Point-of-Care Diagnostic Devices for Detection of Escherichia coli O157:H7 Using Microfluidic Systems: A Focused Review

Bacterial infections represent a serious and global threat in modern medicine; thus, it is very important to rapidly detect pathogenic bacteria, such as Escherichia coli (E. coli) O157:H7. Once treatments are delayed after the commencement of symptoms, the patient’s health quickly deteriorates. Hence, real-time detection and monitoring of infectious agents are highly critical in early diagnosis for correct treatment and safeguarding public health. To detect these pathogenic bacteria, many approaches have been applied by the biosensors community, for example, widely-used polymerase chain reaction (PCR), enzyme-linked immunosorbent assay (ELISA), culture-based method, and adenosine triphosphate (ATP) bioluminescence. However, these approaches have drawbacks, such as time-consumption, expensive equipment, and being labor-intensive, making it critical to develop ultra-sensitive and highly selective detection. The microfluidic platform based on surface plasmon resonance (SPR), electrochemical sensing, and rolling circle amplification (RCA) offers proper alternatives capable of supplementing the technological gap for pathogen detection. Note that the microfluidic biochip allows to develop rapid, sensitive, portable, and point-of-care (POC) diagnostic tools. This review focuses on recent studies regarding accurate and rapid detection of E. coli O157:H7, with an emphasis on POC methods and devices that complement microfluidic systems. We also examine the efficient whole-body detection by employing antimicrobial peptides (AMPs), which has attracted growing attention in many applications.


Introduction
Ingesting contaminated food such as chicken, beef, milk, juice, vegetables, and tap water not only endangers human health but also results in significant financial losses, posing a serious threat to public health [1]. Among the pathogenic bacteria, Escherichia coli (E. coli) O157:H7 is a quite important foodborne or waterborne infectious pathogen of global concern, as well as the most common and lethal serotype of enterohemorrhagic E. coli [2]. In addition, contamination of water sources with this pathogenic bacteria is a big problem in both developing and developed countries [3]. Infants, children, immunocompromised individuals, and the elderly are the most vulnerable to waterborne infections. Ingestion and response to E. coli O157:H7 contamination events. Furthermore, it is necessary to design and fabricate a highly precise and cost-effective pathogenic bacteria diagnostic test that can be easily utilized in underdeveloped nations to combat infectious bacterial/viral infections without causing financial burden. POC devices aimed at clinical diagnostics are a good example of an interface where biomedical and engineering disciplines meet [14][15][16].
The incorporation of leading-edge technology into the microfluidic platform may provide very useful solutions for the relevant issues of diagnostic devices. A microfluidbased system offers advantages of relatively simple operation, low cost, multiple-target testing, ease of automation, portability, and compactness. As a consequence, surface plasmon resonance (SPR), electrochemical, and rolling circle amplification (RCA) have been revitalized due to rapid analysis, highly sensitive and selective performance, and consistent detection limit. Furthermore, these methods incorporated with a microfluidic system can be used in any location or POC, even if there is no electricity accessible, due to the ubiquitous concept.
This review focuses on the detection of E. coli O157:H7 using microfluidic systems, such as POC diagnostic devices including SPR, electrochemical sensing, and RCA. The most important performances for these methods are the limit of detection (LOD) and the detection time. The LOD and detection time were compared with each other from recent literature information, and then the best results of LOD and detection time were discussed for the detection of pathogenic bacteria. In addition, this review also presents the actually advanced E. coli detection by employing antimicrobial peptides (AMPs), which were introduced as emerging alternatives to antibodies. Among the different types of POC, AMP-based microfluidic platforms for pathogen sensing have gained much attention in view of whole-body detection with selectively binding affinity.

