Biomolecule-Assisted Synthesis of Blue-Emissive Gold Nanoclusters for Sensitive Analysis of Tetracycline Residues in Food and Feed Authors: Xianling Chen, Shengnan Han, Ruiyun Tong, Jinyan Xie, Li Song, Jie Chen, Zesheng Yang, Cunshan Zhang, Xiaofang Chen, Shaoguang Li Published: 2026 ABSTRACT Tetracycline antibiotics (TCs), which are frequently used in animal production, may remain in food products and thus require reliable analytical monitoring. In this study, we developed a fluorescence-based method for detecting three kinds of TCs in animal-derived foods and feed, including tetracycline hydrochloride (TC), chlortetracycline hydrochloride (CTC), and oxytetracycline (OTC). l-proline (Pro) served as a reducing and stabilizing ligand, and sodium alginate (SA) functioned as a costabilizer and surface-passivating biomolecule. By providing a biomolecularly coordinated interface that may be associated with restriction of intramolecular motion, these components enabled the one-pot synthesis of water-soluble blue-fluorescent gold nanoclusters (Pro-SA-AuNCs). Compared with systems using Pro alone, the synergistic stabilization of Pro and SA enhanced the fluorescence intensity and stability of the resulting NCs. The resulting probes allowed the detection of TCs through inner filter effect (IFE) and aggregation-induced quenching (static quenching). These dual quenching pathways and their combination with economical and easily operable solid-phase extraction for sample pretreatment contributed to the observed relatively high sensitivity and wide linear response range (5–800 μM). The obtained limits of detections (0.16–0.54 μM) are comparable to reported probes (Table 1). Application to representative real samples, including milk, honey, and animal feed, yielded high recovery rates, which supports the method’s applicability to complex samples. These results suggest that Pro-SA-AuNCs and related methods provide a feasible platform for the development of fluorescence-based detection formats for TCs. Introduction Tetracycline antibiotics (TCs), a class of broad-spectrum antibiotics derived from actinomycetes, include tetracycline hydrochloride (TC), chlortetracycline hydrochloride (CTC), and oxytetracycline (OTC). These compounds can effectively inhibit Gram-positive and Gram-negative bacteria by reversibly binding to the 30S ribosome subunit, thereby blocking bacterial protein synthesis. While clinically used for respiratory, urinary, and skin infections, , their low cost has contributed to their extensive use in livestock production. Residual TCs accumulate in aquaculture soil, animal-derived foods (meat, milk, and honey), and water systems, and they ultimately enter humans via the food chain. These residues are associated with reported cytotoxic effects, including liver injury, and may contribute to bacterial resistance, raising concerns regarding food safety and public health. Current detection methods of TCs in food primarily include polymer-based sensors, high-performance liquid chromatography (HPLC), ultraviolet–visible spectrophotometry (UV–Vis), liquid chromatography–mass spectrometry (LC/MS), and enzyme-linked immunosorbent assay (ELISA). , Specifically, UV–Vis provides moderate sensitivity, and HPLC and LC/MS involve relatively high instrument costs and operational requirements. ELISA incurs relatively substantial costs due to its reliance on antibody-based detection. By contrast, fluorescence analysis offers high selectivity, sensitivity, and operational simplicity and has been proven effective in the detection of medicinal substances, heavy metals, low-mass organic compounds, pathogens, and biological macromolecules. , Owing to these features, fluorescence-based sensing has received increasing research attention. Various fluorescent nanomaterials, such as semiconductor quantum dots, dye-doped nanoparticles, and carbon nanodots, have been developed and applied in practical scenarios, and they demonstrated reliability and versatility. Among these nanomaterials, fluorescent metal nanoclusters (NCs), particularly gold NCs (AuNCs), exhibit distinct advantages including low toxicity, facile surface modification, strong and tunable fluorescence stability, and good biocompatibility. Thus, functionalizing AuNCs and expanding their application scope hold importance for the advancement of analytical techniques. In this study, AuNCs were efficiently synthesized via a green “bottom-up” approach. Common ligands, including histidine (His), cysteine, methionine (Met), L-proline (Pro), 11-mercaptoundecanoic acid, N-acetylcysteine, and other compounds, , were screened as reducing agents, with sodium alginate (SA) added as a co-protection agent. SAa naturally occurring polysaccharide rich in carboxyl and hydroxyl groupsacted as an efficient co-stabilizer and surface-passivating agent. Such polysaccharide ligands can restrict the intramolecular motion (RIM) of surface-bound small molecules via hydrogen bonding and electrostatic interactions, thereby suppressing nonradiative relaxation and significantly enhancing the fluorescence quantum yield of metal nanoclusters. Among these, L- Pro was selected because the resulting Pro-SA-AuNCs exhibited the strongest fluorescence emission under identical synthesis conditions, indicating its superior performance. Pro-SA-capped AuNCs (Pro-SA-AuNCs) were successfully prepared and synthesized via a heating method (Figure A). Under 365 nm UV illumination, the Pro-SA-AuNCs emitted a bright blue fluorescence. Their fluorescence intensity was significantly enhanced compared with that of Pro-capped AuNCs (Pro-AuNCs) synthesized by Mu et al. Pro–SA–AuNCs synthesis, quenching mechanism,and visual TC detection. (A) Schematic illustration of the synthesis process of Pro-SA-AuNCs (using Pro as reducing and stabilizing agent and SA as costabilizer) and their fluorescence quenching mechanism toward TCs. (B) Photographs depicting the fluorescence quenching effect of Pro-SA-AuNCs induced by different concentrations of TCs (0, 100, 500, and 1000 μM). Previous experiments also revealed Pro-SA-AuNCs with responsive fluorescence quenching toward TCs and a linear response within the tested concentration range (Figure B). However, the probe’s preparation conditions still require further optimization, and its application scope and operational conditions call for additional exploration. To address these aforementioned gaps, this study aimed to develop an analytical method featuring simple operation, high sensitivity, and superior selectivity, achieved by optimizing the preparation conditions of Pro-SA-AuNCs and applying the optimized probe to the fluorescence analysis of TCs. Furthermore, the proposed reliable method for TCs detection in complex matrices (e.g., real-world food and feed) was established, which will enhance the probe’s practical potential for safeguarding food safety and protecting public health. Material and Methods Materials and Instrumentation L-proline, Sodium Alginate, sodium chloride (NaCl), chloroauric acid trihydrate (HAuCl4·3H2O), phosphate-buffered saline (PBS), chlortetracycline (CTC, ≥80%), oxytetracycline (OTC, 98%) and tetracycline hydrochloride (TC, 96 %) were purchased from Aladdin Reagent Co. Ltd. (Shanghai, China). n-Hexane, trichloroacetic acid, chloroform, ethanol, hydrochloric acid, and sodium hydroxide were obtained from the National Pharmaceutical Group Corporation (Shanghai, China). Unless otherwise specified, all reagents were of analytical grade and used without further purification. Milk