Inflection Point of the Localized Surface Plasmon Resonance Peak: AGeneral Method to Improve the SensitivityPeng Chen,†,‡ Nhung Thi Tran,†,‡ Xinglin Wen,§ Qihua Xiong,§ and Bo Liedberg*,†,‡

†Centre for Biomimetic Sensor Science, School of Materials Science and Engineering, Nanyang Technological University, 50 NanyangDrive, Singapore 637553‡School of Materials Science and Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798§School of Physical and Mathematical Sciences, Division of Physics and Applied Physics, Nanyang Technological University, 21Nanyang Link, Singapore 637371

*S Supporting Information

ABSTRACT: The shift of the localized surface plasmon resonance (LSPR) spectrum is widely usedin bio- and chemical sensing. Traditionally, the shift is monitored at the peak maximum of theextinction spectrum. We demonstrate that the inflection point at the long wavelength side of the peakmaximum shows better refractive index sensitivity than the peak maximum. A consistent improvementin bulk refractive index sensitivity of 18−55% is observed for six different nanoparticles such asspherical particles of different sizes, nanostar and nanorods with different aspect ratios. Local refractiveindex changes induced by molecular adsorption confirm the superior performance of the method. Wecontribute this improvement in sensitivity to the change in shape of the LSPR peak in response to anincrease of the local refractive index. We further illustrate the advantage of using the inflection pointmethod for analyzing DNA adsorption on U-shaped metamaterials, and for using 17 nm spherical goldnanoparticles for detection of matrix metalloprotease 7 (MMP-7), a biomarker that is heavily up-regulated during certain cancers. With the inflection point, the limit of detection (LOD) for MMP-7 isimproved to 0.094 μg/mL from 0.22 μg/mL. This improvement may facilitate early diagnosis ofsalivary and colorectal cancers. We also envision that this generic method can be employed to trackminute optical responses in other analytical areas.

KEYWORDS: inflection point of LSPR peak, plasmonic nanoparticles, metamaterials, mathematic method, improved sensitivity

The LSPR resonance frequency or wavelength of nanosizednoble metal particles depends on their size, shape,

composition, interparticle distance, and the medium in whichthey are embedded/dissolved.1 Recently, LSPR of noble metalnanoparticles has attracted tremendous interest in bio- andchemical sensing.2 LSPR sensors for many target moleculesincluding DNAs,3 proteins,4 small biological molecules,5 metalions,6 enzymes,7 explosives,8 toxins,9 and gases10 have beendeveloped. Sensing with LSPR most often relies on the red shiftof the extinction spectrum induced by changes in eitherrefractive index or interparticle distance occurring in responseto molecular recognition and/or attachment/bridging.Although useful, the refractive index sensitivity of LSPRsensors is still several orders of magnitude smaller than thesensitivity obtained using propagating SPR sensors.11 As aresult, detecting molecules at ultralow concentrations usingLSPR is very demanding especially for molecules of lowmolecular weight. To solve this problem, numerous efforts havebeen made to improve the sensitivity of LSPR sensors. Oneapproach involves synthesis of novel nanomaterials with higherpolarizability such as nanorods,4a nanostars,12 nanoshells,13

mushrooms,14 nanorings,15 nanotriangles,16 nanodisks,17 nano-holes,18 alloy nanoparticles,10a,19 closely packed nanoparticles,20

plasmonic composites,21 and plasmonic nanoparticles of otherelements19,22 (a chemical/materials approach). Alternatively,

physicists also explored the potential of meta-materials23 orFano resonance24 by fabricating novel submicron plasmonicpatterns using lithography techniques (a physical approach).Although both efforts improve the refractive index sensitivity ofLSPR sensors to some extent, they also bring many problemssuch as high cost, low yield, and tedious processing. In addition,with more and more nanostructures being synthesized andtested, further improvements utilizing the aforementionedstrategies become very difficult. Therefore, a facile and generalmethod to improve the refractive index sensitivity is of highsignificance.In this contribution, we report a simple mathematical

approach by tracking shifts at inflection points of the LSPRspectrum, which is completely independent to both afore-mentioned chemical and physical approaches. The inflectionpoint approach is not an alternative but a complementarytechnique that can be used to further improve the performanceof LSPR assays. A few alternative mathematical methods toevaluate the peaks of SPR/LSPR have been reported before,25

such as centroid, curvature, and A/D ratio, which all

Received: October 12, 2016Accepted: January 30, 2017Published: January 30, 2017

Article

pubs.acs.org/acssensors

© 2017 American Chemical Society 235 DOI: 10.1021/acssensors.6b00633ACS Sens. 2017, 2, 235−242

contributed to an improved sensitivity, signal-to-noise, andLOD in bio- and chemical sensing.As mentioned before, the traditional way of monitoring

changes in the LSPR spectrum relies on the shift of the peakmaximum. However, we noticed that the magnitude of the shiftis not constant across the spectrum, because of the changes inthe shape of the LSPR peak (broadening, asymmetry). Thus, itis critical to reproducibly identify unique and easily tractablelocations in the LSPR spectrum that possess maximum shifts inresponse to refractive index changes/molecular interactions.We recognized two such points in addition to peak maximum,namely, the inflection points around the peak maximum.Mathematically, the inflection points appear as local maxima/minima in the first order derivative spectrum and as zero-crossings in the second order derivative spectrum. Therefore,they are easily located with the help of the first and secondorder derivative spectra (Figure 1a−c). We demonstrate hereinthe performance of the method by studying minute shiftinduced by refractive index changes in the bulk as well as on thesurface of gold nanoparticles and gold metamaterials.

