Patterned Plasmonic Nanoparticle Arrays for Microfluidic and

Subscriber access provided by GAZI UNIV

Article

Patterned Plasmonic Nanoparticle Arrays For Microfluidic And Multiplexed Biological Assays Jie He, Michelle Boegli, Ian R Bruzas, William Lum, and Laura B. Sagle Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b02870 • Publication Date (Web): 23 Oct 2015 Downloaded from http://pubs.acs.org on October 28, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Patterned Plasmonic Nanoparticle Arrays For Microfluidic And Multiplexed Biological Assays Jie He, Michelle Boegli, Ian Bruzas, William Lum, and Laura Sagle* Department of Chemistry, College of Arts and Sciences, University of Cincinnati, 301 West Clifton Court, Cincinnati OH 45221-0172 *Corresponding author Tel: +1 513 556 1034; Fax: +1 513 556 9239. E-mail: [email protected]

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 23

ABSTRACT

For applications ranging from medical diagnostics and drug screening to chemical and biological warfare detection, inexpensive, rapid-readout, portable devices are required. Localized surface plasmon resonance (LSPR) technologies show substantial promise towards meeting these goals, but the generation of portable, multiplexed and/or microfluidic devices incorporating sensitive nanoparticle arrays is only in its infancy. Herein, we have combined photolithography with Hole Mask Colloidal lithography to pattern uniform nanoparticle arrays for both microfluidic and multiplexed devices.

The first proof-of-concept study is carried out with 5 and 7-channel

microfluidic devices to acquire one-shot binding curves and protein binding kinetic data. The second proof-of-concept study involved the fabrication of a 96-spot plate that can be inserted into a standard plate reader for the multiplexed detection of protein binding. This versatile fabrication technique should prove useful in next generation chips for bioassays and genetic screening.

Keywords: point-of-care diagnostics, microfluidic, multiplexed, LSPR, patterned nanoparticle arrays, screening


Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION Current point-of-care diagnostics often require simultaneous testing of several biomarkers to reduce the risk of false diagnosis.1, 2 Multiplexed and microfluidic devices provide platforms with which to do these assays, often reducing sample drift, reaction time and minimizing sample volume. To extend these technologies to resource limited environments, additional requirements must be met, such as simple, fast readout in addition to a robust, inexpensive and portable platform.3 Many of the currently used multiplexed platforms, such as fluorescence and electrochemical assays, are prohibitively expensive and too fragile for use in the field.4, 5 Localized surface plasmon resonance (LSPR) biosensing offers label-free, sensitive, colorimetric detection at a relatively low cost.6,

7

With sensitivities comparable to surface

plasmon resonance (SPR) and quartz crystal microbalance (QCM), LSPR offers colored, transparent substrates which can be measured using simple instrumentation or even by eye.8, 9 Moreover, this nanoparticle-based technology is particularly amenable to miniaturized devices since the sensing elements themselves are small.

It is therefore ideal to combine LSPR

biosensing with miniaturized on-chip devices for portable point-of-care diagnostics. Towards this goal, recent studies have combined SPR and LSPR-based sensors into miniaturized, on-chip devices. Multiplexed spot plates, composed of planar gold and core-shell gold structures, have been fabricated and successfully applied towards DNA hybridization and biomarker characterization.10,

11

Moreover, recent reports have utilized multiplexed and

microfluidic gold and silver nanohole arrays, which are sensitive, transparent substrates capable of amplified detection particularly when substrates are flowed through the nanoholes.12,

13

In 3

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 23

addition, encasing the nanohole arrays in silica allow not only LSPR substrates to be interfaced with supported lipid bilayers, but also enable facile microfluidic device fabrication.14 However, the detection of molecular binding events of these substrates mentioned above often involves confocal microscopy and projection onto a CCD camera, which is not commonly accessible in many research laboratories. Lastly, an unprecedented degree of multiplexing has been created using single nanoparticles of different shapes immobilized in microfluidic channels or spots of nanoparticle arrays, made using e-beam lithography, within microfluidic channels.15,

16

Unfortunately, all of these studies involve either expensive, unscalable fabrication of the substrates or complex, expensive detection schemes. Herein, we have combined Hole Mask Colloidal Lithography17 with photolithography to pattern nanoparticle arrays onto a substrate.

This procedure allows for facile, scalable,

inexpensive fabrication combined with detection that can be achieved with a standard UV-Vis spectrometer or plate reader common to most research laboratories. In order to demonstrate the utility of this technique, nanoparticle arrays were patterned to generate both 5 and 7-channel microfluidic devices, as well as a 96-well spot plate which can be measured with a stand plate reader in absorption mode.

