Article | Open

B cells Using Calcium Signaling for Specific and Rapid Detection of Escherichia coli O157:H7

  • Scientific Reports 5, Article number: 10598 (2015)
  • doi:10.1038/srep10598
  • Download Citation
Published online:


A rapid and sensitive detection technology is highly desirable for specific detection of E. coli O157:H7, one of the leading bacterial pathogens causing foodborne illness. In this study, we reported the rapid detection of E. coli O157:H7 by using calcium signaling of the B cell upon cellular membrane anchors anti-E. coli O157:H7 IgM. The binding of E. coli O157:H7 to the IgM on B cell surface activates the B cell receptor (BCR)-induced Ca2+ signaling pathway and results in the release of Ca2+ within seconds. The elevated intracellular Ca2+ triggers Fura-2, a fluorescent Ca2+ indicator, for reporting the presence of pathogens. The Fura-2 is transferred to B cells before detection. The study demonstrated that the developed B cell based biosensor was able to specifically detect E. coli O157:H7 at the low concentration within 10 min in pure culture samples. Finally, the B cell based biosensor was used for the detection of E. coli O157:H7 in ground beef samples. With its short detection time and high sensitivity at the low concentration of the target bacteria, this B cell biosensor shows promise in future application of the high throughput and rapid food detection, biosafety and environmental monitoring.


There are estimated 48 million cases of foodborne illness resulting in 3,000 deaths, and an estimated cost of 78 billion dollars per year1,2,3. It has been continuously a major public health and economic burden for the United States and worldwide4. In particular, foodborne bacteria such as Campylobacter jejuni, Listeria monocytogenes, Salmonella enterica, E. coli O157:H7 and other shiga-toxin producing E. coli strains (non-O157 STEC), and Vibrio spp. are leading causes of foodborne diseases5. E. coli O157:H7 has been identified as a major etiologic agent, which is one cause of foodborne illness and has been found to contaminate spinach, lettuce, cider, ground beef, and cantaloupe, which is also one of the top six pathogens contributing to domestically acquired foodborne illnesses resulting in hospitalization (38%)6. Therefore, rapid detection of microbial pathogens in food is the solution to the prevention and recognition of problems related to health and safety.

Cell-based biosensors (CBBs) present promises of equally reliable results in much shorter times7. The sensing elements of CBBs could be vegetative cells of bacteria8,9, eukaryotic10 or mammalian cells11,12,13,14,15. The ability of cells to recognize and respond to stimuli has made them attractive for incorporating them into biosensors. Besides neurons, cardiac cells, adenocarcinoma cell line, more and more lymphocytes have been studied in CBBs, such as B cells, mast cells, T cells, etc16,17,18,19,20,21,22,23,24. Among them, B cells showed the superiority of pathogen detection in speed and sensitivity due to their capability of antigen internalization through BCR (B Cell Receptors) for processing and presentation to T cells25. Murine hybridoma B cells (Ped-2E9) have been used for rapid detection of pathogenic Listeria, the toxin listeriolysin O and the enterotoxin from Bacillus species13,25,26. Rider et al.11 used the genetically engineered B lymphocyte cells for a rapid identification of pathogens and could detect E. coli O157:H7 as little as 500 CFU/g in 5 min or less11. B cells are known as the fastest pathogen identifiers (intrinsic response in <1 s)27,28, since calcium ion plays a pivotal role in the regulation of various cellular processes in eukaryotic cells. One of the primary consequences of this identification in a B cell is that a molecule binding to BCR induces a change of Ca2+ flux, which is a critical event in the response of a B cell to antigen stimulation29.

A series of fluorescent calcium indicator dyes have been developed for measurement of free intracellular calcium in eukaryotic cells and prokaryote. Fura-2 has been known as an indicator dye for measuring the concentration of free calcium ([Ca2+]i) within living cells30,31. The ratio of fluorescence emission at the two excitation wave lengths (340:380) is considered a reliable indicator of [Ca2+]i32,33. It has been widely used but not limited in immunology, cytology and neurology for interrogating ion channels34,35,36. and calcium signaling37,38,39. However, to the best of our knowledge, there is no report on utilization of Fura-2 in B cells for the bacterial pathogen detection in the food sample.

