Dual model biosensor integrated with peroxidase-like activity and self-assembly for uric acid detection

Abstract

Uric acid (UA), the final product of purine metabolism, is a crucial biomarker for gout diagnostics and highly related to various metabolic diseases. Precise detection of UA levels in serum and urine enables disease diagnosis and guides treatment. Combining the advantages of colorimetry and laser desorption/ionization mass spectrometry (LDI-MS), we developed a dual-model biosensor based on hollow Cu₂O@Au nanocubes (h-Cu₂O@Au NCs) for UA detection. The h-Cu₂O@Au NCs demonstrated excellent peroxidase (POD)-like activity and were used to rapidly detect UA by colorimetric assay, with a linear range of 0.05–2 mM and limit of detection (LOD) of 35.71 μM. Moreover, the h-Cu₂O@Au NCs achieved enrichment and detection of UA via the liquid–liquid interface self-assembly-assisted LDI-MS, with a linear range of 0.01–0.5 mM, LOD of 15.6 μM, and reproducibility of <5%. In view of its advantages, the dual-model nanoplatform based on h-Cu₂O@Au NCs achieved UA detection in serum samples by colorimetry assay and in urine samples by LDI-MS, obtaining results consistent with the commercial UA assay kit (72–511 μM for serum, R² = 0.956 and 2–9 mM for urine, R² = 0.876), presenting potential in the rapid and sensitive detection of UA in clinic.

---

Yuning Wang

Yuning Wang received her PhD in Analytical Chemistry from Fudan University in 2021 under the supervision of Prof. Baohong Liu. Currently, she is an assistant professor at the School of Biomedical Engineering, Shanghai Jiao Tong University. Her research interests focus on the development of nanoplatforms for biomolecule detection.

Introduction

Uric acid (UA) is the final product of purine metabolism in the human body,¹ but abnormal UA levels can induce gout, gouty arthritis, renal disease, and cardiovascular-related diseases.² Common diseases associated with elevated UA are characterized by hyperuricemia (UA > 385 μM);³ individuals with UA ≥ 535 μM have a 4.5% annual incidence of gout, compared to 0.5% for those between 415–530 μM.⁴ When UA in serum is saturated, the excess UA can be excreted from the body through the urine, maintaining the metabolic homeostasis in the human body.⁵ Moreover, UA in serum and urine has served as a significant biomarker for disease diagnosis and reflects the state of specific diseases.⁶ Therefore, measurement of UA in serum and urine is significant for the clinical diagnosis of related diseases.

To date, several techniques have been applied in UA detection, including colorimetry, fluorescence, electrochemistry, and mass spectrometry.⁷ In view of the need for convenience, economy and precision, colorimetry based on enzymes has attracted much attention.⁸ With the development of nanomaterials, nanozymes, as excellent enzyme mimics, present advantages in clinical medicine, due to their low cost, high stability, ease of preparation, and lower susceptibility to inactivation.⁹ Currently, nanozyme-based colorimetry has been successfully used in the detection of biomarkers, such as H₂O₂,¹⁰ glucose,¹¹ and UA,¹² and has been further advanced through structural innovations, including multi-enzyme platforms and encoded sensing systems.^13,14 Nevertheless, colorimetry faces a challenge in the analysis of low-abundance biomarkers, due to its limited sensitivity. Nanoparticle-assisted laser desorption/ionization mass spectrometry (NALDI-MS) offers a sensitive and accurate technique in low-abundance molecule analysis, with the advantages of high throughput, low sample consumption, and rapid detection speed.¹⁵ Various inorganic nanomaterials, including semiconductors, noble metal nanoparticles (NPs), other metal and metal oxide NPs, and silicon- and carbon-based nanostructures, have been explored as matrix materials to enhance LDI-MS performance.¹⁶ However, the uneven distribution between nanoparticles and the analyte inevitably affects reproducibility.¹⁷ In this context, liquid–liquid interface self-assembly, as a bottom-up strategy, provides a promising way to fabricate orderly nanoarrays to assist LDI-MS analysis with high reproducibility and accuracy.¹⁸ During the self-assembly of nanomaterials, the analyte could be adsorbed on the surface of the nanomaterial via specific interaction for direct detection.¹⁹ Given this, it is meaningful to develop a dual-model biosensor integrated with colorimetry and LDI-MS for rapid and accurate detection of UA. The primary concern is finding a nanostructure possessing excellent catalytic activity and high LDI enhancement capability.

Cuprous oxide nanoparticles (Cu₂O NPs), a p-type semiconductor, have attracted much attention in various fields, such as catalysis,²⁰ sensing,²¹ and solar energy.²² Their peroxidase (POD)-like activity has been shown to be superior to that of CuO and Fe₃O₄ NPs due to the more efficient H₂O₂ reduction catalyzed by Cu(I).²³ Compared to pure Cu₂O NPs, hybrid nanostructures of noble metal NPs (such as Au-, Pt-, Pb-) and Cu₂O NPs exhibit higher catalytic activity,²⁴ showing promise in being applied as nanozymes. The loaded noble metal NPs present unique properties, such as strong catalytic activity and localized surface plasmon resonance (LSPR), which could inhibit the electron–hole complexation as electron traps.^25,26 Meanwhile, the hollow structure also contributes to enhancing the catalytic activity, owing to electron and hole separation.^27,28 Moreover, hybrid Cu₂O nanomaterials possess characteristics such as a large specific surface area and rough surface, which can lead to nonspecific adsorption of small-molecule analytes,²⁹ making them promising for as an excellent LDI-MS matrix. Therefore, a hybrid Cu₂O nanomaterial with high performance is of great potential in the detection of biomarkers.

