Fluorescent Assay for Salmonella Detection Based on Triangle Multivalent Aptamer-Initiated Catalytic Hairpin Assembly

Abstract

Salmonella poses a severe global threat to food safety and public health, necessitating rapid, sensitive, and reliable detection methods. Conventional techniques often suffer from complexity, time consumption, cost, or limited sensitivity. To address this, we developed a novel enzyme-free fluorescence detection platform, termed the MTAI-CHA system, integrating magnetic nanoparticle-based triangle multivalent aptamer-initiators (MTAI) with catalytic hairpin assembly (CHA) signal amplification. The triangular DNA nanostructure contained significantly enhanced binding affinity of multivalent aptamers, increasing the sensitivity compared to monovalent aptamers. The optimized MTAI-CHA system demonstrated exceptional performance: a low detection limit of 10 CFU/mL and excellent specificity against non-target pathogens. This sensitive, specific, and robust strategy, leveraging multivalent aptamer recognition and enzyme-free signal amplification, holds significant potential for rapid pathogen screening in food safety, clinical diagnostics, and environmental monitoring, with adaptability to other targets via aptamer substitution.

1. Introduction

Salmonella, a ubiquitous but invisible foodborne pathogen in the daily food supply chain, poses significant threats to global public health. Its pathogenicity must be considered, as consuming Salmonella-contaminated food without precaution can unfortunately lead to a range of symptoms, including gastroenteritis, sepsis, and typhoid fever [1,2]. Annually, Salmonella causes millions of illnesses worldwide, and in severe cases, it can lead to hospitalization or even death [3]. Compounding the challenge, Salmonella exhibits antimicrobial resistance [4] and is difficult to visually distinguish in food. Consequently, developing efficient and reliable detection methods is critical for ensuring food safety. The conventional methods for identifying Salmonella must go through processes such as enrichment culture, selective isolation, observation of morphological characteristics, physiological and biochemical reactions, and final serological identification, which are complex, time-consuming, and labor-intensive. Alternative immunoassays require costly antibodies [5]. Molecular biology methods such as PCR [6] and LAMP [7], while faster, necessitate DNA extraction and amplification, making them susceptible to aerosol contamination and prone to false-positive or false-negative results. Therefore, there is an urgent need for sensitive, rapid, and simple Salmonella detection methods.

Aptamers represent a viable antibody alternative, widely employed in biomarker discovery due to their specificity, affordability, remarkable stability, ease of synthesis, and chemical versatility [8,9]. Aptamers as single-stranded nucleic acid oligomers often undergo significant structural changes after binding onto the target and enhance aptasensor-switching activity after hybridization to the aptamer’s complementary DNA strand [10]. The aptamer-complementary DNA functions as an aptamer blocker, while the aptamer itself serves as the ligand binder. Upon ligand binding, the aptamer undergoes dehybridization, releasing the free complementary DNA. Then this released DNA is coupled with downstream detection applications, such as fluorescence [11], electrochemistry [12], and signaling cascades [13]. However, aptamers still face challenges in detecting bacteria due to the reduced binding affinity leading to the decreased ligand dependence and fewer free complementary DNA. In recent years, multivalent aptamers have emerged as a focal point of research owing to their ability to engage in multivalent interactions. These interactions, in contrast to monovalent interactions, can significantly enhance the effective concentration of the ligand at the target site, thereby strengthening binding affinity and contributing to improved selectivity in target recognition [14,15]. And to improve avidity owing to increased local aptamer density and facilitating synergistic binding, multivalent aptamers have been modified on DNA nanostructures [16]. Nowadays, multivalent binding has been popularly adopted for enhancing the affinity of aptamers in various fields, such as bacteria detection [17], medical imaging [18], and drug delivery [19]. For example, Qiao et al. [20] successfully constructed a tetrahedron multivalent aptamer DNA nanostructure, resulting in a 3.5-fold increase in binding affinity compared to monovalent aptamers. Thus, DNA nanostructure-based multivalent aptamer-complementary DNA shows great potential in binding affinity improved in downstream applications.