In this study, we combined the properties of B cells and Fura-2 to develop a B cell biosensor with a low detection limit of E. coli O157:H7 and short detection time. The innovative approach in this study is the use of a Ca2+-indicator, Fura-2 for detecting BCR-induced Ca2+ change due to an interaction between B cells and pathogens. The reaction mechanisms underlying intracellular calcium measurement using an Ca2+-indicator Fura-2 are discussed elsewhere31. It has been long known that the excitation spectrum of Fura-2 shifts rightward upon the binding of Ca2+. The excitation wavelength of Fura-2 shifts from 340 nm in the presence of Ca2+ to 380 nm in the absence of Ca2+. It was found that a fluorescence ratio at 340/380 is correlated to a concentration of free intracellular Ca2+30.This technology has a great potential to provide a practical alternative for detection of E. coli O157:H7 and other pathogens.


Principles of the B cell biosensor

Combining advantages of the sensitivity of Fura-2 to Ca2+ and the rapidness of response of B cell to antigens, a Ca2+-indicator based B cell biosensor was designed for E. coli O157:H7 detection. As shown in Fig. 1a, Fura-2 loaded B cells were placed in 96-wells, and a fluorescence ratio (340:380) was measured after the addition of analytes. Fura-2 pentaacetoxymethyl (AM) ester was used in our experiments because this form of dye is Ca2+ insensitive and nonpolar. Once inside the cell, esterase enzymes sequentially cleave the AM groups to leave Fura-2-free acid (Ca2+ sensitive, polar) trapped inside the cell and are ready for Ca2+ binding (Fig. 1b).

Figure 1
Figure 1

Principles of the B cell biosensor. (a). The response signal of the B cell biosensor was measured by a microplate reader. (b). Structural changes of Fura-2 by esterase activity and Ca2+ binding. Fura-2 AM ester is Ca2+ insensitive and nonpolar. Once inside the cell, esterase enzymes sequentially cleave the AM groups to leave Fura-2-free acid (Ca2+ sensitive, polar) trapped inside the cell, where it is able to bind Ca2+. (c). The cellular working principle of the B cell biosensor. When the target pathogen is attached to its specific receptors on the B cell surface, the cross-linking of B cell receptors (BCRs) will produce a signal, and the signaling pathways will be activated, resulting in the release of Ca2+ within seconds. Fura-2 exhibits a calcium dependent excitation spectral shift to report the 340/380 ratio and indicate the presence of the target pathogen.

The specific B lymphocyte plays a major role in the recognition and capture of E. coli O157: H7. The BCR-induced Ca2+ flux occurs in seconds, and then B cells will exhibit increased BCR-evoked Ca2+ influx29. As shown in Fig. 1c, the B lymphocyte immune response is initiated by the binding of an antigen to the BCR25. The BCR is a multisubunit protein complex composed of a membrane form of immunoglobulin (Ig) that is noncovalently associated with heterodimers of Igα and Igβ. Crosslinking the BCR by the binding of bivalent or multivalent antigen leads to both transmembrane signaling and antigen internalization for presentation40. BCR-induced tyrosine phosphorylation of PLC-γ2 is responsible for an increase in its activity, allowing the conversion of phosphatidylinositol 4,5-bisphosphate (PIP2) to the second messengers inositol 1,4,5-trisphosphate (IP3) and diacylglycerol (DAG). IP3 binds IP3 receptor (IP3R), which is localized primarily on the endoplasmic reticulum (ER) and stimulates the release of calcium from intracellular stores41. When ER Ca2+ depletion occurs, Orai, a plasma membrane protein and a pore subunit of the calcium-release activated channel (CRAC), triggers CRAC activation. Thus, localized Ca2+ influx is activated following Ca2+ depletion of ER Ca2+ stores29,42,43. IP3 mediates a transient calcium release from intracellular stores, which leads to a sustained influx of calcium through Ca2+ channels in the plasma membrane, a process termed store-operated calcium entry44,45,46.

Optimization of the B cell biosensor

The experimental conditions such as B cell concentration, measurement time and existence of Ca2+ were interrogated to optimize the sensor. MARC 29F8 with membrane has monoclonal antibodies (MAbs) specific to the LPS of E. coli O157:H747. Upon arriving, the B cells were tested for the presence of antibodies through ELISA (Enzyme-linked immunosorbent assay). Whole cell ELISA analysis procedure is given in the Supplementary Information. ELISA results (Supplementary Fig. S1) confirmed the activity of antibodies on the membrane of selected B cells and the antibodies still have activity even after 12 passages of B cells. In this study, the early-passage numbers (passage numbers between 5 and 12) of B cells were used for B cell biosensor development. The binding of E. coli O157:H7 to the B cells was further confirmed through scanning electron microscopy (SEM). Representative SEM micrographic images are shown in Fig. 2b.

