Scanning Evanescent Fields Using a pointlike Light Source and a

Dec 9, 2011 - field. Recording the emission of the quantum dot within the evanescent field as well as under homogeneous illumination allows one to dir...
0 downloads 0 Views 1MB Size
Letter pubs.acs.org/NanoLett

Scanning Evanescent Fields Using a pointlike Light Source and a Nanomechanical DNA Gear Hergen Brutzer,†,‡ Friedrich W. Schwarz,†,‡ and Ralf Seidel*,† †

Biotechnology Center, Technische Universität Dresden, Dresden 01062, Germany S Supporting Information *

ABSTRACT: The characterization of three-dimensional inhomogeneous illumination fields is a challenge in modern microscopy. Here we use a four-arm DNA junction as a nanomechanical translation stage to move a single fluorescent quantum dot through an exponentially decaying evanescent field. Recording the emission of the quantum dot within the evanescent field as well as under homogeneous illumination allows one to directly obtain the intensity distribution of the excitation field without additional deconvolution. Our method will allow the characterization of a broad range of illumination fields and to study near-field effects between small optical probes. KEYWORDS: Evanescent field, TIRF microscopy, point spread function, magnetic tweezers, DNA, Holliday junction

T

in the object, their extinction coefficients, and quantum efficiencies, one obtains

he past two decades have boosted a broad range of techniques in modern fluorescence microscopy that use sophisticated illumination schemes and geometries. This includes confocal techniques such as fluorescence correlation spectroscopy,1 two-photon laser scanning fluorescence microscopy,2 and super-resolution stimulated emission depletion microscopy.3 In addition, in widefield microscopy the use of inhomogeneous illumination profiles has gained increasing popularity and promoted the development as well as the application of total internal reflection fluorescence (TIRF) microscopy,4−6 fluorescence interference contrast microscopy,7,8 surface plasmon assisted optics,9 super-resolution microscopy using layered metallic structures,10 and structured illumination microscopy.11,12 The unifying principle of such excitation schemes is a narrowed excitation volume in one or more dimensions to increase the signal-to-noise ratio by orders of magnitudes and in part to even bypass the Abbe/Rayleigh-limit for the spatial resolution. When applying nonhomogeneous illumination often knowledge about the exact shape of the applied illumination profile is essential, for example, for quantitative data analysis,1,13 threedimensional (3D) image reconstructions in microscopy,3,12 and absolute distance measurements.8,14 The characterization of excitation profiles in all three dimensions is up to now still a challenging task when using microscopy in particular for the axial direction. It is mainly limited by the fact that the measured intensity Im emitted from a probe results from a convolution of its emitted signal Iem(x,y,z) with the three-dimensional point spread function (PSF) of the optical measurement system.13,15 Typically one can assume proportionality between Iem(x,y,z) and the excitation intensity Iex(x,y,z) at a given position. Using the fluorescent object function F(x,y,z) as proportionality factor, which describes the distribution of fluorescent molecules © 2011 American Chemical Society

Im = Iem ⊗ PSF = [IexF ] ⊗ PSF (1) To extract the distribution of the excitation light from the measured intensity requires therefore a deconvolution of eq 1, which, for example, has been applied previously to characterize evanescent fields.15,16 This demands precise knowledge of the fluorescent object function and most problematic of the PSF, which is an intrinsic characteristic of each experimental setup. In particular, the PSF variation along the optical axis is difficult to obtain, since it arises from a combination of an effective reduction of the numerical aperture for image planes that are out of focus14 and a decreased near-field coupling of the emission light into the glass substrate.17 In order to circumvent an elaborate deconvolution, one can, however, use a single pointlike light source, such as a small fluorescent molecule or a quantum dot, instead of an extended fluorescent object. In this case F(x,y,z) becomes proportional to a delta-function such that the measured intensity becomes proportional to the product of the excitation intensity and the PSF at the position of the light source (Im ∼ IexPSF). For conventional widefield (CW), that is, for a homogeneous illumination, the measured intensity of a pointlike light source becomes even directly proportional to the PSF of the optical system (ImCW ∼ PSF). Combining the last two expressions for an inhomogeneous and a homogeneous excitation field one obtains Im ∼ IexImCW and with this Iex ∼ Im/ImCW. Thus, the relative distribution of the inhomogeneous illumination can be Received: November 3, 2011 Revised: December 7, 2011 Published: December 9, 2011 473

