Class-Specific Immunochromatographic Assay Enabled by Mesoporous Nanozyme-Catalyzed Signal Amplification for On-Site Screening of Sulfonylureas

Abstract

Conventional immunochromatographic assays (ICAs) face limitations in sensitivity and dynamic range, hindering their application in on-site, class-specific screening of sulfonylurea (SU) adulteration in functional foods. To address this, a signal amplification strategy was developed by engineering high-density platinum nanozymes on a mesoporous metal–organic framework (PCN-224). The mesoporous architecture of PCN-224 facilitated high-density and stable loading of catalytically active Pt sites. The established PCN-224@Pt-based ICA achieved detection limits of 0.52–7.94 μg/kg in tea and 0.69–7.02 μg/kg in capsules, with linear ranges of 1.69–513.01 μg/kg and 2.05–716.47 μg/kg, respectively. Compared with traditional colloidal gold immunochromatographic assays (CG-ICAs), sensitivity was improved by up to 57-fold, while the linear detection range was expanded by over 5-fold relative to the previously reported PCN-224@PDA- ICA. The method demonstrated recovery rates of 81.8–119.8% and coefficients of variation between 2.5% and 11.4%. Validation against LC-MS/MS using 20 real samples showed excellent agreement (R² > 0.99). This work not only provides a sensitive and rapid tool for the surveillance of SU adulteration in functional foods but also establishes a generalizable nanozyme design strategy applicable to enhancing the performance of a wide range of ICA-based detection platforms.

1. Introduction

Sulfonylureas (SUs), a category of core prescription drugs for type 2 diabetes, are widely used in clinical practice due to their potent hypoglycemic effects [1]. However, in recent years, unscrupulous manufacturers have illegally adulterated functional foods (e.g., tea beverages and capsules) with these prescription drugs to falsely advertise “natural hypoglycemic effects”, severely disrupting market integrity and posing significant public health risks [2]. This illegal practice not only violates food safety laws but also potentially causes irreversible health issues, including acute hypoglycemia, gastrointestinal disorders, anaphylactic shock, and schistocytic hemolytic anemia [3,4]. Notably, conventional analytical methods lack the high-throughput capacity and on-site screening capability required to fulfill regulatory requirements. Consequently, developing novel analytical strategies that allow both high-sensitivity and class-specific screening is urgently needed in food safety testing.

The detection of SUs is primarily dependent on sophisticated instrumental techniques such as liquid chromatography–tandem mass spectrometry (LC-MS/MS) [5] and high-performance liquid chromatography (HPLC) [6]. Although these methods exhibit excellent sensitivity and reproducibility, making them suitable for standardized laboratory settings and method validation, their high instrument costs, complex sample preparation procedures, and reliance on specialized personnel severely restrict their application in large-scale field screening and consumer self-testing. In contrast, immunoassays based on antigen–antibody (Ag-Ab)-specific recognition have emerged as core tools in the field of rapid detection due to their operational simplicity, rapid response times, high-throughput processing capabilities, and diverse signal output modes [7,8]. For example, an enzyme-linked immunosorbent assay (ELISA) was developed by Li et al. to simultaneously detect six SUs, with LODs ranging from 0.02 to 1.0 ng/mL [9]. However, this method required multiple incubation and washing steps, which hindered its suitability for on-site analysis of large batches of samples. Xie et al. [10] also developed a colloidal-gold-based immunochromatographic assay (CG-ICA) capable of detecting nine SUs (with IC₅₀ values of 4.8–104 ng/mL and cut-off values of 96–3200 ng/mL). Nevertheless, the inherent limitations of colloidal gold in terms of stability and sensitivity [11] highlighted the urgent need to develop novel high-performance signal-labeling carriers to provide a critical technological breakthrough for constructing detection systems that better meet the demands of on-site regulatory monitoring.

