Lectins have multivalent and non-covalent interactions with glycoproteins, which are important in distinguishing bacterial pathogens from contaminated food. This feature of lectins has been exploited to isolate gut bacteria by conjugating them with magnetic beads (Gorakshakar and Ghosh, 2016). Lectins have been immobilized on magnetic beads by adsorption, covalent immobilization on tosylactivated beads or biotinylated lectins were coupled to magnetic particles with immobilized streptavidin. Magnetic particles can be separated later using a free competing sugar (Safarik and Safarikova, 1999). A battery of pathogenic and commensal bacteria was tested using lectins extracted from plants such as Canavalia ensiformis, Datura stramonium, Vicia faba, Solanum tuberosum, Helix aspersa, Cytisus scoparius, Erythrina cristagall, Bauhnia purpurea, Limulus polyphemus, Caragana arborescens. Lycopersicon Con A lectin, for example, has a higher ability to bind with E. coli and S. aureus at rates of 0–98% and 50–100%, respectively (Cheung et al., 2015; Malik et al., 2016). Hyun et al. (2020) devised a method for selectively eliminating harmful bacteria by using bacterial lectin-targeting glycoconjugates containing an epitope or a photosensitizer to enhance antibody-dependent cellular cytotoxicity (ADCC) or photodynamic treatment (PDT). Furthermore, Yang et al. (2019a) employed wheat germ agglutinin lectin (WGA) as a bio-recognition agent for S. aureus enrichment and found that even at 100–1000 CFU/ml of S. aureus, WGA demonstrated high enrichment efficiency. Later, this unique enrichment was applied in a quantitative and selective detection technique to detect S. aureus with a detection limit of 3.5 102 CFU/ml using double stranded DNA (dsDNA) stabilized gold nanoparticles (AuNPs). To improve the efficacy of magnetic nanoparticles in investigating the scope of bacterial pathogen detection based on bio-recognition, a lectin magnetic separation (LMS) technique was developed for S. aureus (Yang et al., 2019b) and the main aim was to improve the efficiency of magnetic nanoparticles and to expand the scope of bacterial recognition. In conventional method, i.e., using Baird–Parker (B-P) method, S. aureus can be detected at limit of 3 × 10⁰ CFU·mL⁻¹ within 15 h; the polymerase chain reaction (PCR) method can be finished within 4 h, with the limit of detection (LOD) of 3 × 10² CFU·mL⁻¹. The LOD of HRP-pig IgG-based colorimetric method is 3 × 10⁵ CFU·mL⁻¹, and the method only lasts for 2 h. If combined with specific detection methods, it meets different needs for rapid detection of S. aureus.

Listeria monocytogenes is categorized into four serological groups and 13 serovars based on somatic and flagellar antigen components (Lopez Valladares, 2019). These are Serogroup 1 (1/2a and 1/2b serovars), serogroup 2 (serovar 1/2c), serogroup 3 (serovars 3a, 3b, and 3c), and serogroup 4 (serovars 4a, 4 ab, 4b, 4c, 4) (Li et al., 2020). Nonetheless, epidemiologic studies benefit little from serotyping because only three serovars, 1/2a, 1/2b, and 4b account for over 90% of all clinical isolates (Braga et al., 2017). Changes in glycosidic substituents coupled to a polyribitol phosphate (Rbop) backbone cause the most variation in wall teichoic acids (WTA) (Brown et al., 2013). WTAs in L. monocytogenes are made up of repetitive Rbop units with variations in carbohydrate molecules such as galactose, rhamnose, glucose, and N-acetyl glucosamine, which vary a lot according to the strain and serovar. On the C-2 and C-4 sites of the thiol group, serovars 12 and 3 have rhamnose sugars and GlcNAc. Serovars 4, 5, and 6 have a more complex WTA assembly in the centre of the GlcNAc chain, which is fused as a chunk of the polyribitol chain and may contain Glc and/or Gal substituents. Listeria WTA does not have any D-alanine substituents (Bielmann et al., 2015). Through a GlcNAc moiety, wheat germ agglutinin (WGA) binds to WTAs (Murakami et al., 2014). Flannery (2019) used an agglutination assay and a florescence-based assay to investigate cell wall diversity of Listeria spp., Canavalia ensiformis, Griffonia simpliciforlia lectin I, Helix pomatia agglutinin, Ficinus cummunis agglutinin, and Ricinus communis agglutinin in clinical, food, and environmental isolates.

