Sensitive Immunochromatographic Determination of Salmonella typhimurium in Food Products Using Au@Pt Nanozyme

In this study, we developed a sensitive immunochromatographic analysis (ICA) of the Salmonella typhimurium bacterial pathogen contaminating food products and causing foodborne illness. The ICA of S. typhimurium was performed using Au@Pt nanozyme as a label ensuring both colorimetric detection and catalytic amplification of the analytical signal due to nanozyme peroxidase-mimic properties. The enhanced ICA enabled the detection of S. typhimurium cells with the visual limit of detection (LOD) of 2 × 102 CFU/mL, which outperformed the LOD in the ICA with traditional gold nanoparticles by two orders of magnitude. The assay duration was 15 min. The specificity of the developed assay was tested using cells from various Salmonella species as well as other foodborne pathogens; it was shown that the test system detected only S. typhimurium. The applicability of ICA for the determination of Salmonella in food was confirmed in several samples of milk with different fat content, as well as chicken meat. For these real samples, simple pretreatment procedures were proposed. Recoveries of Salmonella in foodstuffs were from 74.8 to 94.5%. Due to rapidity and sensitivity, the proposed test system is a promising tool for the point-of-care control of the Salmonella contamination of different food products on the whole farm-to-table chain.


Introduction
A growing global population and increased food needs lead to the intensification and industrialization of the crop and livestock sectors, which creates not only new opportunities but also new threats and risks to food safety [1].These threats include foodborne illnesses-infectious diseases or intoxications caused by bacteria, viruses, or parasites that enter the body through contaminated water or food [2].Foodborne diseases are not only harmful to human health but are a major impediment to socioeconomic development as they put pressure on health systems and harm the national economies, tourism, and trade sectors [3].One of the relevant foodborne pathogens is Salmonella-a Gramnegative rod-shaped enterobacterium [4].The main symptoms of salmonellosis include fever, headache, nausea, vomiting, abdominal pain, and diarrhea [4].Foodstuffs associated with Salmonella outbreaks include milk, domestic animal meat, poultry, eggs, and other animal products [5,6].Salmonella can pass through the entire food chain-from animal feed and primary production to the home or food service establishments-and persist in food and the environment for a long time.One of the highly pathogenic Salmonella species is Salmonella typhimurium, which causes infectious diseases with intestinal symptoms [7,8].
The serious consequences of diseases caused by Salmonella make it necessary to prevent, identify, and eliminate the risks associated with contaminated food.Prevention of foodborne infections requires control measures at all stages of the food chain-from agricultural production to food processing, production, and cooking.Early detection and adequate response to salmonellosis will prevent its further spread and ensure food safety.The traditional methods for detecting Salmonella cells include microbiological (cultivation of the test material on nutrient media) and microscopic (study of stained material) techniques [9,10].Although these approaches provide effective diagnostics, they are time-and labor-intensive and not applicable in the context of quick monitoring of many samples for the presence of pathogens.Another modern method for detecting bacterial pathogens is molecular genetic analysis [9].Despite the high sensitivity and accuracy, this approach requires high-tech equipment, qualified specialists, and special reagents.
Alternative analytical methods are those based on the immunoanalytical detection of foodborne pathogens using specific antibodies, first of all, enzyme-linked immunosorbent assay (ELISA), which provides sensitive and selective detection of bacteria, including Salmonella [11][12][13][14].However, a typical ELISA contains several steps and requires some laboratory equipment; therefore, it cannot be classified as a rapid on-site method.Immunochromatographic analysis (ICA), which is widely used for the detection of pathogens of various foodborne infections does not have this disadvantage [15,16].The merits of ICA are the simplicity of the analytical procedure, rapidity, and low cost.All reagents are preliminarily applied to the test strip and the analysis consists of a simple incubation of the strip with the tested sample followed by a visual qualitative (yes/no) or instrumental quantitative (how much) assessment of the pathogen.In addition, ICA often does not require complex sample preparation before analysis, and for liquid food samples, simple dilution with buffer is often carried out, which contributes to the rapidity of testing [17,18].
Thus, for rapid and sensitive mass control of Salmonella in food, ICA is a promising and demanded method, which is evidenced by a large number of relevant publications (see Section 3.6).On average, the detection sensitivity of Salmonella is 10 4 -10 6 CFU/mL [19][20][21].In several studies, the authors use special approaches to reduce the limit of detection (LOD) by using new markers that can enhance the detected signal [22] including using single-atom nanozymes [23], fluorescent labels [24,25], aggregation or extension of labels [26,27], etc.In this case, LODs decrease to 10 2 -10 3 CFU/mL.Among the tested matrices, mainly liquid samples (milk, drinking water, juice) can be mentioned, the assay duration varies from 5 min to several hours.
In the present study, an enhanced ICA of S. typhimurium was for the first time developed in a sandwich format using core@shell bimetallic Au@Pt nanozyme as a label gaining high popularity due to high activity and enzymatic specificity along with high stability, and low cost [28,29].The catalytic properties of the marker allowed for the amplification of the optical signal and a significant reduction in the LOD of S. tiphymurium was achieved.The assay was validated for S. typhimurium detection in a panel of food samples-milk and chicken meat.