Metal–organic frameworks (MOFs), with their ultra-high porosity and highly tunable chemical properties, have emerged as ideal platforms for loading drugs and biomolecules. Their unique structures can provide excellent “molecular armor” for encapsulated biomolecules (such as Abs and enzymes), effectively resisting thermal, chemical, and mechanical stresses and ensuring the stability of biological activity [12]. These core advantages have enabled MOFs to show their great potential in the field of rapid immunoassays. In recent years, studies have successfully applied MOFs as novel Ab carriers in food safety determination [13]. For example, Pang et al. developed an ICA based on UiO-66-NH₂@Au for the detection of carbofuran in vegetables. The method not only improved sensitivity 2–10-fold compared with the traditional colloidal gold method but also significantly expanded the pH tolerance range to 4–10 (that for the traditional method is 6–8), and organic solvent tolerance was increased to 40% (the traditional value is 20%) [14]. Huang et al. introduced the lanthanide metal Eu³⁺ into UiO-66-(COOH)₂ through post-synthetic modification and constructed a UiO-66-Eu probe for the detection of sodium pentachlorophenate in animal-derived foods. Sensitivity improved more than 10-fold, and the pH tolerance window was expanded to 3–11 (the traditional range is 3–8) [15]. The above cases strongly confirm the significant advantages of using MOFs as high-performance Ab carriers in terms of improving detection sensitivity, environmental robustness, and biocompatibility. For quantitative immunoassays, a wide linear range is crucial, as it is directly related to detection efficiency, flexibility, method robustness, reliability, cost-effectiveness, and the ability to analyze complex matrices [16,17]. Nanozyme-mediated signal amplification has become an effective strategy for expanding the linear range. For instance, Liang et al. used magnetic Prussian-blue nanozymes for signal amplification and successfully induced a 4-fold expansion in the linear range of the ICA [18]. Lin et al. developed an ICA based on Ti₃C₂Tₓ@Pt for the detection of chloramphenicol, and its linear range also widened 2.5-fold compared with that before amplification [19]. In summary, the combination of MOFs and nanozymes to construct MOF@nanozyme composite probes can give full play to the synergistic effect: MOFs contribute excellent biomolecule loading and protection (“molecular armor”), while nanozymes provide strong signal amplification capabilities. This synergy can not only better protect Ab activity and significantly improve detection sensitivity but also effectively expand the linear range of detection, providing a promising material platform for the development of high-performance rapid food safety detection technologies.

Therefore, in this study, PCN-224 with high porosity and a large specific surface area was employed as a substrate to load high-density PtNPs via an in situ reduction strategy. By further conjugating the resulting composite with SUs Abs, we fabricated multifunctional nanoprobes, integrating excellent catalytic activity with high Ab loading capacity. Based on these nanoprobes, an ICA was developed for the detection of SUs in functional foods. Through systematic evaluation of the method’s performance, including LOD, linear range, and applicability to real samples, this work not only provides an effective technical approach for the monitoring of SU residues but also establishes a valuable theoretical foundation for the design and application of MOF@nanozyme composite probes in the field of rapid food safety determination.

2. Materials and Methods

2.1. Materials and Equipment

Glipizide (GP, 98%), glimepiride (GM, 99%), glyburide (GB, 99%), tolbutamide (TB, 99%), gliquidone (GQ, 98%), gliclazide (GL, 99%), carbutamide (CB, 98%), tolazamide (TLZ, 96%), acetohexamide (AH, 98%), chlorpropamide (CPM, 99%), glibornuride (GBN, 98%), repaglinide (RGLN, 99%), rosiglitazone (RGLT, 98%), phenformin (PF, 95%), metformin (MFM, 97%), octahydrate zirconium oxychloride (ZrOCl₂·8H₂O, 98%), dopamine (DA, 98%), 4-carboxyphenyl porphyrin (H2TCPP, 97%), N,N-Dimethylformamide (DMF), potassium bromide (KBr), and 3,3′,5,5′-tetramethylbenzidine (TMB) were provided by Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Hexahydrate chloroplatinic acid (H₂PtCl₆·6H₂O), L-ascorbic acid (Vc), and sodium cyanoborohydride (NaBH₃CN) were supplied by Shanghai Macklin Biochemical Technology Co., Ltd. (Shanghai, China). Sodium borohydride (NaBH₄) was obtained from Merck KGaA, Germany (Darmstadt, Germany). Citric acid (CA), trisodium citrate dihydrate (sodium citrate, SC), and bovine serum albumin (BSA) were bought from Sigma-Aldrich (St. Louis, MO, USA). Coating antigen (SAL-OVA, 2 mg/mL) and sulfonylurea drug monoclonal antibody (anti-SU mAb) were prepared in the laboratory [10]. Secondary antibody (Goat anti-mouse IgG, 17.5 mg/mL) was sourced from Tianjin Sanjian Biotechnology Co., Ltd. (Tianjin, China). Glass cellulose membranes (SB08, SB06, and RB65), absorbent paper (CH37K), PVC substrate plate (SMA31-40), and nitrocellulose membrane (UniSart CN 140) were acquired from Shanghai Liangxin Co., Ltd. (Shanghai, China).

