Supporting information Permanently Fixing or Reversible Trap-and ...
Supporting information
Permanently Fixing or Reversible Trap-and-Release of DNA Micropatterns on a Gold Nanostructure using Continuous-wave or Femtosecond Pulsed Near-Infrared Laser Light.
Tatsuya Shoji,† Junki Saitoh,‡ Noboru Kitamura,†,‡ Fumika Nagasawa,‡ Kei Murakoshi,†,‡ Hiroaki Yamauchi,¶ Syoji Ito,¶ Hiroshi Miyasaka,¶ Hajime Ishihara§ and Yasuyuki Tsuboi1,†,‡,||,*
†
Department of Chemistry, Graduate School of Science, Hokkaido University, Sapporo 060-0810, Japan
‡
Graduate School of Chemical Sciences and Engineering, Hokkaido University, Sapporo 060-0810,
Japan ¶
Division of Frontier Materials Science, Graduate School of Engineering Science, and Center for
Quantum Materials Science under Extreme Conditions, Osaka University, 1-3 Macikaneyama-cho, Toyonaka, Osaka 560-8531, Japan §
Department of Physics and Electronics, Osaka Prefecture University, 1-1, Gakuen-cho, Nakaku, Sakai,
Osaka 599-8531, Japan ||
JST, PRESTO, Japan.
* Corresponding Author e-mail: [email protected] (Yasuyuki Tsuboi)
S1
1. Optical setup and sample preparation For samples, we used λ-DNA (48.5 kbp, Nippon gene) stained with YOYO-1 (λabs = 490 nm, λem = 509 nm, Invitrogen Co. Ltd.) in a Tris-borate-EDTA buffer solution (45 mmol/l Tris base, 45 mmol/l boric acid, 1 mmol/l EDTA, pH 8.0). The concentration of the λ-DNA in the solution was 2.0 x 10-6 mol/l in nucleotide concentration. To suppress photobleaching of YOYO-1, glucose (30 mmol/l), glucose oxidase (1 mg/ml) and catalase (1 mg/ml) were added to the solution in the usual manner.1,2 For the plasmonic substrate, we fabricated gold nano-pyramidal-dimer arrays on a glass substrate by means of angular-resolved nanosphere lithography (AR-NSL substrate). We have already described the structure and optical properties of the AR-NSL substrate in detail elsewhere.3,4 It should be noted here that the substrate has a broad absorption band in the near-infrared (NIR) region (≥ 700 nm), which corresponds to the resonant LSP excitation. For LSP excitation, we used a cw NIR laser (λ = 808 nm, Spectra Physics, CWA0400-SXD-808-30-E) or a fs NIR laser (λ = 770 nm, repetition rate 80 MHz, pulse width 120 fs in FWFM, Toptica photonics AG, FemtoFiber pro). Both lasers can excite LSPs on the substrate around the absorption maximum. The absorbance of AR-NSL substrate at 770 nm is approximately same as that at 808 nm. We obtained the fluorescence micrographs using Hg lamp illumination, while bright-field micrographs were obtained as green colored background images because of S2
a bandpass filter. For the fluorescence microspectroscopy, the YOYO-1 in the DNA was excited with a cw visible Ar+ laser (λ = 488 nm, 5 – 10 µW (2.5 – 5.0 kW/cm2), Ion Laser Technology, 5490ASL). In such laser intensity of Ar+ laser, DNA was never trapped by only Ar+ laser irradiation on a plasmonic substrate. The cw and fs NIR lasers were loosely focused to a circular spot (diameter d ∼ 5 µm) on the AR-NSL substrate, while the visible laser was tightly focused approximately at the diffraction limit (d ∼ 0.5 µm) at the center of the circular spot. Accordingly, we observed fluorescence only from the center of the LSP excitation area. All the experiments were performed under ambient conditions at room temperature.