dx.doi.org/10.1021/nl203876w | Nano Lett. 2012, 12, 473−478

Nano Letters

Letter

obtained by scanning the excitation volume with the pointlike light source and recording its intensity under inhomogeneous as well as conventional widefield illumination. The deconvolution is then readily obtained via the ratio of both intensities. This conceptually simple approach is experimentally difficult to realize, since the emitter has to be attached to an external translation stage for scanning the illumination field. Previously this was realized by using an atomic force microscope to linearly translate a fluorescent quantum dot that had been attached to the tip of the silicon nitride cantilever of the device.18,19 The attachment of the fluorescent probe to a solidstate surface, however, is itself problematic, because it influences both the distribution of the excitation light as well as the emission from the probe by introducing an additional large boundary with differing optical properties. Here we introduce a new method to linearly translate a single pointlike light source in free solution, that is, practically in the absence of a disturbing interface, in order to probe the intensity distribution of an inhomogeneous illumination in axial direction. For this, we attach a fluorescent probe to an extended piece of double-strand DNA and use the DNA itself as a linear translation stage with nanometer precision. Because of its composition and size, double-strand DNA represents a minimal disturbance. It negligibly scatters and absorbs visible light20 and does not significantly quench the emission from a nearby emitter by energy transfer.21 As a proof of principle, we use our method to scan the evanescent fields resulting from sample illumination in TIRF geometry with a quantum dot as a pointlike light source (Figure 1). The central piece of our DNA-based device is a so-called Holliday junction with homologous arms, which is a central intermediate of double-strand break repair by homologous recombination in prokaryotes and eukaryotes22 and transports crossovers between neighboring chromosome pairs over large distances. It is a four-arm DNA junction in which the opposing arms possess identical sequences with respect to the junction center. In the absence of external constraints, this junction is mobile. Thus, one pair of homologous arms can expand at the expense of the other pair. Since DNA has a helical structure, branch migration causes the arms to twist with one turn moved per helical pitch (of ∼3.6 nm). Vice versa branch migration can directly be induced and controlled by adjusting the DNA twist as demonstrated in recent single-molecule measurements using magnetic23 and optical tweezers.24 We adapted this single-molecule branch migration assay to build a translation stage for a fluorescent probe. We constructed an 8.9 kbp (3000 nm) linear DNA molecule (see Supporting Information for a detailed preparation procedure of the DNA construct) of which the initial 4.4 kbp (1500 nm) are composed of two 2.2 kbp long fragments of identical sequence that are arranged in an inverted orientation to each other in order to form a long palindrome (Figure 1a). From the center of this palindromic region a Holliday junction can be extruded by applying external twist (see below).25 The remaining 4.5 kbp stretch that serves as a spacer is labeled internally26,27 7 bp (2.4 nm) upstream of the inverted repeat with two biotin molecules that are 11 bp apart. The biotin modifications serve as attachment sites for the fluorescent probe for which we use a streptavidin-coated quantum dot (Qdot 625, Invitrogen). The 8.9 kbp molecule carries two additional ∼600 bp fragments at its ends that comprise either multiple digoxigenin- (for the end at the repeat region) or fluorescein-labeled bases (for the other end).

Figure 1. Experimental scheme. (a) Used DNA construct consisting of an ∼600 bp long fluorescein labeled tail (Fluo) followed by a 4.5 kbp spacer region, a 4.4 kbp inverted repeat (orientation of the two 2.2 kbp regions with identical sequence indicated by the arrows), and an ∼600 bp long digoxigenin labeled tail (Dig). Two internal biotin modifications (Bio) are introduced upstream of the inverted repeat.26,27 A detailed procedure for the preparation of the construct is given in the Supporting Information. (b, left) A streptavidin conjugated quantum dot is bound to the internal label of the DNA and stretched in a magnetic tweezers setup. (b, middle) After inducing 600 negative turns in the DNA, a Holliday junction is spontaneously extruded where the branches cover the full length of the inverted repeat (Supporting Information, Figure S2). (b, right) Adding sufficient positive turns drives branch migration of the junction and the quantum dot is lifted from its position proximal to the surface upward by about the length of the helical pitch per turn. Monitoring the emission of the quantum dot with an EMCCD camera during branch migration allows thus the scanning of the excitation field distribution in axial direction, for example, of an evanescent field resulting from TIRF illumination as shown here.

