Quantum Dot and Gold Nanoparticle Immobilization for Biosensing

Publication Date (Web): September 7, 2012. Copyright .... Mussel-Inspired Direct Immobilization of Nanoparticles and Application for Oil–Water Separ...
1 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/Langmuir

Quantum Dot and Gold Nanoparticle Immobilization for Biosensing Applications using Multidentate Imidazole Surface Ligands Eleonora Petryayeva and Ulrich J. Krull* Chemical Sensors Group, Department of Chemical and Physical Sciences, University of Toronto Mississauga, 3359 Mississauga Road North, Mississauga, Ontario, L5L1C6, Canada S Supporting Information *

ABSTRACT: A facile approach for modification of solid substrates with multidentate imidazole ligands was developed for immobilization of high densities of quantum dots (QDs) that were capped with hydrophilic thiol-based ligands, and for immobilization of noble metal nanoparticles. Imidazole polymer was synthesized using poly(acrylic acid) as a backbone, and grafted on amine functionalized substrate in a two-step approach. The polymer-modified surface was characterized using ellipsometry, water contact angle, and Xray photoelectron spectroscopy. Fluorescence spectroscopy and scanning electron microscopy were used to evaluate nanoparticle immobilization. Homogeneous, high density (ca. 5 × 1011 cm−2) QD films formed via self-assembly were obtained within 4−6 h. Similarly, the imidazole polymer was also shown to be effective for immobilization of gold nanoparticles as a uniform film. By making use of the pH-sensitive affinity of the imidazole rings to zinc on the surface of QDs, it was possible to achieve regeneration of functional ligands suitable for subsequent immobilization of new QDs. Immobilized QDs were used as a platform for bioconjugation with oligonucleotides and peptides. The transduction of nucleic acid hybridization and enzyme activity using QDs as energy donors in interfacial fluorescence resonance energy transfer (FRET) indicated that the immobilization strategy preserved the functional properties of the QDs. The multidentate imidazole ligands used for QD immobilization offer the highest denticity of binding in comparison to the currently available approaches without compromise in their optical properties and ability to interact with biomolecules in solution.

■

INTRODUCTION Various unique optical properties of quantum dots (QDs) offer opportunities for the development of sensing strategies that make use of fluorescence resonance energy transfer (FRET) as a transduction mechanism. Numerous examples of QD-FRET bioassays suitable for use in bulk solution have been reported, and there has been an emphasis on development of such nanoconstructs for applications in intracellular analysis.1 A less explored area involves immobilization to place QDs at an interface or within a matrix for development of biosensors, labon-a-chip, or optoelectronic devices. Previous studies to develop photonic and electronic devices often rely on QD assembly using polymer films (e.g., PMMA) as a host matrix,2 layer-by-layer immobilization using polyelectrolyte thin films,3 and sol−gels.4 Assays designed using electrochemical detection of QDs often rely on direct adsorption or drop casting of nanoparticles onto the surface of the electrode.5,6 These approaches, however, do not generally allow for control of nanoparticle surface homogeneity, nor do they offer the accessibility and stability necessary for subsequent bioconjugation and washing steps. Strategies to immobilize QDs on surfaces for biosensing applications are strongly correlated with more general methods © XXXX American Chemical Society

that have been developed for bioconjugation. A number of QD immobilization strategies include the use of oligonucleotides,7 proteins,8 and antibodies9 as tethers, biotinylated surfaces for streptavidin-coated QDs,10,11 and multidentate surface thiol ligand exchange for thiol-alkyl-acid capped QDs.12,13 Amide coupling has been reported to immobilize carboxyl or amineterminated QDs on surfaces.14 Surface coupling by means of oligonucleotide hybridization can offer a chemical “switch” for release of QDs upon carefully controlled dehybridization of double-stranded DNA.7 The stability of QDs immobilized using proteins or antibodies is limited to the intrinsic structural integrity of the biomolecules, and any conditions that impose denaturation ultimately compromise the immobilization of QDs at such an interface.11 Assembly of thiol-alkyl-acid coatings on QDs using bidentate or tetradentate thiols was demonstrated to offer sufficient mechanical stability of immobilized QDs that further in situ bioconjugation could be achieved.13 Alternatively, immobilization of streptavidin-coated QDs on biotin functionalized substrate allowed subsequent in Received: July 24, 2012 Revised: September 4, 2012

A

dx.doi.org/10.1021/la302985x | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

