Interparticle Chiral Recognition of Enantiomers - American Chemical

Dec 15, 2008 - Nanoparticle-Based Regulation Strategy ... herein a nanoparticle-based strategy to regulate interparticle ... Southern Illinois Univers...
1 downloads 0 Views 497KB Size
Anal. Chem. 2009, 81, 689–698

Interparticle Chiral Recognition of Enantiomers: A Nanoparticle-Based Regulation Strategy I-Im S. Lim,† Derrick Mott,† Mark H. Engelhard,‡ Yi Pan,§ Shalini Kamodia,† Jin Luo,† Peter N. Njoki,† Shuiqin Zhou,§ Lichang Wang,| and Chuan Jian Zhong*,† Department of Chemistry, State University of New York at Binghamton, Binghamton, New York 13902, EMSL, Pacific Northwest National Laboratory, Richland, Washington 99352, Department of Chemistry, City University of New York, Staten Island, New York 10314, and Department of Chemistry & Biochemistry, Southern Illinois University, Carbondale, Illinois 62901 The ability to regulate how molecular chirality of enantiomeric amino acids operates in biological systems constitutes the basis of drug design for specific targeting. We report herein a nanoparticle-based strategy to regulate interparticle chiral recognition of enantiomers using enantiomeric cysteines (L and D) and gold nanoparticles as a model system. A key element of this strategy is the creation of a nanoscale environment either favoring or not favoring the preferential configuration of the pairwise zwitterionic dimerization of the enantiomeric cysteines adsorbed on gold nanoparticles as a footprint for interparticle chiral recognition. This recognition leads to interparticle assembly of the nanoparticles which is determined by the change in the nanoparticle surface plasmonic resonance. While the surface density and functionality of cysteines on gold nanoparticles are independent of chirality, the interparticle chiral recognition is evidenced by the sharp contrast between the interparticle homochiral and heterochiral assembly rates based on a firstorder kinetic model. The structural properties for the homochiral and heterochiral assemblies of nanoparticles depend on the particle size, the cysteine chirality, and other interparticle binding conditions. The structural and thermodynamic differences between the homochiral and heterochiral interactions for the interparticle assemblies of nanoparticles were not only substantiated by spectroscopic characterizations of the adsorbed cysteine species but also supported by structures and enthalpies obtained from preliminary density functional theory calculations. The experimental-theoretical correlation between the interparticle reactivity and the enantiomeric ratio reveals that the chiral recognition is tunable by the nanoscale environment, which is a key feature of the nanoparticle-regulation strategy for the interparticle chiral recognition. While the importance of enantiomeric chirality in protein interactions is widely recognized in the broad interdisciplinary * To whom correspondence should be addressed. E-mail: cjzhong@ binghamton.edu. † State University of New York at Binghamton. ‡ Pacific Northwest National Laboratory. § City University of New York, Staten Island. | Southern Illinois University. 10.1021/ac802119p CCC: $40.75  2009 American Chemical Society Published on Web 12/15/2008