Biotinylation of MAb
MAb were biotinylated according to [30].For this, the MAb solution (100 µM, 200 µL) in 50 mM K-phosphate buffer with 100 mM NaCl, pH 7.4 (PBS) was mixed with a solution of N-hydroxysuccinimide ester of biotin (1 mM in DMSO) and incubated for 2 h at room temperature with continuous stirring.Excess unreacted low molecular weight reagents were removed by the dialysis against PBS.

ELISA of S. typhimurium
MAb (1 µg/mL, 100 µL in PBS) were immobilized in the microplate wells overnight at 4 • C.Then, the microplate was washed four times with PBS containing 0.05% Triton X-100 (PBST).Next, solutions of S. typhimurium (concentration range from 1 × 10 7 to 1 × 10 4 CFU/mL, 50 µL in PBST) and biotinylated MAb (1 µg/mL, 50 µL in PBST) were added to the wells.The microplate was incubated for 1 h at 37 • C and washed as described above.After that, STR-HRP (1:5000 dilution, 100 µL in PBST) was added to the wells and incubated for 1 h at 37 • C.After washing, the activity of the enzyme label was determined.To do this, 0.4 mM TMB solution in 40 mM sodium citrate buffer, pH 4.0, containing 3 mM hydrogen peroxide (100 µL) was added to the microplate wells and incubated for 15 min at room temperature.The reaction was stopped by adding 1 M sulfuric acid (50 µL) and the optical density was measured at 450 nm (OD 450 ) on a Zenyth 3100 microplate spectrophotometer (Anthos Labtec Instruments, Wals, Austria).

Synthesis and Characterization of AuNPs and Au@Pt Nanozyme
Synthesis of two preparations of AuNPs (of smaller and larger diameters) was performed following the procedure of HAuCl 4 reduction by sodium citrate [31].To obtain smaller/larger AuNPs, HAuCl 4 (97.0/98.8mL of 0.01% solution) was heated to 100 • C and then, sodium citrate (3.0/1.2 mL of 1% solution) was added.The solutions (OD 520 = 1) were kept at 100 • C for 20/30 min under vigorous stirring and then cooled.
Au@Pt was synthesized using AuNPs with a smaller diameter as described by Gao et al. with modifications [32].AuNPs (40 mL) were mixed with 10 mM Na 2 PtCl 6 (8 mL) and deionized water (10.6 mL).The resultant solution was heated in a water bath to 80 • C.After that, 50 mM sodium ascorbate (8 mL) was added with a flow rate of 450 µL/min using a peristaltic pump.After heating the reaction mixture for another 20 min at 80 • C and cooling it to room temperature, the resultant Au@Pt was centrifuged at 12,000× g for 20 min.The sediment was resuspended in deionized water.
To characterize the size and aggregation of the obtained nanodispersed labels, transmission electron microscopy (TEM) was utilized.AuNPs and Au@Pt nanozyme were applied to 300-mesh grids (Pelco International; Redding, CA, USA) coated with a poly(vinyl formal) film.The images were obtained with a JEM CX-100 microscope (Jeol, Tokyo, Japan) at 80 kV and analyzed by the Image Tool 3.00 software (University of Texas, Health Science Center, San Antonio, TX, USA).
For elemental analysis, energy-dispersive spectroscopy (EDS) of Au@NPs and Au@Pt was performed.For this, the transmission electron microscope Jeol JEM-1400 (Jeol, Tokyo, Japan) equipped with the energy-dispersive spectrometer (INCA Energy TEM 350, Oxford Instruments, High Wycombe, UK) was utilized.AuNPs and Au@Pt were centrifuged twice (for 20 min at 20,000× g and 15,000× g, respectively) and redispersed in ultrapure water.The aliquots of AuNPs and Au@Pt (10 µL both) were incubated on a formvar-coated copper grid for 10 min.EDS spectra were recorded in an area of 0.5 µm 2 .
The complexation of Au@Pt with MAb was confirmed by the asymmetrical flow fieldflow fractionation (AF4) coupled with UV-vis detector, multi-angle laser light scattering (MALLS), and dynamic light scattering (DLS).For this, AF4 platform that included Wyatt Eclipse 3+ Separation System with Eclipse Long Channel (Wyatt Technology, Goleta, CA, USA), 1260 Infinity LC System (Isocratic Pump, Autosampler, Variable Wavelength Detector, Agilent Technologies, Santa Clara, CA, USA), Dawn HELEOS II multi-angle light scattering detector with a WyattQELS DLS module, and Optilab T-Rex refractometer (Wyatt Technology, Goleta, CA, USA) was applied.The characterization of Au@Pt and MAb-Au@Pt conjugate was carried out according to [33]; the details on the liquid flow program are presented in Table S1.The data were collected and analyzed with ChemStation v.B.04.03 (Agilent Technologies, Santa Clara, CA, USA) and Astra v.6.1.7.17 (Wyatt Technology, Goleta, CA, USA) software.