situ attachment of biomolecules.11,15 One limitation in this instance was the inability to regenerate the surface to replace QDs. The imidazole chemistry presented herein to immobilize QDs that were capped with thiol-based small molecule ligands was inspired by the well-established high-affinity interactions of biomolecules that are modified with His-tags. Since the first description in the literature in 2005 by Medintz and coworkers,16−18 this approach has become one of the most versatile bioconjugation strategies for use with QDs. His-tag typically contains six imidazole rings connected via amino acid peptide linkages. It has been demonstrated that increasing the number of histidine amino acids within a tag can increase the affinity to QDs.17 Binding of His-tag modified biomolecules proceeds virtually quantitatively, eliminating the typical need for additional reagents and subsequent purification steps that arise when using other conjugation methods.16,17 Rapid assembly at room temperature is also feasible using QDs that are coated with bidentate thiol ligands, dihydrolipoic acid (DHLA), and derivatives thereof.19 Polymer coatings developed by Bawendi and co-workers that are based on 50% imidazole appended ligands have allowed for ligand exchange with organic QDs in under 10 min at room temperature.20 This is very favorable when compared to typical mercaptopropionic acid (MPA) or DHLA ligand exchange that is done over 2−6 h at elevated temperatures. In addition, one important advantage of nitrogen-based ligands is that they do not introduce photoluminescence (PL) quenching of QDs, or photooxidation upon UV irradiation unlike thiol-based ligands.20 This consideration is significant when developing a sensing interface since the sensitivity and detection limits will be a function of QD brightness if the assay is operating in FRET mode. Recently, Zhang et al. described an imidazole-based polymer coating to obtain water-soluble QDs with small hydrodynamic radius and improved colloidal stability over a broad pH range.21 The versatility of such appended peptides and proteins was also extended to small gold nanoparticles (Au NPs), where coordination through imidazole rings was confirmed.22 Direct synthesis of Au NPs on imidazole functionalized latex beads via an electroless plating process generated a well-dispersed population of nanoparticles on the surface indicating favorable interactions for surface immobilization.23 The advantages of using imidazole affinity as well as robustness of multidentate ligands for nanoparticle immobilization prompted us to further develop a facile protocol for stable and cost-effective fabrication of surface chemistry. In the present study, we describe and characterize a synthetic method to build multidentate imidazole surface ligands for immobilization of QDs and Au NPs. An organosilane film bonded to the surface of glass substrates (e.g., fused silica fibers, coverslips, glass slides, glass beads) served as a platform for coupling with polyimidazole ligands. Core/shell QDs capped with a range of thiol-based small molecule ligands were immobilized at high density via surface assembly at room temperature. As a result of multiple coordination points with a single QD, this immobilization chemistry can serve as a suitable approach for the development of sensing platforms that are capable of withstanding multiple washings and denaturing conditions. The observed changes in optical properties of immobilized QDs (PL spectra and PL lifetime) were attributed to the passivation effects of imidazole ligands and the change in local environment. Exposure of immobilized nanoparticles to acidic pH promoted disruption of imidazole−zinc interactions

and provided for regeneration of surface ligands. Subsequent immobilization of the same type of QDs, different QDs, or a mixture of two colors of QDs was also demonstrated. The surface of immobilized QDs remained accessible for bioconjugation of oligonucleotides and peptides, and functional for FRET-based biosensing applications. As an extension of application, the immobilization of gold nanoparticles (Au-NPs) was demonstrated at densities suitable for sensing applications by localized surface plasmon resonance (LSPR). Immobilization of Au NPs capable of withstanding harsh washing conditions and elevated temperatures necessary for removal of adsorbed biomolecules was examined.

■

EXPERIMENTAL SECTION

For detailed experimental procedures, materials and instrumentation used, data collection and analysis, the reader is referred to the Supporting Information. Synthesis of Imidazole Polymer. The polymer with appended imidazole ligands (PAAI) was synthesized using poly(acrylic acid) (PAA) as a backbone. Typically, 0.72 g of PAA (10 mmol of acrylic acid units) and 1.16 g (10 mmol) of N-hydroxysuccinimide (NHS) were dissolved in 150 mL of anhydrous dimethylformamide (DMF) and cooled to 0 °C. Next, 1.54 mL (10 mmol) of N,N′diisopropylcarbodiimide (DIC) dissolved in 20 mL of anhydrous DMF was added dropwise over 1 h. The reaction mixture was gradually brought to room temperature and stirred for 6 h under argon. Without purification, 895 μL of 1-(3-aminopropyl)imidazole (7.5 mmol) was added and the reaction mixture was stirred for an additional 24 h. Surface Modifications of Glass Substrates with Multidentate Ligands. Fused silica fibers, glass beads, and silicon wafers were cleaned using two-step RCA (Radio Corporation of America) protocol to obtain silanol groups.24 The RCA protocol removes organic and ionic contaminants through successive treatment in strong base and acid in the presence of a strong oxidant, hydrogen peroxide. Dried, cleaned substrates were submerged in approximately 200 mL of isopropanol containing 2 mL (1% v/v) of 3-aminopropyltrimethoxysilane (APTMS) and 200 μL (0.1% v/v) of glacial acetic acid. Substrates were incubated for 6 h with gentle shaking. Subsequently, the solution was decanted and the substrates were sonicated three times for 20 s in fresh isopropanol, rinsed once with methanol, dried with nitrogen air, and annealed for 2 h in the oven at 115 °C for crosslinking. Amine modified substrates were further treated with PAAI activated with 50 μL (0.28 mmol) N-(3-dimethylaminopropyl)-N′ethylcarbodiimide (EDC) prior to use. The reaction vessel was incubated for 24−36 h with gentle shaking. Modified substrates were then washed once with DMF, three times with methanol, then dichloromethane and ethyl ether. PAAI functionalized substrates were stored in a desiccator until further use. Immobilization of Quantum Dots, DNA Hybridization, and Proteolytic Experiments. Fibers functionalized with multidentate imidazole ligands were rinsed with borate buffer (pH 9.25, 50 mM) and immersed for 4 h in aqueous solutions of QDs (0.1−0.2 μM) capped with mercaptopropionic acid (MPA), glutathione (GSH), or dihydrolipoic acid (DHLA) ligands. For clarity throughout the text, the QD notation is used in the form “xxx Ligand”, where xxx corresponds to the PL wavelength maxima of organic QDs in toluene, followed by the ligand used to generate water-soluble nanoparticles. Protocols to obtain water-soluble QDs are described in the Supporting Information. The immobilization of two colors of QDs was done from aqueous mixtures of green and red QDs at ratios as indicated in the text. Substrates with immobilized QDs were washed three times with borate buffer, once with 0.1% SDS in borate buffer, and finally rinsed with borate buffer. The bioconjugation of disulfide (dithiophosphoramidite, DTPA) modified probe (Table 1) was done by immersing fibers for 6 h with gentle shaking in oligonucleotide solution (1 μM) that contained borate buffer (pH 8.3, 100 mM) with 10 mM tris(2carboxyethyl)phosphine (TCEP). The oligonucleotide sequence is B