In this work, we constructed a hollow Cu₂O@Au nanocube (h-Cu₂O@Au NC)-based biosensor for dual-model UA detection, integrating POD-like activity for colorimetry and self-assembly for LDI-MS (Scheme 1). The h-Cu₂O@Au NCs were prepared via a simple redox approach and were verified to have POD-like activity, which was further utilized to detect UA using colorimetry with the auxiliary of uricase. Moreover, owing to its rough surface, large specific surface area, and positive charges, h-Cu₂O@Au NCs could adsorb some small molecules with negative charges, such as UA. Thus, h-Cu₂O@Au NCs self-assembled with UA at oil–water interfaces into a two-dimensional nanofilm for the LDI-MS analysis of UA. To this end, the dual-model h-Cu₂O@Au NCs-based biosensor achieved colorimetric detection of UA in serum and LDI-MS detection of UA in urine. The results obtained from the dual-model biosensor were consistent with the biochemistry assay, demonstrating that our developed technique could be reliable in UA detection for clinical diagnosis.

Scheme 1 The schematics of the dual-model UA biosensor based on h-Cu₂O@Au NCs for colorimetric detection and LDI-MS analysis. (a) Synthesis of h-Cu₂O@Au NCs via galvanic replacement reaction. (b) The h-Cu₂O@Au NCs with POD-like activity for colorimetric detection of UA. (c) The h-Cu₂O@Au NCs self-assembled with UA for LDI-MS analysis of UA.

Results and discussion

Characterization of h-Cu₂O@Au NCs

The h-Cu₂O@Au NCs were prepared by a one-pot method using L-ascorbic acid as a reducing agent under alkaline condition. Then, by adding HAuCl₄ solution, the h-Cu₂O@Au NCs were obtained. The morphology of h-Cu₂O@Au NCs was characterized using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). As shown in Fig. 1a and b, the SEM images of the products illustrated the uniform distribution of AuNPs on the surface of Cu₂O NCs, showing the Cu₂O NC-like shape with rough surface nanostructure. The rough surface of the nanostructure provided a large specific surface area, which was conducive to the non-specific adsorption of small-molecule analytes. The TEM images (Fig. 1c and Fig. S1a, ESI†) exhibit h-Cu₂O@Au NCs with hollow core structure, which was formed by HAuCl₄ etching. In the progress of reaction, the hole increasingly expanded and AuNPs were deposited on the surface of Cu₂O NCs, forming the h-Cu₂O@Au NCs with the diameter of about 500 nm. From the elemental mapping analysis shown in Fig. 1d, the Cu and Au elements were homogeneously distributed on the surface of the h-Cu₂O@Au NCs, demonstrating that h-Cu₂O@Au NCs were successfully synthesized. According to the ultraviolet-visible (UV-vis) absorbance spectra in Fig. S1b (ESI†), the absorption of h-Cu₂O@Au NCs covers the incident wavelength at 355 nm of the Nd:YAG laser used in LDI-MS and mitigated background interference of Cu₂O at 652 nm for application in colorimetric detection.³⁰ Furthermore, the typical high-resolution transmission electron microscopy (HRTEM) image in Fig. 1e displays the nanostructure with the lattice fringe spacings of 2.42 Å (Cu₂O, [111]) and 2.49 Å (Au, [111]).³¹ The clearly visible lattice fringes in the image indicated good crystallinity of the h-Cu₂O@Au NCs, suggesting that they were stable enough as the LDI-MS matrix.³² Moreover, due to the separation of electrons and holes, h-Cu₂O@Au NCs are positively charged, and they could easily adsorb some negatively charged small-molecule analytes via electrostatic interaction, such as UA (Fig. 1f).

Fig. 1 Characterization of Cu₂O and h-Cu₂O@Au NCs. (a) Scanning electron microscopic images (SEM) of Cu₂O NCs. (b) SEM and (c) transmission electron microscopy (TEM) images of the h-Cu₂O@Au NCs. (d) Element mapping images of h-Cu₂O@Au NCs, containing Cu (green), O (yellow), and Au (red). (e) High-resolution TEM (HRTEM) image of the h-Cu₂O@Au NCs. The inset shows the selected-area electron diffraction pattern. (f) Zeta potential of Cu₂O NCs (gray), h-Cu₂O@Au NCs (black) and h-Cu₂O@Au NCs@UA (red). Each error bar represents the standard deviation calculated from three replicates.

The POD-like ability of h-Cu₂O@Au NCs

The above characterization results confirmed that h-Cu₂O@Au NCs have unique structure and positive charge, facilitating their catalytic activity. Due to the catalytic action of uricase, UA could be converted into H₂O₂ and allantoin. Therefore, H₂O₂ could further oxidize 3,3′,5,5′-tetramethylbenzidine (TMB) to the oxidation product oxTMB under catalysis of the POD-like nanozyme, thus indirectly detecting UA (Fig. 2a). In view of this, we further explored the inherent POD-like activity of h-Cu₂O@Au NCs, and TMB was selected as the chromogenic substrate for colorimetric determination.³³ As shown in Fig. 2b, compared with pure Cu₂O NCs and AuNPs, we observed that h-Cu₂O@Au NCs catalyzed oxTMB with the highest absorption peak at 652 nm in the presence of H₂O₂. In the control experiment without any nanomaterial, the absorbance at 652 nm did not change significantly. These results demonstrated that h-Cu₂O@Au NCs had prominent POD-like activity, superior to that of pure Cu₂O NCs and AuNPs. The excellent POD-like activity of h-Cu₂O@Au NCs could be attributed to the hollow structure and AuNPs loaded on their surface, which prompt the separation of electrons and holes and provide more catalytic sites.³⁴