powder, honey, and feed samples were acquired from local supermarkets. All aqueous solutions were prepared using deionized water. UV–vis absorption spectra were recorded on a UV-2450 spectrophotometer (Shimadzu, Kyoto, Japan). Fourier-transform infrared (FT-IR) spectra were collected using a Nicolet Avatar 300 spectrometer (Thermo Fisher Scientific, Waltham, USA) with KBr pellets. Steady-state fluorescence spectra were measured on a Cary Eclipse fluorescence spectrophotometer (Agilent Technologies, Santa Clara, USA). Fluorescence lifetime measurements were performed using an FS5 spectrofluorometer (Edinburgh Instruments, Livingston, UK). X-ray photoelectron spectroscopy (XPS) analysis was conducted on a Thermo Fisher Scientific ESCALAB 250Xi spectrometer (Waltham, USA). Hydrodynamic diameter and zeta potential were determined using a Litesizer 500 particle size and zeta potential analyzer (Anton Paar, Graz, Austria). Transmission electron microscopy (TEM) images were obtained on a Tecnai G2 F20 field-emission transmission electron microscope (FEI, Hillsboro, USA) operated at an accelerating voltage of 200 kV. Adjustable pipettes (Thermo Fisher Scientific, Waltham, USA) were used for sample and reagent dispensing. A desktop high-speed centrifuge (Shanghai Medical Analysis Instrument Factory, Shanghai, China) was employed for sample pretreatment. An ultrasonic cleaner (KQ218, Kunshan Ultrasonic Instrument Co., Ltd., Jiangsu, China) was used for solution dispersion. MALDI-TOF-MS analysis was performed on a microflex LRF mass spectrometer (Bruker Daltonics, Bremen, Germany) equipped with a nitrogen laser (337 nm, 3 ns pulse width). Optimization of Preparation Conditions for Pro-SA-AuNCs The preparation conditions of Pro-SA-AuNCs were systematically optimized by evaluating the effects of key parameters on the fluorescence intensity of the product: First, the volume ratios of Pro (2 M) to HAuCl4·3H2O (5 mM) were optimized. The two solutions were mixed at volume ratios of 1:6, 1:7, 1:7.5, 1:8, 1:8.5, and 1:9. Each mixture was incubated in a constant-temperature water bath at 90 °C (as a benchmark temperature for preliminary optimization) for 10 min, followed by centrifugation at 8000 rpm for 10 min to remove precipitates. The fluorescence intensity of the resulting supernatant was measured. On the basis of the optimized Pro/HAuCl4 volume ratio, the concentration of SA (as a co-stabilizing agent) was further optimized. Aliquots of 50 μL SA solutions with different concentrations (0, 2.5, 5.0, 7.5, 10, 12.5, and 15 mM) were separately added to the Pro/HAuCl4 mixture. Each reaction system was incubated at 90 °C (as a benchmark temperature for preliminary optimization) for 10 min and centrifuged at 8000 rpm for 10 min. The fluorescence intensity of the supernatant was measured, and the SA concentration corresponding to the strongest fluorescence emission was selected as the optimal concentration. Moreover, the reaction temperature and time during Pro-SA-AuNC synthesis were investigated. For temperature optimization, the optimized reaction system was incubated at various temperatures (25 °C, 35 °C, 45 °C, 55 °C, 65 °C, 75 °C, 85 °C, 95 °C, and 100 °C) for 10 min. For time optimization, the system was incubated at the optimal temperature of 95 °C for various durations (4, 6, 8, 10, 12, 14, 16, 20, and 40 min). After each reaction, the fluorescence intensity of the supernatant was measured. The reaction conditions that resulted in the highest fluorescence intensity of Pro-SA-AuNCs were selected as the optimal conditions. Characterization of General Properties of Pro-SA-AuNCs General Protocol for Spectroscopic and Structural Characterization The structural morphology, elemental composition, optical properties, and surface chemistry were systematically characterized using the following techniques. The nanostructure and particle size were observed using a high-resolution transmission electron microscope operated at a voltage of 200 kV. Fluorescence excitation and emission spectra were recorded on a fluorescence spectrophotometer, and the slit widths were set to 10 nm. UV–vis absorption spectra were measured using a UV spectrophotometer with a standard quartz cuvette. The scanning range was set from 200 to 800 nm. FT-IR spectroscopy was used to characterize the surface functional groups. XPS analyses were performed on a Thermo Fisher Scientific ESCALAB 250Xi spectrometer (USA) at an X-ray power of 150 W. All binding energies were calibrated against the C 1s core-level peak at 284.8 eV (assigned to sp2-hybridized carbon) to correct for surface charging effects. High-resolution spectra of C 1s, N 1s, O 1s, Au 4f, and Cl 2p were collected to determine the elemental composition and chemical states of Pro-SA-AuNCs. Stability and Selectivity of Pro-SA-AuNCs The zeta potential of Pro-SA-AuNCs was measured to assess its stability. To systematically evaluate the environment stability of Pro-SA-AuNCs, we conducted three types of stability experiments, including pH stability, storage stability and photostability, as follows: Pro-SA-AuNCs were dispersed in PBS buffers (10 mM) with pH ranging from 1.0 to 11.0 and incubated at room temperature in the dark for 30 min to investigate the effect of pH on probe stability. To assess the photostability of Pro-SA-AuNCs, we subjected the probe to continuous measurement for 3600 s at an excitation wavelength of 322 nm in the Time Course Parameters mode of a fluorescence spectrophotometer (RF-5301PC, Shimadzu Corporation, Japan). Fluorescence intensities were monitored at regular intervals (0,5,10, 15,20, 25, and 30 days) to evaluate long-term storage performance. The selectivity of the probe toward TCs (TC, CTC, and OTC) was evaluated by comparing its fluorescence response to TCs and common interfering substances, including common cations (Na+, K+, Mg2+, and Ca2+; final concentration: 1 mM), small-molecule interferents (ascorbic acid [AA], l-glutathione [L-GSH], arginine [Arg], dopamine [DA], His, and lysine [Lys]; final concentration: 1 mM), and other antibiotics (sulfamethoxazole [SMX], vancomycin [VA], amoxicillin [AM], and streptomycin [Str]). The final concentrations of TC, CTC, and OTC were adjusted to 500 μM. All analytes were dissolved in PBS buffer (pH 4.5, 10 mM). For each test, 500 μL of Pro-SA-AuNCs solution was mixed with 1.5 mL analyte solution (or PBS as blank). After incubation at 25 °C for 7 min, the fluorescence intensities of each mixture were measured. Determination of TCs Using Pro-SA-AuNCs Response Principle Under the optimized preparation conditions of Pro-SA-AuNCs, the probe was incubated with TCs at gradient concentrations. To deduce the response principle of Pro-SA-AuNCs toward TCs, we first analyzed the fluorescence quenching behavior using the Stern–Volmer eq (eq )­ 1 I0/I=1+KSV[Q] where I 0 and I represent the fluorescence intensities of Pro-SA-AuNCs in the absence and presence of TCs, respectively; K SV is the Stern–Volmer quenching constant (μM–1), and [Q] denotes the concentration of TCs (μM). To further verify the quenching mechanism, we conducted fluorescence lifetime measurements of Pro-SA-AuNCs before and after TCs addition using an FS5 spectrofluorometer (Edinburgh Instruments, UK). UV–Vis absorption spectra of the Pro-SA-AuNCs-TCs system were recorded to examine spectral overlap between the probe’s emission and TCs’ absorption. , The fluorescence quenching degrees at different temperatures (25 °C, 45 °C and 65 °C) were recorded to distinguish whether the quenching mechanism