Cutting machine (ZQ2000) was bought from Shanghai kinbio Tech. Co., Ltd. (Shanghai, China). XYZ plotting instrument (HM-3060) was purchased from BioDot, Inc. (Irvine, CA, USA). A field-emission scanning electron microscope (Talos F200S) and an ultra-high-resolution field-emission transmission electron microscope (Apreo 2) were purchased from Thermo Fisher Scientific, Inc. (Waltham, MA, USA). Fourier transform infrared spectrometer (Vertex 700) was acquired from Bruker Corporation (Billerica, MA, USA). Ultra-low-temperature high-speed centrifuge was obtained from Hunan Xiangyi Technology Co., Ltd. (Changsha, China). Water purification system (UNIQUE-R10) was supplied by Guangzhou Yuwei Biotechnology Instrument Co., Ltd. (Guangzhou, China).

2.2. Synthesis of PCN-224@Pt

The synthesis of PCN-224 was carried out according to the method reported by He et al. [2]. Briefly, a DMF solution containing H2TCPP was added dropwise to a DMF solution containing benzoic acid and ZrOCl₂·8(H₂O); this was followed by reaction at 80 °C for 5 h. After being washed sequentially with DMF and ethanol three times, each PCN-224 sample was collected and dried under vacuum at 80 °C for 5 h. Then, 100 mg of PCN-224 was accurately weighed and dispersed in 50 mL of ultrapure water. Subsequently, 2.5 mL of 1% H₂PtCl₆ solution was added. After 30 min of stirring, 2.5 mL of 20 mM NaBH₄ solution was added, and the reaction was allowed to proceed for another 30 min. Then, centrifugation was performed at 12,000 rpm for 5 min to obtain dark-brown PCN-224@Pt. After being washed twice with ultrapure water, the product was vacuum-dried at 80 °C overnight to obtain PCN-224@Pt powder for use [20] (Figure 1A).

2.3. Preparation of PCN-224@Pt-mAb Immunoprobe

A total of 2 mg of PCN-224@Pt was accurately weighed in 1 mL of 0.02 M MES (pH 6.5). After ultrasonic mixing, 100 μL of Ab solution (0.05 mg/mL, diluted with 0.5% BSA) was added. The reaction was stirred at 300 rpm for 30 min at room temperature. Then, 100 μL of 10% BSA solution was added for a blocking reaction for another 30 min. Then, centrifugation was performed at 14,000 rpm for 10 min at 4 °C. The supernatant was discarded; the resulting reddish-brown precipitate was resuspended in 200 μL of solution (0.02 M PB, pH 7.4, 0.3% PVP, 0.5% Tween^®-20, 0.5% BSA) and stored at 4 °C for use.

2.4. The Peroxidase-like Activity of PCN-224@Pt

The absorbance values of the PCN-224@Pt-mAb probe at 652 nm were measured under different concentrations of H₂O₂ (0.004, 0.008, 0.015, 0.031, 0.062, 0.125, 0.5, 1, and 2 mol/L), and the Km value of the probe was calculated using GraphPad Prism 9. The peroxidase-like catalytic activity of the PCN-224@Pt-mAb probe was verified using the following Michaelis–Menten equation [21]:

$${\displaystyle \frac{1}{\mathrm{V}}}={\displaystyle \frac{{\mathrm{K}}_{\mathrm{m}}}{{\mathrm{V}}_{\max }\left[\mathrm{S}\right]}}+{\displaystyle \frac{1}{{\mathrm{V}}_{\max }}}$$