To scan evanescent fields, we incubated the DNA construct with an excess of quantum dots and subsequently with 1 μm antifluorescein coated magnetic beads (MyOne, Invitrogen) in order to allow their binding to the biotin and fluorescein modifications of the molecule, respectively. The construct was flushed into the fluidic cell of a magnetic tweezers apparatus,28,29 in which the digoxigenin-labeled DNA end was allowed to attach to the antidigoxigenin-coated bottom. A pair of permanent magnets (W-05-N50-G, Supermagnete) was used to generate a force on the magnetic bead and to stretch the DNA construct (Figure 1b, left). The length of the DNA, that is, the vertical position of the magnetic bead with respect to a nonmagnetic reference bead attached to the surface of the fluidic cell, was determined from videoimages using a Pulnix TM-6710CL CCD camera and real-time 3D particle tracking with sub-nm accuracy.30,31 The magnetic tweezers setup was additionally equipped with a sensitive EMCCD camera (iXon DU897-COO-BV, Andor Technology) for fluorescence detection and a 488 nm laser (Sapphire 488-50, Coherent) that allowed illumination of the sample both in conventional widefield and TIRF geometry through the high numerical aperture objective of the setup (Nikon, CFI Apo TIRF 100×, NA = 1.49) (Supporting Information, Figure S1). The considerable anisotropy of the used magnetic beads30 allows the twisting of the DNA molecule by rotating the magnet configuration.32 To start an experiment we first applied at a force of ∼3 pN ∼600 negative turns with respect to the 474

dx.doi.org/10.1021/nl203876w | Nano Lett. 2012, 12, 473−478

Nano Letters

Letter

throughout the measurement, the distance change of the quantum dot to the substrate surface can be directly calculated from the length change of the DNA (Figure 2). Fitting the DNA extension versus added turns with a linear relation provides a slope of 3.5 nm. This is slightly less than the actual helical pitch, since the DNA is only extended to 92% of its full contour length at a force of 3 pN due to thermal fluctuations. These fast fluctuations cause also the quantum dot to fluctuate around its average axial position with a root-mean-square displacement of ∼2.5 nm for a surface distance of 100 nm at 3 pN force.30 Nonetheless, it shows that the Holliday junction allows one to linearly couple macroscopic rotations into translation on the nanometer scale, where the DNA pitch even serves as an intrinsic ruler. The effects of incomplete stretching can be easily compensated by using the well characterized force− extension behavior of DNA.33 In order to quantitatively evaluate the detected fluorescence emission, the images of the quantum dot were fit with a symmetric two-dimensional Gaussian function whose volume provides the detected intensity. While it remains almost constant before branch migration, it decreases exponentially with the quantum dot distance from the substrate (Figure 2). The measured intensity decay with distance does not directly reflect the intensity distribution of the evanescent field. This is due to the changing PSF when the quantum dot moves out of focus. As detailed above, the intensity decay of the evanescent field can, however, be obtained by dividing the recorded signal in TIRF illumination by the one recorded in conventional widefield illumination, because a quantum dot is a pointlike light source. The signal recorded in conventional widefield illumination is shown in Figure 3. The intensity decay with

helicity of the DNA (Figure 1b, left). Subsequently, the force was lowered in order to stimulate the spontaneous extrusion of the Holliday junction from the linear inverted repeat.25 Junction extrusion is seen as a DNA shortening that covers the full length of the repeat at elevated force (Figure 1b, middle and Supporting Information, Figure S2). In this configuration, the quantum dot is in close proximity to the surface of the fluidic cell. We then started to slowly add positive turns at 2 Hz while simultaneously recording DNA length and fluorescent images of the quantum dot under TIRF illumination at 120 and 10 Hz, respectively (Figure 2). The objective was positioned such that