Fig. 2 The POD-like activity of h-Cu₂O@Au NCs. (a) Schematic illustration of the catalytic system using h-Cu₂O@Au NCs. (b) UV-vis absorbance spectra of different systems, including TMB + H₂O₂ (black curve), AuNPs + TMB + H₂O₂ (red curve), Cu₂O + TMB + H₂O₂ (blue curve), and h-Cu₂O@Au NCs + TMB + H₂O₂ (green curve). (c) Reaction velocity (v) and time-dependent absorbance changes at 652 nm (inset) for the TMB + H₂O₂ system in the presence of h-Cu₂O@Au NCs in acetate buffer (10 mM, pH = 4) containing H₂O₂ with different concentrations and TMB (10 mM). (d) Reaction velocity (v) and time-dependent absorbance changes at 652 nm (inset) for the TMB + H₂O₂ system containing TMB with different concentrations and H₂O₂ (100 mM) in the above same condition. (e) The UV-vis absorption spectra for the detection of UA with different concentrations (0–2 mM). (f) A linear relationship between absorbance at 652 nm and the concentration of UA (0.05–2 mM). (g) Selectivity of the h-Cu₂O@Au NC-based colorimetry biosensor for UA detection. (h) Quantitative analysis of UA in serum samples using both the h-Cu₂O@Au NC-based colorimetric platform (purple) and biochemical assay (orange). (i) The linear correlation between UA concentration in serum obtained from the h-Cu₂O@Au NC-based colorimetric biosensor and the commercial UA assay kit (R² = 0.956). Each error bar represents the standard deviation calculated from three replicates.

To further investigate the effects of hollow structures and AuNPs on catalytic performance, we synthesized h-Cu₂O NCs and Cu₂O@Au NCs formed by loading AuNPs on the surface of Cu₂O NCs via electrostatic adsorption^24,35 as comparison. The characterization results of the above nanostructures are shown in Fig. S2 and S3 (ESI†), including the SEM, TEM and elemental mapping analysis. The h-Cu₂O NC had a larger specific surface area and more cavities, while the Cu₂O@Au NC exhibited a more homogeneous distribution of AuNPs. Then, we compared the POD-like activities of h-Cu₂O@Au NCs, h-Cu₂O NCs, and Cu₂O@Au NCs. As shown in Fig. S4 (ESI†), we found that h-Cu₂O@Au NCs exhibited enhanced efficiency in catalyzing the oxidation reaction of TMB (oxTMB) in the presence of H₂O₂. This nanocomposite presented the highest absorption peak at 652 nm, surpassing the catalytic activities of both Cu₂O@Au NCs and h-Cu₂O NCs. These results indicated that the synergistic effect of hollow structure and the loading of AuNPs significantly enhanced the catalytic activity of the nanozyme.

In order to obtain the best catalytic performance, we optimized the reaction conditions, including pH, the concentration of h-Cu₂O@Au NCs, the concentration of TMB, and incubation time. As shown in Fig. S5a (ESI†), the POD-like activity of h-Cu₂O@Au NCs was pH-dependent, exhibiting the highest catalytic activity at pH 4. With the concentration of TMB increasing from 0 to 6 mM, we found the absorbance at 652 nm increased gradually, reaching its maximum in the presence of 2 mM TMB (Fig. S5b, ESI†). When excess amount of TMB was further added, the catalytic sites could be blocked, resulting in the decline of catalytic activity. As the concentration of h-Cu₂O@Au NCs increased from 0.2 mg mL⁻¹ to 4 mg mL⁻¹, the amount of oxTMB reached its maximum with 0.6 mg mL⁻¹ of h-Cu₂O@Au NCs (Fig. S5c, ESI†). Meanwhile, as shown in Fig. S5d (ESI†), with the increase of incubation time, the absorbance at 652 nm increased quickly. After incubating for 30 min, the absorbance increased slowly, indicating the complete oxidization of TMB. Taken together, the above optimal parameters were selected for further applications.

Under the optimized conditions, steady-state kinetics assays were conducted to characterize the POD-like ability of h-Cu₂O@Au NCs by varying the substrate concentrations of TMB or H₂O₂. First, time-dependent absorbance at 652 nm was monitored while maintaining a TMB concentration of 10 mM and altering the H₂O₂ concentration in acetate buffer (Fig. 2c). Meanwhile, when TMB served as the substrate, the time-dependent absorbance at 652 nm was monitored with fixed H₂O₂ concentration of 100 mM and varying TMB concentrations (Fig. 2d). In this reaction system, by applying the Beer–Lambert law (A = εlc) and Lineweaver–Burk plot, we could obtain the Michaelis–Menten constant (K_m) and maximum velocity (V_max). The K_m value was 336.57 mM and V_max was 39.10 × 10⁻⁸ M s⁻¹ for H₂O₂ as the substrate, while the K_m value was 5.23 mM and V_max was 18.08 × 10⁻⁸ M s⁻¹ for TMB as the substrate, demonstrating the excellent catalytic activity of h-Cu₂O@Au NCs in comparison with other catalysts (Table S1, ESI†).