A proof-of-concept study was carried out with the 7-channel

microfluidic device to yield a single shot binding curve of streptavidin to a biotinylated nanoparticle array. Next, kinetic measurements were obtained with a 5-channel microfluidic device for streptavidin binding to the biotinylated nanoparticle surfaces. Last, binding curves for six different proteins as well as control experiments were simultaneously measured using our fabricated 96-well plate and a plate reader.


Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


EXPERIMENTAL METHODS Fabrication of the patterned nanoparticle arrays on glass substrates.

First,

photolithography is carried out to define regions of the substrate where the nanoparticle arrays will remain versus be removed. The photolithography was carried out by spin-coating Shipley 1818 photoresist (Microchem) with a speed of 2500 rpm for 30 seconds. After spin-coating, the samples were soft-baked at 110 °C for 5 minutes. Next, the patterned masks (printed onto plastic transparency film) were taped to the photoresist-coated substrates and exposed to UV light (Solitec, 365 nm irradiation) for 15 seconds. The samples were subsequently developed in Solution 351 (Dow Electronic Materials) for approximately 1 minute until the pattern became apparent. The samples were then hard-baked at 90 °C overnight to remove excess solvent. On top of these photolithographically patterned substrates, hole mask colloidal lithography17 was then carried out. This procedure first involves spin coating a layer of poly(methylmethacrylate) (PMMA) (350K, Sigma-Aldrich) at 3000 rpm for 30 seconds and baked at 180 °C for 5 minutes. The PMMA-coated substrates were then rendered hydrophilic through oxygen plasma etching (March CS-1701 Reactive Ion Etcher, Nordson Corp.) for 5 seconds, followed by the drop coating for 30 seconds of 0.2% poly(diallyldimethylammonium chloride) in water (SigmaAldrich), which attracted the negatively charged polystyrene spheres to bind to the surface. After washing residual PDDA off the surface with doubly distilled water and drying the substrates with nitrogen, 0.08% carboxyl latex spheres in water (Life Technologies, Inc., 130 nm in diameter), were added to the surface and allowed to sit for at least 2 minutes. The substrates were again dried in nitrogen and a layer of 5 nm of gold is deposited on top of the whole assembly, using e-beam deposition. The polystyrene spheres that stuck to the surface were then removed by tape stripping, leaving ~130 nm gaps in the gold film. Exposure to oxygen plasma



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 23

etch for 2 minutes created holes in the mask that extend down to the underlying substrate of either glass or photoresist. A second deposition of 30 nm of gold filled these holes. Last, the polymer mask and photoresist were lifted off simultaneously through sonication in acetone, leaving behind only the gold which filled the holes in the mask, and thus, a gold nanoparticle array. Because the hole mask procedure was carried out on top of the patterned photoresist, the regions of the substrate containing photoresist were removed completely, leaving the nanoparticle arrays only in regions where the photoresist was absent, see Figure 1. Sealing the patterned substrates to PDMS. For both the microfluidic and spot plate devices, the desired pattern was first made on glass substrates using photolithography. Since the photoresist is resistant to hydrogen fluoride (HF), the substrates were etched with 5% HF, buffered with ammonium fluoride (Fisher Scientific), until the patterns were permanently etched into the glass itself.

The patterned glass was then treated for 40 minutes with

undecyltrichlorosilane (Sigma-Aldrich) under vacuum to allow facile removal of the Polydimethylsiloxane (PDMS) polymer. Next, a thoroughly mixed combination of PDMS and curing agent (10:1 ratio) (Dow Corning Corp.) was degassed under vacuum for 1 hour to remove bubbles.

This degassed PDMS mixture was then added to the patterned, silanized glass

substrates. The PDMS was cured overnight at 70 °C and peeled off the substrate the next day. For the microfluidic devices, 1.5 mm holes were punched on the edge of each channel using a biopsy punch (Miltex, Inc.). The holes in the patterned PDMS for the spot plates were made using a 9-piece hollow leather hole punch (SE 7909LP, Micro-Tools). After holes were punched in the PDMS for both devices, the PDMS is cleaned with acetone, isopropanol, doubly distilled water and thoroughly dried in an oven. Tight sealing of the PDMS to the glass substrates containing patterned nanoparticle arrays is accomplished by oxygen plasma etching the glass


