Rapid, Sensitive, and Selective Quantification of Bacillus cereus Spores Using xMAP Technology

Bacillus cereus is a spore-forming ubiquitous bacterium notable as a food poisoning agent. Detection of B. cereus spores using selective media is laborious and non-specific. Herein, the quantitative detection of B. cereus spores was investigated with commercial antibodies and published aptamer sequences. Several detection reagents were screened for affinity to Bacillus collagen-like protein A (BclA), an abundant exosporium glycoprotein. Sensitivity and selectivity toward B. cereus spores were tested using immunoassays and multi-analyte profiling (xMAP). A recombinant antibody developed in llama against BclA protein showed B. cereus spore selectivity and sensitivity between 102 and 105 spores/mL using xMAP. DNA aptamer sequences demonstrated sensitivity from 103 to 107 spores/mL and no cross-reaction to B. megaterium and B. subtilis. Selectivity for B. cereus spores was also demonstrated in a mixture of several diverse microorganisms and within a food sample with no compromise of sensitivity. As proof of concept for multiplexed measurement of human pathogens, B. cereus and three other microorganisms, E. coli, P. aeruginosa, and S. cerevisiae, were simultaneously detected using xMAP. These data support the development of a rapid, sensitive, and selective system for quantitation of B. cereus spores and multiplexed monitoring of human pathogens in complex matrices.


Introduction
The genus Bacillus covers a phenotypically large, heterogeneous collection of Grampositive, spore-forming, aerobic, or facultative anaerobic bacteria. These organisms exhibit a wide range of physiological capacities for the production of enzymes, antibiotics, and metabolites that allow them to persist in diverse natural environments. As such, Bacillus species are frequently used for medical, pharmaceutical, and industrial applications, but they can also be contaminants of plant-based food products [1,2].
The majority of Bacillus species rarely cause disease. However, among pathogenic Bacillus species, several are genetically related and belong to the Bacillus cereus group (BC group). Of these, Bacillus cereus causes food poisoning, Bacillus anthracis is the etiological agent of anthrax, and Bacillus thuringiensis is pathogenic to insect larvae and used as a bioinsecticide. The spores of the BC group are extremely resilient and exhibit resistance to heat, cold, radiation, desiccation, and disinfectants [2].
B. cereus sensu stricto (s.l.) can replicate and produce toxins in a wide variety of foods and is able to form heat-resistant endospores. Its toxins directly cause food poisoning when viable cells/spores grow in improperly stored foods [3]. However, consumption of as little as 10 3 viable cells or spores can also cause food poisoning [4]. Ingested spores (ATCC 13525), Rhodococcus rhodochrous (ATCC 53968), and Pseudomonas aeruginosa PA01 were obtained from the American Type Culture Collection. Yeast strain BY4743 was a kind gift from Dr. Ashkan Golshani (Golshani laboratory/Carleton University) and used for specificity experiments with xMAP technology. Plasmids were maintained in Escherichia coli DH5α, and recombinant proteins were expressed in E. coli BL21 (DE3) (NEB Inc., Ipswich, MA, USA). The BL21 (DE3) was also used in all xMAP experiments with E. coli. The microbial mixture contained B. megaterium, B. subtilis, A. globiformis, P. fluorescens, R. rhodochrous, and E. coli BL21 (DE3) strain. A gene encoding B. cereus spore BclA protein was cloned into a plasmid (pET151/D-TOPO) (Invitrogen, Waltham, MA, USA). This plasmid was constructed for the expression and isolation of recombinant protein with an N-terminal tag holding the V5 epitope and a hexahistidine tag. Tobacco etch virus (TEV) nuclear-inclusion-a endopeptidase was used to remove the amino-terminal histidine tag. The plasmid also contained an ampicillin-resistant gene that was used as a selectable marker for the isolation of transformed E. coli cells.

Microorganism Preparations and Plasmid Miniprep
Bacterial growth was conducted in lysogeny broth (LB) or LB with 2% agar. YP (1% Yeast extract, 2% Peptone) with 2% dextrose as a source of carbohydrates was used as a culture medium for yeast [40]. B. cereus, B. subtitlis, and B. megaterium spores were grown and purified according to Lin et al., 2003 [41] with~90% purity for B. cereus spores and~70% for the spore of other bacilli, as confirmed by microscopy. Bacillus and E. coli cells were grown at 37 • C, and recombinant protein expression from a pET151/D-TOPO plasmid in E. coli was induced before harvest. All plasmid extractions were performed by using the GeneJET plasmid miniprep kit (Thermofisher, Waltham, MA, USA) following the manufacturer's instructions.