(1)

2.5. Assembly of the Test Strip

The assembly procedure for the test strip was the same as that reported in a previous study [2]. The specific parameters are detailed in Table S1. Briefly, SU-OVA (0.3 mg/mL) and goat anti-mouse IgG (0.15 mg/mL) were dispersed on the NC membrane to develop the test line (T-line) and control line (C-line). The NC membrane was dried at 37 °C for 12 h. The sample pad was submerged in a sample pad pretreatment solution for 30 s and dried for 12 h at 37 °C. Ultimately, the absorbent pad, pretreated NC membrane, and sample pad were stuck onto the adhesive backing pad with a 2 mm overlap between each other and cut into 3.5 mm wide strips.

2.6. Sample Pretreatment and Detection Steps

Functional tea and capsule samples purchased from a local market (Guangzhou, China) were confirmed to be free of SUs via LC-MS/MS prior to unpacking and being ground into powder. Then, 1.00 ± 0.05 g of the powder was accurately weighed, 1 mL of methanol was added, and the mixture was vortexed for 3 min. After centrifugation at 10,000 rpm for 5 min, the supernatant was filtered through a 0.22 μm organic filter membrane and then diluted 5-fold with 0.02 M PB (pH 7.4) for subsequent analysis.

2.7. Detection Steps

A total of 120 μL of treated sample solution was added to a microwell, and 2 μL of PCN-224@Pt-mAbs immunoprobe was added and mixed in. After incubation for 3 min, the test strip was vertically inserted into the microwell. After reaction for 7 min, the test strip was taken out and soaked in TMB solution (0.19%, pH 4.5) for 5 min. The test strip was then taken out again, and the sample pad was removed. Qualitative results were obtained through visual observation, and quantitative results were acquired with a test strip reader (Figure 1B).

2.8. Performance Evaluation of PCN-224@Pt-ICA

2.8.1. Sensitivity

Tea and capsule samples were spiked with the four most important SUs (GP, GM, GB, and TB) in a sequence of concentrations. Each concentration was measured in triplicate to determine the cut-off value, LOD, limit of quantification (LOQ), and linear range of the method for evaluating the sensitivity of the PCN-224@Pt-ICA. The cut-off value was defined as the lowest drug concentration at which the T-line of the test strip completely disappeared [22]. A calibration curve was plotted, using the SU concentration as the x-axis and B/B₀ (the proportion of T-line intensity/C-line intensity value when SUs were present or absent in standard/sample solutions) as the y-axis. In accordance with the practical and standardized reporting conventions for the technical details of ICA, based on the calibration curve, the LOD, LOQ, and linear range were determined as the analyte concentrations at which the inhibition of B/B₀ reached 10%, 20%, and 20–80%, respectively [23].

2.8.2. Selectivity

The selectivity of PCN-224@Pt-ICA was evaluated based on cross-reactivity (CR, %). The developed PCN-224@Pt-ICA and indirect competitive enzyme-linked immunosorbent assay (icELISA) were employed to detect 11 SUs (GP, GM, GB, TB, GQ, GL, CB, TLZ, AH, CPM, and GBN) and 4 commonly used oral hypoglycemic drugs (RGLN, RGLT, PE, and MFM). The CR value was calculated using the following formula:

$$\mathrm{CR} (\%)={\displaystyle \frac{{\mathrm{IC}}_{50} (\mathrm{GP}, \mathsf{\mu }\mathrm{g}/\mathrm{kg})}{{\mathrm{IC}}_{50 }(\mathrm{Other} \mathrm{drugs}, \mathsf{\mu }\mathrm{g}/\mathrm{kg})}}$$

(2)

2.8.3. Accuracy and Precision

The accuracy and precision of PCN-224@Pt-ICA were evaluated based on the recovery rate and coefficient of variation (CV). Tea and capsule samples were spiked with three known concentrations of GP, GM, GB, and TB. All samples were treated in triplicate on three separate days.

2.8.4. Application of PCN-224@Pt-ICA

A total of 20 functional foods (10 teas and 10 capsules) were bought from a local market (Guangzhou, China). They were detected in triplicate using both the PCN-224@Pt-ICA established in this study and LC-MS/MS. The consistency and correlation between the two detection methods were compared.

