Article pubs.acs.org/ac
Secondary Ions Mass Spectrometric Signal Enhancement of Peptides on Enlarged-Gold Nanoparticle Surfaces Young-Pil Kim*,†,‡ and Tae Geol Lee*,§ †
Department of Life Science and ‡Research Institute for Natural Sciences, Hanyang University, Seoul 133-791, Republic of Korea § Center for Nano-Bio Convergence, Korea Research Institute of Standards and Science (KRISS), Daejeon 305-600, Republic of Korea ABSTRACT: A high surface coverage of gold nanoparticles (AuNPs) is a prerequisite for enhancing the peptide signal intensity in time-of-flight secondary ion mass spectrometry (TOF-SIMS). Here, we demonstrate the TOF-SIMS signal amplification of peptides on a surface by enlarging surfaceconfined AuNPs using the NH2OH/Au3+ seeding method. Because of the increased surface area and spherical structure of the Au, the SIMS intensity of the peptides became significantly enhanced on the enlarged-AuNPs surface, especially at high concentrations of peptide solution (>10 μM), compared to that of the bare gold surface or submonolyer of AuNPs. We are confident that this will be a useful method for diagnosis and bioassay with high sensitivity in a label-free manner.
T
as compared to a bare gold substrate or a monolayer of AuNPs. In this work, secondary ion signals of adsorbed peptides were analyzed in terms of the size of the enlarged-AuNP and peptide concentrations.
ime-of-flight secondary ion mass spectrometry (TOFSIMS) has been recognized as a useful tool, allowing the investigation of surface-immobilized biomolecules at the molecular level due to its chemical specificity and analytical sensitivity.1−3 Nonetheless, the ability to detect biomolecules of high molecular weight has not been fully realized because their signals are weak.4−6 Recently, in TOF-SIMS analysis, we found that gold nanoparticles (AuNPs) contributed to the ion signal enhancement of peptides by attaching themselves to the aminemodified self-assembled monolayers (SAMs) onto various solid substrates to function as a signal enhancer.7 In particular, we discovered that the vertical orientation of the peptides on the monolayer of AuNPs was a significant factor in the signal enhancement.8,9 On the basis of these observations, peptidebased bioassays, such as kinases and proteases, have been easily produced on the monolayer of AuNPs.8,9 To expand the versatility of nanoparticle-enhanced TOFSIMS, referred to as NE-SIMS, the AuNPs require a higher surface density. However, this effect cannot be achieved by directly depositing the colloidal AuNPs because of the repulsive interparticle interactions that not only induce sparse interparticle distribution but also inhibit further particle adsorption onto the substrate. A new method of controlling the surface coverage of AuNP, its morphology, as well as interparticle distance would allow for more biological applications of NE-SIMS to reliably detect the desired analytes. Here we describe a method of enhancing the TOF-SIMS signal of peptides by enlarging colloidal AuNPs on a surface. The AuNP-catalyzed reduction was induced by additional Au ions and NH2OH, as reported elsewhere.10−12 The reduction allows small particles to be grown into larger particles of a predetermined size, resulting in a larger surface density of gold, © 2012 American Chemical Society
■
EXPERIMENTAL METHODS Materials. The following materials were obtained from Sigma-Aldrich: hydrogen tetrachloroaurate(III) trihydrate (99.9% HAuCl4·3H2O), sodium citrate dihydrate (trisodium salt, C6H5Na3O7·2H2O), sodium borohydride (99%, NaBH4), hydroxylamine hydrochloride (99%, NH2OH), 3-aminopropyltriethoxysilane (APTES), and angiotensin I peptide (NRVYIHPFHL, Mr = 1294.69). Cysteine-tethered Abl peptide (Ac-IYAAPKKGGGGC, Mr = 1162.58) was synthesized by Peptron, Inc. (Korea). All aqueous solutions were made using distilled water. All of the chemicals were of analytical grade and were used as received. Sample Preparation. Gold nanoparticles (AuNPs) were synthesized by reduction and stabilization with citrate as described elsewhere.13 Briefly, 0.1 g of HAuCl4·3H2O (0.1%, w/w) was dissolved into 100 mL of distilled water and vigorously stirred for 1 min. To this solution, 0.02 g of sodium citrate dihydrate (2-hydroxy-1,2,3-propanetricarboxylic acid) was added, and again the solution was stirred. After 1 min, for the reduction and formation of gold colloids, 85 μL of a stock solution containing 11.4 mg of NaBH4 in 1 mL of distilled water was quickly added to the reaction solution, followed by Received: February 2, 2012 Accepted: April 17, 2012 Published: April 17, 2012 4784
dx.doi.org/10.1021/ac300336h | Anal. Chem. 2012, 84, 4784−4788
Analytical Chemistry
Article
Figure 1. SEM images (top) and side-on schematic (bottom) of 3 nm-AuNP (A) and its enlarged-AuNP surface at 1 min (B) and 3 min (C) after adding gold ion and NH2OH onto Si/SiO2 substrates. The scale bar is 200 nm.