Purification of Recombinant BclA Protein
Purification of recombinant BclA protein was performed according to Moteshareie et al., 2022 [42], where N-terminal histidine-tagged recombinant BclA protein carrying the amino-terminal HIS-tag was purified from plasmid-transformed BL21 (DE3) cells using 3 mL HisPur™ Cobalt Spin Columns (Thermo Fisher Scientific, Waltham, MA, USA). E. coli cells were grown in LB at 37 • C to an optical density of 0.4 at 600 nm. After cooling the cultures to 30 • C, recombinant BclA protein expression was induced by 400 µM isopropyl β-d-1-thiogalactopyranoside (IPTG) for 14 h at 30 • C before harvest. Bacteria were sedimented by centrifugation (5000 rpm, 20 min, 4 • C) and re-suspended in binding buffer (40 mM Tris-HCl, pH 7.5, 200 mM NaCl, 10 mM imidazole) supplemented with protease inhibitors and lysozyme (Thermo Fisher Scientific™, Waltham, MA, USA). Bacteria sonicated on ice (2 min with 50% power) in the presence of 25U/mL Pierce Universal Nuclease (Thermo Fisher Scientific™, Waltham, MA, USA). Cell debris was pelleted by centrifugation (15,000 rpm, 20 min, 4 • C). The supernatant was loaded onto a 3 mL column pre-packed with cobalt (Co 2+ ) immobilized metal chelate affinity (IMAC) beads (Thermo Fisher Scientific™, Waltham, MA, USA) and pre-equilibrated with binding buffer. Following washing with 10 mL of binding buffer, proteins were eluted with elution buffer (binding buffer plus 250 mM imidazole). Fractions containing the recombinant protein were confirmed by SDS-PAGE (Supplementary Figure S1), pooled, and dialyzed against storage buffer (25 mM Tris-HCl, pH 7.5, 200 mM NaCl). The terminal histidine tag was removed with 3000 U/mg protein of TEV protease (NEB Inc.) incubated at 4 • C for 1 h. Protein concentrations were measured using a NanoDrop at 280 nm (Thermo Fisher Scientific™, Waltham, MA, USA), and the final purified BclA protein was stored in aliquots at −80 • C.

Immunoblotting
Total proteins were purified in the absence of detergent [43]. Briefly, samples containing spores were pelleted by centrifugation and rinsed with 1× PBS. Total protein concentration was measured using the bicinchoninic acid assay (BCA) using protocols from the manufacturer (Thermo Fisher Scientific, Waltham, MA, USA). Equal amounts of protein were electrophoretically separated in 10% polyacrylamide-SDS gels using a Mini-PROTEAN Tetra cell electrophoresis unit (Bio-Rad ®, Hercules, CA, USA) [40]. The proteins were transferred to 0.45 µm nitrocellulose (Bio-Rad ® , Hercules, CA, USA) with a Trans-Blot Semi-Dry Transfer system (Bio-Rad ® , Hercules, CA, USA). Table 1 shows the tested antibodies. Following transfer, immunoblots were viewed with enhanced chemiluminescence (ECL) substrates (Bio-Rad ® , Hercules, CA, USA) using a Vilber Lourmat gel doc Fusion FX5-XT (Vilber ® , Marne-la-Vallee, France). Densitometry was conducted with FUSION FX software (Vilber ® , Marne-la-Vallee, France), using total protein to normalize measurements. Data represent three separate experiments from three different total protein isolations.

Paratope Blocking Assay
A paratope blocking approach was used to study the accessibility of antigenic target proteins on the surface of B. cereus spores [42]. The premise logic of this assay was that pre-incubating a B. cereus spore pAb with BclA recombinant protein would block the BclA paratope of the pAbs, resulting in a notable reduction in BclA-associated banding on immunoblots. For this, 10 µg of the isolated recombinant BclA was pre-incubated with 10 µM of pAb1 in 1× PBS (pH 7.5) for 24 h at 4 • C. The pre-incubated antibodies were then used in immunoblotting procedures as previously described [42].