■ EXPERIMENTAL SECTIONMaterials. Sodium citrate tribasic dehydrate (99%), gold(III)

chloride trihydrate (99.9%), silver nitrate (99%), ascorbic acid (99%),cetyltrimethylammonium bromide (96%), chitosan (low molecularweight), acetic acid (99.7%), and glycerol (99%) were purchased from

Sigma-Aldrich. Hydrochloric acid (37%) was purchased from MerckMillipore, Singapore. Spherical AuNPs with the size of 40 and 60 nmwere purchased from BBI solutions. Peptide (CALNN) was purchasedfrom GL Biochem China. Neutravidin was purchased from ThermoFisher Scientific Inc., USA. Purified oligonucleotide with a thiolmodifier (C6 S-S-GGT GGT GGT GGTTGT GGT GGT GGT GG)was purchased from Integrated DNA Technologies, Singapore.

Synthesis and Characterization. The synthesis of polypeptideJR2EC with the sequence of (NAADLEKAIEALEKHLEAKGPC-DAAQLEKQLEQAFEAFERAG) has been describe elsewhere.26 Goldnanoparticles (AuNPs) of size 13 and 17 nm were synthesized byreduction of HAuCl4 with sodium citrate following the procedurespublished elsewhere.27 Gold nanostar (AuNS) was prepared by theprevious protocol in the literature with little modification.28 Briefly,180 μL HAuCl4 (0.01 M), 50 μL AgNO3 (0.01 M), and 30 μL HCl (1M) were consecutively added to 3 mL of Milli-Q water in 20 mL glassvial. Subsequently, 100 μL of gold seed of 20 nm (sodium citratecapped) was added to the solution and the mixture was stirred forfurther 5 min. After that, 150 μL of ascorbic acid (1 wt %) was rapidlyadded. The color of the mixture immediately changed to a greenishblue color as the sign of the formation of AuNS. The reaction finishedwithin 30 s. To stabilize the as-prepared from truncation of the sharptip, chitosan solution (1 wt % in 1% v/v acetic acid, dissolved bysonication until getting transparent solution) was added dropwise intothe freshly prepared AuNS solution under gentle magnetic stirring.The final concentration of chitosan is 0.3% wt. Centrifugation andwash were repeated twice to remove the unbound chitosan. Goldnanorod (AuNR) of aspect ratio of ∼3.3 (AuNR1) was prepared bythe method published in the literature29 and AuNR with aspect ratio of

Figure 1. (a) Extinction, (b) first derivative, and (c) second derivative spectra of 40 nm AuNPs. The three points are indicated with red dots and thered dashed lines serves as a guide for the eye. Points 1 and 3 are two inflection points of the extinction spectrum around the peak maximum, whichshows a 0 value in the second order derivative spectrum. (d) Extinction spectra of 40 nm AuNPs exposed in solvents with refractive indices 1.333,1.362, 1.399, 1.432, and 1.460, respectively. (e) Shift of Point 1, Point 2, and Point 3 with refractive index. (f) Refractive index sensitivity of the threepoints.

ACS Sensors Article

DOI: 10.1021/acssensors.6b00633ACS Sens. 2017, 2, 235−242

236

∼4.7 (AuNR2) was prepared by the method published in theliterature.30

Fabrication of U-Shaped Metamaterials. The U-shapedmetamaterials were fabricated by e-beam lithography following theprocedure published elsewhere.31 Briefly, PMMA was spin-coated onthe ITO glass and then electron beam was used to define the structure.After exposure, the sample was developed in MIBK solution, followedby the deposition of metal. The metamaterials were obtained after alift-off process in acetone.Measurement of Spectra. Bulk refractive index was tuned from

1.333 to 1.46 by changing the fraction of glycerol and water. Localrefractive index change was obtained by the adsorption of peptide andneutravidin on the 60 nm AuNPs. The adsorption process wasconducted at 100 μM of the molecules dissolved in 10 mM sodiumcitrate buffer at pH 6 to prevent aggregation of AuNPs.9b Extinctionspectra were recorded in a PMMA cuvette containing 500 μL AuNPs.The transmission spectra of the metamaterials were collected using amicrospectrophotometer (Craic 20) in the range 400−2100 nm. Thetransmission spectra were measured without any polarizer.Adsorption of Molecules and MMP-7 Sensor. The peptide

CALNN and neutravidin were adsorbed to the AuNPs by introducingthe molecules of 100 μM concentration into the suspension of AuNPs.The thiolated DNA of 100 μM was first deprotected at the terminalthiol group by mixing with 150 mM dithiothreitol at roomtemperature for 2 h. Then the thiolated DNA was desalted andpurified using a NAP-5 column. After that, 0.4 μL DNA molecule (10μM) was spotted on clean metamaterials and incubated overnight in ahumidity chamber. The samples were subsequently washed with DNAfree water to remove any unbound DNA.Functionalization of the AuNPs with JR2EC peptide was performed

by incubation of the AuNPs in the 10 μM peptide solution at pH 6 (10mM sodium citrate) overnight. The unbound peptide was removed by5 repeated centrifugations followed by washing with PBS buffer at pH7.4. The MMP-7 mediated hydrolysis of the immobilized peptides wasperformed at 25 °C and real-time extinction spectra were recordedwith the Lambda 35 spectrometer.7a

Calculation of Derivative Spectrum. First and second orderderivative spectrum are calculated from the extinction spectrum using19 data points. The details of calculation methods are described in theprevious work.25b Point 2 was obtained from the 0 value of the firstderivative spectrum while Point 1 and Point 3 were obtained from the0 value of second order derivative spectrum; see Figure S1. Theinfluence of the number of points on the final noise in the secondorder derivative spectrum are also shown in Figure S1. Similarcalculation was done for the transmittance spectrum for themetamaterials. It is worthwhile to mention that performing the first/second derivative with the 19 points method produces spectra thatsignificantly reduces the propagation of noise.