In recent years, signal amplification has emerged as a dominant strategy for addressing sensitivity limitations in detection systems. Catalytic hairpin assembly (CHA) is an enzyme-free, isothermal signal-amplification technique for analyte detection based on DNA circuits. DNA circuits utilize the negative free energy change inherent in base pairing to generate amplified duplex DNA from two complementary nucleic acid hairpins [21,22]. Owing to attributes like simple design, high turnover rates and low background signal, CHA finds broad application in detecting trace levels of diverse analytes, particularly nucleic acids [23] and proteins [24]. However, CHA systems operating independently often achieve insufficient signal amplification and sensitivity for low-abundance targets. To enhance performance, cascade amplification approaches that combine CHA with complementary methods, such as DNA nanostructure scaffolds, have emerged, facilitating the creation of highly sensitive detection platforms [25,26].

Herein, a sensitive method based on magnetic nanoparticles-triangle multivalent aptamer-initiator (MTAI) and CHA amplification for Salmonella detection was developed, where the detection platform is termed as a MTAI-CHA system. The principle of the detection strategy is shown in Scheme 1. We synthesized a triangular multivalent aptamer-initiator (Tri-MAI), which has two sides modifying the Salmonella aptamer [27] (Scheme 1. pink DNA) partially blocked with an initiator (Scheme 1. blue DNA) and the other side modifying a link DNA (Scheme 1. yellow DNA). The linker DNA embedded Tri-MAI binds to Streptavidin-coated magnetic nanoparticles (SA-MNPs) to produce MTAI via biotin-streptavidin interaction. Upon introduction of Salmonella, the bacteria and the initiator competed for aptamer binding sites. Due to the superior affinity of Salmonella for the aptamer, the initiator was displaced. Subsequent magnetic separation then isolated the released initiator within the supernatant. The free initiators in supernatant were used to trigger the CHA circuit. Two hairpins H1 (labeled with a FAM fluorophore and BHQ1 quencher) and H2 were involved in the CHA circuit. Initiators can hybridize with H1, leading to the successive hybridization reaction between H1 and H2. Then the FAM fluorophore on H1 was recovered and then released the initiator to trigger the next CHA circuit to generate numerous H1-H2 duplexes. The H1-H2 duplexes achieved intense fluorescence signal owing to the CHA amplification. Therefore, by coupling MTAI with CHA, a convenient, and sensitive method for the detection of Salmonella was achieved.

2. Materials and Methods

2.1. Materials and Apparatus

Phosphate-buffered saline (PBS), TAE buffer, agarose, and 4S GelRed, were sourced from Shanghai Sangon Biotechnology Co., Ltd. (Shanghai, China). Luria Broth (LB) was acquired from Hangzhou Microbial Reagent Co., Ltd. (Hangzhou, China). Brain Heart Infusion Broth (BHI) was sourced from Qingdao Hope Bio-Technology Co., Ltd. (Qingdao, China). 1×TMS buffer was composed of 10 mM Tris, 80 mM MgCl₂. The pH of the 1×TMS buffer was adjusted to 7.5 using glacial acetic acid. Fluorescence was measured under monochromatic blue light excitation (460–485 nm). The UV-Vis and fluorescence data were acquired using a SpectraMax i3 multimode microplate reader (Molecular Devices, San Jose, CA, USA).