Enzyme-Linked Immunosorbent Assay (ELISA)
A sandwich enzyme-linked immunosorbent assay (ELISA) method was conducted to measure B. cereus spores [42]. In summary, maleimide-activated microtiter plates (Thermo Fisher Pierce) were coated with 10 µg/mL capture Abs (mAb2, rAb3, and rAb4) in 200 µL of 1× PBS (pH 7.5) and incubated at 4 • C for 24 h. Wells were washed twice with wash buffer (1× PBS plus 1% Tween) and then blocked at 30 • C for 2 h with blocking buffer containing 200 µL of 5% non-fat dry milk in 1× PBS. Following 3 washes with washing buffer, the wells were incubated with 200 µL of serially diluted spore samples for 2 h at 30 • C. After 3 more washes with washing buffer, the wells were incubated with 10 µg/mL of HRP-conjugated detection Ab (pAb1) in 200 µL of 1× PBS (pH 7.5) for 2 h at 30 • C to detect the presence of spores attached to the capture Abs. Following 3 more washes, each well was incubated with detection reagent 3, 3 , 5, 5 -Tetramethylbenzidine (TMB) (Thermo Fisher Scientific) for 15 min. The reaction was arrested with 100 µL of stopping solution (2 M H 2 SO 4 ) to record the optical density at 450 nm. To adapt this assay for aptamer capture reagents, the aptamers had to be modified by 5 thiolation to facilitate their immobilization in the wells. Wells were coated and incubated with 200 µL/well of each aptamer at 10 µM concentration in 1× PBS (pH 7.5) at 4 • C for 24 h.

xMAP Multiplexing Assay
A multi-analyte profiling (xMAP) technology (Bio-Rad ® , Hercules, CA, USA) was performed according to the Bio-Rad manuals, guides, and protocols. Briefly, magnetic beads were first coupled with specific capture bioreagents (antibody or aptamer), and a sandwich strategy was performed to detect the microorganisms. For this, analytes were captured with a bead-antibody complex, and a biotinylated antibody (Table 1) conjugated with Bio-Plex Streptavidin-PE was used to detect the presence of the analytes.

Magnetic Bead Assembly for Detection by xMAP Technology
Four different Bio-Plex Pro Magnetic COOH Beads (Bio-Rad ® , Hercules, CA, USA) were coupled with specific capture reagents for each organism using a Bio-Plex Amine Coupling Kit (Bio-Rad ® , Hercules, CA, USA) according to the manufacturer's instructions. The coated beads were then treated with either B. cereus at different concentrations (10, 10 2 , 10 3 , 10 4 , 10 5 , and 10 6 cells/mL) alone, in the presence of the five diverse microorganisms (Table 1), or with other pathogens. Biotinylated polyclonal antibodies (Table 1) were used as a secondary antibody for detection. Streptavidin-phycoerythrin (Bio-Rad ® , Hercules, CA, USA) was conjugated to the biotin of the secondary antibody to generate a quantifiable signal.

Statistical Data Analysis
Statistics to analyze data were performed using Graphpad Prism TM (Graphpad Software, San Diego, CA, USA). Differences in the paratope blocking assay were made using a Student's t-test. Differences in treatments were confirmed using one-way analysis of variance (ANOVA) followed by post-hoc analysis using either Tukey's or Dunnett's multiple comparison test. A p < 0.05 was considered statistically different.

Antibody Affinity to B. cereus Spore BclA Protein
To test if the pAb1 could bind spore BclA, a paratope blocking assay was performed ( Figure 1A). For this, pAb1 was pre-incubated with rBclA to block its associated paratope, and then the antibody-rBclA complex was used for Western blotting with the entire spore extract. The results showed a significant reduction in the binding of the antibody-rBclA complex to a major spore protein with the same mobility as rBclA compared to unconjugated pAb1. This suggests that immunogenicity to the spore was largely related to BclA ( Figure 1A). This was also statistically verified by densitometry ( Figure 1B).

Statistical Data Analysis
Statistics to analyze data were performed using Graphpad Prism TM (Graphpad Software, San Diego, CA, USA). Differences in the paratope blocking assay were made using a Student's t-test. Differences in treatments were confirmed using one-way analysis of variance (ANOVA) followed by post-hoc analysis using either Tukey's or Dunnett's multiple comparison test. A p < 0.05 was considered statistically different.