2. cw NIR intensity dependence on DNA assembly on an AR-NSL substrate We carried out cw NIR intensity dependence on DNA assembly formation on an AR-NSL substrate as shown in Fig. S1. Figure S1(a) and (b) shows a series of optical micrographs of representative DNA trapping behavior with cw NIR laser irradiation at (a) 4.0 and (b) 5.0 kW/cm2. At 5.0 kW/cm2 (Figure S1(b)), we did not observe anything during several tens of seconds from LSP excitation. During several minutes of LSP excitation, a micro-ring structure was formed and gradually grew in the LSP excitation area. The diameter of the micro-ring is slightly smaller than that of the LSP excitation S3
area. Even after switching off the LSP excitation the micro-ring remained stable on the substrate, keeping its shape and size. On the other hand, below 4.0 kW/cm2 (Figure S1(a)), however, we observed neither such micro-ring structure nor sign of assembly formation. Figure S1(c) shows cw NIR laser intensity dependence on the fluorescence intensity that was measured at the center of LSP excitation area. The fluorescence intensity decreased with increasing cw NIR laser intensity, suggesting that DNA was never trapped at the center of cw NIR laser light even though cw NIR laser intensity decreased to reduce photothermal effect. Furthermore, these results supported that the trap-and-release of DNA was achieved only with fs NIR pulsed-laser light. Pulsed-excitation would prevent DNA from being fixed on the substrate due to the intervals between the pulses.
S4
a 0s
LSP 1 min on
3 min
5 min
1 min
3 min
5 min
LSP 6 min off
5 µm
b 0s
LSP on
LSP off
6 min
5 µm c
Fluorescence intensity
1.0
0.5
0.0
0
5
10
cw NIR laser intensity / kWcm-2
Figure S1. Bright-field micrographs of DNA on a plasmonic AR-NSL gold substrate using cw NIR laser irradiation ((a) 4.0, (b) 5.0 kW/cm2) for LSP excitation. The time shown in each image is the cw laser irradiation time, with 0 s being just before switching on the laser. The cw laser irradiation was turned off at 5 min. The cw laser illumination covered the white dotted circle. (c) Relative fluorescence intensity measured at the center of LSP excitation are as a function of cw NIR laser intensity.
S5
3. Nonlinear optical effect on fluorescence behaviors In order to show no fluorescence emission of YOYO-1 in DNA via 2-photon absorption, we measured the fluorescence on a plasmonic substrate excited by cw and fs NIR laser light. As seen in the Fig.S2, the fluorescence of YOYO-1 was not completely observed in both of (a) cw and (b) fs NIR laser irradiation, although baselines were slightly rising. We quantified the intensity in the spectra under cw and fs illumination, respectively. The slight increase is negligible for fluorescence measurements of LSP-OT of DNA, because the intensity of the rising baseline was quite smaller than that of fluorescence from YOYO-1; fluorescence: raised baseline = 1: < 0.01 under cw NIR laser irradiation and 1: < 0.002 under fs NIR laser irradiation, respectively. The rising baseline would be ascribed to luminescence of the AR-NSL substrate via 2-photon absorption (two-photon luminescence, TPL).5 Thus, we safely concluded that YOYO-1 in DNA was not excited by cw and fs NIR laser irradiation.
S6
a
b 0.4
Fluorescence intensity
Fluorescence intensity
cw NIR off cw NIR on
0.2
0.0 500
550
Wavelength/ nm
600
fs NIR off fs NIR on
0.004
0.002
0.000 500
550
Wavelength/ nm
600
Figure S2. Fluorescence spectra of YOYO-1 in DNA on a plasmonic substrate excited by (a) cw and (b) fs NIR laser light (INIR ∼ 7.0 kW/cm2) without visible laser irradiation. Fluorescence spectra were acquired with the exposure time of 10 s.