Sun et al. (2017) isolated and separated enteric bacteria using lectin-labeled magnetic beads. They utilised sixteen different types of lectins and found that magnetic beads labelled with Con A produced the most agglutinations Figure 2. From river water and laboratory cultures, this approach could extract pure and culturable cells. The findings are equivalent to those obtained using antibody-labeled beads. The findings are encouraging, implying that lectins could be employed as an alternate cell discriminator. Sun et al. (2017) separated coliform from fresh water and effluent from a water treatment plant using lectin-labeled magnetic beads. Direct magnetic capture was shown to be less reliable than indirect magnetic capture. Harito et al. (2017) employed the lectin magnetic separation (LMS) approach to isolate Toxoplasma gondii oocysts from water sample concentrates, with microscopy or molecular methods used for identification. Castillo-Torres et al. (2019) used lectin-functionalized and aptamer-functionalized microdiscs to demonstrate selective separation of generic coliforms and E. coli cells, respectively. They also discovered a direct association between bacterial concentration and low signal in control samples in the form of fluorescence signals, which are directly correlated to bacterial concentrations and visible in control samples (negative samples). Using Con A–linked magnetic beads to detect Hepatitis A Virus (HAV), Ko et al. (2018) discovered that it had a significantly higher binding affinity for HAV than reverse transcriptase PCR, with a detection concentration of 10⁻⁴ of the virus stock. Magnetic beads containing fragment crystalizable-mannose-binding lectin (FcMBL) were used to capture and magnetically remove bacteria from purified cultures of 12 S. aureus strains, as well as from 8 articular fluid samples and 4 synovial tissue samples collected from patients with osteoarthritis or periprosthetic infections at 85% efficiency (Bicart-See et al., 2016). Tateno et al. (2018) employed lectin to bind malignant cells based on cell aggregation, which resulted in cell death, according to a US patent.

Hendrickson et al. (2019) used a ligand–bioreceptor interaction to develop a flow cytometer analysis of the bacterial pathogens E. coli and S. aureus. Plant lectins with fluorescent labels were used as natural receptors that could precisely bind with bacteria’s cell wall polysaccharides. Additionally, epifluorescence microscopy was employed to validate and observe lectin–carbohydrate interactions. The binding specificity of plant lectins to E. coli and S. aureus cells was investigated, and wheat germ agglutinin was chosen as a receptor since it provided high-affinity interactions. Bacterial pathogens can be detected up to 10⁶ cells/mL in 5 min using this approach. Lectin reagents are a promising method for controlling a wide spectrum of bacteria due to their accessibility and universality.

The inclusion of 4 tryptophan, 12 cysteine, and glycine residues in the amino acid sequence of wheat germ agglutinin (WGA) may explain its natural florescent ability (Bogoeva et al., 2004). WGA has a characteristic tryptophan emission at 348 nm in its fluorescence spectra. The results of the fluorescence study led to the conclusion that fluorescence and phosphorescence are caused by at least two of the three tryptophan residues in WGA (Liener, 2012). With increased amounts of log cfu/ml of L. monocytogenes cells, the binding constant of WGA was increased. The affinity of carbohydrate binding domains (CBDs) to their carbohydrate ligands is equivalent to affinity-matured antibodies to bacterial cell surface antigens (Bi and Zimmer, 2020). Another study by Loessner et al. (2002) found that at increasing carbohydrate concentrations, CBDs have a higher association constant with their carbohydrate ligands. Another study found that E. coli O157:H7 that binds to lectins had a higher association constant (Ka) (Wang et al., 2013). In the presence of particular sugar and L. monocytogenes, Raghu (2018) detected maximum fluorescence in the spectra using 290 nm as excitation and 350 nm as emission wavelength with increasing WGA concentration. Therefore, the principle of WGA fluorescence could also be a technique for developing fluorometric tools for detecting L. monocytogenes.

Using platforms such as electrical oscillation (Bhalla et al., 2016), piezoelectric crystal, and micro-calorimetrics, several modules of lectin-based biosensors have been constructed (Vereshchagina et al., 2015). Table 2 shows an overview of biosensors created for the detection of infections employing lectins. Biosensors are seen as a promising alternative to the traditional procedures that have been used in the past. The biorecognition element must be fixed on the sensor surface to provide high selectivity. Lectins are a type of biorecognition element that is more stable and covers greater surface area than other biomolecules like antibodies (Sondhi et al., 2020). The sections that follow will go over the many types of transducers that can be utilized to detect bacterial infections using lectins.