Page 7 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


substrates for 2 minutes followed by another 30 second oxygen plasma etch with both the glass substrates and PDMS together. Within 1 minute of oxygen plasma etching the patterns in both the PDMS and glass substrates were aligned and pressed together, creating a permanent chemical seal. Making flat, wall-less spot plates. Spot plates containing patterned nanoparticle arrays on glass were treated with a 1mM solution of octadecyltrichlorosilane (Sigma-Aldrich) in hexanes for 20 minutes. A mask, made by punching holes in a transparency film, was then clamped to the treated glass substrate to expose only the spots containing nanoparticle arrays. The assembly was then subjected to oxygen plasma etch for 5 seconds, to removes octadecyltrichlorosilane in the spots containing nanoparticles, rendering the areas in between the spots hydrophobic, thus preventing solution leaking from one spot to the next. Characterization of the substrates using Atomic Force Microscopy (AFM). AFM measurements were carried out to confirm the patterning of the uniform nanoparticle arrays. These measurements were carried out in tapping mode using a Veeco Dimension 3100 instrument (Bruker, Corp.) and N-type Si AFM probes with aluminum backing (Mikromasch, Inc.). Typical scan rates were 1 kHz with an amplitude set point of 1V. LSPR measurements. All measurements of the localized surface plasmon resonances were carried out using a USB2000+VIS-NIR (Ocean Optics, Inc.) configured with fiber optic cables and a cuvette holder. The data was collected using the SpectraSuite software provided by Ocean Optics and each spectrum was constructed by averaging 100 scans each collected with 400 millisecond integration time. Frequency values for the individual plasmon resonances were obtained by fitting the peaks of interest to a Gaussian function using the Origin 9.0 software (OriginLab, Inc.).



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 23

For refractive index sensitivity measurements, samples were placed in a liquid transmission cell (Pike Technologies, Inc.) and filled with solutions of varying glycerol (Fisher Scientific) and water (generally between 0 to 80% glycerol). Measurements with the microfluidic devices. When measuring the streptavidin binding curve, different concentrations of streptavidin (Sigma-Aldrich) in phosphate buffered saline were added to the channels using a pipette, allowed to equilibrate for 2 hours, washed with water and dried in nitrogen. The LSPR spectra of the channels were compared, in air, for three different streptavidin concentrations; no streptavidin, an intermediate streptavidin concentration, and a saturating streptavidin concentration. The LSPR shifts observed at the intermediate streptavidin concentration were then normalized by dividing by the highest shift obtained (with saturating streptavidin concentrations) to produce ‘fraction bound’ values plotted in the binding curve shown in Figure 4. For kinetic measurements, the microfluidic devices were connected to a syringe pump (New Era Pump Systems, Inc.) by first inserting 1.5 mm capillary tubes (Fisher Scientific, Inc.) through the PDMS and into each hole at the end of channels. The junction with the capillary tube was sealed by dripping PDMS-curing agent mixture along the sides of the junctions followed by baking at 70 °C for at least 4 hours. Flexible tygon tubing (OD =1/8”, ID =1/16”) (Fisher Scientific, Inc.) was then rolled over the other end of the capillary tube and connected to a syringe fastened to the pump. A 1 mM solution of biotin-PEG (5000)-thiol (Nanocs, Inc.) in water was slowly pumped through the channels overnight. The next day, the biotin solution was removed and the channels were rinsed with doubly distilled water and dried in nitrogen for LSPR measurements.

For streptavidin binding measurements, the protein was slowly pumped (3

µL/min) into a channel containing biotinylated nanoparticle arrays and spectra measured every


Page 9 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3-5 seconds in solution. In order to generate the kinetic data shown in Figure 4, the intensity value at one wavelength (generally close to the peak observed before streptavidin addition) was plotted with time and yielded an exponential decay. Measurements with the 96-well spot plates. For these studies, custom glass pieces were acquired from Ryan’s All Glass in Cincinnati, Ohio, so that each plate would fit directly into the plate reader. Patterned arrays of nanoparticles 130 nm in diameter and 30 nm in height were fabricated as described above. Protein-antibody pairs for three proteins of interest, interleukin10 (IL-10), matrix metalloproteinase-3 (MMP-3), and tumor necrosis factor alpha (TNF-α) were purchased from Life Technologies, Inc. The last three proteins tested, streptavidin, anti-biotin, bovine serum albumin (BSA) and anti-BSA were all purchased from Sigma Aldrich. Each 96well spot plate contained 8 rows of 12 spots each. The first and last row of 12 spots were devoted to control experiments designed to measure non-specific protein binding to either the bare gold nanoparticles or nanoparticles coated with self-assembled monolayer. In the first row, gold nanoparticle arrays in each spot were annealed in water overnight. The last row, along with 4 of the other rows, involved soaking the nanoparticle arrays overnight in a 1:3 mixture of 1 mM mercaptoundecanoic acid (MUA) (Sigma Aldrich) and 1 mM octanethiol (OT) (Sigma Aldrich) in ethanol. Antibodies to attract four of the six different target proteins (BSA, IL-10, MMP-3 and TNF-α) to bind were then coupled using 2-5 mM of N-Hydroxysuccinimide and N-(3Dimethylaminopropyl)-N-ethylcarbodiimide hydrochloride (EDC) (Peirce Biotechnology, Inc.). In rows 2 and 3, 1 mM biotin-PEG (5000)-thiol (Nanocs, Inc.) in ethanol was used as a selfassembled monolayer and soaked overnight. The entire spot plate was then washed with ethanol and water, dried in nitrogen, and measured with the Biotek Cytation 3 Multi-Mode plate reader (Biotek, U.S.). Then, concentrations ranging from 10-18 M to 10-6 M of all six proteins in