Figure 2
Figure 2

Optimization of the B cell biosensor. (a). SEM image of a B cell in the absence of any bacteria (×10000). (b). SEM image of a B cell with E.coli O157:H7 cells (×10000). (c). Real time response profile of Fura-2 340/380 ratio. (d). Comparison of the Fura-2 ratio for each group defined in the text. Data points represent the ratio of the stimulation by each concentration of E. coli O157:H7 (101-109 CFU/mL) in the group at the same time. Bars indicate the medians and p-value obtained by paired t-test. (two-tailed, nsp > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, n = 9) NF6: B cells (no Fura-2, Ca2+ and Mg2+ HBSS), 106 cells/mL. NF5: B cells (no Fura-2, Ca2+ and Mg2+ HBSS), 105 cells/mL. NC6: B cells (Fura-2, Ca2+ and Mg2+-free HBSS), 106 cells/mL. NC5: B cells (Fura-2, Ca2+ and Mg2+-free HBSS), 105 cells/mL. C6: B cells (Fura-2, Ca2+ and Mg2+ HBSS), 106 cells/mL. C5: B cells (Fura-2, Ca2+ and Mg2+ HBSS), 105 cells/mL.

Fura-2 ratios (FR) versus time were monitored using a microplate reader within 1 hour. After 103 CFU/mL of E. coli O157:H7 was added, the time point started as 0 min. As shown in Fig. 2c, a clearly increase in FR value was observed for E. coli O157:H7, when compared to the control (HBSS without target bacteria), which confirmed the sensing principle. In order to check the validation, further experiments were conducted to obtain the ratio of Ca2+-free (Camin) and Ca2+-bound (Camax) conditions. Results showed that the ratios of FR and Control were between the ratios of Camin and Camax, indicating that the obtained result was valid.

The gradual increase of Fura-2 ratio was observed in both FR and Control (Fig. 2c) during the detection time from 0 to 60 min. This might be due to the leaking out of Fura-2 from the cells, which should be subtracted before the calibration is performed32.

To investigate the variation of the biosensor whether there was the presence of Ca2+ in the solution, intracellular Ca2+ levels were measured with and without Ca2+ in the solution using different concentrations of B cells with adding different concentrations of E. coli O157:H7 (101-109 CFU/mL). Paired t-test was conducted, as shown in Fig. 2d. In group NF (no Fura-2), B cells were not loaded with Fura-2, which was set as a blank control. The ratio obtained in NF6 (no Fura-2, 106 cells/mL) was significantly higher than that obtained in NF5 (no Fura-2, 105 cells/mL) (p < 0.0001). In group NC (no Ca2+), B cells were loaded with Fura-2 in the Ca2+ and Mg2+-free solution. There was no significant difference (p > 0.05) between NF5 (no Fura-2, 105 cells/mL) and NF6. In group C (with Ca2+), B cells were loaded with Fura-2 in the Ca2+ and Mg2+ solution. The ratio obtained in C6 (with Ca2+, 106 cells/mL) was much higher than that obtained in C5 (with Ca2+, 105 cells/mL) (p = 0.0025). Since higher Fura-2 ratios were observed when the 106 cells/mL of B cells was tested in different groups, 106 cells/mL of B cells were used to carry out the comparison among the groups. The Fura-2 ratios obtained from NF6 were much lower than that obtained from NC6 (no Ca2+, 106 cells/mL) (p = 0.0324) and C6 (with Ca2+, 106 cells/mL) (p < 0.0001). A pronounced decrease in Ca2+ response (p = 0.0006) was observed in NC6 when the detection was carried out in the Ca2+ and Mg2+-free solution as compared to C6, where the Ca2+ and Mg2+ were present in the solution.