Antibody Affinity to B. cereus Spore BclA Protein
To test if the pAb1 could bind spore BclA, a paratope blocking assay was performed ( Figure 1A). For this, pAb1 was pre-incubated with rBclA to block its associated paratope, and then the antibody-rBclA complex was used for Western blotting with the entire spore extract. The results showed a significant reduction in the binding of the antibody-rBclA complex to a major spore protein with the same mobility as rBclA compared to unconjugated pAb1. This suggests that immunogenicity to the spore was largely related to BclA ( Figure 1A). This was also statistically verified by densitometry ( Figure 1B).  The selectivity of mAb2 was tested with spores from B. cereus, B. subtilis, and B. megaterium. As shown in Figure 1C, mAb2 selectively bound B. cereus spores, but not those from B. subtilis and B. megaterium. The peptide detected using this mAb was equivalent in size to that of rBclA at 72 kDa ( Figure 1C). Furthermore, two commercially available rAbs derived from hamsters and llamas were tested against B. cereus spores ( Figure 1D,E). In this experiment, BclA protein of B. cereus spores was detected using both commercial rAbs (rAb3 and rAb4). In contrast, neither rAbs demonstrated cross-reactivity with any component of B. subtilis or B. megaterium spores ( Figure 1D,E).

Detection Range of B. cereus Spores with rAbs
As both rAbs showed strong selectivity, further experiments to determine their range of detection were conducted by using them as capture reagents in a sandwich ELISA strategy ( Figure 2). As indicated in Figure 2, the rAbs showed a detection range between 10 3 and 1.9 × 10 4 spores/mL for hamster rAb3 and detection between 10 2 and 2.1 × 10 4 spores/mL for llama rAb4. B. subtilis spores were used as a control for selectivity in this experiment. As indicated in Figure 2, neither rAbs showed any significant cross-reaction to B. subtilis spores nor to vegetative cells (data not presented). The selectivity of mAb2 was tested with spores from B. cereus, B. subtilis, and B. megaterium. As shown in Figure 1C, mAb2 selectively bound B. cereus spores, but not those from B. subtilis and B. megaterium. The peptide detected using this mAb was equivalent in size to that of rBclA at 72 kDa ( Figure 1C). Furthermore, two commercially available rAbs derived from hamsters and llamas were tested against B. cereus spores ( Figure 1D,E). In this experiment, BclA protein of B. cereus spores was detected using both commercial rAbs (rAb3 and rAb4). In contrast, neither rAbs demonstrated cross-reactivity with any component of B. subtilis or B. megaterium spores ( Figure 1D,E).

Detection Range of B. cereus Spores with rAbs
As both rAbs showed strong selectivity, further experiments to determine their range of detection were conducted by using them as capture reagents in a sandwich ELISA strategy ( Figure 2). As indicated in Figure 2, the rAbs showed a detection range between 10 3 and 1.9 × 10 4 spores/mL for hamster rAb3 and detection between 10 2 and 2.1 × 10 4 spores/mL for llama rAb4. B. subtilis spores were used as a control for selectivity in this experiment. As indicated in Figure 2, neither rAbs showed any significant cross-reaction to B. subtilis spores nor to vegetative cells (data not presented).   (Table 1) to investigate their potential to be used as capture reagents for the detection of B. cereus spores. An aptamer-antibody sandwich ELISA strategy was developed, in which aptamers were immobilized on ELISA plates to capture B. cereus spores, and a biotinylated commercial Ab against B. cereus spores was used to detect the presence of the captured spores. Figure 3 shows that all aptamers strongly bound to B. cereus spores. Apt2, which was developed against B. anthracis spores, demonstrated the strongest binding affinity for B. cereus spores ( Figure 3B).
Apt2 and Apt3 developed against B. anthracis spores also demonstrated strong binding affinities for B. cereus spores ( Figure 3B,C). Apt2 showed no detectable signal for either B. subtilis or B. megaterium ( Figure 3B). However, minimal reaction to the spores of B. subtilis or B. megaterium was observed, indicating a lower level of selectivity with non-BC group Bacillus spores compared to Apt1 and Apt2 ( Figure 3C).
investigate their potential to be used as capture reagents for the detection of B. cereus spores. An aptamer-antibody sandwich ELISA strategy was developed, in which aptamers were immobilized on ELISA plates to capture B. cereus spores, and a biotinylated commercial Ab against B. cereus spores was used to detect the presence of the captured spores. Figure 3 shows that all aptamers strongly bound to B. cereus spores. Apt2, which was developed against B. anthracis spores, demonstrated the strongest binding affinity for B. cereus spores ( Figure 3B). Apt2 and Apt3 developed against B. anthracis spores also demonstrated strong binding affinities for B. cereus spores ( Figure 3B,C). Apt2 showed no detectable signal for either B. subtilis or B. megaterium ( Figure 3B). However, minimal reaction to the spores of B. subtilis or B. megaterium was observed, indicating a lower level of selectivity with non-BC group Bacillus spores compared to Apt1 and Apt2 ( Figure 3C).

