The Detection of Food Additives Using a Fluorescence Indicator Based on 6– p–Toluidinylnaphthalence-2-sulfonate and Cationic Pillar[6]arene

The current study investigated host-guest complexation in 6-p-toluidinylnaphthalene-2-sulfonate (TNS), a fluorescence probe used to investigate hydrophobic regions that contain the water-soluble cationic pillar[6]arene (CP6). After complexation with CP6, the fluorescence intensity of TNS was significantly increased. The decreases in the fluorescence intensity of the TNS•CP6 complex when phenolic food-additives are added have been used in indicator displacement assays to detect food additives in the water.

Phenolic food additives have been widely used in the food industry for their significant antioxidant, antimicrobial, and flavor-enhancing properties (Vinson, et al., 2012;Zhang, et al., 2014). Whereas the insolubility of food additives in water and their long-term stability contribute to their excessive use, ultimately resulting in their accumulation and negative effects on the biosphere (Tobacman, et al., 2001;Savjani, et al., 2012). Encapsulating smallmolecule food-additives in non-toxic, water-soluble macrocyclic hosts improves their bioavailability and solubility by regulating their physical and chemical properties (Munin, et al., 2011). 2-Hydroxypropyl β-cyclodextrin (HP-β-CD) is one representative example, with a binding affinity of~10 2 M −1 to food-additives (Pal, et al., 2016). It is critical to investigate artificial receptors with extremely high affinity for food additives to improve sensitivity and detection efficiency in compound detection.
We developed a new FID assay with a water-soluble cationic pillar[6]arene (CP6) for the detection of three important phenolic food additives, namely p-coumaric acid (CA), trans-ferulic acid (FA), and gallic acid (GA). Because of its enhanced fluorescence in non-polar environments, the widely used fluorescent probe, 6p-toluidinylnaphthalene-2-sulfonate (Dotsikas, et al., 2000) (TNS in Scheme 1), was used as the fluorescent indicator in our FID system. Due to the complexation of TNS and CP6, we use an FID strategy to perform sensitive fluorescence detection on CA, FA, and GA.

MATERIALS AND METHODS
The reagents used were marketable and applied directly without further purification. CP6 (Duan et al., 2019) was synthesized by following the known procedures. Nuclear magnetic resonance (NMR) spectra were obtained using the Bruker Avance III HD 400 spectrometer with the deuterated solvent as the lock and the residual solvent as the internal reference. Fluorescence spectra were obtained by using the Agilent Cary Eclipse fluorescence spectrophotometer. To prevent the dilution effect during titration, CP6 stock solutions were produced using the same TNS solution. The measurement was repeated three times for each experiment. Displacement assays for CA, FA, and GA were performed at pH 6.8 with CP6 at varying concentrations of CA, FA, and GA, respectively. All the experiments were conducted at room temperature (298 K).

RESULTS AND DISCUSSION
Fluorescent Probe 6-p-Toluidinylnaphthalene-2-Sulfonate Complexed With Cationic Pillar[6]Arene UV-vis absorption spectroscopy was used to confirm the hostguest complexation of fluorescent probe TNS with CP6.
Following successive additions of CP6 to the phosphatebuffered solution (PBS) of TNS at pH 6.8, hyperchromic effects at the maximum absorption wavelengths of 223, 263, and 318 nm occurred with a significant bathochromic shift ( Figure 1A). The variations appeared to be greater than those when α-CD (Nishijo et al., 1992) or β-CD (Nishijo et al., 1995;Dotsikas et al., 2000) addition was used. The results indicate that TNS can form a stable complex with CP6. Additionally, a fluorescence titration on TNS with an increased CP6 concentration was performed in PBS with a pH of 6.8 at room temperature. According to Figure 1B, as the concentration of CP6 increased, a significant increase in fluorescence intensity was observed, along with a shift in the fluorescence maximum to shorter wavelengths. The results indicate that TNS molecules exist in a hydrophobic environment. Encapsulating TNS in CP6 protects it from solvent collisions while also providing a distinct local environment for TNS in terms of polarity, which significantly enhances (approximately 400 times) fluorescence intensity. Additionally, the significant pale blue fluorescence was evident in UV light (the inset of Figure 1B).
1 H NMR tests were used to investigate the host-guest complexation. Because the solubility of the complex in neat D 2 O was insufficient to reach the mM level, DMSO-d 6 cosolvents were added. As illustrated in Figure 2, the naphthyl proton signals of TNS in the inclusion complex underwent varying degrees of upward shifts. The largest shift occurred in the direction away from the sulfonate group, whereas the smallest shift occurred in the direction toward the sulfonate group. Proton signal variations in the methylphenyl group are insignificant, indicating that this group may be located outside the cavity. When combined with the protons' shift and broadening in the sulfonatenaphthyl group, it is concluded that the fluorescence probe molecule is partially in the CP6 cavity, where the shielding effects of the aromatic host produce the characteristic signal broadening (Li et al., 2010). Additionally, the 2D ROESY data  Figure S1) establish a correlation between the naphthyl protons (H a-f ) in the entrapped TNS and the aromatic proton H 1 in CP6, revealing the interpenetrated geometry.
To quantify the binding of TNS to CP6, the association constant (K a ) was determined to be (4.51 ± 0.90) × 10 5 M −1 using titration tests and a non-linear curve-fitting to the fluorescence spectra (Supplementary Figure S4). The complex formed by CP6 and TNS had a 1:1 binding stoichiometry (Supplementary Figure S3A). We deduced that the complex formed between CP6 and TNS in aqueous solution as a result of multiple electrostatic interactions between the cationic ammonium groups on CP6 and the sulfonate anion on TNS, hydrophobic interactions, and π-π stacking interactions between the benzene rings on host CP6 and naphthalene ring on guest TNS. The cooperativity of these non-covalent interactions is attributed to the binding affinity in the host-guest system.

