Chiral Fe_xCu_ySe nanoparticles as peroxidase mimics for colorimetric detection of 3, 4-dihydroxy-phenylalanine enantiomers

Poly(N-vinylpyrrolidone) (PVP), L/D-Histidine (L/D-His), L-3,4-dihydroxy-phenylalanine (L-dopa) and D-3,4-dihydroxyphenylalanine (D-dopa), L/D-Phenylalanine (L/D-Phe), L/D-Cysteine (L/D-Cys) were obtained from Aladdin (Shanghai, China). L-ascorbic acid (AA) was purchased from Tianjin Fengchuan Chemical Reagent Co., Ltd. CuSO₄·5H₂O was ordered from Tianjin Yongda Chemical Reagent Co., Ltd. FeCl₃·6H₂O was obtained from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). Levodopa tablets were purchased from Beijing Shuguang Pharmaceutical Co., Ltd. H₂O₂ was bought from Laiyang Kangde Chemical Co., Ltd. 3,3',5,5'-Tetramethylbenzidine (TMB) was brought from damas-beta. All reagents and solvents were used as received from commercial suppliers without further purification. Deionized water (D.I. water, 18.2 MΩ cm) was utilized to prepare all the solutions.

Transmission electron microscopy (TEM) images were acquired using a JEM-2100 (JEOL) operating at 120 kV. Fourier translation infrared (FTIR) spectra were collected with a Shimadzu FT-IR-8400S. The UV–vis absorption spectra were monitored by a Shimadzu UV-2700 spectrophotometer (Japan). The x-ray diffraction (XRD) patterns were recorded using a SmartLab Focus diffractometer (Rigaku) with Cu Kα source (λ = 1.541862 Å). CD measurement was carried out on a J-1500 CD spectrophotometer (JASCO) with the scanning speed of 10 nm min⁻¹ and data pitch of 0.2 nm. A FEI Super-X microscope was used to obtain TEM elemental-mapping images. X-ray photoelectron spectroscopy (XPS) analysis was carried on ESCALAB 250xi photoelectron spectrometer equipped with an Al Kα monochromated x-ray source (Thermo Scientific, USA). N₂ adsorption–desorption isotherms were recorded on a Micromeritics ASAP 2460 gas sorption analyzer. The element content was analysed by Agilent 7800 inductively coupled plasma massspectrometry (ICP-MS). The molar concentration of the nanomaterials was measured by nanoparticle-tracking analysis system using Nanosight NS300.

Fe_x Cu_y Se NPs were prepared according to the previously reported procedure with some modification [26]. In brief, 10 ml of 80 mg ml⁻¹ PVP solution was firstly added into 60 ml water, which was followed by addition of Na₂SeO₃ solution (1 ml, 0.2 M) and AA solution (3 ml, 0.4 M) in sequence. After stirring for 15 min, CuSO₄·5H₂O (1 ml, 0.4 M) and AA solution (4 ml, 0.4 M) were added into the above red Se solution. The mixed system was allowed to react at room temperature under vigorous stirring until it turned dark green. Finally, FeCl₃·6H₂O (2 ml, 0.1 M) and L/D-His solution (3.5 ml, 0.229 M) were added into 50 ml Cu_2–x Se NPs. The L/D-Fe_x Cu_y Se NPs was obtained after reacting for 1 h with stirring.

The method of preparing L/D-Cys-Fe_x Cu_y Se NPs and L/D-Phe-Fe_x Cu_y Se NPs was same with the synthesis corresponding L/D-Fe_x Cu_y Se NPs while the L/D-His was replaced by L/D-Cys and L/D-Phe, respectively.

To characterize the peroxidase activity of Fe_x Cu_y Se NPs, H₂O₂ (10 mM) and TMB (0.42 mM) were added to 400 μl phosphate buffer saline (PBS) (different pH values) containing the Fe_x Cu_y Se NPs (25 μg ml⁻¹) and mixed immediately. After incubating for 15 min, the UV–vis absorption spectroscopy was used to measure the absorbance at 652 nm. All reactions were performed at room temperature, except for special instructions.

The kinetic measurements were carried out in time course mode by monitoring the absorbance change at 652 nm. Briefly, Cu_2–x Se NPs or Fe_x Cu_y Se NPs (15 μg ml⁻¹) was added to PBS (pH 4.0). The kinetic analysis was performed with H₂O₂ (10 mM) or TMB (0.42 mM) as the substrate and the total reaction volume was 400 μl. The Michaelis–Menten constant was calculated from the Lineweaver–Burk plot:

$$\begin{eqnarray*}&&1/V=\left({K}_{m}/{V}_{\max }\right)/\left[{\rm{S}}\right]+1/{V}_{\max },\end{eqnarray*}$$

where K_m is the Michaelis constant, V is the rate of conversion, V_max is the maximum rate of conversion, and [S] is the substrate concentration.