0.5 pA (Bi1+) at 5 kHz. A pulse with of 0.7 ns from the bunching system resulted in mass resolution exceeding M/ΔM = 104 (full width at half-maximum [fwhm]) at m/z > 500 in both the positive and negative modes. The analysis area (500 × 500 μm2) was randomly rastered by the primary ions, and the primary ion dose was maintained below 1012 ions cm−2 to ensure static SIMS conditions. Positive ion spectra were internally calibrated by using the H+, H2+, CH3+, C2H3+, and C3H4+ signals. Each peak was normalized by using the relative value of total intensity to reduce the surface variation.
stirring for 5 min. The completely reduced solution containing 254 μM Au was stored at 4 °C. The clustering of AuNPs was estimated by UV−visible spectroscopy (UV-2550, Shimadzu), and the size of AuNPs was confirmed to be 3.2 nm (±0.4 nm SD, n = 100) using energy-filtering transmission electron microscopy (EF-TEM, EM912 Omega, Carl Zeiss, Germany). Monolayers of AuNPs were formed by depositing AuNPs for 30 min on the amine-functionalized Si surface, followed by washing with distilled water. The peptides, dissolved in water, were directly adsorbed onto bare gold or AuNPs monolayer or enlarged-AuNPs assembly for 60 min with the same range of peptide concentration (0.01−10 μM), and the surfaces were sequentially washed with distilled water and dried under a stream of N2. Formation of Colloidal Gold Monolayers and Enlargement by NH2OH/Au3+ Seeding. Si or Au coated wafer was cut into 10 mm × 10 mm pieces and cleaned for 20 min in a piranha solution (1:4) 30% H2O2/concentrated H2SO4 (v/v). (Caution: piranha solution reacts violently with most organic materials and must be handled with extreme care.) The cleaned Si substrates were rinsed thoroughly with deionized water and methanol, then placed in a 10% dilute solution of APTES in methanol for 12 h, and rinsed with copious amounts of methanol upon removal. The APTES-coated slides were subsequently immersed in colloidal gold solution for 2 h for AuNPs assembly. The gold monolayer was rinsed with water and used immediately for plating. Substrates having a monolayer of nanosized gold particles were immersed in 1 mL of aqueous 0.4 mM (5 μL of 80 mM) hydroxylamine hydrochloride and 0.3 mM (1 μL of 300 mM) HAuCl4·3H2O at a given time. The deposition was stopped by removing the sample and rinsing it with water. Field Emission Scanning Electron Microscope (FESEM). Scanning electron microscopy was used to study the surface structures. The monolayers of AuNPs were attached to a sample stub with double-sided sticky tape and examined using a JEOL JSM-5410 LV scanning electron microscope (JEOL UK Ltd., Welwyn Garden City, Hertfordshire, England) at an accelerating voltage of 10 or 15 kV. TOF-SIMS Analysis. Ion spectra measurements by TOFSIMS were carried out with a TOF-SIMS V instrument (IONTOF GmbH, Germany) using a 25 keV Bi1+ primary ion beam source. The ion currents were measured using a Faraday cup located at the grounded sample holder and determined to be
■
RESULTS AND DISCUSSION As shown in Figure 1A, the large areas of interparticle distance between the colloidal AuNPs that make up the submonolayer is clearly visible in the SEM image when 3 nm-AuNPs were deposited onto the APTES-modified Si surface. The spaces are the result of electrostatic repulsion between the negatively charged AuNPs, although the negatively charged AuNPs are attaching to the positively charged amino groups via electrostatic attraction. Since a repulsive force between the negatively charged AuNPs prevents them from closely packing together or piling up on top of each other, the surface fractional coverage, which is a relative percentage divided by the two-dimensional hexagonal packing density of gold, was estimated