Sonia
Maniappan
a,
Camelia
Dutta
a,
Arunima
Cheran
a,
Diego M.
Solís
b and
Jatish
Kumar
*a
aDepartment of Chemistry Indian Institute of Science Education and Research (IISER) Tirupati, Tirupati 517507, India. E-mail: jatish@iisertirupati.ac.in
bDepartamento de Tecnología de los Computadores y de las Comunicaciones, University of Extremadura, 10003 Cáceres, Spain
First published on 16th April 2024
The formation of chiral nanosystems and their subsequent enantioselective interaction with chiral amino acids are vital steps in many biological processes. Due to their potential to mimic biological systems, the synthesis of chiral nanomaterials has garnered significant attention over the years. Despite the emergence of diverse nanomaterials showcasing strong chiral responses, the in-depth understanding of the mechanism of plasmonic chirality in copper nanoparticles and their subsequent application in various fields are least explored. Herein, we demonstrate a facile approach for the synthesis of chiral copper nanoparticles using cysteine as a chiral precursor and capping ligand. Ligand-mediated chiral induction, established through experimental findings and a theoretical model, is ascribed as the major contributor to the origin of plasmonic chirality. The enantioselective recognition of chiral copper nanoparticles towards histidine, an amino acid with vast biological functions, was meticulously investigated by leveraging the strong copper–histidine binding ability. Ligand-induced dissolution, a unique phenomenon in nanoparticle reactions, was identified as the underlying mechanism for the nanoparticle-to-complex conversion. Understanding the mechanism of chiral induction in copper nanoparticles coupled with their enantioselective recognition of biomolecules not only holds promise in biomedical research but also sheds light on their potential as catalysts for asymmetric synthesis.
The generation of chirality and its intriguing applications are well explored in plasmonic nanomaterials, such as gold and silver.25–27 However, the fundamental understanding and applicability of chiral copper nanoparticles (CuNPs) is an area that remains underdeveloped. Owing to their excellent electrical and thermal properties, comprehended by their unique optical response in the visible and near-infrared regions, copper nanomaterials that are relatively inexpensive are emerging as viable alternatives to precious metals like gold and silver in various plasmonic applications.28–32 Despite their advantages, the synthesis of robust copper-based nanomaterials has remained a challenging task due to the low standard redox potential of the metal leading to surface passivation through formation of oxides.33,34 In this context, dedicated efforts to develop facile strategies for the synthesis of stable CuNPs using appropriate capping ligands are a pressing priority.35,36 While there are a few recent reports on the synthesis and chiroptical investigations of CuNPs, the existing literature on this topic remains relatively limited, and comprehensive studies are required to fully understand the origin of chirality and to further explore their potential applications.37–40 Herein, we establish a straightforward and efficient method to generate CuNPs from amino acids as a chiral precursor and capping agent. The optical activity of synthesized CuNPs was confirmed by circular dichroism (CD), which gave mirror image signals for the particles synthesized using L- and D-cysteine.
Furthermore, the major focus of investigations has been the enantioselective recognition of optically active CuNPs towards histidine. Histidine is a vital amino acid that is crucial for human growth and development as it serves as the precursor for histamine, which is essential for triggering inflammatory responses in the body. Histidine plays an important role in many biological functions like immune responses and is present in the active site of many metallozymes.41,42 The significance of histidine extends beyond biological functions. It has proven useful in asymmetric synthesis and has been utilized as a catalyst for cross-aldol reactions involving enolizable aldehydes.43 Unlike L-histidine, which is included in food supplements in the form of capsules, tablets, or injections, D-histidine does not have any functional role in the body, rendering the enantioselective recognition of histidine imperative.44 Therefore, investigating the enantioselective interaction between chiral CuNPs and biomolecules, such as histidine, is of considerable interest in various biomedical applications. Taking advantage of strong Cu–histidine binding ability, the enantioselective interaction of the CuNPs with histidine was investigated, and the ligand-induced transformation of the CuNPs into metal complexes was probed. To the best of our knowledge, this is the first attempt toward using chiral CuNPs for establishing the enantioselective recognition of a specific amino acid.