is static or dynamic. These combined analyses enabled the confirmation of the response mechanism between Pro-SA-AuNCs and TCs. Optimization of TCs Detection Conditions To improve the sensitivity and reliability of TCs detection using Pro-SA-AuNCs, we optimized key experimental conditions, including incubation temperature, reaction system pH (3, 4, 4.5, 5, 6 and 7), incubation temperature (25, 35, 45, 55, 65 and 75 °C), and time (1, 3, 5, 6, 7, 9 and 10 min), by monitoring the fluorescence quenching ratio. Standard Curve Construction A total of 36 centrifuge tubes (2 mL) were divided into three groups (12 tubes per group) corresponding to TC, CTC, and OTC, with tubes in each group numbered 0–11. For each group (TC, CTC, OTC), a series of gradient concentrations of TCs was prepared using a 2 mM TCs stock solution. Specifically, different volumes of the TCs stock solution were added to tubes 1–11 to achieve final TCs concentrations of 5, 10, 20, 40, 60, 80, 100, 200, 400, 600, and 800 μM, respectively, whereas tube 0 contained no TCs. Then, 500 μL Pro-SA-AuNCs and PBS (pH 4.5) buffer were added to make a total volume of 2 mL in each tube, and the mixtures were shaken well. All tubes were placed in a thermostatic incubator at 25 °C for 7 min to ensure sufficient interaction between Pro-SA-AuNCs and TCs. After incubation, the fluorescence intensity of each mixture was measured. The fluorescence quenching ratio (I 0/I, where I 0 is the fluorescence intensity of the blank group, tube 0; I is the intensity of the TCs-containing groups) was calculated for each TCs concentration. Relevant linear regression equations were fitted with I 0/I as the vertical axis and TCs concentration as the horizontal axis. Pretreatment of Complex Samples Honey Sample A total of ∼0.5 g honey (precision: 0.002 g) was weighed accurately into a 50 mL centrifuge tube. Next, 5 mL ethanol was added. The mixture was sonicated for 10 min to fully dissolve the honey matrix, and then spiked with TCs standard solutions at different concentrations (24, 30, and 36 μM) to simulate real spiked samples. The mixture was diluted to 100 mL with PBS buffer (pH 4.5, 10 mM, 20-fold volume dilution), sonicated for an additional 5 min to ensure homogeneous dispersion of TCs, and centrifuged at 4000 rpm for 20 min to remove insoluble impurities. The supernatant was collected and filtered through a 0.45 μm microporous membrane, and the filtrate was reserved for subsequent detection. Milk Samples Milk powder (∼0.5 g, precision: 0.002 g) was weighed accurately and dissolved in 10 mL of PBS buffer (pH 4.5) with gentle stirring. The milk was spiked with TCs standard solutions (24, 30, and 36 μM) to prepare spiked milk samples. To remove interfering matrix components (protein and lipids) while retaining TCs in milk, we transferred 1 mL spiked milk solution to a 15 mL tube, added it with 2 mL chloroform and 7 mL 10% (w/v) trichloroacetic acid, and vortexed it vigorously for 3 min. The mixture was centrifuged at 5000 rpm for 10 min to separate the phases, and the supernatant was collected and filtered through a 0.45 μm microporous membrane to remove large particles. Animal Feed Samples Animal feed (∼1 g, precision: 0.002 g) was weighed accurately, added with 10 mL ethylenediaminetetraacetic acid (EDTA)-McIlvaine buffer (pH 4.5) (optimized for TCs extraction from complex solid matrices), and spiked with TCs standard solutions (24, 30, 36 μM). The suspension was vortexed for 10 min, sonicated for 10 min, and then centrifuged at 10,000 rpm for 10 min to separate the extract from the residue. The supernatant was transferred to a clean container, and the residue was washed twice with 5 mL EDTA-McIlvaine buffer under the same conditions. All supernatants and filtrates were combined and stored for further analysis. Pretreatment of Solid-Phase Extraction (SPE) Column SPE fulfilled two core functions in the sample pretreatment process: First, it enriched TCs residues from complex matrices to meet the requirements for limit of detection (LODs) of the Pro-SA-AuNCs fluorescence sensing system; second, it removed interfering substances (e.g., proteins, lipids, and other small molecules) that may trigger nonspecific fluorescence quenching of Pro-SA-AuNCs or result in false-positive detection outcomes. The SPE column was rinsed with 3 mL methanol at a flow rate of ∼1 mL/min to activate the stationary phase and then washed with 3 mL deionized water to remove residual methanol. This aqueous washing step also effectively removed hydrophilic potential interferents (e.g., inorganic cations such as Fe3+, Na+, and K+) that were not retained on the reversed-phase C18 sorbent. The pH of the pretreated sample filtrate was adjusted to pH 7.0. Then, the solution was loaded onto the activated SPE column. A bulb syringe was used to apply slow pressure, controlling the flow rate at ∼1 mL/min, to ensure sufficient adsorption of TCs onto the C18 phase. Subsequently, the enriched TCs from the column was eluted with 1.5 mL methanol. A bulb syringe was used to control the elution flow rate at ∼0.5 mL/min, with a total elution time of 2 min to ensure complete recovery of TCs. The eluate was collected and analyzed immediately using the Pro-SA-AuNCs fluorescence assay. To evaluate the efficiency of the water washing step in the removal of cationic interferents, we selected Fe3+ as a model interferent due to its strong fluorescence quenching effect on Pro-SA-AuNCs. The SPE column was loaded with 600 μM Fe3+ solution, followed by sequential washes with 3 mL deionized water for four cycles. The fluorescence intensity of the eluate after each wash was measured under the same conditions as TCs detection. The recovery of fluorescence intensity was calculated relative to the deionized water blank. Recovery Experiment of TCs in Complex Samples To evaluate the accuracy and reliability of the proposed method for TCs detection in complex matrices, we conducted recovery experiments on spiked honey, milk, and animal feed samples. A total of 500 μL pretreated filtrate from milk, honey, and animal feed samples was collected into a 2 mL centrifuge tube. Then, 500 μL Pro-SA-AuNCs solution was added to each tube. The mixture was mixed thoroughly by vortexing and incubated in the dark at 25 °C for 7 min to ensure sufficient interaction between Pro-SA-AuNCs and TCs. Subsequently, the fluorescence intensity of each sample was measured. Each sample was analyzed in triplicate to ensure data reproducibility. The recovery rate and relative standard deviations (RSDs) were calculated to assess method precision. Results Feasibility Verification and Preparation Condition Optimization of Pro-SA-AuNCs The one-pot strategy simplifies the synthesis process, avoiding complex post-modification steps, which is consistent with the “green, efficient” synthesis concept of fluorescent nanomaterials reported in previous studies. The light-blue emission of Pro-SA-AuNCs was visually observable under UV irradiation, which confirmed the successful formation of the fluorescent probe. To maximize the fluorescence intensity and stability of Pro-SA-AuNCs, we systematically optimized key synthesis parameters, with the fluorescence intensity of the product as the evaluation index (Figure ). The optimized results are as follows. Optimization of synthesis conditions for Pro-SA-AuNCs. (A) Fluorescence intensity vs volume ratio of Pro (2 M) to HAuCl4 (5 mM); optimal ratio = 1:8 (v/v). (B) Fluorescence intensity vs SA concentration (0–15 mM); optimal concentration = 7.5 