Range of Detection with ELISA
The aptamer-antibody sandwich ELISA method was used to investigate the concentration range of detection of B. cereus spores. Following optimization of parameters involved in this strategy, such as capture aptamer and detection antibody concentrations, Apt1 showed a detection range between 10 3 and 2.25 × 10 5 spores/mL, and Apt2 showed a detection range from 10 4 to 2.5 × 10 5 spores/mL (Figure 4). All three aptamers showed binding affinity to B. cereus spores. This experiment was repeated thrice. Error bars represent standard deviations. Asterisks indicate statistically significant differences with non-BC group spores or controls and were determined with one-way ANOVA followed by Tukey post-hoc analysis (* p < 0.005). Abbreviations: Bc, Bacillus cereus; Bm, Bacillus megaterium; Bs, Bacillus subtilis.

Range of Detection with ELISA
The aptamer-antibody sandwich ELISA method was used to investigate the concentration range of detection of B. cereus spores. Following optimization of parameters involved in this strategy, such as capture aptamer and detection antibody concentrations, Apt1 showed a detection range between 10 3 and 2.25 × 10 5 spores/mL, and Apt2 showed a detection range from 10 4 to 2.5 × 10 5 spores/mL (Figure 4).  Apt3 also detected B. cereus spores at a similar range as Apt1, but Figure 3C shows that it also non-specifically bound reaction to B. subtilis spores (Figure 4). The low crossreaction signal from increasing concentrations of B. subtilis is also illustrated in Figure 5. Apt3 also detected B. cereus spores at a similar range as Apt1, but Figure 3C shows that it also non-specifically bound reaction to B. subtilis spores (Figure 4). The low cross-reaction signal from increasing concentrations of B. subtilis is also illustrated in Figure 5. Apt3 also detected B. cereus spores at a similar range as Apt1, but Figure 3C shows that it also non-specifically bound reaction to B. subtilis spores (Figure 4). The low crossreaction signal from increasing concentrations of B. subtilis is also illustrated in Figure 5.

Antibody-Based Detection of B. cereus Spores
The xMAP technology was paired with llama rAb to detect B. cereus spores. Initial experiments were performed with B. cereus spore lysate in 1× PBS buffer ( Figure 6). Followup experiments were performed in the presence of six other microorganisms to test for cross-reactivity and sensitivity in the presence of a microbial mixture ( Figure 6). Different concentrations of spores at 50, 5 × 10 2 , 5 × 10 3 , and 5 × 10 4 spores/mL were used to test the consistency of the standard curve under these conditions. Results showed a range of detection between 10 2 spores/mL and 10 5 spores/mL with a low standard deviation for the four tested spore concentrations ( Figure 6). As indicated in Figure 6, the standard curves were similar for the detection of B. cereus spores alone compared to B. cereus in the presence of other microorganisms.

Antibody-Based Detection of B. cereus Spores
The xMAP technology was paired with llama rAb to detect B. cereus spores. Initial experiments were performed with B. cereus spore lysate in 1× PBS buffer ( Figure 6). Followup experiments were performed in the presence of six other microorganisms to test for cross-reactivity and sensitivity in the presence of a microbial mixture ( Figure 6). Different concentrations of spores at 50, 5 × 10 2 , 5 × 10 3 , and 5 × 10 4 spores/mL were used to test the consistency of the standard curve under these conditions. Results showed a range of detection between 10 2 spores/mL and 10 5 spores/mL with a low standard deviation for the four tested spore concentrations ( Figure 6). As indicated in Figure 6, the standard curves were similar for the detection of B. cereus spores alone compared to B. cereus in the presence of other microorganisms. Figure 6. Range of detection using xMAP system. Detection of B. cereus (Bc) spore lysate with llama rAb4 in 1× PBS (blue curve), in food sample/5% baby cereal (green curve) and the presence of 6 different microorganisms as a microbial mixture in 1× PBS (red curve). This experiment was performed in triplicate, and no significant difference was observed. The error bars represent standard deviations for tested samples (red dots) at 50, 5 × 10 2 , 5 × 10 3 , and 5 × 10 4 spores/mL.