Figure 2. Linear translation of a quantum dot through an evanescent field using branch migration of a Holliday junction. The evanescent field was generated by illuminating the sample in TIRF geometry. Shown are the time courses of applied turns, DNA length and detected fluorescence emission intensity of the quantum dot. After extrusion of the Holliday junction (see Figure 1), where the quantum dot is in a proximal position to the surface (see sketch), positive turns were added to the DNA at a rate of 2 Hz. As soon as excess turns are removed (marked by a vertical dashed line, see text), branch migration occurs that is seen as a DNA length increase with ∼3.5 nm per turn and a simultaneous decrease of the quantum dot emission. Snapshot images of the quantum dot are shown for selected times indicated by the dotted lines. As the spacing between the magnetic bead and the quantum dot remains unchanged, the distance of the quantum dot to the surface can be calculated from the DNA length by subtracting an offset of ∼1.5 μm. The DNA length was measured at a frequency of 120 Hz, and fluorescence images were acquired at 10 Hz. A constant force of 3 pN was applied throughout the experiment. The angle of incidence is 70.5°.

Figure 3. Intensity decay measured for conventional widefield illumination (black curve). Experimental conditions are as in Figure 2. The change in intensity is due to the change of the point spread function when the quantum dot moves out of focus. The signal recorded in TIRF illumination (from Figure 2) is depicted in gray. The onset of branch migration is indicated by the vertical dashed line.

distance is much smaller than the one recorded in TIRF illumination, indicating that the correction due to the changing PSF is relatively small. We finally obtain the normalized axial intensity distribution of the evanescent field Ie̅ x(z) from the ratio of the measured intensities for TIRF Im̅ (z) and conventional widefield illumination ImCW ̅ (z) that were previously normalized by the corresponding intensities obtained for the proximal position of the quantum dot to the surface Im(0) and ImCW(0)

its focus in the fluorescence channel coincided with the bottom of the fluidic cell. First the excess of negative turns was removed while the DNA length remained constant, since only ∼420 turns can be stored in the arms of the Holliday junction according to the number of base-pairs forming the inverted repeat. Subsequently, the onset of branch migration (indicated by the dashed line in Figure 2) could be seen as a linear increase of the DNA length with the number of turns added. The DNA lengthening caused also the quantum dot to decrease its emission (Figure 2) because it moves together with the magnetic bead away from the surface and thus out of the illumination field (Figure 1c). As the average distance between the magnetic bead and the quantum dot remains unchanged

Iex̅ (z) =

475

Im(z) Im(0) CW (z) Im CW Im (0)

=

Im̅ (z) CW Im ̅ (z )

(2) dx.doi.org/10.1021/nl203876w | Nano Lett. 2012, 12, 473−478

Nano Letters

Letter

Figure 4. Dependence of the excitation intensity profiles in TIRF illumination on the angle of incidence of the laser. (a) Normalized excitation intensity profiles as function of surface distance obtained for different angles of incidence. Data was obtained as described in the text by correcting the measured intensity profiles in TIRF illumination for the variation of the point spread function. Red lines are fits to the data with a single exponential function according to eq 2. Periods of occasional, sudden intensity drops due to quantum dot blinking were removed from the profiles. (b) Dependence of the penetration depth, that is, the mean length of the exponential intensity decay of the evanescent excitation field, on the angle of incidence (filled symbols). Each color-symbol combination corresponds to data taken for an individual quantum dot-DNA construct. Open symbols represent the corresponding decay lengths of the measured, that is, uncorrected, intensity profiles, which are shown for comparison. The theoretically expected dependence of the penetration depth according to eq 3 is shown as a solid black line, the critical angle for total internal reflection as a gray dashed line.