Detection of Food-Additives Using Fluorescent Indicator
TNS complexed with CP6 exhibits a significant fluorescence response, allowing it to be used for FID detection. The binding affinities of CP6 to CA, FA, and GA were determined in this study using FID detection (Scheme 1), and the fluorescent indicator TNS was first reversibly bound to the receptor CP6. The solution was then added with a weakly fluorescent or non-fluorescent analyte, which competitively displaced the highly fluorescent TNS from the indicator CP6 cavity, altering the optical signal (You et al., 2015). Notably, titration of a preformed TNS•CP6 complex with increasing concentrations of competitor food additive molecules results in a reversal of the fluorescence intensity, which is used to determine the competitor molecules' binding affinity. Competitive displacement was used to investigate the binding of three major phenolic food-additives, namely CA, FA, and GA, to CP6. Figure 3A illustrates a typical fluorescence displacement titration with CA as a strong competitor. The quenching of fluorescence in the presence of CA was easily observed with the naked eye using a simple UV-lamp (the inset of Figure 3A).
To avoid the effect of changes in pH and dilution on the displacement assay, the pH of complex and food-additive solutions was set to 6.8, and the concentrations of TNS and CP6 in food-additive solutions were kept constant. We used that previously reported competitive binding formula to fit the reduced fluorescence intensities at the band maximum against the concentration of competitor food additives (Bakirci et al., 2006). Using TNS•CP6 as the reporter pair, we determined the association constants (K a ) for CA, FA, and GA to be (1.24 ± 0.29) × 10 4 M −1 , (1.19 ± 0.16) × 10 4 M −1 , and (2.78 ± 0.18) × 10 2 M −1 (Supplementary Figures S5-S7), respectively. Except for GA, the binding affinities are approximately two orders of magnitude greater than those of previously studied HP-β-CD to the other two food additives, which are around 10 2 M −1 (Pal et al., 2016).
Additionally, the observed fluorescence response can also be used to quantify CA, FA, and GA. The fluorescence intensity plots increase linearly as the CA, FA, and GA concentrations increase (Supplementary Figure  S8), respectively. 0.047-2.3 mM, 0.047-0.14 mM, and 0.047-2.5 mM were the linear ranges. The results indicate that the limit of detection (LOD) values was 0.012, 0.08, and 0.17 µM, respectively, using a 3σ/slope method (MacDougall et al., 1980). NMR research with GA, CA, and FA was used to determine the complexation of food additives. Supplementary Figure S9 illustrates the 1 H NMR spectra for GA in the presence of CP6. As illustrated in the figure, shielding caused a shift in the benzene proton signal of GA, conclusively confirming the inclusion complex between CP6 and GA. Additionally, the 2D NOESY data (Supplementary Figure  S2) show NOE cross-peaks between the benzene proton (H a ) in entrapped GA and the protons H 1-4 in CP6, indicating the inclusion of a benzene ring in the CP6 cavity. The signals in the NMR spectra of CA and FA changed similarly upon the addition of CP6 (Supplementary Figures S10,S11).

CONCLUSIONS
To summarize, we demonstrated a new fluorescence activation switch based on host-guest complexation between the fluorescent indicator probe TNS and cationic pillar[6]arene CP6. In TNS solution, the complexation significantly enhanced the fluorescence. A fluorescence switch-off displacement assay was used to detect three commonly used non-fluorescence phenolic food additives in the water. The study used molecular recognition and fluorescence indicator displacement assays to develop a prospective strategy for phenolic food additive detection.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

AUTHOR CONTRIBUTIONS
QD designed the work. YX and KG made contributions to the experiments and collective data. The paper was written by QD. All authors extensively discussed the results, reviewed the manuscript, and approved the final version of the manuscript to be submitted.

FUNDING
This work was supported by Henan Province Science and Technology Research Program (No. 222102320067).