Colloidal Stability of Gold Nanoparticles Modified with Thiol

Jan 25, 2012 - (11, 14-18) It has been reported that the scattering light intensity of a 60 nm GNP is equivalent to that of 5 × 105 fluorescein molec...
1 downloads 11 Views 4MB Size
Download PDF
Article pubs.acs.org/Langmuir

Colloidal Stability of Gold Nanoparticles Modified with Thiol Compounds: Bioconjugation and Application in Cancer Cell Imaging Jie Gao, Xiangyi Huang, Heng Liu, Feng Zan, and Jicun Ren* College of Chemistry & Chemical Engineering, State Key Laboratory of Metal Matrix Composites, Shanghai Jiaotong University, Shanghai 200240, People’s Republic of China ABSTRACT: Gold nanoparticles (GNPs) are attractive alternative optical probes and good biocompatible materials due to their special physical and chemical properties. However, GNPs have a tendency to aggregate particularly in the presence of high salts and certain biological molecules such as nucleic acids and proteins. How to improve the stability of GNPs and their bioconjugates in aqueous solution is a critical issue in bioapplications. In this study, we first synthesized 17 nm GNPs in aqueous solution and then modified them with six thiol compounds, including glutathione, mercaptopropionic acid (MPA), cysteine, cystamine, dihydrolipoic acid, and thiolending polyethylene glycol (PEG-SH), via a Au−S bond. We systematically investigated the effects of the thiol ligands, buffer pH, and salt concentrations of the solutions on the colloidal stability of GNPs using UV−vis absorption spectroscopy. We found that GNPs modified with PEG-SH were the most stable in aqueous solution compared to other thiol compounds. On the basis of the above results, we developed a simple and efficient approach for modification of GNPs using a mixture of PEG-SH and MPA as ligands. These biligand-modified GNPs were facilely conjugated to antibody using 1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide and N-hydroxysulfosuccinimide as linkage reagents. We conjugated GNPs to epidermal growth factor receptor antibodies and successfully used the antibody−GNP conjugates as targeting probes for imaging of cancer cells using the illumination of a dark field. Compared to current methods for modification and conjugation of GNPs, our method described here is simple, has a low cost, and has potential applications in bioassays and cancer diagnostics and studies.

1. INTRODUCTION In the past several decades, various bioimaging techniques have been emerging with the rapid development of biomedicine science and technology, and they have become very important tools in basic studies of biology and biomedicine and clinical diagnosis. Among bioimaging techniques, optical imaging modes provide great potential for cell and small animal imaging with their outstanding advantages such as no harm to the tissues, high sensitivity and spatial resolution, etc. So far, optical imaging of living cells and tissues is widely utilized for microscopic structure analysis, in vivo determination of specific binding properties, and sensitive diagnosis of cancers. However, optical imaging methods are dramatically dependent on the chemical and physical properties of labeling probes. Currently, the most commonly used optical labels are organic fluorescent dyes, but the drawback of the organic dyes is limited observation time owing to rapid photobleaching, which restricts their applications in this field.1 Recently, some nanoparticles have been used as labeling probes in optical imaging, which mainly include quantum dots (QDs),2,3 carbon nanotubes,4,5 magnetic nanoparticles,6−8 and gold nanoparticles (GNPs).9−11 Although QDs have high brightness and photostability over organic dyes,2,3 they are still unsuitable for some in vivo © 2012 American Chemical Society

applications due to their certain degree of photodecomposition, strong biological toxicity, and difficulty in functionalizing in a controlled way.12,13 Carbon nanotubes and magnetic nanoparticles are also limited for their toxicity and opsonization.5,8 Compared with other optical probes, GNPs have such outstanding advantages as noncytotoxicity, excellent biocompatibility, ease of synthesis and surface functionalization, strong light absorption and scattering effect, high photothermal conversion rate and photostability, and so on.11,14−18 It has been reported that the scattering light intensity of a 60 nm GNP is equivalent to that of 5 × 105 fluorescein molecules.19 Due to their excellent features, GNPs as labeling probes have been used in immunoassays, single particle tracking, and cell imaging.9,20−22 However, colloidal GNPs have a tendency to easily aggregate particularly in the presence of high salts and certain biological molecules such as nucleic acids and proteins. Although the aggregation of GNPs is useful for certain events of biomolecular recognition, GNPs must be stably dispersed in Received: November 1, 2011 Revised: January 23, 2012 Published: January 25, 2012 4464

