Analyst

PAPER

Publ

ishe

d on

13

Oct

ober

201

4. D

ownl

oade

d by

Bos

ton

Uni

vers

ity o

n 24

/10/

2014

21:

48:4

9.

View Article OnlineView Journal

Nanoscale chara

aDepartment of Biomedical Engineering, B

Boston, MA 02215, USA. E-mail: selim@bu.bDepartment of Electrical and Computer

Mary's St., Boston, MA 02215, USAcDivision of Materials Science and Engineerin

Brookline, MA 02446, USA

† Electronic supplementary informationsequences and nomenclature. (2) UniforDNA microarrays on the polymer sucustomized ow cell assembly. (4) Simuaxial heights from expected valuesthicknesses. (5) Calculation of the averadsDNA. (6) Validating the dual-color SSFMsteps. (7) Simulation of the deviation of ausing dual-color SSFM. See DOI: 10.1039/

Cite this: DOI: 10.1039/c4an01425a

Received 3rd August 2014Accepted 11th October 2014

DOI: 10.1039/c4an01425a

www.rsc.org/analyst

This journal is © The Royal Society of

cterization of DNA conformationusing dual-color fluorescence axial localization andlabel-free biosensing†

Xirui Zhang,a George G. Daaboul,b Philipp S. Spuhler,a David S. Freedman,b

Abdulkadir Yurt,c Sunmin Ahn,a Oguzhan Avcib and M. Selim Unlu*ab

Quantitative determination of the density and conformation of DNA molecules tethered to the surface can

help optimize and understand DNA nanosensors and nanodevices, which use conformational or motional

changes of surface-immobilized DNA for detection or actuation. We present an interferometric sensing

platform that combines (i) dual-color fluorescence spectroscopy for precise axial co-localization of two

fluorophores attached at different nucleotides of surface-immobilized DNA molecules and (ii)

independent label-free quantification of biomolecule surface density at the same site. Using this

platform, we examined the conformation of DNA molecules immobilized on a three-dimensional

polymeric surface and demonstrated simultaneous detection of DNA conformational change and

binding in real-time. These results demonstrate that independent quantification of both surface density

and molecular nanoscale conformation constitutes a versatile approach for nanoscale solid-biochemical

interface investigations and molecular binding assays.

1. Introduction

The immobilization or synthesis of DNA molecules on a solidsurface has stimulated the development of DNA sensors andnanodevices in wide-ranging biomedical applications.1–7 Forinstance, DNA microarrays, which allow for massively parallelmultiplexed analysis, have become routine for gene-expressionproling in both research and clinical practice.8–10 In recentyears, utilizing surface-immobilized DNA, researchers havedeveloped highly sensitive switching-based biological detectorsand dynamic programmable DNA motors.3,6,11–14 To understandand optimize the sensing and actuation performance of theseDNA sensors and nanodevices, it is necessary to characterizetwo parameters: the DNA surface density and conformation,

oston University, 44 Cummington Mall,

edu

Engineering, Boston University, 8 Saint

g, Boston University, 15 Saint Mary's St.,

(ESI) available: (1) Oligonucleotidem and consistent spot morphology ofrface. (3) Schematic illustration oflation of the deviation of uorophoreon substrates with different SiO2

ge orientation of surface-immobilizedsystem with nanometer scale surface

xial step heights from expected valuesc4an01425a

Chemistry 2014

which dictate the behavior of the surface-immobilized DNAmolecules.15–20

Various optical, electrical, and mechanical techniques havebeen developed to characterize the surface density, conforma-tion, or thickness of the DNA molecules immobilized on two-dimensional (2-D) functionalized surfaces such as gold, quartzcrystal, silicon, silicon dioxide or diamond.21–29 In recent years,three-dimensional (3-D) polymeric coatings have beencommonly used for surface functionalization given their highimmobilization specicity and capacity, optimized biologicalactivity, and simple immobilization procedure.30–35 The axialswelling of the 3-D polymeric coating in aqueous solutionsincreases the complexity for precise characterization of thebiophysical properties of the solution-phase interfacial micro-environment. Here, we report a simple, non-destructive inter-ferometric spectroscopy method to directly determine both thenanoscale conformation and surface density of DNA immobi-lized on a 3-D polymeric surface.

Spectral self-interference uorescence microscopy (SSFM)localizes the height of monolayer uorophores above a reect-ing surface from the spectral oscillations caused by the inter-ference between direct and reected waves of uorescenceemission.36 SSFM enables the determination of axial heights ofuorophore tags on surface-immobilized DNA molecules on atwo layer SiO2–Si substrate with subnanometer resolution.36,37

In this work we have extended SSFM by designing the thicknessof the SiO2 layer to employ two distinct spectral ranges thatenables the nanoscale characterization of DNA conformation by

Analyst

Analyst Paper

Publ

ishe

d on

13

Oct

ober

201

4. D

ownl

oade

d by

Bos

ton

Uni

vers

ity o

n 24

/10/

2014

21:

48:4

9.

View Article Online

determining the average axial height difference between twouorophore positions on a single DNA spot (Fig. 1).

We combined dual-color SSFM with white light reectancespectroscopy (WLRS) to quantify the DNA surface density toevaluate ensemble conformational measurements (Fig. 2).WLRS has been used for label-free detection of bimolecularinteractions by monitoring the thickness of the biological layer

Fig. 1 Dual-color SSFM detection principle. (a) Amino-modified DNA mfluorophores) on opposite ends are covalently immobilized on a polymheight difference (Dh) between the DNA surface-distal end (green fluomeasured fluorescence interference spectrums of Atto647N and Atto532an average axial height of the fluorophores to the SiO2–Si interface wenvelope of the two fluorophores given by the manufacturer.

Fig. 2 SSFM combined with LED-based white light reflectance spectrosSSFM and LED-RS. Yellow dashed line indicates the thickness of the bio-height of red fluorophores measured by SSFM. (b) Simultaneous acquisitiblack line represents the interference spectrum for the LED and the red fluthe emission spectral envelope of the LED while the red dotted line is theby the manufacturer.

Analyst

(bio-layer).37–39 The surface density of bound biomolecules canbe quantied by calibrating the thickness of the bio-layer to anamount of known deposited biomass and molecular weight.40

We incorporated WLRS into the dual-color SSFM system byusing a light-emitting diode (LED) instead of a white lightsource traditionally used for illumination, which will bereferred to as LED-RS. The interference of the LED light

olecules labeled with Atto647N (red fluorophores) and Atto532 (greener-functionalized SiO2–Si substrate. Dual-color SSFM measures therophores) and the surface-proximal end (red fluorophores). (b) Theon the layered substrate. A few oscillations of each spectrum are fit intoith subnanometer accuracy.36 Dotted lines are the emission spectral

copy (referred to as LED-RS). (a) Compare the detection principles oflayer measured by LED-RS while red dotted line indicates the averageon of the average fluorophore axial height and bio-layer thickness. Theorophores taken from a single acquisition. Yellow dashed line indicatesfluorescence emission spectral envelope of the red fluorophore given

This journal is © The Royal Society of Chemistry 2014

Fig. 3 Schematic illustration of the optical setup combining the dual-color SSFM andWLRS using a LED as illumination source (referred to asLED-RS). Components of the system shown are not to scale.

