically pressed against an acrylic flow cell with a Mylar gasket. A polychromatic light beam is directed through the prism (21) Boozer, C. L., Yu, Q. M., Chen, S. F., Lee, C. Y., Homola, J., Yee, S. S., and Jiang, S. Y. Sens. Actuator, B 2003, 90, 22-30.
and the glass substrate and excites surface plasma waves at the metal-dielectric interface. The reflected light is analyzed with a spectrograph. DNA and Protein Binding Experiments. All SPR experiments began with a buffer baseline (TE-NaCl: 10 mM Tris-HCl with 1 mM EDTA, NaCl concentration as indicated, pH 7.2), and buffer was run again between each protein or DNA step. Solutions were flowed at a rate of 50 µL/min. Antibody conjugate and secondary antibody solutions were prepared at 0.02 and 0.05 mg/ mL, respectively. For the surface optimization experiments, the antigen concentration was 1.0 µg/mL. For the hCG detection experiments, 1 mg/mL BSA was added to the buffer to minimize protein losses due to protein degradation and nonspecific adsorption to tubing. RESULTS AND DISCUSSION Sequence Specificity and Protein Resistance. Control experiments were performed to test the sequence specificity and protein resistance of the DNA/OEG SAM surface. To test the specificity of hybridization to the surface-bound ssDNA probes, the SPR response to both complement and noncomplementary control oligonucleotide strands was measured. While the complement hybridized to the surface, there was no detectable binding of the noncomplementary control sequence. Depending on the density of ssDNA probes, the complements hybridized with coverage in the range of (1-5) × 1012 duplex/cm2. These data are not shown because they are in agreement with previous results for DNA SAMs backfilled with mercaptohexanol.18 High-concentration solutions of BSA were used to check the protein resistance of the DNA SAM surfaces. Figure 2 shows the SPR response of DNA SAMs with MCH and OEG backgrounds to a 1 mg/mL solution of BSA in TE-NaCl. Whereas the DNA/ MCH SAM rapidly and irreversibly adsorbs high amounts of BSA, the DNA/OEG SAM is protein resistant. Finally, Figure 3 shows the immobilization of complementary and noncomplementary DNA-antibody conjugates onto a ssDNA (sequence A) probe surface. The antibody conjugates with the complimentary sequence, anti-hCG/c-A, bind to the surface due to sequence-specific hybridization. The other antibody conjugates, anti-lysozyme/c-B, have a noncomplementary sequence and do not bind to the surface. In a similar manner, ssDNA/OEG SAMs prepared with the B sequence probes bind the anti-lysozyme/ c-B conjugates and resist the anti-hCG/c-A conjugates (not shown). These control experiments clearly demonstrate that immobilization of the conjugates is controlled exclusively by sequencespecific hybridization. The nonfouling OEG background prevents the protein segment of the conjugates, or any other protein for that matter, from nonspecifically binding to the surface, and the Analytical Chemistry, Vol. 76, No. 23, December 1, 2004
6969
Figure 2. Testing protein resistance. BSA (1 mg/mL) flowed over ssDNA SAMs prepared with MCH (solid line) and OEG (dashed line) diluent thiols. The surface prepared with OEG is completely protein resistant.
Figure 3. Control experiment to test for conjugate immobilization specificity. Both anti-hCG/c-A (solid line) and anti-lysoyzme/c-B (dashed line) conjugates flowed over a sequence A ssDNA/OEG SAM, but only the conjugate with the complimentary sequence bound to the surface.