dx.doi.org/10.1021/la204289k | Langmuir 2012, 28, 4464−4471

Langmuir

Article

After the color changed, the solution was refluxed for an additional 15 min, allowed to cool to room temperature, and finally placed in a refrigerator for further use. All solutions were prepared with ultrapure water (18.2 MΩ) purified on the Millipore Simplicity system. Transmission electron microscopy (TEM) images were taken with a JEM-100CX transmission electron microscope. UV−vis absorption spectra of the GNPs were obtained by using a UV-3501 spectrophotometer (Tianjin Gangdong Science & Technology Development Co. Ltd., China). The hydrodynamic diameter of the GNPs was measured by RLSCS. The principle and setup of RLSCS are similar to those of fluorescence correlation spectroscopy (FCS), which is based on an inverted Olympus IX 71 microscope (Tokyo, Japan). The details of the experimental setup can be found elsewhere.19 In brief, an argon laser with a 488 nm wavelength (ILT Technology, Shanghai, China) was reflected by a dichroic mirror (505DRLP, Omega Optical, Brattleboro, VT) and then focused into the sample solution by a water immersion objective (UplanApo, 60×, NA (numerical aperture) = 1.2, Olympus). The sample was placed on a coverslip (thickness 170 μm). The scattering signal was collected after the 35 μm pinhole was passed by an avalanche photodiode (SPCMAQR14, Perkin-Elmer EG&G, Canada). The signals obtained were recorded by a real time digital collector (Flex02-12D/C, Correlator.com). 2.3. Modification of GNPs with Thiol Compounds. Six kinds of thiol compounds, including PEG-SH, glutathione, MPA, cysteine, cystamine, and DHLA, were used to modify GNPs in this study. In the modification approach, a solution of PEG-SH (5 × 10−4 M, 100 μL) was added into 1 mL of GNP solution (about 3.2 nM), and then the resulting solution was stirred for 30 s. This mixture was allowed to react at 4 °C overnight. Modification of GNPs with other ligands was carried out in the same way as described above.40 The ζ potentials of modified GNPs were measured by using a Zetasizer Nano (Malvern, ZS90, Malvern Instruments Ltd., Worcestershire, U.K.). The hydrodynamic diameters of modified GNPs were determined by RLSCS and the Zetasizer Nano. 2.4. Colloidal Stability of Modified GNPs to pH and Electrolytes. We investigated the effects of pH and electrolyte changes on the colloidal stability of GNPs in aqueous solution by using UV−vis absorption spectroscopy. The solutions with the desired pH values were prepared by adding a few drops of HCl or NaOH to 1× phosphate buffer (PB; 0.01 M, pH 7.4; 29.6 mg of NaH2PO4 and 111 mg of Na2HPO4 dissolved in 50 mL of ultrapure water) stock solutions. The solutions with electrolytes of different concentrations were prepared by adding a given amount of NaCl solution to 1× PB stock solutions. Small volumes of the nanoparticles were added to the solutions with final nanoparticle concentrations of 1 nM for GNPs. 2.5. Conjugation of GNPs with Antibody. Conjugation of GNPs with antibody was carried out using EDC and Sulfo-NHS as linking reagents according to the protocol described in ref 41. Briefly, 15 μL of EDC (1 mg/mL) and 15 μL of Sulfo-NHS (1 mg/mL) were first added into 1.2 mL of GNP solution modified with MPA (or GSH) and PEG-SH, which amidated the carboxyl terminus of MPA on the GNPs, and then a given amount of antibody was added to the solution above. The mixed solution was stirred for 2 min, reacted for 2 h at room temperature, and stored for 12 h at 4 °C. The final GNP− antibody suspension was centrifuged to remove unbound antibody. The deposition was redissolved in phospate-buffered saline (PBS) and was stored at 4 °C for use. 2.6. Cell Culture. SiHa cervical cancer cells were used as a model for cell imaging. SiHa cells showed overexpression of EGFR in epithelial precancers.42 The culture of SiHa cells was preformed using a reported method with a minor modification.9 The cells were cultured with high Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum (Hyclone) in a humidified incubator (MCO15AC, Sanyo, Osaka, Japan) at 37 °C under 5% CO2. In our experiments, a 1 mL solution with approximately 8000 cells was seeded onto the coverslip, and then GNP−antibody conjugates were added into the medium grown for 1 h at 37 °C under 5% CO2. The GNP−antibody conjugates can bind to the EGFR of SiHa cells specifically by immunoreactions. The cells on the coverslip were first