SPR-Based E. coli O157:H7 Detection
SPR-based biosensors are rapidly emerging as a direct label-free technology for the rapid and highly sensitive detection of chemical and biological analytes in critical fields, such as medical diagnostics, food safety, and environmental sensing, due to the unique optical properties they possess [17][18][19]. SPR utilizes the simple phenomenon of refractive index changes in conjunction with the binding of target biomolecules (either antigens or protein samples) to identify these changes. The SPR phenomena occur due to the excitement and collective vibration of free electrons in metal and dielectric junctions. This excitation is caused by the interaction of electromagnetic waves in the visible area with the plasmon of gold and silver nanoparticles [20][21][22]. When combined with well-developed microfluidic technology, SPR offers numerous advantages, which are explained and summarized in the concluding section of this review article. The complete binding mechanism is illustrated in Figure 1a,b, where the binding event is indicated by a change in reflectivity ∆R at a fixed observed angle. Before antibody binding, θ 1 (blue) is shown in both panels (panel A and panel B), while after the binding of antibody, θ 2 (red) is observed. The change in reflectance ∆R represents the binding event or sensitivity at the fixed observed angle. In the 1990s, the United States Department of Agriculture (USDA) introduced a commercial SPR apparatus for the detection of E. coli O157:H7, where the antibodies were sufficiently immobilized on the surface of a substrate. Since then, SPR has become a popular label-free tool in the biosensors society for detecting a variety of chemical and biological samples [23], serving as a POC technique in microfluidics.
Numerous researchers have worked in the SPR domain as a POC diagnostic device for the accurate, rapid, and highly sensitive detection of E. coli O157:H7 with no cross-reaction. To achieve this goal, Tokel et al. [24] developed a microfluidic SPR biochip capable of detecting and quantifying E. coli and Staphylococcus aureus. A microfluidic channel with a volume of 4 µL was created by laser cutting a microchannel form in a double-sided adhesive (DSA) layer with a thickness of 50 µm. The DSA layer held together the polymethyl methacrylate (PMMA) layer and the gold-coated glass substrate. The sample was injected into the channel and adhered to the gold surface after opening a single inlet and outflow port on the PMMA layer. For SPR measurement, the prism was placed on the PMMA-DSA-glass setup. Figure 1c,d depicts the custom-made SPR platform for microfluidic integration based on a complementary metal-oxide-semiconductor (CMOS) sensor. The lower part of Figure 1c shows an enlarged view of the microfluidic chip with a gold coating. Figure 1e illustrates the schematic of the microfluidic integrated SPR system, in which the gold surfaces were modified with different activators, including 11-mercaptoundeconoic acid (MUA), N-(3-dimethylaminopropyl)-N -ethylcarbodiimide hydrochloride (EDC), Nhyroxysuccinimide (NHS), and anti-liposaccharide (LPS) antibodies, to capture E. coli. The antibodies in the microchannel capture bacteria, resulting in a change in the local refractive index. This change is detected by the reflected light, collected by the sensor, and delivered to a computer for analysis. The microfluidic SPR sensor was used to investigate the capture of E. coli with a LOD of 10 6 CFU/mL. A detection time of 20 min was observed using this POC microfluidic system, which is considered a good reaction/detection time in biosensor design and fabrication. Previous studies have reported a lower LOD of 94 CFU/mL with SPR, which is approximately 10 4 times better than the previous system [24,25]. Table 1 summarizes the most recent studies, including various parameters such as material type, method, LOD, detection time, detection range for linear analysis (to calculate R 2 value using regression analysis), sample type, and sensor response. thyl methacrylate (PMMA) layer and the gold-coated glass substrate. The sample was injected into the channel and adhered to the gold surface after opening a single inlet and outflow port on the PMMA layer. For SPR measurement, the prism was placed on the PMMA-DSA-glass setup. Figure 1c,d depicts the custom-made SPR platform for microfluidic integration based on a complementary metal-oxide-semiconductor (CMOS) sensor. The lower part of Figure 1c shows an enlarged view of the microfluidic chip with a gold coating. Figure 1e illustrates the schematic of the microfluidic integrated SPR system, in which the gold surfaces were modified with different activators, including 11-mercaptoundeconoic acid (MUA), N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC), N-hyroxysuccinimide (NHS), and anti-liposaccharide (LPS) antibodies, to capture E. coli. The antibodies in the microchannel capture bacteria, resulting in a change in the local refractive index. This change is detected by the reflected light, collected by the sensor, and delivered to a computer for analysis. The microfluidic SPR sensor was used to investigate the capture of E. coli with a LOD of 10 6 CFU/mL. A detection time of 20 min was observed using this POC microfluidic system, which is considered a good reaction/detection time in