Furthermore, we examined TPL measurement of YOYO-1 in DNA upon simultaneous NIR and visible laser irradiation without a plasmonic substrate. In Fig. S3, the black line shows fluorescence spectrum of YOYO-1 in DNA during visible laser irradiation, while the red dotted line shows the spectrum during visible and fs NIR laser irradiation. Comparing these spectra, TPL of YOYO-1 in DNA was not observed during these laser irradiations. Thus, TPL of YOYO-1 in DNA was never observed in our experimental system.
S7
Fluorescence intensity 500
550
Wavelength/ nm
600
Figure S3. Fluorescence spectra of YOYO-1 in DNA without a plasmonic substrate excited by visible (black) and visible+fs lasers (INIR ∼ 7.0 kW/cm2) (red dotted line). Fluorescence spectra were acquired with the exposure time of 10 s.
4. FCS measurement for estimation of temperature elevation during fs NIR laser irradiation Using fluorescence correlation spectroscopy (FCS), we determined the local temperature elevation (∆T) in the vicinity of AR-NSL surface (5 μm distance from the surface) during fs NIR laser irradiation. Fig. S4 shows a representative example of FCS. The fundamental experimental method and analysis were described in the previous literatures.4,6 In Fig. S4(a), autocorrelation function traces are shown as a function of fs NIR laser intensity (λ = 800 nm). An analysis of the trace provides a diffusion time of the dye molecules (rhodamine 123) (Fig. S4(b)). The diffusion time can be converted
S8
into ΔT based on the molecular diffusion equation. In Fig. S4(c), ∆T is plotted against fs laser intensity. A linear fs-laser intensity-dependence was clearly observed below 5.0 kW/cm2. Using this relation, we can determine the temperature elevation at the surface of gold nanostructure upon LSP excitation using a method that is same as in our previous work (extrapolation based on a quasi-1D-steady state thermal conduction model)4 and it was evaluated to be 7.2 K at 1.0 kW/cm2. Considering the extinction coefficient ratio of AR-NSL substrate at 800 and 770 nm, ∆T is estimated to be 7-8 K at 1.0 kW/cm2 of 770 nm fs laser. These values are appropriate in comparison with the previous literatures on the ∆T on the surface of AR-NSL or other plasmonic substrates.7– 9
S9
c
a
b
Figure S4. FCS for the evaluation of temperature rise upon LSP excitation with fs NIR laser. The detecting position is 5 μm distance from the gold surface. (a) Dependence of the autocorrelation curve on the power of the incident fs laser light. (b) Averaged diffusion time of the probe dye in the observation volume as a function of excitation intensity of LSP. (c) Temperature at the focal point of the NIR laser light as a function of the incident laser power for the solution.
5. Molecular diffusion of DNA during the interval between two adjacent fs laser pulses Considering molecular diffusion, we should estimate the diffusive distance of λ-DNA (chain length L = 16.4 µm)2,10 during the interval time t between two adjacent fs
S10
laser pulses (t = 12 ns). According to Robertson et al.,11 the diffusion coefficient D [µm2/s] of DNA with the chain length L [µm] is obtained as following equation, D ∼ L-(0.589 ± 0.018) ....................................................................(1) . Using equation (1), we determined the diffusion coefficient D to be 0.19 ± 0.01 µm2/s whose value was approximately consistent with that reported in a literature.12 Thus, we obtained averaged diffusion length <x> in the interval t,
<x> = 2(Dt/π)1/2 .....................................................................(2) .
Using equation (2), <x> is approximately 50 pm. This value means that DNA hardly escaped from the plasmonic substrate during the period.
Photothermal effect during LSP excitation, however, induces thermal convection and thermophoresis, which complicate such molecular diffusion motion. In order to reveal the overall mechanism of LSP-OT, we should consider not only gradient force and molecular diffusion but also non-equilibrium thermodynamics induced by photothermal effect.
References (1) Gurrieri, S. Nucleic Acids Res. 1996, 24, 4759–4767. (2) Ichikawa, M.; Matsuzawa, Y.; Koyama, Y.; Yoshikawa, K. Langmuir 2003, 19, 5444–5447.