Paper Analyst

Publ

ishe

d on

13

Oct

ober

201

4. D

ownl

oade

d by

Bos

ton

Uni

vers

ity o

n 24

/10/

2014

21:

48:4

9.

View Article Online

reected from the top solution/bio-layer interface and that fromthe bottom SiO2–Si interface results in spectral oscillations inthe collected LED spectrum that shis upon biomoleculeaccumulation on the substrate. The emission spectrum of theLED is selected to be distinct from that of the red uorophores.Thus as shown in Fig. 2(b) we can simultaneously measure thebio-layer thickness and the red uorophore height from a singlespectrum measurement. The bio-layer thickness providesinformation regarding the molecular surface density while theuorophore heights indicate the conformation of the DNAmolecules at the same location on the surface. The capability ofSSFM combined with LED-RS to quantify both biomoleculedensity and conformation is not only desired for the study ofspecic molecular binding mechanisms,41 but is also needed tocharacterize the biophysical properties of a biosensingsurface.16,42–45

2. ExperimentalMaterials

All buffer solutions were prepared with deionized water (DIwater) ltered by Barnstead Nanopure Diamond (18.2 MU cm�1,Thermo Scientic, Waltham, MA). Buffer used for wetmeasurements was NaCl/Tris solution prepared with Tris buff-ered saline tablets and NaCl (Tris: 10 mM, NaCl: 50 mM, pH 7.6,Sigma-Aldrich) unless specied. DNA sequences were designedwith Oligo Analyzer (Integrated DNA Technologies, Inc., Coral-ville, IA) to minimize secondary structures. Single-stranded 60-bp DNA molecules were synthesized by IBA GmbH (Goettingen,Germany), and single-stranded 20-bp and 40-bp DNA moleculeswere synthesized by Integrated DNA Technologies, Inc. Allsingle-stranded DNA (ssDNA) molecules were puried withhigh-performance liquid chromatography (HPLC) aersynthesis. For the purpose of immobilization, ssDNA moleculeswere modied with amine-C6 linkers on the 50 or 30 end. Fluo-rophores Atto647N or Atto532 were labeled on the other end ofthe amino-modied strand or on the 50 or 30 end of the

This journal is © The Royal Society of Chemistry 2014

complementary strand. The DNA sequences and modicationsused in different experiments are shown in Table S1 of ESI.† ThedsDNA molecules were hybridized at 30 mM in 150 mM sodiumphosphate buffer (prepared with sodium phosphate monobasicand sodium phosphate bibasic, pH 8.5, Sigma-Aldrich) beforespotting. Microarray substrates were prepared from siliconwafers that had 17.5 mm thick thermally grown oxide (SiliconValley Microelectronics, Santa Clara, CA), which were cut into 15mm � 15 mm square chips. Before use, chips were rinsed withacetone, methanol, subjected to sonication in acetone for 5min, and oxygen plasma ashing (M4LTM, PVA TePla AmericaInc., Corona, CA) for 10 min.

DNA microarray and polymeric coating preparation

Clean chips were functionalized with a commercial polymerproduct MCP-2 from Lucident Polymers (Sunnyvale, CA), whichwill be referred to as the polymer. The polymer is obtained fromradical copolymerization of N,N-dimethylacrylamide (DMA), N-acryloyloxysuccinimide (NAS), and 3-(trimethoxysilyl)-propylmethacrylate (MAPS), and covalently binds to amino-modiedDNA as described elsewhere.35,46 The polymer has been shownto have many advantages for DNA immobilization, such as highstability and reproducibility, fast and cost-effective preparationprocedures, and minimal non-specic binding.32,35,47 The chipswere immersed in a 1 : 1 mixture of 1% (w/v) MCP-2 in DI waterand ammonium sulfate at 40% saturation concentration for 30min, subsequently washed thoroughly with DI water, dried withargon gas, and then baked for 15 min at 80 �C. DNA moleculeswere spotted in micro-arrayed fashion in 150 mM sodiumphosphate (pH 8.5) buffer using a robotic spotter (sci-FLEXARRAYER, Scienion, Monmouth Junction, NJ), and thenimmobilized for twelve hours in 65% humidity at roomtemperature. The chips were then washed in a petri dish on anorbital shaker three times in 2� saline-sodium citrate buffer(SSC) for 10 minutes, three times in 0.2� SSC for 5 minutes,once in 0.1� SSC for 1 minute, dried with argon gas, and kept invacuum until measurements were performed. The diameter ofeach individual DNA spot was about 150 mm. Low backgroundand consistent spot morphology were previously shown for DNAmicroarrays on glass slides functionalized with the polymer.DNA spot morphology was optimized by changing spottingconditions, such as concentration, temperature and humidity,and examined with a uorescence scanner before measure-ments. Within the measured area at the center of the DNAspots, all spots possessed satisfactory uniform morphology(ESI,† Section 2).

Measurements in a customized ow cell

Both DNA orientation and hybridization measurements wereperformed in a customized ow cell, where the chip surface wasincubated in buffer (ESI,† Section 3). A silicone rubber sheet(McMaster-Carr, Robbinsville, NJ), a glass window (EdmundOptics, Barrington, NJ), and the ow cell bottom formed theow chamber, which were clamped and sealed using fourscrews. The custom-made glass window has anti-reectioncoatings in the visible and near-infrared spectrum specic for

Analyst

Analyst Paper

Publ

ishe

d on

13

Oct

ober

201

4. D

ownl

oade

d by

Bos

ton

Uni

vers

ity o

n 24

/10/

2014

21:

48:4

9.

View Article Online

each side so as not to affect spectral signals. The workingvolume of the ow cell was 100 mL. Buffer solutions were lteredwith 0.1 mm Millex-LG syringe lter (Millipore, Billerica, MA)before introduction to remove impurities. A peristaltic pump(Control Company, Friendswood, TX) drove buffer at a constantow rate of 240 mL min�1. The ow cell was xed onto a 2-axispositioning micro-stage (Mad City Labs, Madison, WI), and thescanning of the DNA microarray was implemented by movingthe stage by the DNA spot center-to-center distance with sub-micrometer accuracy.

