pubs.acs.org/Langmuir © 2009 American Chemical Society
Lysine-Functionalized Silver Nanoparticles for Visual Detection and Separation of Histidine and Histidine-Tagged Proteins Doo Ri Bae,† Won Seok Han,† Jung Mi Lim,‡ Sunwoo Kang,§ Jin Yong Lee,*,§ Dongmin Kang,*,‡ and Jong Hwa Jung*,† † Department of Chemistry and Research Institute of Natural Sciences and Environmental Biotechnology National Core Research Center, Gyeongsang National University, Jinju 660-701, Korea, ‡Division of Life and Pharmaceutical Sciences, Ewha Womans University, Seoul 120-750, Korea, and §Department of Chemistry and Institute of Basic Science, Sungkyunkwan University, Suwon 440-746, Korea
Received July 22, 2009. Revised Manuscript Received August 9, 2009 A new chromogenic chemosensor based on lysine-functionalized silver nanoparticles 1 was prepared and characterized by transmission electron microscopy (TEM), Fourier transform Raman, and ultraviolet-visible (UV-vis) spectroscopy. The color changes of nanoparticles 1 in the absence and the presence of metal ion were observed upon addition of various amino acids and proteins in aqueous solution. Among the various amino acids, the sensor 1 in the absence of metal ion shows a novel colorimetric sensor with capability to probe histidine and histidine-tagged proteins. On the other hand, the color changes of 1 in the presence of metal ions such as KCl or NiCl2 did not occur with any amino acids. Therefore, the sensor 1 in the absence of metal ion responds selectively to histidine, a response which can be attributed to its aggregation induced by histidine with high numbers of electrostatic interactions. This highly selective sensor 1 allows a rapid quantitative assay of histidine to concentrations as low as 5.0 μM, providing a new tool for the direct measurement of histidine and histidine-tagged proteins in vitro system. Furthermore, we examined the effect of pH on absorbance (A520) of 1 in the presence of histidine (pH 4-12). The absorbance under basic conditions was higher than that under acidic or neutral conditions, in accord with the stronger aggregation of 1 with histidine by electrostatic interaction between the carboxylate anion of 1 and ammonium protons of histidine under basic conditions.
*To whom correspondence should be addressed. E-mail: jinylee@ skku.edu (J.Y.L.);
[email protected] (D.K.);
[email protected] (J.H.J.).
events.5 Willner and co-workers reported that a cascade of DNAzymes enabled the colorimetric or chemiluminescent detection of Pb2þ ion or histidine.5 As biological molecules, histidine and histidine-tagged proteins play a critical biological role in the human body. An abnormal level of histidine-tagged protein is an indicator for many diseases, such as advanced liver cirrhosis,6 AIDS,7 renal disease,7 asthma,8 pulmonary disorders,9 thrombotic disorders,8,10 and malaria.11 Some analytical methods for the detection of histidine and histidine-tagged proteins have been developed in conjunction with immunoassays, fluorimetric, and colorimetric detection.12 Most of these methods require complicated and expensive instrumentation, involve cumbersome laboratory procedures, and suffer from low throughput, which limit the scope of their practical application. However, among these methods, colorimetric sensors are especially promising due to their simple nakedeye applications with less labor and less expensive equipment than other closely related methods such as fluorimetric sensors. Although numerous studies have dealt with the detection of histidine or histidine-tagged proteins, studies of the use of colorimetric or fluorimetric probes in conjunction with functionalized
(1) (a) Lioubashevski, O.; Chegel, V. I.; Patolsky, F.; Katz, E.; Willner, I. J. Am. Chem. Soc. 2004, 126, 7133. (b) Wessels, J. M.; Nothofer, H. G.; Ford, W. E.; Wrochem, F. V.; Scholz, F.; Vossmeyer, T.; Schroedter, A.; Weller, H.; Yasuda, A. J. Am. Chem. Soc. 2004, 126, 3349. (c) Elghanian, R.; Storhoff, J. J.; Mucic, G. C.; Letsinger, R. L.; Mirkin, C. A. Science 1997, 277, 1078. (d) Gao, J.; Fu, J.; Lin, C.; Lin, J.; Han, Y.; Yu, X.; Pans, C. Langmuir 2004, 20, 9775. (2) (a) Stoeva, S. I.; Lee, J.-S.; Smith, J. E.; Rosen, S. T.; Mirkin, C. A. J. Am. Chem. Soc. 2006, 128, 8378. (b) Darbha, G. K.; Singh, A. K.; Rai, U. S.; Yu, E.; Yu, H.; Ray, P. C. J. Am. Chem. Soc. 2008, 130, 8038. (c) Durocher, S.; Rezaee, A.; Hamm, C.; Rangan, C.; Mittler, S.; Mutus, B. J. Am. Chem. Soc. 2009, 131, 2475. (3) (a) Dai, Q.; Liu, X.; Coutts, J.; Austin, L.; Huo, Q. J. Am. Chem. Soc. 2008, 130, 8138. (b) Lee, J.-S.; Lytton-Jean, A. K. R.; Hurst, S. J.; Mirkin, C. A. Nano Lett. 2007, 7, 2112. (4) Lee, J.-S.; Ulmann, P. A.; Han, M. S.; Mirkin, C. A. Nano Lett. 2008, 8, 529. (5) Elbaz, J.; Shlyahovsky, B.; Willner, I. Chem. Commun. 2008, 1569.
