DNA Hybridization Efficiency on Concave Surface Nano-Structure in

Department of Life Sciences, Faculty of Life Sciences, Toyo University, 1-1-1 Izumino, Itakura-machi, Oura-gun, Gunma374-0193, Japan. Langmuir , 2014,...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/Langmuir

DNA Hybridization Efficiency on Concave Surface Nano-Structure in Hemispherical Janus Nanocups Hyonchol Kim,† Hideyuki Terazono,†,‡ Hiroyuki Takei,†,§ and Kenji Yasuda*,†,‡ †

Kanagawa Academy of Science and Technology, KSP East 310, 3-2-1 Sakado, Takatsu-ku, Kawasaki, Kanagawa 213-0012, Japan Department of Biomedical Information, Tokyo Medical and Dental University, Division of Biosystems, Institute of Biomaterials and Bioengineering, 2-3-10 Kanda-Surugadai, Chiyoda-ku, Tokyo 101-0062, Japan § Department of Life Sciences, Faculty of Life Sciences, Toyo University, 1-1-1 Izumino, Itakura-machi, Oura-gun, Gunma374-0193, Japan ‡

ABSTRACT: We examined the effect of a concave structure on DNA hybridization efficiency using an inner surface of hemispherical Janus nanocups in the range from 140 to 800 nm. Target DNA was specifically immobilized onto the inner cup surface, hybridized with complementary DNA-attached 20 nm Au probes, and the number of the hybridized probes was counted by scanning electron microscopy. The hybridization density of the attached Au probes on 800 nm nanocups was 255 μm−2, which was 0.57 times that on a flat surface, 449 μm−2, and increased to 394 μm−2 on a 140 nm cup, 0.88 times of a flat surface, as the cup size decreased. The local density of attached Au probes within the central 25% at the bottom of the 800 nm nanocups was 444 μm−2, which was closer to that on a flat surface, and the tendency was the same for all sizes of cups, indicating that the size dependency of DNA hybridization efficiency on the concave structures were mostly affected by the lower efficiency of side wall hybridization.



INTRODUCTION The nanoparticle is one of the most widely used tools in life science. It is used as a label for visualizing target molecules1−5 and as a probe for the purification of target molecules or cells.6,7 It is also used in cancer therapy.8 One practical challenge in nanoparticle use is the fabrication of nanoparticles, such as core−shells or void structures,9−12 that can carry small molecules attached within them. These particles can be used as artificial casts for molecular reactions colocalized in nanospace that mimic reactions in living systems.13−17 Understanding how the efficiency of the target-ligand reaction is affected by the nanoscale surface structure to which the target is attached is therefore important. Understanding this is also important for significantly improving biomolecular sensing. In the DNA microarray assay, for example, the sensitivity with which target DNA captured on a chip could be detected was increased a thousand times or more by using gold nanoparticles instead of conventional fluorescent probes as sensing probes.2,18,19 Although its strict mechanism is still unclear, one possibility is the structure of gold nanoparticle contributed © 2014 American Chemical Society

for condensed immobilization of probe DNA onto the particle surface, which yielded high affinity and low detachment for its target DNA. To clarify the contribution of nanoscale substrate roughness to such reactions, we previously examined the improvement of DNA immobilization due to a convex curvature on gold nanoparticle surfaces and found that as the curvature increased, the density of immobilized DNA increased to 130 times that on a flat surface.20 Although the efficiency of molecular immobilization on rough and smooth convex surfaces has been widely studied,20−25 the efficiency on concave surfaces has not. If concave nanostructures affect the efficiency of molecular sensing, the sensitivity with which target molecules can be detected could be further improved not only by using gold nanoparticles as probes but also by forming concave gold nanostructures on the substrate in a DNA microarray. Received: April 1, 2013 Revised: January 12, 2014 Published: January 17, 2014 1272

dx.doi.org/10.1021/la403557g | Langmuir 2014, 30, 1272−1280

Langmuir

Article

Figure 1. Immobilization of target DNA on nanocups and its hybridization with Au probes. (a) Schematic image of the reaction. Thiolated target DNA fragments were immobilized on inner Au surfaces of Au/Ni Janus nanocups 140 to 800 nm in diameter, and the Au probes to which complementary DNA for targets was immobilized hybridized with their targets on nanocups. Success of reactions was confirmed by comparing results with those obtained using nanocups on which noncomplementary (control) DNA had been immobilized. (b−i) FE-SEM images of (b) 140, (c) 200, (d) 250, (e) 300, (f) 400, (g) 500, (h) 600, and (i) 800 nm cups after the hybridization reactions. (j) FE-SEM images of flat surfaces after the hybridization reactions. Left: Complementary pairs. Right: Noncomplementary pairs. Bar: 200 nm.