S11
(3) Tsuboi, Y.; Shoji, T.; Kitamura, N.; Takase, M.; Murakoshi, K.; Mizumoto, Y.; Ishihara, H. J. Phys. Chem. Lett. 2010, 1, 2327–2333. (4) Toshimitsu, M.; Matsumura, Y.; Shoji, T.; Kitamura, N.; Takase, M.; Murakoshi, K.; Yamauchi, H.; Ito, S.; Miyasaka, H.; Nobuhiro, A.; Mizumoto, Y.; Ishihara, H.; Tsuboi, Y. J. Phys. Chem. C 2012, 116, 14610–14618. (5) Ghenuche, P.; Cherukulappurath, S.; Taminiau, T. H.; Hulst, N. F. van; Quidant, R. Phys. Rev. Lett. 2008, 101, 116805. (6) Ito, S.; Sugiyama, T.; Toitani, N.; Katayama, G.; Miyasaka, H. J. Phys. Chem. B 2007, 111, 2365– 2371. (7) Baffou, G.; Quidant, R. Laser & Photonics Reviews 2012, 17, n/a–n/a. (8) Coronado, E. A.; Encina, E. R.; Stefani, F. D. Nanoscale 2011, 3, 4042–4059. (9) Takase, M.; Nabika, H.; Hoshina, S.; Nara, M.; Komeda, K.-I.; Shito, R.; Yasuda, S.; Murakoshi, K. Phys. Chem. Chem. Phys. 2013, 15, 4270–4274. (10) Quake, S. R.; Babcock, H.; Chu, S. Nature 1997, 388, 151–154. (11) Robertson, R. M.; Laib, S.; Smith, D. E. Proc. Natl. Acad. Sci. 2006, 103, 7310–7314. (12) Pluen, A.; Netti, P. A.; Jain, R. K.; Berk, D. A. Biophys. J. 1999, 77, 542–552.
S12
Permanently Fixing or Reversible Trap-and-Release of DNA Micropatterns on a Gold Nanostructure using Continuous-wave or Femtosecond Pulsed Near-Infrared Laser Light.
Tatsuya Shoji,† Junki Saitoh,‡ Noboru Kitamura,†,‡ Fumika Nagasawa,‡ Kei Murakoshi,†,‡ Hiroaki Yamauchi,¶ Syoji Ito,¶ Hiroshi Miyasaka,¶ Hajime Ishihara§ and Yasuyuki Tsuboi1,†,‡,||,*
†
Department of Chemistry, Graduate School of Science, Hokkaido University, Sapporo 060-0810, Japan
‡
Graduate School of Chemical Sciences and Engineering, Hokkaido University, Sapporo 060-0810,
Japan ¶
Division of Frontier Materials Science, Graduate School of Engineering Science, and Center for
Quantum Materials Science under Extreme Conditions, Osaka University, 1-3 Macikaneyama-cho, Toyonaka, Osaka 560-8531, Japan §
Department of Physics and Electronics, Osaka Prefecture University, 1-1, Gakuen-cho, Nakaku, Sakai,
Osaka 599-8531, Japan ||
JST, PRESTO, Japan.