community of pharmaceutical science and technology,1 the ability to template chiral interactions2 at the molecular level has been a difficult challenge in specific drug targeting. The exploitation of the nanoscale size effect and interparticle specificity of nanoparticles represents an advanced solution to address the challenge. In contrast to the nanoparticle-assisted complementary binding of DNAs for diagnostics and detection,3,4 our nanoparticleregulated chiral recognition strategy stems from the creation of a nanoscale templating environment for pairwise zwitterionic interactions of R-amino acids in aqueous solution. This strategy could potentially lead to the development of a highly effective route for controlling the enantiomeric specificity.5,6 For example, while L-cysteine plays an important role in living systems and its deficiency is associated with a number of clinical situations (liver damage, skin lesions, AIDS, and certain neurodegenerative diseases), the role of L-cysteine in the central nervous system is not well understood.5 On the other hand, D-cysteine is believed to interfere with many targets inside the cell, but the corresponding sites of action of D-cysteine are unknown, and little information is available on the occurrence and roles of D-amino acids.6 The nanoparticle-regulated chiral recognition strategy differs from those involving the association of chirality with single-crystal surface properties,7 chiral structures on metal nanoparticles,8-10 (1) (a) Izake, E. L. J. Pharm. Sci. 2007, 96, 1659. (b) McConathy, J.; Owens, M. J. J. Clin. Psychiatry 2003, 5, 70. (c) Wang, W.; Rusin, O.; Xu, X.; Kim, K. K.; Escobedo, J. O.; Fakayode, S. O.; Fletcher, K. A.; Lowry, M.; Schowalter, C. M.; Lawrence, C. M.; Fronczek, F. R.; Warner, I. M.; Strongin, R. M. J. Am. Chem. Soc. 2005, 127, 15949. (d) Zhang, J.; Albelda, M. T.; Liu, Y.; Canary, J. W. Chirality 2005, 17, 404. (2) Castronuovo, G.; Elia, V.; Niccoli, M.; Strollo, D.; Velleca, F. Phys. Chem. Chem. Phys. 1999, 1, 5653. (3) (a) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin, C. A. Science 1997, 277, 1078. (b) Rosi, N. L.; Giljohann, D. A.; Thaxton, C. S.; Lytton-Jean, A. K. R.; Han, M. S.; Mirkin, C. A. Science 2006, 312, 1027. (4) (a) Alivisatos, A. P.; Johnsson, K. P.; Peng, X.; Wilson, T. E.; Loweth, C. J.; Bruchez, M. P.; Schultz, P. G. Nature 1996, 382, 609. (b) Liu, G. L.; Yin, Y.; Kunchakarra, S.; Mukherjee, B.; Gerion, D.; Jett, S. D.; Bear, D. G.; Gray, J. W.; Alivisatos, A. P.; Lee, L. P.; Chen, F. F. Nat. Nanotechnol. 2006, 1, 47. (5) Jana´ky, R.; Varga, V.; Hermann, A.; Saransaari, P.; Oja, S. S. Neurochem. Res. 2000, 25, 1397. (6) Soutourina, J.; Blanquet, S.; Plateau, P. J. Biol. Chem. 2001, 276, 40864. (7) (a) Ku ¨hnle, A.; Linderoth, T. R.; Hammer, B.; Besenbacher, F. Nature 2002, 415, 891. (b) Ku ¨ hnle, A.; Linderoth, T. R.; Besenbacher, F. J. Am. Chem. Soc. 2006, 128, 1076. (c) Ku ¨ hnle, A.; Linderoth, T. R.; Hammer, B.; Besenbacher, F. J. Am. Chem. Soc. 2003, 125, 14680. (8) Li, T.; Park, H. G.; Lee, H. S.; Choi, S. H. Nanotechnology 2004, 15, S660.