Optical setup (Fig. 3)

The light beams of a red laser (helium–neon, 633 nm, MellesGriot, Carlsbad, CA) and a green laser (diode-pumped solid-state, 532 nm, Laserglow Technologies, Toronto, Canada) werecombined with a dichroic beamsplitter. The combined laserbeams were collimated and expanded with two achromaticlenses (Thorlabs Inc., Newtown, NJ). A mechanical shutter thatopens and closes via a transit TTL signal (+5 V or 0 V) controlledthe red laser. A current control module regulated the green laserupon receiving a TTL signal (+5 V or 0 V). A printed circuit board(PCB) was designed to receive digital signals from the computerand send TTL signals to the shutter and the controller. The laserbeams were then reected by a dual-edge dichroic beam splitter(545/650 nm BrightLine, Semrock, Inc. NY), which specicallyreects the wavelengths of both lasers and transmits over 90%of the emission spectrums of the two uorophores. A Nikon 5�objective with a numerical aperture (NA) of 0.13 focused thelaser beams to diffraction-limited spots at the center of eachDNA spot on the sensor surface. Emitted uorescence wascollected by the objective and focused onto a spectrometerconnected to a CCD camera (SP-2150i, Princeton Instruments,Trenton, NJ). A yellow LED (M565L2, Thorlabs, Newton, NJ) wasused as the illumination source for LED-RS measurements. Apinhole with a diameter of 100 mm was placed in front of theLED and a reduced image of the pinhole was focused to thesame location as the laser spots. The reected LED light wasthen collected and focused onto the spectrometer with the samefocus as the laser beams. The spectrometer, the micro-stage,and the lasers were controlled via custom MATLAB sowarethat performs automatic data acquisition. The interferencespectrums consisted of three parts: the uorophore or the LEDemission spectral envelope, the oscillatory interferencecomponent, and the shot noise collected from the CCD camera.Both uorescence and LED interference spectrums were t withcustom algorithms to obtain uorophore heights and bio-layerthicknesses as described elsewhere.48

Calculation of dsDNA orientation using uorophore heightmeasurements

The average orientation of dsDNA was calculated by applyingsimple trigonometry using the uorophore height difference(Dh) and the end-to-end length of the dsDNA. The persistencelength (l) of dsDNA is approximately 50 nm in our buffer solu-tions.49 Thus a short dsDNAmolecule (<150-bp) can be modeledas a rigid rod tethered to the sensor surface. The length of the

Analyst

rod is dened by the root-mean-square (rms) end-to-enddistance ð ffiffiffiffiffiffiffiffihr2ip Þ of the dsDNA based on the worm-like chainmodel.50 Therefore, the average orientation of dsDNA relative tothe surface is hqi ¼ arcsinðhDhi= ffiffiffiffiffiffiffiffihr2ip Þ; where Dh is ensembleaverage of the uorophore height difference between thesurface-distal and surface-proximal ends of the dsDNA mole-cules resided within the measured area (ESI,† Section 5).

Quantication of DNA surface density

The DNA layer thickness measured by LED-RS and a calibrationcoefficient (1 ng mm�2 nm�1)40,51 were used to calculate theDNA mass surface density. Furthermore, using the DNAmolecular weight and the Avogadro constant, the molecularsurface density can be calculated from themass surface density.For example, a 1 nm thick dsDNA spot (60-bp, molecular weight:38 700 g mol�1) renders a molecular surface density of 1.56 �1012 molecules per cm2.

3. Results and discussionDesign of the SiO2 layer thickness and validation of dual-colorSSFM and LED-RS

Employing the interference spectrum at two different spectralranges by dual-color SSFM and LED-RS on the same chiprequires us to use substrates with a thicker SiO2 layer. Previ-ously for single-color SSFM using a substrate with 5 mm SiO2

layer, at least two periods of oscillations that spanned a band-width of about 1000 cm�1 within the uorescence emissionspectrum were needed to precisely determine uorophore axialheights with less than 1 nm deviation.52 However, for thecombined dual-color SSFM and LED-RS, the bandwidths usablefor tting without sacricing signal-to-noise ratio were muchnarrower, constrained by the emission spectrums of the our-ophores and the LED and the passbands of the availabledichroic beam splitters and notch lters. Since the interferenceoscillation period is inversely proportional to the thickness ofthe SiO2 layer,36 within the narrower bandwidths, we need to usethicker SiO2 layer to obtain sufficient oscillation periods forprecise height determination. We simulated the uorophoreaxial height deviation from expected values on substrates withdifferent SiO2 thicknesses under the available bandwidths inthe system (red uorophore: 700 cm�1, green uorophore: 450cm�1, LED: 800 cm�1). Based on the simulation, we selected asubstrate with 17.5 mm thick SiO2 layer that allowed for at leasttwo periods of oscillations within each spectral bandwidth usedby dual-color SSFM and LED-RS (ESI,† Section 4).

Table 1 shows that the uorophore axial heights measuredon the newly designed substrate utilizing a 17.5 mmSiO2 layer bydual-color SSFM were as precise as traditional single-colorSSFM. We spotted an array of 60-bp dsDNA with differentmodications, measured the average height difference (Dh)between the surface-distal and surface proximal ends of thedsDNA, and calculated the dsDNA orientations to the surface(see Experimental section). In Table 1, the schemes (a) and (b)represent single-color SSFM while schemes (c) and (d) representdual-color SSFM. We also measured nanometer-scale steps

This journal is © The Royal Society of Chemistry 2014

Table 1 Compare quantification of surface-immobilized dsDNAorientation in different detection schemes. Data presented is themeanand standard deviation of the height difference (Dh) between thesurface-distal and surface-proximal ends of the dsDNA on 10 separateDNA spots or spot pairs

Detection schemes

DNA name Seq1 Seq2 Seq3 Seq4 Seq5 Seq6

hDhi (nm) 11.9 11.0 11.7 11.7SDhDhi 1.0 1.3 0.8 0.6hqi (�) 38.6 35.2 37.7 37.2SDhqi 3.0 3.9 2.5 1.9

Paper Analyst

Publ

ishe

d on

13

Oct

ober

201

4. D

ownl

oade

d by

Bos

ton

Uni

vers

ity o

n 24

/10/

2014

21:

48:4

9.

View Article Online

etched on the SiO2 surface by single-color SSFM, dual-colorSSFM and LED-RS. The results show that all modalities candetermine the nanometer axial height differences with compa-rable accuracy both in air and in solution (ESI,† Section 6).