strict size control, just the same quality with the strict size control of polystyrene spheres. We have already used this method to make “Janus”26,27 nanocups, where two layers of different metals constitute the outer and inner surfaces of the cuplike structure, having various diameters. Biomolecules can be immobilized on only the inner layer of a Janus nanocup by

In previous studies, we fabricated strictly size-controlled metal nanoparticles having a characteristic spherical bowl shape, which we call nanocups, by using polystyrene spheres as a template and coating them with various thin metal layers by thermal deposition.9,10 One advantage of this method is the ease of fabrication of particles having multiple metal layers with 1273

dx.doi.org/10.1021/la403557g | Langmuir 2014, 30, 1272−1280

Langmuir

Article

Figure 2. Quantification of attached Au probes. (a) Relationship between cup diameter and the number of attached Au probes on a nanocup. Complementary: target DNAs were immobilized on nanocups. Noncomplementary: control DNAs were immobilized on nanocups. Error bars show standard deviation (S.D.). (b) Detailed numbers of attached Au probes on a cup for noncomplementary pairs. The substrate was then placed in a vacuum evaporator (VPC-1100, Ulvac Kiko), and 5 nm of Au and 20 nm of Ni were sequentially deposited on the spheres. The metal-deposited polystyrene spheres were then placed in a UV-ozone cleaner (UV253HR, Filgen), and the spheres were removed from the deposited metal by UV-ozone oxidization, leaving Janus nanocups whose inner layer was Au and outer layer was Ni. Polydimethylsiloxane (PDMS, SYLGARD 184 silicon elastomer, Dow Corning) sol was then spread onto the nanocups and hardened by curing at 95 °C for 1 h. After the curing, the PDMS sheet was peeled off along with the nanocups that were embedded on the surface. As a result, the opening of the nanocavity faced the exterior. The nanocups on the PDMS sheet were treated with oxygen plasma in a plasma etching system (FA-1, Samco) for 5 min at 100 W in order to remove excess PDMS. For the evaluation of DNA hybridization efficiency, a singlestranded DNA fragment having a thiol group attached at its 5′ end through a short alkyl chain (C6), referred to hereafter as “target DNA,” was immobilized on the inner Au layer of the cups. The target DNA

making the inner layer a bioreactive element like gold, and such a cup can be used as an artificial nanocast to study molecular interactions on concave surfaces. 10 In this study, we investigated the relationship between DNA hybridization efficiency and the curvature of concave nanostructures by using a set of differently sized Janus nanocups.



EXPERIMENTAL SECTION

Janus nanocups, each consisting of a Au inner layer and a Ni outer layer, were prepared according to the following procedure, which has been reported previously.9,10 Polystyrene spheres having eight different diameters (STADEX series with nominal diameters of 143, 196, 254, 309, 402, 506, 603, and 814 nm JSR) were mixed with 1-ethyl-3-(3dimethylaminopropyl) carbodiimide (EDC) in water, with the EDC concentration adjusted to 20 mM. A small volume of each suspension of spheres was placed on a flat Si substrate, incubated for 10 min, washed with water, and dried to form a monolayer on the substrate. 1274