Analytical Chemistry, Vol. 81, No. 2, January 15, 2009

689

and chiral nanoparticles11 in that the interparticle chiral reactivity plays an important role in chiral recognition. In the study by Besenbacher and co-workers on the adsorption of enantiomeric cysteines on a flat Au (110) surface using STM,7 the homochiral cysteine dimers are proposed to adsorb on the surface via sulfur to the bridge site between adjacent layers of atoms with a weaker bond formation between the amino groups and the gold surface. These enantiomer cysteines interact through hydrogen bonding between carboxylic groups of the two dimers. In the recent report on the adsorption of N-acetyl-L-cysteine (NAC) on a Au substrate studied using in situ ATR-IR,9b the amide group of the adsorbed NAC is proposed to be tilted pointing away from the surface, whereas the carboxylic group interacts with the Au surface. On the basis of the adsorption chemistry of thiol-containing amino acids on gold nanoparticles,8-18 we report herein that the chiral selectivity of the interparticle chiral interaction of the amino acids such as cysteines can be regulated and propelled by the surface specificity of nanoparticles in terms of homochiral and heterochiral interactions. In contrast to the prior studies of cysteines, which involve interactions between adsorbed neighboring molecules on a flat surface and do not involve any interparticle reactivity, the work reported herein deals entirely with the interparticle chiral interaction and reactivity of enantiomeric cysteines adsorbed on nanoparticles in a solution. This important focus is also distinct from our earlier report on the assembly of nanoparticles in the presence of homocysteine,15 which describes the interparticle zwitterionic interaction and reactivity using DL-homocysteine. The present report focuses on the role of chirality in the interparticle reactivity using enantiomeric cysteines (L- and D-cys). To our knowledge, this is the first report on how nanoparticles in solutions can be used for regulating chiral recognition of enantiomeric amino acids. Starting from the pairwise zwitterionic interactions of R-amino acids in aqueous solution as reported in a previous microcalorimetric study2 and in our previous study of homocysteine-mediated assembly of nanoparticles,15 a basic element of our novel nanoparticle-regulation strategy for the chiral recognition is the creation of a nanoscale environment either favoring or not favoring the preferential configuration of the pairwise chiral dimerization of (9) (a) Gautier, C.; Bu ¨ rgi, T. J. Am. Chem. Soc. 2006, 128, 11079. (b) Bieri, M.; Bu ¨ rgi, T. J. Phys. Chem. B 2005, 109, 22476. (c) Gautier, C.; Bu ¨ rgi, T. J. Am. Chem. Soc. 2008, 130, 7077. (10) (a) Yao, H.; Miki, K.; Nishida, N.; Sasaki, A.; Kimura, K. J. Am. Chem. Soc. 2005, 127, 15536. (b) Nishida, N.; Yao, H.; Ueda, T.; Sasaki, A.; Kimura, K. Chem. Mater. 2007, 19, 2831. (11) (a) Schaaff, T. G.; Whetten, R. L. J. Phys. Chem. B 2000, 104, 2630. (b) Schaaff, T. G.; Knight, G.; Shafigullin, M. N.; Borkman, R. F.; Whetten, R. L. J. Phys. Chem. B 1998, 102, 10643. (12) Sudeep, P. K.; Joseph, S. T. S.; Thomas, K. G. J. Am. Chem. Soc. 2005, 127, 6516. (13) Aryal, S.; Bahadur, K. C. R.; Bhattarai, N.; Kim, C. K.; Kim, H. Y. J. Colloid Interface Sci. 2006, 299, 191. (14) Naka, K.; Itoh, H.; Tampo, Y.; Chujo, Y. Langmuir 2003, 19, 5546. (15) Lim, I-I. S.; Ip, W.; Crew, E.; Njoki, P. N.; Mott, D.; Zhong, C. J.; Pan, Y.; Zhou, S. Langmuir 2007, 23, 826. (16) Zhang, F. X.; Han, L.; Israel, L. B.; Daras, J. G.; Maye, M. M.; Ly, N. K.; Zhong, C. J. Analyst 2002, 127, 462. (17) Lu, C.; Zu, Y. Chem. Commun. 2007, 3871. (18) Lo, C. K.; Xiao, D.; Choi, M. M. F. J. Mater. Chem. 2007, 17, 2418. (19) (a) Lim, I-I. S.; Goroleski, F.; Mott, D.; Kariuki, N.; Ip, W.; Luo, J.; Zhong, C. J. J. Phys. Chem. B 2006, 110, 6673. (b) Njoki, P. N.; Lim, I-I. S.; Mott, D.; Park, H. Y.; Khan, B.; Mishra, S.; Sujakumar, R.; Luo, J.; Zhong, C. J. J. Phys. Chem. C 2007, 111, 14664. (c) Yang, J.; Lee, J. Y.; Too, H.; Chow, G.; Gan, L. M. Chem. Phys. 2006, 323, 304.

690

Analytical Chemistry, Vol. 81, No. 2, January 15, 2009

Scheme 1. Structural Model for the Nanoparticle-Regulated Pairwise Zwitterionic Interactionsa

a Not to scale. The light-green represents a hypothetical quasiplane for the purpose of illustrating the relative positions of the amino acid groups (C*, NH3+ and CO2- are approximately in the quasi-plane) and the thiolate-anchored gold nanoparticles (The particles are either in the same or the opposite side of the quasiplane). The drawing of only one pair of cysteines in this proposed preferential model of the interparticle pairwise zwitterion interaction is only for illustrative purpose, which could consist of a number of cysteine pairs in reality. (L-, D-cysteines: HSCH2C* H(NH3+)CO2-, C*: chiral center).