dx.doi.org/10.1021/la302985x | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

increase by 0.7 ± 0.1 nm after APTMS deposition and 3.1 ± 0.5 nm after coupling with PAAI. The thickness of the expected organosilane layer is consistent with formation of a monolayer, typically reported in the range of 0.5−1.1 nm.27,28 Coupling of polyimidazole ligands was associated with an increase of a few nanometers suggesting that ligands were collapsed along the surface for these samples in the dry state. Considering that PAAI is anchored to the surface at about 5−7 points, assuming >99% coupling efficiency, it is anticipated that imidazole ligands can extend farther away from the surface in the presence of nanoparticles. This configuration of surface ligands could increase the overall strength of interaction by wrapping around the curved surface of nanoparticle. For this particular reason, a low-molecular-weight PAA was chosen as a backbone. An average of 21 carboxylic acid units in the polymer chain derivatized with imidazole ligands is expected to have greater flexibility to mimic His-tag behavior. Furthermore, it is anticipated that the PAAI ligands can provide a high denticity and, therefore, a high binding affinity. Comparison of stability of immobilized QDs based on use of mono- and tetra-dentate thiol surface ligands have been shown to offer greater stability with increase in denticity of binding.29 Characterization of PAAI synthesis was carried out using FTIR and 1H NHR spectroscopy (for details, see Supporting Information). Formation of intermediate PAA-NHS was confirmed by the disappearance of the sharp carboxylic acid peak at 1713 cm−1 and the appearance of three characteristic CO stretch vibrations of NHS-ester at 1812, 1782, and 1736 cm−1. Subsequent coupling with 1-(3-aminopropyl)imidazole was associated with the appearance of a characteristic amide stretch at 1655 cm−1. Modifications in surface chemistry were also indirectly confirmed with an increase in static water contact angle of 10 ± 4°, 42 ± 5°, and 56 ± 6° from cleaned silicon wafers, to APTMS and PAAI ligands, respectively. An increase in hydrophobicity of the substrates is consistent with introduction of more hydrophobic functional groups on the surface. The RCA-treated surfaces were fully wetted, indicating availability of silanol groups. Treatment with APTMS generated a hydrophobic surface as a result of the alkane core of this molecule and cross-linking of acidic surface silanols. Immobilization of imidazole-appended polymer resulted in the most hydrophobic substrates. Although the imidazole ring has some hydrophilic character, the overall contribution of underlying

Table 1. Oligonucleotide and Peptide Sequences abbreviation DTPAa-5′-ATT TTG TCT GAA ACC CTG T-3′ SMN1 Probe 3′-TAA AAC AGA CTT TGG GAC A-5′-Cy3 5′-Cy3-TRG Cy3-3′-TAA AAC AGA CTT TGG GAC A-5′ 3′-Cy3-TRG noncomplementary oligonucleotide 5′-ATT TTG TCT GAA ACC CTG T-3′-Cy3 NC-TRG peptide sequence Ac-His-His-His-His-His-His-Gly-Leu-Aib-Ala-Ala-Gly-Gly-Trp-Gly-Cys-NH2

Probe Target

a

DTPA = dithiophosphoramidite.

diagnostic of the H. sapiens survival motor neuron protein coding gene (SMN 1), which is associated with the disorder known as spinal muscular atrophy.25 Subsequently, fibers were rinsed with borate buffer and immersed for 50 min in denatured bovine serum albumin (dBSA) 0.5 mg/mL in borate buffer (pH 8.3). Hybridization was done for 1 h using target (3′-Cy3-TRG or 5′-Cy3-TRG) solution prepared in Tris-borate buffer (pH 7.4, 50 mM NaCl). A noncomplementary oligonucleotide target sequence (NC-TRG) was used for control experiments. The peptide sequence is used to characterize chymotrypsin activity (Table 1) and was coupled with maleimide functionalized Alexa Fluor 647, as described previously.26 QD-modified fibers were incubated in 100 nM peptide solution in borate buffer (pH 8.3) for 5 min and rinsed three times with buffer. Surfaces coated with QD-peptide conjugates were submerged in enzyme solution (1 mg/mL) for the indicated period of time and rinsed in buffer prior to measurement. Immobilization of Au NPs. Surfaces coated with multidenate imidazole ligands were immersed overnight in as-synthesized citratecapped Au NPs solution. Samples were then rinsed with water and additional washing steps were incorporated. Surface Regeneration. QD-modified substrates were sonicated (20 min × 3) in Gly-HCl buffer and subsequently were immersed for 4 h with shaking in borate buffer (pH 9.25) containing 0.5 M imidazole. Substrates were rinsed in borate buffer and sonicated for 15 min in buffered fresh imidazole solution and then rinsed thoroughly in borate buffer (pH 9.25), dried, and used for subsequent experiments.

■

RESULTS AND DISCUSSION Characterization of Surface Ligands. The two-step surface modification shown in Figure 1 was characterized using ellipsometry, XPS, and water contact angle measurements. Ellipsometry results obtained from analysis of silicon wafers indicated an average homogeneous film thickness

Figure 1. Schematic presentation of two-step synthesis of multidentate imidazole surface ligands (a) and design of interfacial QD-FRET oligonucleotide hybridization (b) and protease assays (c). C

dx.doi.org/10.1021/la302985x | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

of QDs on the surface. Substrates functionalized with only PAA were used as controls and displayed no detectable QD PL, while substrates functionalized with PAAI readily immobilized QDs at room temperature. Figure 3a shows PL spectra of fibers

alkane linkers, and the acrylic acid backbone increased the hydrophobicity. The XPS results were used to further confirm chemical modifications of substrates and are summarized in Table 1S (Supporting Information). Following APTMS silanization, a significant increase in both nitrogen and carbon content was observed relative to substrates that had been cleaned using the RCA protocol. The intensity of the silicon signal decreased by more than 10% following APTMS treatment, consistent with an overcoating of organic film. Further decrease of the silicon signal was observed with addition of the PAAI layer. Consistent with the chemical structure of PAAI, a further increase was observed in relative atomic percent of nitrogen, oxygen, and carbon. In the final synthetic step, the relative percent composition of nitrogen increased by 78% in comparison to the APTMS coated surfaces, and this is consistent with the coupling of the imidazole ligand. In order to confirm the identity of functional groups, high-resolution XPS spectra were obtained for C 1s and N 1s. As shown in Figure 2a, APTMS