biosensor design and fabrication. Previous studies have reported a lower LOD of 94 CFU/mL with SPR, which is approximately 10 4 times better than the previous system [24,25]. Table 1 summarizes the most recent studies, including various parameters such as material type, method, LOD, detection time, detection range for linear analysis (to calculate R 2 value using regression analysis), sample type, and sensor response. Figure 1. The principle of SPR as a label-free biosensing technology and SPR-based microfluidic platform for E. coli detection as a POC. (a) The incident light travels along the dielectric interface and excites plasmon at this interface. (b) Accurate binding of a biomolecule (antibody) on the surface of a gold film results in a shift in SPR resonance angle with before antibody binding θ1 (blue) and after binding θ2 (red). The occurrence of binding is denoted as the reflectivity angle (ΔR) with the measured resonance angle fixed. Reprinted with permission from [26]. (c) The disposable microfluidic chips, which are surface activated, are placed on the top of the device. (d) The electronic configuration of the device is depicted from bottom to top. A light-emitting diode (LED) illuminates a cylindrical lens, which focuses the light onto a rectangular prism. The reflected light is captured by a CMOS sensor, and the image is transmitted to a portable computer via control circuitry. (e) The schematic shows a microfluidic integrated SPR platform. Reprinted with permission from [24].
In recent years, numerous studies have aimed at developing SPR biosensors for the fast and accurate detection of E. coli O157:H7. However, these studies often demonstrate the ability to identify bacterial pathogens only at higher concentrations and longer incubation times using expensive equipment, which is inadequate for most hazardous infections. Figure 1. The principle of SPR as a label-free biosensing technology and SPR-based microfluidic platform for E. coli detection as a POC. (a) The incident light travels along the dielectric interface and excites plasmon at this interface. (b) Accurate binding of a biomolecule (antibody) on the surface of a gold film results in a shift in SPR resonance angle with before antibody binding θ 1 (blue) and after binding θ 2 (red). The occurrence of binding is denoted as the reflectivity angle (∆R) with the measured resonance angle fixed. Reprinted with permission from [26]. (c) The disposable microfluidic chips, which are surface activated, are placed on the top of the device. (d) The electronic configuration of the device is depicted from bottom to top. A light-emitting diode (LED) illuminates a cylindrical lens, which focuses the light onto a rectangular prism. The reflected light is captured by a CMOS sensor, and the image is transmitted to a portable computer via control circuitry. (e) The schematic shows a microfluidic integrated SPR platform. Reprinted with permission from [24].
In recent years, numerous studies have aimed at developing SPR biosensors for the fast and accurate detection of E. coli O157:H7. However, these studies often demonstrate the ability to identify bacterial pathogens only at higher concentrations and longer incubation times using expensive equipment, which is inadequate for most hazardous infections. According to the literature, several assays have been employed for the detection of E. coli using the SPR method, including monoclonal antibody, polyclonal antibody, T4 bacteriophage, and AMP with magainin 1-C-based SPR detection. Each method possesses its own advantages and disadvantages, depending on factors such as the LOD, detection time, cost, and commercialization feasibility. The minimum detection time achieved was 10 min using a polyclonal antibody (rabbit anti-goat IgG), while a detection time of 120 min was observed when utilizing a chicken antibody for E. coli detection. Furthermore, the best LOD achieved using the SPR method was 10 CFU/mL. This notable achievement was made possible by detecting E. coli using murine anti-E. coli O157:H7 monoclonal antibodies, utilizing magnetic nanoparticles coated with an Au shell. In conclusion, SPR-based detection of E. coli O157:H7 offers a sensitive and real-time approach for the identification and quantification of the bacteria. This advancement contributes to enhanced food safety measures and improved public health.
The specificity of E. coli detection plays a crucial role in SPR-based biosensors. When specific antibodies or aptamers designed to recognize and bind to E. coli antigens are employed as capture molecules in an SPR assay, the specificity can be significantly high. These capture molecules are carefully selected to have minimal cross-reactivity with other bacterial species or contaminants, ensuring their specific interaction with the target E. coli bacteria. Numerous researchers have investigated the specificity of E. coli O157:H7 using the SPR method. For instance, Li et al. [27] determined the specificity of E. coli O157:H7 against E. coli, S. typhimurium, S. aureus, and V. parahemolyticus. They evaluated the signal intensities and found that the intensities associated with E. coli O157:H7 were almost fivefold stronger than those of the non-target pathogens. The most recent studies that focused on accurate detection using the SPR method are summarized in Table 2.