the carboxylate and ammonium groups. Such preferential configuration could be due to kinetic or thermodynamic reasons. In this environment, the side chains (i.e., the methylene) of the polar groups (i.e., the amino acid) undergo hydrophobic interactions2 whereas the zwitterionic character constitutes the driving force for the formation of diastereomeric interactions. A fundamental question is how this type of interaction can be guided or regulated by nanoparticles for achieving chiral recognition. Scheme 1 shows a structural model to illustrate our nanoparticle-regulated pairwise zwitterionic interactions using enantiomeric cysteines adsorbed on gold nanoparticles as a model system. With respect to the hypothetical quasi-plane illustrating the interparticle zwitterion interaction of cysteines of different chiralities (L and D), there are three possible interaction modes: two of them are homochiral (LL and DD) and one is heterochiral (DL). As illustrated in Scheme 1, the homochiral case would involve the nanoparticles being aligned on the same side of the quasiplane because of homochiral pairing, for example, L (or D) interacting with L (or D), which is abbreviated as “LL” or “DD”. The heterochiral case would involve the nanoparticles being aligned on the opposite side of the quasi-plane because of heterochiral pairing, for example, L (or D) interacting with D (or L), which is abbreviated as “DL”. In an actual interparticle reaction, the L- (or D-) cys capped nanoparticles can only have homochiral interaction, LL (or DD), whereas the DL-cys capped nanoparticles could have both homochiral (LL or DD) and heterochiral (DL) interactions. A difference between interparticle homochiral and heterochiral interactions is expected based on kinetic or thermo-

dynamic considerations and computational modeling. The demonstration of this novel interparticle chiral recognition concept would parallel some of the important aspects in the initial demonstration of nanoparticle-directed DNA recognition. EXPERIMENTAL SECTION Chemicals. Hydrogen tetrachloroaurate (HAuCl4, 99%), sodium citrate (99%), sodium acrylate (97%), sodium chloride (NaCl, 99%), L-cysteine (cys, 97%), D-cys (99%), DL-cys (97%), DL-methionine (Met, 99%), L-cystine (cys-cys, 99%), N-acetyl-Lcysteine (NAC, 99%), bis(p-sulfonatophenyl)phenylphosphine dihydrate dipotassium salt (BP, 97%), and agarose were purchased from Sigma-Aldrich and used as received. Phosphate buffer (pH 4, 7, 8, 10) were purchased from either Fisher Scientific or VWR. Water was purified with a Millipore Milli-Q water system. For the accurate comparison of reactivities, the weighted mass was corrected by considering the difference in purity for L-cysteine (97%) and D-cysteine (99%). Synthesis. The synthesis of citrate-capped gold nanoparticles of 13 nm diameter (Au13nm) followed the reported procedures described previously.19a Briefly, freshly prepared sodium citrate (38.8 mM dissolved in 5 mL of deionized water) is added to a boiling solution containing 1 mM of HAuCl4 (in 45 mL of deionized water). The solution is heated for an additional 30 min, and the nanoparticle solution displays a ruby-red color. The particle size determined by transmission electron microscopy (TEM) was 13.3 ± 0.9 nm. The synthesis of acrylate capped gold nanoparticles of 30 and 60 nm diameters (Au30nm and Au60nm) followed our previously reported procedures.19b The particle sizes were 32.6 ± 1.6 nm and 62.0 ± 1.9 nm. The synthesis of citrate-capped gold nanoparticles of 6 nm diameter followed the reported procedure,19c and the particle size was 6.1 ± 1.1 nm. Measurements. UV-visible spectra were acquired with an HP 8453 spectrophotometer. Spectra were collected over the range of 200-1100 nm. A cuvette with a path length of 1.0 cm was utilized. Briefly, Aunm solution and NaCl are quantitatively mixed and allowed to react for 15 min, upon which a quantitative amount of cysteine was added, and the reaction was monitored via UV-vis. The change in ionic strength of the solution is negligible since the proton concentration is much less than the salt (NaCl) concentration (CH+ , Csalt). Unless otherwise stated, the concentrations used for the study were [Au13nm] ) 2.4 nM; [Au60nm] ) 0.021 nM; [NaCl] ) 5.0 mM; and [cys] ) 10.0 µM. Dynamic light scattering (DLS) was performed on a standard laser light scattering spectrometer (BI-200SM) equipped with a BI-9000 AT digital time correlator and a He-Ne laser (35 mW, 633 nm) to determine the evolution of the hydrodynamic diameter for the nanoparticle assemblies in solution. All solutions were filtered using 0.45 or 0.22 µm filters. The typical concentrations are [Au13nm] ) 2.5 nM; [NaCl] ) 5.0 mM; and [cys] ) 5.0 µM. Gel electrophoresis data were obtained using a Pharmacia voltammeter. A typical experiment used 1% agarose gel at 70 V (7.8 V/cm) and phosphate buffer (pH 10) as the running buffer with a run time of 20 min to 1 h. The typical concentrations were [Au13nm] ) 12 nM and [cys] ) 23 µM. Briefly, 17 µL of the assembled solution (nanoparticles and cysteines) and 3 µL of loading buffer (contains glycerol) were transferred to a 2 mL