Figure 3. (a) PL emission spectra of fibers modified with PAAI (red) and PAA (blue). The affinity of the QDs for the imidazole surface is clearly visible over native PAA ligands indicating immobilization (b) SEM image of 525 MPA immobilized QDs on a silicon wafer functionalized with multidentate imidazole ligands, and (c) streptavidin coated QDs on a biotinylated surface.

modified with PAA or PAAI after exposure to red MPA coated QDs. Typically, rapid increase in PL intensity was observed within 30 min, with the intensity becoming maximized after 4− 6 h. Occasionally, some nonspecific adsorption of thiol-alkyl acid capped QDs to amine modified substrates was noted. However, the contrast ratio as estimated from PL maxima between PAAI and APTMS modified substrates was 263:1. The morphology of surfaces after QD immobilization was investigated using SEM. The SEM images indicated uniform films of immobilized QDs at high density on the surfaces (Figure 3b). This observation of relatively uniform films is consistent with previous observations collected using atomic force microscopy.24 No corrections were therefore made to introduce roughness parameters and account for threedimensional distributions of QDs. The surface density for QDs with an average diameter 5.5−7.0 nm was estimated as (4.8−5.2) × 1011 per cm2. Proteins that are roughly on the same size scale typically immobilize at comparable densities 1011 to 1012 cm−2.34,35 It is anticipated that smaller nanoparticles will be immobilized at higher number density. One of

Figure 2. High-resolution XPS spectra. (a) N 1s spectra of APTMS (i) and PAAI modified substrates (ii). (b) C 1s spectra after PAAI coupling.

deposition exhibited a single N 1s peak (400.0 eV) consistent with primary amines. The binding energy of N 1s of an amide bond (400.5 eV, 26%) appeared after coupling with PAAI.30,31 The signal at 399.7 eV (74%) was assigned to imide and tertiary amine functional groups found in imidazole ligand.31,32 The high-resolution C 1s spectra for PAAI modified substrates (Figure 2b) confirmed the presence of the amide bond (288.8 eV), as well as CN (CO) (287.8 eV) and CN (286.3 eV).33 Efficacy of QD Immobilization and Photoluminescence Properties of Immobilized QDs. QD affinity to surfaces at each step of synthesis was monitored in parallel using QDs capped with different ligands (MPA, GSH, or DHLA). PL measurements from modified fibers indicated that only the presence of imidazole ligands correlated with retention D

dx.doi.org/10.1021/la302985x | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

obtain water-soluble QDs. MPA capped QDs have shown an increase in PL lifetime that was attributed to an additional passivation effect caused by the imidazole surface ligands. Analogous trends have been reported previously in LBL assembly.45 Only a modest increase in average lifetime was noted with DHLA capped QDs. The more dense and stable film formed by use of bidentate DHLA likely reduced the extent of interaction between the imidazole ligands and QD shell. A different trend was observed with GSH coated QDs, where the lifetime of immobilized QDs either remained unchanged or decreased slightly. Transfer of organic QDs into aqueous media using GSH ligands produced the least decrease of brightness in comparison to MPA or DHLA (Table 2S, Supporting Information), consistent with other reports in the literature.46 The retention of higher quantum yield is associated with minimal damage to the ZnS shell, while harsh conditions used for MPA and DHLA promote etching of Zn2+ ions generating trap states. Preservation of the integrity of the ZnS shell during ligand exchange with GSH ultimately results in minimal passivation effects by imidazole ligands upon immobilization. Surface Regeneration. The affinity of imidazole groups to the Zn-rich QD shell is strongly pH dependent. Polymer coatings designed with appended imidazole ligands have previously shown colloidal stability over the pH range 6−11. Below pH 5.5, QD aggregation in solution was observed due to the dissociation of ligands.20 A controlled pH-triggered release of immobilized QDs to regenerate multidentate surface ligands is therefore possible. Multiple cycles of removal and regeneration of QD films on surfaces, including reimmobilization of the same color of QDs or alteration between different colors is shown in Figure 4. Furthermore, simultaneous immobilization of two colors of QDs was achieved by incubating substrates in a mixture of water-soluble QDs at a desired ratio. It is interesting to note that two-color immobilization is strongly influenced by the relative sizes of QDs. The mixture of alloy CdSxSe1‑x/ZnS QDs (525 MPA and 623 MPA) had similar size of 5−6 nm, irrespective of emission wavelength. These QDs exhibited direct correlation between bulk solution and interfacial PL (Figure 4a). A mixture of green and red QDs prepared in 1:1 molar ratio produced comparable PL intensities in bulk solution as a result of similar molar absorption coefficients at the excitation wavelength (ε(green) = 9 × 105 M−1 cm−1 and ε(red) = 1.2 × 106 M−1 cm−1 at 405 nm). In contrast to alloy QDs, the immobilization of a mixture of CdSe/ZnS green (518 MPA) and red emitting CdSe/ZnS QDs (605 MPA) with relative sizes 2 and 6 nm, respectively, resulted in greater immobilization efficiency of smaller QDs. A molar ratio of 6:1 green-to-red MPA coated QDs produced a normalized QD PL ratio in bulk solution of 1:1.4 versus 1:0.9 at an interface. The normalization of PL spectra to green QD PL maxima was used to monitor relative changes in PL intensities. A higher concentration of green QDs relative to red QDs was necessary to obtain comparable PL intensities due to significantly stronger absorption of light by red QDs at the excitation wavelength (ε(green) = 9 × 105 M−1 cm−1 and ε(red) = 2.6 × 106 M−1 cm−1 at 405 nm). The decrease in red QD PL maxima from 1.4 to 0.9 relative to green QD PL (normalized to one) was correlated to a difference by ca. 35% between solution and surface concentrations of red QDs. Considering the high affinity of the imidazole ligands, it is proposed that over time smaller QDs can fill gaps on the surface between larger QDs, increasing their relative surface density.