dx.doi.org/10.1021/la403557g | Langmuir 2014, 30, 1272−1280

Langmuir

Article

was commercially synthesized (Tsukuba Oligo) and had the following sequence of 25 bases: 5′-AAAAAGAATGATGCTCACTTGTTGC-3′ (Tm = 69.3 °C). The target DNA was immobilized on the inner Au layer of Au/Ni Janus nanocups10 by dissolving it in 4× saline sodium citrate (SSC) containing 0.5% sodium dodecyl sulfate (SDS) at a concentration of 1 μM and placing the DNA solution on the nanocups on a substrate at 40 °C. After 16 h of incubation, reacted nanocups were washed with ultrapure water 3 times. For control experiments, another DNA fragment having the sequence 5′-AAAAATGACCAGCACCATGGCGGTT-3′ (Tm = 74.2 °C) was immobilized on the nanocups in the same manner. After the immobilization, the targets were labeled with 20 nm Au nanoparticle probes to which complementary DNA had been attached (referred to hereafter as “Au probe”). To prepare the Au probes, a colloidal solution of Au nanoparticles 20 nm in diameter at a concentration of 1.2 nM was obtained from a commercial source (British BioCell). A 5′-thiolated DNA fragment having 20 bases complementary to the 3′ end of the target DNA was attached to the Au probe in the same method as in previous studies.2,10,28 The prepared Au probes were suspended in 4× SSC containing 0.5% SDS, placed on Janus nanocups on which target DNA had already been immobilized, and then incubated at 60 °C for 2 h for the hybridization reaction between the target DNA and the probe. After incubation, reacted nanocups on the substrate were washed with 6× SSC and 2× SSC at room temperature. Finally, the nanocups were washed with ultrapure water to remove any salts. The reacted Janus nanocups were observed using a field emission scanning electron microscope (FE-SEM, JEOL JSM 6701-F). Typical observation conditions were as follows: 5 kV acceleration voltage, 10 μA emission current, 200 pA probe current, 50000× magnification, 8 mm working distance, and the backscattered electron detection mode. Results of hybridization reactions were evaluated by using the FE-SEM and counting the numbers of Au probes attached to nanocups (averages from at least 50 cups for each condition).

nanocup was counted in the FE-SEM pictures. The results of counting complementary and noncomplementary pairs of DNA samples are shown in Figure 2a. Although for complementary pairs, the numbers of captured Au probes increased with increasing cup diameter because of the increase of surface area of each single cup; for noncomplementary pairs the numbers were almost negligibly small throughout the range of cup diameters. The numbers of captured Au probes in a nanocup are listed in Table 1 for both complementary and nonTable 1. Numbers of Captured Au Probes on a Janus NanoCupa nanocup diameter (nm) 140 200 250 300 400 500 600 800

for complementary pairs (mean ± S.D.) 12 21 24 28 65 72 93 256

± ± ± ± ± ± ± ±

2.4 3.4 4.6 4.8 8.0 10 12 24

for noncomplementary pairs (mean ± S.D.)

S/N ratio

± ± ± ± ± ± ± ±

120 1050 600 70 3250 720 930 853

0.1 0.02 0.04 0.4 0.02 0.1 0.1 0.3

0.3 0.1 0.2 0.6 0.1 0.3 0.2 0.5

a

N = 50 for 140, 200, 250, 400, 600, and 800 nm, and N = 100 for 300 and 500 nm.

complementary pairs. In cups from 140 to 600 nm in diameter, the standard deviation (S.D.) of the number of attached Au probes was within 20% of the mean value, and in 800-nm nanocups, it was within 10%. This smaller S.D. in the largest cups is partly because of the greater total number of attached Au probes: more than 250 Au probes were attached on an 800 nm nanocup, which was 2.5 times more than that on a 600 nm nanocup and 10 times more than that on a 200 nm nanocup. The numbers listed in Table 1 for noncomplementary pairs are less than 1, which is due to averaging of the numbers for more than 50 cups. Figure 2b also shows the detail of captured Au probe numbers on the Janus nanocups having noncomplementary target DNA. As shown there, the probe was not detected in over 70% of cups of all diameters, and there were no cups on which more than 4 were captured. The result shows that nonspecific adhesion in a cup is negligibly small. The S/N ratio defined as the number of Au probes captured on cups on which target DNA had been immobilized divided by the number captured on cups on which control DNA had been immobilized ranged from 70 to 3250, indicating the success of specific hybridization reaction in the nanocups. To evaluate the contribution of concave curvature to the DNA hybridization efficiency, we calculated the density of Au probes along the inner surfaces of the nanocups (Figure 3). These densities are also summarized in Table 2, where the value obtained using the same experimental procedure on a flat substrate is also listed (449 ± 4 μm−2). As shown in Figure 3, the average density of attached Au probes on a 140 nm nanocup was 394 μm−2, which was 0.88 times that on a flat substrate, and the values on nanocups more than 250 nm in diameter were 0.37 to 0.58 times that on a flat substrate [e.g., 0.57 times for 800 nm nanocups (255 μm−2 on average)]. These results indicated that hybridization on a concave structure more than 250 nm in diameter is less efficient than that on a flat surface. We then examined the contribution of concave structure to hybridization efficiency in more detail by measuring the