specificity of DNA hybridization ensures proper placement of the conjugates. As a result, it should be possible to apply a cocktail of different DNA-protein conjugates to a patterned ssDNA probe surface and let sequence-specific hybridization direct the conjugates to the appropriate spots on the surface. Surface Optimization. DNA probe surface density was optimized to achieve maximal sensor performance. As described previously, the mixed ssDNA/OEG SAMs were prepared by coadsorption from a mixed thiol solution containing ssDNA thiol and OEG thiol. The ssDNA thiol solution concentration was held constant, while the OEG thiol concentration was varied over a wide range of concentrations. Due to the competition between the two thiol components, the concentration of OEG thiol in the assembly solution determines the surface density of ssDNA probes; i.e., low OEG concentrations lead to high ssDNA probe 6970 Analytical Chemistry, Vol. 76, No. 23, December 1, 2004
Figure 4. Optimization of surface composition for maximum conjugate binding. Mixed ssDNA/OEG SAMs were prepared with various amounts of OEG thiol, and the amount of conjugate (squares) and antigen (triangles) that bound to each surface was measured. A ssDNA probe surface with 100 nM ssDNA thiol and 10 µM OEG thiol was found to bind the most conjugate.
densities and vice versa. To determine the probe density that would immobilize the maximum amount of antibody conjugate, a series of ssDNA/OEG SAMs was prepared and conjugate hybridization was measured in situ by SPR. As shown in Figure 4, at high OEG concentrations, very little conjugate is immobilized because there is a low amount of ssDNA probes available on the surface. Once the OEG concentration drops below 25 µM, there is sufficient DNA on the surface to immobilize nearly a monolayer of antibody conjugates. As the OEG concentration is reduced further, the DNA probe density increases, until eventually, the ssDNA strands are packed so tight that the conjugates cannot penetrate the DNA SAM and hybridize. As a result, less DNAantibody conjugate can be immobilized. To maximixe conjugate coverage and sensor sensitivity, all subsequent ssDNA/OEG SAMs were prepared with 100 nM ssDNA thiol and 10 µM OEG thiol. Note that the optimal DNA probe density determined in this work depends on the size and shape of the protein immobilized. If a protein with a different footprint were used, the optimal DNA probe density would likely shift to reflect the change in protein packing. Detection of hCG. Figure 5 shows the calibration curve for the detection of hCG. The DNA-antibody conjugates were immobilized on a ssDNA/OEG surface and then used to detect hCG. A secondary antibody, polyclonal anti-hCG, was used for signal amplification. The lower limit of detection was 0.1 ng/mL, and the response was linear for concentrations less than 100 ng/ mL. For comparison, Figure 5 also shows the detection of hCG using biotinylated anti-hCG, which was immobilized via streptavidin on a mixed biotin and OEG SAM.21 For this system, the lower detection limit was 5 ng/mL, 50-fold higher. The superior sensitivity of the DNA platform is likely due to improved antibody immobilization efficiency combined with the increased flexibility afforded by DNA. With ssDNA/OEG SAMs, it is possible to control the surface coverage of DNA probes in a continuous manner, for optimal antibody coverage. Immobilization of biotinylated antibodies on streptavidin, on the other hand, is
Figure 5. (a) Calibration curve for the detection of hCG. Amplified response refers to the SPR shift due to the binding of a secondary antibody. The lowest detection was 0.1 ng/mL hCG for the DNA conjugates (squares) and 5 ng/mL for the biotinylated antibodies (triangles). (b) Linear region of detection curve. Table 2. Comparison of SPR Response for Detection of HCG on Fresh and Regenerated Surface
Figure 6. Recycling of DNA probe surface. A typical hCG detection experiment was run, and 50 mM NaOH was used to dehybridize the DNA duplex. After dehybdridization, all conjugate and protein is removed from the surface and a second detection was performed.