centrifuging tube and spun briefly using a vortex shaker. The 20 µL solution was then carefully pippeted into the wells of the freshly prepared agarose gel. X-ray Photoelectron Spectroscopy (XPS) measurements were performed using a Physical Electronics Quantum 2000 Scanning ESCA Microprobe. This system uses a focused monochromatic Al KR X-ray (1486.7 eV) source and a spherical section analyzer. The instrument has a 16 element multichannel detector. The X-ray beam used was a 100 W, 100 µm diameter beam that was rastered over a 1.3 mm by 0.2 mm rectangle on the sample. The X-ray beam is incident normal to the sample, and the photoelectron detector was at 45° off-normal. Wide scan data was collected using a pass energy of 117.4 eV. The binding energy (BE) scale is calibrated using the Cu2p3/2 feature at 932.62 ± 0.05 eV and Au 4f at 83.96 ± 0.05 eV for known standards. The vacuum chamber pressure during analysis was 1 monolayer. At submonolayer cysteine levels, where the surface structure consists of mixed cysteines and citrates, we believe that they are homogeneously mixed, though the dispersity is a subject of ongoing investigation.21 The gray-shaded region indicates the concentration range estimated for a monolayer coverage of cysteines on the particle surface. The estimated minimum-maximum range is 1800-3000 cysteines per particle (13 nm). As shown by the

Figure 2. UV-vis spectral evolution for LL-, DD-, and DL-cysteines mediated assemblies of gold nanoparticles of (A) 13 nm, (B) 30 nm, and (C) 60 nm. Spectra were collected over the course of 1 h (A) and 30 min (B, C). For cysteine-Au13nm: ([Au13nm] ) 2.4 nM; [NaCl] ) 5.0 mM; [cys] ) 10.0 µM). pH ) 6.2 (LL), 6.2 (DD), and 6.2 (DL). For cysteine-Au30nm: ([Au30nm] ) 0.14 nM; [NaCl] ) 10.0 mM; [cys] ) 10.0 µM). pH ) 6.4 (LL), 6.5 (DD), and 6.4 (DL). For cysteine-Au60nm: ([Au60nm] ) 0.01 nM; [NaCl] ) 20.0 mM; [cys] ) 16.0 µM). pH ) 6.7 (LL), 6.8 (DD), and 6.8 (DL).

Figure 3. Correlation between the interparticle reactivity represented by normalized rate or rate constant (k) and the enantiomeric percentage of cysteines (%) in the presence of gold nanoparticles of two different sizes Au60nm (blue squares) (pH ) 5), Au13nm (black circles) (pH ) 6), and Au13nm (green triangles) (pH ) 7). A data point for Au13nm reacting with NAC (red diamonds) is included, and its inactivity is represented by the lower-right scheme (not to scale). The red dot-dash line represents the theoretical rate derived from modeling LL, DD, and DL dimerization kinetics (eq 1), whereas the other gray dashed lines show the trends of the experimental data points (not fitting data). The schemes shown in the top and bottom panels are for the purpose of illustrating the different interparticle chiral interactions, whose details are further discussed in the text of this and latter sections.

surface composition analysis obtained by XPS described later, samples for the monolayer level coverage of cysteines on the particles all showed practically identical surface coverage for the adsorbed cysteines, regardless of the chirality. While it is possible that the interparticle linkages may include LL, DD, and DL interactions, the systematic change shown by the

relationship of the apparent rate constants versus the racemic ratio and its consistency with theoretical considerations (Figure 3) provide substantial evidence for the operation of the heterochiral interaction. The fact that the homochiral interparticle reactivities are greater than the heterochiral ones, even at a submonolayer coverage or with a mixed monolayer (e.g., L- and D-cysteines Analytical Chemistry, Vol. 81, No. 2, January 15, 2009

693

Figure 4. Plot of the apparent rate constants (k, s-1) for the LL- (black circles), DD- (red squares), and DL- (blue triangles) assemblies at different cysteine concentrations (µM) while the concentration of nanoparticles and salt are kept constant. The gray area corresponds to the concentration range for monolayer coverage of cysteines on the particles.