Conjugation of MAb with AuNPs and Au@Pt
Before conjugation, MAb were dialyzed against 10 mM Tris-HCl buffer, pH 9.0, for 2 h at 4 • C, and AuNPs and Au@Pt nanozyme were adjusted to pH 9.0 by 0.1 M K 2 CO 3 .AuNPs/Au@Pt nanozyme were added to MAb solutions (4.2/20 µg/mL) and incubated for 30/60 min at room temperature with stirring.Then, an aqueous solution of BSA was added to the reaction mixtures to a final concentration of 0.25%.After that, Mab-AuNPs/MAb-Au@Pt were centrifuged at 17,000×/12,000× g for 15/20 min at 4 • C. The precipitates were concentrated (by 6 and 10 times for Mab-AuNPs and MAb-Au@Pt, respectively) by resuspending in small volumes of 0.01 M Tris-HCl, pH 9.0, containing 1% BSA, 1% sucrose, and 0.01% sodium azide (TBSA), and stored at 4 • C.

Production of Immunochromatographic Tests
Two types of test strips were produced to perform AuNPs-based and Au@Pt nanozymebased ICAs.In the first case, the control (C) zone and the test (T) zone were formed on the working membrane by application of GAMI (0.5 mg/mL in PBS) and MAb (1 mg/mL in PBS), respectively, using an Iso-Flow dispenser (Image Technology, Lebanon, NH, USA) at a loading of 0.12 µL/mm of the membrane.MAb-AuNPs conjugate containing 1% of Tween 20 (OD 520 = 6) was manually applied to the fiberglass conjugate pad (0.8 µL/mm).
In the second case, C and T zones were also formed by the immobilization of GAMI (0.5 mg/mL in PBS) and MAb (2 mg/mL in PBS).MAb-Au@Pt was 13 times diluted by TBSA containing 0.05% of Tween-20 and applied to the conjugate pad (with a loading of 32 µL/cm of the membrane).The same glass fiber membrane was used both as a sample and a conjugate pad.The conjugate pad was preliminary soaked by the PBS containing 1% of Triton X-100.All membranes and pads were dried at room temperature for 20 h.Then, multimembrane sheets were composed consisting of the working membrane, sample, conjugate, and absorbent pads.Finally, test strips (of 3.5 mm width) were obtained by cutting sheets with an automatic Index Cutter-1 guillotine (A-Point Technologies, Gibbstown, NJ, USA).Test strips could be stored at room temperature in zipper bags with desiccant (silica gel) for at least 3 months without changing the analytical performance of the ICA.

Sample Preparation before the ICA and ELISA
As real samples, cow lactose-free milk of 1.5% fat (sample 1), cow milk of 3.2% fat (sample 2), cow-baked milk of 4% fat (sample 3), and raw chicken meat (sample 4) were purchased in local supermarkets.Milk was diluted by PBS containing 0.1% of Triton X-100 (PBST 0.1 ) 10, 20, and 25 times for samples 1-3, respectively.For chicken meat, a procedure of sample preparation based on that described in [34] was used.Briefly, the meat sample was minced and 5 mL of the extraction buffer (PBST 0.1 containing 0.5 M KCl) was added to 250 mg of homogenized meat.The mixture was intensively shaken for 20 min and then left for 15 min to settle the solid component.The upper layer was used for the analysis.

ICA of S. typhimurium Using AuNPs
Test strips were placed horizontally and the S. typhimurium standard solutions (concentration range of 3 × 10 7 -1 × 10 3 CFU/mL, 60 µL) were applied to the sample pads.After 10 min incubation at room temperature, test strips were scanned using a CanoScan 9000F scanner (Canon, Tochigi, Japan).The obtained digital images were processed by TotalLab TL120 software (Nonlinear Dynamics, Newcastle, UK) to measure the intensity of zones' coloration.

Common and Enhanced ICAs of S. typhimurium Using Au@Pt Nanozyme
Test strips were immersed vertically in the standard solutions of S. typhimurium in PBST 0.1 (concentration range of 2.3 × 10 8 -23 CFU/mL, 75 or 40 µL for the common and enhanced ICAs, respectively) or real samples (milk and chicken extracts) and incubated for 10 min at room temperature.Then, in the case of the common ICA, test strips were taken out and blotted.For the enhanced ICA, test strips were washed (by placing the bottom edge of the strip to 25 µL of PBST 0.1 for 3 min) and the DAB-based substrate solution (1 µL) was added directly to the T zone and incubated for 2 min.After both ICA modes, the images of the test strips were obtained and processed as described above.

Evaluation of the ELISA and ICA Results and Statistical Analysis
To fit the dependencies of OD 450 (for the ELISA) or color intensity of the T zone (for the ICA) (y) versus S. typhimurium concentrations (x), OriginPro 9.0 software from OriginLab (Northampton, MA, USA) was applied.The analytical performance was determined following [35].In the ELISA, the instrumental LOD of S. typhimurium corresponded to 10% binding with the immobilized MAb.In the ICA, visual LOD was estimated as the minimum S. typhimurium concentration causing a visible coloration in the T zone.That is, the visual LOD was assessed as the concentration of the analyte at which the coloration of the T zone on the test strip is noticeable to the naked eye (usually this corresponds to approximately 600-700 RU and more).The linear range of the detectable S. typhimurium concentrations corresponded to the diapason where linear approximation allows obtaining reliable quantitative results.
All measurements were made in triplicate.For the ELISA and ICA calibration curves, the Means ± SE (standard errors) of S. typhimurium concentration were calculated.To evaluate the repeatability and reproducibility of the assay, intra-assay coefficients of variation (CV) and inter-assay CV were assessed (n = 6).For the analysis of milk and meat samples, 5 repeats were made for each sample.