experimentally obtained and the theoretical predicted penetration depths. In summary, we showed here how a nanomechanical DNA gear can be conveniently used as a linear nanometer-precise translation stage for a small fluorescent probe to axially scan intensity distributions of inhomogeneous illumination profiles. Using DNA as a carrier for a fluorescent probe provides minimal optical interference. In addition, the DNA helical pitch serves as an intrinsic ruler by coupling one macroscopic rotation to about 3.5 nm linear translation. The determination of illumination profiles with our method does not require additional knowledge of the optical properties of the microscope, because the deconvolution correction is directly obtained from recordings in conventional widefield illumination. It is therefore potentially less error-prone compared to the deconvolution of microscopy images of extended objects. For example, field distributions for TIRF illumination following a double exponential dependency have been obtained from imaging 9 μm-sized fluorescent beads,16 while other studies reported a single exponential relation15,18 in agreement with our results and the prediction from classical electrodynamics.34 Recently, an alternative, deconvolution-free technique that allows to determine absolute intensity distributions of laser foci has been introduced. It is not microscopy-based and relies on ionic current measurements through a nanopore.36 This technique is, however, limited to rather large laser powers and confocal setups, while the present measurements have been obtained at excitation intensities, which are typically used for detection of single-molecule fluorescence. Beyond characterizing the electromagnetic field distribution in a TIRF microscope that is important for various applications,14,18 our method should be a versatile tool to experimentally characterize other intensity distributions, that is, near metal surfaces,9 metallic multilayers,10 and in structured illumination microscopy.11 Mounting the fluidic cell on an XY-nanopositioning system will enable the three-dimensional scanning of intensity distributions. Our measurement scheme can also be used to experimentally determine the threedimensional PSF of an optical system as function of distance from the typically present glass interface, which can currently only be obtained from theoretical calculations.13

For noise reduction the widefield data was smoothed to 1 Hz prior division. Figure 4a shows the obtained normalized intensity profiles of the evanescent field recorded for the same quantum dot at different angles of incidence θ of the laser, which were adjusted by displacing the laser beam from the optical axis of the objective (Supporting Information, Figure S3). As expected, as the angle of incidence increases, the evanescent field becomes shallower. For each angle of incidence, the excitation intensity profile can be well fitted by a single exponential function (red lines in Figure 4a)

⎡ z ⎤ Iex̅ (z , θ) = exp⎢ − ⎥ ⎣ d (θ ) ⎦

(3)

from which the penetration depth d(θ) of the evanescent field is obtained. The obtained dependence of the penetration depth on the angle of incidence is shown in Figure 4b (filled symbols). Experiments for different DNA constructs demonstrate the reproducibility of the measurements. The theoretical dependence of the penetration depth on the angle of incidence according to classical electrodynamics is given by34,35

λ

d (θ ) = 4π

n12

sin 2 θ − n22

(4)

where λ is the wavelength of the incident light and n1 > n2 are the indices of refraction of the media at the interface where the light is totally internally reflected. As we used an oil immersion objective, n1 corresponds to the index of refraction of the immersion oil (Immersol 518F immersion oil, Zeiss) with n1 = 1.5229 at 488 nm, and n2 to the index of refraction of the buffer (n2 = 1.33). Using these values, the theoretical dependence of the penetration depth on the angle of incidence reproduces well the experimentally obtained values (filled symbols in Figure 4b). We attribute the slight increase of the measured penetration depth at larger angles to the fact that optical abberations become more important close to the maximum angle of incidence, which is 78° as calculated from the numerial aperture of the objective. Fitting the uncorrected measured intensity profiles (as shown in Figure 2) with an exponential relation results in decay lengths (open symbols in Figure 4b) that are only slightly smaller than the 476