Detection of E. coli O157:H7 in pure culture

The developed B cell biosensor was evaluated for the detection of E. coli O157:H7 ranging from 101 to 105 CFU/mL in pure culture. The control measurement was carried out using HBSS to replace E. coli O157:H7. Comparing to the control, the means of the ratio were significantly different when the concentration of E. coli O157:H7 varies from 102-105 CFU/mL. As shown in Fig. 3a the ratio was found to increase linearly with the increase of the number of bacteria presented in the solution from 101 to 103 CFU/mL. A calibration curve was fitted by linear regression, yielding Y=0.0565X + 0.6753 (R2  = 0.96). When equivalent amounts of extracted LPS were used to stimulate the B cell (101-107 CFU/mL), the linear relation between the amount of LPS and the ratio was observed (Fig. 3b). A calibration curve was fitted by linear regression, yielding Y=0.03183X + 0.6532 (R2 = 0.83).

Figure 3
Figure 3

Detection of E. coli O157:H7 in pure culture. (a). Plot of the biosensor’s response, Fura-2 ratios, to different concentrations of E. coli O157:H7 in pure culture. Insert: Linear calibration curve obtained for E. coli O157:H7 in the range of concentrations from 101 to 103 CFU/mL. All data shown is mean±s.e.m. and p-value obtained by unpaired t-test. (two-tailed, nsp > 0.05, *p < 0.05, **p < 0.01, n = 4). (b). Plot of the biosensor’s response, Fura-2 ratios, to different concentrations of LPS extracted from E. coli O157:H7 which is equal to bacterial concentrations from 101 to 107 CFU/mL. All data shown is mean±s.e.m. and p-value obtained by unpaired t-test. (two-tailed, nsp > 0.05, *p < 0.05, **p < 0.01, n = 3). (c). Inclusivity tests and exclusivity tests to compare the B cell biosensor’s response to target and non-target bacteria species at 103 CFU/mL. All data shown is mean±s.e.m. and p-value obtained by unpaired t-test. (two-tailed, nsp > 0.05, *p < 0.05, **p < 0.01, n = 3). (d). ROC curves for all sensitivity sets and specificity sets with three replicates in bacteria pure culture conducted continuously for 30 min.

Selective detection of E. coli O157:H7 is necessary for use of the biosensor in food safety. Therefore, the abilities of the B cell biosensor to detect target organisms and non-target organisms were determined by the inclusivity test and the exclusivity test, respectively. Eight strains of E. coli comprising 2 different serogroups (O157 and non-O157) were tested, and seven of the strains, except for EHEC (non-O157), were sensitive for the detection in inclusivity tests. Exclusivity tests using 7 strains from 6 distinct different genera showed high specificity (Fig. 3c). p-values of the results of inclusivity and exclusivity tests are shown in Supplementary Table S5.

Listeria monocytogenes, S. Typhimurium and V. parahaemolyticus were chosen as representative bacteria in exclusivity tests, because this B cell was constructed to produce monoclonal antibodies (MAbs) specific for the LPS of E. coli O157 and group N Salmonella for use as highly specific diagnostic reagents. Bio X cell (West Lebanon, NH, USA) has developed anti-mouse IgM isotype control (Catalog number: BE0087) using B cell MARC 29F8. Mabs 29F8 was core antigen specific and also was determined to react with the LPS core antigen on the basis of its reactivity with the 12- to 16-kDa bands of the aqueous-phase LPS47. L. monocytogenes and V. parahaemolyticus were used as the representatives of Gram-positive and Gram-negative bacteria respectively. As to S. Typhimurium, it was used to test the specificity of this biosensor to O-antigen. The polysaccharides of group N Salmonella such as S. godesberg belong to chemotype VI (basal sugars with additional galactosamine and fucose). S. Typhimurium belongs to group D (polysaccharides chemotype XVI: basal sugars with additional mannose, rhamnose and tyvelose)48.

The accuracy of this sensor was evaluated quantitatively by Receiver Operating Characteristic (ROC) curve (Fig. 3d). The ratios of Fura-2 in the presence of E. coli O157:H7 ATCC 43888 at different concentrations respectively were detected continuously for 30 min, and each concentration was repeated three times. The collected data was used to obtain the ROC curve. The accuracy of the detection as represented by the area under ROC (AUR) curves was determined to be 0.7319, 0.7690, 0.8484, 0.7817 and 0.7885, respectively, for the concentrations of E. coli O157:H7 from 101, 102, 103,104 to 105 CFU/mL. All values of AUR obtained from different concentrations of E. coli O157:H7 were greater than 0.7 in a range of 0.7319 to 0.8484, indicating that the B cell biosensor was accurate in the detection range. It is observed that AUR value dropped to 0.7817 and 0.7885 when the concentration of E. coli O157:H7 increased to 104 CFU/mL and 105 CFU/mL, respectively. When 103 CFU/mL L. monocytogenes, S. Typhimurium, and V. parahaemolyticus were tested, the AUR values dropped to 0.5689, 0.5028 and 0.6001, respectively.