The h-Cu₂O@Au NC-based colorimetric biosensor for UA detection

In view of the high catalytic activity of h-Cu₂O@Au NCs as a POD mimetic, we constructed a colorimetric biosensor based on h-Cu₂O@Au NCs and uricase for UA detection. With the catalysis of uricase, UA could be oxidized by O₂, generating allantoin and H₂O₂. Meanwhile, H₂O₂ could oxidize TMB to oxTMB, and the color changes to blue in the presence of h-Cu₂O@Au NCs. Based on this principle, the concentration of UA could be measured by recording the variation of UV-vis absorbance at 652 nm. Under the above optimal conditions, the solution changed from colorless to blue and the absorbance of 652 nm increased when the concentration of UA increased (Fig. 2e). As shown in Fig. 2f, when the concentration of UA ranged from 0.05 mM to 2 mM, a linear relationship was obtained between the concentration of UA and the absorbance value at 652 nm (y = 0.357x + 0.141, R² = 0.996), with a limit of detection (LOD) of 35.71 μM.³⁶ This LOD is well below the normal UA concentration range in human serum (150–420 μM), ensuring accurate quantification in clinical samples.³⁷ To evaluate the selectivity of this biosensor for UA detection, we selected various small-molecule metabolites as control, including glutamic acid, lysine, urea, dopamine, glucose, L-ascorbic acid, and guanine (five-fold the concentration of UA). As shown in Fig. 2g, the colorimetric response of these interferents was at least 20-fold lower than that of UA, demonstrating that this biosensor had high selectivity for UA detection. Encouraged by the above results, we further applied the h-Cu₂O@Au NC-based colorimetric biosensor to detect UA in clinical serum samples (n = 13) and obtained results consistent with the commercial UA assay kit (R² = 0.956, Fig. 2h and i). Moreover, the h-Cu₂O@Au NCs, as a POD mimetic enzyme, showed excellent long-term stability when it was used in UA detection over a continuous 14 days, as well as good reusability, retaining over 60% of their catalytic activity after six cycles (Fig. S6, ESI†).

Fabrication of the self-assembled h-Cu₂O@Au NC (SA-Cu₂O@AuNC)-based LDI-MS platform

Interfacial self-assembly is a simple, convenient and cheap strategy to build plasmonic NPs into highly ordered and closely packed two-dimensional nanostructures.³⁸ The highly stable, reproducible, and uniform substrates efficiently enhance electromagnetic fields by reducing the interparticle distances, which have been used in various fields, such as catalysis,³⁹ surface-enhanced Raman spectroscopy (SERS), and LDI-MS.⁴⁰ Given this, we self-assembled h-Cu₂O@Au NCs into a closely packed nanoarray (SA-Cu₂O@Au NCs) at the oil–water interface as LDI-MS substrates. In the process, h-Cu₂O@Au NCs with positive charge and large specific surface area could capture negatively charged small molecules on the surface for direct detection, such as UA.

To demonstrate the feasibility of our SA-Cu₂O@Au NC-assisted LDI-MS platform, UA was chosen as a model. As shown in Fig. 3a, after adding UA to the system, UA could be adsorbed on the surface of h-Cu₂O@Au NCs, as demonstrated in Fig. 1f and Fig. S7 (ESI†), generating electrostatic shielding. After vigorous shaking, the UA-trapped h-Cu₂O@Au NCs could self-assemble into a closely packed nanofilm at the oil–water interface owing to the decreased interfacial energy, forming brilliant nanofilm (SA-Cu₂O@Au NCs). The SA-Cu₂O@Au NCs were further transferred onto a silicon chip to assist in the LDI-MS analysis of UA. As shown in Fig. 3b, we clearly observed the MS peak of UA at m/z of 168 ([UA + e]⁻). To explore the universality of our platform, we further selected a series of metabolites with negative charge, including polyfluorooctane sulfonic acid (PFOs), aspartic acid, glutamic acid, and L-ascorbic acid. With a similar method, we successfully identified the above metabolites and obtained the corresponding MS peaks at m/z of 499 ([PFOs-H]⁻), 132 ([aspartic acid-H]⁻), 146 ([glutamic acid-H]⁻), and 175 ([ascorbic acid-H]⁻), respectively, demonstrating that the SA-Cu₂O@Au NC-based LDI-MS platform could efficiently adsorb and detect various metabolites with negative charges (Fig. 3c–f).

Fig. 3 Fabrication of self-assembled h-Cu₂O@Au NCs (SA-Cu₂O@Au NCs) and SA-Cu₂O@Au NC-assisted LDI-MS performance. (a) Schematic illustration and photos of h-Cu₂O@Au NCs self-assembled with UA at the oil–water interface. (b)–(f) The typical MS spectrum of small molecules analyzed by SA-Cu₂O@Au NC-assisted LDI-MS, including UA, PFOs, aspartic acid, L-ascorbic acid, and glutamic acid (1 mM). (g) MS signal intensities of the UA standard sample (1 mM) using SA-Cu₂O@Au NC (purple) and non-assembled h-Cu₂O@Au NC (orange)-based LDI-MS; each error bar represents the standard deviation calculated from three replicates. (h) Bright field micrographs (left) and corresponding 3D reconstruction images (right) of co-crystallization between urine samples and SA-Cu₂O@Au NCs (bottom) or non-assembled h-Cu₂O@Au NCs (upper). (i) and (j) MS signal intensities of 500 μM UA at m/z of 168 recorded from (i) four different spots on one chip and (j) eight chips at random positions by SA-Cu₂O@Au NCs.

