Article pubs.acs.org/ac
Mapping of Surface-Immobilized DNA with Force-Based Atomic Force Microscopy Yoonhee Lee,† Sung Hong Kwon,†,⊥ Youngkyu Kim,‡ Jong-Bong Lee,‡,§ and Joon Won Park*,†,∥ †
Department of Chemistry, ‡School of Interdisciplinary Bioscience and Bioengineering, §Department of Physics, and ∥Division of Integrative Biosciences and Biotechnology, WCU Program, Pohang University of Science and Technology, San 31 Hyoja-dong, Pohang, 790-784, South Korea S Supporting Information *
ABSTRACT: Single-stranded 50-mer, 100-mer, and 150-mer DNAs were immobilized on a surface, and force-based atomic force microscopy (AFM) was employed to examine their behavior. A complementary 20-mer probe DNA on an AFM tip was used for the measurements. High-resolution maps were generated, and relevant parameters, including the force, stretching distance, unbinding probability, cluster size, and degree of distortion, were analyzed. Due to thermal drift, the cluster shape became increasingly distorted as the scan speed was decreased and as the map area was reduced. The cluster radius increased with the number of base (N), and the radius was proportional to N0.6 (r = 0.977) and N0.53 (r = 0.991). Due to the effect of the pulling angle, the apparent values of the stretching distance and the unbinding force decreased as the AFM probe was moved away from the center position; these values can be described as a function of sin θ.
S
single biomolecule with a microbead attachment6 and although FRET has been similarly applied,7 the superb lateral resolution of AFM would provide additional information on the hydrodynamic behavior of target molecules on a surface. Polyethylene glycol (PEG) is a widely employed linker that can be used to conjugate one end of a capturing agent (DNA or an antibody) to a functional group on the surface. Thus, the captured biomarker is subjected to observable Brownian motion. When the pixel size of the AFM images is smaller than the hydrodynamic distance of the captured biomarker, the corresponding AFM probe finds its counterpart at multiple pixels, forming a cluster, the size of which is related to the hydrodynamic radius.3,4,8 However, a fundamental understanding of key physical parameters, such as the stretching distance, the unbinding force value, the unbinding probability within the cluster, and the correlation between the cluster size and the hydrodynamic radius, has not yet been elucidated. In this study, a DNA-conjugated AFM probe was employed to image ssDNA (50-mer, 100-mer, and 150-mer DNA strands) immobilized on a surface. At each pixel, the force−distance curves were recorded under various conditions. Force maps were generated, and various crucial parameters, such as the cluster radius, the stretching distance, the force value, and the probability of obtaining a specific curve, were obtained. Thus,
ingle-molecule imaging methods have been applied to directly count and quantify individual biomarkers. Of these methods, atomic force microscopy (AFM) has shown great potential. The advantages of AFM include a high spatial resolution and the lack of a requirement for fluorescence labeling. Sahin et al. used AFM to discriminate the mechanical stiffness of double-stranded (ds) DNA from that of singlestranded (ss) DNA and were able to count the copy number of the captured DNA in a designated area without PCR.1 In addition, Hinterdorfer et al. developed simultaneous topography and recognition imaging (TREC) to image target molecules on a surface and measured the reduction in AFM probe oscillations as the probe was bound to its target molecule.2 Recently, Park et al. reported a force mapping protocol to analyze biomarkers captured on a microarrayed surface and demonstrated the inherent capability of AFM to individually “see and count” captured biomarkers, including prostate-specific antigen3 and HCV mRNA.4 It is believed that the current detection limit (ca. 1 fM) can be enhanced to subattomolar concentrations by either reducing the microarrayed spot size or by scanning a larger area. Park et al. also used AFM force mapping to image individual mRNA molecules exposed on the surface of a tissue.5 To precisely interpret the images obtained by force mapping, it is necessary to understand the physicochemical behaviors of the biomarkers when captured on a surface. Brownian motion should be taken into consideration when performing highresolution imaging. Although the tethered particle motion (TPM) method has been used to monitor the trajectory of a © 2013 American Chemical Society
Received: December 27, 2012 Accepted: March 19, 2013 Published: March 19, 2013 4045
dx.doi.org/10.1021/ac3037848 | Anal. Chem. 2013, 85, 4045−4050
Analytical Chemistry
Article
with hybridization buffer (2X SSPE buffer pH 7.4 containing 7.0 mM sodium dodecyl sulfate). The solution was stirred at 45 °C for 20 min, and the holder was then moved into a new beaker filled with fresh buffer. The solution was then stirred at 65 °C for 3 min. Subsequently, the slide was washed with an excess amount of water to remove any nonspecifically bound target ssDNA. Finally, the printed slides were stored in a nitrogen atmosphere at 4 °C. AFM Force Mapping Experiment. All of the AFM force measurements and mapping experiments were performed with a ForceRobot 300 apparatus (JPK Instruments AG, Germany) at room temperature. The AFM probe was calibrated in solution using the thermal fluctuation method, and the measured spring constant was in the range 0.015−0.025 N/ m. The force measurements were conducted in PBS (pH 7.4, 10 mM phosphate buffer, 2.7 mM potassium chloride, and 137 mM sodium chloride). The AFM tip was programmed to contact the surface with a force between 80 and 100 pN, rest at the surface for 0.10 s, and then, move a vertical distance of 200 nm. To obtain the force−distance curve, a total of 400 data points were recorded. To measure the effect of the pulling angle with a retract velocity of 2.0 μm/s, 25 force−distance curves were recorded at each pixel. The scan area was then divided into 8 × 8 (64) pixels. The AFM probe raster scanned the surface at intervals of 8.75 nm (150-mer), 7.50 nm (100mer), or 6.25 nm (50-mer), and the observed values were fitted with a Gaussian distribution. To perform a cluster size analysis, the scan area was divided into 20 × 20 (400) pixels and five force−distance curves were acquired at each pixel. The AFM probe raster scanned the surface at intervals of 4.00, 4.50, and 5.00 nm (150-mer), 4.00 nm (100-mer), or 3.00 nm (50-mer). Data Analysis. The collected rupture force−distance curves were processed and analyzed using JPK data processing software. The stretching distance and unbinding force were determined from the peak value of each force−distance curve. Any curves with a characteristic linear profile (rupture distance of