Characterization of the conformation of dsDNA immobilizedon the polymer surface

We used dual-color SSFM combined with LED-RS to evaluate theeffects of surface density, buffer ionic strength, and DNA lengthon the conformation of dsDNA on the polymer surfaceimmersed in buffer solution. First, we observed that the orien-tation of dsDNA to the surface positively correlated to its surface

Fig. 4 Characterization of DNA molecules immobilized on a polymericaverage orientation of each dsDNA spot positively correlates to its surfacedsDNA negatively correlates to buffer ionic strength. (c) Long dsDNA spothe polymer surface. (d) Schematic illustration of the effect of surface dpenetrate the polymeric surface and orient downwards, resulting in low

This journal is © The Royal Society of Chemistry 2014

density on the polymer surface. We measured an array of 60-bpuorophore-labeled dsDNA immobilized at concentrationsranging from 8 mM to 10 mM to generate dsDNA spots of variedsurface densities. We determined the surface densities of thedsDNA spots with LED-RS and their ensemble average orienta-tions with dual-color SSFM. For SSFM measurements, DNAspots of different surface densities were exposed for differentlengths of time, ranging from 0.1 second to 1 second. The dual-color SSFM optical setup operates in the shot noise limitedregime where the signal-to-noise ratio (SNR) is proportional tothe square root of signal counts. The exposure time was selectedto ensure sufficient signal counts, corresponding to at least aSNR of 30 for accurate tting and axial localization. Fig. 4(a)shows that dsDNA orientation has a positive correlation withsurface density. This correlation was expected because as thedensity of dsDNA layer increased, the distance between adjacentdsDNA molecules became smaller, resulting in stronger elec-trostatic repulsion and more steric hindrance between thedsDNA molecules. As illustrate in Fig. 4(d), the random rota-tional freedom for each dsDNA molecule under thermal uc-tuation decreased, and thus the ensemble average orientationof the dsDNA from the surface increased.

The measured orientations of dsDNA shown in Fig. 4(a) werelower than the values predicted by trigonometric calculationsusing a rigid rod model of dsDNA on a 2-D surface, which areover 76� to the surface, much higher than our results. Weattribute this discrepancy to the 3-D geometry of the polymersurface. The actual distance between the dsDNAmolecules werepresumably larger than that calculated based on 2-D surfacegeometry. The polymeric coating swells by about 7 to 20 nm

surface. Error bar shows standard deviation of 10 dsDNA spots. (a) Thedensity. The red dashed line is a guide to the eye. (b) The orientation ofts have larger average orientations than those of short dsDNA spots onensity on the orientation of dsDNA. (e) Short dsDNA molecules mayer average orientation.

Analyst

Analyst Paper

Publ

ishe

d on

13

Oct

ober

201

4. D

ownl

oade

d by

Bos

ton

Uni

vers

ity o

n 24

/10/

2014

21:

48:4

9.

View Article Online

upon hydration and its functional groups for immobilizationare distributed in the axial dimension.32,35 As a result, thedsDNA molecules may penetrate the polymer scaffold and alsobe distributed axially depending on the polymer pore size andthe dsDNA length. Hence, the axial penetration and distributionof the dsDNA molecules resulted in reduced ensemble averageheight differences between surface-distal and surface-proximalends, and consequently reduced average orientations. More-over, axial distribution of the immobilized dsDNA moleculespotentially added more space for random rotation, which couldalso lead to lower calculated orientations compared to thosebased on a 2-D surface model.

Next, we examined the effects of buffer ionic strength onthe orientation of 60-bp dsDNA molecules by measuring theaverage orientation of 10 dsDNA spot replicates in buffers ofdifferent NaCl concentrations (Fig. 4(b)). We observed thataverage orientation of dsDNA on the polymer surface nega-tively correlated to the buffer ionic strength, which agreeswith Manning's counterion condensation theory.53 Thetheory states that cations in the buffer can condense onto thenegatively charged dsDNA backbones. Therefore, increasedconcentration of cations in the buffer shielded the electro-static repulsive forces between the dsDNA molecules, whichresulted in more random rotational freedom of the dsDNAand therefore lowered the average orientation. Fig. 4(b)shows that when the salt concentration was sufficiently high,the effect of the electrostatic repulsion between dsDNAmolecules on their orientation became minimal. The orien-tation of 60-bp dsDNA molecule approached 33�, the statis-tical average orientation of dsDNA with full random rotationfreedom under thermal uctuations on a 2-D surface (ESI,†Section 5). However, the intermolecular distance calculatedfrom the surface density was about 6.8 nm, where the DNAmolecules should be sterically restricted from random freerotation. This observation again implied that under the samesurface density used in the measurement, the dsDNA mole-cules had more free rotational space on the 3-D polymersurface than on a 2-D surface.

Finally, by measuring orientations of three dsDNA species ofdifferent lengths, we provide evidence that shorter dsDNAmolecules have more freedom of random thermal rotations onthe polymer surface. 20-bp, 40-bp, and 60-bp dsDNA spots wereimmobilized of the same surface density and measured inbuffer of the same NaCl concentration. Fig. 4(c) shows that theaverage orientation of longer dsDNA molecules was larger thanthat of shorter dsDNA molecules. This trend was anticipatedsince shorter dsDNA molecules, when immobilized at a similardensity, have less steric hindrance and electrostatic repulsionbetween them than longer dsDNA molecules. However, theaverage orientation of 20-bp dsDNA was smaller than thestatistical average (33�) of dsDNA freely rotating on a planarsurface.54 The length of 20-bp dsDNA (�6.5 nm) is comparableto the diameter of the polymeric pore size (�5 nm) in hydratedstate.32 Hence as illustrated in Fig. 4(e) the 20-bp dsDNA mayorient downwards in the 3-D polymer surface. The result indi-cates that short dsDNA of length comparable to the polymeric

Analyst

pore size can have 3-D random rotational freedom when thepolymer is hydrated and swells.

Real-time quantication of conformational change andbinding during DNA hybridization on the polymer surface

We demonstrate the use of dual-color SSFM and LED-RS forsimultaneous quantication of DNA conformational changeand complementary strand binding during DNA surfacehybridization. We immobilized 60-bp ssDNA modied withamine-C6 linker at the 30 end (surface-proximal end) and the reduorophore at the 50 end (surface-distal end) on a polymer-functionalized chip. The chip was xed in a customized ow cellwith its surface immersed in buffer solution, which was drivenby a peristaltic pump (ESI,† Section 3). Aer owing onlyhybridization buffer for a baselinemeasurement, we introducedhybridization buffer containing complementary strands at aconcentration of 500 nM at 45 �C. The surface-distal red uo-rophore height and the thickness of the ssDNA layer weremonitored at a time interval of 21 seconds before and duringhybridization while the proximal green uorophore height wasmeasured aer hybridization reached equilibrium. The timeresolution was set so that over 100 uorescence measurementswere taken during the entire hybridization process before theuorophores became completely photobleached.

