HAuCl₄.3H₂O (≥49.0% Au basis), H₂PtCl₆.6H₂O (≥37.50% Pt basis), hydrogen peroxide (H₂O₂, 30%) and Methyl thiazolyl tetrazolium (MTT, 98%) were purchased from Sigma-Aldrich (Louis, MO, United States). Glucose was purchased from Macklin. Trifluoroacetic acid (CF₃COOH, 99%) was purchased from Alfa Aesar. The ATP assay kit, Enhanced BCA protein assay kit and Calcein/PI cell viability assay kit were from Beyotime Institute of Biotechnology, China. All chemical agents in this work were utilized without further purification. The Milli-Q water was obtained from the Milli-Q System.

The transmission electron microscope (TEM) images of the nanoparticles were obtained on an FEI Tecnai G2 F30 microscope. The crystalline phases of the materials were collected with X-ray powder diffraction (XRD; Rigaku Ultima Ⅳ), using copper Kɑ radiation (λ = 0.154056 nm) in the 2θ range of 20°–90° at a scan rate of 20°/min. Fourier transform infrared spectroscopy (FT-IR) spectra was measured by a BRUKER TENSOR Ⅱ in the range of 4,000–400 cm⁻¹. Dynamic light scattering (DLS) particle size analyzer (Malvern, United States) was used to measure the hydrophilic diameters of the particles. The concentration of Au and Pt was measured by Agilent Technologies 5100 inductively coupled plasma optical emission Spectrometry (ICP-OES). Shimadzu, AXIS SUPRA was used to detect X-ray photoelectron spectroscopy (XPS). The position of the C 1s peak at 284.8 eV was used as a calibration reference to determine the accurate binding energies. UV-vis absorption spectra was recorded on Shimadzu UV 3600 plus. The oxygen content in water was measured by dissolved oxygen tester JPSJ-605F, Leici China. The confocal laser scanning microscopy (CLSM) images were obtained using NIKON A1 R.

We added 5 ml of HAuCl₄.3H₂O (20 mmol) and H₂PtCl₆.6H₂O (20 mmol) in different volume ratios to 95 ml water, stirred it for 5 min until it was evenly mixed, and then heated the mixture to boiling. Then, 10 ml sodium citrate aqueous solution with a concentration of 20 mg/ml was added to the boiling liquid and stirred for 10 min. Next, turned off the heat and continued to stir vigorously to room temperature. AuPt alloys with different proportions were obtained by centrifugal washing. Au NPs and Pt NPs were synthesized by similar methods. The composition of AuPt alloys was determined by ICP-OES.

The catalytic performance of the alloys was characterized by the formation of gluconic acid. 50 μL AuPt alloys (2 mg/ml) and 950 μL glucose (1 M) were mixed evenly for 4 h to verify the effect of AuPt alloys on glucose catalytic oxidation. After the reaction, AuPt alloys were removed by centrifugation to obtain supernatant containing gluconic acid.

Solution A (5 mM EDTA and 0.15 mM triethylamine aqueous solution), solution B (3 M hydroxylamine, NH₂OH), and solution C (1 M HCl, 0.1 M FeCl₃ and 0.25 M CCl₃COOH aqueous solution) were respectively prepared. Then, 250 μL solution A and 25 μL solution B were successively added to the clear liquid. After 25 min of incubation, 125 μL solution C was added, and the mixture was evenly mixed for 5 min before spectral test (Huang et al., 2017).

10 μL AuPt alloys (2 mg/ml) were mixed with an aqueous solution of H₂O₂ (10 mM), with a total volume of 5 ml, and the amount of oxygen produced by the decomposition of H₂O₂ was recorded by a dissolved oxygen recorder (Luo et al., 2021). All the reactions were carried out in deoxidized water at 37 °C.

In order to evaluate the cell compatibility of AuPt alloys, PC12 cells were seeded into 96-well cell culture plates at 1×10⁴/well for 24 h. Then Au_0.75Pt_0.25 alloys with different concentrations were added into 96-well plates and cultured for 24 h. After the culture medium was sucked out, 100 μL MTT (phosphate buffered saline (PBS) buffer solution 0.6 mg/ml) was added to each well and cultured for 4 h. Finally, removed the MTT solution, added 100 μL dimethyl sulfoxide (DMSO), and incubated for 10 min. The absorbance of the solution at 490 nm was measured with a microplate reader.

PC12 cells were inoculated in 1 ml CLSM dish at 1×10⁵/ml and incubated overnight to ensure firm cell adhesion. The cells were incubated with AuPt alloys of different concentrations for 24 h, and then the cells in the corresponding groups were dyed alive and dead with Calcein (green) and PI (red).