less accommodating due to the predefined distance between pockets in the streptavidin. The amount of streptavidin on the surface can be varied, but the distance between biotin binding spots cannot be changed. Furthermore, the 24-base pair-long DNA used in this work acts like a molecular tether and provides the immobilized antibody with greater mobility than a biotin/streptavidin linkage. Surface Regeneration. The DNA double helix can be dehybridized, or melted, by either chemical or thermal means. By dehybridizing the DNA duplex, the protein-DNA conjugates can be removed from the surface, thereby regenerating the ssDNA probe surface. In this work, we use 0.05 M NaOH to dehybridize the DNA duplex and remove the conjugates from the surface. Figure 6 shows a typical antigen detection and surface regeneration experiment. After equilibrating the surface with buffer, the conjugates are immobilized, antigen is detected, and the response is amplified with a secondary antibody. Once the detection is complete, the NaOH solution flows for 5 min to
chip surface
anti-hCG conjugate
SPR shift (nm) hCG (5 µg/mL)
polyclonal anti-hCG
fresh regenerated efficiency (%)
16.6 16.8 101
3.6 3.2 89
34.2 28.3 83
dehybridize the DNA duplex and remove the conjugate and all associated proteins. The baseline returns to its original position, indicating that the DNA duplex has been dehybridized and the antibody conjugates and all associated protein material have been removed from the surface. The surface is completely regenerated. At this point, the antibody conjugate is immobilized a second time, and the hCG detection is repeated. Table 2 compares the SPR response of the fresh DNA surface with the regenerated surface for each step in the hCG detection. Note that the regenerated DNA/OEG surface immobilized the same amount of antibody conjugate as the fresh sensor surface and that the responses to hCG and the secondary antibody are also comparable. Based on these preliminary results, it appears that further optimization of the recycling procedure could likely lead to 100% sensor regeneration. Surface regeneration is a desirable feature for any sensor that is operated in a remote or potentially dangerous environment. Instead of requiring a person to manually change the sensor chip, it can be regenerated with NaOH and used again for a new detection. In the experiment described above, we used a sensor chip for two consecutive detections of the same analyte, but a regenerated sensor could also be used for detection of an entirely different analyte. CONCLUSIONS In this work, we have presented a method for using DNA directed assembly to convert a ssDNA probe surface into a protein biosensor surface using DNA-protein conjugates. The conjugates are used in conjunction with a newly developed mixed DNA/OEG SAM that presents a protein-resistant background and allows for rational control over the DNA probe surface density. We have detected 0.1 ng/mL hCG, 50-fold lower than we previously Analytical Chemistry, Vol. 76, No. 23, December 1, 2004
6971
measured using biotinylated anti-hCG. Control experiments verify the specificity of the ssDNA/OEG SAM platform, and dehybridization of the DNA duplex allows for regeneration of the sensor surface. The ssDNA/OEG SAM developed in this work is unique because it is designed and optimized for use with both DNA and proteins. Most DNA surfaces are used exclusively with DNA, where nonspecific adsorption of proteins is not a concern. Since this DNA surface is used to produce a biosensor, the proteinresistant OEG background is crucial for the proper placement of the antibody conjugates and minimizes the chance of false detections caused by nonspecific adsorption. Due to the diversity of available DNA sequences and the high specificity of base pairing, DNA directed immobilization holds great potential for producing multichannel biosensors, or protein arrays. Although this paper focused on using the DNA-antibody conjugates and DNA/OEG SAM to prepare a single-channel biosensor surface, our control experiments clearly show that this system easily could be used to spatially assemble multiple proteins into an array. If a ssDNA chip was patterned with distinct probe
6972
Analytical Chemistry, Vol. 76, No. 23, December 1, 2004
sequences, and the desired proteins were conjugated to the corresponding compliments, site-directed hybridization could be used to immobilize the proteins on their respective spots. In a single step, a DNA array could be converted into a protein array. Furthermore, this surface could be used with either the DNAstreptavidin conjugates8,12-14 or the mRNA-protein fusions15,16 discussed earlier. ACKNOWLEDGMENT We thank Esmaeel Naeemi from Buddy Ratner’s group at the University of Washington for synthesis of the OEG thiols. This work has been supported by the Defense Advanced Research Projects Agency (DAAL01-96-K-3614) and the Food and Drug Administration (FD-U-002250). C.B. was partially supported by a fellowship from the Ford Motor Company.
Received for review July 26, 2004. Accepted September 13, 2004. AC048908L