Detection of E. coli O157:H7 in ground beef

The B cell biosensor was further evaluated with bacteria inoculated ground beef extract samples. Based on the experiments performed using pure culture, we chose ground beef extract that were inoculated with 102 CFU/mL, 103 CFU/mL and 104 CFU/mL E. coli O157:H7 for the test. The B cells’ response to the ground beef extract inoculated with E. coli O157:H7 ATCC 43888 exhibited a similar response to the pure culture. A control measurement was carried out in the absence of E. coli O157:H7. As shown in Fig. 4a, significant differences, as compared to a control, were observed when the E. coli O157:H7 concentrations varied from 102-104 CFU/mL (p = 0.0420, 0.0002 and 0.0001 respectively).

Figure 4
Figure 4

Detection of E. coli O157:H7 in ground beef. (a). Plot of the biosensor’s response, Fura-2 ratio, to different concentrations of E. coli O157: H7 in ground beef. All data shown is mean±s.e.m. and p-value obtained by unpaired t-test. (two-tailed, *p < 0.05, ***p < 0.001, n = 3). (b). Biosensor’s response to different pathogens with the same concentration of 103 CFU/mL in ground beef. All data shown is mean±s.e.m. and p-value obtained by unpaired t-test. (two-tailed, nsp > 0.05, **p < 0.01, ***p < 0.001, n =3)

L. monocytogenes, S. Typhimurium and V. parahaemolyticus were used in ground beef samples as non-target bacteria to evaluate the specificity of the biosensor. As shown in Fig. 4b, when the same concentration (103 CFU/mL) of different pathogen was detected, the ratio of the sample with E. coli O157:H7 was extremely significant compared to the control (p = 0.0002), but ratios of Fura-2 with non-target bacteria weren’t found the significant differences compared to the control (p > 0.05) in the ground beef extract samples. The very significant difference between the detection of E. coli O157:H7 and L. monocytogenes was found (p = 0.0095), and the extremely significant differences between the detection of E. coli O157:H7 and S. Typhimurium and V. parahaemolyticus were found (p = 0.0001 and 0.0002 respectively).


The proposed Ca2+ indicator based B cell biosensor for detecting E. coli O157:H7 with better combined detection time and sensitivity was conceived, tested and validated in this study. BCR-induced Ca2+ flux can be triggered in seconds, and the Ca influx will happen instantly29, leading to Fura-2 (a Ca2+ sensitive divalent metal ion chelator) complexing with Ca2+33, all of them happened in a very short time. In the optimization of the B cell biosensor, our data suggested that the non-subtracted ratios obtained from the earlier time would be more accurately correlated to E. coli O157:H7. To determine the optimum time for the detection, the ratio of FR obtained from the different time was subtracted by the ratio of Control at the same time. It was found that after four to six min the response signal became stable. Thus, the ratio at 10th min was used as a detection ratio in the subsequent experiments. ROC curves drawn using the data collected from continuous 30 min in the bacterial pure culture (Fig. 3d) showed the same results as that at the 10th min.

From Fig. 2d, it is very clear that the B cells with the higher concentration combined with relatively the same concentration of E. coli O157:H7 had higher ratio produced by autofluorescence signal of bacteria, and the ratio produced by the autofluorescence of B cells without loading Fura-2 combined with E. coli O157:H7 had significant differences from the ratio produced by the B cells loaded with Fura-2 when the detection solution was with or without Ca2+ and Mg2+. Considering the previous result obtained from the comparison between group NC5 and NC6, these data divulged that a proportion of the total BCR induced Ca2+ flux observed is due to extracellular Ca2+ influx. The previous research on CD20 induced cytosolic Ca2+ flux also indicated a proportion of the total Ca2+ flux observed is due to extracellular Ca2+ influx in B cells49. The aforementioned results showed that the autofluorescence signal from the organism was negligible in comparison with the Fura-2 response signal. The higher response signal could be obtained from the higher concentrations of the B cells because more Ca2+ influx from the extracellular would increase the ratio, which would increase the biosensor sensitivity. Therefore, 106 cells/mL B cells loaded with Fura-2 in the solution with Ca2+ were used in the later tests.