biological fluids in most applications, for example, in drug delivery, cell and animal imaging, and diagnostic assays. Up to now, some compounds such as surfactants, cyclodextrin, and thiol compounds have been used to modify nanoparticles (such as GNPs) to improve their stability, dispersibility, and biocompatibility.23−27 Compared to other compounds, thiol compounds can efficiently improve the stability and dispersity of colloidal GNPs in aqueous solution. This is mainly attributed to the fact that the thiol groups of these compounds can bind covalently to the surface of GNPs via a Au−S bond. Currently used thiol compounds mainly include glutathione (GSH), mercaptopropionic acid (MPA), cysteine, cystamine, dihydrolipoic acid (DHLA), thiol-ending polyethylene glycol (PEG-SH), some derivatives, etc.28−39 In fact, thiol-containing compounds with specific chemical structures show different affinities for GNPs, which probably affect the colloidal stability of GNPs in aqueous solution. To the best of our knowledge, we have not found a systematic investigation on the effects of various thiol compounds on the colloidal stability of GNPs in aqueous solution. In this study, we want to choose some commonly available thiol-containing compounds as modification ligands of GNPs and evaluate the abilities of these thiol ligands for stabilizing colloidal GNPs in aqueous solutions with different pH values and salt concentrations. Furthermore, on the basis of the results obtained, we develop a simple and efficient method for modification of GNPs. We first modified GNPs with six thiol compounds, including PEG-SH, glutathione, MPA, cysteine, cystamine, and DHLA, and then systematically investigated the effects of the thiol ligands and pH and salt concentrations of the solutions on the colloidal stability of the GNPs using UV− vis spectroscopy and resonance light scattering correlation spectroscopy (RLSCS). We found that PEG-SH-modified GNPs were most stable in aqueous solution compared to other thiol compounds. On the basis of the results above, we developed a simple and efficient approach for modification of GNPs using a mixture of PEG-SH and MPA as mixed ligands. The biligand-modified GNPs showed good stability in aqueous solution, and they were facilely conjugated to the antibody of epidermal growth factor receptor (EGFR) using 1-ethyl-3-[3(dimethylamino)propyl]carbodiimide hydrochloride (EDC) as a coupling reagent. Furthermore, we successfully applied GNP−antibody conjugates to target the EGFR of SiHa cancer cells for dark field imaging.

2. EXPERIMENTAL SECTION 2.1. Chemicals and Reagents. PEG-SH (5 kDa), MPA, DHLA, GSH, EDC, and N-hydroxysulfosuccinimide (Sulfo-NHS) were products of Sigma-Aldrich Chemical Co. (Milwaukee, WI). Anti epidermal growth factor receptor (anti-EGFR) antibodies (Erbitux) were from Merck Co. (Darmstadt, Germany). Sodium citrate, disodium hydrogen phosphate, sodium dihydrogen phosphate, and hydrogen tetrachloroaurate(III) hydrate (HAuCl4) were purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China). Other chemical reagents used in this study are from Sigma-Aldrich. Ultrapure water (18.2 MΩ) was obtained from the Millipore Simplicity system (Millipore, Bedford, MA). 2.2. Synthesis and Characterization of GNPs. Approximately 17 nm GNPs were synthesized using a reported method with a minor modification.14 Briefly, all glassware was cleaned in aqua regia (3 parts HCl, 1 part HNO3), rinsed with H2O, and then oven-dried prior to use. A mixture containing 5 mL of HAuCl4 (0.2%, w/w) and 90 mL of water was brought to reflux while being stirred, and then 5 mL of sodium citrate trihydrate solution (1%, w/w) was added quickly, which resulted in a change in the solution color from pale yellow to deep red. 4465

dx.doi.org/10.1021/la204289k | Langmuir 2012, 28, 4464−4471

Langmuir

Article

Figure 1. Characterization of as-prepared GNPs. (a, b) TEM image of GNPs and the corresponding statistical result of GNP sizes. The size of the GNPs used in this study was about 17 nm. (c) UV−vis absorption spectrum of GNPs. (d) RLSCS curve of GNPs. The measurement time was 300 s.

Table 1. ζ Potentials and Hydrodynamic Diameters of Unmodified and Modified GNPsa

rinsed with PBS (0.01 M, pH 7.4; 0.8 g of NaCl, 0.02 g of KCl, 0.144 g of Na2HPO4, and 0.024 g of KH2PO4 dissolved in 100 mL of ultrapure water) and then directly observed with dark field microscopy. 2.7. Cellular Imaging. Dark field images were taken using an inverted Olympus IX 71 microscope equipped with a condenser (NA = 0.3−1.2) which delivers a very narrow beam of white light from a 100 W halogen tungsten lamp on the top of the samples. A 60× objective (NA = 0.75) was used to collect only the scattered light from the samples. When the light beam direction is optimized, the center illumination light beam does not enter the light collection cone of the microscope objective, and only the scattered light of the side beam by the sample is collected. This presents an image of a bright object in a dark background. The setup of dark field imaging is similar to that of the system described in ref 43. The unit type of our charge-coupled device (CCD) is JVC TK-C921EC (A) with an exposure duration of 1/24 s. SiHa cells labeled with anti-EGFR antibody provide a model for cancerous cells. In a control experiment, the imaging of SiHa cells was preformed in the same conditions using bovine serum albumin (BSA) instead of anti-EGFR antibody. GNP−BSA conjugates were prepared according to the same procedure as GNP−anti-EGFR conjugates.