(6) (a) Saito, H.; Goodnough, L. T.; Boyle, J. M.; Heimburger, N. Am. J. Med. 1982, 73, 179. (b) Leebeek, F. W.; Kluft, C.; Knot, E. A.; De Matt, M. P.; Lab, J. Clin. Med. 1989, 113, 493. (7) Jones, A. L.; Hulett, M. D.; Parish, C. R. Immunol. Cell Biol. 2005, 83, 106. (8) Engesser, L.; Kluft, C.; Briet, E.; Brommer, Br. E. J. J. Haematol. 1987, 67, 355. (9) Morgan, W. T. Biochem. Med. Metab. Biol. 1986, 36, 210. (10) Kuhli, C.; Scharrer, I.; Koch, F.; Hattenbach, L. O. Am. J. Ophthalmol. 2003, 135, 232. (11) Sullivan, D. J., Jr.; Gluzman, I. Y.; Goldberg, D. E. Science 1996, 271, 219. (12) (a) Jelinek, T.; Grobusch, M. P.; Harms, G. J. Infect. Dis. 2001, 33, 752. (b) Sahal, D.; Kannan, R.; Sinha, A.; Babbarwal, V.; Prakash, B. G.; Singh, G.; Chauhan, V. S. Anal. Biochem. 2002, 308, 405.
Introduction Nanoparticles modified with noble metals, especially silver and gold, have attracted much attention due to their optical, electrical, and chemical properties.1 In particular, ligand-modified metal nanoparticles have been explored as analytical tools in many biological fields.2 The introduction of organic ligands onto the surfaces of nanoparticles not only stabilizes these nanostructures in different solvents but also provides desirable surface functionalities. Noble metal nanoparticles modified with DNA have been the subject of recent research.3 Mirkin and co-workers reported a DNA-gold nanoparticles-based colorimetric assay, in which the DNA had to be specifically modified for the detection of cysteine.4 Furthermore, DNA-modified gold nanoparticles were only detectable with a color change at temperatures above 50 °C. Recently, the horseradish peroxidase-mimicking DNAzymes have been conjugated to aptamers, and the formation of the aptamer-analyte complex released the DNAzyme, which acted as a biocatalyst for the amplified readout of the recognition
Langmuir 2010, 26(3), 2181–2185
Published on Web 08/28/2009
DOI: 10.1021/la9026865
2181
Article
nanoparticles for this purpose remain sparse. In this article, we show that N,N-bis(carboxymethyl)-L-lysine-modified silver nanoparticles 1 act as a selective colorimetric probe for histidine and histidine-tagged proteins. In comparison with ligand systems employed in the previous studies, N,N-bis(carboxymethyl)-Llysine (BCML) is a simple, small, and commercially available functional ligand. Herein, we report the preparation of 1 and its application as a novel colorimetric sensor for histidine and histidine-tagged proteins.