Electrochemical-Based E. coli O157:H7 Detection
Electrochemical biosensors are commonly produced and widely employed for the detection of food-borne and water-borne pathogens due to the possibilities of miniaturization and the ability to build disposable, flexible, and affordable sensing systems. Over the last decade, various methodologies have been developed for the rapid quantification of E. coli O157:H7. According to the International Union of Pure and Applied Chemistry (IUPAC), an electrochemical biosensor is a self-contained device that utilizes a bioreceptor in conjunction with an electrochemical transduction component to provide quantitative or semi-quantitative analytical data [36]. Many researchers have made efforts to detect E. coli O157:H7 as a POC device using electrochemical methods. For this purpose, Dastider et al. [37] developed a microfluidic platform for accurate and sensitive detection, known as the MEMS sensor, which operates on the principle of dielectrophoresis force.
The key components of this MEMS device are the design of the focusing region and the sensing region. The focusing region generates a positive dielectrophoresis force that moves the cells to the edges of the tilted thin-film electrode fingers in the center of the microchannel. The fluidic drag force then transports the focused cells to the sensing region, which contains multiple pairs of three interdigitated electrode arrays (IDEAs) embedded inside the microchannel. This technique allows for highly concentrated samples in the sensing region. For specific sensing of E. coli 0157:H7, the sensing IDEAs are functionalized with an anti-E. coli antibody. When E. coli binds to the antibody, there is an impedance change. The complete design of this MEMS microfluidic biosensor is shown in Figure 2. The detection time is 120 min and the LOD is 39 CFU/mL. Despite the good LOD, the long detection time makes this device less efficient for a POC application.   In addition, to further improve the LOD and detection time for accurate detection, Yao et al. [38] designed and fabricated a microfluidic biosensor based on electrochemical impedance spectroscopy (EIS) combined with magnetic nanoparticles (MNPs). MNPbacteria-GNP-urease complexes were created by combining MNP-bacteria with gold nanoparticles (GNPs) modified with urease and aptamers against E. coli O157:H7. The complete development is shown in Figure 2d. A satisfactory linear relationship between the relative rate of impedance change of the catalysate and the concentration of the bacteria was established with a low detection limit of 12 CFU/mL and a detection period of 15 min. This represents an excellent LOD and detection time for POC application. The most recent studies on the detection of E. coli O157:H7 are summarized in Table 3, including the LOD, detection time, material type, methodology, assay structure, sample type, and detection range for linearity checking (R 2 using regression analysis). In addition, the most recent studies that have focused on the accurate detection (specificity) using electrochemical methods are summarized in Table 4.   Rolling circle amplification (RCA) driven by DNA polymerase is an effective enzymatic isothermal process. It involves the synthesis of long single-stranded (ss) DNA molecules on a short circular ssDNA template using a single DNA or RNA primer. In addition, RCA can replicate circularized oligonucleotide probes with either linear or geometric kinetics. By modifying the template, introducing functional sequences during the reaction, and hybridizing the RCA products to oligonucleotides with specific functions, RCA can serve multiple purposes. Due to these innovative features, RCA-based nanotechnology has found wide applications in biological detection, medication delivery, and other fields [52]. A brief description of the RCA is as follows: First, a primer is conjugated to the necessary surface by employing surface activation and functionalization. The primer is then hybridized with a padlock probe, which is carefully designed to include a primer-binding site, a pathogen-binding site, and a selfassembly area that generates a molecular dumbbell shape. By regulating the length of each binding area, the padlock probe is designed to possess thermal stability at room temperature and specificity for primer and pathogen hybridization. After annealing, the padlock probe takes an asymmetric dumbbell shape. Subsequently, the probe is hybridized with the primer immobilized on the required surface. Hybridization occurs when the padlock probe comes into contact with a target pathogen. The opening padlock probe is ligated with a ligase to create a closed-loop template that can be utilized in the RCA procedure. During the RCA process, complementary single-stranded DNA gradually extends in a dumbbell form over time by DNA polymerase. The amplification of long DNAs, owing to the dumbbell-shaped template, leads to their easy entanglement with each other, resulting in aggregation with surrounding DNAs and the formation of a DNA gel. This DNA hydrogel has been developed for the visualized, simple, and rapid detection of E. coli O157:H7.