The average height of the distal red uorophores increasedfrom 9 nm to about 15 nm and the average thickness of the DNAlayer increased from 1.5 nm to 2.7 nm as ssDNA becamehybridized into dsDNA till binding equilibrium was reached(Fig. 5). The initial thickness of the ssDNA layer corresponded toan immobilization density of 3.8 � 1012 cm�2. The DNAhybridization efficiency was 0.8 at steady state obtained bycalculating the fraction of additional thickness on the surface.Based on the studies of DNA hybridization regimes by Levickyand Melosh,20,55 the surface density under our experimentalcondition was in the pseudo-Langmuir regime (2 � 1012 cm�2

to 6 � 1012 cm�2). We thus t each dynamic hybridization datato the rst-order Langmuir kinetic equation (Fig. 5):

x ¼ Xeq(1 � e�keffCtt)

For LED-RS measurement, x denotes the hybridization effi-ciency, and Xeq is the steady state hybridization efficiency. ForSSFM measurement, x denotes the average ensemble red uo-rophore height at the surface-distal end of the ssDNA, and Xeq isthe average ensemble red uorophore height at steady state. Forboth measurements, keff is the effective rate constant, Ct is theconcentration of the target complementary strands, and t istime. The keff of DNA spot mass density change, indicated by thehybridization efficiency change, and of DNA conformationchange, indicated by the red uorophore height change, wereboth around 1 � 104 M�1 S�1.

The height increase at the surface-distal end of the ssDNAprobes signied that the DNA conformation changed fromhighly exible single-stranded coils to rigid double-strandedhelices (Fig. 5(a)). The average uorophore height increase wasapproximately 6 nm (Fig. 5(b)), which agrees with thedisplacement range between 3 and 10 nm determined by a

This journal is © The Royal Society of Chemistry 2014

Fig. 5 Real-time detection and quantification of DNA conformationand hybridization using dual-color SSFM and LED-RS. (a) Schematicillustration of the DNA hybridization process on the polymer surface.Average fluorophore heights are indicated as dotted lines. (b)Hybridization efficiency calculated from the fraction of additionalthicknesses of the DNA spot measured by LED-RS with a time reso-lution of 21 seconds. (c) Simultaneous detection of the average heightincrease of the red fluorophores at the surface-distal end of the ssDNAprobes during hybridization. A first-order Langmuir kinetic equationwas fit to both measurements and both observed effective rateconstants were around 1 � 104 M�1 S�1.

Paper Analyst

Publ

ishe

d on

13

Oct

ober

201

4. D

ownl

oade

d by

Bos

ton

Uni

vers

ity o

n 24

/10/

2014

21:

48:4

9.

View Article Online

single-molecule micromechanical method.56 The complemen-tary strands had green uorophores tagged to the 50 end, whichresided on the surface-proximal end of the immobilized ssDNAmolecules upon hybridization. Aer hybridization reachedequilibrium, we replaced the hybridization solution in the owcell with pure buffer and measured the height of the greenuorophores at about 2 to 3 nm. If we subtract the averageheight of the green uorophores from that of the red uo-rophores before hybridization, we then obtained the heightdifference between the surface-distal and surface-proximal endsof the ssDNAmolecules before hybridization, which was about 6to 7 nm as shown in Fig. 5(a). ssDNA is oen modeled asrandom coils in solution,57,58 whose exibility is characterizedby a persistence length of about 1 to 4 nm. This larger thanpersistence length height difference implies that rather than a

This journal is © The Royal Society of Chemistry 2014

complete random coil, the immobilized ssDNA molecules hadan extended conformation on the polymer surface. Thisextended conformation was also observed on short ssDNAtethered to a 2-D gold surface.25

Fig. 5(b) shows that the DNA hybridization efficiency atequilibrium was smaller than one. Many factors can affect theefficiency of DNA surface hybridization.20 We rst attribute tothe electrostatic barrier created by the DNA layer negativepotential, which was enhanced along the incorporation of morenegatively charged complementary strands. Moreover, hybrid-ized dsDNA and the polymeric surface might also stericallyhinder the incorporation of the complementary strands. Thehybridization efficiency can be improved by both lowering thesurface density and increasing the buffer ionic strength.20

We demonstrated the implementation of a dual-color SSFMcombined with LED-RS for the nanoscale characterization ofDNA conformation on a 3-D polymer surface. The systemimproved upon traditional single-color SSFM and white lightreectance spectrometry by using a substrate with a thicker SiO2

layer (17.5 mm) and a LED illumination source. The newlydesigned substrate along with additional optical componentsenabled the use of multiple spectral bandwidths for preciseaxial localization of two different uorophores and quantica-tion of DNA surface density. We note that SSFM determines theaxial location of uorophores from the spectral oscillations, notfrom uorescence intensity variations. Thus experimentalconditions that can potentially cause uorescence intensityvariations do not affect the results.

Our results demonstrate that dual-color SSFM is advanta-geous over traditional single-color SSFM. For example, DNAsurface density affects the orientation of surface-immobilizedDNA. Previously, single-color SSFM determined orientations ofsurface-immobilized dsDNA by measuring the height differ-ences between surface-distal and surface-proximal ends onseparate DNA spots, which might have different immobilizationdensities and axial distributions. Also, dual-color SSFMincreases the throughput of the height measurements underlimited number of arrayed spots. Additionally, dual-color SSFMoffers new assay designs. For instance, the location of one u-orophore tagged to a DNA-binding protein can be compared toanother labeled at a reference location on the DNA molecule todetermine the protein binding site along the DNA sequence.Comparing to uorescence resonance energy transfer (FRET),59

dual-color SSFM does not require a complex uorescencenormalization procedure in spectrum processing. Moreover, theaxial heights measurable using dual-color SSFM ranges fromsubnanometer to more than 100 nanometers36 (ESI,† Section 7)whereas the distance measurable using FRET is less than 10nm.59 Comparing to acoustic wave sensors, the results of dual-color SSFM are directly interpreted as the geometric features ofDNA conformation rather than mechanical properties of DNAlms, such as modulus and viscosity.60 Moreover, LED-RSprovides orthogonal information of the DNA surface density,which is oen coupled with the signal generated by DNAconformational change for mechanical and optical resonancebased sensors. Notably, the nanoscale DNA conformationalchange from exible coils to rigid double helices was

Analyst

Analyst Paper

Publ

ishe

d on

13

Oct

ober

201

4. D

ownl

oade

d by

Bos

ton

Uni

vers

ity o

n 24

/10/

2014

21:

48:4

9.

View Article Online

simultaneously monitored during the addition of complimen-tary sequence to ssDNA probes.