ligand

ζ potential (mV)

citrate PEG-SH MPA GSH cysteine DHLA

−25.2 ± 0.4 −3.8 ± 0.9 −35.7 ± 1.2 −39.7 ± 1.3 −45.5 ± 0.7 −39.8 ± 2.0

hydrodynamic diameter (nm), Zetasizer Nano 24.9 51.4 28.4 28.3 35.4 28.5

± ± ± ± ± ±

0.2 1.3 0.8 0.6 0.8 0.9

hydrodynamic diameter (nm), RLSCS 22.8 40.3 27.5 26.3 29.2 29.6

± ± ± ± ± ±

1.2 2.6 2.6 1.6 1.0 1.2

The ζ potentials were measured by a Zetasizer Nano-ZS90, and hydrodynamic diameters were measured by the Zetasizer Nano-ZS90 and RLSCS. The concentrations of the thiol ligands were 0.5 mM, and the GNP concentration was 1 nM. Citrate was the original ligand from the preparation of GNPs (referred to as unmodified GNPs). The measurement of each sample was repeated three times. a

good monodispersion and their size was about 17 nm. Furthermore, the hydrodynamic diameter of GNPs was measured by RLSCS similar to FCS. The autocorrelation curve of GNPs and the fitting curve are shown in Figure 1d. This result documented that the autocorrelation curve was well fitted with the theoretical mode of the autocorrelation function and the correlation coefficient (R2) is 0.9998. The hydrodynamic diameter of GNPs obtained is about 23 nm, and this result is basically identical with that of TEM imaging when the contribution of the hydrated outer shell of the GNPs was stripped away. The modification of GNPs is described in the Experimental Section. To confirm the successful modification of GNPs with thiol compounds, we measured the ζ potentials and the hydrodynamic diameters of unmodified and modified

3. RESULTS AND DISCUSSION 3.1. Synthesis, Characterization, and Modification of GNPs. GNPs used in this study were synthesized by using reduction of the HAuCl4 solution with sodium citrate, and the synthesis approach is described in the Experimental Section. The as-prepared GNPs were characterized by UV−vis absorption spectroscopy and TEM methods. The absorption spectrum and TEM image of the as-prepared GNPs are shown in Figure 1 a,b. UV−vis absorption spectroscopy documented that the maximum absorption of GNPs was at about 520 nm. The TEM results illustrated that the as-prepared GNPs had 4466

dx.doi.org/10.1021/la204289k | Langmuir 2012, 28, 4464−4471

Langmuir

Article

Figure 2. UV−vis absorption spectra of GNPs modified with thiol compounds in different pH solutions. PB was used in this study, and the buffer pH values were 5.4, 6.4, 7.4, 8.4, and 9.4. The thiol compounds include PEG-SH (a), GSH (b), MPA (c), DHLA (d), and cysteine (e). The different pH values are distinguished with curves of diverse colors shown in the legends. (f) Increase in the absorbance at 600 nm of GNPs at pH 6.4 corresponding to the data from previous figures. The GNP concentration was about 1 nM, and the concentrations of all thiol compounds were 0.5 mM. The black curve (GNPs) expresses the original solution of GNPs, which is used as a contrast sample.

3.2. Colloidal Stability of GNPs Modified with Thiol Compounds. Surface-modified or -functionalized GNPs are necessary in bioapplications. Surface modification not only improves the stability and biocompatibility of GNPs, but also functionalizes the GNP surface, which is facilely linked with biomolecules such as antibodies and enzymes. Thiol compounds have strong affinity for GNPs due to the formation of a Au−S bond, and in this study, six kinds of thiol compounds (such as PEG-SH, glutathione, MPA, cysteine, cystamine, and

GNPs, and the results obtained are shown in Table 1. These data indicated the ζ potentials and the hydrodynamic diameters of modified GNPs were significantly different from those of unmodified GNPs. These results illustrated that the GNPs were successfully modified with thiol compounds. Compared to other thiol compounds, PEG-SH-modified GNPs showed larger changes in ζ potentials and hydrodynamic diameters, which was mainly attributed to the fact that PEG-SH was a polymer of 5 kDa molecular mass and carried no charge in aqueous solution. 4467

dx.doi.org/10.1021/la204289k | Langmuir 2012, 28, 4464−4471

Langmuir

Article

Figure 3. UV−vis absorption spectra of GNPs modified with thiol compounds in solutions of different NaCl concentrations. The NaCl concentrations in PB were 0.005, 0.01, 0.02, 0.04, and 0.08 M. The thiol compounds include PEG-SH (a), GSH (b), MPA (c), DHLA (d), and cysteine (e). The different concentrations of NaCl are distinguished with curves of diverse colors shown in the legends. (f) Increase in the absorbance at 600 nm of GNPs in 0.04 M NaCl solution corresponding to the data from previous figures. The GNP concentration was about 1 nM, and the concentrations of all thiol compounds were 0.5 mM. The black curve (GNPs) expresses the original solution of GNPs, which is used as a contrast sample.