1 ml of L- or D-Fe_x Cu_y Se NPs (2 mg ml⁻¹) was dialyzed against a 10 ml solution containing a 1:1 mixture of L-dopa and D-dopa (5 mM). After 24 h dialysis, CD spectra was used to characterize the dialysate.

The kinetic measurements were carried out in time course mode by monitoring the absorbance change at 475 nm. Experiments were carried out using L/D-Fe_x Cu_y Se NPs (15 μg ml⁻¹) in a reaction volume of 400 μl PBS (pH 4.0) with 10 mM H₂O₂ and L/D-dopa as the substrate.

The K_m and V_max values were calculated from the Lineweaver–Burk plot. While, k_cat was calculated according to the equation:

$$\begin{eqnarray*}&&{k}_{cat}={V}_{\max }/\left[{\rm{E}}\right],\end{eqnarray*}$$

where [E] refers to the concentration of catalyst. The comparison of catalytic activity for different nanozymes was estimated according to their k_cat/K_m values [27].

All simulations were performed using the ORCA software [28]. Structure optimizations were carried out with density functional theory utilizing the well-known B3LYP hybrid functional in combination with the def2-SVP basis set. All structures were optimized with default parameters. The interaction energy was calculated according to the follow equation:

$$\begin{eqnarray*}&&{E}_{int}={E}_{tot}-{E}_{His}-{E}_{dopa}.\end{eqnarray*}$$

In order to optimize the experimental conditions for L-dopa oxidation, the pH value and incubation time were optimized, respectively. 4 μl L-Fe_x Cu_y Se NPs (1.5 mg ml⁻¹), 4 μl H₂O₂ (1 M) and 40 μl freshly prepared L-dopa solution (10 mM) were added into 352 μl PBS (different pH values). The mixture solution was incubated at room temperature for 30 min, followed by the spectral measurements. The absorption intensities at 475 nm were recorded. Furthermore, the same method was used to explore the influence of incubation time on the detection system.

To investigate the effect of chiral ligands on the enantioselectivity of chiral Fe_x Cu_y Se NPs towards dopa enantiomers, Cys and Phe were used as chiral ligands to prepare L/D-Cys-Fe_x Cu_y Se NPs and L/D-Phe-Fe_x Cu_y Se NPs. For the detection of dopa enantiomers, 4 μl H₂O₂ (1 M) and 40 μl freshly prepared L- or D-dopa solution (10 mM) were added into 356 μl PBS (pH = 4.0) containing different nanoparticles and mixed immediately. The mixture solution was then incubated for 30 min before the spectral measurements.

To detect L-dopa, L-Fe_x Cu_y Se NPs (15 μg ml⁻¹), H₂O₂ (10 mM) and freshly prepared L-dopa solution with different concentrations (0.005, 0.01, 0.025, 0.05, 0.075, 0.1, 0.125, 0.25, 0.5, 0.75, 1.0, 1.25, 1.5 and 2.0 mM) were added to 400 μl PBS solution (pH 4.0). After incubating the suspension for 30 min, UV–vis absorption spectrum was recorded at 475 nm.

To determinate L-dopa level in commercially available levodopa tablets, the standard addition method was used. Ten levodopa tablets were accurately weighed and ground into fine powder. The levodopa fine powder (40.39 mg) was then transferred into a 100 ml volumetric flask. PBS (pH 4.0, 100 ml) was added to dissolve the L-dopa. The solution was filtered with microporous membrane and the subsequent filtrate was detected. According to the L-dopa content (250 mg per tablet) provided by the supplier, the concentration of L-dopa in the filtrate was converted into 1.6865 mM after the above treatment.

Typically, L-Fe_x Cu_y Se NPs (15 μg ml⁻¹), H₂O₂ (10 mM) and 9.5 μl sample solution were mixed in PBS (pH 4.0) with the total volume of 400 μl. After incubating at room temperature for 30 min, the UV–vis absorption intensity at 475 nm was recorded.

Next, different amounts of L-dopa or D-dopa standard solution (0.04, 0.08, 0.36, 0.76 mM) was added to the sample solution respectively. The suspension was incubated for 30 min under the room temperature and then measured by UV–vis spectroscopy at 475 nm.