the most common approaches for QD immobilization in the literature is through streptavidin−biotin interactions. This method of coupling was used in this work to compare the density that could be achieved. Figure 3c shows streptavidin coated QDs (em. 525 nm) immobilized on a biotinylated surface, and the density was estimated to be (2.2−2.4) × 1011 QDs per cm2. The density in comparison to Figure 3b is decreased by roughly a factor of 2, likely due to the bulky protein coating where a single streptavidin measures ca. 5.5 × 4.5 × 5.0 nm3.36 High surface density of QDs is particularly important in the development of multiplexed solid-phase QDFRET assays, where dilution effects may lead to decreased sensitivity.37 In the development of solid-phase assays, it is desirable that QDs retain their optical properties upon immobilization. It is anticipated that unpredictable changes will result from any variability in synthesis including the quality of the protective ZnS shell. However, for a consistent batch of QDs, a question arises whether there are trends of spectral changes when using a variety of ligands to coat the QDs. Immobilization introduced bathochromic shifts of 0.8−3.5 nm for all of the QDs used in this work regardless of the type of coating (Table 2S, see Supporting Information). It has been reported previously that immobilization of QDs at an interface or within a solid matrix results in changes of PL properties.24,38−40 Similarly, transfer of QDs with organic coatings into aqueous media by ligand exchange typically results in PL shifts, and significant decreases in quantum yield and lifetime. Spectral shifts were previously reported for a monolayer of QDs in high density as a result of energy transfer from smaller to larger nanoparticles.40 Such nanoassemblies have been based on QD solids,41 layer-by-layer assembly,40,42 or self-assembly on a modified surface.24 Redshifts of 14 nm were observed in drop-cast films of organic CdSe/ZnS QDs.43 The evidence of homo-FRET in QD assemblies has been previously reported by comparison of lifetime within the blue and red regions of PL spectra.44 In the present work, PL decay measured for immobilized 525 MPA QDs showed a noticeable increase from the blue side (τ515 = 21.7 ns) to the red side (τ540 = 25.1 ns), whereas average lifetime measured at the PL maxima was an intermediate of both (τ527 = 23.5 ns). The QD−QD energy transfer contributes to the wavelength-dependent changes in lifetimes and to the observed changes in optical properties, and confirms the close proximity of QDs. QD immobilization by use of imidazole introduced only a very small change in full width at half-maxima (fwhm), with changes for MPA, DHLA, and GSH coated QDs being 2 nm or less. These data in aggregate suggest relative homogeneity of QDs at high density on the surfaces. As expected, the PL lifetime was influenced by immobilization using multidentate imidazole ligands (Table 2). Changes in PL lifetime between bulk solution and surface immobilization were found to be strongly dependent on the ligands used to Table 2. PL Lifetimes of Immobilized QDs QDs 525 525 525 623 623 623

MPA DHLA GSH MPA DHLA GSH

τav (soln), ns 18.5 15.9 21.1 16.6 17.8 20.2

± ± ± ± ± ±

0.5 0.3 0.3 0.4 0.3 0.5

τav (surf), ns 23.5 17.5 19.8 22.0 20.7 19.7

± ± ± ± ± ±

0.8 0.2 0.9 0.2 0.8 0.8 E

dx.doi.org/10.1021/la302985x | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Figure 4. Removal and regeneration of QD film on multidentate imidazole surface. (a) Removal (ii and iv) of QDs and reimmobilization of red QDs (iii) or mixture of two colors of QDs (v) on a glass bead; (vi) shows corresponding spectra to (v). Insets in (ii) and (iv) show white light glass bead as seen in the field of view. (b) PL intensity collected from a fused silica fiber upon washing off and then reimmobilization of 525 MPA QDs.

Figure 5. Nanoparticle immobilization using multidentate surface ligands for sensing applications. (a) 525 MPA QD-FRET hybridization assay with 3′-Cy3-TRG (blue), 5′-Cy3-TRG (red), NC-TRG (dashed black), and QD only PL (green). (b) Protease assay using 624 DHLA QDs and Alexa 647 labeled peptide; (i) to (viii) corresponds to the exposure to chymotrypsin for 0, 3, 8, 12, 30, 60, 90, and 120 min. The spectrum of only the QDs (ix) is shown for reference. The inset illustrates the corresponding FRET ratios. (c) Absorption spectra of immobilized Au NPs after sonication in water for 5 min. The inset shows a SEM image of the silicon wafer with immobilized Au NPs; scale bar is 50 nm.