We first investigated the effects of the solution pH on the stability of GNPs modified with thiol compounds. The pH range of PB solutions used was from 5.4 to 9.4. UV absorption spectroscopy was used to efficiently characterize the colloidal stability of modified GNPs. The UV absorption spectra of GNPs were significantly dependent on the sizes of the GNPs. With an increase in the diameters of the GNPs, their absorption spectra showed a dramatic red shift. It was reported that the aggregation of GNPs led to a dramatic red shift in the absorption spectra of the GNPs due to the significant

DHLA) were used to modify GNPs. The modification details are described in the Experimental Section. However, in initial experiments, we found that an extremely low concentration of cystamine (5 × 10−8 M) dramatically induced the aggregation of GNPs in solution. This result implied that cystamine failed to provide colloidal stability to GNPs in either buffers of diverse pH or different electrolytes. Therefore, cystamine was not used further in the following experiments. 4468

dx.doi.org/10.1021/la204289k | Langmuir 2012, 28, 4464−4471

Langmuir

Article

Figure 4. UV−vis absorption spectra of GNPs modified with biligands. The ratio of PEG-SH to MPA was 1:10, the nanoparticle concentration was about 1 nM, and the concentrations of MPA and PEG-SH were 0.5 mM. The black curve (GNPs) expresses the original solution of GNPs, which is used as a contrast sample.

Figure 5. UV−vis absorption spectra of GNPs before and after conjugation of antibody (anti-EGFR). The GNP concentration was about 1 nM, and the antibody and ligand concentrations used in this section were 1 mg/mL and 0.5 mM, respectively. The antibody was dissolved in 1× PB before use. The black curve (GNPs) expresses the original solution of GNPs, which is used as a contrast sample.

shortening of the distance between the GNPs,44−46 which was similar to an increase in the size of the GNPs. Parts a−e of Figure 2 show the UV absorption spectra of GNPs modified with a thiol compound in PB of pH ranging from 5.4 to 9.4. As shown in Figure 2a, GNPs modified with PEG-SH have almost no change in their absorption spectra when the buffer pH is from 5.4 to 9.4. This result documented that GNPs modified with PEG-SH were stable enough against either certain acidic environments or alkaline environments. Compared to PEG-SH, the absorption spectra of GNPs modified with other thiol compounds considerably changed with the change of the buffer pH, especially at pH 5.4. The data illustrated that GNPs modified with these thiol compounds were not as stable as GNPs modified with PEG-SH. Furthermore, we chose pH 6.4 to quantify the increase in the absorbance of GNPs at 600 nm wavelength corresponding to the data from previous figures, and the data are shown in Figure 2f. This further supported the above result that GNPs modified with PEG-SH were the most stable in buffer of pH 5.4−9.4. Next, we study the effects of salt concentrations on the stability of GNPs modified with thiol compounds. In this study, the concentration of NaCl used ranges from 0.005 to 0.08 M, and the absorption spectra of GNPs modified with thiol compounds are shown in Figure 3a− e. From Figure 3, we observed that the absorption spectra of GNPs modified with PEG-SH were relatively stable when the concentration of sodium chloride increased from 0.005 to 0.08 M. This result indicated that GNPs modified with PEG-SH were very stable and no aggregation of GNPs was observed in a

solution of sodium chloride. However, GNPs modified with other thiol compounds showed poor stability in aqueous solution, and their absorption spectra, as shown in Figure 3b−e, significantly broadened with the concentration of sodium chloride. Meanwhile, we also chose the concentration of sodium chloride (0.04 M) to quantify the increase in the absorbance of GNPs at 600 nm wavelength corresponding to the data from previous figures. This result further documented that PEG-SH was the most efficient ligand to improve the colloidal stability of GNPs in aqueous solution. The stability of nanoparticles in solution is mainly dependent on the surface properties of nanoparticles such as surface charge and ligand structure. Generally, the increase in the electrostatic repulsion of nanoparticles and steric hindrance of the nanoparticles' surfaces can significantly improve the stability of nanoparticles in solution. In this case, the surface charge of nanoparticles considerably decreased with an increase in the electrolyte concentration (NaCl), which led to aggregation of nanoparticle due to the reduction in the electrostatic repulsion of nanoparticles. Compared to other thiol compounds, PEG-SH is a polymer, and the surface on GNPs modified with PEG-SH was coated with a lot of polymers, which resulted in an increased stability of GNPs in solution due to an increase in the steric hindrance of GNPs. In fact, we found that the PEG-SHmodified GNP colloid remained very stable even if in 1 M NaCl solution (data not shown), which was far higher than the concentration of physiological saline solution. This opens up