RESULTS AND DISCUSSION We first examined the microscopic improvement of DNA hybridization caused by the concave surface structure in single cups. We fabricated Janus nanocups of eight different strictly controlled sizes: 143 ± 2, 196 ± 5, 254 ± 8, 309 ± 9, 402 ± 6, 506 ± 12, 603 ± 14, and 814 ± 19 nm (we hereafter refer to them as 140, 200, 250, 300, 400, 500, 600, and 800 nm cups). As we have reported previously, their size variations, in terms of the coefficient of variation, were within 5% regardless of the size.9 We next examined the efficiency of hybridization between the target DNA fragments immobilized on the Janus nanocups and the probe DNA fragments attached to the Au probes (see Figure 1a). First a thiolated target DNA fragment was immobilized on the inner Au surface of the Janus nanocup through Au−S bonding, and then the Au probe was reacted with the target DNA fragment on the nanocup. The conjugate was then dried and observed using a FE-SEM. The success of the DNA hybridization was evaluated by counting the number of Au probes captured. For quantitative confirmation of specificity, we also used probes to which noncomplementary DNA fragments had been attached. Figure 1 (panels b−i) show typical micrographs of the results of complementary (left) and noncomplementary (right) reactions of Au probes with Janus nanocups with diameters ranging from 140 to 800 nm. Figure 1j shows the results obtained when the target DNA was immobilized on a flat substrate. Captured Au probes are seen in the micrographs as small white dots. To evaluate specific hybridization reactions on nanocups quantitatively, the number of captured Au probes in each single 1275

dx.doi.org/10.1021/la403557g | Langmuir 2014, 30, 1272−1280

Langmuir

Article

Figure 3. Relationship between nanocup diameter and the density of Au probes captured along the inner surface of Janus nanocups. Error bars show standard deviation (S.D.).

Equation 3 predicts that Dx decreases with increasing x and finally approached to the half of D0 at x is 100% (whole cup). To evaluate the validity of our assumption, we compared the densities of attached Au probes at the bottom of 800 nm cups in 10, 25, and 50% areas and compared them with the number of Au probes on the whole inner surface (100% area). Figure 4c shows the local densities of captured Au probes (Dx) in 10, 25, and 50% areas at the bottom of 800 nm cups, as well as in the whole hemisphere. Also shown in Figure 4 are the density on a flat surface, DFL, and the line calculated from eq 3 when using DFL as D0 (the same substitution is used hereafter). The average Dx values for the central 10 and 25% areas of the cups were 481 and 444 μm−2, close to the Dx for a flat surface (dashed-dotted line). Dx values slightly larger than DFL in this region might be due to the surface area increase yielded by the concave curvature (see the Figure 4c inset). On the other hand, the average density for the central 50% of the cup decreased to 295 μm−2, which is 0.66 times the Dx for a flat surface, and approached 255 μm−2 for the whole cup area. As shown in Figure 4c, all data points were close from the line calculated using eq 3), indicating the validity of our assumption (i.e., the concave surface structure), especially the lateral surface of the nanostructure, contributed to the reduction of hybridization efficiency. For additional evaluation of the validity of the above discussion, the densities of attached Au probes on the bottom 25% area of the cups, D25%, were calculated for all sizes of cups in order to evaluate the effect of the lateral surface of the concave structure in detail. The D25% values for all nanocup diameters are shown in Figure 5 as a function of cup diameter and are listed in Table 2. In accordance with eq 3, D25% should be 0.875 times DFL (dashed line in Figure 5), and all the data points we obtained experimentally were spread around that value, which also indicates that the concave surface structure, especially the lateral surface of the nanostructure, contributed to the reduction of hybridization efficiency. We also calculated the ratio of the attached probe density in the total area of the cup, D100%, to the probe density in the bottom 25% of the cup, D25%, which reflect the reduction of hybridization efficiency caused by the contribution of lateral

Table 2. Densities of Captured Au Probes on a Janus NanoCup nanocup diameter (nm) 140 200 250 300 400 500 600 800 flat surface

density (μm−2) (mean ± S.D.) 394 326 241 195 259 182 164 255 449

± ± ± ± ± ± ± ± ±

density for 25% center of a cup, D25% (μm−2) (mean ± S.D.)