Fe_x Cu_y Se NPs were prepared according to the previously reported procedure with some modification [26]. L-His or D-His as the chiral ligand was conjugated to the surface of Fe_x Cu_y Se NPs through coordination interactions. As shown in TEM image, Fe_x Cu_y Se NPs with porous spherical morphology were monodispersed and had a uniform size of about 80 nm (figure 1(A)), which was quite similar with that of Cu_2–x Se NPs (figure S1 (available online at stacks.iop.org/NANO/33/135503/mmedia)). Moreover, Fe doping in Fe_x Cu_y Se NPs was validated by TEM elemental-mapping images and XPS. As shown in figure S2, the TEM elemental-mapping images indicated that Fe, Cu and Se were homogeneously distributed in Fe_x Cu_y Se NPs. Fe 2p, Cu 2p and Se 3d were observed in the XPS spectra of Fe_x Cu_y Se NPs, which further confirmed the successful doping of Fe into Fe_x Cu_y Se NPs (figure S3). In addition, the contents of Fe and Cu in Fe_x Cu_y Se NPs were determined by inductively coupled plasma mass spectrometry (ICP-MS), which corresponded to about 20.9 μg mg⁻¹ Fe_x Cu_y Se NPs and 95.1 μg mg⁻¹ Fe_x Cu_y Se NPs, respectively. The porous structure of Fe_x Cu_y Se NPs and Cu_2–x Se NPs was characterized by N₂ adsorption–desorption isotherms. As shown in figure S4, both Fe_x Cu_y Se NPs and Cu_2–x Se NPs exhibited a typical Type IV isotherm with a hysteresis loop, demonstrating their porous structure.

Zoom In Zoom Out Reset image size

The characteristic absorbance peak of Fe_x Cu_y Se NPs at approximately 850 nm was also observed in the UV–vis–NIR spectrum (figure 1(B)). The well-defined copper iron selenide (JCPDS no. 81-1959) in Fe_x Cu_y Se NPs was evidenced by x-ray powder diffraction (XRD) patterns, which was quite different from the initial berzelianite (JCPDS no. 06-0680) in Cu_2–x Se NPs (figure S5). The successful modification of L-His or D-His on the surface of Fe_x Cu_y Se NPs was validated by the appearance of the broad absorbance band at 3500–3000 cm⁻¹ in FTIR spectra, which belonged to the N–H stretching in the amino groups of His (figure 1(C)). Meanwhile, the typical bands of imidazole stretching at 1800–1400 cm⁻¹ were also observed (figure 1(C)). UV–vis–NIR spectra and XRD patterns indicated that introduction of L-His or D-His could not disturb the structure of Fe_x Cu_y Se NPs. All above characteristics demonstrated that the L-His or D-His was successfully conjugated onto the surface of Fe_x Cu_y Se NPs. CD was used to measure the chirality of L/D-Fe_x Cu_y Se NPs. As shown in figure 1(D), the CD signals of L/D-Fe_x Cu_y Se NPs showed equal in intensity but opposite in the ultraviolet region, further proofed of the grafting of His.

The peroxidase-mimicking activity of the nanozyme was evaluated via the oxidation of TMB by H₂O₂. As shown in figure 2(A), Fe_x Cu_y Se NPs can catalyze the oxidation of TMB into oxidized TMB (oxTMB) with a color change from colorless to blue in the presence of H₂O₂. The produced oxTMB exhibited a typical absorption peak at 652 nm. In contrast, the control experiments demonstrated that only H₂O₂ or Fe_x Cu_y Se NPs can hardly induce the color change, indicating that Fe_x Cu_y Se NPs possessed an intrinsic peroxidase-mimicking activity similar to HRP [15]. Like other peroxidase mimics [29, 30], the catalytic activity of Fe_x Cu_y Se NPs was deeply affected by pH value and the concentration of Fe_x Cu_y Se NPs (figures 2(B) and (C)). Based on the above experimental results and considering the practical application for drug analysis, pH value of 4.0, room temperature and 15 μg ml⁻¹ were chosen as the optimal pH, temperature and concentration of Fe_x Cu_y Se NPs for the subsequent experiments. Critically, for the achiral substrate TMB, L/D-Fe_x Cu_y Se NPs displayed the same catalytic effect as the unmodified Fe_x Cu_y Se NPs under the same conditions (figure 2(D)). This result manifested that the modification of His ligand had little effect on peroxidase-like activity of Fe_x Cu_y Se NPs.