Although h-Cu₂O@Au NCs could serve as LDI-MS matrix for the direct detection of these small molecules, notably, the SA-Cu₂O@Au NC-based LDI-MS platform significantly enhanced the MS signal by about 3–4 folds in comparison with using the h-Cu₂O@Au NCs as matrix (Fig. 3g). Moreover, due to the inevitable “coffee ring” effect during analyte localization, the reproducibility of the MS signal was a key issue to be studied.⁴¹ Considering this, 3D laser scanning microscopy was used to examine the homogeneity of co-crystallization of the SA-Cu₂O@Au NCs with samples. As shown in Fig. 3h, the samples were uniformly distributed on the surface of SA-Cu₂O@Au NCs, with a surface roughness (S_a) of 0.385 μm, which was much lower than the non-assembled h-Cu₂O@Au NCs as matrix (S_a = 0.751 μm). The results showed that SA-Cu₂O@Au NCs with uniformly distributed nanostructures efficiently weakened the “coffee ring” effect to a great extent, which was beneficial to improve the reproducibility of MS signals in LDI-MS analysis.⁴² In order to evidence the excellent reproducibility of the SA-Cu₂O@Au NC-assisted LDI-MS, we recorded the mass spectra of UA from eight different SA-Cu₂O@Au NC chips (one spot per chip). The relative standard deviation (RSD) of MS intensity at m/z of 168 ([UA + e]⁻) was 3.9%, which was about 5-folds superior to non-assembled h-Cu₂O@Au NCs (Fig. 3i and Fig. S8a, ESI†). Meanwhile, under the same condition, we collected the mass spectra from four different positions on a SA-Cu₂O@Au NC chip, obtaining the RSD of 4.7%, which was about 8-folds superior to non-assembled Cu₂O@Au NCs (Fig. 3j and Fig. S8b, ESI†). Taken together, the SA-Cu₂O@Au NC-assisted LDI-MS exhibited great advantages in LDI-MS performance, including the enhancement of MS signal and reproducibility.

The SA-Cu₂O@Au NC-based LDI-MS platform for UA detection

Encouraged by the excellent performance of SA-Cu₂O@Au NCs in LDI-MS analysis, our platform was further utilized to quantitatively detect UA using the internal standard (IS) method. UA samples at different concentrations (0.01–1 mM) were self-assembled with h-Cu₂O@Au NCs and analyzed by the SA-Cu₂O@Au NC-based LDI-MS platform by spotting PFOs as IS (4 μM, 1.5 μL). The results in Fig. 4a show the MS peaks of UA (m/z = 168) and PFOs (m/z = 499), and the relative intensity (I₁₆₈/I₄₉₉) increased as the UA concentration increased. As shown in Fig. 4b, a good linear relationship was obtained between I₁₆₈/I₄₉₉ and UA concentration in the range of 0.01–0.5 mM (R² = 0.996) with the LOD of 15.6 μM. The achieved sensitivity also covers the physiological UA concentration range in human urine (1.44–4.43 mM), demonstrating its suitability for the direct analysis of urinary samples.³⁷ The above results demonstrated the feasibility of the SA-Cu₂O@Au NC-based LDI-MS platform in the detection of low-abundance UA, showing promise in the UA analysis of clinical samples. Thus, we selected 13 urine samples, self-assembled with h-Cu₂O@Au NCs, for the analysis of UA by LDI-MS. The quantitation results displayed consistency with the commercial UA assay kit, affording a correlation coefficient of 0.876 (Fig. 4c and d). This slightly lower correlation compared with colorimetric assays for serum analysis (R² = 0.956) can be attributed to the matrix effects in urine, where high salt and metabolite content interfere with LDI-MS detection by affecting ionization efficiency and surface crystallization behaviour.^43,44 Taken together, our SA-Cu₂O@Au NC-based LDI-MS platform is reliable for UA detection in urine, presenting great potential for clinical disease diagnosis.

Fig. 4 SA-Cu₂O@Au NC-based LDI-MS for the detection of UA. (a) MS intensities of UA (0.01–1 mM) at m/z of 168 ([UA + e]⁻) and PFOs (4 μM) at m/z of 499 ([PFOs-H]⁻). (b) Linear relationship between the ratio of MS intensity of I₁₆₈/I₄₉₉ and the concentrations of UA with R² of 0.996 (0.01–0.5 mM). (c) Quantitative analysis of UA in urine samples using both the SA-Cu₂O@Au NC-based LDI-MS platform (purple) and biochemical assay (orange). (d) Linear correlation between UA concentration in urine obtained from the SA-Cu₂O@Au NC-based LDI-MS platform and the commercial UA assay kit with R² of 0.876. Each error bar represents the standard deviation calculated from three replicates.