Obtaining and Characterization of the Reagents
The antigen-binding properties of Salmonella-specific MAb were characterized in the sandwich format of the ELISA, where the detected antigen occurs in a triple complex with antibodies of the same or different clones.Anti-Salmonella antibodies of the 1E6cc clone were used for immobilization on the solid phase and the same biotinylated antibodies in combination with a STR-HRP conjugate were used to label the resulting Salmonella-MAb complexes.For bacteria, the use of the same clone is acceptable due to the recurrence of antigenic determinants on the cell surface.Optimization of the ELISA included choice of the concentration of the immobilized and biotinylated Mab (the ELISA optimization results are presented in Figure S1; the varied and finally selected parameters-in Table S2).Finally, the following values were selected: Mab were absorbed at 1 µg/mL and biotinylated Mab was added at the concentration of 2 µg/mL.The calibration curve of S. typhimurium is shown in Figure 1.The LOD was 2.2 × 10 5 CFU/mL and the linear range was 4.3 × 10 5 -4.0 × 10 6 CFU/mL.sandwich format of the ELISA, where the detected antigen occurs in a triple com antibodies of the same or different clones.Anti-Salmonella antibodies of the 1 were used for immobilization on the solid phase and the same biotinylated an combination with a STR-HRP conjugate were used to label the resulting Salmo complexes.For bacteria, the use of the same clone is acceptable due to the rec antigenic determinants on the cell surface.Optimization of the ELISA included the concentration of the immobilized and biotinylated Mab (the ELISA optim sults are presented in Figure S1; the varied and finally selected parameters-in Finally, the following values were selected: Mab were absorbed at 1 µg/mL an ated Mab was added at the concentration of 2 µg/mL.The calibration curve murium is shown in Figure 1.The LOD was 2.2 × 10 5 CFU/mL and the linear ran × 10 5 -4.0 × 10 6 CFU/mL.Two types of nanodispersed carriers, AuNPs and Au@Pt nanozymes, we labels in the analyses.AuNPs are widely used in various ICA formats as an easil and stable marker that provides a high and reproducible colorimetric signal on [36].Two preparations of AuNPs were synthesized by the citrate method varyin centrations of the reagents and the incubation time for the preparation of smaller particles.The size and homogeneity of AuNPs were characterized by results are presented in Figure 2a-d.The observation showed that the shape of in both samples was close to spherical and no aggregates were found.Based o tron microscopic images of AuNPs, the mean diameters/ellipticity coefficients w 1.0 nm/1.1 ± 0.1 (according to measurements of 113 particles) and 42.0 ± 5.6 n (according to measurements of 111 particles) for particles of smaller (Figure larger (Figure 2c,d) diameters.A nanozyme label was obtained using AuNPs w eter of 14 nm by reducing the platinum salt with sodium ascorbate on a gold sur core@shell Au@Pt nanoparticles with peroxidase-mimic catalytic properties tained.According to the TEM data (Figure 2e,f), the average size/ellipticity of A ticles was 27.1 ± 7.2 nm/1.2 ± 0.01 (according to measurements of 74 particles).black color, the nanozyme can form high-contrast lines on the test strip and c used as a colored label.Two types of nanodispersed carriers, AuNPs and Au@Pt nanozymes, were used as labels in the analyses.AuNPs are widely used in various ICA formats as an easily obtained and stable marker that provides a high and reproducible colorimetric signal on test strips [36].Two preparations of AuNPs were synthesized by the citrate method varying the concentrations of the reagents and the incubation time for the preparation of larger and smaller particles.The size and homogeneity of AuNPs were characterized by TEM; the results are presented in Figure 2a-d.The observation showed that the shape of the AuNPs in both samples was close to spherical and no aggregates were found.Based on the electron microscopic images of AuNPs, the mean diameters/ellipticity coefficients were 14.1 ± 1.0 nm/1.1 ± 0.1 (according to measurements of 113 particles) and 42.0 ± 5.6 nm/1.3 ± 0.2 (according to measurements of 111 particles) for particles of smaller (Figure 2a,b) and larger (Figure 2c,d) diameters.A nanozyme label was obtained using AuNPs with a diameter of 14 nm by reducing the platinum salt with sodium ascorbate on a gold surface.Thus, core@shell Au@Pt nanoparticles with peroxidase-mimic catalytic properties were obtained.According to the TEM data (Figure 2e,f), the average size/ellipticity of Au@Pt particles was 27.1 ± 7.2 nm/1.2 ± 0.01 (according to measurements of 74 particles).Due to its black color, the nanozyme can form high-contrast lines on the test strip and can also be used as a colored label.
To confirm the composition of AuNPs and Au@Pt nanozyme, their elemental analysis was performed using EDS.The obtained spectra are presented in Figure 2f; they contain peaks characteristic of Au or/and Pt and, consequently, confirm the formation of objects composed of these elements.The detected peaks of copper refer to the copper grid used as a solid support for Au@Pt or AuNPs during EDS measurements.Conjugates of MAb with AuNPs and Au@Pt nanozyme were obtained by adsorption immobilization.The choice of MAb concentration for conjugation with 42 nm AuNPs was based on the dimensions of the carrier and immunoglobulin G (IgG) molecule and the maximum surface area occupied by one IgG molecule as described in [37].Monolayer immobilization of MAb on the surface of AuNPs was ensured by the MAb concentration of 4.2 µg/mL.The choice of MAb concentration for conjugation with a nanozyme label was made based on our previous results; it was shown that for a nanozyme of such composition, the conjugate with an antibody load of 20 µg/mL was the most effective in the ICA [38].Au@Pt and Mab-Au@Pt were characterized spectrophotometrically (Figure 3a) and by DLS on Zetasizer Nano ZS 90 (Malvern, UK) (Figure 3b).UV-vis spectra demonstrated a complete attenuation of the surface plasmon resonance band of gold after plating with platinum.According to the obtained DLS data, the diameters of Au@Pt and Mab-Au@Pt were ~55 nm and ~85 nm.The difference in the dimension values obtained by TEM and DLS is explained by the fact that DLS takes into account the hydration and protein shell on the surface of the studied particles.The zeta potentials of Au@Pt and Mab-Au@Pt were measured.The zeta potential of Au@Pt was −33.6 ± 0.5 mV, which correlated well with data obtained in other studies [39].Conjugation of Au@Pt with Mab led to the converting of zeta potential to −8.17 ± 0.4 mV, which could be attributed to the positive charge of antibodies.
peaks characteristic of Au or/and Pt and, consequently, confirm the formation of objects composed of these elements.The detected peaks of copper refer to the copper grid used as a solid support for Au@Pt or AuNPs during EDS measurements.
Conjugates of MAb with AuNPs and Au@Pt nanozyme were obtained by adsorption immobilization.The choice of MAb concentration for conjugation with 42 nm AuNPs was based on the dimensions of the carrier and immunoglobulin G (IgG) molecule and the maximum surface area occupied by one IgG molecule as described in [37].Monolayer immobilization of MAb on the surface of AuNPs was ensured by the MAb concentration of 4.2 µg/mL.The choice of MAb concentration for conjugation with a nanozyme label was made based on our previous results; it was shown that for a nanozyme of such composition, the conjugate with an antibody load of 20 µg/mL was the most effective in the ICA [38].Au@Pt and Mab-Au@Pt were characterized spectrophotometrically (Figure 3a) and by DLS on Zetasizer Nano ZS 90 (Malvern, UK) (Figure 3b).UV-vis spectra demonstrated a complete attenuation of the surface plasmon resonance band of gold after plating with platinum.According to the obtained DLS data, the diameters of Au@Pt and Mab-Au@Pt were ~55 nm and ~85 nm.The difference in the dimension values obtained by TEM and DLS is explained by the fact that DLS takes into account the hydration and protein shell on the surface of the studied particles.The zeta potentials of Au@Pt and Mab-Au@Pt were measured.The zeta potential of Au@Pt was −33.6 ± 0.5 mV, which correlated well with data obtained in other studies [39].Conjugation of Au@Pt with Mab led to the converting of zeta potential to −8.17 ± 0.4 mV, which could be attributed to the positive charge of antibodies.To confirm the conjugation between Au@Pt and MAb, the AF4 technique was used.UV-vis absorbance, MALLS, and DLS measurements were conducted at two characteristic wavelengths (280 nm-for the protein and nanozyme components and 400 nm-only for the nanozyme).The obtained AF4 fractograms are presented in Figure 4.As can be seen, the AF4 absorbance fractogram of Au@Pt-MAb conjugate and TBSA as its medium (which contained BSA) comprises the first peak (retention time is 15 min) of the conjugate at 280 nm due to TBSA (Figure 4a).The second peak (retention time is 23 min) of the conjugate Figure 3. UV-Vis spectra of Au@Pt (1) and Mab-Au@Pt (2) (a) and the results of DLS measurements for Au@Pt (1) and Mab-Au@Pt (2) (b).
To confirm the conjugation between Au@Pt and MAb, the AF4 technique was used.UV-vis absorbance, MALLS, and DLS measurements were conducted at two characteristic wavelengths (280 nm-for the protein and nanozyme components and 400 nm-only for the nanozyme).The obtained AF4 fractograms are presented in Figure 4.As can be seen, the AF4 absorbance fractogram of Au@Pt-MAb conjugate and TBSA as its medium (which contained BSA) comprises the first peak (retention time is 15 min) of the conjugate at 280 nm due to TBSA (Figure 4a).The second peak (retention time is 23 min) of the conjugate corresponds to the absorbance at both 280 nm and 400 nm indicating the presence of Au@Pt.However, the absorption at 280 nm is much more intense than that at 400 nm (the peak area is smaller by ~25%) while the scattering for both fractograms was almost the same (Figure 4b).This confirms the presence of protein immobilized on the surface of the nanozyme.Au@Pt nanoparticles were not stable during fractionation and probably aggregated at the focusing stage.
corresponds to the absorbance at both 280 nm and 400 nm indicating the presence of Au@Pt.However, the absorption at 280 nm is much more intense than that at 400 nm (the peak area is smaller by ~25%) while the scattering for both fractograms was almost the same (Figure 4b).This confirms the presence of protein immobilized on the surface of the nanozyme.Au@Pt nanoparticles were not stable during fractionation and probably aggregated at the focusing stage.