PC12 cells were inoculated at a density of 5 × 10⁵ cells/well and incubated in 6-well plates for 12 h. Then AuPt alloys with different proportions were added and incubated for 24 h. After that, the cells were lysed with ATP lysis buffer, and the supernatant was collected by centrifugation. The protein concentration of each sample was determined by BCA protein concentration kit, and ATP concentration was detected by ATP detection kit (Liu et al., 2021).

AuPt alloys with different proportions were prepared by one-step reduction of AuCl₄ ⁻ and PtCl₆ ²⁻ with sodium citrate in water. As shown in Figure 1A, TEM images showed that these AuPt alloys with different proportions were uniformly spherical. High resolution TEM (HRTEM) images and the corresponding FAST Fourier transform (FFT) further revealed the lattice plane of the Au_0.75Pt_0.25 alloys (Supplementary Figure S1). The calculated lattice spacing of the (111) plane of the alloys was 0.230 nm, which was between the lattice spacing of the pure Au (111) plane (0.235 nm) and that of the pure Pt (111) plane (0.228 nm). The lattice spacing in the middle proved the formation of AuPt alloys (He et al., 2017).

In addition, the average hydrodynamic diameters of AuPt alloys synthesized with different AuPt ratios by DLS test were approximately 10–20 nm and had good dispersibility (Figure 1B) (Zhang et al., 2019b). The stability of the Au_0.75Pt_0.25 was further evaluated. TEM images showed that the morphological characteristics of Au_0.75Pt_0.25 did not change significantly after 3 days of soaking in PBS (Supplementary Figure S2), indicating that Au_0.75Pt_0.25 had excellent stability. The Zeta potential was about -30 mV due to the -COOH in sodium citrate on the surface of the AuPt alloys (Figure 1D). Supplementary Figure S3 showed the characteristic peaks of sodium citrate, such as –C=O characteristic absorption peak at 1,600 cm⁻¹ and -CH₂ shear vibration peak at 1,400 cm⁻¹, indicating the successful preparation of sodium citrate stabilized AuPt alloys (Sanches et al., 2011; Zhang et al., 2018).

It was well known that the lattice of two metals would change after they formed alloys, and their diffraction patterns would change correspondingly according to the basic principle of crystal diffraction. Therefore, XRD pattern was used to further confirm the formation of AuPt alloys. Figure 1C showed XRD patterns of alloys with different proportions. Diffraction peaks of Au, Pt and AuPt alloys with different AuPt ratios indicated that the prepared samples all had face-centered cubic (fcc) phases. According to Figure 1C, the 2θ peaks (111) and (200) of AuPt NPs fell between the corresponding 2θ peaks of Au and Pt NPs (Liu et al., 2011). With the change of AuPt ratios, the positions of these peaks changed gradually, indicating the change of alloying degree. UV-vis absorption spectrum (Supplementary Figure S4) also confirmed the formation of AuPt alloys. With the gradual addition of Pt component, the typical absorption of Au NPs at 520 nm due to surface plasmon resonance would gradually disappear. The vanishing peak also represented the gradual formation of AuPt alloys (Zhao et al., 2014).

The relative abundance of Au and Pt in AuPt alloys detected by ICP and XPS was shown in Supplementary Table S1. Because the standard redox potential E⁰ for AuCl₄ ⁻/Au⁰ couple (+0.99 V) was higher than that of PtCl₆ ²⁻/Pt⁰ couple (+0.74 V), the Au was reduced first (De and Rao, 2005). The percentage of Au element in the alloys obtained from ICP was slightly higher than that in the raw solution, and the results of XPS measurement further confirmed that Pt element was enriched on the surface of the alloys.

Since the prepared AuPt alloys included different catalytic properties, in which Au components possessed the intrinsic GOx-like activity (Luo et al., 2010), while Pt components owned the intrinsic CAT-like activity [Figure 2A(1)] (Fan et al., 2011), it was expected that AuPt alloys could catalyze the cascade reaction of glucose oxidation [Figure 2A(2)]. To prove it, the GOx-like activity of AuPt alloys was first investigated. AuPt alloys in different proportions reacted with O₂ in the reaction solution to catalyze the oxidation of glucose, resulting in H₂O₂ and gluconic acid. The reaction solution was centrifuged to remove AuPt alloys, and the supernatant obtained contained H₂O₂ and gluconic acid. In a typical experiment, with gluconic acid as substrate, NH₂OH and Fe³⁺ were successively added to the solution, and the reaction of glucose to gluconic acid catalyzed by AuPt alloys were detected by a colorimetric assay. As shown in Figure 2B, the apparent absorbance band at 450–550 nm confirmed that gluconic acid was the product of glucose oxidation catalyzed by AuPt alloys. Next, we made statistics on the absorbance value at 505 nm (Figure 2C), and it could be seen that Au_0.75Pt_0.25 had the best catalytic effect compared with other alloys and physical mixing group. The above experimental results confirmed the GOx-like activity of the AuPt alloys.