dx.doi.org/10.1021/nl203876w | Nano Lett. 2012, 12, 473−478

Nano Letters

Letter

(11) Mertz, J. Optical sectioning microscopy with planar or structured illumination. Nat. Methods 2011, 8, 811−819. (12) Lemmer, P.; Gunkel, M.; Baddeley, D.; Kaufmann, R.; Urich, A.; Weiland, Y.; Reymann, J.; Muller, P.; Hausmann, M.; Cremer, C. SPDM: light microscopy with single-molecule resolution at the nanoscale. Appl. Phys. B 2008, 93, 1−12. (13) Nasse, M. J.; Woehl, J. C. Realistic modeling of the illumination point spread function in confocal scanning optical microscopy. J. Opt. Soc. Am. A 2010, 27, 295−302. (14) Singh-Zocchi, M.; Dixit, S.; Ivanov, V.; Zocchi, G. Singlemolecule detection of DNA hybridization. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 7605−7610. (15) Gell, C.; Berndt, M.; Enderlein, J.; Diez, S. TIRF microscopy evanescent field calibration using tilted fluorescent microtubules. J. Microsc. 2009, 234, 38−46. (16) Mattheyses, A. L.; Axelrod, D. Direct measurement of the evanescent field profile produced by objective-based total internal reflection fluorescence. J. Biomed. Opt. 2006, 11, 014006. (17) Enderlein, J.; Bö hmer, M. Influence of interface-dipole interactions on the efficiency of fluorescence light collection near surfaces. Opt. Lett. 2003, 28, 941−943. (18) Sarkar, A.; Robertson, R. B.; Fernandez, J. M. Simultaneous atomic force microscope and fluorescence measurements of protein unfolding using a calibrated evanescent wave. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 12882−12886. (19) Puchner, E. M.; Kufer, S. K.; Strackharn, M.; Stahl, S. W.; Gaub, H. E. Nanoparticle Self-Assembly on a DNA-Scaffold Written by Single-Molecule Cut-and-Paste. Nano Lett. 2008, 8, 3692−3695. (20) Sutherland, J. C.; Griffin, K. P. Absorption spectrum of DNA for wavelengths greater than 300 nm. Radiat. Res. 1981, 86, 399−409. (21) Sindbert, S.; Kalinin, S.; Nguyen, H.; Kienzler, A.; Clima, L.; Bannwarth, W.; Appel, B.; Müller, S.; Seidel, C. A. M. Accurate Distance Determination of Nucleic Acids via Förster Resonance Energy Transfer: Implications of Dye Linker Length and Rigidity. J. Am. Chem. Soc. 2011, 133, 2463−2480. (22) Sigal, N.; Alberts, B. Genetic recombination: the nature of a crossed strand-exchange between two homologous DNA molecules. J. Mol. Biol. 1972, 71, 789−793. (23) Dawid, A.; Guillemot, F.; Brème, C.; Croquette, V.; Heslot, F. Mechanically controlled DNA extrusion from a palindromic sequence by single molecule micromanipulation. Phys. Rev. Lett. 2006, 96, 188102. (24) Forth, S.; Deufel, C.; Patel, S. S.; Wang, M. D. Direct measurements of torque during Holliday junction migration. Biophys. J. 2011, 101, L5−L7. (25) Ramreddy, T.; Sachidanandam, R.; Strick, T. R. Real-time detection of cruciform extrusion by single-molecule DNA nanomanipulation. Nucleic Acids Res. 2011, 39, 4275−4283. (26) Luzzietti, N.; Brutzer, H.; Klaue, D.; Schwarz, F. W.; Staroske, W.; Clausing, S.; Seidel, R. Efficient preparation of internally modified single-molecule constructs using nicking enzymes. Nucleic Acids Res. 2011, 39, e15. (27) Luzzietti, N.; Knappe, S.; Richter, I.; Seidel, R. Nicking enzymebased internal labelling of DNA at multiple loci. Nat. Protoc. 2012. (28) Revyakin, A.; Ebright, R. H.; Strick, T. R. Single-molecule DNA nanomanipulation: improved resolution through use of shorter DNA fragments. Nat. Methods 2005, 2, 127−138. (29) Kauert, D. J.; Kurth, T.; Liedl, T.; Seidel, R. Direct Mechanical Measurements Reveal the Material Properties of Three-Dimensional DNA Origami. Nano Lett. 2011, 11, 5558−5563. (30) Klaue, D.; Seidel, R. Torsional stiffness of single superparamagnetic microspheres in an external magnetic field. Phys. Rev. Lett. 2009, 102, 028302. (31) Otto, O.; Czerwinski, F.; Gornall, J. L.; Stober, G.; Oddershede, L. B.; Seidel, R.; Keyser, U. F. Real-time particle tracking at 10,000 fps using optical fiber illumination. Opt. Express 2010, 18, 22722−22733. (32) Maffeo, C.; Schöpflin, R.; Brutzer, H.; Stehr, R.; Aksimentiev, A.; Wedemann, G.; Seidel, R. DNA-DNA interactions in tight

In addition, our technique will allow one to directly investigate the distance-dependent near-field coupling between two optically active probes, for example, fluorescence resonance energy transfer (FRET) between a fluorescent donor and an acceptor molecule37 or the near-field enhancement of the emission of a fluorophore near a gold nanoparticle.38 The probes can be conveniently attached below and above the Holliday junction using the here applied internal labeling method.26,27 The distance between the two probes can then be changed continuously by branch migration while the emission is recorded simultaneously. We therefore think that our nanomechanical DNA gear will become a versatile tool for anchoring and moving small optically active probes with minimal optical interference.