mixed with non-zwitterionic NAC to form mixed monolayers), rules out the possibility of preferential interactions of DL on the same particle under these experimental conditions. Moreover, model analysis of L and D cysteines on the same particle surface indicates that the pairwise zwitterionic interaction is energetically less favorable because it would require a rotation of the amino acid groups in one of the chiral molecules. (Scheme 1) This is in fact supported by earlier calorimetric-based data of the pairwise enthalpic interaction coefficients of cysteines in aqueous solution which showed that LL or DD . DL.2 Factors Affecting Interparticle Chiral Reactivity. Both pH and ionic strength effects on the rate constant supported the zwitterionic nature for cysteines in the molecular recognition process. Figure 5 shows a representative set of data comparing the pH and salt concentration effects on the rate constants. While the rate constants were found to decrease with pH (e.g., k(LL) ) 8.2 × 10-2 s-1 (pH 5) and 7.8 × 10-5 s-1 (pH 7)), the difference between homochiral and heterochiral assemblies is dramatically enlarged as pH increases (as large as 6-7 orders of magnitude was observed for the case of pH ∼7, Figure 5A). On the basis of the pKa values (pKa1, 2, 3 ) 1.92 (-CO2H); 8.37 (-SH); and 10.70 (-NH2)) and isoelectric point (pI ) 5.02) of cysteine,22 the -CO2H groups are mostly deprotonated to -CO2at pH ∼ 5 while -NH2 is protonated to -NH3+, maximizing the zwitterionic electrostatic interactions. On the other hand, the observation that significant reactivity requires the presence of salt is consistent with the zwitterionic nature of cysteines. The salt concentration dependence of the reaction rates (Figure 5B) reflects the fact that the stabilization of zwitterions is linked to the thickness of the diffuse layer surrounding the charged particle (∝ (Csalt)-1/2).15 We also attempted to isolate L- and D-cysteine capped Au nanoparticles separately before the occurrence of assembly so that their zwitterionic reactivities can be assessed under a different condition. However, upon the adsorption of cysteines onto the particle surface, the favorable homochiral reactivity led to the immediate formation of LL or DD linked assemblies of nanoparticles, which made it nonviable to assess the zwitterionic reactivities of the isolated particles under these experimental condi(21) (a) Hostetler, M. J.; Templeton, A. C.; Murray, R. W. Langmuir 1999, 15, 3782. (b) Tracy, J. B.; Kalyuzhny, G.; Crowe, M. C.; Balasubramanian, R.; Choi, J.; Murray, R. W. J. Am. Chem. Soc. 2007, 129, 6706. (22) Gao, Z. N.; Zhang, J.; Liu, W. Y. J. Electroanal. Chem. 2005, 580, 9.

694

Analytical Chemistry, Vol. 81, No. 2, January 15, 2009

tions, whereas the non-zwitterionic NAC-capped nanoparticles do not. The idea of using pre-made biomolecule capped nanoparticles has also been tried in our laboratory, as reported by other researchers (e.g., glutathione capped nanoparticles,11 penicillamine capped nanoparticles,10 N-isobutyryl cysteine capped nanoparticles,9a and cysteine capped nanoparticles 14). The resulting nanoparticles are often too small (2-5 nm) to exhibit sufficient SP band intensity for assessing the assemblies. Unlike the previous demonstrated homocysteine-mediated assembly,15 the cysteine nanoparticle assemblies can only be partially disassembled by manipulating pH and temperature (Figure 6). In a typical experiment, the nanoparticle assemblies were allowed to precipitate overnight. After a brief sonication, a blue solution was observed (arrow 1). The pH was then adjusted to 11.5 and the solution was heated in an oven at 75-90 °C for a duration of 30-80 h. The spectra showed some indication of partial reversal of the SP band evolution (arrow 2). The increase in temperature accelerates the speed of the reversal, as demonstrated for the case of homocysteine.15 At room temperature there was no significant indication of the partial reversal for days. Note that the disassembly also depends on the overall concentrations of cysteines and nanoparticles in the solution. There are two important implications from this set of data. First, the pH- and temperature-induced disassembly show almost the same degree of partial reversal for LL- (A) and DL- (B) assemblies, which may imply similar interparticle binding energies. One possible scenario for this to happen is that the actual interparticle linkages involved homochiral (LL and DD) interactions in the DL assembly case. Second, the difficulty to achieve complete reversal seemed to be counterintuitive based on considerations of pKa values for the acid and amino groups (1.92 and 10.70).22 One possible explanation for this is that the actual pKa value of the amino acid groups might have changed to a higher value in the microenvironment of the zwitterion-linked nanoparticle assembly. Such a change may require a higher pH to disassemble the assembly, which is beyond the stability limit of the nanoparticles. The fact that the difference of only one methylene unit between homocysteine15 and cysteine can have such a profound impact on the binding strength substantiates the important role of the zwitterion interaction in the chiral recognition. It is also possible that there are multiple zwitterionic interactions which are more complicated than the idealized dimerization models used here to explain the basic concept. The interaction for a DL assembly may include a combination of LL, DD, and DL interactions. Part of our ongoing work involves the use of specifically designed nanoparticles to separate the products and to provide the realistic details. Spectroscopic Identification of Adsorbed Cysteines of Different Chirality. The chemical identities of cysteines adsorbed on Au nanoparticles responsible for the zwitterionic interactions were confirmed by XPS analysis for samples deposited on gold thin film or molybdenum (Mo) substrates. Figure 7 shows a typical set of XPS spectra in the BE regions of S2p, N1s, O1s, and Au4f for LL-, DD-, and DL-cysteine assemblies of gold nanoparticles. Note that the nanoparticle assembly samples were precipitated and washed by deionized water to remove any free cysteines. The BE of S2p bands detected at 163.4 and 162.3 eV