* Corresponding Author e-mail: [email protected] (Yasuyuki Tsuboi)
S1
1. Optical setup and sample preparation For samples, we used λ-DNA (48.5 kbp, Nippon gene) stained with YOYO-1 (λabs = 490 nm, λem = 509 nm, Invitrogen Co. Ltd.) in a Tris-borate-EDTA buffer solution (45 mmol/l Tris base, 45 mmol/l boric acid, 1 mmol/l EDTA, pH 8.0). The concentration of the λ-DNA in the solution was 2.0 x 10-6 mol/l in nucleotide concentration. To suppress photobleaching of YOYO-1, glucose (30 mmol/l), glucose oxidase (1 mg/ml) and catalase (1 mg/ml) were added to the solution in the usual manner.1,2 For the plasmonic substrate, we fabricated gold nano-pyramidal-dimer arrays on a glass substrate by means of angular-resolved nanosphere lithography (AR-NSL substrate). We have already described the structure and optical properties of the AR-NSL substrate in detail elsewhere.3,4 It should be noted here that the substrate has a broad absorption band in the near-infrared (NIR) region (≥ 700 nm), which corresponds to the resonant LSP excitation. For LSP excitation, we used a cw NIR laser (λ = 808 nm, Spectra Physics, CWA0400-SXD-808-30-E) or a fs NIR laser (λ = 770 nm, repetition rate 80 MHz, pulse width 120 fs in FWFM, Toptica photonics AG, FemtoFiber pro). Both lasers can excite LSPs on the substrate around the absorption maximum. The absorbance of AR-NSL substrate at 770 nm is approximately same as that at 808 nm. We obtained the fluorescence micrographs using Hg lamp illumination, while bright-field micrographs were obtained as green colored background images because of S2
a bandpass filter. For the fluorescence microspectroscopy, the YOYO-1 in the DNA was excited with a cw visible Ar+ laser (λ = 488 nm, 5 – 10 µW (2.5 – 5.0 kW/cm2), Ion Laser Technology, 5490ASL). In such laser intensity of Ar+ laser, DNA was never trapped by only Ar+ laser irradiation on a plasmonic substrate. The cw and fs NIR lasers were loosely focused to a circular spot (diameter d ∼ 5 µm) on the AR-NSL substrate, while the visible laser was tightly focused approximately at the diffraction limit (d ∼ 0.5 µm) at the center of the circular spot. Accordingly, we observed fluorescence only from the center of the LSP excitation area. All the experiments were performed under ambient conditions at room temperature.
2. cw NIR intensity dependence on DNA assembly on an AR-NSL substrate We carried out cw NIR intensity dependence on DNA assembly formation on an AR-NSL substrate as shown in Fig. S1. Figure S1(a) and (b) shows a series of optical micrographs of representative DNA trapping behavior with cw NIR laser irradiation at (a) 4.0 and (b) 5.0 kW/cm2. At 5.0 kW/cm2 (Figure S1(b)), we did not observe anything during several tens of seconds from LSP excitation. During several minutes of LSP excitation, a micro-ring structure was formed and gradually grew in the LSP excitation area. The diameter of the micro-ring is slightly smaller than that of the LSP excitation S3
area. Even after switching off the LSP excitation the micro-ring remained stable on the substrate, keeping its shape and size. On the other hand, below 4.0 kW/cm2 (Figure S1(a)), however, we observed neither such micro-ring structure nor sign of assembly formation. Figure S1(c) shows cw NIR laser intensity dependence on the fluorescence intensity that was measured at the center of LSP excitation area. The fluorescence intensity decreased with increasing cw NIR laser intensity, suggesting that DNA was never trapped at the center of cw NIR laser light even though cw NIR laser intensity decreased to reduce photothermal effect. Furthermore, these results supported that the trap-and-release of DNA was achieved only with fs NIR pulsed-laser light. Pulsed-excitation would prevent DNA from being fixed on the substrate due to the intervals between the pulses.
S4
a 0s
LSP 1 min on
3 min
5 min
1 min
3 min
5 min
LSP 6 min off
5 µm
b 0s
LSP on
LSP off
6 min
5 µm c
Fluorescence intensity
1.0
0.5
0.0
0
5
10
cw NIR laser intensity / kWcm-2
Figure S1. Bright-field micrographs of DNA on a plasmonic AR-NSL gold substrate using cw NIR laser irradiation ((a) 4.0, (b) 5.0 kW/cm2) for LSP excitation. The time shown in each image is the cw laser irradiation time, with 0 s being just before switching on the laser. The cw laser irradiation was turned off at 5 min. The cw laser illumination covered the white dotted circle. (c) Relative fluorescence intensity measured at the center of LSP excitation are as a function of cw NIR laser intensity.