Experimental Section Synthesis of N,N-Bis(carboxymethyl)-L-lysine-Modified Silver Nanoparticle (1). Silver nanoparticles with average diameter of 3-5 nm were prepared by the citrate-mediated reduction of AgNO3. An aqueous solution of AgNO3 (5.0 mM, 1.0 mL) was stirred at room temperature, and then sodium citrate solution (30 mM, 1.0 mL), sodium borohydride (50 mM, 1.0 mL), and polyvinylpyrrolidone (5 mg) were added quickly, resulting in a change in solution color from pale yellow to deep red. After the color change, the solution was heated under reflux for an additional 30 min and allowed to cool to room temperature. Ligand exchange reactions were performed under stirring at room temperature for 24 h by mixing a given volume of as-prepared silver colloids with water solution containing an excess of N, N-bis(carboxymethyl)-L-lysine. The solutions of functionalized silver nanoparticles were centrifuged for 30 min and redispersed in aqueous solution after the supernatant was removed. The particles were washed three more times and finally redispersed in the detection buffer. The solution was then adjusted to pH 9 by NaOH or 2-(cyclohexylamino)ethanesulfonic acid (CHES) buffer. Colorimetric Detection of Amino Acids and Proteins. For the colorimetric detection of amino acids and proteins, the histidine stock solution in the detection buffer was mixed with the probe solution prepared as described above at room temperature to a final volume of 1 mL. (The final concentration of silver nanoparticles probes was 10.0 μM.) The color changes of 1 in the absence and the presence of KCl and NiCl2 with 5.0 μM of amino acids or proteins were observed. To evaluate the sensitivity of the assay, the final concentrations of histidine were varied from 5.0 to 30.0 μM. The selectivity for histidine was confirmed by the addition of other amino acid stock solutions to 1, using the same conditions, with a final amino acid concentration of 50.0 μM. In addition, to the color change of 1, each 1.5 μg of proteins was added to 1 (5.0 mM) solution. Once again, the color changes of 1 with amino acids and proteins were carried out at the same condition including the pH. SDS-PAGE. SDS-PAGE analysis was conducted according to Bio-Rad mini-PROTEIN 3 Cell Instruction Manual using a 4% stacking gel and 16% resolving gel. Aliquots (1.5 μg) of each protein (ubiquitin from Sigma, histidine-Cdc20 purified using Sf9 cell overexpression system) were incubated with 1 (10.0 μM, 1.0 mL) for 2 h at room temperature with rotary mixing. 1 was removed by centrifugation for 20 min at 4000 rpm. The supernatant and premixed proteins were resolved on the SDSPAGE gel and visualized with Coomassie Brilliant Blue (CBB) staining. Instruments. 1H and 13C NMR spectra were measured with a Bruker 300 apparatus. IR spectra were obtained for KBr pellets, in the range 400-4000 cm-1, with a Shimadzu FT-IR 8400S instrument, and the MS spectrum was obtained with a JEOL JMS-700 mass spectrometer. All UV-vis absorption spectra were recorded in RF-5301PC spectrophotometer. Time-of-flight second ion mass spectrometer (TOF-SIMS) was analyzed on a Model PHI 7200 equipped with Cs and Ga ion guns for positive and negative ion mass detection. FT-Raman spectra were measured by a Model LabRAM HR800. 2182 DOI: 10.1021/la9026865
Bae et al. Scheme 1. Schematic Illustration of the Fabrication of 1 and Its Aggregation in the Presence of Histidine in Aqueous Solution
Scheme 2. Chemical Structure of BCML (N,N-Bis(carboxymethyl)L-lysine)
Results and Discussion Sensor 1 was prepared as shown in Scheme 1. Silver nanoparticles were prepared as described previously.12 The surface of the silver nanoparticles was functionalized with the BCML at 25 °C for 2 h. The well-structured carboxyl groups of BCML perform steric effect to shield these silver nanoparticles from agglomeration.13 The synthetic compound 1 was well characterized by transmission electron microscopy (TEM), FT-Raman spectroscopy, time-of-flight second ion mass spectroscopy (TOF-SIMS), and UV-vis spectroscopy. To verify the immobilization of BCML (Scheme 2) onto the silver nanoparticles, we took FT-Raman and TOF-SIMS of 1. Peaks at 3263, 1684, and 1284 cm-1 were present in Raman spectrum of 1 (Figure S1). These originated from the BCML, indicating that BCML was located on the silver nanoparticles. The TOF-SIMS spectrum of 1 displayed the molecular mass for BCML (m/z = 262.26), confirming that BCML was anchored onto the surface of the silver nanoparticles (Figure S2). We also observed the UV-vis spectrum of free 1 (Figure 1). An absorption peak was observed at 400 nm that originates from the surface plasmon absorption of 1 without amino acids. The spectrum recorded after 5 min displayed the narrowest full width at half-maximum (fwhn), indicating that the synthesized silver nanoparticles were monodispersed and uniform. A variety of amino acids were added to the solution of 1 in order to investigate the