Our results of the orientation and surface density of dsDNAmolecules indicated that dsDNA immobilized on a 3-D poly-meric coating have more random rotation freedom underthermal uctuations than those immobilized on a 2-D surfacein hydrated environment. A precise model describing the rela-tionship between the average orientation and surface density isunattainable at the moment due to lack of the exact axialdistribution of dsDNA molecules using average ensemblemeasurements. Yet our results suggested that this larger degreeof free random thermal rotation of the DNA molecules, andpossibly of protein molecules, could cause the previouslydescribed high performance of the polymeric coating, such ashigh surface density, bioactivity retaining, and easy accessibilityof the immobilized molecular probes.35,47

Although dual-color SSFM and LED-RS allowed for real-timesimultaneous quantication of molecular binding and confor-mational change, a few limitations need to be overcome in futuredevelopments. First, other uorescence labels with narrowspectrums, such as Quantum Dots (QDs) should be used toreplace the uorophores for real-time measurements. In thiswork, the green and red uorescence spectrums were measuredseparately. The tail of the green uorophore emission spectrumoverlapped with that of the red uorophore, which can affect theaccuracy of red uorophore localization if measured at the sametime. Therefore, during DNA hybridization, we only measuredthe green uorophore height aer hybridization reached steadystate in pure buffer without complementary sequences. More-over, the uorophores photobleach aer certain number ofmeasurements, limiting the time resolution for monitoring real-time binding processes. Also, FRET occurred between the greenuorophores and the red uorophores, and expedited the pho-tobleaching of the red uorophores. Non-blinking quantum dots(QDs)61 possessing narrow non-overlapping spectrums can be agood substitute. The emission of QDs can be excited at the samelaser wavelength and they have extraordinary brightness andphotostability as compared to organic uorophores, which havealready been exploited in cell imaging and biological sensing.62

Second, the ow cell employed for real-time hybridizationmeasurements operated in a diffusion-limited regime.63 Thus,the observed effective rate constants of DNA binding andconformational change were indistinguishable, both of whichwere limited by the transportation rate of the complementarystrands to the surface-immobilized ssDNA. Thus, the ow cellneeds to be optimized to a reaction-limited regime, where therates of conformational change and binding can help determinethe biophysical mechanisms of the reaction. Additionally, thecurrent optical setup is congured in a point-to-point scanningmodality where only one DNA spot can be measured at a time. A2-D spectral imaging modality using spatial line scanning arecurrently being implemented to image and measure an array ofDNA spots simultaneously. This spectral imaging modality candramatically improve the measurement throughput and reducethe time resolution for measuring an array of DNA spots in real-time assays.

Analyst

4. Conclusion

In summary, we have demonstrated the development of a dual-color SSFM combined with independent biomolecular sensingfor the characterization of nanoscale conformation of DNA on a3-D polymer surface. A newly designed substrate with a thickSiO2 layer (17.5 mm) allowed for the use of multiple spectralbandwidths for precise axial localization of two different uo-rophores and quantication of biomolecule surface density.Using dual-color SSFM, we have shown that the conformation ofsurface-immobilized dsDNA on a 3-D polymeric surface wassubject to surface density, buffer ionic strength, and dsDNAlength. Furthermore, dsDNA immobilized on a polymericsurface possess more conformational freedom than those on a2-D surface, providing additional evidence for the high densityand easy accessibility of biomolecules on 3-D polymeric coatingin molecular binding assays. Particularly, we have shownsimultaneous and independent real-time monitoring of DNAconformational change and complementary strand bindingduring DNA surface hybridization. The overall optical setup issimple and compatible with regular uorescence microscopes.All measurements were performed in a ow cell on DNAmicroarrays, compatible with multiplexed parallel molecularbinding assays. The versatility, simplicity, and robustness of thepresented platform make it a useful tool that can be broadlyused for quantitative nanoscale characterization of solid-biochemical interfaces for the optimization of various DNAsensors and nanodevices.

Acknowledgements

The authors would like to acknowledge the National ScienceFoundation (Grant 0933670) for funding support. Xirui Zhangwould like to thank Professor Bennett Goldberg for fruitfuldiscussions during manuscript preparation.

References

1 G. R. Abel, Jr, E. A. Josephs, N. Luong and T. Ye, A SwitchableSurface Enables Visualization of Single DNA HybridizationEvents with Atomic Force Microscopy, J. Am. Chem. Soc.,2013, 135, 6399–6402.

2 S. F. J. Wickham, J. Bath, Y. Katsuda, M. Endo, K. Hidaka,H. Sugiyama and A. J. Turbereld, A DNA-Based MolecularMotor That Can Navigate a Network of Tracks, Nat.Nanotechnol., 2012, 7, 169–173.

3 K. W. Plaxco and H. T. Soh, Switch-Based Biosensors: A NewApproach Towards Real-Time, In Vivo Molecular Detection,Trends Biotechnol., 2011, 29, 1–5.

4 Y. Mao, D. Liu, S. Wang, S. Luo,W.Wang, Y. Yang, Q. Ouyangand L. Jiang, Alternating-Electric-Field-Enhanced ReversibleSwitching of DNA Nanocontainers with pH, Nucleic AcidsRes., 2007, 35, e33.

5 U. Rant, K. Arinaga, S. Scherer, E. Pringsheim, S. Fujita,N. Yokoyama, M. Tornow and G. Abstreiter, SwitchableDNA Interfaces for the Highly Sensitive Detection of Label-

This journal is © The Royal Society of Chemistry 2014

Paper Analyst

Publ

ishe

d on

13

Oct

ober

201

4. D

ownl

oade

d by

Bos

ton

Uni

vers

ity o

n 24

/10/

2014

21:

48:4

9.

View Article Online

Free DNA Targets, Proc. Natl. Acad. Sci. U. S. A., 2007, 104,17364–17369.

6 W. Shu, D. Liu, M. Watari, C. K. Riener, T. Strunz,M. E. Welland, S. Balasubramanian and R. A. McKendry,DNA Molecular Motor Driven Micromechanical CantileverArrays, J. Am. Chem. Soc., 2005, 127, 17054–17060.

7 T. G. Drummond, M. G. Hill and J. K. Barton,Electrochemical DNA Sensors, Nat. Biotechnol., 2003, 21,1192–1199.

8 J. D. Hoheisel, Microarray Technology: Beyond TranscriptProling and Genotype Analysis, Nat. Rev. Genet., 2006, 7,200–210.

9 M. F. Berger and M. L. Bulyk, Universal Protein-BindingMicroarrays for the Comprehensive Characterization of theDNA-Binding Specicities of Transcription Factors, Nat.Protoc., 2009, 4, 393–411.

10 J. S . Reis-Filho and L. Pusztai, Gene Expression Proling inBreast Cancer: Classication, Prognostication, andPrediction, Lancet, 2011, 378, 1812–1823.

11 A. Vallee-Belisle, F. Ricci, T. Uzawa, F. Xia and K. W. Plaxco,Bioelectrochemical Switches for the Quantitative Detectionof Antibodies Directly in Whole Blood, J. Am. Chem. Soc.,2012, 134, 15197–15200.