ICA of S. typhimurium Using AuNPs
First, the ICA based on traditional AuNPs as a label was developed.In the sandwich format, the first capture antibodies are immobilized in the T zone.During the assay, the detected antigen from the test sample interacts with the second capture antibodies labeled by a marker and immobilized on the conjugate pad.The resulting complex moves to the T zone and binds there forming the first colored band.An excess of labeled antibodies moves on and binds to anti-species antibodies in the C zone to form a second colored band.Thus, in the presence of an analyte in the sample, a colored band occurs in the T zone and its intensity is directly proportional to the analyte concentration.In our case, both the first and second capture antibodies were of the same clone.To achieve the minimum LOD, the ICA was optimized by varying the concentrations of immunoreagents and the assay duration (Table S2).To form the T zone, the concentration of Mab was varied from 0.25 to 2 mg/mL.It was shown that the signal intensity increases in the concentration range of 0.25-1 mg/mL and remains constant at concentrations above 1 mg/mL, so the concentration of 1 mg/mL was chosen as the optimal one.For the formation of the C zone, the optimal concentration of GAMI was 0.5 mg/mL, which provided approximately the same intensity of the colorimetric signal in both zones at high cell concentrations.OD520 of the MAb-AuNPs conjugate was varied in the range of 1-8.In this interval, the signal intensity in the T zone increased, but at OD520 > 6, nonspecific background coloration was observed, therefore, OD520 = 6 was chosen as optimal.Labeled antibodies were applied with the loading of 16 µL/cm of the glass fiber membrane.This ensured the formation of intensely colored zones during the analysis combined with the complete washout of the reagent from the start of the strip and the absence of nonspecific coloration of the working membrane.
The calibration curve of S. typhimurium cells obtained under optimized conditions is presented in Figure 5.The visual LOD for S. typhimurium cells was 3 × 10 4 CFU/mL and the linear range was 4.5 × 10 5 -7.7 × 10 6 CFU/mL.