molecular recognition ability of 1. Figure 1 illustrates the color changes and R values (A520/A400) of 1 after the addition of solutions (1.0 10-4 M) of various amino acids. Over 10 min, the solution containing histidine changed from yellow to red, with a dramatic increase in the absorbance ratio A520/A400. The addition of other amino acids, in contrast, had no effect on the color or absorption spectrum of 1. Thus, the sensor 1 responds selectively to histidine, a response which can be attributed to its aggregation induced by histidine. To determine whether the specificity in the detection of histidine is compromised by complex mixtures of other amino acids, we observed the color changes of 1 with histidine in binary systems (Figure S3). The sensor 1 was added as a stable dispersion in mixtures of these amino acids. However, the color of 1 changed from yellow to red only when a solution of histidine was added to the mixture. Thus, we believe that the highly selective nature of the color change of 1 is due to the high selectivity of the assay for histidine. This result indicates that the detection of histidine can (13) Aubel, F. M., Brent, R., Kingson, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., Struhl, K., Albright, L. M., Coen, D. M., Varki, A., Eds. Current Protocols in Molecular Biology; John Wiley & Sons: New York, 2003.
Langmuir 2010, 26(3), 2181–2185
Bae et al.
Article
Figure 1. (a) UV-vis spectra and (b) R (A520/A400) of 1 (10.0 μM) in the presence of amino acids (1.0 10-4 M) at pH 9. (c) Photographic images of 1 containing different amino acids.
Figure 2. TEM images of 1 (10.0 μM) (a) before and (b) after addition of histidine (1.0 10-4 M). The scale bars are both 10 nm. (c) Size distribution of 1 (a) before and (b) after addition of histidine in aqueous solution.
be performed in solution even when a mixture of other amino acids is present. To investigate the color change of 1 with histidine, we observed the TEM image of 1 before and after addition of histidine (Figure 2A,B). The TEM image shows that prior to the addition of histidine the particles of 1 are highly dispersed and uniform in aqueous solution with diameters of 3-5 nm. After the addition of histidine, the particles of 1 increased in size and became highly aggregated into aggregates of 16-20 nm of diameter (Figure 2B,C). This aggregation resulted in the yellow-to-red color change reflecting to the interparticle coupled plasmon excitons in the aggregated states. To evaluate the minimum concentration of histidine aqueous solution detectable by the color change, we added histidine into Langmuir 2010, 26(3), 2181–2185
the mixture of 1 over the concentration range 5.0-30.0 μM. Histidine was detectable by eye with 1 at levels as low as 5.0 μM (Figure S4). In addition, a linear response between R values (A520/ A400) and concentration of 1 was observed between 5.0 and 30.0 μM with a detection limit of ca. 5.0 μM. We observed the color changes of 1 upon addition of amino acids in the presence of KCl (200 mM with respect to BCMLattached silver nanoparticles). As shown in Figure S5, when KCl was present, the color of 1 did not change upon addition of various amino acids. In addition, it is well-known that BCML forms a 1:1 complex with Ni2þ ion.14 Therefore, we investigated the effect of the presence of Ni2þ (1.0 equiv with respect to BCML-attached silver nanoparticles) on color changes of 1 when amino acids were added. No significant color changes were induced by any amino acids if Ni2þ was present (Figure 3). These findings indicate that the 1-Ni2þ complex could not selectively recognize histidine. Apparently, 1 forms coordination bonds with the amino acid in the presence of the metal ion, and thus, additional electrostatic interactions cannot occur. Therefore, the numbers of electrostatic interactions between 1 and amino acids are the important factors to effect selective color change. To observe color change of 1 for histidine in the presence of interference compounds, we observed the color changes of 1 with amino acids in the presence of imidazole (Figure S6). The yellow color of 1 was changed into red color upon the addition of histidine. The addition of other amino acids, in contrast, had no effect on the color or absorption spectrum of 1. The results support the view that 1 selectively recognize only histidine in the presence of other compounds. Furthermore, the color change of 1 was observed upon the addition of nucleic acids such as thymine, cytosine, adenosine, and guanine (Figure S7). As expected, the addition of nucleic acids had no effect on the color or absorption spectrum of 1. We examined the effect of pH on absorbance (A520) of 1 in the presence of histidine (pH 4-12). As shown in Figure S8, the absorbance under basic conditions was higher than that under (14) (a) Graff, R. A.; Swanson, T. M.; Strano, M. Chem. Mater. 2008, 20, 1824. (b) Gu, H.; Xu, K.; Xu, B. Chem. Commun. 2006, 941.