2.8.5. Stability

The stability of PCN-224@Pt-ICA was evaluated through low-temperature (4 °C), accelerated (37 °C), and long-term (room temperature) stability tests. The test strips and probes were stored under the above three temperature conditions, and the T-line intensity and inhibition ratio were measured. The low-temperature and accelerated stability tests were conducted on days 0, 1, 7, 15, and 30, while the long-term stability tests were conducted on days 0, 15, 30, and 60.

2.9. Statistical Analysis

Each experiment was repeated three times, and the data are presented as the mean ± standard deviation (SD). Statistical analysis was performed using GraphPad Prism 9 and Origin 2021. The Pearson correlation coefficient was used for the correlation analysis (p < 0.05). p < 0.05 indicates a statistically significant difference.

3. Results

3.1. Identification of PCN-224@Pt and PCN-224@Pt-mAb

Scanning electron microscope (SEM) analysis (Figure 2A) showed that PCN-224@Pt exhibited a homogeneous spherical morphology with a size of approximately 100 nm, and PtNPs were dispersed on the surface of the PCN-224 framework [20]. Through high-resolution transmission electron microscopy (HRTEM) (Figure 2B), it was observed that PtNPs were anchored on the surface of the PCN-224 matrix, with lattice spacings of 0.226 nm and 0.198 nm, corresponding to the (111) and (200) crystallographic planes of platinum nanoparticles, respectively [24]. The Fourier-transform infrared (FT-IR) spectrum of PCN-224@Pt (Figure 2C) was not completely consistent with that of PCN-224. After the in situ growth of PtNPs on PCN-224, the vibrational peaks of -(O-C-O)- were shifted to 1580 ${\mathrm{cm}}^{-1}$ and 1400 ${\mathrm{cm}}^{-1}$. In the spectrum of PCN-224@Pt-mAb, an absorption peak of the amide bond appeared at 1630 ${\mathrm{cm}}^{-1}$ [25]. These results confirmed that the PCN-224@Pt labeling material and probe had been successfully synthesized. Zeta potential measurements (Figure 2D) demonstrated that the potentials of PCN-224, PCN-224@Pt, and PCN-224@Pt-mAb were +34.1 mV, −10.9 mV, and −24.1 mV, respectively. This potential change further confirmed the successful preparation of the probe and indicated that the stability of the probe had been enhanced.

3.2. Determination of the Catalytic Activity of PCN-224@Pt-mAb

The catalytic activity of the probe was measured under different concentrations of H₂O₂. The data were fitted using the Michaelis–Menten equation in GraphPad Prism 9 (Figure S1). The Km value of the measured probe was 0.363 mM, which is lower than the Km value of HRP (3.702 mM) [21,26]. This indicated that the probe exhibited strong substrate affinity, which was conducive to enhancing the signal through catalytic deposition. Strict normalization (e.g., regarding catalyst mass and active site density) is essential when comparing kinetic parameters across different catalysts to evaluate intrinsic catalytic efficiency. However, the comparative analysis in this study is focused on the affinity between the probe and HRP at the substrate level. Therefore, cross-system comparisons of Km values among catalysts with different structures should be interpreted with consideration of system-specific differences. Moreover, this analytical context does not affect the practical applicability of the probe’s catalytic performance within the established ICA system.

3.3. Optimization of the Synthesis of PCN-224@Pt

3.3.1. Optimization of the Concentration of H₂PtCl₆

PtNPs were loaded on the surface of PCN-224 substrates, which can immobilize Abs through electrostatic adsorption and perform catalytic functions. Therefore, the concentration of H₂PtCl₆ precursor for synthesizing PtNPs was particularly important [27]. As can be seen from Figure 3A, when the concentration of H₂PtCl₆ was 2.5 mL, the colorimetric and inhibition ratios were optimal. Accordingly, the optimal concentration of H₂PtCl₆ was determined to be 5.0 mM.