78 54 47 34 32 29 21 24 4

404 334 266 262 433 340 251 444 449

± ± ± ± ± ± ± ± ±

155 71 58 51 58 51 41 44 4

distribution of Au probe attachment on the concave inner surface of an 800 nm Janus nanocup because the number of probes attached there was large enough for localization measurement, and its S.D. was less than 10% of its mean value. We assumed that one main cause for the reduction of hybridization efficiency on a nanocup was less binding of Au probes on the lateral surface of the cup and that this binding depends on the solid angle θ from the cup bottom. When we assume the local density of Au probes at θ, Dθ·local is described as Dθ ·local = D0 cos θ

(1)

where D0 is local density of Au probes at the center of the bottom of the cup (see Figure 4a). The ratio of Dθ·local to D0 decreases with increasing θ, as shown in Figure 4b. The average density of the Au probes within θ, Dθ (see Figure 4(a)), is then described using D0 as Dθ = D0 sin 2 θ /2(1 − cos θ )

(2)

If we consider eq 2, a function of the area ratio x for the total cup area instead of a function of θ (i.e., gauge transformation with x = 1 − cos θ), the density of Au probes in a central x × 100% area, Dx, can be written Dx = D0(1 − x /2)

(3) 1276

dx.doi.org/10.1021/la403557g | Langmuir 2014, 30, 1272−1280

Langmuir

Article

Figure 4. Dependence of captured probe density on the position of the inner cup surface. (a) Schematic sectional view of a Janus nanocup with diameter d. Local densities are designated D0 at the cup bottom and Dθ•local at the position of solid angle θ from the cup bottom. Averaged density from the bottom to θ is designated Dθ. (b) Dependence of the ratio Dθ·local/D0 on θ. (c) Densities of captured Au probes at the bottom x × 100% area of 800 nm cups, Dx. The densities of 10 (D10%), 25 (D25%), 50 (D50%), and 100% (whole cup, D100%) are plotted. Dashed-dotted line indicates the density on a flat substrate, DFL, and dashed line indicates the density expected from eq 3. The inset is a schematic image showing the small density increase caused by the concave curvature at small x. Error bars show standard deviation (S.D.).

corresponding to cups 200−400 nm in diameter, the values of the ratio became smaller gradually in a sigmoidal curve and approach 57% (the value expected from eq 3), indicating that this phase is a transition state between phases 1 and 3. In phase 3, corresponding to cups more than 400 nm in diameter (more than twenty times the diameter of the Au probes), the ratio is almost constant and is close to the value expected from eq 3, indicating a clear contribution of the concave structure in this phase. From these results one can conclude that (1) a concave structure 20 times the size of the approaching probe can decrease hybridization efficiency 43%, (2) a concave structure less than 10 times the size of the approaching probe does not decrease hybridization efficiency significantly below that on a

surface on concave structure and plotted it as a function of cup diameter (Figure 6a). The ratio expected from eq 3, 57%, is also shown in Figure 6a, where the ratio D100%/D25% is subdivided into three phases. Phase 1 corresponds to cups less than 200 nm in diameter (i.e., less than ten times the diameter of the Au probes), and in in this phase the ratio is almost constant at 98%, indicating first that the ratio does not exceed 100% (i.e., the probe density on the lateral surface is not greater than that on the center portion) and second that even though the density on the lateral surface is less than that at the center, the contribution of the structure is small in this phase because the depth of the structure is relatively small in comparison with the size of the approaching Au probe. In phase 2, 1277

dx.doi.org/10.1021/la403557g | Langmuir 2014, 30, 1272−1280

Langmuir

Article

Figure 5. Relationship between cup diameter and the densities of the attached Au probes in the bottom 25% area of a cup, D25%. The density on a flat surface, DFL, was calculated using eq 3.

flat substrate, and (3) a concave structure between 10 and 20 times the size of the probe partially inhibits hybridization efficiency. From a macroscopic point of view, fabrication of a hemispherical nanostructure increases the total surface area to twice that of the projection area (i.e., 2πr2/πr2, where r is the radius of the hemispherical nanostructure). This suggests that the DNA hybridization efficiency in a projection area can be increased about 2-fold by a hemispherical nanostructure whose diameter is less than 10 times that of the approaching probe. The projection density of the Au probe, calculated according to the above discussion from the results in Figure 3, is shown as a function of cup diameter in Figure 6b, where improved projection density is not clear in phase 2 but the projection density in phase 1 is clearly higher than DFL. This shows that the fabrication of small concave nanostructures improved hybridization efficiency from a macroscopic projection viewpoint. Another possible explanation for less attachment of Au probes on the inner side walls of hemispherical nanocups is nonuniform formation of the Au layer of the nanocup because the metal layers of nanocups were formed on the polystyrene template by vapor deposition, which yielded less Au deposition on the side walls (especially on the almost vertical portion of the sphere) and, therefore, resulted in less immobilization of the target DNA on the lateral surface of large nanocups. However, we set the thickness of the Au layer to 5 nm, and thus at least 50% of the hemisphere’s bottom surface (i.e., the part within 45° of the bottom) should be 5/(21/2) = 3.5 nm thick, which is 75% more than the thickness required for stable attachment of target DNA, 2 nm.29 Hence, the density of the attached probe DNA should be the same throughout the central 50% of the total hemisphere area. In 800 nm cups, however, the density of attached Au probes in this area was only 66% of that on a flat surface. This indicated that concave curvature reduced the DNA hybridization efficiency. Another possible factor contributing to the reduction of hybridization efficiency is the target DNA conformation on the nanocup surface. According to previous reports by Parak and his colleagues,24,25 there are three different configurations of single-stranded DNAs attached to Au nanoparticles: (i)