mM. (C) Fluorescence intensity vs reaction temperature (25 °C–100 °C); optimal temperature = 95 °C. (D) Fluorescence intensity vs reaction time (4–40 min); optimal time = 10 min. All experiments were performed in triplicate (n = 3), and data are presented as mean ± SD. The volume ratio of Pro (2.0 M) to HAuCl4 (5 mM) directly affected the reduction efficiency of Au3+ and the surface passivation of AuNCs. As shown in Figure A, the fluorescence intensity of Pro-SA-AuNCs first increased with the increase in Pro dosage, reaching a maximum at a volume ratio of 1:8 (Pro: HAuCl4), and then decreased with further increases. This phenomenon was attributed to the fact that insufficient Pro concentration led to incomplete reduction of Au3+, and its excess may cause ligand aggregation on the AuNC surface, quenching fluorescence via nonradiative transitions. SA, as a negatively charged polysaccharide, enhanced the colloidal stability of Pro-SA-AuNCs via electrostatic repulsion and hydrogen bonding. As displayed in Figure B, the fluorescence intensity of Pro-SA-AuNCs increased with the increase in SA concentration from 2.5 mM to 7.5 mM, and remained stable at concentrations above 7.5 mM. When SA concentration was less than 7.5 mM, insufficient surface coverage led to AuNC aggregation and fluorescence quenching; when SA exceeded 7.5 mM, excess SA did not further improve stability but increased background interference. Reaction time and temperature regulated the nucleation and growth kinetics of AuNCs. As revealed in Figure C, the fluorescence intensity increased with temperature from 25 to 95 °C, attained a maximum at 95 °C, and decreased slightly at 100 °C (possibly due to AuNC sintering). Figure D shows that the fluorescence intensity reached a plateau after 10 min of reaction at 95 °C, which indicates complete formation of Pro-SA-AuNCs. Prolonged reaction time (>10 min) did not increase the fluorescence intensity but caused energy waste. Thus, the optimal reaction conditions were 95 °C for 10 min. Based on the above optimization results, we obtained the standardized synthesis steps of Pro-SA-AuNCs. A total of 4.6052 g Pro powder was weighed accurately, dissolved in 20 mL deionized water, and stirred until complete dissolution to obtain a 2.0 M Pro solution. Then 0.0162 g SA powder was weighed, dissolved in double-distilled water, and diluted to 10 mL to prepare a 7.5 mM SA solution. Next, 1 g chloroauric acid powder was dissolved in deionized water to prepare a 25.39 mM HAuCl4 stock solution (stored in the dark at 4 °C). Under continuous magnetic stirring, Pro (2.0 M, 100 μL), HAuCl4 (5 mM, 800 μL), and SA solution (7.5 mM, 50 μL) were sequentially added into a 5 mL centrifuge tube. The mixture was incubated at 95 °C for 10 min and then retrieved, and the reaction solution was centrifuged at 8000 rpm for 10 min. The supernatant was collected and stored in the dark at 4 °C for subsequent experiments. Characterization of Pro-SA-AuNCs Structural, Morphological, and Spectroscopic Characteristics Water-soluble Pro-SA-AuNCs were synthesized using Pro as a reducing agent and co-stabilizer and SA as an additional stabilizer to enhance fluorescence performance. Pro-SA-AuNCs exhibited a significantly higher fluorescence intensity than Pro-AuNCs (synthesized without SA) under the same experimental conditions (Figure A, a and b). This enhancement was attributed to SA’s capability to restrict intramolecular motion of Pro on the AuNCs surface (via hydrogen bonding and electrostatic interactions), which reduced nonradiative energy loss, consistent with the “RIM” mechanism reported for fluorescent nanoclusters. The fluorescence spectra of Pro-SA-AuNCs showed an optimal excitation wavelength of 322 nm and a strong emission peak at 398 nm. Visually, the Pro-SA-AuNC solution appeared nearly colorless under ambient light while emitting intense blue fluorescence under UV irradiation (Figure A,c and d). Characterization of the synthesized Pro–SA–AuNCs. (A) Optical properties. (a) Fluorescence excitation (λex = 322 nm) and emission (λem = 398 nm) spectra of the Pro-AuNCs; (b) fluorescence excitation (λex = 322 nm) and emission (λem = 398 nm) spectra of Pro-SA-AuNCs. (c, d) Photographs of the Pro-SA-AuNC solution under ambient light and 365 nm UV illumination, respectively, demonstrating its strong blue fluorescence. (B) Morphology and size distribution. Representative TEM (scale bar: 10 nm) and HRTEM (scale bar: 5 nm; inset: 2 nm) images revealing the ultrasmall, quasi-spherical, and well-dispersed morphology of the AuNCs. The corresponding particle size histogram (e) confirms a narrow size distribution centered at ∼2.41 nm. (C) Compositional verification via UV–vis spectroscopy. Comparison of the absorption spectra of the precursors (HAuCl4, Pro, and SA) and the final Pro-SA-AuNC product. The distinct absorption profile of the AuNCs, differing from its precursors, confirms the formation of a new nanomaterial. (D) Surface chemical analysis by FT-IR. FT-IR spectra of Pro, SA, and Pro-SA-AuNCs. The shifts and changes in characteristic vibrational bands (e.g., –COOH, –NH2) suggest the successful coordination and capping of AuNCs by Pro and SA molecules. (E,F) Elemental composition and chemical state by XPS. (E) Full XPS survey spectrum confirmed the presence of C, N, O, Cl, and Au. (F) High-resolution Au 4f XPS spectrum. The dominant low-binding-energy spin–orbit doublet (Au 4f5/2 at ∼83.4 eV and Au 4f5/2 at ∼87.0 eV) is assigned to metallic Au(0). The minor high-binding-energy shoulder corresponds to a small fraction of Au­(I) species, which are stabilized by coordination with N/O atoms from the Pro and SA ligands. TEM was used to characterize the morphology and size of Pro-SA-AuNCs. As shown in Figure B, Pro-SA-AuNCs exhibited a spheroidal shape with uniform size and good dispersibility, with an average particle size of 2.41 nm. High-resolution TEM (HRTEM) images revealed clear lattice fringes with a lattice spacing of 0.23 nm, corresponding to the crystal plane of metallic Au. The lattice structure indicates the good crystallinity of Pro-SA-AuNCs, which is beneficial for stable fluorescence emission, avoiding the fluorescence quenching caused by amorphous Au aggregation. The UV absorption spectrum of Pro-SA-AuNCs (Figure C) showed a broad absorption band in the range of 240–400 nm, and pure Pro exhibited negligible absorption in this region. The spectral profile confirmed the successful formation of fluorescent AuNCs, aligning with the characteristic emission of small-sized AuNCs (<3 nm) reported in previous studies. This finding arose from electronic transitions between discrete energy levels of the Au core (molecular-like properties) rather than the surface plasmon resonance (SPR) effect of larger Au nanoparticles (>10 nm). FT-IR spectroscopy (Nicolet Avatar 300, Thermo Fisher Scientific, USA) was employed to identify surface functional groups of Pro-SA-AuNCs and verify the binding of Pro and SA to the Au core (Figure D). The broad absorption band at ∼3470 cm–1 was attributed to the O-H stretching vibrations (ν­(O-H)) from SA and the overlapped N-H stretching vibrations (ν­(N-H)) from Pro. Compared with pure Pro, the attenuation/disappearance of the broad peak at ∼3070 cm–1 (corresponding to the N-H stretching vibrations of the protonated amino group (-NH2 +) in Pro zwitterion) suggested coordination of the amino nitrogen atoms with Au ions on the cluster surface, consistent with the formation of Au-N bonds. The intense peak at ∼1630 cm–1 was assigned to the