Microfluidics for foodborne bacteria analysis: Moving toward multiple technologies integration

Food security related to bacterial pathogens has attracted global attention. Accurate, early-stage screening and ultrasensitive detection of foodborne bacteria is crucial to guarantee food safety and reduce the risk of those health threats. Broad-based, multiplexed diagnostic tests of foodborne pathogenic bacteria are urgently needed for the unknown samples. While genome-based approaches have been a promising tool for diagnosing unknown foodborne hazards, tracking the evolution of foodborne hazards, and managing foodborne diseases. They still cannot output all the contaminated information related to the pathogenic bacteria in food samples.¹ Many methods have been developed for detecting suspected or known major contaminated pathogen(s). Among these methods, traditional culture-based methods are not suited for some micro-organisms that are unculturable or difficult to grow in culture. Current detection methods for identifying specific species or strains, including Enzyme-Linked Immunosorbent Assay (ELISA), Polymerase Chain Reaction (PCR), and Mass Spectrometry (MS), lack the means to detect foodborne pathogens in a simple, rapid, high throughput, and high sensitivity way to some extent.² Microfluidics, a promising tool that effectively addresses the limitations of commonly used detection methods, is garnering increasing attention from scientific researchers.

Microfluidics have the advantages of miniaturization, high throughput, simple operation, lower reagent consumption, and cost-effectiveness, which are suitable for designing microfluidic sensors to detect foodborne bacteria. In the microfluidic sensing system, the biomolecules or nanomaterials were used for specifically recognizing the bacterial targets and transferring them into measurable physical/chemical signals. When using microfluidics for foodborne bacteria analysis, it is necessary to simultaneously consider the reaction process for recognition of the target and the record for different detection signals. The signal amplification strategies are also applied to improve detection sensitivity. During this process, the above process should be coupled with a suitable type of microfluidic chip. Here, we briefly summarize the commonly used microfluidic types, recognition strategies, and detection signal amplification technologies in the microfluidic system in the field of foodborne bacteria analysis (Fig. 1), and we proposed the future directions of microfluidics for foodborne bacterial analysis.

Guided by the theories of hydrodynamics, microfluidics is designed to manipulate fluid flow on a miniature scale to achieve complex and versatile operations including mixing, separating, reacting, and detecting, which simplifies the operations and reduces the reagent consumption.³ In addition, microfluidics can convert chemical or biomolecular reactions into measurable physical or chemical signals, enabling the sensing of foodborne bacteria. The additional features that assist the structure of the chip are usually chosen for controlling fluids, including magnetic, electric, and acoustic fields, fluid pumping, and piping systems. Various types of chips including lateral flow strip (LFS),⁴ volumetric bar-chart chips (V-Chip),⁵ Slipchip,⁶ finger-actuated chips,⁷ centrifugal microfluidic chips,⁸ digital microfluidics chip,⁹ droplet microfluidics¹⁰ are proposed for fluid manipulation to analyze bacterial pathogens. Among these chip formats, droplet microfluidics enable high-throughput detection of multiple pathogens, while powerless forms including LFS, V-Chip, slipchip, and finger-actuated chip are cost-effective and better suited for point-of-care (POC) applications. The selection of microfluidic types should also take into account the detection costs in order to promote commercial applications.

Target biorecognition is crucial in the microfluidic analysis of foodborne bacteria, directly affecting the analytical performance of the sensing process. Direct whole bacteria detection minimizes complex/time-consuming sample preparation, attracting widespread attention. Antibodies or aptamers are mostly used as high-quality biorecognizers for target bacteria at the whole bacteria level. Surface imprinted polymers (SIPs) have been developed for the direct detection of whole bacteria, utilizing micro-contact imprinting or positive master stamp mimicking the template bacteria.¹¹ Phages library composed of phages, peptides, antibodies, and nanobodies can be readily modified through genetic engineering or chemical processes, showing a favorable probe for detecting foodborne pathogens.¹² Species-specific targets (DNAs or RNAs) of the common kinds of foodborne bacteria have been mined, and the probe design has shifted toward the target nucleic acid. Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) and CRISPR-associated proteins (CRISPR/Cas) technologies as the third generation of gene editing tools have emerged to recognize target nucleic acids through CRISPR RNAs (crRNAs),¹³ and some amplification-free CRISPR/Cas-based technologies have emerged.¹⁴ Argonaute (Ago), as a next-generation gene editing technology, effectively identifies and cleaves complementary DNA or RNA targets guided by gDNA, showing single-base recognition sensitivity.¹⁴ These gene editing techniques can provide nucleic acid amplification-free detection of multiplexed target DAN of foodborne bacteria. Furthermore, certain types of CRISPR/Cas (such as Cas12a, Cas13a and Cas14) exhibit both cis- and trans-cleavage properties,^15,16 generating various forms of signal readout. Trans cleavage can be employed as an amplifier to amplify signals and improve sensitivity. Among these two types of identification strategies, direct whole bacteria detection avoids the DNA extraction process compared to the detection of the specific target DNAs of bacteria, but lacks detection sensitivity and multiplexed bacterial detection.