S5
3. Nonlinear optical effect on fluorescence behaviors In order to show no fluorescence emission of YOYO-1 in DNA via 2-photon absorption, we measured the fluorescence on a plasmonic substrate excited by cw and fs NIR laser light. As seen in the Fig.S2, the fluorescence of YOYO-1 was not completely observed in both of (a) cw and (b) fs NIR laser irradiation, although baselines were slightly rising. We quantified the intensity in the spectra under cw and fs illumination, respectively. The slight increase is negligible for fluorescence measurements of LSP-OT of DNA, because the intensity of the rising baseline was quite smaller than that of fluorescence from YOYO-1; fluorescence: raised baseline = 1: < 0.01 under cw NIR laser irradiation and 1: < 0.002 under fs NIR laser irradiation, respectively. The rising baseline would be ascribed to luminescence of the AR-NSL substrate via 2-photon absorption (two-photon luminescence, TPL).5 Thus, we safely concluded that YOYO-1 in DNA was not excited by cw and fs NIR laser irradiation.
S6
a
b 0.4
Fluorescence intensity
Fluorescence intensity
cw NIR off cw NIR on
0.2
0.0 500
550
Wavelength/ nm
600
fs NIR off fs NIR on
0.004
0.002
0.000 500
550
Wavelength/ nm
600
Figure S2. Fluorescence spectra of YOYO-1 in DNA on a plasmonic substrate excited by (a) cw and (b) fs NIR laser light (INIR ∼ 7.0 kW/cm2) without visible laser irradiation. Fluorescence spectra were acquired with the exposure time of 10 s.
Furthermore, we examined TPL measurement of YOYO-1 in DNA upon simultaneous NIR and visible laser irradiation without a plasmonic substrate. In Fig. S3, the black line shows fluorescence spectrum of YOYO-1 in DNA during visible laser irradiation, while the red dotted line shows the spectrum during visible and fs NIR laser irradiation. Comparing these spectra, TPL of YOYO-1 in DNA was not observed during these laser irradiations. Thus, TPL of YOYO-1 in DNA was never observed in our experimental system.
S7
Fluorescence intensity 500
550
Wavelength/ nm
600
Figure S3. Fluorescence spectra of YOYO-1 in DNA without a plasmonic substrate excited by visible (black) and visible+fs lasers (INIR ∼ 7.0 kW/cm2) (red dotted line). Fluorescence spectra were acquired with the exposure time of 10 s.
4. FCS measurement for estimation of temperature elevation during fs NIR laser irradiation Using fluorescence correlation spectroscopy (FCS), we determined the local temperature elevation (∆T) in the vicinity of AR-NSL surface (5 μm distance from the surface) during fs NIR laser irradiation. Fig. S4 shows a representative example of FCS. The fundamental experimental method and analysis were described in the previous literatures.4,6 In Fig. S4(a), autocorrelation function traces are shown as a function of fs NIR laser intensity (λ = 800 nm). An analysis of the trace provides a diffusion time of the dye molecules (rhodamine 123) (Fig. S4(b)). The diffusion time can be converted
S8
into ΔT based on the molecular diffusion equation. In Fig. S4(c), ∆T is plotted against fs laser intensity. A linear fs-laser intensity-dependence was clearly observed below 5.0 kW/cm2. Using this relation, we can determine the temperature elevation at the surface of gold nanostructure upon LSP excitation using a method that is same as in our previous work (extrapolation based on a quasi-1D-steady state thermal conduction model)4 and it was evaluated to be 7.2 K at 1.0 kW/cm2. Considering the extinction coefficient ratio of AR-NSL substrate at 800 and 770 nm, ∆T is estimated to be 7-8 K at 1.0 kW/cm2 of 770 nm fs laser. These values are appropriate in comparison with the previous literatures on the ∆T on the surface of AR-NSL or other plasmonic substrates.7– 9
S9
c
a
b
Figure S4. FCS for the evaluation of temperature rise upon LSP excitation with fs NIR laser. The detecting position is 5 μm distance from the gold surface. (a) Dependence of the autocorrelation curve on the power of the incident fs laser light. (b) Averaged diffusion time of the probe dye in the observation volume as a function of excitation intensity of LSP. (c) Temperature at the focal point of the NIR laser light as a function of the incident laser power for the solution.