asymmetric stretching vibration of carboxylate groups (νas­(COO–)) from both Pro and SA, while the peak at ∼1400 cm–1 corresponded to the symmetric stretching vibration of carboxylate groups (νs­(COO–)). The presence of these carboxylate bands, together with the absence of a peak at ∼1710–1730 cm–1 (typical for protonated carboxylic acid C=O stretching), confirmed that the carboxyl groups of Pro and SA existed predominantly in their deprotonated forms and participated in the stabilization of AuNCs through Au-O coordination or electrostatic interactions. These spectral features collectively indicate the successful anchoring of Pro and SA onto the AuNC surface through coordination interactions (Au-N and Au-O) and interligand hydrogen bonding. XPS was used to investigate the elemental composition and chemical state of Au in Pro-SA-AuNCs. The full XPS survey spectrum (Figure E) showed characteristic peaks of Cl 2p (197 eV), C 1s (285 eV), N 1s (400 eV), and O 1s (530 eV), and Au 4f, confirming the presence of these elements in Pro-SA-AuNCs (Cl originated from unreacted HAuCl4). High-resolution Au 4f XPS spectrum (Figure F) deconvoluted into two spin–orbit doublets. The dominant low-binding-energy doublet (Au 4f7/2 at ∼83.4 eV and 4f5/2 at ∼87.0 eV) is assigned to metallic Au(0), while the minor high-binding-energy doublet corresponds to Au­(I) species stabilized by coordination with N/O atoms from the Pro and SA ligands. This Au(0)/Au­(I) mixed-valence state, stabilized by dual ligands, enhanced fluorescence stability and colloidal dispersibility compared with single-valence AuNCs. The above-mentioned findings confirm the successful synthesis of Pro-SA-AuNCs, with Pro and SA costabilizing the Au core. The probe exhibited uniform size (∼2.41 nm), good crystallinity, intense blue fluorescence, and stable surface functionalization. To verify the gold core composition of Pro-SA-AuNCs, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry was conducted on a microflex LRF instrument (Bruker Daltonics) equipped with an N2 laser (337 nm), using DCTB as the matrix in positive ion mode. The linear mode spectrum (Figure S1A)­exhibited a base peak at m/z 759.795, assigned to [Au4(Pro)3(SA)1 + Na]+, establishing Au4 as the predominant gold core species of Pro-SA-AuNCs. Two additional envelope peaks at m/z 1710.27 and 3490.655 were reasonably attributed to SA-bridged Au4 assemblies or matrix-associated adducts, rather than discrete high-nuclearity gold clusters. High-resolution reflectron mode further captured fragment ions at m/z 602.994 and 663.052, corresponding to Au2 and Au3 species, corroborating the formation pathway with Au4 as the fundamental building block. Collectively, the mass spectrometric data reveal that Pro-SA-AuNCs comprise Au4 as the core unit, with sodium alginate functioning as a costabilizer that modifies the cluster surface via hydrogen bonding and electrostatic interactions without altering the gold core sizeconsistent with the probe’s characteristic single blue emission at 398 nm. Stability and Selectivity of Pro-SA-AuNCs Zeta potential of Pro-SA-AuNCs was measured in ultrapure water using a nano laser particle size analyzer. The results showed an average surface potential of −17.0 ± 0.80 mV (Figure A), which indicated that the Pro-SA-AuNCs carried a negative surface charge. This negative charge induced electrostatic repulsion between adjacent clusters, which prevented aggregation and ensured good colloidal stability, which was consistent with the TEM revealing uniform dispersion. Evaluation of stability and selectivity of Pro-SA-AuNCs. (A) Colloidal stability. The zeta potential distribution of Pro-SA-AuNCs showed an average value of −17.0 ± 0.80 mV, which denoted good colloidal stability in aqueous solutions due to sufficient electrostatic repulsion. (B) Chemical stability. The fluorescence intensity of Pro-SA-AuNCs remained stable over a wide pH range (1–11), which demonstrated its robustness under varied chemical conditions. (C) Photostability. The fluorescence intensity of Pro-SA-AuNCs maintained over 103.77% of its initial value after continuous UV light (322 nm) irradiation for 60 min, which proved their high photobleaching resistance. (D) Long-term stability. The fluorescence intensity of Pro-SA-AuNCs displayed a negligible decline after storage at 4°C in the dark for 30 days, which confirmed its excellent shelf life for practical applications. (E) Selectivity toward TCs. The fluorescence quenching response [(I 0–I)/I] of Pro-SA-AuNCs was significantly higher toward three target TCs (TC, CTC, OTC; 500 μM each) than toward various potential interfering substances (1 mM), confirming their high specificity. (F) Anti-interference capability. The quenching response toward each TC (500 μM) remained largely unchanged in the presence of a high concentration (1 mM) of interfering substances, demonstrating reliable detection in complex matrices. All data were presented as mean ± SD (n = 3). The fluorescence intensity of Pro-SA-AuNCs was monitored in PBS buffered with pH ranging from 1 to 11. As illustrated in Figure B, no significant fluctuation in fluorescence intensity was observed across this pH range, which implies that the Pro-SA-AuNCs maintained a stable fluorescence in neutral, weakly acidic, and weakly alkaline environments. Such stability expands the probe’s applicability to complex samples with varying pH (e.g., acidic honey, neutral milk). The photostability of Pro-SA-AuNCs was evaluated by continuous excitation at 322 nm for 60 min. After 60 min continuous irradiation, the fluorescence intensity remained at 103.77% of the initial intensity (Figure C) and then slightly increased (possibly due to transient ligand rearrangement). This photostability avoided signal attenuation during prolonged measurement, which is critical for reliable TCs detection. Pro-SA-AuNC solutions were stored in the dark at 4 °C, and their fluorescence properties were monitored every 5 days for 1 month. As shown in Figure D, the Pro-SA-AuNCs exhibited good and long-term storage stability. Under the same experimental conditions as TCs detection, the fluorescence quenching effect of TCs (500 μM) and other interferents (1 mM) was investigated. The interferents were introduced partially to simulate the types of matrix interferences typically encountered in real samples, including other antibiotics (SMX, VA, AM, and Str), small molecules (AA, L-GSH, Arg, DA, and Lys), and common cations (Na+, K+, Mg2+, and Ca2+). As shown in Figure E, TCs significantly quenched the fluorescence of Pro-SA-AuNCs, and some cations (Na+, K+, and Ca2+) and antibiotics, such as VA at 1 mM (2-fold higher than TCs), exhibited certain interference effects. However, given that the SPE column was used to enrich TCs from samples, these aqueous interferents were removed during the washing step. Their final concentrations were far below 1 mM and thus did not interfere with the detection. To further verify the probe’s reliability in complex matrices, mixtures of TCs (500 μM) and each interferent (1 mM) were added to Pro-SA-AuNCs, and their fluorescence responses were compared with that of TCs alone. As shown in Figure F, the fluorescence quenching ratio (I 0–I)/I of the mixed systems was nearly identical to that of TCs alone, with no significant deviation (<5%). This finding indicates that the presence of interferents did not disrupt the specific interaction between TCs and Pro-SA-AuNCs, implying the probe’s reliable anti-interference capability. Response Principles and