The ability to reimmobilize QDs on the various substrates indicates that the imidazole ligands remained largely unchanged and stable at the conditions that were used. However, some decrease in efficacy was noted over multiple cycles (Figure 4b). Typically, loss of QD PL was 25% or less after four cycles of regeneration. This could be associated with some quenched QDs being retained, or by dislodged zinc ions from QD shell

material being complexed with imidazole. In order to evaluate any presence of Zn ions from QDs, the substrates were washed with ethylenediaminetetraacetic acid (EDTA) solution, which introduces a Zn-EDTA dissociation constant on the order of 10−17 M. Although KD for Zn-His tag is orders of magnitude larger (nanomolar range), a similar trend of QD PL decrease F

dx.doi.org/10.1021/la302985x | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

upon exposure to chymotrypsin in comparison to an untreated control sample. The proteolytic digestion of peptide suggests utility of this platform for screening and investigation of enzyme activity in a multiplexed format. Immobilization of Au NPs. Gold nanoparticles continue to be significant in the development of methods for real-time colorimetric and optical assays that monitor molecular processes. A solution of Au NPs was characterized with UV− vis spectroscopy and exhibited a narrow plasmon band centered at 522 nm. The nominal size of the synthesized NPs was estimated at 13 nm, and this was in agreement with the size estimated from SEM images.50 Glass substrates modified with multidentate imidazole ligands appeared pink to red after treatment with a solution containing Au NPs. Figure 5c presents a spectrum of immobilized NPs with a characteristic plasmon band centered at 526 nm. The absence of a band at lower energy (600−700 nm) excludes formation of aggregates on the surface. SEM imaging of immobilized Au NPs suggests a uniform film with an average surface density 1.1 × 1011 cm−2 as shown in Figure 5c. Harsh washing protocols including up to five minutes of sonication in water, as well as submersion of slides in hot water (>95 °C) did not induce changes in optical properties of immobilized Au NPs. Retention of the characteristic plasmon band suggests the suitability of the polyimidazole chemistry for Au-NP immobilization in sensing applications, as similar washing steps are often used to remove adsorbed biomolecules (e.g., oligonucleotides). In contrast to regeneration of QD functionalized substrates, a similar washing protocol did not result in the removal of Au NPs. Immobilized Au NPs remained stable upon subsequent coating with MPA and DHLA ligands, which was done by treatment with 1 mM ligand solution for 2−6 h.

was observed with subsequent reimmobilization. This suggests that residual zinc ion was not responsible for the trend. Bioconjugation and FRET-Based Applications. FRETbased applications have been used to demonstrate the utility of the imidazole polymer for immobilization of QDs. An example of a simple oligonucleotide hybridization assay is described. Bioconjugation of bidentate thiol-modified oligonucleotide probe proceeded efficiently with MPA and GSH coated QDs via ligand exchange.13 The same strategy was ineffective with DHLA coated QDs. It is interesting to note that, while MPA coated QDs exhibited strong nonspecific adsorption and required passivation with denatured bovine serum albumin (dBSA), as described previously,29 GSH coated QDs exhibited minimal to nondetectable nonspecific interactions with oligonucleotides. The partially zwitterionic character of the small peptide used as a coating ligand could significantly minimize hydrogen bonding that has been reported for MPA ligands.47 The exposure of immobilized green QD-probe conjugates to complementary Cy3 labeled oligonucleotide target resulted in localization of dye in close proximity for FRET-based detection (Figure 5a). The FRET ratios for 5′-Cy3-TRG and 3′-Cy3TRG were found to be 2.2 and 6.3, respectively. The estimated FRET ratio for NC TRG was 0.01, suggesting effective passivation with dBSA to ameliorate nonspecific adsorption of MPA-coated QDs. Lifetimes calculated from PL decay curves for QD-probe conjugates before and after passivation with dBSA were 18 and 14 ns, respectively. Further decrease in QD lifetime upon hybridization was associated with additional relaxation pathways via FRET, such that hybridization with 5′Cy3-TRG and 3′-Cy3-TRG resulted in average QD lifetime 6 and 1 ns, respectively. Since QD donors and dye acceptors do not form discrete FRET pairs at an interface, reporting conventional FRET efficiency is not possible.37 Nevertheless, the estimate of apparent FRET efficiency can be useful to correlate energy transfer as a function of the location of the dye acceptor as observed in steady-state experiments. The apparent FRET efficiencies estimated from changes in PL lifetime were close to 57% and 93% for distal and proximal targets, respectively. Despite the possibility that surface ligands wrapped around the surface of nanoparticles offering high binding denticity, QDs were still capable of interacting with biomolecules. Furthermore, the increased lifetimes of immobilized QDs using the imidazole polymer are consistent with the passivating nature of imidazole ligands, and are highly beneficial for interfacial FRET applications. Longer lifetimes and higher quantum yields of the donors correlate to larger Förster distances, improving sensitivity and limits of detection of bioassays. In addition to the oligonucleotide hybridization assay, assembly of functional QD-peptide conjugates was also demonstrated. QD-substrate conjugates in bulk solution provide enhanced enzymatic activity purportedly due to a hopping mechanism, and it was of interest to evaluate the ability of enzymes to function in a more sterically hindered environment with immobilized QDs.48 Protease sensing using a solution-phase QD-FRET approach is well-characterized in the literature.26,49 Here, we demonstrate that QD-peptide conjugates at an interface remain accessible to enzyme, as observed in a decrease of FRET sensitized emission of the acceptor dye (Alexa 647). Figure 5b shows the time-dependent response of red DHLA coated QDs bioconjugated with dye labeled peptide via His-tag. The initial FRET ratio was 0.46 and rapidly decayed