Aptamer-Based Detection of B. cereus Spores
The possibility of using an aptamer with xMAP technology was tested. Due to its higher sensitivity ( Figure 3B), Apt2 was used for this experiment. This aptamer was developed against B. anthracis spores and successfully detected B. cereus spores with ELISA. Using xMAP technology and pairing Apt2 with the magnetic beads, B. cereus spores were successfully detected in a microbial mixture of six other microorganisms (Figure 7), with a range of detection between 10 3 and 10 7 spore/mL (Figure 7). This experiment was carried out with total spore lysate, and lower performance may be due to the fact that aptamers in this study were developed using the entire spore. Figure 6. Range of detection using xMAP system. Detection of B. cereus (Bc) spore lysate with llama rAb4 in 1× PBS (blue curve), in food sample/5% baby cereal (green curve) and the presence of 6 different microorganisms as a microbial mixture in 1× PBS (red curve). This experiment was performed in triplicate, and no significant difference was observed. The error bars represent standard deviations for tested samples (red dots) at 50, 5 × 10 2 , 5 × 10 3 , and 5 × 10 4 spores/mL.

Aptamer-Based Detection of B. cereus Spores
The possibility of using an aptamer with xMAP technology was tested. Due to its higher sensitivity ( Figure 3B), Apt2 was used for this experiment. This aptamer was developed against B. anthracis spores and successfully detected B. cereus spores with ELISA. Using xMAP technology and pairing Apt2 with the magnetic beads, B. cereus spores were successfully detected in a microbial mixture of six other microorganisms (Figure 7), with a range of detection between 10 3 and 10 7 spore/mL (Figure 7). This experiment was carried out with total spore lysate, and lower performance may be due to the fact that aptamers in this study were developed using the entire spore.

Multi-Analyte Detection of Four Different Microorganisms
Commercial antibodies were paired with different magnetic beads for simultaneous multi-analyte detection of four different microorganisms in the same sample. The antibodies were claimed to be selective to P. aeruginosa, S. cerevisiae, and E. coli (Table 1), but a comprehensive evaluation of specificity was not conducted in this study. All four organisms demonstrated acceptable detection ranges, with minimal cross-reactivity to the microorganisms (Figure 8). The rAb retained its selectivity and sensitivity between 10 2 and 10 5 spores/mL during multi-analyte detection. P. aeruginosa and E. coli showed a range of detection between 10 3 and 10 5 cfu/mL. S. cerevisiae showed sensitive detection between 10 2 and 10 4 cfu/mL. Despite some cross-reactivity, E. coli could be detected between 10 3 and 10 6 cfu/mL (Figure 8).

Figure 7.
Range of detection using xMAP system. Detection of B. cereus (Bc) spores with Apt2 in the presence of 6 different biotechnology microorganisms as a microbial mixture. This experiment was performed in triplicate, and no significant difference was observed.

Multi-Analyte Detection of Four Different Microorganisms
Commercial antibodies were paired with different magnetic beads for simultaneous multi-analyte detection of four different microorganisms in the same sample. The antibodies were claimed to be selective to P. aeruginosa, S. cerevisiae, and E. coli (Table 1), but a comprehensive evaluation of specificity was not conducted in this study. All four organisms demonstrated acceptable detection ranges, with minimal cross-reactivity to the microorganisms ( Figure 8). The rAb retained its selectivity and sensitivity between 10 2 and 10 5 spores/mL during multi-analyte detection. P. aeruginosa and E. coli showed a range of detection between 10 3 and 10 5 cfu/mL. S. cerevisiae showed sensitive detection between 10 2 and 10 4 cfu/mL. Despite some cross-reactivity, E. coli could be detected between 10 3 and 10 6 cfu/mL (Figure 8).

Discussion
A rapid, sensitive, and selective detection system was developed for B. cereus spores, which are significant causes of health and economic impacts. Control measures should be focused on spore detection since they are the environmentally stable form of the bacterium that germinates and produce toxins in food matrices. The xMAP system was demonstrated to be robust and amenable to simultaneous detection of several pathogens without interference from a model food matrix of other diverse microorganisms. Although more experiments with other food poisoning pathogens could further validate the use of xMAP technology in the food sector, a proof of concept for multiplex measurement was demonstrated by detecting three different bacteria and yeast.
Due to the strong antigenicity of BclA protein [42], initial efforts for B. cereus spore detection focused on the BclA protein. This protein is a hairpin-shaped trimeric glycopro- Figure 8. Multi-analyte detection using xMAP system. Antibody-based detection of B. cereus (Bc; rAb4) spore lysate along with P. aeruginosa (Pa; mAb5), S. cerevisiae (Sc; mAb6), and E. coli (Ec; mAb7) cell lysates in the presence of other microorganisms as a microbial mixture. This experiment was performed in triplicate, and no significant difference was observed.