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 23

phosphate buffered saline were added to the corresponding rows 2-7 of the spot plate and allowed to soak for 2 hours in order to acquire binding curves. To investigate the non-specific binding of all six proteins of interest to either the bare gold arrays (row 1) and the MUA/OT selfassembled monolayer (row 8), 10-11 M of IL-10, MMP-3 and TNF-α and 10-7 M of BSA, antibiotin and streptavidin were added to the first 6 spots in these two rows and allowed to soak for 2 hours. Upon washing with doubly distilled water and drying with nitrogen, the entire spot plate was measured again in the plate reader.

RESULTS AND DISCUSSION Hole mask colloidal lithography offers an inexpensive, scalable means to pattern dense, uniform nanoparticle arrays in which the size and sensitivity of the nanoparticles can be easily tuned.17 As shown in Figure 1, a pattern for either microfluidic channels or a spot plate is first made on the substrate using photolithography. Next, the hole mask procedure is carried out on top of the patterned photoresist to make the desired nanoparticle array. Briefly, this procedure involves spin coating a layer of poly(methylmethacrylate) (PMMA) followed by a layer of poly(diallyldimethylammonium chloride) (PDDA). Next, polystyrene spheres of a desired size are drop-coated onto the top of the polymer layers and a layer of 5 nm of gold is deposited over the whole assembly. Polystyrene spheres are removed by tape-stripping, and the masks are oxygen plasma etched to generate holes the same size as the polystyrene spheres. Metals are then deposited into these holes followed by in the last step, where both the PMMA/PDDA polymer mask and the photoresist are removed simultaneously using acetone.

Finally,

nanoparticle arrays are left on the substrate in the desired pattern. In this study, glass was used as a substrate and bound to similarly patterned PDMS through oxygen plasma etching, creating a


Page 11 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


chemical, tight seal in the regions that do not contain nanoparticles. The PDMS layer was used to create the channels for the microfluidic devices and to prevent solution migration from spot to spot in the spot plates. Likewise, it is also possible to make flat, wall-less spot plates which do not need to be sealed to PDMS to prevent solution migration. This is accomplished through silanization with a hydrophobic molecule, such as octadecyltrichlorosilane, and selective oxygen plasma etching to remove silane from the nanoparticle arrays. As shown in Figure 1c, the hydrophobic silane molecules greatly increase the contact angle outside of the nanoparticle spots, making solution migration difficult. This technique is not only scalable and inexpensive, but also allows for the versatile fabrication of nanoparticle arrays ranging from 20 to 400 nm in diameter in any desirable diffraction-limited pattern. As shown in Figure 1, nanoparticle arrays were patterned to generate both 5 and 7-channel microfluidic devices, as well as a 96-well spot plate. In principle, one could pattern a significantly larger number of microfluidic channels or spot plates containing 1536 or more spots. In addition, the nanoparticle size and shape could be tuned such that colorimetric detection by eye for each channel or spot can be enabled.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 23

Figure 1. Fabrication of the microfluidic devices, a), spot plates, b), and wall-less spot plates, c), containing nanoparticle arrays. This procedure allows for the fabrication of uniform plasmonic nanoparticle arrays in any diffraction-limited pattern. The patterned nanoparticle arrays depicted above were first characterized using atomic force microscopy (AFM). As shown in Figure 2, the patterning was effective at producing nanoparticles only in defined regions of the substrates. Moreover, the nanoparticles that make up the dense array were uniform in diameter and height on the surface of the substrate, unlike many in situ methods used to generate nanoparticles in microfluidic channels which rely on the reduction of gold salt in solution.18, 19 Due to the high density of the arrays, the color was vivid and can be easily observed by eye.

In addition, since the nanoparticle arrays are substrate-

bound, washing and regeneration of the sensor surface is facile.