■ RESULTS AND DISCUSSION

The extinction spectrum of 40 nm AuNPs and itscorresponding first order and second order derivative spectra25b

are shown in Figure 1a,b,c, respectively. The peak maximum(Point 2) and two inflection points (Points 1 and 3) flankingthe peak maximum (Point 2) are highlighted in Figure 1a,b,c byred dots. The red dashed lines serve as guidelines for the eye.The two inflection points are not obvious in the extinctionspectrum, but appear as local maxima/minima in the first orderderivative and have a 0 value in the second order derivativespectrum. Therefore, the inflection points can be easily trackedin the same way as the peak maximum in the extinctionspectrum. It is worthwhile to mention that calculating the first/second derivative with the 19 points method produces aspectrum that is almost as smooth as the original one. Toillustrate the shift of the three points with changes in bulkrefractive index, the extinction spectra of 40 nm AuNPs indifferent solvents are measured, see Figure 1d. The

corresponding first and second order derivative spectra areshown in Figure S1. The shifts of Point 1, Point 2, and Point 3with refractive index are summarized in Figure 1e. It is clearfrom Figure 1e that Point 3 shows the highest slope(sensitivity) with bulk refractive index. The refractive indexsensitivity of 40 nm AuNPs tracked at the three points isplotted in Figure 1f. The refractive index sensitivity for Point 1,Point 2, and Point 3 are 41.2, 61.8, and 91.4 nm/RIU,respectively. Interestingly, by simply tracking the shift of LSPRspectrum by Point 3, the refractive index sensitivity for 40 nmAuNPs could be improved by 48% with respect to the peakmaximum, Point 2.The LSPR spectra measured in solvents with different

refractive indices were carefully evaluated and we found that theLSPR peak becomes broader and more intense, and shifts tolonger wavelengths when the refractive index in thesurrounding solution increases. The broadening influences theinflection points and the peak maximum to a varying degree.For example, for the point at the right side of the peakmaximum (point 3, Scheme 1), the broadening of LSPR peak

will induce an additional shift to longer wavelengths, while forthe points at the left side of the LSPR peak (Point 1) it willinduce a negative shift (to shorter wavelengths; see Scheme 1).Therefore, the reason for the better sensitivity with inflectionpoint 3 is that it accounts for both the shift of the LSPR peakand the broadening, whereas the broadening counteracts theshift of point 1.To verify the observation of higher sensitivity of Point 3 over

Point 2, a few other gold nanoparticles with different sizes andshapes were tested. Figure 2a shows the extinction spectra ofspherical AuNPs of 13, 40, and 60 nm diameter, Au nanostar(AuNS), and Au nanorods (AuNR) with aspect ratios of ∼3.3and ∼4.7, respectively. The TEM images of these nanostruc-tures are shown in Figure 2b−g, respectively. Extinction spectraof all the nanostructures in Figure 2 were recorded in a mixtureof glycerol and water with different refractive indices (seeSupporting Information, Figure S2).The refractive index sensitivity of all the nanostructures

tracked at Point 3 (black) and Point 2 (white) are summarizedin Figure 3a. Obviously, for all the nanostructures tested, Point

Scheme 1. Illustration of the Sources Contributing to Shift ofLSPR Spectra (Blue Spectrum to Red Spectrum)a

a(a) Pure shift of spectrum; (b) broadening of spectrum; (c)asymmetric broadening of spectrum; (d) typical scenario in LSPRsensing, where a shift, increase in height, and broadening coexist. Shiftsat points 1, 2, and 3 are illustrated visually and with equations.

ACS Sensors Article

DOI: 10.1021/acssensors.6b00633ACS Sens. 2017, 2, 235−242

237

3 demonstrates superior performance in refractive sensitivitywith respect to Point 2. The percentage of improvement isplotted in Figure 3b. A consistent improvement in refractiveindex sensitivity from 18% to 55% was observed for all sixdifferent nanostructures. With these observations, we proposethat tracking at the inflection point (Point 3) could be used as a

general method to enhance the refractive index sensitivity formany different plasmonic nanostructures. It is worthwhile tomention that in biosensing, the performance is determined notonly by the sensitivity, but also by the noise of the sensor. Asystematic evaluation of the noise using 19 points method forthe three points are shown in Supporting Information Table SI,Table SII, and Table SIII. From our study, there is no differenceobserved in the noise between the three points, which weattribute to the high quality (low noise level) of the originalUV−vis data and the 19 points method. Therefore, theimprovement in LOD or S/N ratio of our sensor is representedby the improvement in sensitivity.In biological sensing, the change in local refractive index

induced by molecular binding is of significant interest.Therefore, in addition to the bulk refractive index, the responsefrom local refractive index change was studied using twomolecules, peptide (CALNN) and neutravidin. The CALNNpeptide consists of 5 amino acids and it has a cysteine group atthe end, allowing for end-point attachment to AuNPs. Themolecular weights of the peptide and neutravidin are 506 g/moland ∼60 000 g/mol, respectively. Both molecules arecommonly used to functionalize AuNPs for biosensingapplications.9b,32 The 60 nm AuNPs capped with sodiumcitrate were used for the experiments with the peptide andneutravidin in solution. Figure 4a shows the normalized

extinction spectra of the AuNPs before (red) and afteradsorption of peptide (blue) and neutravidin (black),respectively. The neutravidin molecule induces a larger redshift of the LSPR spectrum than peptide as expected because ofits higher molecular weight. The shift upon adsorption ofpeptide and neutravidin at Point 2 (white) and Point 3 (black)is plotted in Figure 4b. For both molecules, tracking at Point 3demonstrates a larger shift than that of Point 2 and the

Figure 2. (a) Extinction spectra of AuNPs of 13 nm (red), 40 nm(blue), 60 nm (black) size and AuNS (green), AuNR1 (cyan), andAuNR2 (pink), respectively. The labeling of the peak in (a) refers tothe TEM images (b) to (g). TEM images of AuNPs of 13 nm (b), 40nm (c), 60 nm in diameter; (d) AuNS (e), AuNR1 aspect ratio 3.1 (f),and AuNR2 aspect ratio 4.7 (g), respectively. Scale bars for all TEMimages are 50 nm.