Zoom In Zoom Out Reset image size

The synergistic effect of multi-metal mixing could enhance the catalytic activity of nanozymes [31, 32]. To demonstrate the critical role of Fe doping in the enhanced catalytic activity of Fe_x Cu_y Se NPs, TMB was employed as the substrate to monitor the catalytic activities of Cu_2–x Se NPs and Fe_x Cu_y Se NPs. As shown in figure S6, compared with Cu_2–x Se NPs, the catalytic activity of Fe_x Cu_y Se NPs enhanced significantly under the same conditions. The steady state kinetic study was also conducted to further investigate the peroxidase-like property of Fe_x Cu_y Se NPs [27]. As indicated in figure 3, within the certain range of H₂O₂ and TMB concentrations, typical Michaelis–Menten curves were obtained for both Fe_x Cu_y Se NPs and Cu_2–x Se NPs. The apparent Michaelis–Menten constant (K_m) and maximum initial velocity (V_max) were calculated from Lineweaver–Burk plot (table 1). The K_m values of Fe_x Cu_y Se NPs for TMB and H₂O₂ were close to that of HRP [15], indicating that Fe_x Cu_y Se NPs exhibited an excellent catalytic activity. It was worth noting that although the V_max values of Fe_x Cu_y Se NPs for both TMB and H₂O₂ were similar to that of Cu_2–x Se NPs, the K_m values of Fe_x Cu_y Se NPs were much smaller than that of Cu_2–x Se NPs, indicating that the porous structure and synergistic effects from the multi-metal mixing enhanced the binding affinity of Fe_x Cu_y Se NPs with H₂O₂ [29, 33–35]. All these phenomena contributed to the excellent catalytic activity of Fe_x Cu_y Se NPs, endowing them with the great potential to construct nanozyme-based colorimetric sensors.

Zoom In Zoom Out Reset image size

Based on the principles of chiral recognition, we explored the feasibility of detecting L/D-dopa by L/D-Fe_x Cu_y Se NPs-based system. L/D-Fe_x Cu_y Se NPs can catalyze the oxidation of dopa enantiomers with H₂O₂ to corresponding dopachrome (L/D-dc), which has a characteristic absorption at 475 nm (figure S7(A)). In addition, the experimental conditions including pH value and incubation time for dopa oxidation were optimized (figures S7(B) and (C)). In order to simultaneously obtain a higher catalytic activity and a shorter detection time in practical application, pH value of 4.0 and 30 min were chosen as the optimal pH and incubation time. As illustrated in figure 4(A), achiral Fe_x Cu_y Se NPs exhibited the same catalytic ability towards L/D-dopa. However, after conjugated with L/D-His, the chiral nanozymes displayed enantioselectivity in catalyzing the oxidation of dopa enantiomers. In particular, L-Fe_x Cu_y Se NPs had better catalytic activity for L-dopa than D-dopa, while, D-Fe_x Cu_y Se NPs showed reversed selectivity. All these results demonstrated that the enantioselectivity of L/D-Fe_x Cu_y Se NPs arose from the chiral His units. To further investigate the effect of chiral ligands on the enantioselectivity of chiral Fe_x Cu_y Se NPs towards dopa enantiomers, we also employed Cys and Phe as chiral ligands to prepare chiral L/D-Cys-Fe_x Cu_y Se NPs and L/D-Phe-Fe_x Cu_y Se NPs (figures S8(A) and (B)). As indicated in figures S8(C) and (D), L/D-Cys-Fe_x Cu_y Se NPs and L/D-Phe-Fe_x Cu_y Se NPs showed negligible enantioselectivity for catalyzing the oxidation of dopa enantiomers. Therefore, L/D-His were chosen as the chiral ligands for our subsequent experiments.

Zoom In Zoom Out Reset image size

The apparent kinetic parameters for dopa enantiomers based on the Michaelis–Menten equation were next measured to elucidate the different catalytic efficiency of the chiral nanozymes towards L-dopa and D-dopa (figure S9). As summarized in table 2, according to the k_cat/K_m values, the efficiency of L-His modified nanoparticle (L-Fe_x Cu_y Se NPs) towards L-dopa was 1.56 times higher than that of D-dopa. The higher k_cat/K_m values of L-Fe_x Cu_y Se NPs towards L-dopa arose from the relative lower K_m but higher k_cat values for L-dopa, indicating that L-Fe_x Cu_y Se NPs had a stronger binding affinity with L-dopa and higher catalytic activity for L-dopa oxidation. In contrast, D-dopa was the preferred oxidation substrate for D-Fe_x Cu_y Se NPs, which showed a selectivity factor of 1.49. The obtained selectivity factors were generally higher than or comparable with the records of other chiral nanozymes reported recently (table 3) [23, 36–38]. The relatively high catalytic activity and enantioselectivity of the L/D-Fe_x Cu_y Se NPs endowed them with great potential for constructing colorimetric sensor for detection of dopa enantiomers.