2.2. Bacterial Culture

The Salmonella typhimurium (ATCC 14028) as target bacteria was cultivated in BHI liquid culture medium, while the Salmonella enteritidis (CICC 10982), E. coli O157:H7 (ATCC 43889), Vibrio parahaemolyticus (ATCC 17802), S. aureus (ATCC 25923), Listeria monocytogenes (CICC 21540), Vibrio parahaemolyticus (CICC 21617), Vibrio alginolyticus (ATCC 17749), and Vibrio vulnificus (ATCC 27562) as non-target bacteria were cultured with LB liquid culture medium at 37 °C for 20 h in a shaking incubator. The bacterial cultures were washed thrice by centrifugation at 6000 rpm for 8 min, followed by serial dilution in sterile PBS to achieve concentrations ranging from 10 to 10⁷ CFU/mL. For Salmonella quantification, 200 mL of selected dilutions were spread-plated on Xylose-Lysine-Desoxycholate (XLD) agar plates. And then, they were incubated at 37 °C for 24 h. Colonies on the plates were recorded to calculate viable cell concentrations in terms of colony forming units per milliliter (CFU/mL). Finally, the obtained bacteria were subsequently inactivated by boiling for 5 min and refrigerated at 4 °C for storage.

2.3. Preparation of Hairpin Probes

In our design, the initiator not only participates in the CHA reaction but also is complementary to partial sequences of the aptamer. NUPACK software (NUPACK 4.0.) was used to develop and test all DNA oligonucleotides, and Sangon Co., Ltd. (China) provided HPLC-purified DNA oligonucleotides. The initiator and two hairpin probes (H1, H2. The 5′ end of hairpin H1 was marked with FAM, and the 3′ end was marked with BHQ1.) dissolved in PBS were denatured at 95 °C for 10 min and then gradually cooled to room temperature over a 4 h period. All annealed probes were stored at 4 °C for further use.

2.4. Gel Electrophoresis and Fluorescence Analysis

The result of CHA signal amplification was demonstrated with agarose gel electrophoresis under optimized experimental conditions. The trigger H1 and H2 were mixed according to experimental needs. The CHA reactions were performed at room temperature for 30 min. 1 μL loading buffer was added into 5 μL of initiator (0.5 μM), 5 μL of H1 (1 μM), 5 μL of H2 (1.5 μM), 5 μL H1/H2, 5 μL H1/H2/initiator, and DNA marker, respectively. The mixture was loaded onto agarose gel and ran in 1× tris-acetate-EDTA (TAE) buffer at 130 V for 30 min. The results of gel electrophoresis were developed under Bio-Rad Gel Doc XR+ system (Bio-Rad, Richmond, CA, USA) instrument.

2.5. Modification and Detection of Monovalent Aptamer-Initiator (Mono-AI) on the Surface of Streptavidin Magnetic Beads (SA-MNPs)

First, 10 μM Salmonella aptamers and 15 μM initiator were mixed and prehybridized at 95 °C for 5 min and gradually cooled down to form AI. Second, 4 μL of the 10 mg/mL SA- MNPs w washed with 50 μL of PBS and then resuspended in 25 μL Mono-AI mixture and incubated for 0.5 h at room temperature. Then the supernatant with excess Mono-AI was discarded through magnetic separation and washed three times with 200 μL of PBS. After that, added 25 μL of the pre-treated Salmonella solution 10⁶ to the Mon-AI modified magnetic beads and incubate at room temperature for 1 h. Finally, the supernatant containing initiator were separated. 1 μM H1 and 1.5 μM H2 were added into the supernatant incubated at room temperature for 1 h to perform the CHA reaction. The generated fluorescent signals were recorded from 510 nm to 550 nm.

2.6. Modification and Detection of Triangle Multivalent Aptamer-Initiator (Tri-MAI) on the Surface of Streptavidin Magnetic Beads (SA-MNPs)

First, we synthesized a triangular planar multivalent aptamer-initiator, and all the DNA sequences were dissolved in 1×TMS buffer to a final concentration of 50 μM before operation. Four equal amounts of DNA sequences with a final concentration of 1 μM and an initiator with a final concentration of 4.5 μM were added in 1×TMS buffer. The solution (25 μL) was heated to 95 °C for 4 min and rapidly cooled to 4 °C, and the solution was kept at 4 °C for 1 min to fabricate Tri-MAI. The free strands were removed by ultrafiltration to obtain the pure Tri-MAI. Then, the way of the Tri-MAI binds to the magnetic beads is consistent with the above monovalent, and the method for subsequent detection of bacteria is also consistent. All the Nucleic acid sequences used in MTAI-CHA system were shown in

Table 1. Nucleic acid sequences used in MTAI-CHA system.