Scheme 1. Schematic Illustration for Modification of GNPs with PEG-SH and MPA and Conjugation with Antibody

4469

dx.doi.org/10.1021/la204289k | Langmuir 2012, 28, 4464−4471

Langmuir

Article

Figure 6. Dark field images of SiHa cancer cells: (a) SiHa cells incubated with GNP−antibody conjugates using biligands of MPA and PEG-SH; (b) SiHa cells without any treatment; (c) SiHa cells incubated with GNP−antibody conjugates using biligands of GSH and PEG-SH; (d) SiHa cells incubated with GNP−BSA conjugates.

clearly distinguished from the image in Figure 6b. In this case, we used GNP−anti-EGFR antibodies prepared by using GNPs modified with biligands of MPA and PEG-SH. Furthermore, GNP−anti-EGFR antibodies were prepared by using GNPs modified with biligands of GSH and PEG-SH and successfully used in imaging of SiHa cancer cells, which is shown in Figure 6c. We can observe the distribution of many GNPs on the cell surface, which is similar to Figure 6a. To further confirm GNP−anti-EGFR conjugates specifically bound on the cell membrane, the contrast experiments were carried out at the same conditions using GNP−BSA conjugates instead of GNP− anti-EGFR conjugates. GNP−BSA conjugates were prepared according to the same procedure as GNP−anti-EGFR conjugates. Figure 6d shows the image of SiHa cells using GNP−BSA conjugates as probes, which is significantly different from the images in Figure 6a,c. As expected, only a little unspecific binding of GNP−BSA conjugates to cell membranes was observed. Furthermore, we observed that certain GNPs were nonspecifically bound to cell membranes when GNP− BSA conjugates were directly prepared by the adsorption method (data not shown). This result was mainly attributed to GNPs incompletely covered with BSA in this case. These preliminary results further documente that our approaches for modification and conjugation of GNPs were efficient and suggest that the functionalized GNPs may be potentially useful as labeling probes in cancer diagnostics and studies.

possibilities for using PEG-SH-modified GNPs as stable probes in a variety of bioapplications. 3.3. Modification of GNPs with a Mixture of MPA and PEG-SH and Conjugation of Antibody. The results above revealed that GNPs modified with PEG-SH were very stable in aqueous solution. However, PEG-SH used in this study has no functional group to conjugate biomolecules. To resolve this problem, we prepared a mixture solution of PEG-SH and MPA (a ratio of 1:10) to modify GNPs in our subsequent research. In the same way, we also investigated the effects of buffer pH and salt concentrations on the stability of GNPs in solution. The partial results are shown in Figure 4. From the absorption spectra, we observed that GNPs modified with PEG-SH and MPA were very stable too. The strategy of bioconjugation of GNPs to antibody is shown in Scheme 1. In this approach, EDC and NHS were used as linkage reagents, which were widely used in biolabeling fields. Figure 5 implies that the absorption spectra of GNPs have no significant change before and after the conjugation of antibody. This result indicated that no obvious aggregation of GNPs was generated in the bioconjugation of GNPs. 3.4. Cellular Imaging. Finally, we explore the possibility of cell imaging using GNP−antibody conjugates as labeling probes. In the experiments, we first conjugated GNPs to antiEGFR antibodies according to the procedure described above and then targeted the resulting conjugates to cells for specific dark field imaging. Herein, SiHa cells were used as a model in cell imaging. SiHa cells are derived from a human epithelial carcinoma, and it has been reported that they exhibit significantly elevated EGFR expression. The high level of EGFR expression is often associated with enhanced aggressiveness of epithelial cancers and poor prognosis. Parts a and b of Figure 6 show the images of SiHa cancer cells with and without GNP−anti-EGFR conjugates, respectively. From Figure 6a, we can observe the distribution of many GNPs on the cell surface due to the specific binding of overexpressed EGFR on the cancer cells with the GNP−anti-EGFR antibodies, which is