to be only 16.5% according to our previous report.7 To attain a higher surface density of gold, the surface-confined AuNPs were subjected to further growth after treating with NH2OH/Au3+ seeding. The result by an enlargement of the gold area was shown as a function of time (Figure 1B,C). From SEM images, the surface coverage of Au on each surface was determined to be 18% (control AuNP surface in Figure 1A), 77% (1 min enlargement in Figure 1B), and 95% (3 min enlargement in Figure 1C) when normalized to that of the bare gold surface (100%). To get some insight into how different surfaces affect the secondary ion (SI) intensity, four different surfaces including bare gold were treated with a high concentration (10 μM) of peptide solution and analyzed. A cysteine-tethered peptide was chosen because the −SH group attaches strongly onto gold, resulting in a higher surface density, compared to peptides without the −SH group.8 Previously, we reported observing quasimolecular secondary-ion signals of [MH−SH−COOH]+ from the peptide at m/z 1085.61 rather than molecular ion signals, [MH]+. In Figure 2, the SI intensity of [Au3]+ which 4785
dx.doi.org/10.1021/ac300336h | Anal. Chem. 2012, 84, 4784−4788
Analytical Chemistry
Article
Figure 2. Positive secondary ion spectra and corresponding ion intensity from Au (A and C) and cysteine-tethered Abl peptide (B and D) on different surfaces: bare Au (black), AuNPs (light gray), 1 min-enlarged (dark gray) and 3 min-enlarged AuNPs surface (white). [Au3]+ and [MH− SH−COOH]+ were used as gold and peptide characteristic peaks, respectively. The full width at half-maximum intensity of the peak was normalized by multiplying the ratio of sample total SI to control total SI (bare Au) in order to eliminate systematic differences between spectra.
corresponds to m/z 591 increased with AuNP-enlargement (the second−fourth panels in Figure 2A), but it was relatively lower than that of bare gold (the first panel in Figure 2A). This result reveals that the SI from Au on the surface was attributed to the increased gold surface area. In contrast to the SI intensity of [Au3]+, the quasi-molecular SI intensity of the peptides ([MH−SH−COOH]±) was most distinct for 1 min-enlarged AuNP (the third panel in Figure 2B) compared to the other surfaces (the first, second, and fourth panels in Figure 2B). On the basis of this result, it is presumed that the peptideadsorptive process and relevant SI intensity of the peptide on a surface were critically affected by both the 3D gold nanostructure and the gold surface area. In the previous study,7 it was reported that the SI emission of peptide adsorbed on the rough evaporated gold was relatively high compared to that on the flat mica gold, due to the increased adsorption of peptides by the globular structure of gold. Likewise, a threedimensional surface coverage of gold is likely to be maximized in 1-min-enlarged AuNP surfaces, leading to the maximum binding density of peptides. Further enlargement of the AuNP surface, despite the increased amount of gold, may lose the three-dimensional structures like flat bare gold, resulting in low SI intensity of the peptide (Figure 2C). When the longer duration (5 min or more) for AuNP enlargement was tested