Then, the CAT-like activity of AuPt alloys was studied. Since it could catalyze H₂O₂ to produce H₂O and O₂, we explored the catalytic process of AuPt alloys for H₂O₂ by testing the content of dissolved oxygen. A large number of bubbles were observed in the tubes containing H₂O₂ after the addition of AuPt alloys in different proportions (except pure Au NRs), demonstrating that Pt in AuPt alloys could play a CAT-like effect (Figure 2D). Next, we used a solution-oxygen meter to further monitor the oxygen generation process and found that Au_0.75Pt_0.25 catalyzed the decomposition of H₂O₂ to generate oxygen at the fastest rate.

As mentioned above, AuPt alloys had GOx-like and CAT-like activities, and different AuPt ratios exhibited different nanozymes catalytic activities, among which Au_0.75Pt_0.25 had the best catalytic effect.

To further explore the reasons for the enhancement of the catalytic activity of AuPt alloys as nanozymes, we conducted the following analysis. We realized that it has been reported that the electronic structure of metal nanoparticles played a critical role in their catalytic activity (Duchesne et al., 2018). Therefore, we investigated whether the alloying of Au and Pt would change the electronic structures of Au and Pt components, thus affecting the catalytic performance (Figure 3A). The electron binding energies of XPS for Au and Pt in AuPt alloys with different proportions were used to reflect the charge distribution of AuPt in alloys (Supplementary Figure S5). Among them, in AuPt alloys system, the peak between 60 and 90 eV corresponded to Pt 4f_7/2, Pt 4f_5/2 and Au 4f_7/2, Au 4f_5/2, respectively (Figure 3B) (Yamauchi et al., 2012). As shown in Figure 3C, the binding energy of Au 4f in Au NPs was 83.7 and 87.4 eV, respectively. With the addition of Pt component, the 4f binding energy of Au component increased in AuPt alloys (Figures 3D–F). However, the 4f binding energy of Pt component in Pt NPs was 72.8 and 76.1 eV (Figure 3J), and the binding energy shifted to lower with the decrease of Pt component content (Figures 3G–I). The above analysis results proved that with the formation of AuPt alloys, the electron on Au component in AuPt alloys transfered to Pt component, resulting in lower electron density of Au component in AuPt alloys than Au NPs, and higher electron density of Pt component than Pt NPs.

Based on the above interesting results, we provided a theoretical basis for the change of electronic structure of AuPt alloys with different components. For bimetallic alloys, the work function of different metals was the decisive factor affecting the regulation of electronic structure. Then, for AuPt alloys, the work functions of Au and Pt were 5.54 and 6.13 eV, respectively. The difference in the work function between Au and Pt components caused the electrons on Au surface to be transfer towards Pt. Coincidentally, in the current study on the mechanism of Au nanozymes catalyzing glucose oxidation, Au components needed to maintain positive valence, and Pt surface was rich in electrons, which was conducive to its catalase properties. Therefore, electron transfer due to work function would synergistically enhance the effect of the two components of AuPt as nanozymes.

In order to be applied in subsequent biological studies, we evaluated the biological activity of AuPt alloys. MTT assay (Figure 4B) was used to verify the toxicity of PC12 cells co-cultured with AuPt alloys, and it was found that AuPt alloys had no obvious toxicity for cells. Subsequently, the survival of cells co-cultured with AuPt alloys for 24 h was observed in confocal images by live/dead staining experiment. Cells were stained with calcein/PI, in which living cells were stained green with calcein and dead cells were stained red with PI. Confocal images (Figure 4A) could be observed that AuPt alloys materials with different concentrations had no obvious killing effect on cells, which also facilitated the application of AuPt alloys for further research.

Encouraged by the high catalytic performance of AuPt alloys in aqueous solution for glucose and H₂O₂, we carried out cell level experiments to observe the catalytic effect of AuPt alloys. We verified whether AuPt alloys could affect the cellular metabolism process because glucose was the main energy source for cell activity. Intracellular ATP levels were measured using ATP detection kits (Supplementary Figures S6, S7). As shown in Figure 4C, by catalyzing glucose oxidation, AuPt alloys could effectively reduce intracellular ATP levels by affecting energy supply. And it was worth noting that Au_0.75Pt_0.25 also had the best catalytic effect at the cellular level. Therefore, AuPt alloys, which affected cell energy metabolism, were considered as a treatment for glucose-related diseases.