■

CONCLUSION In this study, we systematically investigated the effects of thiol compounds on the colloidal stability of GNPs in various solutions with different pH values and salt concentrations. Our data showed that GNPs modified with PEG-SH were most stable in solution compared to glutathione, MPA, cysteine, cystamine, and DHLA. We developed a simple approach for modification of GNPs using a mixture of PEG-SH and MPA (or GSH) as ligands. Our results illustrated that GNPs 4470

dx.doi.org/10.1021/la204289k | Langmuir 2012, 28, 4464−4471

Langmuir

Article

(17) Shi, X. Y.; Wang, S. H.; Meshinchi, S.; Antwerp, M. E. V.; Bi, X. D.; Lee, I.; Baker, J. R. Small 2007, 3, 1245−1252. (18) Curry, A. C.; Crow, M.; Wax, A. J Biomed. Opt. 2008, 13, 1−7. (19) Lan, T.; Dong, C. Q.; Huang, X. Y.; Ren, J. C. Analyst 2011, 136, 4247−4253. (20) Fan, A. P.; Lau, C. W.; Lu, J. Z. Anal. Chem. 2005, 77, 3238− 3242. (21) He, H.; Ren., J. C. Talanta 2008, 77, 166−171. (22) Zhou, R.; Xiong, B.; He, Y.; Yeung, E. S. Anal. Bioanal. Chem. 2011, 399, 353−359. (23) Zhao, Y. Y.; Wang, Z.; Zhang, W.; Jiang, X. Y. Nanoscale 2010, 2, 2114−2119. (24) Sharma, D. Colloids Surf., B 2011, 85, 330−337. (25) Tarantola, M.; Pietuch, A.; Scheneider, D.; Rother, J.; Sunnick, E.; Rosman, C.; Pierrat, S.; Sönnichsen, C.; Wegener, J.; Janshoer, A. Nanotoxicology 2011, 5, 254−268. (26) Ortiz, M.; Fragoso, A.; O’Sullivan, C. K. Anal. Chem. 2011, 83, 2931−2938. (27) Joseph, D.; Geckeler, K. E. Langmuir 2009, 25, 13224−13231. (28) Tsai, D. H.; DelRio, F. W.; MacCuspie, R. I.; Cho, T. J.; Zachariah, M. R.; Hachley, V. A. Langmuir 2010, 26, 10325−10333. (29) Lee, H.; Choi, S. H.; Park, T. G. Macromolecules 2006, 39, 23− 25. (30) Ruan, C. M.; Wang, W.; Gu, B. H. Anal. Chim. Acta 2006, 567, 114−120. (31) Zhang, S. X.; Wang, N.; Yu, H. J.; Niu, Y. M.; Sun, C. Q. Bioeletrochemistry 2005, 67, 15−22. (32) Roux, S.; Garcia, B.; Bridot, J. L.; Salomĕ, M.; Marquette, C.; Lemelle, L.; Gillet, P.; Blum, L.; Perriat, P.; Tillement, O. Langmuir 2005, 21, 2526−2536. (33) Mei., B. C.; Susumu, K.; Medintz, I. L.; Delehanty, J. B.; Mountziaris, T. J.; Mattoussi, H. J. Mater. Chem. 2008, 18, 4949−4958. (34) Chai, F.; Wang, C. G.; Wang, T. T.; Li, L.; Su, Z. M. ACS Appl. Mater. Interfaces 2010, 2, 1466−1470. (35) Ghosh, P.; Han, G.; De, M.; Kim, C. K.; Rotello, V. M. Adv. Drug Delivery Rev. 2008, 60, 1307−1315. (36) Kaufman, E. D.; Belyea, J.; Johnson, M. C.; Nicholson, Z. M.; Richs, J. L.; Shah, P. K.; Bayless, M.; Pettersson, T.; Feldotö, Z.; Blomberg, E.; Claesson, P.; Franzen, S. Langmuir 2007, 23, 6053− 6062. (37) Hu, F. Q.; Ran, Y. L.; Zhou, Z. A.; Gao, M. Y. Nanotechnology 2006, 17, 2972−2977. (38) Huang, H. W.; Liu, X. Y.; Hu, T.; Chu, P. K. Biosens. Bioelectron. 2010, 25, 2078−2083. (39) Stewart, M. H.; Susumu, K.; Mei, B. C.; Medintz, I. L.; Delehanty, J. B.; Blanco-Canosa, J. B.; Dawson, P. E.; Mattoussi, H. J. Am. Chem. Soc. 2010, 132, 9804−9813. (40) Tang, L. C.; Dong, C. Q.; Ren, J. C. Talanta 2010, 81, 1560− 1567. (41) Huang, T.; Nallathamby, P. D.; Gillet, D.; Xu, X. H. N. Anal. Chem. 2007, 79, 7708−7718. (42) Pfeifer, D.; Stellwag, B.; Pferfer, A.; Borlinghaus, P.; Meier, W. Gynecol. Oncol. 1989, 33, 146−150. (43) El-Sayed, I. H.; Huang, X. H.; El-Sayed, M. A. Nano Lett. 2005, 5, 829−834. (44) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin, C. A. Science 1997, 277, 1078−1081. (45) R. A. Reynolds, R. A.; Mirkin, C. A.; Letsinger, R. L. J. Am. Chem. Soc. 2000, 122, 3795−3796. (46) Du, B. A.; Li, Z. P.; Liu, C. H. Angew. Chem., Int. Ed. 2006, 45, 8022−8025.