Discussion
A rapid, sensitive, and selective detection system was developed for B. cereus spores, which are significant causes of health and economic impacts. Control measures should be focused on spore detection since they are the environmentally stable form of the bacterium that germinates and produce toxins in food matrices. The xMAP system was demonstrated to be robust and amenable to simultaneous detection of several pathogens without interference from a model food matrix of other diverse microorganisms. Although more experiments with other food poisoning pathogens could further validate the use of xMAP technology in the food sector, a proof of concept for multiplex measurement was demonstrated by detecting three different bacteria and yeast.
Due to the strong antigenicity of BclA protein [42], initial efforts for B. cereus spore detection focused on the BclA protein. This protein is a hairpin-shaped trimeric glycoprotein that is found in abundance on the exosporium layer of BC group spores [44]. Recombinant BclA protein was synthesized and purified to test affinity to a pAb that was developed against B. cereus spores in rabbits ( Figure 1A). The pAb was used to determine the antigenicity and accessibility of BclA protein as a target for spore detection. Several mAbs were demonstrated to provide increased selectivity compared to pAbs (data not shown). As indicated in Figure 1C, the strong relatedness of BC group members was exploited to demonstrate that a mAb (mAb2) developed against B. anthracis spores (Table 1) could also be selective for B. cereus spores. Not unexpectedly, the spore protein recognized by the mAb was identical in migratory molecular weight to that of both spore-associated and purified BclA ( Figure 1C). The mAb could also successfully detect the recombinant purified BclA, which confirmed its strong antigenicity. Although the results showed selectivity for the mAb with strong recognition of BclA protein, the large-scale production cost, poor clonal reproducibility, and problems with long-term maintenance of mAbs make them inferior compared to rAbs [45,46]. Therefore, two specific rAbs developed in hamster and llama (Table 1) against purified BclA protein were tested ( Figure 1D,E). These results indicated that there was no significant cross-reaction with B. subtilis or B. megaterium, supporting BC group specificity of BclA protein and selectivity of rAbs developed against BclA. Both rAbs were tested for their range of detection with a sandwich ELISA technique to determine their sensitivity for subsequent application using the xMAP strategy. The sensitivity of rAbs with ELISA was shown to be as low as 10 3 to 2.1 × 10 4 spores/mL ( Figure 2). It should be noted that although the selectivity of BclA rAbs against B. subtilis and B. megaterium spores demonstrated that they were highly specific for B. cereus spores, investigation with other microorganisms (pathogenic or non-pathogenic) that may be present in food and dairy products should be performed to determine any potential cross-reaction.
The rAbs surely provide an excellent source of capture bioreagent, but initial production may be costly. As an alternative, DNA-based aptamers were tested for their high specificity and affinity to capture BC group spores ( Table 1). The application of aptamers in magnetic bead-based assays and xMAP technology has been previously illustrated in multiple studies [47][48][49]. In this study, three published aptamer sequences (Table 1) were tested for their ability to capture B. cereus spores. As indicated in Figure 3A,B, two of the three aptamers produced no significant signal for B. subtilis and B. megaterium; however, Apt3 showed small cross-reactivity with B. subtilis and B. megaterium ( Figures 3C and 5). All three aptamers demonstrated a relatively wide range of detection between 10 3 and 2.25 × 10 5 spores/mL (Figure 4). The application of Apt1 for the detection of B. thuringiensis at 10 3 spores/mL has recently been reported with a digital photocorrosion biosensor [42].
The xMAP technology is applicable for multiplex detection of different analytes within a single complex sample. xMAP multiplex assays are currently available in various nucleic acid and immunoassay formats, enabling simultaneous detection and typing of pathogenic viruses, bacteria, parasites, and fungi and also antigen or antibody interception [32,50,51]. As an open architecture platform in the xMAP technology, magnetic microsphere sets that are internally labelled with two spectrally different fluorophores can be paired with selective and sensitive capture molecules such as antibodies [32] and aptamers [49,50] for the detection of multiple foodborne pathogens, which is significantly beneficial for food and dairy industries. This technology could dramatically increase the flexibility of pathogen detection in food products [32,[50][51][52].