Page 13 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. Characterization of the patterned arrays using atomic force microscopy. Large scale images revealed nanoparticles were only present in defined regions of the substrate, a). A zoomed in region containing nanoparticles with cross sectional analysis showed the height of the nanoparticles is uniformly 30 nm, b). In addition, analysis of the particle diameters for this image, bottom inset, shows a distribution centered at 130 nm with a FWHM of 14 nm. Next, the refractive index sensitivity of the nanoparticle arrays was characterized by subjecting the fabricated samples to increasing concentrations of glycerol, see Figure 3. As the index of refraction of the solution is increased, the plasmon resonance peak absorption shifts to the red, as expected.20, 21 Plotting the shift in LSPR peak frequency versus refractive index for three independent samples reveals a linear trend with a slope of 196 ± 31 nm/RIU. This slope or m-value is quite high when compared to typical values of 100 nm/RIU for 100 nm gold colloids in solution, which is attributed to the asymmetric nanodisk-like shape of the nanoparticles where the diameter is roughly four times the height.21,

22

Indeed, these gold nanodisk arrays have

similar m-values as those reported for gold nanorod films on glass.23 In addition, one can 13 ACS Paragon Plus Environment


calculate a figure of merit (FOM) value, in which the refractive index sensitivity is divided by the spectral linewidth. The FOM value for these nanodisks is 1.03, which is significantly greater than that observed for traditional gold colloids (FOM = 0.6) but not as high as those reported for other types of asymmetric gold particles such as gold nanorods (FOM = 1.7), gold nanostars (FOM = 1.4) and triangular silver nanoprisms (FOM = 4.6).21, 24-26 Using Hole Mask Colloidal lithography, it is possible to pattern uniform nanoparticle arrays of larger diameter and more

1.0

0.8

Water 20% glycerol/water 30% glycerol/water 40% glycerol/water 50% glycerol/water 60% glycerol/water

0.6

0.4 700

750

800

850

Wavelength (nm)

900

LSPR Peak Shift (nm)

asymmetric dimensions for even greater sensitivity over that of colloidal particles.

Normalized Absorbance

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 23

12 10 8 6 4 2

Slope = 196 ± 31 nm/RIU

0 1.34

1.35

1.36

1.37

1.38

1.39

1.40

Refractive Index

Figure 3. Sensitivity of the patterned nanoparticle arrays. The nanoparticle arrays were exposed to solutions of increasing glycerol concentration and a sensitivity of 196 nm/RIU was observed.

The first proof-of-concept study was carried out with a 7-channel microfluidic device in which different concentrations of streptavidin protein were added to biotinylated self-assembled monolayers attached to nanoparticle arrays in each channel. Thus, a 7-point binding curve was obtained through the measurement of each channel using a standard UV-Vis spectrometer, see Figure 4a. The Kd value obtained when fitting the binding curve to a standard Langmuir isotherm, 6 x 10-11 M was in agreement with that observed for streptavidin binding to nanoparticle arrays generated using nanosphere lithography.27 Next, a 5-channel microfluidic 14 ACS Paragon Plus Environment

Page 15 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


device was made to study the kinetics of three different concentrations of streptavidin binding to the same biotinylated nanoparticle arrays. These measurements were carried out by pumping the desired concentration of streptavidin through a channel using a syringe pump and monitoring the UV-Vis spectra with time. The resulting kinetic curves, shown in Figure 4b, for three different streptavidin concentrations were fit to an exponential decay and the rate constants, k, were obtained in a rapid, facile manner.

Figure 4. Measurements in both the 5-channel and 7-channel microfluidic devices containing nanoparticle arrays. A 7-channel microfluidic device was used to measure a oneshot binding curve of streptavidin binding to biotinylated nanoparticle arrays in the channels, a). The ‘fraction bound’ values on the y-axis are obtained by soaking each channel in 0 M, intermediate concentrations, and saturating concentrations of streptavidin. The ‘fraction bound’ values are then calculated as the shift obtained at intermediate concentrations divided by largest shift obtained at saturating concentrations. Kinetic measurements of streptavidin binding to biotinylated nanoparticle arrays was achieved with a 5-channel microfluidic device, b). The data acquired every 5 seconds was fit to an exponential decay to obtain rate constants. A second proof-of-concept study was carried out using glass/PDMS plates containing 96 spots of nanoparticle arrays. These plates were applied towards multiplexed protein binding measurements, see Figure 5. The surface-bound nanoparticle arrays in each well allowed for facile functionalization and washing of each spot with varying concentrations and protein solutions.

Using the 96-spot plate, the binding curves of 6 different protein pairs were



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 23

simultaneously measured using a standard plate reader: streptavidin-biotin, anti-biotin-biotin, BSA-anti-BSA, TNFα-anti-TNFα, MMP-3-anti-MMP-3, and IL-10-anti-IL-10.

Figure 5. The fabricated spot plates can be directly measured using a standard plate reader. Our fabricated 96-well plate containing nanoparticle arrays in each of the 96 wells, alongside the commercially available 96-well plates, a). Insertion of our devices into a laboratory plate reader, b).