Name	Sequence (5′–3′)
Apt1(linker1-apt).	CTATCATGGCAAGGTCCTACGCAAAACTCCTCTGACTGTAACCACGGTGGTTTGATCACTATTGGGCCTTTGTGATGTCGGTAGT
Apt2(linker2-apt)	CTACGTCAGCGTTAGACTGAACAAAACTCCTCTGACTGTAACCACGGTGGTTTGATCACTATTGGGCCTTTGTGATGTCGGTAGT
Tri	CGAGTCATCTGTCTACTGAGCCAAAAAAAAA-biotin
Linker	GATGACTCGTTTCAGTCTAACGCTGACGTAGTCGTAGGACCTTGCCATGATAGTGCTCAGTAGACA
H1	FAM-GGTTACAGTTATGTGTACCACTGTAACCACGGTG-BHQ1
H2	TATGTGTACCTGGTTACAGTGGTACACATAACTGTAACC
Initiator	CACCGTGGTTACAGT

.

2.7. Preparation of Spiked Samples

Locally sourced milk, eggs, and chicken were selected as food samples. Milk was centrifuged (6000 rpm, 10 min) to remove some impurities, and the resulting supernatant was spiked with Salmonella. Eggshells were cleaned with 70% (v/v) ethanol and broken with sterilized forks. The egg white was then collected by centrifuging at 1000 rpm for 5 min. 25 g of chicken meat were added to 225 mL of PBS and mixed evenly for about 2 min in homogenizer. The suspension was then centrifuged at 2000 rpm for 10 min, and the supernatant was collected as the chicken meat matrix. Different concentrations (10¹–10⁷ CFU/mL) of Salmonella were added to the milk, eggs, and chicken meat through dilution. The samples with added Salmonella were tested using the proposed detection method. In addition, a traditional plate counting method was used to verify the bacterial concentrations in the samples with added Salmonella.

3. Results and Discussion

3.1. Construction and Characterization of the MTAI

As shown in Figure 1A, agarose gel electrophoretic method was performed to demonstrate the Tri-MAI structure self-assembly. Following incremental addition of strands from lane 1 to lane 3, a marked decline in electrophoretic mobility could be observed, resulting from the increased valence of the aptamer. The triangle multivalent aptamer hybridized with the linker DNA (lane 4) exhibited reduced electrophoretic mobility compared to the three single-stranded sequences (lane 1, lane 2 and lane 3). By comparing lane 5 (the DNA hybridization between aptamer and initiator) with lane 4, the slower band could be observed in lane 5, demonstrating that the Tri-MAI had been successfully formed. The geometric structure and size of the Tri-MAI were characterized by atomic force microscopy (AFM) measurement as shown in Figure 1B. The triangular geometry is roughly observed, and the average side length (ASL) of Tri-MAI is 8.15 nm, which is little higher than the theoretical value (7.14 nm) because of the well-known tip-broadening effect [28]. The cross-sectional profile of the triangle taken along the red line in Figure 1B showed the height of the Tri-MAI is about 1.6 nm (Figure 1C), which is consistent with the theoretical value of dsDNA. Next, the hydrated diameter of the MTAI was characterized by dynamic light scattering (DLS). The average hydrated diameter of streptavidin MNPs was 310.7 nm and it was enlarged to 568.3 nm, indicating the successful preparation of MTAI (Figure 1D). For the evolution of zeta potentials of MNPs, MTAI were also conducted, and the results were determined as—20.01 mV and—32.85 mV in Figure 1E, demonstrating the negative charged Tri-MAI had been successfully attached on the surface of MNPs to increase the zeta potential. Then the feasibility of triggered initiator release based on aptamer recognition owing to the competition between Salmonella and initiator for the binding site of the aptamer was validated in Figure 1F.