Figure 3. (a) Refractive index sensitivity of Point 2 (white) and point3 (black) for all the nanoparticles shown in Figure 2. (b) %improvement of Point 3 over Point 2 for all nanoparticles.

Figure 4. (a) Normalized extinction spectra of 60 nm AuNPs withsodium citrate layer (red) and AuNPs functionalized with CALNN(blue) and Neutravidin (black). (b) Shift of Point 2 (white) and Point3 (black) upon adsorption of molecules.

ACS Sensors Article

DOI: 10.1021/acssensors.6b00633ACS Sens. 2017, 2, 235−242

238

improvement if especially pronounced for the low molecularweight peptide CALNN.As mentioned before, in addition to the traditional plasmonic

nanoparticles synthesized by the bottom-up approach, meta-materials have also attracted tremendous interest in biosensingdue to the improved sensitivity.33 Metamaterials are virtuallywell-defined, densely packed periodic structures/patterns,whose optical properties go beyond the limitations of thenaturally occurring materials or composites.34 To furtherexplore the applicability of this inflection point method, wetested it with gold metamaterials. A SEM image of the U-shaped nanosized metamaterial made of gold is shown in Figure5a. The transmission spectra of the metamaterials before(black) and after (red) modification with DNA molecules areshown in Figure 5b.

The DNA molecules were chemically adsorbed onto the goldsurface via the thiol group at the terminal. The adsorption ofDNA shift the transmittance spectrum of metamaterials towardthe red direction. The shifts of the spectra were monitored atthe two inflection points around the valley and the valley(minimum) (Figure 5c). Interestingly again, with the sameDNA adsorption, the amounts of shift at point 1 (left inflection

point), point 2 (minimum), and point 3 (right inflection point)are calculated to be 39.9, 47.6, and 54.8 nm, respectively. Thisobservation is consistent with that observed on plasmonicnanoparticles that the inflection point at the right side of thepeak of the extinction/valley of transmittance spectrum yields alarger shift. Therefore, we believe that this novel mathematicalmethod opens an avenue to enhance the sensitivity ofplasmonic sensors.From these observations, it is evident that tracking the shift

at Point 3 is better than Point 2, regardless of the type ofnanoparticles and the cause of the shift (bulk or monolayerrefractive index change). Therefore, it is reasonable to expectthat the method of tracking shift at Point 3 could further yield abetter performance for applications in biological sensing.An assay for matrix metalloproteinase 7 (MMP-7) previously

developed in our laboratory7a was employed to evaluate theperformance by tracking Point 3. MMP-7 plays a significantrole in cell development and in the metastasis of cancers.35 It issignificantly up-regulated in many cancers, e.g., lung cancer,gastric cancer, colorectal cancer, and bladder cancer.36 Theconcentration of MMP-7 also varies with the stages of cancerdevelopment; therefore, potentially it is a useful early biomarkerfor cancer prognosis.37 In addition, it is also an important targetfor anticancer drugs.38 A synthetic polypeptide (JR2EC) withtwo cleavage sites (A-L and A-Q-L) for MMP-7 had beensynthesized and its sequence was modified to enable reliableimmobilization to AuNPs through a cysteine (SH-containing)residue at position 22.7a,37 Digestion of the polypeptide byMMP-7 reduces its net charge from −5 to −1, which in turnreduces the repulsive force between AuNPs, resulting inmassive aggregation of the AuNPs (Figure 6a). The red shiftof LSPR spectrum upon aggregation of AuNPs was used toquantify the concentration and activity of MMP-7. Figure 6bshows the extinction spectra of JR2EC functionalized AuNPsincubated with different concentrations of MMP-7 for 4 min.The shift of the extinction spectrum induced by MMP-7 wasanalyzed with both Point 3 and Point 2. Figure 6c shows thekinetics of the shifts of extinction spectrum at Point 3 (solidsquares) and Point 2 (open squares) incubated with MMP-7 ata concentration of 2 μg/mL.Both response curves were fitted with exponential decay

function (red curves). It is clear that Point 3 shows largerresponses during the catalytic event. Figure 6d shows the shiftof extinction spectrum at Point 3 (solid squares) and Point 2(open squares) for different concentrations of MMP-7incubated for 4 min. Both calibration curves were fitted wellwith exponential decay function (red curves). From the twocalibration curves, Point 3 clearly demonstrates a largerresponse than Point 2 across all concentrations of MMP-7.The LOD (limit of detection) calculated using 3 times standarddeviation (0.15 nm) for Point 3 and Point 2 are 0.094 and 0.22μg/mL, respectively. The noise for peak maximum andinflection point are calculated from the 3 times of the standardvariations of the LSPR spectra of blank samples. Both peakmaximum and inflection point show a standard deviation of∼0.05 nm. This is an encouraging finding given that theconcentration of MMP-7 in cancer patients typically variesbetween 0.01 and a few μg/mL.36,39 Therefore, the lowering ofthe LOD offers a significant advantage in early diagnosis.Further improvement on LOD can be made by combining thisanalysis method with better sensing platforms.

Figure 5. (a) SEM image of the U-shaped metamaterial, scale bar is 1μm and (b) transmittance spectra of the U-shaped metamaterial before(black) and after (red) exposure to DNA. (c) Shift at different pointsinduced by DNA adsorption calculated from the spectra in (b).