To achieve this goal, many researchers have made efforts to design and develop microfluidics RCA sensors for the rapid and accurate detection of E. coli O157:H7. Recently, Jiang et al. [53] developed an aptamer-based microfluidic platform for the detection of whole cells by applying the dual RCA technique. The complete design of dual-RCA microfluidic biosensor design is illustrated in Figure 3. The detection time and LOD are 120 min and 39 CFU/mL, respectively, which are better results than those with an incubation time of 120 min and LOD of 4.0 × 10 2 CFU/mL [54]. The most recent studies on E. coli detection, including LOD and detection time, material type, methodology, assay structure, and sample type, are summarized in Table 5. Further exploration is still needed for the identification/detection of E. coli in order to improve the detection time and LOD. whole cells by applying the dual RCA technique. The complete design of dual-RCA microfluidic biosensor design is illustrated in Figure 3. The detection time and LOD are 120 min and 39 CFU/mL, respectively, which are better results than those with an incubation time of 120 min and LOD of 4.0 × 10 2 CFU/mL [54]. The most recent studies on E. coli detection, including LOD and detection time, material type, methodology, assay structure, and sample type, are summarized in Table 5. Further exploration is still needed for the identification/detection of E. coli in order to improve the detection time and LOD. In situ capturing RCA (cRCA) is performed to generate repeating aptamers that capture the target cells. Subsequently, detection signals are amplified using signaling RCA (sRCA). The immobilized dendrimers on the PDMS surface provide several handles, allowing for more copies and higher density of dispersed cRCA on the capturing surface. Reprinted with permission from [53].  A schematic illustrates a sandwich detection system where the surface of a microfluidic channel is modified with polyamidoamine (PAMAM) dendrimers. In situ capturing RCA (cRCA) is performed to generate repeating aptamers that capture the target cells. Subsequently, detection signals are amplified using signaling RCA (sRCA). The immobilized dendrimers on the PDMS surface provide several handles, allowing for more copies and higher density of dispersed cRCA on the capturing surface. Reprinted with permission from [53]. PCR, the gold standard method, amplifies specific DNA sequences unique to E. coli O157:H7, enabling rapid and sensitive detection of the bacteria in food samples or clinical specimens. PCR-based assays provide results within hours, making them invaluable for timely identification and response to potential outbreaks. PCR has been extensively employed in laboratory settings for several decades, leading to significant advancements across various fields. When comparing RCA and PCR, both techniques serve as powerful molecular methods for nucleic acid amplification. While PCR is widely established and extensively used, RCA offers distinct advantages in specific applications. The following section outlines a comparative analysis of RCA and PCR, highlighting the advantages of RCA: Amplification Mechanism: The fundamental amplification mechanisms of RCA and PCR differ. PCR utilizes a thermal cycling process to amplify a specific DNA segment, employing heat-stable DNA polymerases. On the other hand, RCA employs a rolling circle mechanism, where a circular template is exponentially amplified by a DNA polymerase with strand-displacement activity. This mechanism enables RCA to generate long, single-stranded DNA products, which can be advantageous for various downstream applications, such as DNA sequencing or in situ hybridization techniques. Sensitivity: In certain scenarios, RCA has demonstrated higher sensitivity compared to PCR. Due to its isothermal nature, RCA can produce a larger number of amplification products, resulting in increased sensitivity for detecting low-abundance targets. This sensitivity advantage has proven particularly useful in applications such as detecting rare genetic mutations, single-cell analysis, or amplifying targets with low copy numbers. Simplified Workflow: RCA offers a simplified workflow compared to PCR. RCA reactions can be performed under isothermal conditions, eliminating the need for sophisticated thermal cycling equipment. This simplification can reduce overall costs and technical complexity associated with amplification procedures, making RCA an appealing option for resource-limited settings or POC applications. Product Length: As mentioned earlier, RCA can generate long, single-stranded DNA products. This feature proves advantageous in applications where longer DNA fragments are desired, such as generating templates for DNA sequencing or studying DNA-protein interactions. In contrast, PCR typically produces shorter amplicons due to limitations inherent in the polymerase enzyme and primer design considerations. Enzyme Selection: RCA can be conducted using various DNA polymerases, including both strand-displacing and nick-translating enzymes. This flexibility allows researchers to choose an appropriate enzyme based on their specific requirements, such as amplification efficiency, fidelity, or compatibility with specific detection methods. In contrast, PCR primarily relies on thermostable DNA polymerases, which may have limitations in certain applications, such as amplifying challenging templates or incorporating modified nucleotides.