Figure 5. (A) Comparison of pH effect on the relative rate constants (k) while the concentration of nanoparticles, electrolytes, and cysteines are kept constant. The bar color: black for LL, red for DD, and green for DL-cysteine assemblies. (B) Comparison of salt concentration effect on the relative reaction rates while the concentration of nanoparticles and cysteines are kept constant (pH ) 6.0). The bar color: brown for LL and orange for DL-cysteine assemblies. The rate constants are normalized to the highest k value in each case.

Figure 6. Disassembly of the LL- (A) and DL- (B) assemblies of Au13nm. ([Au13nm] ) 9.8 nM; [cys] ) 41.0 µM). Arrow 1: assembly of nanoparticles after 1 day. Arrow 2: after 30 h of heating at 75 °C in the presence of NaOH. The color evolved from red (nanoparticles, pH 6.0) to bluish (assembly) and to purple/reddish (after heating at 75 °C and pH 10.8).

(Figure 7A) is identical for LL, DD, and DL assemblies of gold nanoparticles demonstrating that cysteines are bound to the gold surface via a gold-thiolate bond.23 The detection of the N1s band at 401.6 eV (Figure 7B) indicates that the amino group is present predominantly as -NH3+ species, in sharp contrast to the dominance of -NH2 species for the adsorption of cysteines on single crystal Au surfaces.23 The detection of an identical O1s band at 531.5 eV (Figure 7C) demonstrates that the acid group is present predominantly as -CO2- species.24,25 Similar data were obtained for the samples deposited on a molybdenum (Mo) substrate. In this case, not only identical S2p, N1s, and O1s bands but also the Au4f bands (BE (Au4f5/2 and 7/2) ) 87.8 and 84.2 eV) were observed, which is a characteristic of the Au nanoparticles (Figure 7D). These results, along with similar data for samples deposited on a molybdenum substrate (not shown), eliminate any possibility of interference due to adsorption of residual free cysteines on the Au substrate and establish the identities of the adsorbed cysteines in the assemblies as gold-bound thiolates and zwitterions (Figure (23) Schillinger, R.; Sˇljivancˇanin, Zˇ.; Hammer, B.; Greber, Phys. Rev. Lett. 2007, 98, 136102. (24) Wu, N.; Fu, L.; Su, M.; Aslam, M.; Wong, K. C.; Dravid, V. P. Nano Lett. 2004, 4, 383. (25) Zubavichus, Y.; Fuchs, O.; Weinhardt, L.; Heske, C.; Umbach, E.; Denlinger, J. D.; Grunze, M. Radiat. Res. 2004, 161, 346.