ACS Sensors Article

DOI: 10.1021/acssensors.6b00633ACS Sens. 2017, 2, 235−242

239

■ CONCLUSIONSWe demonstrate that the inflection point at the red side of theLSPR peak maximum (Point 3) shows higher refractive indexsensitivity than the traditionally used peak maximum (Point 2).A consistent improvement in bulk refractive index sensitivity of18−55% was observed for six different nanoparticles. Localrefractive index changes induced by molecular adsorption togold nanoparticles and gold metamaterials also confirm thesuperior performance of the inflection point. The betterperformance of the inflection point at the red side (Point 3)originates from the change in shape of LSPR peak in responseto an increase of the local refractive index. We furtherdemonstrate with a colorimetric sensor for Matrix Metal-loProteinase 7 that the inflection point approach offers a LODis 0.094 μg/mL while with the peak maximum shift gives aLOD of 0.22 μg/mL. This improvement in LOD is veryencouraging as it offers early diagnosis of certain cancers. Webelieve that this method also can be used as a general methodfor tracking minute shifts/responses in other analyticalapplications.

■ ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acssen-sors.6b00633.

First and second order derivative spectra of 40 nmAuNPs in different solvents, extinction spectra of all thenanoparticles measured in solvents with differentrefractive indices, noise analysis (PDF)

■ AUTHOR INFORMATIONCorresponding Author* E-mail: bliedberg@ntu.edu.sg. Tel: (+65) 6316 2957.ORCIDBo Liedberg: 0000-0003-2883-6953NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work was supported by the School of Materials Scienceand Engineering and the Provost office and the Nanomedicineinstitute and iFood institute, Nanyang Technological Univer-sity, Singapore and Ministry of Education, Singapore throughMOE-2014-T1-001-133.

■ REFERENCES(1) Odom, T. W.; Schatz, G. C. Introduction to Plasmonics. Chem.Rev. 2011, 111 (6), 3667−3668.(2) (a) Anker, J. N.; Hall, W. P.; Lyandres, O.; Shah, N. C.; Zhao, J.;Van Duyne, R. P. Biosensing with plasmonic nanosensors. Nat. Mater.2008, 7 (6), 442−453. (b) Mayer, K. M.; Hafner, J. H. LocalizedSurface Plasmon Resonance Sensors. Chem. Rev. 2011, 111 (6), 3828−3857.(3) (a) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.;Mirkin, C. A. Selective colorimetric detection of polynucleotides basedon the distance-dependent optical properties of gold nanoparticles.Science 1997, 277 (5329), 1078. (b) Cheng, X. R.; Hau, B. Y. H.; Endo,T.; Kerman, K. Au nanoparticle-modified DNA sensor based onsimultaneous electrochemical impedance spectroscopy and localizedsurface plasmon resonance. Biosens. Bioelectron. 2014, 53 (0), 513−518.

Figure 6. (a) Schematic illustration of the principle of colorimetric sensor for MMP-7. (b) Normalized extinction spectra of peptide functionalizedAuNPs incubated with the MMP-7 concentrations of 0 (black), 0.25 (red), 0.5 (blue), 2 (green), and 4 μg/mL (pink), respectively. (c) Shift of Point3 (solid squares) and Point 2 (open squares) with 2 μg/mL MMP-7 and exponential decay fitted curves (red lines); R2 for Point 2 and Point 3 are0.9961 and 0.9912, respectively. (d) Shift of Point 3 (solid squares) and Point 2 (open squares) with concentrations of MMP-7 and exponentialdecay fitted curves (red lines) over time; R2 for Point 2 and Point 3 are 0.9821 and 0.9906, respectively. The error bar is calculated from threeindependent experiments.