In this study, BclA llama rAb was paired with Bio-Plex Pro Magnetic COOH Beads and then tested for their ability to capture B. cereus spores in the absence and presence of other microorganisms (i.e., pathogens or microorganisms used in biotechnology) ( Figure 6). When detection of B. cereus was performed in baby cereal as a model complex food matrix, no interference was observed ( Figure 6). After optimization of all parameters involved in a sandwich immunoassay with xMAP technology using rAb, the range of detection was improved compared to ELISA to~10 2 spores/mL. The results demonstrated both sensitivity and selectivity with xMAP when paired with rAb to detect B. cereus spores in both pristine conditions with PBS buffer and in the presence of other microorganisms ( Figure 6). These results showed negligible interference from PBS alone, food sample, or other microorganisms ( Figure 6). In addition, this suggests high selectivity in the microbial mixture for llama rAb developed using BclA protein.
The possibility of using an aptamer conjugated to magnetic beads was tested for the detection of B. cereus spores in a microbial mixture. Using this strategy, B. cereus spores were successfully detected in the presence of other microorganisms, and sensitivity was improved by~10-fold following optimization (Figure 7). The inconsequential presence of other microorganisms and low background signal demonstrated that Apt2 was highly selective (Figure 7). Unfortunately, due to the limited availability of known selective aptamer sequences, not all microorganisms may be analyzed by multiplex analysis. The availability of aptamer with strong affinity and selectivity is an obvious limitation for their widespread use as capture reagents for all pathogens, but the possibility of multiplex sensing using a combination of Abs and aptamers can be explored.
Although both Apt2 and llama rAb showed a wide range of detection for B. cereus spore, the rAb demonstrated more sensitivity and a lower range of detection at 10 2 spores/mL versus 10 3 spores/mL for Apt2. However, Apt2 was able to detect~100-times more spores compared to rAbs (Figures 6 and 7). It is possible that the smaller size of aptamers permitted greater immobilization on the spore surface, leading to less stearic interference, higher signal, and more sensitivity. Since 10 3 spores/g or spores/mL of food or dairy products is sufficient for infection in humans, the lower detection limit of 10 2 spores/mL is in the appropriate range to serve as a screen for contaminated foods. In case higher sensitivity is required, samples can be filtered and further condensed [53,54]. Nevertheless, improved sensitivity compared to ELISA was observed, which may be due to a combination of instrumentation interpolations for the emission signal, and better accessibility of the suspended capture reagent-coupled microspheres compared to immobilized capture reagents on the ELISA plate surface.
Using xMAP technology with bimolecular quantitative detection of B. cereus spores is a novel concept demonstrating attractive sensitivity for analysis of starchy foods and dairy samples. Other sensitive methods have been developed, such as quantitative real-time PCR and biosensing devices. These systems may also quantify B. cereus at 10 2 bacteria or spores/mL, comparable to the xMAP strategy; however, the latter has the superior advantage of multiplexing [55][56][57][58]. The xMAP technology has been successfully developed for multiplex detection and quantification of foodborne pathogens with an adequate range of sensitivity based on the infectious dose of the pathogen following human consumption [50][51][52].
As a proof of concept, three diverse organisms (P. aeruginosa, S. cerevisiae, and E. coli) were selected based on the availability of commercial monoclonal-specific antibodies. Following antibody-magnetic bead coupling, the sensitivity of each was tested using xMAP technology (Figure 8). Standard curves for each of the three other microorganisms were achieved with detection ranges from 10 3 to 10 5 cells/mL for P. aeruginosa and S. cerevisiae, whereas E. coli showed sensitivity up to 10 6 cells/mL (Figure 8).

Concluding Remarks
The ability of several commercially available antibodies and published DNA aptamer sequences to detect B. cereus spores was investigated. Based on the results of immunoassays (ELISA and Western blotting), rAbs generated in hamsters and llamas and two DNA aptamers have been demonstrated to be specific and sensitive for B. cereus spore detection. Compared to ELISA, the xMAP approach improved the detection range and could be used for the detection of B. cereus spores alone in a food sample and in the presence of a mixture of diverse organisms. Most importantly, xMAP technology showed utility as a multiplexing tool that can be developed for selective detection of pathogens in the food and dairy industry. This technique was also successfully demonstrated with aptamers. Collectively, these results indicate that the benefits of using aptamer-based detection can also be extended to other capture reagent-based detection methods such as biosensing devices.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/microorganisms10071408/s1, Figure S1: Recombinant BclA protein, purified from E. coli and tested by SDS-PAGE and Coomassie Blue staining.