Implementing signal amplification in microfluidic sensing processes can further enhance the overall detection capability. The signal amplification techniques are considered during the integration of microfluidics and optical, electrical, and POC detection methods. Background-free detection methods such as chemiluminescence have inherent advantages. When designing POC sensors, specific detection hardware (such as a photomultiplier tube, operational amplifier, and lock-in amplifier) is selected to minimize detection noise and enhance the signal-to-noise ratio. Nucleic acid amplification methods can be applied to detect DNA or RNA targets of foodborne bacteria, indirectly amplifying the detection signals. The combination of isothermal nucleic acid amplification technology with CRISPR can further amplify the signals.¹⁷ The amplification strategies during the chemical or biological reaction process for bacterial detection are the most important. Nanozymes can convert chemical or biological reactions into measurable signals and selectively amplify the signals during this process.¹⁸ Some fluorescent nanomaterials contain nitrogen-vacancy (NV) centers that enable the separation of the signal from a high background under microwaves, improving the sensitivity.¹⁹ During data processing, machine learning technology has been used as a powerful tool for feature extraction and classification to identify weak signals.²⁰ Fully considering these signal amplification techniques coupled with the microfluidic types and target recognition reactions can effectively enhance the sensitivity of the microfluidic system.

In this perspective, we clarified microfluidics for foodborne bacteria analysis based on their functional implementation in flue control. Target biorecognition and implementing signal amplification enhance the detection capability of microfluidics. Microfluidics have a bright future in food safety monitoring. However, there are still some limitations and challenges of microfluidics for foodborne bacteria analysis, which are mainly focused on the following aspects: (1) Not all sensing reactions can be realized on microfluidic chips. The characteristics of some reactions make it impossible to achieve the entire reaction process using microfluidics, resulting in fewer methods developed using microfluidics compared to the traditional ways. (2) There is a lack of multiplex detection of different kinds of bacteria simultaneously. During this, the specificity and mutual cross interference for multi bacteria detection are the most challenging. (3) It is hard to detect unknown food samples directly. Most methods are developed for detecting suspected or known major contaminated bacteria, lacking detection methods for unknown contamination situations.

In the field of foodborne bacteria analysis, microfluidics will be developed to realize more different types of reactions and combine with more new detection methods. Here, we propose the following perspectives of microfluidics for further investigation.

1. High selectivity: High selectivity ensures interference from nontargets during the analysis process and primarily depends on the development of target bacterial recognition probes. The probes will focus on minding species-specific elements of the micro-organisms themselves, while also developing chemical and biological identifiers for these elements. Droplet microfluidics will be a useful tool for probe screening to improve the detection selectivity.

2. High sensitivity: Some bacteria with strong pathogenic ability require detection limits to achieve as low as a single-cell level. High sensitivity has always been an important indicator for the development of microfluidics. The signal amplification technologies will be considered throughout the entire detection process of the microfluidic analysis. The main focus of the research will be on the chemical/biological reaction processes of targets, the selection of detector hardware, and data processing. Another research focus will be on droplet microfluidics, where the domain-limiting effect of droplets can greatly improve detection sensitivity at the single-molecule and single-cell levels.

3. Multiplexed detection: Some food samples are easily contaminated simultaneously by multiple bacteria and by different types of pollutants including heavy metals and toxins. Microfluidic methods for the simultaneous detection of multiplex pathogens or different food contaminants will be promoted, obtaining all contamination information in food samples at one time. In addition, the enrichment and other pre-processing processes are considered on the microfluidic chip, aiming for “sample-in-answer-out” detection.

4. High integration: High selectivity, high sensitivity, and multiplexed detection in the microfluidic sensing system will be achieved through the manipulation of fluids and the integration of various technologies including chemistry, biology, mechanics, optics, electronics, and computers. During the integration, the connection between microfluidics and other technologies should be selected according to the detection methods, mainly using electronic and 3D printing technologies. By integrating advanced technology, microfluidics is evolving toward greater intelligence and miniaturization, promising enhanced specificity, sensitivity, high-throughput capabilities, affordability, and rapid detection of targets.

The authors acknowledge the National Key R&D Program of China (No. 2021YFF0600700) and the Startup Fund for Advanced Talents of Putian University (No. 2024046).

The authors have no conflicts to disclose.

Ethics approval was not required.

Gaowa Xing: Writing – original draft (equal). Jin-Ming Lin: Writing – review & editing (equal).

Data sharing is not applicable to this article as no new data were created or analyzed in this study.