7E). The results of the surface analysis of S-to-Au or N-to-Au atomic ratios (Supporting Information, Table S2) further indicated that the surface coverages of the adsorbed cysteines are practically independent of the chirality. There is no difference for the identity and coverage of the adsorbed cysteines. Characterization data from Surface-enhanced Raman scattering (SERS) spectra and Circular Dichroism (CD) spectra further substantiate the adsorption of cysteines on the surface. From SERS spectra for Au nanoparticles in the presence and in the absence of cysteine, peaks characteristic of the adsorbed cysteine were detected (for description, see Supporting Information, Figure S1).26 CD spectra (Supporting Information, Figure S2) provide additional information.8-10,27 In contrast to the observation of positive and negative bands in the far-UV region (∼200 nm) for L- and D-cysteines and a small positive band for DL-cysteine, no band was detected for the LL-, DD-, and DL-cysteine mediated assemblies. This observation is consistent with a previous report.8 The lack of difference for the nanoparticle assemblies mediated by cysteines of different chirality suggests that the close association of the chiral molecules with the nanoparticles likely diminished the detection of CD spectra (see Supporting Information for detail discussion). Note that the absorbance of the peak in the CD spectra corresponding to free cysteines was greatly reduced in the presence of nanoparticles, indicative of the strong adsorption of cysteines on gold nanoparticles. It is important to emphasize that the cysteines did not lose their chirality after attaching onto the nanoparticle surface. They remain chiral, but their adsorption on the gold nanoparticles likely diminishes the observation of the optical activity because of the nanoparticle size effect. While the adsorption of cysteines on flat gold substrates is known to preserve their biological functions, and recently biological activities were also observed for cysteine-related molecules upon adsorption on gold nanoparticles,28 the understanding of how their adsorption on gold nanoparticles affects the biological functions will need further investigations. Size Changes for the Interparticle Assembly. The chiral recognition process for the enantiomeric cysteine-mediated as(26) Podstawka, E.; Ozaki, Y.; Proniewicz, L. M. Appl. Spectrosc. 2004, 58, 570. (27) Oi, H.; Hegmann, T. J. Mater. Chem. 2006, 16, 4197. (28) (a) Lee, J.; Ulmann, P. A.; Han, M. S.; Mirkin, C. A. Nano Lett. 2008, 8, 529. (b) Huang, C.; Tseng, W. Anal. Chem. 2008, 80, 6345. (c) Gates, A. T.; Fakayode, S. O.; Lowry, M.; Ganea, G. M.; Murugeshu, A.; Robinson, J. W.; Strongin, R. M.; Warner, I. M. Langmuir 2008, 24, 4107.

Analytical Chemistry, Vol. 81, No. 2, January 15, 2009

695

Figure 7. Identification of functional groups of cysteines adsorbed on the Au nanoparticle surface. XPS spectra in the BE regions of (A) S2p1/2 (163.4 and 162.3 eV), (B) N1s (401.6 and 399.9 eV), and (C) O1s (533.6 and 531.5 eV) for LL- (black circles), DD- (red squares), and DL(blue triangles) assemblies of nanoparticles (samples were cast on Au/glass substrates). (D) XPS in Au4f (BE (Au4f5/2 and 7/2) ) 87.8 and 84.2 eV) region for the same samples cast on Mo substrates. The curve fitting (dash lines) was based on spectral deconvolution using full width at half-maximum (fwhm) for S2p ∼ 1.0 eV, N1s ∼ 1.5 eV, and O1s ∼ 2.0 eV, and Lorentzian for S2p and N1s and Gaussian for O1s bands. (E) An illustration (not to scale) of the functional group identities of cysteine adsorbed on the Au nanoparticle surface.

and 3/2

semblies of gold nanoparticles involves a change in the assembly size. Such changes and related properties were determined by DLS,15,29,30 and were also characterized by gel electrophoresis (GE) measurements 31-33 and TEM analysis. Figure 8 shows a representative set of DLS data monitoring the growth in the hydrodynamic diameters (Dh) as a function of time for the assemblies of gold nanoparticles mediated by L-, D-, and DLcysteines. While the overall growth rate of the hydrodynamic diameters is largely linear in all three cases, the rates for the homochiral assemblies are much greater than that for the heterochiral assembly, demonstrating the important role of chiral recognition in the size growth of the nanoparticle assembly. A detailed examination revealed that the initial rates for the two homochiral cases were similar for the first ∼60 min, after which the LL assembly appeared slower than the DD assembly, which was likely due to the partial precipitation of the LL assembly in the solution. The initial subtle decrease in Dh (