ICA of S. typhimurium Using AuNPs
First, the ICA based on traditional AuNPs as a label was developed.In the sandwich format, the first capture antibodies are immobilized in the T zone.During the assay, the detected antigen from the test sample interacts with the second capture antibodies labeled by a marker and immobilized on the conjugate pad.The resulting complex moves to the T zone and binds there forming the first colored band.An excess of labeled antibodies moves on and binds to anti-species antibodies in the C zone to form a second colored band.Thus, in the presence of an analyte in the sample, a colored band occurs in the T zone and its intensity is directly proportional to the analyte concentration.In our case, both the first and second capture antibodies were of the same clone.To achieve the minimum LOD, the ICA was optimized by varying the concentrations of immunoreagents and the assay duration (Table S2).To form the T zone, the concentration of Mab was varied from 0.25 to 2 mg/mL.It was shown that the signal intensity increases in the concentration range of 0.25-1 mg/mL and remains constant at concentrations above 1 mg/mL, so the concentration of 1 mg/mL was chosen as the optimal one.For the formation of the C zone, the optimal concentration of GAMI was 0.5 mg/mL, which provided approximately the same intensity of the colorimetric signal in both zones at high cell concentrations.OD 520 of the MAb-AuNPs conjugate was varied in the range of 1-8.In this interval, the signal intensity in the T zone increased, but at OD 520 > 6, nonspecific background coloration was observed, therefore, OD 520 = 6 was chosen as optimal.Labeled antibodies were applied with the loading of 16 µL/cm of the glass fiber membrane.This ensured the formation of intensely colored zones during the analysis combined with the complete washout of the reagent from the start of the strip and the absence of nonspecific coloration of the working membrane.
The calibration curve of S. typhimurium cells obtained under optimized conditions is presented in Figure 5.The visual LOD for S. typhimurium cells was 3 × 10 4 CFU/mL and the linear range was 4.5 × 10 5 -7.7 × 10 6 CFU/mL.