■

CONCLUSIONS In this work, we described a facile synthetic approach to derivatize various substrates with multidentate imidazole surface ligands. QDs capped with thiol-based small molecule ligands readily immobilized at an interface at high density (4.8− 5.2) × 1011 QDs cm−2. The changes in optical properties of immobilized QDs were consistent with films of high density at a level that would facilitate intraensemble energy transfer. Bathochromic shifts of up to 3.5 nm were observed upon immobilization, while fwhm was increased by at most 2 nm, suggesting formation of homogeneous films. The surface passivation effect of imidazole ligands was noted by an increase in average QD lifetime upon immobilization. The most pronounced effect was observed with MPA capped QDs, while inherently bright GSH coated QDs did not exhibit significant change in PL. Regeneration of surface ligands triggered by pH change to induce protonation of the imidazole ring allowed for subsequent immobilization of fresh QDs. The immobilized QDs were able to form functional bioconjugates with oligonucleotides and peptides for FRET-based biodetection. The versatility of this immobilization strategy was also demonstrated for Au NPs, which formed a uniform film of isolated nanoparticles with retention of the characteristic plasmon peak. Immobilization of QDs offers advantages in enhancing FRET efficiency, eliminates concerns about precipitation from solution, facilitates sample delivery and subsequent washing to remove nonspecifically adsorbed materials, and traps nanoparticles from entry into the environment. Overall, this work provides an effective immobilization strategy for development of multiplexed QDG

dx.doi.org/10.1021/la302985x | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

(15) Tavares, A. J.; Petryayeva, E.; Algar, W. R.; Chen, L.; Krull, U. J. Proc. SPIE 2010, 7750, 775003. (16) Sapsford, K. E.; Pons, T.; Medintz, I. L.; Higashiya, S.; Brunel, F. M.; Dawson, P. E.; Mattoussi, H. J. Phys. Chem. C 2007, 111, 11528− 11538. (17) Medintz, I. L.; Berti, L.; Pons, T.; Grimes, A. F.; English, D. S.; Alessandrini, A.; Facci, P.; Mattoussi, H. Nano Lett. 2007, 7, 1741− 1748. (18) Goldman, E. R.; Medintz, I. L.; Hayhurst, A.; Anderson, G. P.; Mauro, J. M.; Iverson, B. L.; Georgiou, G.; Mattoussi, H. Anal. Chim. Acta 2005, 534, 63−67. (19) Dennis, A. M.; Sotto, D. C.; Mei, B. C.; Medintz, I. L.; Mattoussi, H.; Bao, G. Bioconjugate Chem. 2010, 21, 1160−1170. (20) Liu, W. H.; Greytak, A. B.; Lee, J.; Wong, C. R.; Park, J.; Marshall, L. F.; Jiang, W.; Curtin, P. N.; Ting, A. Y.; Nocera, D. G.; Fukumura, D.; Jain, R. K.; Bawendi, M. G. J. Am. Chem. Soc. 2010, 132, 472−483. (21) Zhang, P. F.; Liu, S. H.; Gao, D. Y.; Hu, D. H.; Gong, P.; Sheng, Z. H.; Deng, J. H.; Ma, Y. E.; Cai, L. T. J. Am. Chem. Soc. 2012, 134, 8388−8391. (22) Kogot, J. M.; England, H. J.; Strouse, G. F.; Logan, T. M. J. Am. Chem. Soc. 2008, 130, 16156−+. (23) Kim, H.; Daniels, E. S.; Dimonie, V. L.; Klein, A. J. Polym. Sci., Part A: Polym. Chem. 2008, 46, 912−925. (24) Algar, W. R.; Krull, U. J. Langmuir 2008, 24, 5514−5520. (25) Watterson, J. H.; Raha, S.; Kotoris, C. C.; Wust, C. C.; Gharabaghi, F.; Jantzi, S. C.; Haynes, N. K.; Gendron, N. H.; Krull, U. J.; Mackenzie, A. E.; Piunno, P. A. E. Nucleic Acids Res. 2004, 32. (26) Medintz, I. L.; Clapp, A. R.; Brunel, F. M.; Tiefenbrunn, T.; Uyeda, H. T.; Chang, E. L.; Deschamps, J. R.; Dawson, P. E.; Mattoussi, H. Nat. Mater. 2006, 5, 581−589. (27) Kurth, D. G.; Bein, T. Langmuir 1995, 11, 3061−3067. (28) Petri, D. F. S.; Wenz, G.; Schunk, P.; Schimmel, T. Langmuir 1999, 15, 4520−4523. (29) Algar, W. R.; Krull, U. J. Langmuir 2009, 25, 633−638. (30) Hendrick., Dn; Hollande., Jm; Jolly, W. L. Inorg. Chem. 1969, 8, 2642−&. (31) Jansen, R. J. J.; Vanbekkum, H. Carbon 1995, 33, 1021−1027. (32) Andresa, J. S.; Reis, R. M.; Gonzalez, E. P.; Santos, L. S.; Eberlin, M. N.; Nascente, P. A. D.; Tanimoto, S. T.; Machado, S. A. S.; Rodrigues-Filho, U. P. J. Colloid Interface Sci. 2005, 286, 303−309. (33) Briggs, D.; Brewis, D. M.; Dahm, R. H.; Fletcher, I. W. Surf. Interface Anal. 2003, 35, 156−167. (34) Williams, R. A.; Blanch, H. W. Biosens. Bioelectron. 1994, 9, 159− 167. (35) Johnsson, B.; Lofas, S.; Lindquist, G. Anal. Biochem. 1991, 198, 268−277. (36) Darst, S. A.; Ahlers, M.; Meller, P. H.; Kubalek, E. W.; Blankenburg, R.; Ribi, H. O.; Ringsdorf, H.; Kornberg, R. D. Biophys. J. 1991, 59, 387−96. (37) Algar, W. R.; Krull, U. J. Langmuir 2010, 26, 6041−6047. (38) Achermann, M.; Petruska, M. A.; Crooker, S. A.; Klimov, V. I. J. Phys. Chem. B 2003, 107, 13782−13787. (39) Lunz, M.; Bradley, A. L.; Chen, W. Y.; Gun’ko, Y. K. Superlattices Microstruct. 2010, 47, 98−102. (40) Lunz, M.; Bradley, A. L.; Chen, W. Y.; Gerard, V. A.; Byrne, S. J.; Gun’ko, Y. K.; Lesnyak, V.; Gaponik, N. Phys. Rev. B: Condens. Matter 2010, 81. (41) Kimura, J.; Uematsu, T.; Maenosono, S.; Yamaguchi, Y. J. Phys. Chem. B 2004, 108, 13258−13264. (42) Kim, D.; Okahara, S.; Nakayama, M.; Shim, Y. Phys. Rev. B: Condens. Matter 2008, 78. (43) Shcherbatyuk, G. V.; Inman, R. H.; Ghosh, S. J. Appl. Phys. 2011, 110. (44) Lunz, M.; Bradley, A. L.; Gerard, V. A.; Byrne, S. J.; Gun’ko, Y. K.; Lesnyak, V.; Gaponik, N. Phys. Rev. B: Condens. Matter 2011, 83. (45) Komarala, V. K.; Rakovich, Y. P.; Bradley, A. L.; Byrne, S. J.; Corr, S. A.; Gun’ko, Y. K. Nanotechnology 2006, 17, 4117−4122.