The optical properties of the synthesized CuNPs were analyzed using UV-visible and CD spectroscopy. The UV-visible spectra showed typical features in the region corresponding to the LSPR and interband transition of CuNPs: a broad absorption peak covering 500–800 nm with a shoulder at around 620 nm (Fig. 2a). The nanoparticles showed mirror image CD spectra for the particles synthesized using opposite enantiomers of the amino acid with a gabs value of −1.1 × 10−3 and 1.21 × 10−3 (at 613 nm) for L- and D-CuNPs, respectively (Fig. 2b). Positive and negative couplets were observed for the nanoparticles synthesized using L- and D-cysteine, respectively. CD signals observed in the visible and NIR region of the electromagnetic spectrum must originate from the interactions of the CuNPs with the chiral field of the ligand.52 The stability of the CuNPs was assessed by monitoring spectral changes over a period of several days. The consistency in the position and intensity of the absorption and CD peaks revealed that the nanoparticles remained stable without any degradation over the entire period of analysis (Fig. S5†). Moreover, the stability of the CuNPs was evaluated in solvents of varying polarity. Among different solvents analyzed, nanoparticles dispersed in polar solvents, such as water, ethanol, and DMSO, were found to retain their spectral profile confirming their stability in these solvents (Fig. S6†). The pH-dependent stability analysis revealed that at lower pH the CuNPs exhibited a broadened UV-visible signal and weak CD due to the aggregation of cysteine (Fig. S7†).53 Notably, the CuNPs displayed relatively high stability at pH 9 due to the absence of electrostatic interaction between the uncharged amino group and negatively charged carboxylic group, thus preventing their aggregation.54 To demonstrate the optical properties of the synthesized nanoparticles in the solid state, self-standing films of the CuNPs in polyvinyl alcohol (PVA) were fabricated. CD plots of transparent polymeric films exhibited spectral signatures similar to those in solution, albeit with a slight broadening (Fig. S8†) pointing to the applicability of the nanoparticles as optically active solid-state material. Moreover, incorporation in a polymer matrix can be adopted as a viable strategy to protect the particles from surface oxidation.
Fig. 2 (a) Extinction and (b) CD spectra of CuNPs synthesized using L-cysteine (blue trace) and D-cysteine (red trace) as capping ligands. |
To further understand the physical origin and nuances of chirality in our synthesized CuNPs, a theoretical analysis was carried out: the roughly spherical shape of these shells justifies the assumption of two ideally spherical surfaces as interfaces separating the copper core, cysteine layer and water background. This allows for a completely analytical and exact solution to Maxwell's equations based on an expansion of vector spherical harmonics (Mie theory), with no approximations other than the above-mentioned assumption of perfect sphericity.55 In Fig. 3, we apply this model to a single Cu sphere of radius 29.1 nm, surrounded by a 2 nm-thick cysteine coating with isotropic chiroptical activity (details on its parameterization can be found in the ESI†). Fig. 3a shows the extinction efficiency Qe (in terms of extinction cross section divided by area: ECS/πr2) vs. wavelength for incident L- (solid lines) and R-handed (dashed lines) plane-wave circular polarizations, considering different Cu-to-Cys volume fractions f(Cu/Cys). These curves are consistent with the experimental observation of a damped resonance around 600 nm: in the theoretical model, a sharper resonance emerges at 583 nm, together with an additional peak that redshifts as f(Cu/Cys) increases. Importantly, the first resonance not only is invariant with the Cu core radius, but it also shows up without the cysteine layer (see panels (c) and (d) of Fig. S9 of the ESI,† respectively); these findings, together with the isotropic nature of the sphere, lead us to conclude that the 583 nm resonance is mainly due to the interband transitions of copper, which rule its dielectric function for wavelengths below approximately 600 nm (above this threshold, a simple Drude model is sufficient to describe its metallic response, as shown in Fig. S9a†). As for the f(Cu/Cys)-dependent peak, this can be tracked down to the 225 nm resonance of the model chosen for the permittivity of cysteine56 (Fig. S9b†): in the effective medium approach, f(Cu/Cys) thus acts as a controlling factor that brings the position of this resonance to longer wavelengths. In Fig. 3a, there is a clear resonance splitting, e.g. for f(Cu/Cys) = 0.35, but somewhere in the range [0.20, 0.30] (blue, red, green) these two modes become hybridized, which can already explain by itself, to some extent, the smoothed out experimental peak. This hybridization mechanism adds to the zero average anisotropies and Gaussian size distribution of synthesized quasi-spherical particles (polydispersity), which all contribute to this smoothing.