Piezoelectric biosensors are analytical devices that work on the idea of analyte and ligand molecule affinity, which causes a change in mass/weight on vibrating quartz crystals, changing the frequency of oscillation. The change in oscillation frequency is a measure of the analyte concentration in the samples (Skladal, 2016), and Figure 3 is a graphical depiction of a piezoelectric lectin-based sensor. A weight indicator made from a vibrating quartz crystal is extremely sensitive and accurate. Mass changes can be detected utilizing a crystal-controlled oscillator and the piezoelectric biosensor’s resonance frequency (Petr, 2015). Serra et al. (2008) suggested a piezoelectric biosensor for the detection and quantification of bacteria using an electrochemical quartz crystal microbalance (EQCM). Con A and E. coli were used to test the efficacy of lectin-based quartz crystal microbalance. Shen et al. (2007) employed a non-labeled quartz crystal microbalance (QCM) biosensor to detect bacterial infections using lectins as bio-receptors. The bio-recognition of O-antigen using a carbohydrate non-labeled mass sensor packed with lectin could be a promising technique with a reduced detection limit and maximum selectivity for high molecular weight bacterial targets. With the use of Con A lectin as the bio-recognition molecule, Shen et al. (2007) developed a QCM biosensor. It did, however, reveal a low detection limit for E. coli (limited to a few hundred bacterial cells). These proteins were employed by Serra et al. (2008) to make biosensors for detecting bacterial infections. Using an EQCM, gold-plated quartz crystals were immobilized with lectins. The Con A was used to test the efficiency of the lectin immobilization approach and the biosensor performance for E. coli detection. Adsorption of polarized and non-polarized (0.200 V) gold-empaneled quartz crystals through avidin–biotin binding was used to ensure attachment of Con A. By measuring the change in frequency with E. coli concentration 1 h after bacterial binding, a linear calibration plot in the range of 5.0 × 10⁶ to 2.0 × 10⁷ cfu/mL was obtained, with the limit of detection of this developed sensor at around 1.0 × 10⁴ cfu/mL.

Using four different carbohydrate binding lectins, namely WGA, MAL, LCA, and Con A, a QCM biosensor was applied to distinguish seven Campylobacter jejuni strains (Yakovleva et al., 2011). The frequency change in the Con A lectin was higher at 34 Hz (Wang et al., 2013). The use of a carbohydrate epitope in conjunction with a lectin-bacterial O-antigen interaction in the development of sensitive and specific non-labeled biosensors for the detection of bacterial pathogens has been demonstrated using this patented technology for “Microorganism Detection and Analysis Using Carbohydrate and Lectin Recognition” (Zeng and Shen, 2016). Many other exo- or endogenous toxins produced by or present in bacterial cells may detect the carbohydrate ligand and generate increased cross-reactivity, leading to false positive and negative results.

Liedberg was the first to propose the SPR sensor for use in biosensors (Prabowo et al., 2018). It has now been thoroughly investigated and is gradually evolving into an inherent label-free device for facilitating chemical interactions between the target analyte and ligand molecules. When a photon of light strikes a metal surface (gold or silver), it interacts with and excites the metal’s electrons, causing them to move as a single electrical unit called a Plasmon, which oscillates parallel to the metal surface (Mitchell and Wu, 2010). Surface Plasmon is a phenomenon in which a Plasmon oscillates and generates an electric field that extends about 300 nm out from the metal surface (Homola, 2003; Hoa et al., 2007). It is based on the fact that at a specific incidence angle and wavelength, electrons in a metal layer become excited and absorb light. Surface Plasmons (SPs) are electron density waves that develop at the interface of a metal (such as gold) and a dielectric (such as glass). Changes in the refractive index caused by an analyte binding to a prism’s metal-coated surface characterize these waves (Mullett et al., 2000; Homola, 2003). A resonance occurs at a specific angle where the energy shift from the photon to the Plasmon is well matched, resulting in significantly increased energy shift and a commensurate decrease in reflectance (Mitchell and Wu, 2010).

If mass is attached to the surface within this 300 nm region, it perturbs the Plasmon and alters the resonance angle for a photon of fixed wavelength. If the metal surface’s plane is fixed, the resonance wavelength can be determined by scanning a range of photon incidence angles and determining the resonance angle (Zeng et al., 2017). An SPR biosensor operates on the same concept as a surface Plasmon resonance sensor, measuring minute changes in refractive index that correspond to binding interactions on the sensor surface (Homola, 2008). In most cases, SPR biosensors can achieve resolutions of 1 resonance unit (RU), or 1 × 10⁻⁶ refractive index units (Prabowo et al., 2018). SPR biosensors can detect binding interactions between molecules as small as 2 kDa, but smaller molecules produce insufficient changes in bound mass, which cannot be monitored directly (Perumal and Hashim, 2014).