stretched, (ii) random coiled, and (iii) wrapped forms. The first is the configuration of DNA densely immobilized on a solid surface and is especially likely to be formed by DNA comprising less than about 30 bases. The second is an intermediate configuration between the configurations of densely packed DNA and thinly packed DNA, and the probe DNA can approach the target DNA immobilized on a solid surface because this configuration provides enough space for it. The third configuration is that of thinly immobilized DNA, for which the hybridization efficiency is low because the DNA lies on and is attached to the solid surface. For the following reasons, we concluded that the target DNAs immobilized on the nanocup surface in this study were in the stretched configuration. First, the target DNAs were 25 bases long, which is less than the length thought to be needed to form the random coiled configuration.24,25 Second, it is unlikely that the target DNA was in the wrapped configuration because a micromolar-order concentration of target DNA, which can form a densely packed DNA layer on Au nanoparticles,20 was reacted with nanocups for immobilization, and its hybridization would result in a density of attached Au probes that would be of the same order as that reacted with densely packed target DNA on DNA microarrays.19,28 The possibility that the immobilized DNA formed the wrapped configuration is therefore negligible. Yet another reason is that although nonthiolated DNA can be immobilized on a gold surface,30 the binding energy of this nonspecific attachment is expected to be lower than that of the sulfur−Au bond31 used in this study, and therefore, the contribution of nonspecific binding of DNA would also be negligible. The expected density of target DNAs immobilized on the inner solid surface of the cup is less than that of DNA closely packed on a flat surface because of the steric hindrance caused by the concave curvature of the nanocup surface and is of the same order as the density of DNA closely packed on a convex surface having a smaller diameter by twice of DNA contour length than the nanocup in maximum. The contour length of the target DNA used in this study was estimated to be 11.5 nm,32 and the estimated packing density on the 140 nm cup, for example, is of the same order as that on a convex surface, 117 nm in diameter. Although the DNA 1278

dx.doi.org/10.1021/la403557g | Langmuir 2014, 30, 1272−1280

Langmuir

Article

Figure 6. Relationships between cup diameter and the contributions of concave curvature to DNA hybridization efficiency. (a) Cup-diameter dependence of the ratio of the density of probes in the bottom 25% of the cup to that in the whole cup, D100%/D25%. Values of 100% and 57% (expected from eq 3) are indicated by horizontal dashed-dotted and dashed lines, respectively. Boundaries between three different phases are indicated by vertical dashed lines. (b) Relationship between nanocup diameter and calculated projection density of Au probes. DFL: density on a flat surface.

configuration could be the stretched form, whose reaction efficiency is less than that of the random coiled form, the hybridization efficiency is still high enough that about 200 μm−2 of Au probes were attached, which is about 30% of the full packing with hexagonal lattice structure in the 2D surface. This indicated that the reduced attachment of Au probes to the nanocup was due more to the concave-structure-caused binding on the lateral surface than to the DNA configurations on the nanocup. These results indicate the importance of surface structures for the acquisition of stable and reproducible results in hybridization-based experiments using Au probes. The concave roughness larger than 200 nm (phases 2 and 3) has a potential to reduce the local density of attached probes to a level lower than that on a flat surface, and the inhomogeneity of the size of concave structures on sample surfaces like those in cellular systems or tissues can also reduce the local density of Au probes. One promising way to evaluate the correct number of

Au probes hybridized is to make sure the size of the concave structure is less than 200 nm (phase 1).