ASSOCIATED CONTENT

S Supporting Information *

Additional information and figures. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]. Author Contributions ‡ These authors contributed equally to this paper.



ACKNOWLEDGMENTS This work was supported by Grant SE 1646/1-1 from the DFG, a starting grant from the European Research Council (No. 261224) and a NanoSci-E+ collaborative proposal (SE 1646/51) to R.S. as well as the Dresden International Graduate School for Biomedicine and Bioengineering, funded by the DFG, to F.W.S. We thank Anna Grushina for assistance with the measurement protocol.



REFERENCES

(1) Bacia, K.; Schwille, P. Practical guidelines for dual-color fluorescence cross-correlation spectroscopy. Nat. Protoc. 2007, 2, 2842−2856. (2) Denk, W.; Strickler, J. H.; Webb, W. W. Two-photon laser scanning fluorescence microscopy. Science 1990, 248, 73−76. (3) Klar, T. A.; Jakobs, S.; Dyba, M.; Egner, A.; Hell, S. W. Fluorescence microscopy with diffraction resolution barrier broken by stimulated emission. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 8206−8210. (4) Toomre, D.; Manstein, D. J. Lighting up the cell surface with evanescent wave microscopy. Trends Cell Biol. 2001, 11, 298−303. (5) Axelrod, D. Total Internal Reflection Fluorescence Microscopy in Cell Biology. Traffic 2001, 2, 764−774. (6) Tokunaga, M.; Kitamura, K.; Saito, K.; Iwane, A. H.; Yanagida, T. Single Molecule Imaging of Fluorophores and Enzymatic Reactions Achieved by Objective-Type Total Internal Reflection Fluorescence Microscopy. Biochem. Biophys. Res. Commun. 1997, 235, 47−53. (7) Ajo-Franklin, C. M.; Yoshina-Ishii, C.; Boxer, S. G. Probing the Structure of Supported Membranes and Tethered Oligonucleotides by Fluorescence Interference Contrast Microscopy. Langmuir 2005, 21, 4976−4983. (8) Kerssemakers, J.; Howard, J.; Hess, H.; Diez, S. The distance that kinesin-1 holds its cargo from the microtubule surface measured by fluorescence interference contrast microscopy. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 15812−15817. (9) Barnes, W. L.; Dereux, A.; Ebbesen, T. W. Surface plasmon subwavelength optics. Nature 2003, 424, 824−830. (10) Elsayad, K.; Heinze, K. G. Defining a superlens operating regime for imaging fluorescent molecules. PLoS One 2009, 4, e7963. 477

dx.doi.org/10.1021/nl203876w | Nano Lett. 2012, 12, 473−478

Nano Letters

Letter

supercoils are described by a small effective charge density. Phys. Rev. Lett. 2010, 105, 158101. (33) Bouchiat, C.; Wang, M. D.; Allemand, J.; Strick, T.; Block, S. M.; Croquette, V. Estimating the persistence length of a worm-like chain molecule from force-extension measurements. Biophys. J. 1999, 76, 409−413. (34) Jackson, J. D. Classical Electrodynamics, 3rd ed.; John Wiley & Sons, Inc.: New York, 1999; pp 306− 309. (35) Axelrod, D.; Hellen, E.; Fulbright, R. In Topics in Fluorescence Spectroscopy; Lakowicz, J., Ed.; Springer: New York, 2002; Vol. 3; pp 289−343. (36) Keyser, U. F.; Krapf, D.; Koeleman, B. N.; Smeets, R. M. M.; Dekker, N. H.; Dekker, C. Nanopore tomography of a laser focus. Nano Lett. 2005, 5, 2253−2256. (37) Hillisch, A.; Lorenz, M.; Diekmann, S. Recent advances in FRET: distance determination in protein-DNA complexes. Curr. Opin. Struct. Biol. 2001, 11, 201−207. (38) Härtling, T.; Reichenbach, P.; Eng, L. M. Near-field coupling of a single fluorescent molecule and a spherical gold nanoparticle. Opt. Express 2007, 15, 12806−12817.

478

dx.doi.org/10.1021/nl203876w | Nano Lett. 2012, 12, 473−478