While RCA offers several advantages over PCR in certain applications, it is important to note that PCR still maintains its strengths and remains the method of choice for many diagnostic and research purposes. PCR benefits from a well-established protocol, extensive commercial availability of reagents, and a vast body of literature supporting its applications. The selection between RCA and PCR should be based on the specific requirements of the experiment, the desired output, and the available resources. Additionally, the most recent studies focused on this accurate detection (specificity) using RCA method are summarized in Table 6.

Attraction of Antimicrobial Peptide (AMP) for E. coli O157:H7 Detection
The novel antimicrobial compounds have great potential for combating antibiotic resistance in bacteria. As part of the innate immune system, there is a growing demand for these compounds. To address this need, researchers have employed several alternative techniques, including the use of metals, metal oxides, polymer materials, and AMPs [60]. Among these techniques, AMPs show impressive potential for direct detection of both pathogenic and nonpathogenic bacteria. They are also considered strong candidates for whole bacteria detection. AMPs can be found in various natural niches and act as the organism's first line of defense against bacterial attacks, forming an important part of the unique immune system that protects hosts from pathogens [61]. Most AMPs consist of 6 to 50 positively charged amino acid residues and a large number of hydrophobic residues [62]. Due to their excellent stability and low cost, AMPs have been extensively explored as alternative recognition elements in biosensors for bacterial detection. Since the affinity of AMPs is important in detection, the elements that influence the affinity of AMPs will be discussed before addressing various AMP-based approaches. Microfluidics have also been combined with AMP-based approaches to identify foodborne infections, making it a promising new tool. Finally, the future opportunities and challenges in constructing reliable and sensitive AMP-based platforms will be highlighted.
Although the mechanisms of action of AMPs have not been fully elucidated, there is a widely accepted model that explains their mode of action. According to this hypothesis, cationic AMPs bind to bacterial surfaces through electrostatic and hydrophobic interactions, leading to various modifications in membrane structures and disruption of the integrity of the bacterial cytoplasmic membrane, ultimately killing microorganisms. Based on their secondary structures, AMPs' bioactivity toward microbial cells can be categorized into many groups [63]. Many AMPs adopt amphipathic conformations that allow them to target the negatively charged head groups of lipids on the bacterial membrane, utilizing spatial hydrophobic clustering from cationic amino acids. Plant and animal membranes, on the other hand, isolate negative charges from the interior leaflet and contain cholesterols that inhibit AMP activation [64]. AMPs, as antibiotics, have exhibited remarkable resistance to acquired resistance by targeting the fundamental structure of the bacterial cell membrane, making them difficult for proteases to identify. Linear cationic peptides such as magainin, with their small molecular size and inherent stability, are particularly attractive for microbial sensing applications [65]. As a precursor to bactericidal activity, the positively charged AMP magainin I (GIGKFLHSAGKFGKAFVGEIMKS) exhibits selective binding to the pathogenic bacteria E. coli O157:H7 [66,67]. Magainin I also demonstrates broad-spectrum efficacy against other Gram-negative bacteria responsible for a majority of human pathogenic infections.