Conclusions

We developed a dual-mode diagnostic nanoplatform based on h-Cu₂O@Au NCs that integrated POD-like activity for colorimetric sensing and self-assembly in LDI-MS analysis. The h-Cu₂O@Au NCs demonstrated good POD-like catalytic activity, achieving colorimetric detection of UA in serum samples. Moreover, the h-Cu₂O@Au NCs could enrich UA and self-assembled at the liquid–liquid interfaces, forming a closely packed nanoarray, which efficiently enhanced the MS signal and reproducibility in LDI-MS analysis. By using the SA-Cu₂O@Au NC-based LDI-MS, we successfully detected low-abundance UA in urine samples. The results obtained by our dual-mode biosensor were highly consistent with commercial UA assay kits. Both the colorimetric and LDI-MS methods provide rapid detection (30 min and 10 min, respectively), faster than enzymatic assays used in clinic (50–60 min)⁴⁵ and with wide linear ranges (50–2000 μM for colorimetry and 0.01–0.5 mM for LDI-MS). Considering that our method provides sufficient sensitivity for clinical analysis, covering the physiological UA concentration ranges in both serum (150–420 μM) and urine (1.44–4.43 mM), our platform presents strong potential for clinical and point-of-care testing. In view of its simple, fast, highly sensitive and accurate UA detection, we foresee our diagnostic nanoplatform opening up new opportunities for clinical diagnosis of related diseases.

Materials and methods

Synthesis of cubic Cu₂O

Cu₂O NCs were prepared according to the following method.⁴⁶ First, 0.3361 g of CuCl₂ was dissolved in 250 mL of deionized (DI) water. The above mixture solution was stirred, and 25 mL of NaOH solution (2 M) was added dropwise at 55 °C. After stirring for 30 min, 25 mL of L-ascorbic acid (0.6 M) was added dropwise. Then, the mixture was continuously stirred for 3 h and cooled down to room temperature (RT) naturally. Finally, the mixture was centrifuged at 5000 rpm for 10 min and washed several times using DI water and ethanol to remove unreacted chemicals.

Synthesis of h-Cu₂O@Au NCs

To synthesize the h-Cu₂O@Au NCs, 80 mg of the Cu₂O was first dissolved in 158.4 mL of DI water. Then, 1.6 mL of HAuCl₄ (2 wt%) aqueous solution was added dropwise into the above mixture with stirring at RT. After stirring for 30 min, the mixture was centrifuged at 5000 rpm for 10 min and washed several times. Finally, the h-Cu₂O@Au NCs were dissolved in 80 mL of DI water for further use.

Kinetic analysis of h-Cu₂O@Au NCs as nanozyme

Kinetic experiments were processed by recording the variation of absorbance at 652 nm under the time-scan mode of a microplate reader (MD SpectraMax i3x). To investigate the kinetic mechanism of the POD-like activity of the h-Cu₂O@Au NCs, K_m and V_max were calculated by varying the concentration of H₂O₂ and TMB, respectively.⁴⁷ On the one hand, 100 μL of h-Cu₂O@Au NCs (0.6 mg mL⁻¹) was added in 400 μL of HAc–NaAc buffer solution (pH = 4), and 100 μL of TMB (10 mM) was used as the substrate. After adding 100 μL of H₂O₂ (5 mM to 1 M) to the above solution, the absorbance was monitored at 652 nm under the time-scan mode of the MD SpectraMax i3x microplate reader. On the other hand, 100 μL of h-Cu₂O@Au NCs (0.6 mg mL⁻¹) was added in 400 μL of HAc–NaAc buffer solution (pH = 4), and 100 μL of H₂O₂ (100 mM) was used as the substrate. After adding 100 μL of TMB (0.5 mM to 2 mM) to the above solution, the absorbance was monitored at 652 nm under the time-scan mode of MD SpectraMax i3x microplate reader. Subsequently, the values of K_m and V_max were calculated via Michaelis–Menten function (v = V_max [S]/(K_m + [S])), where v is the current velocity, [S] is the current concentration of substrate, V_max is the maximum velocity, and K_m is the Michaelis–Menten constant.

Colorimetric detection of UA by h-Cu₂O@Au NCs

For the quantification of UA in both standard and serum samples, a UA biosensor based on colorimetry was designed by using h-Cu₂O@Au NCs as the POD-like nanozyme.^48,49 This is a two-step reaction. Firstly, different concentrations of UA solutions (0–2 mM, 100 μL) were incubated with 20 μL of uricase (5 U mL⁻¹) in PBS buffer (pH = 8) at 37 °C for 30 min. Then, 700 μL of HAc–NaAc buffer solution (pH = 4), 100 μL of h-Cu₂O@Au NCs (0.6 mg mL⁻¹), and 100 μL of 2 mM TMB were successively added into the above mixture and incubated at 37 °C for 30 min. The obtained solution was further subjected to UV-vis absorbance measurement at 652 nm by a UV-vis spectrometer and MD SpectraMax i3x microplate reader.

LDI-MS detection of UA by self-assembly of h-Cu₂O@Au NCs

The interfacial self-assembly of h-Cu₂O@Au NCs and UA was prepared according to the following method.^50,51 Firstly, 50 μL of h-Cu₂O@Au NCs was added in 500 μL of dichloroethane (DCE). Then, 50 μL of UA solution with different concentrations (10–1000 μM) or urine samples was added into the above mixture and shaken vigorously for 30 s. The h-Cu₂O@Au NCs were self-assembled with UA at the water-DCE interface by electrostatic interaction. To transfer the self-assembled h-Cu₂O@Au NCs to a polished chip, 80 μL of aqueous solution was discarded from the water phase to obtain the droplet with a metallic sheen. Then, the droplet was transferred to the polished chip, forming a black nanofilm. After drying at RT, the nanofilm was used as the substrate for further LDI-MS analysis of UA. For the quantitative detection of UA, 1.5 μL of PFOs was added on the substrate as the IS.