FRET assays as would be useful in applications such as pointof-care diagnostics and high-throughput screening in conjunction with microfluidics.

■

ASSOCIATED CONTENT

* Supporting Information S

Detailed experimental procedures, including preparation of water-soluble nanoparticles, characterization of PAAI polymer and optical properties of QDs, XPS data for relative atomic percent in stepwise synthesis, description of instrumentation used, characterization of FRET pairs and equations used. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge the Natural Sciences and Engineering Research Council of Canada (NSERC) for financial support of this research. E.P. is also grateful to NSERC for provision of graduate fellowship. The authors also would like to thank Dr. Rana Sodhi for assistance with XPS analysis and Dr. Neil Combs for assistance with SEM imaging.

■

ABBREVIATIONS MPA, mercaptopropionic acid; DHLA, dihydrolipoic acid; GSH, glutathione; dBSA, denatured bovine serum albumin; PAA, poly(acrylic) acid; PAAI, poly(acrylic) acid functionalized with appended imidazole ligands; APTMS, 3-aminopropyltrimethoxysilane; FRET, fluorescence resonance energy transfer; PL, photoluminescence

■

REFERENCES

(1) Algar, W. R.; Susumu, K.; Delehanty, J. B.; Medintz, I. L. Anal. Chem. 2011, 83, 8826−8837. (2) Lee, B.; Kim, Y.; Lee, S.; Kim, Y. S.; Wang, D. Y.; Cho, J. Angew. Chem. Int. Ed. 2010, 49, 359−363. (3) Gupta, S.; Uhlmann, P.; Agrawal, M.; Lesnyak, V.; Gaponik, N.; Simon, F.; Stamm, M.; Eychmuller, A. J. Mater. Chem. 2008, 18, 214− 220. (4) Li, C. L.; Murase, N. Langmuir 2004, 20, 1−4. (5) Lu, Q.; Hu, S. S.; Pang, D. W.; He, Z. K. Chem. Commun. 2005, 2584−2585. (6) Jiang, H.; Ju, H. X. Anal. Chem. 2007, 79, 6690−6696. (7) Chen, L.; Algar, W. R.; Tavares, A. J.; Krull, U. J. Anal. Bioanal. Chem. 2011, 399, 133−141. (8) Goldman, E. R.; Balighian, E. D.; Kuno, M. K.; Labrenz, S.; Tran, P. T.; Anderson, G. P.; Mauro, J. M.; Mattoussi, H. Physica Status Solidi B 2002, 229, 407−414. (9) Sapsford, K. E.; Medintz, I. L.; Golden, J. P.; Deschamps, J. R.; Uyeda, H. T.; Mattoussi, H. Langmuir 2004, 20, 7720−7728. (10) Medintz, I. L.; Sapsford, K. E.; Clapp, A. R.; Pons, T.; Higashiya, S.; Welch, J. T.; Mattoussi, H. J. Phys. Chem. B 2006, 110, 10683− 10690. (11) Tavares, A. J.; Noor, M. O.; Vannoy, C. H.; Algar, W. R.; Krull, U. J. Anal. Chem. 2012, 84, 312−319. (12) Algar, W. R.; Krull, U. J. Anal. Chem. 2009, 81, 4113−4120. (13) Algar, W. R.; Krull, U. J. Sensors 2011, 11, 6214−6236. (14) Robel, I.; Subramanian, V.; Kuno, M.; Kamat, P. V. J. Am. Chem. Soc. 2006, 128, 2385−2393. H

dx.doi.org/10.1021/la302985x | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

(46) Zhang, Y.; Zhang, H. Y.; Hollins, J.; Webb, M. E.; Zhou, D. J. Phys. Chem. Chem. Phys. 2011, 13, 19427−19436. (47) Algar, W. R.; Krull, U. J. J. Colloid Interface Sci. 2011, 359, 148− 154. (48) Algar, W. R.; Malonoski, A.; Deschamps, J. R.; Blanco-Canosa, J. B.; Susumu, K.; Stewart, M. H.; Johnson, B. J.; Dawson, P. E.; Medintz, I. L. Nano Lett. 2012, 12, 3793−3802. (49) Algar, W. R.; Wegner, D.; Huston, A. L.; Blanco-Canosa, J. B.; Stewart, M. H.; Armstrong, A.; Dawson, P. E.; Hildebrandt, N.; Medintz, I. L. J. Am. Chem. Soc. 2012, 134, 1876−1891. (50) Haiss, W.; Thanh, N. T. K.; Aveyard, J.; Fernig, D. G. Anal. Chem. 2007, 79, 4215−4221.

I

dx.doi.org/10.1021/la302985x | Langmuir XXXX, XXX, XXX−XXX