Detection Conditions of Pro-SA-AuNC toward TCs Response Principle of Pro-SA-AuNC toward TCs Comprehensive spectral analyses were performed to investigate the fluorescence quenching mechanism of Pro-SA-AuNCs induced by TC. As revealed in Figure A, the UV absorption spectrum of TC partially overlapped with the fluorescence excitation spectrum of Pro-SA-AuNCs, which suggests the possible involvement of fluorescence resonance energy transfer (FRET) or the inner filter effect (IFE). In addition, the fluorescence quenching mechanism may involve static (driven by electrostatic adsorption) and dynamic quenching. − Mechanism of TC-induced fluorescence quenching and analytical performance of Pro-SA-AuNCs. (A) Spectral overlap between Pro-SA-AuNC emission spectrum and TC UV absorption spectrum (supporting IFE mechanism). (B) Fluorescence lifetime decay curves of Pro-SA-AuNCs before and after addition of 100 μM TC at different temperatures (confirming static quenching). (C) Standard curve of Pro-SA-AuNCs for TC. (D) Gradual fluorescence quenching within increased TC concentration. First, fluorescence lifetime measurements (Figure B) revealed no significant changes upon interaction with TCs. The average fluorescence lifetimes of the probe alone at 25 °C, 45 °C, and 65 °C were 4.83, 4.66, and 3.98 ns, respectively. Upon addition of TCs (100 μM), no significant decrease in fluorescence lifetime was observed: 4.80 ns (25 °C), 4.80 ns (45 °C), and 4.39 ns (65 °C). As the quencher (TCs) did not reduce the lifetime of the probe, this result excludes dynamic quenching, which is characterized by a decrease in lifetime. Given that static quenching and IFE did not decrease fluorescence lifetime, the essentially unchanged lifetimes of Pro-SA-AuNCs after the addition of TCs across all temperatures indicate that the quenching process was dominated by static quenching and the IFE rather than FRET. According to the Stern–Volmer equation: The linear equation of Pro-SA-AuNCs for detecting TC is expressed as follows (Figure C) I0/I=0.00278×[TC]+0.98947 Therefore,KSV=0.00278μM−1 For dynamic quenching, the equation must satisfy eq 2 KSV=kq×τ0 where k q denotes the molecular quenching rate constant (M–1·s–1), and τ0 indicates the fluorescence lifetime (ns) of the probe without the addition of TC. Taking τ0 = 4.83 ns and K SV = 0.00278 μM–1 into the equation, k q = 5.76 × 1011 M–1·s–1 was calculated, and it is higher than the maximum value of the molecular quenching rate constant (approximately 1 × 1010 M–1·s–1). Therefore, dynamic quenching was not the mechanism by which TC quenched the probe. Moreover, the absorption spectrum measurements (Figure D) revealed that the probe’s spectrum shifted with the variation in quencher concentrations. Detailed fluorescence lifetime fitting parameters, including lifetime components (τ1, τ2), amplitude contributions (A1, A2), average lifetime (τavg), and fitting quality (χ2), are provided in Table S1 (Supporting Information). Then, the quenching effect decreased with increased temperatures (25 °C–75 °C, Figure A) at the same quencher concentration (100 μM TC), which supports a static quenching mechanism. Optimization of TCs detection conditions. (A) Effect of incubation temperature (25 °C–75 °C, optimal: 25 °C). (B) Hydrodynamic diameter distribution of Pro-SA-AuNCs before and after reaction with TCs (increased diameter suggests nanoaggregation). (C) Effect of reaction pH (3.0–7.0, optimal: 4.5). (D) Effect of incubation time (1–10 min, optimal: 7 min). (E) Time-dependent quenching response at different TC concentrations (50, 100, and 200 μM). All data are presented as mean ± SD (n = 3). Furthermore, to verify the aggregation-induced static quenching mechanism, the hydrodynamic diameter of Pro-SA-AuNCs was monitored before and after reaction with TCs at varying concentrations. As shown in Figure B, the pristine probe exhibited a narrow size distribution centered. Upon reaction with TCs at increasing concentrations (50, 200, and 800 μM), the hydrodynamic diameter progressively increased, accompanied by broader distribution profiles and shifted peak positions toward larger sizes. This concentration-dependent size increase confirms TC-induced nanoaggregation of Pro-SA-AuNCs, providing direct evidence for the static quenching mechanism involving ground–state complex formation. In summary, IFE and static quenching were the main mechanisms underlying the fluorescence quenching of TCs by Pro-SA-AuNCs. The clear identification of this dual mechanism, with elimination of dynamic quenching, advances analytical chemistry by providing a reliable mechanistic basis for the optimization of TCs detection via AuNC probes. Optimization of Detection Conditions of Pro-SA-AuNCs toward TCs The detection conditions for Pro-SA-AuNC/TC system, including reaction temperature, pH, and incubation time, were optimized individually, with the results presented in Figure A,C–E. First, the influence of reaction temperature on the system’s fluorescence intensity was evaluated. The results demonstrate that the optimal experimental temperature was 25 °C (Figure A). The decrease in quenching efficiency at elevated temperatures (25 °C–75 °C) is consistent with a static quenching mechanism, where higher temperatures destabilized the nonfluorescent ground–state complex formed between Pro-SA-AuNCs and TCs. This finding further supports the mechanistic conclusion derived from fluorescence lifetime and Stern–Volmer analyses. Moreover, the optimal temperature of 25 °C is favorable for practical operation, as it requires no additional heating or cooling equipment. Preliminary tests indicated that Pro-SA-AuNCs exhibited superior stability in acidic environments compared with alkaline conditions. Consequently, the pH optimization range was set at 3–7. As illustrated in Figure C, the fluorescence intensity ratio of the system initially increased with rising pH, peaked at pH 4.5, and subsequently decreased. Therefore, the optimal reaction pH was 4.5. This pH-dependent response can be attributed to two factors: the surface charge state of Pro-SA-AuNCs and the molecular form of TCs. At pH 4.5, the carboxyl groups on the Pro-SA-AuNCs surface were partially deprotonated, and TCs predominantly existed as zwitterions. This electrostatic state facilitated strong attraction and stable complex formation between the probe and TCs. Under more acidic conditions (pH < 4.0), excessive protonation of carboxyl groups weakened electrostatic interactions; under near-neutral conditions (pH > 6.0), it increased electrostatic repulsion. In addition, altered TC speciation reduced the quenching efficiency. Therefore, pH 4.5 was selected as the optimal reaction pH. As revealed in Figure D, the quenching ratio increased sharply within the first 7 min and subsequently reached a plateau. To evaluate the robustness of this kinetic behavior, the time-dependent response was investigated at three representative TC concentrations (50, 100, and 200 μM). As shown in Figure E, all three concentrations exhibited similar rapid equilibration kinetics, reaching plateau values at approximately 7 min. This parallel time-course behavior indicates that the binding kinetics are rapid and consistent across the tested concentration range. Notably, the plateau quenching ratios increased progressively with increasing TC concentrations, confirming the concentration-dependent nature of the static quenching response. Such rapid and reproducible equilibration is advantageous for high-throughput screening applications. The 