To support the enantioselective nature of the catalytic efficiency, dialysis experiments were performed to reveal the binding selectivity of these chiral nanozymes with L/D-dopa [6]. Racemic mixtures of dopa were obtained by mixing equimolar amounts of the pure enantiomers. The mixture was dialyzed against L-Fe_x Cu_y Se NPs or D-Fe_x Cu_y Se NPs, and CD was used to monitor the dialysate which indicated the enrichment of the enantiomer with lower binding affinity to L-Fe_x Cu_y Se NPs and D-Fe_x Cu_y Se NPs in the dialysis tube. As shown in figure 4(B), the dialysate was enriched in D-dopa and L-dopa, respectively. These results were consistent with the K_m values, which indicated unambiguously that L-Fe_x Cu_y Se NPs bound preferentially to L-dopa, while D-Fe_x Cu_y Se NPs showed a higher binding ability to D-dopa. All the results demonstrated that conjugation of chiral His units endowed Fe_x Cu_y Se NPs with stereoselectivity towards the oxidation of L/D-dopa and the stereoselectivity was associated with the chirality of His.

For further understanding the role of chiral ligand in the stereoselectivity of nanozymes, the molecular interactions between L/D-His and L/D-dopa were investigated using MD simulations [38]. As shown in figure 5, both L-dopa and D-dopa can bind with His enantiomers via hydrogen bonds. There were three hydrogen bonds formed in the interaction model between L-dopa and L-His. However, L-His formed only two hydrogen bonds with the D-dopa. In addition, the interaction energy obtained by ab initio calculations (E_int) was shown in table S1. The interaction of L-His with L-dopa was stronger than that of L-His with D-dopa. The results were consistent with the stereoselectivity of L-Fe_x Cu_y Se NPs towards L-dopa, which further indicated the important role of the chiral ligand played in enantiomer recognition and chiral selective catalysis.

Zoom In Zoom Out Reset image size

The high catalytic activity of L-Fe_x Cu_y Se NPs for L-dopa oxidation inspired us to explore the possibility of using L-Fe_x Cu_y Se NPs as peroxidase mimics to construct colorimetric sensor for detection of L-dopa. Figure 6(A) showed the response curves of the typical absorbance at 475 nm with various concentration of L-dopa. With the enhancement of the L-dopa concentration from 0 to 1 mM, the absorption intensity of L-dc at 475 nm gradually increased while the solution appeared to brown. There were two linear ranges for L-dopa. One was in the range of 5 μM–0.125 mM with a correlation coefficient of 0.998 (figure 6(B)), the other was from 0.125 mM to 1 mM with a correlation coefficient of 0.996 (figure 6(C)). The linear regression equations were A =0.270[L-dopa] (mM) + 0.069 and A = 0.098[L-dopa] (mM) + 0.093, respectively. When signal to noise was 3 [28], the detection limit (LOD) was calculated to be 1.02 μM. The low LOD and the wide detection range demonstrated the high sensitivity of L-Fe_x Cu_y Se NPs-based system for detecting L-dopa.

Zoom In Zoom Out Reset image size

To assess the feasibility for practical application, the proposed method was then applied to determine the concentrations of L-dopa in tablets. And the detection performance in real pharmaceutical samples was also investigated by using the standard addition method. As displayed in table 4, the concentration of L-dopa was determined to be 0.03999 mM, which was consistent well with the practical value (0.0400 mM) given by manufacturer. In addition, upon addition of certain amounts of L-dopa, the analytical recoveries from medicinal samples ranged within 92.97% to 107.25% and the relative standard deviations (RSD, n = 4) ranged from 2.04% to 9.63%, indicating that the proposed nanozyme-based sensor can be used for qualitative detection of L-dopa in real medicine with good recovery and precision.

Inspired by the enantioselectivity of the chiral nanozymes in oxidation of dopa enantiomers, we also examined whether the system can be used to detect L-dopa in the presence of D-dopa, which was of great significance for drug quality control since only L-dopa has pharmacological activity. As shown in table 4, when replacing L-dopa with the same amount of standard D-dopa solution, the recoveries of the racemic mixture were significantly lower than that of pure L-dopa samples, which was arose from the lower catalytic ability of L-Fe_x Cu_y Se NPs towards D-dopa. The outcomes exhibited an encouraging conclusion that the proposed chiral nanozyme-based sensor was promising for quality control of the clinical used drug, L-dopa.