ACS Sensors Article

DOI: 10.1021/acssensors.6b00633ACS Sens. 2017, 2, 235−242

240

(4) (a) Lambertz, C.; Martos, A.; Henkel, A.; Neiser, A.; Kliesch, T.-T.; Janshoff, A.; Schwille, P.; Sonnichsen, C. Single Particle PlasmonSensors as Label-Free Technique To Monitor MinDE Protein WavePropagation on Membranes. Nano Lett. 2016, 16 (6), 3540−3544.(b) Jeong, H.-H.; Erdene, N.; Park, J.-H.; Jeong, D.-H.; Lee, H.-Y.;Lee, S.-K. Real-time label-free immunoassay of interferon-gamma andprostate-specific antigen using a Fiber-Optic Localized SurfacePlasmon Resonance sensor. Biosens. Bioelectron. 2013, 39 (1), 346−351. (c) Zhou, W.; Ma, Y.; Yang, H.; Ding, Y.; Luo, X. A label-freebiosensor based on silver nanoparticles array for clinical detection ofserum p53 in head and neck squamous cell carcinoma. Int. J. Nanomed.2011, 6 (1), 381−386. (d) Acimovic, S. S.; Ortega, M. A.; Sanz, V.;Berthelot, J.; Garcia-Cordero, J. L.; Renger, J.; Maerkl, S. J.; Kreuzer,M. P.; Quidant, R. LSPR Chip for Parallel, Rapid, and SensitiveDetection of Cancer Markers in Serum. Nano Lett. 2014, 14 (5),2636−2641. (e) Pallarola, D.; Schneckenburger, M.; Spatz, J. P.;Pacholski, C. Real-time monitoring of electrochemical controlledprotein adsorption by a plasmonic nanowire based sensor. Chem.Commun. 2013, 49 (75), 8326−8328. (f) Tan, Y. N.; Su, X.; Zhu, Y.;Lee, J. Y. Sensing of transcription factor through controlled-assemblyof metal nanoparticles modified with segmented DNA elements. ACSNano 2010, 4 (9), 5101−5110. (g) Liu, X. H.; Wang, Y.; Chen, P.;McCadden, A.; Palaniappan, A.; Zhang, J.; Liedberg, B. Peptidefunctionalized gold nanoparticles with optimized particle size andconcentration for colorimetric assay development - detection ofcardiac troponin I. ACS Sensors 2016, 1, 1416.(5) (a) Xie, L.; Yan, X.; Du, Y. An aptamer based wall-less LSPRarray chip for label-free and high throughput detection ofbiomolecules. Biosens. Bioelectron. 2014, 53 (0), 58−64. (b) Brigo,L.; Michieli, N.; Artiglia, L.; Scian, C.; Rizzi, G. A.; Granozzi, G.;Mattei, G.; Martucci, A.; Brusatin, G. Silver Nanoprism ArraysCoupled to Functional Hybrid Films for Localized Surface PlasmonResonance-Based Detection of Aromatic Hydrocarbons. ACS Appl.Mater. Interfaces 2014, 6 (10), 7773−7781.(6) (a) Hall, W. P.; Anker, J. N.; Lin, Y.; Modica, J.; Mrksich, M.; VanDuyne, R. P. A Calcium-Modulated Plasmonic Switch. J. Am. Chem.Soc. 2008, 130 (18), 5836−5837. (b) Du, J.; Zhu, B.; Chen, X. Urinefor Plasmonic Nanoparticle-Based Colorimetric Detection of MercuryIon. Small 2013, 9 (24), 4104−4111.(7) (a) Chen, P.; Selegard, R.; Aili, D.; Liedberg, B. PeptideFunctionalized Gold Nanoparticles for Colorimetric Detection ofMatrilysin (MMP-7) Activity. Nanoscale 2013, 5, 8973−8976.(b) Chandrawati, R.; Stevens, M. M. Controlled assembly ofpeptide-functionalized gold nanoparticles for label-free detection ofblood coagulation Factor XIII activity. Chem. Commun. 2014, 50,5431−5434.(8) Jiang, Y.; Zhao, H.; Zhu, N.; Lin, Y.; Yu, P.; Mao, L. A SimpleAssay for Direct Colorimetric Visualization of Trinitrotoluene atPicomolar Levels Using Gold Nanoparticles. Angew. Chem. 2008, 120(45), 8729−8732.(9) (a) Sun, J.; Ge, J.; Liu, W.; Wang, X.; Fan, Z.; Zhao, W.; Zhang,H.; Wang, P.; Lee, S.-T. A facile assay for direct colorimetricvisualization of lipopolysaccharides at low nanomolar level. Nano Res.2012, 5 (7), 486−493. (b) Liu, X.; Wang, Y.; Chen, P.; Wang, Y.;Zhang, J.; Aili, D.; Liedberg, B. Biofunctionalized Gold Nanoparticlesfor Colorimetric Sensing of Botulinum Neurotoxin A Light Chain.Anal. Chem. 2014, 86 (5), 2345−2352.(10) (a) Wadell, C.; Nugroho, F. A. A.; Lidstrom, E.; Iandolo, B.;Wagner, J. B.; Langhammer, C. Hysteresis-Free Nanoplasmonic Pd−Au Alloy Hydrogen Sensors. Nano Lett. 2015, 15 (5), 3563−3570.(b) Nugroho, F. A. A.; Xu, C.; Hedin, N.; Langhammer, C. UV−Visible and Plasmonic Nanospectroscopy of the CO2 AdsorptionEnergetics in a Microporous Polymer. Anal. Chem. 2015, 87 (20),10161−10165.(11) (a) Liedberg, B.; Nylander, C.; Lunstrom, I. Surface plasmonresonance for gas detection and biosensing. Sens. Actuators 1983, 4,299−304. (b) Haes, A.; Van Duyne, R. A unified view of propagatingand localized surface plasmon resonance biosensors. Anal. Bioanal.Chem. 2004, 379 (7), 920−930.