OUTLOOK We examined the effect of concave nanostructure on DNA hybridization efficiency by using a set of strictly size-controlled nanocups and found three cup-size phases: (i) smaller concave structures, typically less than 10 times the size of the approaching compound, that do not reduce hybridization efficiency and can improve it by increasing the total surface area in the projective view, (ii) larger concave structures, typically more than 20 times the size of the approaching compound, reduce hybridization efficiency by about half because of the lower hybridization efficiency at the side walls of large nanocups, and (iii) middle-size concave structures, 10−20 times the size of the approaching compound, partially reduce the efficiency and act as a transition state between the two phases. The results indicate the importance of the contribution 1279

dx.doi.org/10.1021/la403557g | Langmuir 2014, 30, 1272−1280

Langmuir

Article

the membrane skeleton at the plasma membrane interface by electron tomography. J. Cell Biol. 2006, 174 (6), 851−62. (17) Lingwood, D.; Kaiser, H. J.; Levental, I.; Simons, K. Lipid rafts as functional heterogeneity in cell membranes. Biochem. Soc. Trans. 2009, 37 (Pt 5), 955−960. (18) Park, S. J.; Taton, T. A.; Mirkin, C. A. Array-based electrical detection of DNA with nanoparticle probes. Science 2002, 295 (5559), 1503−1506. (19) Kim, H.; Takei, H.; Yasuda, K. Quantitative evaluation of a goldnanoparticle labeling method for detecting target DNAs on DNA microarrays. Sens. Actuators, B 2010, 144 (1), 6−10. (20) Kira, A.; Kim, H.; Yasuda, K. Contribution of nanoscale curvature to number density of immobilized DNA on gold nanoparticles. Langmuir 2009, 25 (3), 1285−1288. (21) Demers, L. M.; Mirkin, C. A.; Mucic, R. C.; Reynolds, R. A., 3rd; Letsinger, R. L.; Elghanian, R.; Viswanadham, G. A fluorescence-based method for determining the surface coverage and hybridization efficiency of thiol-capped oligonucleotides bound to gold thin films and nanoparticles. Anal. Chem. 2000, 72 (22), 5535−5541. (22) Peterson, A. W.; Wolf, L. K.; Georgiadis, R. M. Hybridization of mismatched or partially matched DNA at surfaces. J. Am. Chem. Soc. 2002, 124 (49), 14601−14607. (23) Hurst, S. J.; Lytton-Jean, A. K.; Mirkin, C. A. Maximizing DNA loading on a range of gold nanoparticle sizes. Anal. Chem. 2006, 78 (24), 8313−8318. (24) Parak, W. J.; Pellegrino, T.; Micheel, C. M.; Gerion, D.; Williams, S. C.; Alivisatos, A. P. Conformation of oligonucleotides attached to gold nanocrystals probed by gel electrophoresis. Nano Lett. 2003, 3 (1), 33−36. (25) Pellegrino, T.; Sperling, R. A.; Alivisatos, A. P.; Parak, W. J. Gel electrophoresis of gold-DNA nanoconjugates. J. Biomed. Biotechnol. 200710.1155/2007/26796. (26) Casagrande, C.; Veyssie, M. Janus Beads: Realization and 1st Observation of Interfacial Properties. C.R. Acad. Sci., Ser. II: Mec., Phys., Chim., Sci. Terre Univers 1988, 306 (20), 1423−1425. (27) Degennes, P. G. Soft Matter. Rev. Mod. Phys. 1992, 64 (3), 645− 648. (28) Kim, H.; Kira, A.; Yasuda, K. Non-amplified quantitative detection of nucleic acid sequences using a gold nanoparticle probe set and field-emission scanning electron microscopy. Jpn. J. Appl. Phys. 2010, 49 (6), 06GK07. (29) Kim, H.; Takei, H.; Yasuda, K. Production of Nanoparticles Using Several Materials for Labeling of Biological Molecules. Jpn. J. Appl. Phys. 2010, 49 (8), 087001. (30) Herne, T. M.; Tarlov, M. J. Characterization of DNA probes immobilized on gold surfaces. J. Am. Chem. Soc. 1997, 119 (38), 8916−8920. (31) Nuzzo, R. G.; Zegarski, B. R.; Dubois, L. H. Fundamentalstudies of the chemisorption of organosulfur compounds on gold(111). Implications for molecular self-assembly on gold surfaces. J. Am. Chem. Soc. 1987, 109 (3), 733−740. (32) Rechendorff, K.; Witz, G.; Adamcik, J.; Dietler, G. Persistence length and scaling properties of single-stranded DNA adsorbed on modified graphite. J. Chem. Phys. 2009, 131 (9), 095103.

of concave nanostructure to the Au probe density in the measurement of samples having rough surface structures, such as cells and tissues.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +81 3 5280 8046. Fax: +81 3 5280 8049. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We thank Ms. M. Murakami for her technical assistance. This work was financially supported by Japan Prize Foundation research grants, JSPS KAKENHI Grant 24681025, and the Kanagawa Academy of Science and Technology.