Only a few reports on how lectins are employed as a bio-receptor for bacterial detection are available to date. Gram-negative bacteria, such as E. coli, can be distinguished by their chemically distinct surface LPS, which can be detected by specific lectins; hence, bio-receptors assist ConA attach to E. coli W1485. Masarova et al. (2004) created a SPR detection system for endotoxins or entire bacterial cells by covalently attaching eight distinct lectins to CM 5 chip surfaces (CM5). These eight lectins were found to be unique. Based on lectin-endotoxin (lipopolysaccharide) interaction, these eight lectins were able to distinguish between the microorganisms. Wang et al. (2013) employed SPR to create a novel type of biosensor for detecting E. coli O157:H7, which utilized lectin as a bio-recognition molecule. Using amine coupling chemistry, lectin was immobilized on the gold chip (CM5) surfaces to capture this bacterial pathogen in the buffer solution. WGA, ConA, UEA, PNA, and MAL were evaluated for the best lectin, one that could efficiently bind to this bacterial pathogen. Among the lectins that were screened, WGA lectin has a wide linear range of 3 × 10⁵ to 3 × 10⁸ cfu/mL with a lower detection limit of 3 × 10³ cfu/mL, demonstrating good selectivity by utilizing SPR. SPR has advantages over QCM and electrochemical measurements, according to Raghu and Kumar (2020). This is because biomolecular interactions can be tracked in real time, allowing kinetic parameters to be determined (KD and KA). The affinity constant KA, in particular, aids in performance analysis of the bacteria and carbohydrate binding protein in terms of specificity and utility. In contrast, there are few publications on lectin based SPR sensors. Wang et al. (2013) and Raghu and Kumar (2020) developed a novel SPR biosensor using lectin as the bio-receptor for fast detection of E. coli O157:H7 and L. monocytogenes. Figure 4 depicts the sensor configuration for the detection of L. monocytogenes. Lectins were immobilised on the CM5 chip with the help of self-assembly monolayers to capture E. coli O157:H7 in the buffer solution (SAM). In another investigation, Leonard et al. (2004) used a Biacore SPR 3000 biosensor with a detection limit of 1105 L. monocytogenes cells/ml within 30 min to detect L. monocytogenes with a polyclonal antibody.

An electrochemical sensor is a device that turns biological events into a signal that can be measured analytically, such as the amount of a certain analyte in a sample or composition, harmful contaminants, or bacterial infections (Thevenot et al., 2001). A distinguishing feature of an electrochemical biosensor is the electrochemical technique used. By using lectins as a bio-recognition molecule, Ertl and Mikkelsen (2001) produced an electrochemical biosensor array, and transduction is assessed by taking the respiratory cycle activity, in which the native respiratory chain of the microbe is interrupted by exogenous oxidants. They discovered a link between agglutination outcomes and the electrochemical assessment of lectin cell binding. To determine how each organism reacts to lectin, chronocoulometric measures were created. Bacillus cereus, S. aureus, Proteus vulgaris, E. coli, Enterobacter aerogenes, and Saccharomyces cerevisiae can all be distinguished using the established method. Ertl et al. (2003) also created an electrochemical biosensor with a screen-printed array whose selectivity was based on lectin affinity for cell-surface lipopolysaccharides. These arrays were used in conjunction with factor analysis to quickly identify four E. coli subspecies (E. coli B, E. coli Neotype, E. coli JM105, and E. coli HB101) in under 40 min (Velusamy et al., 2010). Sa et al. (2020) used Con A and WGA to construct an impedimetric biosensor for detecting pathogenic Candida species. The sensor can distinguish between Candida species with a detection sensitivity of 10²–10⁶ CFU/ml. Lopez-Tellez et al. (2020) established a sensitive electrochemical detection approach for Salmonella spp. employing Hechtia argentea lectin immobilized on a screen-printed gold electrode that exhibited selectivity for D-mannose, which is found on the lipopolysaccharide (LPS) layer of Salmonella spp. With increased concentrations of Salmonella spp., the interaction of bacteria with immobilized lectin on a biosensor surface increased the impedance, achieving a detection range of 15–2.57 × 10⁷ CFU/mL, with a 5 CFU/ml limit of detection.

Lectin biosensors have a limited ability to work in a multiplexed fashion, which is their principal drawback. Only SPR imaging (SPRi) and, to a lesser extent, micro-cantilever-based assays allow for highly parallel analysis. More research on electrochemical approaches that work in an array arrangement is required. It is expected that lectin biosensors will be better suited for diagnostic purposes than for preliminary assays of a large number of samples. There is also the additional problem of lectin biosensors and their use in the examination of real complex materials (Belicky et al., 2016). However, controlled immobilization of lectin for the construction of lectin biosensors is more difficult due to the larger complexity of a protein molecule.

Carbohydrate-protein interactions have been identified as critical for a variety of biological activities, including extracellular and intracellular signaling, as well as cell-to-cell recognition. Lectins are useful tools for studying glycoconjugates and understanding the mechanics of numerous biological processes because of their carbohydrate binding selectivity. Lectins are useful tools in glycobiology and are the greatest model system for studying protein-carbohydrate interactions. Lectins have been linked to the creation of lectin-based high-throughput molecular tools and have established themselves as a major glycan decoding tool in glycobiology. However, there is still a lot of room for lectins to be used in numerous biotechnological and medicinal applications. Furthermore, the identification and characterization of novel lectins, as well as in-depth research on existing lectins, will have a significant impact on future research.