For streptavidin and anti-biotin binding, thiol-PEG-biotin was added to the nanoparticle arrays. In order to attract TNF-α, MMP-3, IL-10, and BSA to bind to the nanoparticle arrays, antibodies specific to these proteins were fixed to the nanoparticle arrays by first binding a mixed selfassembled monolayer of octanethiol and mercaptoundecanoic acid (OT/MUA).

Then, the

surface amine groups on the antibodies were coupled using carbodiimide chemistry to the COOH groups of the mercaptoundecanoic acid moieties on the surface.

After the self-assembled

monolayer or antibodies were coupled to each spot, the plates were dried in air and the LSPR spectra monitored using a plate reader. The protein of interest was then added in buffered solution in varying concentration to different spots along the same row, dried and measured once again in the plate reader to determine the shift in LSPR wavelength accompanying protein 16 ACS Paragon Plus Environment

Page 17 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


binding. Spectra of the nanoparticle array in air, array with self-assembled monolayers in air, and protein bound to the array in air for two representative proteins, streptavidin and BSA, are shown in Figure S3 in the Supporting Information section. As shown in Figure 6, the two 15minute measurements with the plate reader were able to simultaneously measure 6 different protein binding curves. The first and last rows of the 96-well plate were taken to run control measurements for non-specific binding of the proteins being tested to the nanoparticle arrays. In row 1, all six proteins were added to the bare nanoparticle arrays and in row 8, the six proteins were added to nanoparticle arrays containing a mixed OT/MUA self-assembled monolayer. In all cases, typical LSPR shifts observed for these control measurements were 4-11 nm, significantly lower than the LSPR shifts observed when the appropriate antibody is attached to the nanoparticle arrays. Moreover, nanoparticle arrays containing the OT/MUA mixture and the biotin-PEG-thiol were incubated in 100% mouse serum and the LSPR shifts due to nonspecific binding were only somewhat higher than those observed in buffer, indicating these linker molecules are indeed good at repelling unwanted protein binding, see Figure S5 of the Supporting Information section. All binding data was fit to a Langmuir isotherm where it is assumed that one site on the protein of interest is binding to one ligand or antibody site on the surface. The midpoint of each sigmoidal curve yields the Kd values, which represents the protein concentration required to bind 50% of the sites. The Kd values obtained for BSA, streptavidin and anti-biotin agree well with those obtained from either planar gold surfaces or nanosphere lithography fabricated surfaces.28-31 In addition, the binding affinities measured for TNF-α, MMP-3, and IL-10 are within a factor of ten to those observed with planar gold and electrode surfaces, which have Kd values in the picomolar range.32-35 Lastly, the lowest detectable LSPR shift could be determined by first measuring the inherent noise of the plate reader, see Figure S4



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 23

of the Supporting Information, and adding four times this noise value to the shift observed for the nonspecific binding of the protein to the self-assembled monolayers.

The control

experiments in rows 1 and 8 measure this nonspecific binding and it shown to be slightly different for each protein, as shown in the first and last column of the plot in Figure 6. Limit of detection (LOD) values are then calculated as the lowest concentration required to achieve the calculated shift in LSPR frequency (above that of nonspecific binding), see Table S1 in the Supporting Information section. The LOD values are also shown for each protein in Figure 6. Although the LOD values for both anti-biotin and BSA are roughly ten-fold higher than previously published values using SPR,36, 37 our LOD value for streptavidin protein is roughly 100 fold lower than that obtained with similar LSPR biosensing elements.27 Moreover, the LOD values reported in the literature using SPR for MMP-3 and IL-10 are in the 10-100 pM range, and that of TNF-α in the 0.1 pM range, which are all roughly 10 fold higher than that observed herein.38-40


Page 19 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 6. The simultaneous measurement of 6 different binding curves using the fabricated 96-well plate and a plate reader. The first and last rows in the 96-well plate were used for control experiments in which non-specific binding of the different proteins were measured to either the bare nanoparticle array or a nanoparticle array containing a mixed SAM of OT and MUA. Rows two through six contained different concentrations of interleukin-10 protein (IL-10), matrix metalloproteinase-3 protein (MMP-3), tumor necrosis factor alpha protein (TNF-α), bovine serum albumin (BSA), anti-biotin protein, and streptavidin protein respectively, a). The data was used to generate six different binding curves with which Kd and limit of detection (LOD) values could be extracted, b).

In conclusion, this study reports a facile, inexpensive, scalable means to fabricate nanoparticle arrays in any desired pattern on a substrate for multiplexed and/or microfluidic LSPR biosensing measurements.