In AMP-based biosensors, AMPs are often immobilized on substrate surfaces. The immobilized AMPs are believed to interact with the outer membrane through electrostatic contacts, triggering various interactions with the cytoplasmic membrane such as amphipathic conformations and insertion into the membrane, leading to disturbances in the lipid bilayer. Immobilizing peptides may restrict their mobility and ability to interact with cellular membranes compared to their soluble state. However, due to limited research on soluble and immobilized AMPs, it is challenging to determine the similarities and differences in their activities. The strong affinity of AMPs for bacterial surfaces has garnered more attention in biosensing applications than their antibacterial activity because AMP surfaces serve as recognition elements. The AMPs adsorbed on the surface of biosensors are believed to facilitate bacterial attachment through electrostatic and hydrophobic interactions. However, few studies have focused on the interactions of immobilized AMPs with the bacterial surface in biosensors. Despite the unknown recognition processes, AMPs have gained significant attention in biosensing due to their strong affinity for bacteria. The AMP database currently contains over 3000 natural AMPs, with their sequences and structures extensively studied by the research community.
One major advantage of AMPs as recognition elements is their ability to engage with multiple pathogens, thanks to their selective binding nature to target cells. By replacing existing antibody-based probes with more durable and stable antimicrobial peptides in biological sensors, the shelf life of current diagnostic systems can be extended. Due to these advantages, AMPs have been utilized to construct various bacterial biosensors. Numerous researchers have worked on developing microfluidic-based AMPs with specific identification elements for rapid and accurate detection of E. coli O157:H7 [68,69].
Various recognition elements, including magainin I, leucocyte A, clavA, colicin V, and cecropin P1, have been employed for bacterial detection. Among them, magainin I has shown the most promising binding results for reliable detection. Until recently, the majority of AMP research had focused on magainin II, but magainin I has proven to be a suitable bacterial recognition factor. Manoor et al. [69] demonstrated that immobilizing AMP magainin I on a gold microelectrode array via a C-terminal cysteine residue enabled the capture and measurement of the binding effect on pathogenic E. coli. They developed a robust and portable biosensor for the detection of E. coli O157:H7 using impedance spectroscopy for evaluation. The biosensor achieved a detection time of 30 min and a LOD of 1000 CFU/mL in water samples, with a detection range of 10 1 to 10 5 . While this detection time is favorable, the LOD, which is a key factor for POC devices, needs improvement. To further improve the LOD and detection time, Dong and Zhao [70] fabricated the microfluidic E. coli biosensor with an excellent detection time and a lower LOD of 400 CFU/mL, which is 2.5 times better than the previous study [69]. They demonstrated that an AMP specific to E. coli O157:H7, tagged with a cysteine residue at the C-terminal for immobilization on a gold quartz crystal microbalance (QCM) electrode surface, could be used to capture and measure the binding effect. The complete detection mechanism is shown in Figure 4. While these approaches have achieved low bacterial cell detection limits, many of them require extensive sample preparation and, in some cases, multiple reagent and assay steps before detection. Therefore, developing an integrated detection system with high mobility, robustness, sensitivity, and selectivity for harmful organisms remains challenging. Further exploration is needed to improve the detection time and LOD for the identification/detection of E. coli O157:H7.  To further enhance the LOD and decrease detection time, our research group [68] investigated the detection of pathogenic E. coli O157:H7 using a microfluid-based biosensing device embedded with AMP-labeled microbeads. Through effective immobilization of AMP magainin I on the surface of glass microbeads, the LOD as well as the detection time were significantly improved. We achieved an outstanding LOD of 10 CFU/mL with To further enhance the LOD and decrease detection time, our research group [68] investigated the detection of pathogenic E. coli O157:H7 using a microfluid-based biosensing device embedded with AMP-labeled microbeads. Through effective immobilization of AMP magainin I on the surface of glass microbeads, the LOD as well as the detection time were significantly improved. We achieved an outstanding LOD of 10 CFU/mL with a detection time of less than 20 min. In addition, we successfully detected the entire bacteria, which is crucial for sensing pathogenic bacteria. Furthermore, we explored this microfluidic platform for effective regeneration capabilities. The strategic fabrication of our system is illustrated in Figure 5, highlighting the importance of the weir structure in the microfluidic channel. Figure 6 provides an explanation of the binding mechanism between the entire E. coli O157:H7 and the AMP-immobilized glass microbeads.
sor for E. coli O157:H7 detection. (Step I) The AMPs with C-terminal are immobilized on a gold QCM electrode surface. (Step II) The surface is blocked with mercaptohexanol (MCH) to inhibit unnecessary sites for enhancement of the signal−to−noise ratio. (Step III) Implementation and detection of E. coli occur. Reprinted with permission from [70].