DOI: 10.1021/la9026865
2183
Article
Bae et al.
Figure 4. Calculated structure of the L-lysine-histidine complex in aqueous solution.
Figure 3. (a) UV-vis spectra and (b) photographic image of 1
solution (10.0 μM) by addition of amino acids (1.0 10-4 M) in the presence of NiCl2 (10.0 μM) at pH 9.
acidic or neutral conditions in accord with the stronger aggregation of 1 with histidine by electrostatic interaction between the carboxylate anion of 1 and ammonium protons of histidine under basic conditions (Figure S9). In addition, we obtained the same color changes of 1 with histidine at pH=9 by using CHES buffer solution (Figure S10). To obtain direct evidence for electrostatic interaction between the carboxylate anion of 1 and ammonium protons of histidine, we obtained FT-IR spectra of 1 with and without histidine. As shown in Figure S11, the vibrational band for the COO- species of 1 in the absence of histidine showed strong bands at 1398 and 1386 cm-1. In contrast, in the presence of histidine, the vibrational band for the COO- species of 1 showed a strong band at 1404 cm-1. These results are strong evidence for the electrostatic interaction between the ammonium ion of histidine and COOspecies of 1. To shed light on the binding mode of 1 with histidine, we carried out density functional theory (DFT) calculations for the complex between the histidine and the BCML with 6-31G* basis sets using a suite of Gaussian 03 programs.15 In aqueous solution, the histidine and the BCML exist as þ3 and -3 charged species, respectively. The interaction between highly charged species in the solution phase is considerably reduced compared with that in the gas phase due to the dielectric screening. Thus, the solvent effect should be considered for the information on structures as well as the interactions. In this study, the polarizable continuum model (PCM) method is employed for the solvent effect. In the gas phase, two hydrogen bondings and a C-H 3 3 3 O interactions are working between the BCML and the histidine, where three carboxylates of the BCML are involved as noted from the N 3 3 3 O distances (Figure S12). However, the structure in aqueous solution is quite different from that in the gas phase. In aqueous solution, there are two strong charged hydrogen bondings between -NH3þ of the histidine and -COO- of the BCML and between a -NH2þ of the histidine and -COO- of the BCML. In particular, the latter one seems to be electrostatic (15) (a) Kim, K. S.; Oh, K. S.; Lee, J. Y. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 6373. (b) Kim, K. S.; Kim, D.; Lee, J. Y.; Tarakeshwar, P.; Oh, K. S. Biochemistry 2002, 41, 5300. (c) Perrin, C. L.; Nielson, J. B. Annu. Rev. Phys. Chem. 1997, 48, 511. (d) Pacios, L. F.; Gomez, P. C.; Galvez, O. J. Comput. Chem. 2006, 27, 1650.