Multiplexed, label-free measurements are increasingly

important for a wide range of applications such as point-of-care diagnostics and drug screening.41,

42

This approach allows for the fabrication of microfluidic devices with

significantly more channels than the standard commercially available SPR instrumentation, since 19 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 23

the PDMS is chemically sealed to the substrate and can withstand the higher pressures required for smaller channels. Moreover, highly multiplexed spot plates can be fabricated which directly interface with plate readers common to most research laboratories. Lastly, it should be possible to interface these patterned arrays and channels with other on-chip devices for whole cell analysis or gene profiling. We are currently working to extend this fabrication to substrates other than glass, such as flexible plastics or paper, to further reduce cost and increase portability.

SUPPORTING INFORMATION Additional data on standard deviation, limit of detection values and measurements in 100% mouse serum are available free of charge via the Internet at http://pubs.acs.org.

ACKNOWLEDGEMENTS This work was supported by University of Cincinnati start-up funds.

References 1. Zethelius, B.; Berglund, L.; Sundstrom, J.; Ingelsson, E.; Basu, S.; Larsson, A.; Venge, P.; Arnlov, J. New Engl. J. Med. 2008, 358, (20), 2107-2116. 2. Gowda, G. A. N.; Zhang, S. C.; Gu, H. W.; Asiago, V.; Shanaiah, N.; Raftery, D. Expert Rev. Mol. Diagn. 2008, 8, (5), 617-633. 3. Tokel, O.; Inci, F.; Demirci, U. Chem. Rev. 2014, 114, (11), 5728-5752. 4. He, B.; Morrow, T. J.; Keating, C. D. Curr. Opin. Chem. Biol. 2008, 12, (5), 522-528. 5. Chin, C. D.; Linder, V.; Sia, S. K. Lab Chip 2012, 12, (12), 2118-2134. 6. Sagle, L. B.; Ruvuna, L. K.; Ruemmele, J. A.; Van Duyne, R. P. Nanomedicine-Uk 2011, 6, (8), 14471462. 7. Mayer, K. M.; Hafner, J. H. Chem. Rev. 2011, 111, (6), 3828-3857. 8. Anker, J. N.; Hall, W. P.; Lyandres, O.; Shah, N. C.; Zhao, J.; Van Duyne, R. P. Nat. Mater. 2008, 7, (6), 442-453. 9. Larsson, E. M.; Alegret, J.; Kall, M.; Sutherland, D. S. Nano Lett. 2007, 7, (5), 1256-1263. 10. Campbell, K.; McGrath, T.; Sjolander, S.; Hanson, T.; Tidare, M.; Jansson, O.; Moberg, A.; Mooney, M.; Elliott, C.; Buijs, J. Biosens. Bioelectron. 2011, 26, (6), 3029-3036. 20 ACS Paragon Plus Environment