12 A. J. Bonham, K. Hsieh, B. S. Ferguson, A. Vallee-Belisle,F. Ricci, H. T. Soh and K. W. Plaxco, Quantication ofTranscription Factor Binding in Cell Extracts Using anElectrochemical, Structure-Switching Biosensor, J. Am.Chem. Soc., 2012, 134, 3346–3348.

13 E. Farjami, L. Clima, K. Gothelf and E. E. Ferapontova, “Off–on” Electrochemical Hairpin-DNA-Based Genosensor forCancer Diagnostics, Anal. Chem., 2011, 83, 1594–1602.

14 C. M. Strohsahl, B. L. Miller and T. D. Krauss, Preparationand Use of Metal Surface-Immobilized DNA Hairpins forthe Detection of Oligonucleotides, Nat. Protoc., 2007, 2,2105–2110.

15 E. A. Josephs and T. Ye, Nanoscale Spatial Distribution ofThiolated DNA on Model Nucleic Acid Sensor Surfaces,ACS Nano, 2013, 7, 3653–3660.

16 Z.-G. Yu and R. Y. Lai, Effect of Signaling ProbeConformation on Sensor Performance of a Displacement-Based Electrochemical DNA Sensor, Anal. Chem., 2013, 85,3340–3346.

17 A. V. Pinheiro, J. Nangreave, S. Jiang, H. Yan and Y. Liu,Steric Crowding and the Kinetics of DNA HybridizationWithin a DNA Nanostructure System, ACS Nano, 2012, 6,5521–5530.

18 E. Mirmomtaz, M. Castronovo, C. Grunwald, F. Bano,D. Scaini, A. A. Ensa, G. Scoles and L. Casalis,Quantitative Study of the Effect of Coverage on theHybridization Efficiency of Surface-Bound DNANanostructures, Nano Lett., 2008, 8, 4134–4139.

19 M. Castronovo, S. Radovic, C. Grunwald, L. Casalis,M. Morgante and G. Scoles, Control of Steric Hindrance onRestriction Enzyme Reactions with Surface-Bound DNANanostructures, Nano Lett., 2008, 8, 4140–4145.

20 P. Gong and R. Levicky, DNA Surface Hybridization Regimes,Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 5301–5306.

This journal is © The Royal Society of Chemistry 2014

21 N. E. Marotta, K. R. Beavers and L. A. Bottomley, Limitationsof Surface Enhanced Raman Scattering in Sensing DNAHybridization Demonstrated by Label-Free DNA Oligos asMolecular Rulers of Distance-Dependent Enhancement,Anal. Chem., 2013, 85, 1440–1446.

22 G. Papadakis, A. Tsortos, F. Bender, E. E. Ferapontova andE. Gizeli, Direct Detection of DNA Conformation inHybridization Processes, Anal. Chem., 2012, 84, 1854–1861.

23 B. Rezek, D. Shin, H. Uetsuka and C. E. Nebel, MicroscopicDiagnostics of DNA Molecules on Mono-CrystallineDiamond, Phys. Status Solidi A, 2007, 204, 2888–2897.

24 K. Wang, C. Goyer, A. Anne and C. Demaille, Exploring theMotional Dynamics of End-Graed DNA Oligonucleotidesby In Situ Electrochemical Atomic Force Microscopy, J.Phys. Chem. B, 2007, 111, 6051–6058.

25 U. Rant, K. Arinaga, S. Fujita, N. Yokoyama, G. Abstreiter andM. Tornow, Structural Properties of OligonucleotideMonolayers on Gold Surfaces Probed by FluorescenceInvestigations, Langmuir, 2004, 20, 10086–10092.

26 S. N. Patole, A. R. Pike, B. A. Connolly, B. R. Horrocks andA. Houlton, STM Study of DNA Films Synthesized onSi(111) Surfaces, Langmuir, 2003, 19, 5457–5463.

27 D. Y. Petrovykh, H. Kimura-Suda, M. J. Tarlov andL. J. Whitman, Quantitative Characterization of DNA Filmsby X-Ray Photoelectron Spectroscopy, Langmuir, 2004, 20,429–440.

28 D. Y. Petrovykh, H. Kimura-Suda, L. J. Whitman andM. J. Tarlov, Quantitative Analysis and Characterization ofDNA Immobilized on Gold, J. Am. Chem. Soc., 2003, 125,5219–5226.

29 A. Barhoumi, D. Zhang and N. J. Halas, Correlation ofMolecular Orientation and Packing Density in a dsDNASelf-Assembled Monolayer Observable with Surface-Enhanced Raman Spectroscopy, J. Am. Chem. Soc., 2008,130, 14040–14041.

30 Z. Yang, Y. Chevolot, T. Gehin, V. Dugas, N. Xanthopoulos,V. Laporte, T. Delair, Y. Ataman-Onal, G. Choquet-Kastylevsky, E. Souteyrand, et al., Characterization of ThreeAmino-Functionalized Surfaces and Evaluation of AntibodyImmobilization for the Multiplex Detection of TumorMarkers Involved in Colorectal Cancer, Langmuir, 2013, 29,1498–1509.

31 M. Rendl, A. Bonisch, A. Mader, K. Schuh, O. Prucker,T. Brandstetter and J. Ruhe, Simple One-Step Process forImmobilization of Biomolecules on Polymer SubstratesBased on Surface-Attached Polymer Networks, Langmuir,2011, 27, 6116–6123.

32 A. Yalçın, F. Damin, E. Ozkumur, G. di Carlo, B. B. Goldberg,M. Chiari and M. S. Unlu, Direct Observation ofConformation of a Polymeric Coating with Implications inMicroarray Applications, Anal. Chem., 2009, 81, 625–630.

33 F. Rusmini, Z. Zhong and J. Feijen, Protein ImmobilizationStrategies for Protein Biochips, Biomacromolecules, 2007, 8,1775–1789.

34 R. Suriano, M. Levi, G. Pirri, F. Damin, M. Chiari andS. Turri, Surface Behavior and Molecular Recognition inDNA Microarrays from N,N-Dimethylacrylamide

Analyst

Analyst Paper

Publ

ishe

d on

13

Oct

ober

201

4. D

ownl

oade

d by

Bos

ton

Uni

vers

ity o

n 24

/10/

2014

21:

48:4

9.

View Article Online

Terpolymers with Activated Esters as Linking Groups,Macromol.Biosci., 2006, 6, 719–729.

35 G. Pirri, F. Damin, M. Chiari, E. Bontempi and L. E. Depero,Characterization of a Polymeric Adsorbed Coating for DNAMicroarray Glass Slides, Anal. Chem., 2004, 76, 1352–1358.

36 L. Moiseev, C. R. Cantor, M. I. Aksun, M. Dogan,B. B. Goldberg, A. K. Swan and M. S. Unlu, Spectral Self-Interference Fluorescence Microscopy, J. Appl. Phys., 2004,96, 5311.