The monotonically increasing trend of the extinction towards higher energies observed in experimental measurements is not entirely present in these analytical results. Part of this discrepancy might be attributed to the absence of polydispersity in the theoretical model but, given the lack of anisotropy—at least in an average sense—and the above-mentioned invariance of the extinction profile shape with particle size, it most likely can be related to slight divergences in the effective dielectric function of the synthesized copper with respect to tabulated values in ref. 57 in the region where interband transitions become preponderant, and/or some effective dielectric mixing of (Cu–Cys)/H2O compounds.57 This rationale is supported by the fact that, by replacing water with air as the background, extinction curves do match measurements in this low-wavelength region (Fig. S9c†): now, we stress, again, that the shape of the extinction curve is independent of the particle size, and hence, we would obtain similar curves by varying the dielectric function of copper such that ε(Cu,new)/εwaterε ≈ εCu/εair within this high-energy region. In the framework of a Lorentzian model, a similar effect would be achieved in this case by simply reducing the value of the dielectric function by a small constant, i.e., using a smaller effective ε∞.
Fig. 3b and (c) show the quantification of chiroptical responses from L-CuNPs and how the results of approximations considered (see the ESI†), with variations of the order of ∼10 millidegrees in the [600–700] wavelength span, agree very well with the experimental measurements. Fig. 3d–(f) present near-field electric field distributions of a single CuNP at 614 nm, an operating point which in Fig. 3b (red curve) shows a maximum in CD when f(Cu/Cys) = 0.25. Fig. 3d depicts the normalized Ex component—the real part of its phasor, to be more precise—when the incident field is polarized along the x axis with arbitrary magnitude E0. In such a case, a chirally inactive sphere would produce identically zero Ey field component at planes x = 0 and y = 0. Our chiral NP, however, yields normalized Ey values of ∼0.01, as shown in Fig. 3e. Fig. 3f illustrates the norm of the difference of (Ex, Ey, and Ez) electric field vectors with and without chiral response εCys = (εL + εR)/2, which is basically the square root of the optical intensity of such a difference, again normalized to E0. This colour map distinctly shows how the largest magnitudes happen inside the cysteine shell, as expected.
Establishment of the chiral nature of the synthesized CuNPs lays the groundwork for their application in enantioselective recognition. The enantioselective detection of amino acids by L-CuNPs was studied by the sequential addition of amino acids and monitoring the spectral changes. The addition of most amino acids to the CuNPs resulted in minimal/no noticeable changes in CuNP spectra. Interestingly, the addition of L-histidine to L-CuNPs resulted in distinct CD spectral profile; an inversion of the CD signal from a negative peak at 621 nm to a positive peak at 687 nm was the noted signature upon the interaction of L-histidine with L-CuNPs (Fig. 4b). The CD peak maxima displayed a remarkable bathochromic shift of 66 nm with an inversion of sign. In contrast, the addition of D-histidine to L-CuNPs resulted in an increase in the negative CD intensity with a bathochromic shift in peak position (Fig. 4d). The CD spectra after the reaction of L- and D-histidine with L-CuNPs showed mirror image spectral profiles (Fig. 5b). A similar pattern of CD peak reversal was observed with the addition of D-histidine to D-CuNPs (Fig. S10†). Hence, CD spectral changes observed in optically active CuNPs upon the addition of histidine could help distinctly differentiate one enantiomer of histidine from the other (Fig. 4e). However, the introduction of a racemic mixture of histidine led to the disappearance of optical activity of CuNPs (Fig. S11†). In all cases, UV-visible spectra showed the conversion of the dampened band of CuNPs to a relatively structured peak upon interaction with L-histidine (Fig. 4a and c). The spectral changes were accompanied by a visual colour change of the solution from turbid brown to clear blue (inset, Fig. 4a). The studies with other sets of amino acids revealed that the spectral changes are minimal, highlighting the specificity of the CuNP interaction with histidine (Fig. 4f and S12–S14†). The optical measurements reveal that the synthesized chiral CuNPs can be used for selective identification of histidine as well as for its enantioselective recognition. The nature of the metal aids the selectivity towards amino acid (histidine), whereas nanoscale chirality helps in enantioselectivity.