REFERENCES

(1) Reichert, J.; Csaki, A.; Kohler, J. M.; Fritzsche, W. Chip-based optical detection of DNA hybridization by means of nanobead labeling. Anal. Chem. 2000, 72 (24), 6025−6029. (2) Taton, T. A.; Mirkin, C. A.; Letsinger, R. L. Scanometric D. N. A. array detection with nanoparticle probes. Science 2000, 289 (5485), 1757−1760. (3) Csaki, A.; Kaplanek, P.; Moller, R.; Fritzsche, W. The optical detection of individual DNA-conjugated gold nanoparticle labels after metal enhancement. Nanotechnology 2003, 14 (12), 1262−1268. (4) Huber, M.; Wei, T. F.; Muller, U. R.; Lefebvre, P. A.; Marla, S. S.; Bao, Y. P. Gold nanoparticle probe-based gene expression analysis with unamplified total human RNA. Nucleic Acids Res. 2004, 32 (18), e137. (5) Bao, Y. P.; Wei, T. F.; Lefebvre, P. A.; An, H.; He, L.; Kunkel, G. T.; Muller, U. R. Detection of protein analytes via nanoparticle-based bio bar code technology. Anal. Chem. 2006, 78 (6), 2055−2059. (6) Gamper, M.; Viereck, V.; Geissbuhler, V.; Eberhard, J.; Binder, J.; Moll, C.; Rehrauer, H.; Moser, R. Gene expression profile of bladder tissue of patients with ulcerative interstitial cystitis. BMC Genomics 2009, 10, 199. (7) Niwa, A.; Heike, T.; Umeda, K.; Oshima, K.; Kato, I.; Sakai, H.; Suemori, H.; Nakahata, T.; Saito, M. K. A novel serum-free monolayer culture for orderly hematopoietic differentiation of human pluripotent cells via mesodermal progenitors. PLoS One 2011, 6 (7), e22261. (8) Grossman, J. H.; McNeil, S. E. Nanotechnology in Cancer Medicine. Phys. Today 2012, 65 (8), 38−42. (9) Kim, H.; Takei, H.; Yasuda, K. Production of size-controlled nanoscopic cap-shaped metal shells. Jpn. J. Appl. Phys. 2010, 49 (4), 048004−048005. (10) Kim, H.; Hayashi, M.; Terazono, H.; Takei, H.; Yasuda, K. Production of double-layered metal nano-cups for artificial nanospace of biomolecular reaction. Jpn. J. Appl. Phys. 2011, 50 (6), 06GJ03-1− 06GJ03-4. (11) Wagner, B. M.; Schuster, S. A.; Boyes, B. E.; Kirkland, J. J. Superficially porous silica particles with wide pores for biomacromolecular separations. J. Chromatogr., A 2012, 1264, 22−30. (12) Wu, Q.; Liu, C.; Fan, L.; Shi, J.; Liu, Z.; Li, R.; Sun, L. Heparinized magnetic mesoporous silica nanoparticles as multifunctional growth factor delivery carriers. Nanotechnology 2012, 23 (48), 485703. (13) Hynes, R. O. Integrins: versatility, modulation, and signaling in cell adhesion. Cell 1992, 69 (1), 11−25. (14) Harder, T.; Simons, K. Caveolae DIGs and the dynamics of sphingolipid-cholesterol microdomains. Curr. Opin. Cell. Biol. 1997, 9 (4), 534−542. (15) Simons, K.; Ikonen, E. Functional rafts in cell membranes. Nature 1997, 387 (6633), 569−572. (16) Morone, N.; Fujiwara, T.; Murase, K.; Kasai, R. S.; Ike, H.; Yuasa, S.; Usukura, J.; Kusumi, A. Three-dimensional reconstruction of 1280

dx.doi.org/10.1021/la403557g | Langmuir 2014, 30, 1272−1280