3.3.2. Optimization of the Types of Reducing Agents

The degree of reduction of H₂PtCl₆ on PCN-224 surfaces can be significantly modulated using different reducing agents, consequently affecting the status of immobilization with Ab [28]. As illustrated in Figure 3B, although both NaBH₄ and Vc demonstrated optimal colorimetric and inhibition ratios, the probes prepared through Vc reduction exhibited elevated background signals during detection. Therefore, the optimal reducing agent was determined to be 20 mM NaBH₄.

3.3.3. Optimization of the Amount of NaBH₄

The size and optical properties of PtNPs are also affected by the amount of the reducing agent. As shown in Figure 3C, due to an excessive amount of reducing agent, the structure of the PCN-224 carrier was destroyed, which hindered the effective growth of PtNPs on its surface, resulting in a decrease in the colorimetric and inhibition ratios following an increase in the amount of reducing agent. Maximum colorimetric and inhibition ratios were achieved when 2.0 mL of 20 mM NaBH₄ was used. Therefore, the optimal amount of NaBH₄ was determined to be 2.0 mL.

3.4. Optimization of Key Technical Parameters of PCN-224@Pt-ICA

To assess the performance of PCN-224@Pt-ICA, three key technical parameters, namely, the coupling pH value, the diluent of Abs, and coating pH value, were evaluated.

3.4.1. Optimization of the Coupling pH Value

The coupling pH value is a crucial parameter that affects both the binding efficiency between the Ab and the carrier and the activity of the Ab [29]. As shown in Figure 3D, the optimal colorimetric and inhibition ratios were achieved when the coupling pH was set to 6.5. Consequently, the best coupling buffer condition was 0.02 M MES buffer at pH 6.5.

3.4.2. Optimization of the Diluent of Ab

The appropriate Ab dilution buffer can improve the microenvironment for Ab immobilization, thereby facilitating the effective adsorption of Abs onto the surface of nanomaterials [30]. As shown in Figure 3E, among the five diluents, 0.5% BSA exhibited superior colorimetric and inhibition ratios, which may be attributed to the protein-stabilizing effect of BSA, through which the stability and adsorption capacity of Ab were enhanced. Therefore, the best Ab dilution buffer was determined to be 0.5% BSA.

3.4.3. Optimization of the Coating pH Value

The coating buffer employed significantly affects both the immobilization efficiency of SU-OVA on NC membranes and its specific binding to the probe [31]. As shown in Figure 3F, when 0.02 M pH 8.0 BB was used as the coating buffer, SU-OVA was uniformly distributed in the pores of the NC membrane, forming a stable solid-phase Ag layer, and optimal colorimetric and inhibition ratios were observed. Therefore, the optimal coating buffer was 0.02 M pH 8.0 BB.

3.5. Performance Evaluation for PCN-224@Pt-ICA

3.5.1. Sensitivity

As illustrated in Figure 4, the cut-off values for GP/GM/GB/TB in the functional tea and capsule samples were 60/150/150/600 and 90/240/240/960 μg/kg, the LODs were 0.52/1.72/2.94/7.94 and 0.69/3.98/4.99/7.02 μg/kg, the LOQs were 1.69/4.16/6.13/20.39 and 2.05/8.95/10.40/19.99 μg/kg, and the linear ranges were 1.69–52.87/4.16–84.42/6.13–75.30/20.39–513.01 and 2.05–85.59/8.95–142.97/10.40–127.52/19.99–716.47 μg/kg, respectively.

3.5.2. Selectivity

The selectivity results for PCN-224@Pt-ICA are shown in Table S2. Ab exhibited a different CR relative to that of GP (100%), GB (100%), GM (83.3%), GQ (68.2%), TB (23.1%), and CB (17%). This was because the basic structure of SU consists of a benzene ring linked to a sulfonamide group, with two additional substituents at both ends. The differences in these substituents result in variations in the exposed structures, which correspond to differences in antigenic epitopes [32]. Therefore, the prepared anti-SU mAb demonstrated varying recognition capabilities for different drugs, and the selectivity results for PCN-224@Pt-ICA were largely consistent with the trends observed in icELISA. The established method showed CR values greater than 2.9% for nine types of SU drugs, indicating that this method can simultaneously detect nine SU drugs.