(12) Liu, Y.; Carpenter, A.; Yuan, H.; Zhou, Z.; Zalutsky, M.;Vaidyanathan, G.; Yan, H.; Vo-Dinh, T. Gold nanostar as theranosticprobe for brain tumor sensitive PET-optical imaging and image-guidedspecific photothermal therapy. Cancer Res. 2016, 76, 4213−4213.(13) Kawawaki, T.; Shinjo, N.; Tatsuma, T. Backward-scattering-based localized surface plasmon resonance sensors with goldnanospheres and nanoshells. Anal. Sci. 2016, 32 (3), 271−274.(14) Shen, Y.; Zhou, J.; Liu, T.; Tao, Y.; Jiang, R.; Liu, M.; Xiao, G.;Zhu, J.; Zhou, Z.-K.; Wang, X.; Jin, C.; Wang, J. Plasmonic goldmushroom arrays with refractive index sensing figures of meritapproaching the theoretical limit. Nat. Commun. 2013, 4, 2381.(15) Im, H.; Bantz, K. C.; Lee, S. H.; Johnson, T. W.; Haynes, C. L.;Oh, S.-H. Self-Assembled Plasmonic Nanoring Cavity Arrays for SERSand LSPR Biosensing. Adv. Mater. 2013, 25 (19), 2678−2685.(16) Martinsson, E.; Shahjamali, M. M.; Enander, K.; Boey, F.; Xue,C.; Aili, D.; Liedberg, B. Local Refractive Index Sensing Based on EdgeGold-Coated Silver Nanoprisms. J. Phys. Chem. C 2013, 117 (44),23148−23154.(17) Jackman, J. A.; Spackova, B.; Linardy, E.; Kim, M. C.; Yoon, B.K.; Homola, J.; Cho, N.-J. Nanoplasmonic ruler to measure lipidvesicle deformation. Chem. Commun. 2016, 52 (1), 76−79.(18) Jackman, J. A.; Linardy, E.; Yoo, D.; Seo, J.; Ng, W. B.; Klemme,D. J.; Wittenberg, N. J.; Oh, S.-H.; Cho, N.-J. Plasmonic NanoholeSensor for Capturing Single Virus-Like Particles toward Virucidal DrugEvaluation. Small 2016, 12 (9), 1159−1166.(19) Nugroho, F. A. A.; Iandolo, B.; Wagner, J. B.; Langhammer, C.Bottom-Up Nanofabrication of Supported Noble Metal AlloyNanoparticle Arrays for Plasmonics. ACS Nano 2016, 10 (2), 2871−2879.(20) Martinsson, E.; Sepulveda, B.; Chen, P.; Elfwing, A.; Liedberg,B.; Aili, D. Optimizing the Refractive Index Sensitivity of PlasmonicallyCoupled Gold Nanoparticles. Plasmonics 2014, 9 (4), 773−780.(21) Ferhan, A. R.; Kim, D.-H. Nanoparticle polymer composites onsolid substrates for plasmonic sensing applications. Nano Today 2016,11 (4), 415−434.(22) (a) Olson, J.; Manjavacas, A.; Basu, T.; Huang, D.; Schlather, A.E.; Zheng, B.; Halas, N. J.; Nordlander, P.; Link, S. High ChromaticityAluminum Plasmonic Pixels for Active Liquid Crystal Displays. ACSNano 2016, 10 (1), 1108−1117. (b) Pirzadeh, Z.; Pakizeh, T.;Miljkovic, V.; Langhammer, C.; Dmitriev, A. Plasmon−InterbandCoupling in Nickel Nanoantennas. ACS Photonics 2014, 1 (3), 158−162. (c) Sugawa, K.; Tahara, H.; Yamashita, A.; Otsuki, J.; Sagara, T.;Harumoto, T.; Yanagida, S. Refractive Index Susceptibility of thePlasmonic Palladium Nanoparticle: Potential as the Third PlasmonicSensing Material. ACS Nano 2015, 9 (2), 1895−1904.(23) (a) Sreekanth, K. V.; Alapan, Y.; ElKabbash, M.; Ilker, E.;Hinczewski, M.; Gurkan, U. A.; De Luca, A.; Strangi, G. Extremesensitivity biosensing platform based on hyperbolic metamaterials. Nat.Mater. 2016, 15, 621−627. (b) Xu, X.; Peng, B.; Li, D.; Zhang, J.;Wong, L. M.; Zhang, Q.; Wang, S.; Xiong, Q. Flexible Visible−InfraredMetamaterials and Their Applications in Highly Sensitive Chemicaland Biological Sensing. Nano Lett. 2011, 11 (8), 3232−3238.(c) Verellen, N.; Van Dorpe, P.; Huang, C.; Lodewijks, K.;Vandenbosch, G. A. E.; Lagae, L.; Moshchalkov, V. V. Plasmon LineShaping Using Nanocrosses for High Sensitivity Localized SurfacePlasmon Resonance Sensing. Nano Lett. 2011, 11 (2), 391−397.(d) Rodríguez-Lorenzo, L.; de La Rica, R.; Alvarez-Puebla, R. A.; Liz-Marzan, L. M.; Stevens, M. M. Plasmonic nanosensors with inversesensitivity by means of enzyme-guided crystal growth. Nat. Mater.2012, 11 (7), 604−607.(24) King, N. S.; Liu, L.; Yang, X.; Cerjan, B.; Everitt, H. O.;Nordlander, P.; Halas, N. J. Fano Resonant Aluminum Nanoclustersfor Plasmonic Colorimetric Sensing. ACS Nano 2015, 9 (11), 10628−10636.(25) (a) Nusz, G. J.; Marinakos, S. M.; Curry, A. C.; Dahlin, A.;Hook, F.; Wax, A.; Chilkoti, A. Label-Free Plasmonic Detection ofBiomolecular Binding by a Single Gold Nanorod. Anal. Chem. 2008,80 (4), 984−989. (b) Chen, P.; Liedberg, B. Curvature of theLocalized Surface Plasmon Resonance Peak. Anal. Chem. 2014, 86