7 min plateau reflects the establishment of thermodynamic equilibrium for ground–state complex formation between Pro-SA-AuNCs and TCs. Therefore, an incubation time of 7 min was selected as the optimal condition for subsequent experiments. Based on these findings, a broader concentration range (5–800 μM) was examined to establish a quantitative calibration curve. Determination of TCs Using Pro-SA-AuNCs Standard Curve Establishment Within a specific concentration range, TC, CTC, OTC, and mixed TCs exerted a quenching effect on the fluorescence of Pro-SA-AuNCs, accompanied by a red shift in the fluorescence emission wavelength. Moreover, the fluorescence intensity ratio I 0/I showed a linear positive correlation with the concentrations. To enable quantitative detection of TC, CTC, OTC, and TCs in unknown samples, we established standard curves with a linear relationship between I 0/I and the corresponding analyte concentrations (Figure C for TC; Figure D–F for OTC, CTC, and TCs). The linear fitting equations for each analyte and TCs within their respective concentration ranges (5–800 μM) are as follows: I0/I=0.00278×[TC]+0.98947 I0/I=0.00265×[OTC]+1.00876 I0/I=0.00430×[CTC]+1.03489 I0/I=0.00344×[TCs]+1.05079 Fluorescence intensity gradient quenching and standard curves of OTC, CTC, and TCs. (A–C) Fluorescence intensity gradient quenching trends as the concentration of (A) OTC, (B) CTC, and (C) TCs increased from 5 μM to 800 μM. (D–F) Standard curves of Pro-SA-AuNCs for (D) OTC, (E) CTC, and (F) TCs. All data are presented as mean ± SD (n = 3). The calculated LODs of Pro-SA-AuNCs for TC, CTC, OTC, and TCs were 0.23, 0.16, 0.45, and 0.54 μM, respectively. For the quantitative analysis of mixed TCs, TC was used as the reference standard: OTC and CTC concentrations were converted to equivalent TC concentrations, the total converted TC concentration was plotted on the x-axis, and I 0/I was plotted on the y-axis to fit a linear regression curve. A comparison with previously reported detection methods (Table ) − showed that the proposed Pro-SA-AuNC-based method exhibited relatively lower LODs and a wider linear detection range. This performance advantage is a key advancement in analytical chemistry for TCs, as it balances high sensitivity and broad applicability (for varying contamination levels) better than many existing fluorescent probes. 1Comparison of This Work with Other Fluorescent Assay for TCs Detectionfluorescence sensorsynthesis reaction timeanalytessample preparationlinear range (μM)LOD (μM)RSDdetection time (min)stabilityvisual detectionrefCDs3 hTCmatrix-cleaned detection10.0–400.06.02.6%≥10stored at 4 °Cyes CD-AuNCs>50 hTCreal world water0.5–400.0562.1–4.2%7-yes CdTe QDs@SiO2/N-CDs@MIPs>8 daysTCmatrix-cleaned detection0–1400.8461.12–1.86%≤1-no Eu-MOF>12 hOTC, H2PO4 – matrix-cleaned detection2.5–15.000.2252.29–7.73%10-no Eu3+/NH2-MIL-53(Al)>1 dayTCtap water0.5–600.16-1vacuum-dryingno P-NCDs>32 hTC-0–1000.045-1.5-yes 2MP-CuNPs∼3.5 hTCmatrix-cleaned detection0.1–2400.12<3.7%3Stored at 4 °Cno UiO-66-NH2 >1 daysTC, OTC, and doxycycline (DOX)matrix-cleaned detection0–47.50.4490.63–3.40%5Vacuum-dryingno BSA-AuNCs>5 hDOXmatrix-cleaned detection0–17576–12%15-yes BSA-AuNCs (Au25)>40 hCTCmatrix-cleaned detection0.2–100.065<5%60stored at 4 °Cyes Pro-SA-AuNCs∼10 minTC, CTC, OTCmatrix-based enrichment, closer to real matrix5–8000.16–0.54≤8.30%7store at 4 °Cyesthis work Comparison of This Work with Other Fluorescent Assay for TCs Detection Determination of TCs in Agricultural Products and Animal Feed The efficiency of SPE in the removal of cationic interferents was evaluated using Fe3+ as a representative model. As shown in Figure A, the fluorescence intensity of Pro-SA-AuNCs was strongly quenched upon addition of 600 μM Fe3+ (red dashed line). After each successive wash with 3 mL deionized water, the fluorescence intensity gradually recovered. Notably, after the fourth wash, the fluorescence intensity was no longer significantly different from that of the deionized water blank (gray dashed line). Quantitative analysis (Figure B) revealed the removal of more than 95% of Fe3+ after four washes. These results confirm that the SPE water washing step effectively eliminated the cationic interferents that were not retained on the C18 sorbent, which minimized false-positive signals and ensured the reliability of TCs detection in complex matrices. SPE cleanup efficiency validation and matrix-matched standard curves for TCs detection in real-world samples. (A) Fluorescence recovery of Pro-SA-AuNCs during sequential SPE washing steps. Deionized water blank is shown as a gray dashed line (baseline). Addition of 600 μM Fe3+ induced maximum quenching (red dashed line). Fluorescence intensity gradually recovered after four successive water washes (3 mL each). (B) Quantitative removal efficiency of Fe3+ in each wash fraction, calculated as follows: [(F wash – FFe 3+)/(F water – F Fe 3+)] × 100%. More than 95% of Fe3+ was removed after four washes, which confirmed the effective elimination of this potential interferent. (C–E) Matrix-matched calibration curves for TCs detection in (C) milk, (D) honey, and (E) animal feed after SPE pretreatment. The fluorescence intensity ratio (I 0/I) was plotted against TCs concentration. All data are presented as mean ± SD (n = 3). To validate the practical applicability of the Pro-SA-AuNC-based method, we selected honey, milk, and animal feed as representative samples for TCs detection (Figure C–E). SPE pretreatment avoided false signals from food/feed components (e.g., proteins, lipids) and ensured that water-soluble impurities (e.g., Na+, Ca2+, and hydrophilic antibiotics) did not interfere with the detection. After appropriate pretreatment, the extracts of these samples were collected, and the fluorescence quenching degree of the Pro-SA-AuNC system was measured following the optimized experimental procedures described above. To evaluate the matrix effect, we compared the slopes of matrix-matched calibration curves for milk, honey, and animal feed with that of the TCs standard calibration curve in buffer (slope = 0.00344). The matrix effect (%ME) was calculated as follows %ME=(slopmatrix/slopestandard−1)×100% The %ME values were −20.9% for milk, +8.7% for honey, and −38.9% for animal feed, indicating variable matrix effects across all three matrices. These results highlight the necessity of using matrix-matched calibration for accurate quantification of TCs in complex food and feed samples, rather than relying on a simple external calibration curve. Subsequently, spiked recovery experiments were conducted to evaluate the accuracy of the method. The obtained experimental data were substituted into the corresponding matrix-matched regression equations for calculation, and the detailed results are summarized in Table . The spiked recoveries of TCs in honey, milk, and animal feed samples ranged from 94.51% ± 6.90% to 104.40% ± 5.32%, with relative standard deviation (RSD) ≤8.30%. These findings show that the Pro-SA-AuNC-based fluorescent detection method exhibited good accuracy for the determination of TCs residues in complex matrices, such as agricultural products and animal feed, meeting the requirements for practical analytical applications and providing a potential scalable analytical tool for food safety. 