Common and Enhanced ICAs S. typhimurium Using Au@Pt Nanozyme
Common ICA with a nanozyme is similar to that with AuNPs.The only difference is the color of the bands formed on the test strip: in the case of a nanozyme label, bands have a shade from dark brown to black depending on the concentration of the conjugate.It should be noted that when using test strips in the usual configuration with a paper sample pad, nanozymes tend to stick to the sample pad.This effect was shown in our previous work [38] and confirmed in this study.Therefore, in order not to cut the composite under the lower edge of the working membrane, the type of the sample pad was changed, namely, instead of paper, a glass fiber membrane (the same as the conjugate pad) was used.Moreover, it was additionally impregnated with PBST 0.1 to increase the mobility of the labeled conjugate and eliminate non-specific interactions.This replacement ensured the free movement of labeled antibodies along the test strip.(5), 1 × 10 5 (6), 3 × 10 4 (7), 1 × 10 4 (8), and 1 × 10 3 (9) CFU/mL (n = 3).

Common and Enhanced ICAs of S. typhimurium Using Au@Pt Nanozyme
Common ICA with a nanozyme is similar to that with AuNPs.The only difference is the color of the bands formed on the test strip: in the case of a nanozyme label, bands have a shade from dark brown to black depending on the concentration of the conjugate.It should be noted that when using test strips in the usual configuration with a paper sample pad, nanozymes tend to stick to the sample pad.This effect was shown in our previous work [38] and confirmed in this study.Therefore, in order not to cut the composite under the lower edge of the working membrane, the type of the sample pad was changed, namely, instead of paper, a glass fiber membrane (the same as the conjugate pad) was used.Moreover, it was additionally impregnated with PBST0.1 to increase the mobility of the labeled conjugate and eliminate non-specific interactions.This replacement ensured the free movement of labeled antibodies along the test strip.
The optimization of the common ICA with nanozymes consisted of the selection of reagent concentrations and analysis time (see the varied and finally selected parameters of the ICA in Table S2).It was shown that an increase in the concentration of MAb immobilized in the T zone up to 2 mg/mL (concentrations from 0.75 to 2.5 were tested) led to the increase of the analytical signal and, consequently, the decrease of LOD of S. typhimurium.The GAMI concentration of 0.5 mg/mL was chosen from the interval of 0.2-0.6 mg/mL (less concentration led to a decrease in the intensity of the C zone while higher concentration resulted in no change in the color intensity).The dilution of the initial MAb-Au@Pt preparation for application to the conjugate pad was 1:13, which provided highintensity bands and economical consumption of the reagent.A 10 min duration proved sufficient for the ICA; within this period, the entire volume of the reaction mixture (75 µL) was completely absorbed by the test strip, which also contributed to the saving of reagents while maintaining the maximum rapidity of analysis.The calibration curve of S. typhimurium obtained under optimized conditions is shown in Figure 6a.The S. typhimurium LOD was 2 × 10 4 CFU/mL and the linear range was 8 × 10 5 -2 × 10 7 CFU/mL.The estimated LOD was approximately the same as that of the AuNPs-based ICA.The optimization of the common ICA with nanozymes consisted of the selection of reagent concentrations and analysis time (see the varied and finally selected parameters of the ICA in Table S2).It was shown that an increase in the concentration of MAb immobilized in the T zone up to 2 mg/mL (concentrations from 0.75 to 2.5 were tested) led to the increase of the analytical signal and, consequently, the decrease of LOD of S. typhimurium.The GAMI concentration of 0.5 mg/mL was chosen from the interval of 0.2-0.6 mg/mL (less concentration led to a decrease in the intensity of the C zone while higher concentration resulted in no change in the color intensity).The dilution of the initial MAb-Au@Pt preparation for application to the conjugate pad was 1:13, which provided high-intensity bands and economical consumption of the reagent.A 10 min duration proved sufficient for the ICA; within this period, the entire volume of the reaction mixture (75 µL) was completely absorbed by the test strip, which also contributed to the saving of reagents while maintaining the maximum rapidity of analysis.The calibration curve of S. typhimurium obtained under optimized conditions is shown in Figure 6a.The S. typhimurium LOD was 2 × 10 4 CFU/mL and the linear range was 8 × 10 5 -2 × 10 7 CFU/mL.The estimated LOD was approximately the same as that of the AuNPs-based ICA.
The enhanced ICA was carried out under the same conditions as a common one with the addition of a catalytic enhancement step.Nanozyme having peroxidase-like properties can catalyze the oxidation reaction of the peroxidase substrate followed by the formation of an insoluble colored product.The additional coloration contributes to the enhancement of the colorimetric signal revealing those T zones that were invisible to the naked eye in the ICA without enhancement.Thus, an increase in the sensitivity of the determination of S. typhimurium is expected.In the enhanced ICA, the concentrations of reagents and the analysis time were the same as in the common ICA with nanozyme.The two most common peroxidase substrates, TMB and DAB, were used for the catalytic reaction.When applying TMB, non-specific coloration at the zero point (no S. typhimurium cells in the sample) and uneven background spotting of the entire test strip were observed.When using DAB, such undesirable side effects were absent but the volume and location of the substrate applied to the test strip were critical.In addition, a short washing step was added to eliminate residues of labeled antibodies on the working membrane and, accordingly, possible background coloration.It was shown that when the strip was washed for 3 min with buffer and a minimum volume of DAB (1 µL) was applied exactly to the T zone, the coloration of exactly the T zone was observed and no background signal at the zero point was registered.The standard curve of S. typhimurium in the enhanced ICA is shown in Figure 6b.As can be seen, the amplification stage enabled an increase in the signal intensity and reduction of the S. typhimurium LOD by two orders of magnitude down to 2 × 10 2 CFU/mL.The linear range was 3.2 × 10 5 -1.2 × 10 7 CFU/mL.The enhanced ICA was carried out under the same conditions as a common one with the addition of a catalytic enhancement step.Nanozyme having peroxidase-like proper ties can catalyze the oxidation reaction of the peroxidase substrate followed by the for mation of an insoluble colored product.The additional coloration contributes to the en hancement of the colorimetric signal revealing those T zones that were invisible to th naked eye in the ICA without enhancement.Thus, an increase in the sensitivity of th determination of S. typhimurium is expected.In the enhanced ICA, the concentrations o For the developed ICA, intra-assay and inter-assay CV did not exceed 15% in the linear range of S. typhimurium concentrations.The study of the stability of the test strips demonstrated that the functionality of the test system was retained after its storage for at least 3 months at room temperature.For the ELISA, samples were additionally 10-fold diluted to fit the linear range.The recovery values estimated by the ICA were compared with those obtained by the ELISA as a reference method.As can be seen from Table 1, good convergence of the ICA and ELISA results was observed, which confirms the accuracy of the developed method in detecting Salmonella cells in real samples.