Author contributions

Dingyitai Liang: conceptualization, methodology, investigation, writing – original draft. Ziqi Ding: methodology, investigation, writing – original draft. Yushu Ding: methodology, investigation. Wenxuan Tang: methodology, investigation, writing – original draft. Shouzhi Yang: investigation. Xiaoyu Xu: investigation. Yuning Wang: methodology, supervision, writing – review and editing draft, funding acquisition. Kun Qian: conceptualization, review, funding acquisition, all authors contributed to the article and approved the final manuscript.

Data availability

All data are available in the main text or the ESI,† with additional data provided upon request.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was financially supported by the Projects 2022YFC2502800 of the National Key R&D Program of China, the Project 22204103, 82421001, 824B2059 by NSFC, Medical-Engineering Joint Funds of Shanghai Jiao Tong University (YG2023QNA12, YG2023ZD08, YG2024ZD07), and Project 2021-01-07-00-02-E00083 of Shanghai Institutions of Higher Learning. This work was also sponsored by Project 2022XYJG0001-01-16 of Shanghai Jiao Tong University Inner Mongolia Research Institute, Sichuan Science and Technology Program (2024YFHZ0176), the Innovation Research Plan by the Shanghai Municipal Education Commission (ZXWF082101), Innovative Research Team of High-Level Local Universities in Shanghai (SHSMU-ZDCX20210700).