To further enhance the LOD and decrease detection time, our research group [68] investigated the detection of pathogenic E. coli O157:H7 using a microfluid-based biosensing device embedded with AMP-labeled microbeads. Through effective immobilization of AMP magainin I on the surface of glass microbeads, the LOD as well as the detection time were significantly improved. We achieved an outstanding LOD of 10 CFU/mL with a detection time of less than 20 min. In addition, we successfully detected the entire bacteria, which is crucial for sensing pathogenic bacteria. Furthermore, we explored this microfluidic platform for effective regeneration capabilities. The strategic fabrication of our system is illustrated in Figure 5, highlighting the importance of the weir structure in the microfluidic channel. Figure 6 provides an explanation of the binding mechanism between the entire E. coli O157:H7 and the AMP-immobilized glass microbeads.
Most recent studies for the E. coli detection based on AMPs structure are summarized in Table 7 with the LOD and detection time. Further explorations are still needed for the identification/detection of E. coli O157:H7 to improve the detection time and LOD.

Sample Type
Cu phosphate nanocomposites em-  Most recent studies for the E. coli detection based on AMPs structure are summarized in Table 7 with the LOD and detection time. Further explorations are still needed for the identification/detection of E. coli O157:H7 to improve the detection time and LOD.
The accurate detection of pathogenic bacteria is a critical issue for POC devices to ensure better therapy for the corresponding disease and early-stage control. Currently, significant efforts are being directed towards achieving accurate and highly specific detection. To address this, Bai et al. [71] developed a sensor for the precise detection of E. coli O157:H7 using copper phosphate nanocomposites embedded with AMP magainin I and cecropin P1. The highly specific detection of E. coli O157:H7 was evaluated against multiple bacterial samples, with a bacterial density of 10 5 CFU/mL (n = 3). In addition, Qi et al. proposed a novel microfluidic biosensor for the colorimetric detection of foodborne pathogens [80], and Zhang et al. introduced an integrated colorimetric and photothermal lateral flow immunoassay based on bimetallic Ag-Au urchin-like hollow structures for the sensitive detection of E. coli O157:H7 [80,81]. The most recent studies focused on this accurate detection are summarized in Table 8.

Conclusions and Future Challenges
This review provides an overview of recent advancements in the design and development of POC microfluidic-based biosensors for the detection of pathogenic bacteria, especially E. coli O157:H7. These biosensors offer a valuable means of delivering clinically relevant information in a simple, rapid, and cost-effective manner, catering to the demands of POC testing. The LOD and detection time are crucial parameters for detecting these pathogenic bacteria at an early stage of infection. While many of the proposed systems may not match the accuracy of traditional analytical methods, they can still provide sufficient information for routine testing and screening of food and water samples. This review summarizes the detection of E. coli O157:H7 using popular POC methods, such as SPR, RCA, and electrochemical methods. In addition, the impressive role of AMPs in terms of LOD and detection time is evaluated. Developing a simple, fast, and cost-effective method for E. coli detection not only ensures the safety of water and food supplies for consumers but also prevents costly recalls.
AMPs have emerged as a promising alternative to antibodies, and in particular, AMPbased microfluidic platforms for bacteria detection have garnered significant attention due to their simplicity, sensitivity, and high selectivity. In summary, while POC testing still has room for improvement, microfluidic biosensors hold great promise as practical tools for real-world applications. However, there are limitations in terms of LOD, detection time, and specific detection.
Improving the LOD, detection time, and accuracy of microfluidic chips is crucial. Additionally, it is equally important to detect the entire bacteria rather than solely focusing on its DNA or antibodies in order to save time and costs. AMPs present a viable option as they enable rapid and accurate detection of the entire bacteria. However, AMP-based detection faces challenges in terms of repeatability. Therefore, it is necessary to improve repeatability, along with the LOD, detection time, and selectivity, in order to develop efficient and useful biosensors for detecting pathogenic bacteria. Finally, in Table 9, a comparison is provided for widely applied methods used as POC devices for the detection of pathogenic E. coli O157:H7. Both the advantages and disadvantages of each method are summarized in this table.