Page 21 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


11. Endo, T.; Kerman, K.; Nagatani, N.; Hiepa, H. M.; Kim, D. K.; Yonezawa, Y.; Nakano, K.; Tamiya, E. Anal. Chem. 2006, 78, (18), 6465-6475. 12. Barik, A.; Otto, L. M.; Yoo, D.; Jose, J.; Johnson, T. W.; Oh, S. H. Nano Lett. 2014, 14, (4), 20062012. 13. Cetin, A. E.; Coskun, A. F.; Galarreta, B. C.; Huang, M.; Herman, D.; Ozcan, A.; Altug, H. Light-Sci. Appl. 2014, 3. 14. Lee, S. H.; Lindquist, N. C.; Wittenberg, N. J.; Jordan, L. R.; Oh, S. H. Lab Chip 2012, 12, (20), 3882-3890. 15. Ahijado-Guzman, R.; Prasad, J.; Rosman, C.; Henkel, A.; Tome, L.; Schneider, D.; Rivas, G.; Sonnichsen, C. Nano Lett. 2014, 14, (10), 5528-5532. 16. Acimovic, S. S.; Ortega, M. A.; Sanz, V.; Berthelot, J.; Garcia-Cordero, J. L.; Renger, J.; Maerkl, S. J.; Kreuzer, M. P.; Quidant, R. Nano Lett. 2014, 14, (5), 2636-2641. 17. Fredriksson, H.; Alaverdyan, Y.; Dmitriev, A.; Langhammer, C.; Sutherland, D. S.; Zaech, M.; Kasemo, B. Adv. Mater. 2007, 19, (23), 4297-+. 18. Li, Y. T.; Liu, H. S.; Lin, H. P.; Chen, S. H. Electrophoresis 2005, 26, (24), 4743-4750. 19. SadAbadi, H.; Badilescu, S.; Packirisamy, M.; Wuthrich, R. Biosens. Bioelectron. 2013, 44, 77-84. 20. Miller, M. M.; Lazarides, A. A. J. Phys. Chem. B 2005, 109, (46), 21556-21565. 21. Chen, H. J.; Kou, X. S.; Yang, Z.; Ni, W. H.; Wang, J. F. Langmuir 2008, 24, (10), 5233-5237. 22. Jain, P. K.; Ei-Sayed, M. A. J. Phys. Chem. C 2007, 111, (47), 17451-17454. 23. Mayer, K. M.; Lee, S.; Liao, H.; Rostro, B. C.; Fuentes, A.; Scully, P. T.; Nehl, C. L.; Hafner, J. H. ACS Nano 2008, 2, (4), 687-692. 24. Shen, Y.; Zhou, J. H.; Liu, T. R.; Tao, Y. T.; Jiang, R. B.; Liu, M. X.; Xiao, G. H.; Zhu, J. H.; Zhou, Z. K.; Wang, X. H.; Jin, C. J.; Wang, J. F. Nat. Commun. 2013, 4. 25. Xue, B.; Wang, D.; Zuo, J.; Kong, X.; Zhang, Y.; Liu, X.; Tu, L.; Chang, Y.; Li, C.; Wu, F.; Zeng, Q.; Zhao, H.; Zhao, H.; Zhang, H. Nanoscale 2015, 7, 8048-8057. 26. Barbosa, S.; Agrawal, A.; Rodriguez-Lorenzo, L.; Pastoriza-Santos, I.; Alvarez-Puebla, R. A.; Kornowski, A.; Weller, H.; Liz-Marzan, L. M. Langmuir 2010, 26, (18), 14943-14950. 27. Haes, A. J.; Van Duyne, R. P. J. Am. Chem. Soc. 2002, 124, (35), 10596-10604. 28. Riboh, J. C.; Haes, A. J.; McFarland, A. D.; Yonzon, C. R.; Van Duyne, R. P. J. Phys. Chem. B 2003, 107, (8), 1772-1780. 29. Hall, W. P.; Ngatia, S. N.; Van Duyne, R. P. J. Phys. Chem. C 2011, 115, (5), 1410-1414. 30. Wu, P.; Grainger, D. W. Tech. Papers ISA 2004, 449, 243-248. 31. Judy Owen, J. P., Sharon Stranford, Sharon A. Stranford, Janis Kuby, Immunology, 6th edition. W. H. Freeman and Co.: New York, N. Y., 2007. 32. Song, M. Y.; Park, S. K.; Kim, C. S.; Yoo, T. H.; Kim, B.; Kim, M. S.; Kim, Y. S.; Kwag, W. J.; Lee, B. K.; Baek, K. Exp. Mol. Med. 2008, 40, (1), 35-42. 33. Munge, B. S.; Fisher, J.; Millord, L. N.; Krause, C. E.; Dowd, R. S.; Rusling, J. F. Analyst 2010, 135, (6), 1345-1350. 34. Goh, A. X. H.; Bertin-Maghit, S.; Yeo, S. P.; Ho, A. W. S.; Derks, H.; Mortellaro, A.; Wang, C. I. Mabs-Austin 2014, 6, (3), 764-772. 35. Yoon, S. I.; Logsdon, N. J.; Sheikh, F.; Donnelly, R. P.; Walter, M. R. J. Biol. Chem. 2006, 281, (46), 35088-35096. 36. Huang, C. J.; Bonroy, K.; Reekmans, G.; Laureyn, W.; Verhaegen, K.; De Vlaminck, I.; Lagae, L.; Borghs, G. Biomed. Microdevices 2009, 11, (4), 893-901. 37. Arlett, J. L.; Myers, E. B.; Roukes, M. L. Nat. Nanotechnol. 2011, 6, (4), 203-215. 38. Bolduc, O. R.; Pelletier, J. N.; Masson, J. F. Anal. Chem. 2010, 82, (9), 3699-3706. 39. Swanson, S. J.; Jacobs, S. J.; Mytych, D.; Shah, C.; Indelicato, S. R.; Bordens, R. W. Dev. Biol. Stand. 1999, 97, 135-147. 21 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

40. 41. 42.

Page 22 of 23

Martinez-Perdiguero, J.; Retolaza, A.; Bujanda, L.; Merino, S. Talanta 2014, 119, 492-497. Myers, F. B.; Lee, L. P. Lab Chip 2008, 8, (12), 2015-2031. Unser, S.; Bruzas, I.; He, J.; Sagle, L. Sensors (Basel) 2015, 15, (7), 15684-15716.


Page 23 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TOC Figure:


Patterned Plasmonic Nanoparticle Arrays for Microfluidic and

Recommend Documents