References

1. B. N. Ames, R. Cathcart, E. Schwiers and P. Hochstein, Proc. Natl. Acad. Sci. U. S. A., 1981, 78, 6858–6862.
2. N. Dalbeth, T. Bardin, M. Doherty, F. Lioté, P. Richette, K. G. Saag, A. K. So, L. K. Stamp, H. K. Choi and R. Terkeltaub, Nat. Rev. Rheumatol., 2017, 13, 561–568.
3. E. W. Campion, R. J. Glynn and L. O. Delabry, Am. J. Med., 1987, 82, 421–426.
4. A. T. Eggebeen, Am. Fam. Physician, 2007, 76, 801–808.
5. J. W. Zhou, X. B. Zheng, H. S. Liu, B. Y. Wen, Y. C. Kou, L. Zhang, J. J. Song, Y. J. Zhang and J. F. Li, Biosens. Bioelectron., 2024, 251, 116101.
6. F. Li, J. H. Rui, Z. Y. Yan, P. Qiu and X. M. Tang, Microchim. Acta, 2021, 188, 311.
7. C. Y. Ma, N. Jiang, X. Y. Sun, L. B. Kong, T. Liang, X. W. Wei and P. Wang, Biosens. Bioelectron., 2023, 237, 115495.
8. L. Y. Yang, X. Y. Xu, Y. Song, J. Huang and H. Y. Xu, Chem. Eng. J., 2024, 487, 150612.
9. Y. Y. Huang, J. S. Ren and X. G. Qu, Chem. Rev., 2019, 119, 4357–4412.
10. J. Y. Yao, T. T. Wu, Y. Sun, Z. Y. Ma, M. L. Liu, Y. Y. Zhang and S. Z. Yao, Talanta, 2019, 195, 265–271.
11. L. Huang, Y. Zhou, Y. X. Zhu, H. Y. Su, S. Z. Yang, L. Feng, L. Zhao, S. R. Liu and K. Qian, Biosens. Bioelectron., 2022, 210, 114254.
12. J. Jiang, Y. Z. Li, L. L. Liu, L. J. Chen, J. W. Zhao, C. Streb and Y. F. Song, ACS Appl. Mater. Interfaces, 2023, 15, 1486–1494.
13. H. Wang, Z. Zhao, Y. Liu, C. Shao, F. Bian and Y. Zhao, Sci. Adv., 2018, 4, eaat2816.
14. M. Lu, X. Zhang, D. Xu, N. Li and Y. Zhao, Adv. Mater., 2023, 35, 2211330.
15. R. Shi, X. Dai, W. F. Li, F. Lu, Y. Liu, H. H. Qu, H. Li, Q. Y. Chen, H. Tian, E. H. Wu, Y. Wang, R. H. Zhou, S. T. Lee, Y. Lifshitz, Z. H. Kang and J. Liu, ACS Nano, 2017, 11, 9500–9513.
16. Y. Liu, S. Yang, S. Li, Y. Wang, X. Liu, W. Xu, H. Su and K. Qian, Small, 2025, 21, 2409714.
17. M. W. Yang, D. J. Chen, J. Hu, X. Y. Zheng, Z. J. Lin and H. M. Zhu, TrAC, Trends Anal. Chem., 2022, 157, 116752.
18. Y. N. Wang, X. Y. Xu, Y. Z. Fang, S. Z. Yang, Q. R. Wang, W. S. Liu, J. X. Zhang, D. Y. T. Liang, W. Zhai and K. Qian, ACS Nano, 2024, 18, 2409–2420.
19. K. Zhang, J. Ji, Y. X. Li and B. H. Liu, Anal. Chem., 2014, 86, 6660–6665.
20. X. Y. Xiang, W. Huang and Y. He, Appl. Catal., B, 2022, 313, 121450.
21. N. An, T. T. Chen, J. F. Zhang, G. H. Wang, M. Yan and S. K. Yang, Small Methods, 2024, 8, 2300910.
22. S. Brittman, Y. Yoo, N. P. Dasgupta, S. I. Kim, B. Kim and P. D. Yang, Nano Lett., 2014, 14, 4665–4670.
23. B. Yuan, H.-L. Chou and Y.-K. Peng, ACS Appl. Mater. Interfaces, 2021, 14, 22728–22736.
24. A. A. Zhang, J. F. Wu, L. Xue, C. X. Li, S. H. Zeng, D. Caracciolo, S. Wang and C. J. Zhong, ACS Appl. Mater. Interfaces, 2021, 13, 46577–46587.
25. X. Wang, A. Ruditskiy and Y. N. Xia, Natl. Sci. Rev., 2016, 3, 520–533.
26. S. R. Zhang, R. B. Jiang, Y. Z. Guo, B. C. Yang, X. L. Chen, J. F. Wang and Y. F. Zhao, Small, 2016, 12, 4264–4276.
27. L. Xiong, S. W. Li, B. Zhang, Y. C. Du, P. Miao, Y. Ma, Y. X. Han, H. T. Zhao and P. Xu, RSC Adv., 2015, 5, 76101–76106.
28. M. A. Mahmoud, W. Qian and M. A. El-Sayed, Nano Lett., 2011, 11, 3285–3289.
29. Z. J. Li, P. P. Huo, C. C. Gong, C. H. Deng and S. Z. Pu, Anal. Bioanal. Chem., 2020, 412, 8083–8092.
30. B. Liu, Y. L. Li, H. Wan, L. Wang, W. Xu, S. J. Zhu, Y. Y. Liang, B. Zhang, J. T. Lou, H. J. Dai and K. Qian, Adv. Funct. Mater., 2016, 26, 7994–8002.
31. Y. Zhao, W. Zhao, H. Y. Chen and J. J. Xu, Anal. Chim. Acta, 2021, 1162, 338503.
32. L. Huang, L. Wang, X. M. Hu, S. Chen, Y. W. Tao, H. Y. Su, J. Yang, W. Xu, V. Vedarethinam, S. Wu, B. Liu, X. Z. Wan, J. T. Lou, Q. Wang and K. Qian, Nat. Commun., 2020, 11, 3556.
33. J. W. Han, Y. W. Zhang, X. J. Lv, D. Q. Fan and S. J. Dong, Nanoscale, 2024, 16, 1394–1405.
34. X. F. Meng, K. Zhou, Y. Qian, H. J. Liu, X. Y. Wang, Y. F. Lin, X. Y. Shi, Y. Tian, Y. J. Lu, Q. W. Chen, J. C. Qian and H. Wang, Small, 2022, 18, 2107422.
35. H. Liu, Y. Zhou, S. A. Kulinich, J. J. Li, L. L. Han, S. Z. Qiao and X. W. Du, J. Mater. Chem. A, 2013, 1, 302–307.
36. G. T. B. Sanders, A. J. Pasman and F. J. Hoek, Clin. Chim. Acta, 1980, 101, 299–303.
37. L. A. Pachla, D. L. Reynolds, D. S. Wright and P. T. Kissinger, J. – Assoc. Off. Anal. Chem., 1987, 70, 01–14.
38. S. Hou, L. Bai, D. R. Lu and H. W. Duan, Acc. Chem. Res., 2023, 56, 740–751.
39. H. B. Xu, H. Z. Li, L. Xie, D. Y. Zhao and B. A. Kong, Adv. Mater. Interfaces, 2022, 9, 2101998.
40. Y. N. Wang, K. Zhang, T. T. Tian, W. L. Shan, L. Qiao and B. H. Liu, ACS Appl. Mater. Interfaces, 2021, 13, 4886–4893.
41. V. Poulichet, M. Morel, S. Rudiuk and D. Baigl, J. Colloid Interface Sci., 2020, 573, 370–375.
42. R. M. Wang, S. Z. Yang, M. F. Wang, Y. Zhou, X. L. Li, W. Chen, W. S. Liu, Y. D. Huang, J. Wu, J. Cao, L. Feng, J. J. Wan, J. Y. Wang, L. Huang and K. Qian, Nat. Sustainable, 2024, 7, 602–615.
43. D. Sviridov and G. L. Hortin, Clin. Chim. Acta, 2009, 404, 140–143.
44. X. Jiang, F. Gao, X. Chen, Y. Yu, G. Ding and J. Wu, J. Proteome Res., 2021, 20, 4022–4030.
45. Y. Bai, X. Fan, G. Chen and Z. Zhao, Anal. Methods, 2024, 16, 5896–5901.
46. J. Anxin, X. Linlin, T. Yue, C. Qingqiang, L. Xiangdong and C. Ming, Talanta, 2021, 225, 121990.
47. B. Jiang, D. M. Duan, L. Z. Gao, M. J. Zhou, K. L. Fan, Y. Tang, J. Q. Xi, Y. H. Bi, Z. Tong, G. F. Gao, N. Xie, A. Tango, G. H. Nie, M. M. Liang and X. Y. Yan, Nat. Protoc., 2018, 13, 1506–1520.
48. D. Nilanjan and B. Santanu, Anal. Chem., 2017, 89, 10376–10383.
49. Z. J. Peng, X. M. Tang, P. Xu and P. Qiu, ACS Appl. Mater. Interfaces, 2023, 16, 54–65.
50. Z. J. Peng, X. M. Tang, P. Xu and P. Qiu, ACS Appl. Mater. Interfaces, 2023, 16, 54–65.
51. Y. N. Wang, Y. Liu, S. Z. Yang, J. Yi, X. Y. Xu, K. Zhang, B. H. Liu and K. Qian, Small, 2023, 19, 2207190.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5tb00692a

‡ D. Liang and Z. Ding contributed equally to this work.

---