ACS Sensors Article

DOI: 10.1021/acssensors.6b00633ACS Sens. 2017, 2, 235−242

241

(15), 7399−7405. (c) Aili, D.; Gryko, P.; Sepulveda, B.; Dick, J. A. G.;Kirby, N.; Heenan, R.; Baltzer, L.; Liedberg, B.; Ryan, M. P.; Stevens,M. M. Polypeptide Folding-Mediated Tuning of the Optical andStructural Properties of Gold Nanoparticle Assemblies. Nano Lett.2011, 11, 5564−5573. (d) Johansen, K.; Stalberg, R.; Lundstrom, I.;Liedberg, B. Surface plasmon resonance: instrumental resolution usingphoto diode arrays. Meas. Sci. Technol. 2000, 11 (11), 1630.(e) Johansen, K.; Lundstrom, I.; Liedberg, B. Sensitivity deviation:instrumental linearity errors that influence concentration analyses andkinetic evaluation of biomolecular interactions. Biosens. Bioelectron.2000, 15 (9), 503−509.(26) Aili, D.; Enander, K.; Rydberg, J.; Lundstrom, I.; Baltzer, L.;Liedberg, B. Aggregation-induced folding of a de novo designedpolypeptide immobilized on gold nanoparticles. J. Am. Chem. Soc.2006, 128 (7), 2194−2195.(27) Carpay, F.; Cense, W. Controlled nucleation for the regulationof the particle size in monodisperse gold suspensions. Nature, Phys. Sci.1973, 241, 19−20.(28) Yuan, H.; Ma, W.; Chen, C.; Zhao, J.; Liu, J.; Zhu, H.; Gao, X.Shape and SPR Evolution of Thorny Gold Nanoparticles Promoted bySilver Ions. Chem. Mater. 2007, 19 (7), 1592−1600.(29) Sau, T. K.; Murphy, C. J. Room Temperature, High-YieldSynthesis of Multiple Shapes of Gold Nanoparticles in AqueousSolution. J. Am. Chem. Soc. 2004, 126 (28), 8648−8649.(30) Chen, H.; Kou, X.; Yang, Z.; Ni, W.; Wang, J. Shape- and Size-Dependent Refractive Index Sensitivity of Gold Nanoparticles.Langmuir 2008, 24 (10), 5233−5237.(31) Cao, C.; Zhang, J.; Wen, X.; Dodson, S. L.; Dao, N. T.; Wong,L. M.; Wang, S.; Li, S.; Phan, A. T. a.; Xiong, Q. Metamaterials-basedlabel-free nanosensor for conformation and affinity biosensing. ACSNano 2013, 7 (9), 7583−7591.(32) Levy, R.; Thanh, N. T.; Doty, R. C.; Hussain, I.; Nichols, R. J.;Schiffrin, D. J.; Brust, M.; Fernig, D. G. Rational and combinatorialdesign of peptide capping ligands for gold nanoparticles. J. Am. Chem.Soc. 2004, 126 (32), 10076−10084.(33) (a) Kabashin, A.; Evans, P.; Pastkovsky, S.; Hendren, W.; Wurtz,G.; Atkinson, R.; Pollard, R.; Podolskiy, V.; Zayats, A. Plasmonicnanorod metamaterials for biosensing. Nat. Mater. 2009, 8 (11), 867−871. (b) Pryce, I. M.; Kelaita, Y. A.; Aydin, K.; Atwater, H. A.Compliant metamaterials for resonantly enhanced infrared absorptionspectroscopy and refractive index sensing. ACS Nano 2011, 5 (10),8167−8174. (c) Wang, S.; Zhang, Y.; Zhang, R.; Yu, H.; Zhang, H.;Xiong, Q. High-Order Nonlinearity of Surface Plasmon Resonance inAu Nanoparticles: Paradoxical Combination of Saturable and Reverse-Saturable Absorption. Adv. Opt. Mater. 2015, 3 (10), 1342−1348.(d) He, Y.; Lawrence, K.; Ingram, W.; Zhao, Y. Circular dichroismbased refractive index sensing using chiral metamaterials. Chem.Commun. 2016, 52 (10), 2047−2050.(34) (a) Soukoulis, C. M.; Wegener, M. Past achievements and futurechallenges in the development of three-dimensional photonicmetamaterials. Nat. Photonics 2011, 5 (9), 523−530. (b) RoyChoud-hury, S.; Rawat, V.; Jalal, A. H.; Kale, S.; Bhansali, S. Recent advancesin metamaterial split-ring-resonator circuits as biosensors andtherapeutic agents. Biosens. Bioelectron. 2016, 86, 595−608.(35) Egeblad, M.; Werb, Z. New functions for the matrixmetalloproteinases in cancer progression. Nat. Rev. Cancer 2002, 2(3), 161−174.(36) (a) Sier, C. F.; Hawinkels, L. J.; Zijlmans, H. J.; Zuidwijk, K.; deJonge-Muller, E. S.; Ferreira, V.; Hanemaaijer, R.; Mulder-Stapel, A. A.;Kenter, G. G.; Verspaget, H. W. Endothelium specific matrilysin(MMP-7) expression in human cancers. Matrix Biol. 2007, 27 (3),267−271. (b) Aitken, K. J.; Bagli, D. J. The bladder extracellularmatrix. Part I: architecture, development and disease. Nat. Rev. Urol.2009, 6 (11), 596−611. (c) Maurel, J.; Nadal, C.; Garcia-Albeniz, X.;Gallego, R.; Carcereny, E.; Almendro, V.; Marmol, M.; Gallardo, E.;Maria Auge, J.; Longaron, R.; et al. Serum matrix metalloproteinase 7levels identifies poor prognosis advanced colorectal cancer patients.Int. J. Cancer 2007, 121 (5), 1066−1071.

(37) Chen, H.; Chen, P.; Huang, J.; Selegard, R.; Platt, M.;Palaniappan, A.; Aili, D.; Tok, A. I. Y.; Liedberg, B. Detection ofMatrilysin Activity Using Polypeptide Functionalized ReducedGraphene Oxide Field-Effect Transistor Sensor. Anal. Chem. 2016,88 (6), 2994−2998.(38) Sparano, J. A.; Bernardo, P.; Stephenson, P.; Gradishar, W. J.;Ingle, J. N.; Zucker, S.; Davidson, N. E. Randomized phase III trial ofmarimastat versus placebo in patients with metastatic breast cancerwho have responding or stable disease after first-line chemotherapy:Eastern Cooperative Oncology Group trial E2196. J. Clin. Oncol. 2004,22 (23), 4683−4690.(39) Ress, C.; Tschoner, A.; Ciardi, C.; Laimer, M. W.; Engl, J. W.;Sturm, W.; Weiss, H.; Tilg, H.; Ebenbichler, C. F.; Patsch, J. R.Influence of significant weight loss on serum matrix metalloproteinase(MMP)-7 levels. Eur. Cytokine Network 2010, 21, 65−70.

ACS Sensors Article

DOI: 10.1021/acssensors.6b00633ACS Sens. 2017, 2, 235−242

242