2Analysis of Spiked Recovery Experiments for TCs (μM, n = 3) samplesweightspiked in sampledetected (Mean ± SD)recovery ±SD (%)RSD (%) milk 1.0 g0not foundnot foundnot found 2424.72 ± 1.56102.98 ± 6.496.30 3031.32 ± 1.60104.40 ± 5.325.09 3637.09 ± 1.56103.03 ± 4.334.20 honey 0.5 g0not foundnot foundnot found 2423.77 ± 1.9799.05 ± 8.228.30 3029.00 ± 1.7296.63 ± 5.745.94 3634.10 ± 1.5594.72 ± 4.314.55 animal feed 1.0 g0not foundnot foundnot found 2424.10 ± 1.85100.44 ± 7.727.70 3028.35 ± 2.0794.51 ± 6.907.30 3635.41 ± 2.3898.37 ± 6.616.72aAbbreviations: SD, Standard Deviation; RSD, relative standard deviation. Analysis of Spiked Recovery Experiments for TCs (μM, n = 3) Abbreviations: SD, Standard Deviation; RSD, relative standard deviation. Discussion Food and feed safety monitoring of TCs still requires effective and convenient analytical methods. Here, a biomolecule-assisted synthesis of Pro–SA–AuNCs was developed, which is simple, cost-effective, and suitable for complex matrices. The design rationale of the probe is based on the RIM at the cluster interface, a process known to suppress nonradiative energy dissipation and thereby enhance fluorescence emission. In this system, SA enhanced the RIM of Pro ligands through hydrogen bonding and electrostatic interactions, which reduced vibrational and rotational energy losses and consequently improved fluorescence intensity. In line with this mechanism, the fluorescence of Pro-SA-AuNCs was higher than that of Pro-AuNCs prepared without alginate. XPS analysis further unveiled the coexistence of Au(0) and Au­(I) states in the Au 4f spectrum, which suggests a rationally coordinated metal–ligand interface that contributed to fluorescence stability. Comprehensive characterization using TEM, HRTEM, UV–vis, and fluorescence spectroscopy, FT-IR, and XPS revealed the successful formation and photophysical features of the Pro-SA-AuNCs. The NCs, with an average size below ∼3 nm, exhibited molecular-like electronic structures rather than the SPR typical of larger gold nanoparticles. Their negatively charged surface imparted electrostatic repulsion, which effectively prevented aggregation and ensured good colloidal dispersion. The fluorescence intensity remained nearly unchanged over a pH range of 1–11, indicating good environmental stability. Furthermore, after one month of storage at 4 °C in the dark, the fluorescence was largely preserved, with an outstanding photostability. These properties meet the essential requirements for analytical probes intended for complex matrices, such as food and feed samples. Mechanistic studies on the interaction between Pro-SA-AuNCs and TCs revealed overlapping UV absorption and fluorescence excitation spectra, implying possible FRET or IFE processes. Time-resolved fluorescence measurements showed negligible lifetime variation upon TC addition, ruling out dynamic quenching that typically shortens fluorescence lifetimes. The Stern–Volmer plots were linear, and the quenching constant decreased with the increase in temperature, which supports a static quenching mechanism. Moreover, the enlarged hydrodynamic diameter after reaction indicated complex or aggregate formation between the probe and TC, which further confirmed static quenching accompanied by IFE rather than dynamic quenching. Such hybrid quenching behavior is consistent with prior reports on AuNC-based antibiotic sensors. The combination of lifetime and temperature-dependent analyses in this study enhanced the mechanistic rigor and interpretability of the fluorescence response. When coupled with C18-SPE, the achieved LODs (0.16–0.54 μM) of TCs by Pro-SA-AuNCs were comparable to the regulatory values summarized in the literature (e.g., honey, 200 μg/kg for TC, corresponding to 0.45 μM) and the Codex Alimentarius Commission MRL (CAC/MRL 2-2015) , lists (e.g., milk, 100 μg/kg for TC, corresponding to 0.225 μM). Compared with previously reported fluorescence sensors for TCs detection, the Pro-SA-AuNCs developed in this work exhibited several notable advantages (Table ). First, the synthesis of Pro-SA-AuNCs required only approximately 10 min, which is significantly shorter than the preparation times of other sensors, such as CDs (3 h), Eu-MOF (>12 h), or BSA-AuNCs (Au25) (>40 h). This rapid synthesis not only improved efficiency but also reduced energy consumption and potential batch-to-batch variability. Second, Pro-SA-AuNCs demonstrated excellent stability when stored at 4 °C for a month, which eliminates the need for strict storage conditions, such as vacuum-drying, which are required for sensors such as Eu3+/NH2-MIL-53­(Al), UiO-66-NH2. Moreover, Pro-SA-AuNCs allowed for visual detection of multiple TCs with a broad linear range (5–800 μM) and relatively low LODs (0.16–0.54 μM), which make them suitable for the analysis of complex samples, including milk, honey, and animal feed. The method also showed good reproducibility, with RSD values ≤8.30%, and a reasonably short detection time of 7 min. Collectively, these attributesrapid synthesis, ambient stability, visual capability, multianalyte detection, and applicability to real samplesposition Pro-SA-AuNCs as a practical and efficient sensing platform for on-site and routine monitoring of TC residues. In real-sample assays of honey, milk, and feed following SPE pretreatment, recoveries ranging from 94.51% to 104.40% and with low RSD demonstrated reliable analytical performance in complex matrices. Although AuNCs have been increasingly employed in food analysis, matrix effects, stability, and selectivity remain major challenges, issues which were effectively addressed in the present design. Compared to sensor arrays that rely on cross-reactive responses from multiple gold nanocluster probes and require pattern recognition algorithms for antibiotic identification, our single-probe method provides a simpler, more rapid, and cost-effective detection scheme with comparable sensitivity for tetracyclines. The structural precision confirmed by mass spectrometry and the elimination of complex data processing make our approach particularly attractive for practical food safety monitoring. Overall, this work provides a rational framework for the advancement of AuNC probes toward TCs-responsive systems. Future directions include tuning ligand composition for multiwavelength emission or cross-analyte recognition and integrating probes with portable analytical devices, such as paper-based strips, hand-held fluorometers, or microfluidic chips, to enable rapid, low-cost, on-site detection. Further strengthening matrix tolerance and interference resistance, particularly under lipid-rich, proteinaceous, or colloidal environments will deepen the mechanistic understanding and guide probe optimization. Enhancing reproducibility, storage stability, and cost efficiency will be essential steps toward industrial translation. Collectively, this study demonstrated a coherent integration of synthetic design, mechanistic exploration, analytical performance, and real-sample validation, which underscores the promise of Pro–SA–AuNC-related methods as a next-generation sensing platform for antibiotic residue monitoring. Conclusion A biomolecule-assisted synthesis of Pro–SA–AuNCs was developed, and it provides a sensitive platform suitable for TCs monitoring in food and feed. This high-intensity blue-emissive probe recognized TCs through a dual mechanism involving IFE and aggregation-induced (static) quenching. The probe, coupled with SPE pretreatment, exhibited low LODs, broad linear ranges, and satisfactory recoveries in milk, honey, and feed samples. These results highlight its analytical robustness and potential for real-world food and feed safety monitoring. Supplementary Material