2184 DOI: 10.1021/la9026865
interactions between the positive and negative charges due to the inefficient arrangement for H-bonding (Figure 4). Depending on the environment, these two protons may exist in a single potential well between the histidine nitrogen atom and the BCML oxygen atom as is in the low barrier hydrogen bond (LBHB) or short strong hydrogen bond (SSHB). The SSHB is often found in a charged system with very strong binding.15 The interaction energy between the histidine and the BCML in aqueous solution was calculated to be 138.07 kcal/mol. This strong binding originates from not only simple electrostatic interactions but also two SSHB (or LBHB) between them. We also investigated the selective color changes of 1 with ubiquitin (Ub) protein lacking a histidine tag at the N-terminus and Cdc20 protein having six histidine residues tagging the N-terminus. In a typical experiment, the sensor 1 (5.0 mM) was incubated with protein (1.5 μg) for 10 min at room temperature, and then the color changes of 1 were observed by UV-vis spectroscopy (Figure 5A). When Ub (non-histidine-tagged protein) was treated with 1, no significant color change was observed (Figures 5A:b and 5B:b), indicating that the nonhistidine-tagged protein did not bind to 1. However, upon addition of Cdc20 protein to the solution of 1, the color changed from yellow to red color with an increase in the absorbance ratio A520/A400 (Figures 5A:c and 5B:c), indicating that the histidinetagged protein had interacted with 1. When the sensor 1 was incubated with a mixture of both the histidine-tagged protein and the non-histidine-tagged protein, the yellow color of 1 again changed to red, suggesting that the histidine-tagged protein selectively binds to the surface of 1. In addition, the incubation of the histidine-tagged protein-captured silver nanoparticles with concentrated HCl solution caused the proteins to be released from 1, resulting in a 95% recovery of the UV-vis spectrum intensity (Figure S13), a 98% separation of the histidine-tagged protein, and a reversal of the color change from red to yellow in the solution of 1. We carried out sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) experiments to determine whether the sensor 1 can discriminate between proteins with and without histidine residues (Figure 6). To minimize the effects of other amino acid sequences, the isoelectric point (pI), the protein structure, and the presence of particular side chains, ubiquitin (Ub) protein without a histidine tag at the N-terminus and Cdc20 protein modified with six histidine residues tagging the N-terminus were also used in the present study. Figure 6a shows the image of a gel containing the two proteins markers before the addition of 1. After addition and separation of 1 (Figure 6b), the nonhistidine-tagged Ub proteins were still visible on the gel. On the other hand, the histidine-tagged Cdc20 protein disappeared, since the formation of a complex between 1 and Cdc20 produced a Langmuir 2010, 26(3), 2181–2185
Bae et al.
Article
Figure 5. (A) UV-vis spectra and (B) photographic images of 1 (5.0 mM) (a) before and after addition of (b) Ub (1.5 μg) and (c) Cdc20 (1.5 μg) proteins at pH 9.
histidine-tagged protein from a mixture of proteins in vitro system.
Conclusions We have demonstrated the synthesis of 1 and its successful application for the selective visual detection of histidine and histidine-tagged proteins. We believe that our current approach will provide a more convenient method for selective detection of histidine and histidine-tagged proteins, when compared with the current methods, which are multistep and require expensive instrumentation. We also believe that our approach can be employed for anchoring other biomolecules on nanoparticles, potentially producing a range of novel biomedical nanomaterials. Figure 6. (a) Protein stained gel with coomassie brilliant blue (CBB) before addition of 1. Each 1.5 μg of proteins was loaded. (b) The supernatant was loaded after the removal of 1 in the presence of Ub (1.5 μg) and Cdc20 (1.5 μg) proteins by centrifugation. (c) Immunoblots with antihistidine antibody to check the presence of histidine tagging in proteins.
structure too large to enter the matrix of the gel. The presence of the six histidine tagging was confirmed by immunoblots with antihistidine antibody (Figure 6c). The results indicate that histidine-tagged proteins are effectively bound onto the surface of 1. Thus, the sensor 1 can be useful for the separation of a
Langmuir 2010, 26(3), 2181–2185
Acknowledgment. This work was supported by Korea Science and Engineering Foundation Grant (R01-2007-000-20299-0) and the EB-NCRC (Grant #: R15-2003-012-01001-0), Korea Research Foundation Grant (KRF-2008-C00447), and World Class University Project supported from Ministry of Education, Science and Technology, S. Korea. Supporting Information Available: Synthesis, experimental details, and additional spectroscopic data. This material is available free of charge via the Internet at http://pubs.acs.org.
DOI: 10.1021/la9026865
2185