Comparison with Other Studies
Table 2 presents the literature data on the ICA of Salmonella cells, sorted according to the need for additional separation/detection equipment.As can be seen, the first developments in the immunochromatographic determination of Salmonella appeared more than ten years ago.As expected, the assays were based on the use of such a traditional marker as AuNPs and were characterized by rather high LODs-from 10 4 to 10 8 CFU/mL [43,44].AuNPs continued to be used in subsequent studies without significant progress in the sensitivity of Salmonella detection [19,21,45,46].As the method evolved and different approaches were designed and applied to amplify the analytical signal, the analytical performance of ICAs of S. typhimurium was improved.Thus, the use of liposomes with encapsulated sulforhodamine [24], lanthanide label combined with immunomagnetic separation [47], the effect of increasing the concentration of the label due to salt-induced aggregation [27], increased signal based on gold growth [26], signal amplification in the SERS-based ICA [48], and ICA with MoS 2 or graphene as labels [49] allowed reducing LODs down to 10 3 CFU/mL (less often-down to 10 2 CFU/mL).However, in some studies, such new approaches and manipulations as the use of multifunctional labels [20,50] or magnetic particles [51] did not completely give the desired effect leaving the assay sensitivity on par with that when using simple AuNPs (10 4 -10 5 CFU/mL).Only a few complex approaches that require special detection equipment (i.e., SERS signal enhance-ment) or difficult-to-obtain composite labels have reduced the LOD down to 10 2 -tens of CFU/mL [28,52,53].The objective of our research was to create an analytical system based on a new combination of nanozyme/detected analyte, which, in contrast to the works presented in the literature (see Table 2), would be characterized by (1) maximum simplicity of the analytical procedure, (2) non-instrumental detection, (3) simple pre-treatment of real samples, (4) maximum rapidity, (5) high specificity for S. typhimurium, and (6) possibility to detect the pathogen in a panel of detected samples (both liquid and solid).Consequently, obtaining the label with new properties was fast and unsophisticated.Au@Pt nanozyme was easily synthesized and ensured an intense reproducible signal.The analytical procedure was one step.No additional equipment was required for the detection of test results; cell presence could be estimated by the naked eye down to a concentration of 2 × 10 2 CFU/mL.The specificity of the developed test system was very high: among 11 different pathogens of the Salmonella genus and other foodborne, only S. typhimurium could be revealed, which enabled the differentiation of this pathogen among numerous species that cause similar symptoms and clinical manifestations.Despite the introduction of additional steps of test strip washing and catalytic reaction, the analysis time was only 15 min.Moreover, ICA was implemented on an expanded panel of real samples (liquid and solid) including milk with different contents of fat and chicken meat.A simple sample preparation procedure provided sample processing during 3 min for milk (dilution with buffer) and 35 min for chicken meat processing.
The extended matrix panel essentially contributed to the characterization of this nanozyme as a marker because it was confirmed that the components of different real samples did not influence its functionality.As the additional functionality confirmation, the applicability of the MAb-Au@Pt conjugate for the detection of a bacterial polyvalent antigen was shown.In this case, a quite different manner of the assembly of labeled immune complexes than for low molecular weight or protein antigens (and, accordingly, a change in the flow process) can be noted.Therefore, detection of the foodborne pathogen

Figure 2 .
Figure 2. Microphotographs of AuNPs with smaller (a) and larger (c) diameters and Au@Pt nanozyme (e) and the corresponding histograms of size distribution (b,d,f) and EDS spectra of AuNPs and Au@Pt nanozyme (g).

Table 1 .
Recoveries of S. typhimurium from cow milk and chicken meat (n = 5) estimated by the developed ICA and the ELISA as a reference method.

Table 2 .
Characteristics of the ICAs of Salmonella.