modified with PEG-SH and MPA were very stable in buffers with different pH values and high salt concentrations as well. Since the surface of modified GNPs contained certain carboxyl functional groups, GNPs were facilely linked to biomolecules (such as antibodies). On the basis of this approach, we conjugated GNPs to EGFR antibodies and successfully applied the resulting bioconjugates to in vivo targeted cancer imaging. Our method described for modification and conjugation of GNPs is simple and efficient, and this opens up possibilities for using PEG-SH-modified GNPs as stable probes in a variety of bioapplications.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: 0086-21-54746001. Fax: 0086-21-54741297. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was financially supported by the National Natural Science Foundation of China (NSFC; Grants 21075081, 21135004, and 20705019), National Basic Research Program of China (Grant 2009CB930400), and Nano-Science Foundation of Shanghai (Grant 1052 nm04000). We thank Dr. Tao Lan and Dr. Bocheng Zhang for measuring the hydrodynamic diameters of modified GNPs by resonance light scattering correlation spectroscopy.

■

REFERENCES

(1) Huang, H. C.; Barua, S.; Sharma, G.; Dey, S. K.; Rege, K. J. Controlled Release 2011, 06, 004. (2) Smith, A. M.; Duan, H. W.; Mohs, A. M.; Nie, S. M. Adv. Drug Delivery Rev. 2008, 60, 1226−1240. (3) Chan, W. C.; Maxwell, D. J.; Gao, X. H.; Baily, R. E.; Han, M. Y.; Nie, S. M. Curr. Opin. Biotechnol. 2002, 13, 40−46. (4) Liu, Z.; Fan, A. C.; Rakhra, K.; Sherlock, S.; Goodwin, A.; Chen, X. Y.; Yang, Q. W.; Felsher, D. W.; Dai, H. J. Angew. Chem., Int. Ed. 2009, 48, 7668−7672. (5) Poland, C. A.; Duffin, R.; Kinloch, I.; Maynard, A.; Wallae, W. A. H.; Seaton, A.; Stone, V.; Brown, S.; Macnee, W.; Donaldson, K. Nat. Nanotechnol. 2008, 3, 423−428. (6) Weissleder, R.; Stark, D. D.; Engelstad, B. L.; Bacon, B. R.; Compton, C. C.; White, D. L.; Jacobs, P.; Lewis, J. Am. J. Roentgenol. 1989, 152, 167−173. (7) Johannsen, M.; Gneveckow, U.; Taymoorian, K.; Thiesen, B.; Waldofner, N.; Scholz, R.; Jung, K.; Jordan, A.; Wust, P.; Loening, S. A. Int. J. Hyperthermia 2007, 23, 315−323. (8) Bulte, J. W. M.; Kraitchman, D. L. NMR Biomed. 2004, 17, 484− 499. (9) He, H.; Xie, C.; Ren, J. C. Anal. Chem. 2008, 80, 5951−5957. (10) Shukla, R.; Bansal, V.; Chaudhary, M.; Basu, A.; Bhonde, R. R.; Sastry, M. Langmuir 2005, 21, 10644−10654. (11) Link, S.; El-Sayed, M. A. J. Phys. Chem. B 1999, 103, 8410− 8426. (12) Derfus, A. M.; Chan, W. C. W.; Bhatia, S. N. Nano Lett. 2004, 4, 11−18. (13) Wang, X. Y.; Kim, Y.; Drew, C.; Ku, B.; Kumar, J.; Samuelson, L. A. Nano Lett. 2004, 4, 331−334. (14) Storhoff, J. J.; Elghanian, R.; Mucic, R. C.; Mirkin, C. A.; Letsinger, R. L. J. Am. Chem. Soc. 1998, 120, 1959−1964. (15) Han, M. S.; Lytton-Jean, A. K. R.; Oh, B. K.; Heo, J.; Mirkin, C. A. Angew. Chem., Int. Ed. 2006, 45, 1807−1810. (16) Medley, C. D.; Smith, J. E.; Tang, Z. W.; Wu, Y. R.; Bamrungsap, S.; Tan, W. H. Anal. Chem. 2008, 80, 1067−1072. 4471

dx.doi.org/10.1021/la204289k | Langmuir 2012, 28, 4464−4471