In the presence of Salmonella, the fluorescence intensity was enhanced significantly, showing a large amount of FAM labeled initiator entered the supernatant and an almost inconspicuous fluorescence signal in the absence of Salmonella. The result has revealed the competition between Salmonella and initiator for aptamer successfully. Moreover, comparing the monovalent aptamers with multivalent aptamer, as shown in Scheme 2, the Tri-MAI structure shows about twice the sensitivity to Salmonella than to Mono-AI (Figure 1G), indicating the superior multivalence interactions of Tri-MAI.

3.2. Validation of the CHA Performance

The practicality of the proposed CHA performance was confirmed through dual mode verification: DNA gel electrophoresis and fluorescence detection. As Figure 2A demonstrated, through DNA hybridization and unwinding, initiator-mediated unfolding of the H1 hairpin structure led to the cessation of FRET [29] between FAM and BHQ1, thereby causing a significant increase in fluorescence. Then H2 complements with H1, generating the H1-H2 duplex, by which the initiator could be released to catalyze a repeated CHA cycle, generating stronger fluorescence. In the CHA cycle, the electrophoresis image in Figure 2B shows that initiator (lane B), H1 (lane C) and H2 (lane D) were consistent with our design. In the presence of the initiator, the complexes generated by the initiator unfolding the H1 hairpin were observed in lane E, indicating that the formation of complexes led to increased size and slower motion rate. While the mixture of initiator and H2 (lane F) as well as the mixture of H1 and H2 (lane G) showed no change in motion rate, it indicated that H1 and H2 do not react in the absence of the initiator, ensuring a low background sign. The slowest motion rate was observed (lane H) when initiator, H1, and H2 were present, confirming successful complex formation of H1-H2. Additionally, the results of the fluorescence experiment demonstrated that the initiator activated the CHA reaction with an obvious FI increase, and that the H1 + H2 + initiator group was 181.5-fold to the H1 + H2 group (Figure 2C), thereby performing effective signal amplification. To confirm the high signal amplification efficiency of CHA, the whole Salmonella detection procedure using Mono-AI was explored in Figure 2D, and it was found that Salmonella with CHA generated a strong fluorescence signal (black line) and achieved 6.85-fold fluorescence signal higher than Salmonella without CHA in 0.5 h (red line). The background fluorescence signal in Salmonella detection was virtually negligible based on the very weak signal observed in the blank control (without Salmonella; blue line). At the same time, the signal amplification effect of Tri-MAI and Mono-AI was also compared in the detection of bacteria with the participation of CHA. As shown in Figure 2E, the fluorescence intensity of Tri-MAI showed a 1.5-fold of Mono-AI, indicating the higher binding affinity of Tri-MAI over the monovalent. All results demonstrated the high signal amplification efficiency of CHA as well as the good feasibility of MTAI-CHA for Salmonella detection.