(data not shown), the SI intensity of gold increased but the SI intensity of peptide reduced, which was very similar to that of bare gold. In the case of the AuNP surface (Figure 2A), its gold surface area was too low to be fully covered on the measuring surface even if it has a 3D spherical structure, thereby generating a relative low SI intensity of peptide, compared to that of bare Au. This result indicates that the enlarged-AuNPs surfaces (especially for 1-min enlargement), due to the globular nanostructure of the gold, encourage a higher surface density of peptides, leading to high ion intensity of the peptide. This also may reflect one TOF-SIMS characteristic that its signals originate from the uppermost layers (10−15 Å) of a surface, where most of surface-attached peptides are sputtered by the gold. The two peaks ([Au3]+ and [MH−SH−COOH]+) were compared on different surfaces (Figure 2C,D) after the intensities were normalized by multiplying the ratio of total SI of the sample surface to total SI of the bare Au surface in order to eliminate systematic differences between spectra. There were no contaminative peaks (i.e., organic molecules or metal ions except Au) on bare gold (Figure 2A) because it was treated with piranha solution prior to use; therefore, Au secondary ion is likely to be produced without any interference by other molecules. On the other hand, some contaminants were found to be on the AuNP and AuNP-enlarging surface 4786
dx.doi.org/10.1021/ac300336h | Anal. Chem. 2012, 84, 4784−4788
Analytical Chemistry
Article
due to the substrates and modifying chemical groups (i.e., Si, APTES, citrate, and NH2OH-related peaks, data not shown), and these contaminants are engaged in reducing the sputtering effect of Au and peptide ions. For example, the total ion intensity from the Si/APTES/enlarged AuNPs was approximately 2-fold lower than that from bare Au, even if gold surface coverage is relatively high in the enlarged AuNP surface. This observation has been well documented in alkanthiolate SAM on the AuNP, where there was a decrease in the Au-containing fragments due to increased attenuation of Au signals by the thicker overlayer.14 However, considering that the surface layer on AuNPs, in the present study, is not as thick as alkanthiolate SAMs, we believe that the sputtering and ionization process of peptides was affected little by the contaminants on the AuNP regions. In particular, the SI yield of the cysteine-tethered peptides on the enlarged AuNP surface is likely to be higher than observed from bare Au surface, which is very similar to that of alkanthiolates on the AuNP when they were analyzed by combining TOF-SIMS and XPS.14 We believe that, in addition to the surface gold coverage, a globular structure of AuNP would be advantageous for reducing the contaminant effect, due to a faster energy deposition and desorption by a radial direction. A similar result was observed for cysteine-free peptide (NRVYIHPFHL) when its SI intensity was analyzed on different surfaces; that is, the highest intensity was generated from the 1 min-enlarged AuNP surface, regardless of a sulfhydryl group of the peptide, suggesting that this approach is applicable to other peptides (Figure 3). Assuming that the
Figure 4. Changes in TOF-SIMS intensity from peptides on different surfaces: bare Au (black bar), AuNPs (light gray bar), and 1 minenlarged AuNPs (dark gray bar) as a function of peptide concentration. Quasi-molecular secondary ion signal [MH−SH− COOH]+ was used as a characteristic peak of the peptide and was normalized as noted in Figure 2. The standard deviation was obtained from three independent experiments.
concentration increased, but the highest signal intensity of SIMS occurred when a highly concentrated peptide solution (>10 μM) was applied to the enlarged-AuNP surface. Interestingly, however, at peptide concentration levels of 0.1 μM or below, the SI intensity from bare gold surface was greater than that of the AuNP or enlarged-AuNP surfaces, suggesting that a relatively high concentration of peptides is necessary to saturate the peptides on the enlarged-AuNPs surface, which then triggers an amplification of SIMS intensity. Nonetheless, longer deposition of low concentration peptides (