3.5.3. Accuracy and Precision

As illustrated in

Table 1. Recovery of SUs in functional tea and capsule samples by PCN-224@Pt-ICA (n = 3).

Analytes	Tea				Capsule
	Spiked Level (μg/kg)	Detected Level (μg/kg)	Recovery (%)	CV (%)	Spiked Level (μg/kg)	Detected Level (μg/kg)	Recovery (%)	CV (%)
GP	2	2.30 ± 0.15	115.2	6.6	2	1.64 ± 0.09	81.8	5.3
	4	3.51 ± 0.13	87.8	3.9	4	4.73 ± 0.34	118.3	7.1
	8	6.83 ± 0.89	85.4	13.0	20	23.71 ± 2.18	118.6	9.2
GM	4	3.34 ± 0.32	83.5	9.6	9	10.31 ± 0.84	114.6	8.1
	8	7.10 ± 0.76	88.7	10.7	18	21.35 ± 1.45	118.6	6.8
	40	45.18 ± 1.3	112.9	3.6	90	106.44 ± 2.68	118.3	2.5
GB	6	6.99 ± 0.49	116.5	7.0	10	8.67 ± 0.67	86.7	7.7
	12	13.94 ± 1.00	116.1	7.2	20	16.86 ± 1.92	84.3	11.4
	60	71.85 ± 2.09	119.8	2.9	100	87.17 ± 2.83	87.2	3.2
TB	20	23.27 ± 2.33	116.4	11.4	20	17.96 ± 1.99	89.8	11.1
	40	45.53 ± 2.57	113.8	5.7	40	35.45 ± 3.18	88.6	9.0
	200	235.00 ± 6.93	117.5	2.9	200	177.06 ± 5.73	88.5	3.2

, the recovery ranges of GP/GM/GB/TB in the functional tea samples and capsule samples were 85.4–115.2%/83.5–112.9%/116.1–119.8%/113.8–117.5% and 81.8–118.6%/114.6–118.6%/84.3–87.2%/88.5–89.8%, respectively, with corresponding CVs of 3.9–13.0%/3.6–10.7%/2.9–7.2%/2.9–11.4% and 5.3–9.2%/2.5–8.1%/3.2–11.4%/3.2–11.1%, respectively. Although some recoveries approached 120%, this was attributed to the matrix enhancement effect induced by trace components present in the complex matrix of functional foods. Nevertheless, these results still demonstrate the favorable accuracy and precision of the PCN-224@Pt-ICA method.

3.5.4. Application of PCN-224@Pt-ICA

As shown in Figure 2E,F and Table S3, to verify the reliability of the established PCN-224@Pt-ICA, GP was randomly added to the aforementioned 10 tea samples and 10 capsule samples, respectively. The inspectors were unaware of the concentration of GP added. The results show that the detection results for the two methods are basically consistent (R² > 0.99), and the data confirm that the PCN-224@Pt-ICA method has excellent reliability.

3.5.5. Stability

As shown in Figure 2G, after the test strips were stored at a low temperature (4 °C) for 30 d, there was almost no change in T-line intensity or the inhibition ratio, indicating that low temperatures are highly beneficial to the activity and stability of the probe. After 30 d of accelerated storage at 37 °C and 60 d of storage at room temperature, the T-line intensity of the test strips decreased slightly, but their inhibition ratios remained basically unchanged. Based on the Technical Guidelines for Stability Studies of Biological Products and the Arrhenius equation [33], the developed PCN-224@Pt-ICA test strip is expected to remain stable for at least one year under the modelled conditions.

3.6. Comparison of the Methods for the Detection of SUs

As shown in

Table 2. Comparison between the developed PCN-224@Pt-ICA and other methods for SU detection.