3.3. Optimization of Experimental Parameters

To identify the Salmonella-binding region of the aptamer, which could be influenced by initiator blocking potentially affecting final fluorescence intensity, three FAM-labeled probe strands, 20 nucleotides in length, complementary to distinct aptamer regions, were designed (Figure 3A). The selection of the most suitable regions was optimized by changing the complementary pairing position of the initiator and adapter as a variable. The other experimental conditions were the same as the previous experiment of capturing Salmonella with the aptamers. Salmonella was added to the aptamer–initiator complex-modified MNPs and the supernatant was collected for fluorescence detection after incubation. It can be observed in Figure 3B that the fluorescence intensity was higher when the initiator paired with the aptamers (IE2). That was probably owing to the affinity of IE2 and aptamers being weaker, so the IE2 was easier to be released. Then, the distance between two adjacent aptamers in the MAI is considered to be a crucial factor influencing binding affinity [30]. We therefore optimized this distance by precisely controlling the length of the DNA (15 bp, 21 bp, 27 bp and 33 bp). From the results in Figure 3C, optimal fluorescence emission is observed in Salmonella samples with 21 bp length of DNA, whereas considerably weaker signal intensities are recorded for both shorter (15 bp) and longer (27 bp, 33 bp) length of DNA. Consequently, the 21 bp length of DNA was selected for subsequent studies. In addition, the time of CHA reaction also affected the formation yield of H1-H2 duplexes. Figure 3D indicates that fluorescence signal increased with more incubation time and reached a plateau at 30min, suggesting 30min was the best reaction time for CHA amplification. The concentration of Tri-MAI modified to the MNPs also plays an important role in this approach. As we can see in Figure 3E, the fluorescent signal increased gradually until the 3 μM, leading to the sufficient capture of Salmonella. The time of Salmonella incubation with aptamers-modified MNPs affected the detection as well, and the sample signal did not grow higher after 90 min incubation in Figure 3F. Hence, we chose 90 min as the optimal incubation time.

3.4. Sensitivity of the Assay

Under optimized experimental conditions, fluorescence signal intensity was measured at varying Salmonella concentrations to evaluate the quantitative detection capability of the assay. The fluorescence profiles with Salmonella concentrations of 0 ~ 1 × 10⁷ CFU/mL are presented in Figure 4A, and the detectable limit of Salmonella is 10 CFU/mL. We ascribe the improvement of good sensitivity in this detection to the higher apparent affinity of multivalent aptamer binding on the Salmonella surface and the participation of CHA signal amplification. Therefore, it could be reasoned that this proposed detection method had a good sensitivity for the detection of Salmonella and one should consider applying this method to the detection of other pathogenic bacteria to contribute to food safety testing.

3.5. Specificity of the Assay

To confirm whether the increase in the fluorescence signal was specifically induced by Salmonella, other nontarget pathogens including Salmonella enteritidis, E. coli O157:H7, S. aureus, Listeria monocytogenes, Vibrio parahaemolyticus, Vibrio alginolyticus, and Vibrio vulnificus were selected as the target analytes to react with Tri-MAI. As depicted in Figure 4B, only the presence of target Salmonella or the mixture containing all species caused the sensing system to produce a very strong fluorescence signal, while all nontarget analytes only yielded fluorescence signals indistinguishable from the background signal. This marked contrast demonstrated the high specificity of this assay for Salmonella detection, attributed to the use of Salmonella aptamers as selective recognition elements.

3.6. Analysis of Real Samples

To further investigate the practical application of the assay, different concentrations of Salmonella were spiked into milk, egg whites, and chicken meat as simulated practical complex samples. As depicted in Figure 5, the fluorescence intensity increased as the concentration of Salmonella increased. And the detection yielded positive signals when the spiked concentrations of Salmonella was 10 CFU/mL, indicating the suitability of the proposal for Salmonella detection in real samples.

4. Conclusions

In summary, a simple, robust and enzyme-free fluorescence assay for Salmonella detection was developed by combining triangle multivalent aptamer with high binding affinity and CHA signal amplification for the first time. Benefiting from the released initiators that are bound to aptamers and CHA signal amplification, the proposed method exhibits high sensitivity for Salmonella detection with low limit of detection of 10 CFU/mL, suggesting the potential for directly detecting Salmonella from samples. Furthermore, this wide detection range confirms the suitability of this assay for direct Salmonella detection, eliminating the need for sample dilution or enrichment. Consequently, the platform holds significant potential for universal pathogen detection. By simply substituting the aptamer sequence with those specific to different targets, it can be readily adapted to identify diverse pathogens, offering substantial utility in food safety surveillance, environmental monitoring, and clinical diagnostics.