37 L. Moiseev, M. S. Unlu, A. K. Swan, B. B. Goldberg andC. R. Cantor, DNA Conformation on Surfaces Measured byFluorescence Self-Interference, Proc. Natl. Acad. Sci. U. S.A., 2006, 103, 2623–2628.

38 M. Kitsara, P. Petrou, D. Kontziampasis, K. Misiakos,E. Makarona, I. Raptis and K. Beltsios, Biomolecular LayerThickness Evaluation Using White Light ReectanceSpectroscopy, Microelectron. Eng., 2010, 87, 802–805.

39 P. S. Petrou, D. Ricklin, M. Zavali, I. Raptis, S. E. Kakabakos,K. Misiakos and J. D. Lambris, Real-Time Label-FreeDetection of Complement Activation Products in HumanSerum by White Light Reectance Spectroscopy, Biosens.Bioelectron., 2009, 24, 3359–3364.

40 E. Ozkumur, A. Yalçın, M. Cretich, C. A. Lopez,D. A. Bergstein, B. B. Goldberg, M. Chiari and M. S. Unlu,Quantication of DNA and Protein Adsorption by OpticalPhase Shi, Biosens. Bioelectron., 2009, 25, 167–172.

41 C. Cao, J. Zhang, X. Wen, S. L. Dodson, N. T. Dao,L. M. Wong, S. Wang, S. Li, A. T. Phan and Q. Xiong,Metamaterials-Based Label-Free Nanosensor forConformation and Affinity Biosensing, ACS Nano, 2013, 7,7583–7591.

42 H. M. Watkins, A. Vallee-Belisle, F. Ricci, D. E. Makarov andK. W. Plaxco, Entropic and Electrostatic Effects on theFolding Free Energy of a Surface-Attached Biomolecule: AnExperimental and Theoretical Study, J. Am. Chem. Soc.,2012, 134, 2120–2126.

43 P. Akkahat, W. Mekboonsonglarp, S. Kiatkamjornwong andV. P. Hoven, Surface-Graed Poly(Acrylic Acid) Brushes as aPrecursor Layer for Biosensing Applications: Effect of GraDensity and Swellability on the Detection Efficiency,Langmuir, 2012, 28, 5302–5311.

44 C. Xue, N. Yonet-Tanyeri, N. Brouette, M. Sferrazza,P. V. Braun and D. E. Leckband, Protein Adsorption onPoly(N-Isopropylacrylamide) Brushes: Dependence onGraing Density and Chain Collapse, Langmuir, 2011, 27,8810–8818.

45 B. Sanavio, D. Scaini, C. Grunwald, G. Legname, G. Scolesand L. Casalis, Oriented Immobilization of Prion ProteinDemonstrated via Precise Interfacial NanostructureMeasurements, ACS Nano, 2010, 4, 6607–6616.

46 M. Cretich, G. Pirri, F. Damin, I. Solinas and M. Chiari, ANew Polymeric Coating for Protein Microarrays, Anal.Biochem., 2004, 332, 67–74.

47 M. Cretich, G. di Carlo, R. Longhi, C. Gotti, N. Spinella,S. Coffa, C. Galati, L. Renna and M. Chiari, High

Analyst

Sensitivity Protein Assays on Microarray Silicon Slides,Anal. Chem., 2009, 81, 5197–5203.

48 L. Moiseev, Spectral Self-Interference FluorescenceMicroscopy and Its Applications in Biology, Dissertation,Boston University, 2003.

49 A. V. Dobrynin, Electrostatic Persistence Length ofSemiexible and Flexible Polyelectrolytes, Macromolecules,2005, 38, 9304–9314.

50 A. R. Khokhlov, A. Y. Grosberg and V. S. Pande, StatisticalPhysics of Macromolecules (Polymers and Complex Materials),American Institute of Physics, 1994th edn, 1997.

51 G. Gauglitz, Multiple Reectance Interference SpectroscopyMeasurements Made in Parallel for Binding Studies, Rev.Sci. Instrum., 2005, 76, 062224.

52 M. Dogan, A. Yalcn, S. Jain, M. B. Goldberg, A. K. Swan,M. S. Unlu and B. B. Goldberg, Spectral Self-InterferenceFluorescence Microscopy for Subcellular Imaging, IEEE J.Sel. Top. Quantum Electron., 2008, 14, 217–225.

53 G. S. Manning, Limiting Laws and Counterion Condensationin Polyelectrolyte Solutions. V. Further Development of theChemical Model, Biophys. Chem., 1978, 9, 65–70.

54 U. Rant, K. Arinaga, S. Fujita, N. Yokoyama, G. Abstreiter andM. Tornow, Electrical Manipulation of OligonucleotidesGraed to Charged Surfaces, Org. Biomol. Chem., 2006, 4,3448.

55 I. Y. Wong and N. A. Melosh, An Electrostatic Model for DNASurface Hybridization, Biophys. J., 2010, 98, 2954–2963.

56 M. Singh-Zocchi, S. Dixit, V. Ivanov and G. Zocchi, Single-Molecule Detection of DNA Hybridization, Proc. Natl. Acad.Sci. U. S. A., 2003, 100, 7605–7610.

57 B. Tinland, A. Pluen, J. Sturm and G. Weill, PersistenceLength of Single-Stranded DNA, Macromolecules, 1997, 30,5763–5765.

58 M. C. Murphy, I. Rasnik, W. Cheng, T. M. Lohman and T. Ha,Probing Single-Stranded DNA Conformational FlexibilityUsing Fluorescence Spectroscopy, Biophys. J., 2004, 86,2530–2537.

59 A. Hillisch, M. Lorenz and S. Diekmann, Recent Advances inFRET: Distance Determination in Protein–DNA Complexes,Curr. Opin. Struct. Biol., 2001, 11, 201–207.

60 A. Tsortos, G. Papadakis, K. Mitsakakis, K. A. Melzak andE. Gizeli, Quantitative Determination of Size and Shape ofSurface-Bound DNA Using an Acoustic Wave Sensor,Biophys. J., 2008, 94, 2706–2715.

61 X. Wang, X. Ren, K. Kahen, M. A. Hahn, M. Rajeswaran,S. Maccagnano-Zacher, J. Silcox, G. E. Cragg, A. L. Efrosand T. D. Krauss, Non-Blinking SemiconductorNanocrystals, Nature, 2009, 459, 686–689.

62 I. L. Medintz, H. T. Uyeda, E. R. Goldman and H. Mattoussi,Quantum Dot Bioconjugates for Imaging, Labelling andSensing, Nat. Mater., 2005, 4, 435–446.

63 T. M. Squires, R. J. Messinger and S. R. Manalis, Making ItStick: Convection, Reaction and Diffusion in Surface-BasedBiosensors, Nat. Biotechnol., 2008, 26, 417–426.

This journal is © The Royal Society of Chemistry 2014