5. Molecular diffusion of DNA during the interval between two adjacent fs laser pulses Considering molecular diffusion, we should estimate the diffusive distance of λ-DNA (chain length L = 16.4 µm)2,10 during the interval time t between two adjacent fs
S10
laser pulses (t = 12 ns). According to Robertson et al.,11 the diffusion coefficient D [µm2/s] of DNA with the chain length L [µm] is obtained as following equation, D ∼ L-(0.589 ± 0.018) ....................................................................(1) . Using equation (1), we determined the diffusion coefficient D to be 0.19 ± 0.01 µm2/s whose value was approximately consistent with that reported in a literature.12 Thus, we obtained averaged diffusion length <x> in the interval t,
<x> = 2(Dt/π)1/2 .....................................................................(2) .
Using equation (2), <x> is approximately 50 pm. This value means that DNA hardly escaped from the plasmonic substrate during the period.
Photothermal effect during LSP excitation, however, induces thermal convection and thermophoresis, which complicate such molecular diffusion motion. In order to reveal the overall mechanism of LSP-OT, we should consider not only gradient force and molecular diffusion but also non-equilibrium thermodynamics induced by photothermal effect.
References (1) Gurrieri, S. Nucleic Acids Res. 1996, 24, 4759–4767. (2) Ichikawa, M.; Matsuzawa, Y.; Koyama, Y.; Yoshikawa, K. Langmuir 2003, 19, 5444–5447.
S11
(3) Tsuboi, Y.; Shoji, T.; Kitamura, N.; Takase, M.; Murakoshi, K.; Mizumoto, Y.; Ishihara, H. J. Phys. Chem. Lett. 2010, 1, 2327–2333. (4) Toshimitsu, M.; Matsumura, Y.; Shoji, T.; Kitamura, N.; Takase, M.; Murakoshi, K.; Yamauchi, H.; Ito, S.; Miyasaka, H.; Nobuhiro, A.; Mizumoto, Y.; Ishihara, H.; Tsuboi, Y. J. Phys. Chem. C 2012, 116, 14610–14618. (5) Ghenuche, P.; Cherukulappurath, S.; Taminiau, T. H.; Hulst, N. F. van; Quidant, R. Phys. Rev. Lett. 2008, 101, 116805. (6) Ito, S.; Sugiyama, T.; Toitani, N.; Katayama, G.; Miyasaka, H. J. Phys. Chem. B 2007, 111, 2365– 2371. (7) Baffou, G.; Quidant, R. Laser & Photonics Reviews 2012, 17, n/a–n/a. (8) Coronado, E. A.; Encina, E. R.; Stefani, F. D. Nanoscale 2011, 3, 4042–4059. (9) Takase, M.; Nabika, H.; Hoshina, S.; Nara, M.; Komeda, K.-I.; Shito, R.; Yasuda, S.; Murakoshi, K. Phys. Chem. Chem. Phys. 2013, 15, 4270–4274. (10) Quake, S. R.; Babcock, H.; Chu, S. Nature 1997, 388, 151–154. (11) Robertson, R. M.; Laib, S.; Smith, D. E. Proc. Natl. Acad. Sci. 2006, 103, 7310–7314. (12) Pluen, A.; Netti, P. A.; Jain, R. K.; Berk, D. A. Biophys. J. 1999, 77, 542–552.
S12