The spectral changes observed during the addition of histidine to the CuNPs point towards a substantial change to the nanoparticles during the interaction. To further probe the changes occurring at the molecular level, the solution after the addition of histidine was purified and left undisturbed. To our surprise, crystal formation was observed over time. The formation of single crystals helped authenticate the molecular structure using single-crystal X-ray diffraction (SC-XRD). Interestingly, SC-XRD data showed the formation of metal complexes that crystallized in monoclinic space groups (Fig. 5a and Table S2†). The crystal structure of the complex revealed a distorted square planar geometry with a Schiff base ligand formed from histidine (Fig. 5a).58 The formation of the Schiff base ligand was facilitated by the self-aldol condensation of acetone, which occurred through a vapour diffusion process. In the observed structure, one copper atom is bound to two histidine ligands, each through a bidentate binding. One histidine, acting as negatively charged bidentate ligand, binds to the copper(II) center through N (amino) and O (carboxyl) atoms with bond lengths of 1.96 Å and 1.98 Å, respectively. The other histidine coordinates with the copper(II) center utilizing its amino and imidazole nitrogen atoms with bond lengths of 2.02 Å and 1.98 Å. The UV-visible and CD patterns observed during the addition of histidine to the CuNPs closely align with the expected spectral changes during the nanoparticle-to-complex conversion (vide supra). The addition of L- and D-histidine to the CuNPs resulted in opposite spectral features, indicating the conversion of nanoparticles to complexes with opposite handedness (Fig. 5b). The absorption band for the complex is the result of d–d transitions from the Cu2+ center.59
For the better understanding of ground-state photophysical activities, theoretical simulations were performed using Gaussian 09 software (Fig. S15–S17†). The calculations show absorption peaks at 686.36 nm and 350.79 nm involving energies of 1.8064 eV and 3.5345 eV, respectively. The simulated UV-visible and CD spectra of the metal complex match well with experimental plots confirming the structural purity of the complex as well as the nature of excitations (Fig. S16†). To study the reversibility of the process, cysteine was added to the formed complex. However, subsequent addition of cysteine to the complex showed the disappearance of CD signals but not the regaining of plasmonic peaks indicating that the conversion of nanoparticle to complex is an irreversible process (Fig. S18†). The relative binding strengths of cysteine and histidine to copper would act as the deciding factor for forward and reversible reactions, and these aspects were probed further.
Investigations of the mechanism show that the nanoparticle-to-complex conversion is facilitated by the ligand-induced dissolution process as illustrated in Fig. 6. The process involves the utilization of specific molecules or ligands, such as amino acids in this particular case, to guide and control the dissolution of nanoparticles, leading to the subsequent formation of a complex. The addition of a new ligand would displace the surface-bound ligand on the nanoparticles or stabilize intermediate surface species, promoting the release of ions into the surrounding medium. The ligand can then form a complex with released ions, stabilizing them in solution and preventing their re-adsorption onto the nanoparticle surface. The complex formation was further identified by the change in the morphology of the nanoparticles with histidine addition. The nanoparticle size decreased gradually with each successive addition of histidine leading to the absence of visible particles in SEM images following the last addition. These observations may be ascribed to the dissolution of nanoparticles and subsequent liberation of ions and complex formation (Fig. S19†). Ligands with a high binding affinity for the nanoparticle surface and those with the ability to promote redox reactions are more likely to induce dissolution.60,61 The relative binding strengths of amino acids determine their ability to form stable complexes and control the dissolution process.
Fig. 6 Scheme illustrating the nanoparticle -to-complex conversion through the ligand-induced dissolution process. |
The spectral changes of the CuNPs upon treatment with different amino acids indicate that positively charged amino acids such as histidine and lysine interact well with the CuNPs (Fig. S12–S14†). In contrast, addition of L- and D-cysteine to L-CuNPs resulted in the complete loss of chirality due to the aggregation and precipitation of nanoparticles. The overall findings from amino acid addition experiments are that the binding is specific to histidine leading to a noticeable inversion of the CD signal. This outcome is attributed to the exceptional binding capability of histidine to copper, characterized by high binding constants. To substantiate this claim, binding constant values for the interaction of copper with various amino acids were calculated using the Benesi–Hildebrand method (Fig. S20†). The calculated values were found to be 5.43 × 107 M−1, 1.37 × 104 M−1 and 2.33 × 105 M−1 for copper binding with histidine, cysteine and lysine, respectively. The irreversibility of the complex-to-nanoparticle conversion is also a result of this difference in binding strengths. The binding mechanism of histidine to copper differs from that of other amino acids primarily due to its distinctive imidazole side chain. The unshared electrons on the imidazole nitrogen atom enable the formation of coordination bonds with copper ions. This proposition was further supported by the SC-XRD analysis of the complex formed upon addition of histidine to L-CuNPs (vide supra). The observed inversion in CD signals with histidine, in contrast to other amino acids, can be attributed to the involvement of the imidazole group in bonding, resulting in an asymmetric distorted square planar geometry.
Footnote |
† Electronic supplementary information (ESI) available. CCDC 2308273 and 2308274. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4sc00477a |
This journal is © The Royal Society of Chemistry 2024 |