Method	Sample	Detection Time (min)	LOD (μg/kg)	LOQ (μg/kg)	Cut-Off Value (μg/kg)	Linear Ranges (μg/kg)	Reference
HPLC	Herbal Supplement, Health Food	43	216	432	-	-	[6]
HPLC	Milk power, Capsule	65	2450	8170	-	2 × 103–3 × 105	[34]
UPLC	Tablet, Capsule, Granule	25	2.7	-	-	5–500	[35]
Ion-Pair LC	Tablet, Pill, Granule, Capsule	47	330	1010	-	-	[36]
LC-MS/MS	Tablet, Pill, Capsule	35	0.5	15	-	-	[5]
LC-Q/TOF	Tablet, Pill, Capsule	47	1.05	-	-	1–2500	[37]
ELISA	Pill, Capsule	90	0.3	0.7	-	-	[9]
CG-ICA	Tea	18	32	-	96	-	[10]
PCN-224@PDA-ICA	Functional Tea, Capsule	10	Tea: 0.22 (GP) Capsule: 0.4 (GP)	Tea: 0.75 (GP) Capsule: 1.03 (GP)	Tea: 45 (GP) Capsule: 70 (GP)	Tea: 0.75–121.32 Capsule: 1.03–193.04	[2]
PCN-224@Pt-ICA	Functional Tea, Capsule	15	Tea: 0.52 (GP) Capsule: 0.69 (GP)	Tea: 1.69 (GP) Capsule: 2.05 (GP)	Tea: 60 (GP) Capsule: 90 (GP)	Tea: 1.69–513.01 Capsule: 2.05–716.47	This study

-: Unmentioned.

, the currently published detection methods for SUs are mainly based on instrumental analysis techniques [2,5,34,35,36,37], and only three immunoassay methods for the detection of SUs have been reported [2,9,10]. Among the reported ICAs, only CG and PCN-224@PDA have been employed as Ab-labeling materials. In this study, the nanozyme activity of PtNPs was utilized in SU detection for the first time, with substrate catalysis carried out through a soaking approach. The advantages of this approach are reflected in the following three aspects: (1) In contrast to the commonly used ex situ loading strategy in existing MOF-based ICA approaches, the in situ confined synthesis strategy adopted in this study enabled high-density and uniform loading of active sites, along with optimized mass-transfer efficiency. (2) Unlike the enzyme-free catalytic signal amplification method used in PCN-224@PDA-ICA and the non-catalytic signal labeling approach of CG-ICA, we utilized platinum nanozymes with high catalytic activity as the signal core, thereby enhancing the intrinsic catalytic efficiency of the detection system and facilitating cascade signal amplification. (3) Unlike the single-step reaction-based signal generation mode in traditional ICA, the design of a “soak-and-catalyze” two-step signal amplification mode allowed kinetic optimization of the detection mechanism, thereby synergistically overcoming the traditional trade-off between sensitivity and linear range. Based on these advantages, this method not only prevented the diffusion of the T and C lines but also achieved signal amplification, resulting in a 1.75-fold increase in the output signal. The linear ranges of the tea and capsule samples in this study were about 5-fold wider than the range for PCN-224@PDA-ICA, and the sensitivity was 57-fold higher than that of traditional CG-ICA. Additionally, compared with instrumental analysis methods, our approach has obvious advantages in terms of sensitivity and operation steps, with the detection time shortened by an average of 28 min. In summary, the method developed in this study has practical value for the on-site rapid detection of SUs owing to its unique signal amplification mode, wide linear range, high sensitivity, simplicity, and speed.

4. Conclusions

In this study, a high-performance PCN-224@Pt-ICA was developed to enable simultaneous and rapid screening of nine SUs in functional teas and capsules. Through the in situ growth of PtNPs on the surface of PCN-224, the resulting probe significantly enhanced catalytic signal generation while enabling efficient Ab loading, thereby synergistically leveraging the high surface area of MOF materials and the excellent catalytic activity of nanozymes. The method achieved LODs as low as 0.52–7.94 μg/kg for functional teas and 0.69–7.02 μg/kg for capsules, with linear ranges of 1.69–513.01 μg/kg and 2.05–716.47 μg/kg, respectively, approximately 5-fold wider than those of existing analogous methods. Moreover, the analytical sensitivity improved 57-fold compared to conventional CG-ICA. This strategy integrates high sensitivity, a wide dynamic range, class-specific screening capacity, and on-site applicability, providing a feasible method for rapid monitoring of illicitly added drugs in complex matrices, with significant potential for application in ensuring and promoting the safety of functional foods.