In the detection of E. coli O157:H7 in pure culture, the B cell biosensor provided an increasing signal response which is linear in relation to the logarithm of E. coli O157:H7 concentration (101-103 CFU/mL). The ratio began to drop when the concentration of E. coli O157:H7 was 104 CFU/mL and 105 CFU/mL, probably because the high number of E. coli cells triggered the B cell expansion and apoptosis50, subsequently BCR-induced Ca2+ signaling was interfered. To verify it, the equivalent amounts of extracted LPS from E. coli O157:H7 were detected by this B cell biosensor. As shown in Fig. 3a, the ratio increased linearly with an increase of the equivalent amounts of LPS (101-107 CFU/mL). It is consistent with a recent study showing the ability of B cells to present antigen-derived peptides to T cells was also dependent on the amount of antigen tethered on the membrane51.

In the selective detection of E. coli O157:H7 in pure culture, our results showed that nonspecific binding did not lead to significant increase in the ratio with respect to the control in this B cell biosensor. Our results also showed that the MAbs on the surface of B cells are specific to O-antigen and the LPS of Gram-negative bacteria, which might have caused less response to EHEC (non-O157) and Gram-positive bacteria such as L. monocytogenes and Lactobacillus plantarum, respectively. In Fig. 3c, the statistical analysis of p-values (based on t-test) shows significant differences between inclusive bacteria and exclusive bacteria (Supplementary Table S5), however, it is recognized that the signal of all E. coli O157:H7 in the inclusivity test may need to be further amplified to ensure the true positive or true negative results for real food sample testing.

Based on ROC curve in Fig. 3d, it is quite evident that the B cell biosensor has a good specificity and its detection is reproducible and the ratio obtained from the 10th min could represent the results obtained from 10 min to 30 min during the detection. Since there was no significant difference as compared to the control when E. coli O157:H7 concentration was 101 CFU/mL, the detection range of the biosensor for E. coli O157:H7 was 102-105 CFU/mL.

For the detection of bacteria in the pure culture, this B cell biosensor had a detection limit of 5.9×102 CFU/mL. An additional feature of this biosensor is its rapid analysis time of 10 min. It is shorter than most biosensing methods which have the detection limits in the range of 102 to 104 CFU/mL, such as surface plasmon resonance (SPR)52, enzyme-linked immunosorbent assay (ELISA)53, fiber optic54 and quartz crystal microbalance (QCM)55. E. coli O157:H7 bacteria as low as 7 CFU were detected on the screen-printed carbon electrode (SPCE) sensor in 70 min56, and 10 CFU/mL were detected by the long-range surface plasmon-enhanced fluorescence spectroscopy (LRSP-FS) in 40 min57. There are also some biosensors with a low detection limit and a short detection time, such as 67 CFU/mL were detected by a lateral-flow immunobiosensor based on electrospun nanofibers and conductive magnetic nanoparticles (MNPs) in 8 min58, 500 CFU/g in lettuce were also detected by CANARY (cellular analysis and notification of antigen risks and yields) sensor in less than 5 min11. In this study, since only 30 μL test sample was used in each well for detection, the B cell biosensor was actually able to detect 18 cells of E. coli O157:H7. If some concentration procedures were applied in sample preparation, the detection limit of the B cell biosensor could be further improved to the level of several cells per mL. This biosensor can be easily operated in sample handling and preparation for multiplex tests, which has the potential for rapid, high throughput detection of foodborne pathogens.

For the detection of bacteria in ground beef, this B cell biosensor had a detection limit of 8.6×102 CFU/mL (26 CFU in 30 μL test sample) responding to E. coli O157:H7 in the ground beef extract sample. According to the statistical analysis, the specificity significance of target and non-target bacteria detected in the pure culture and the ground beef are slightly different. It may be due to food particulates and high levels of background microbiota from beef extract affectted the sensitivity of this B cell biosensor.

The advantages of the rapid response to the antigen of B cell and the sensitivity to Ca2+ of Fura-2 were combined to develop this Ca2+ -indicator based B cell biosensor which successfully detected E. coli O157:H7 at a low detection limit and in short time in both pure culture and ground beef. The main hurdle that remains is that the B cells themselves are not comparably ‘stable’ to other biosensing elements such as antibodies or nucleic acid probes. The signal response varies from batch to batch, and is influenced mainly by the cell’s viability, cell concentration and the exposing time after loading Fura-2. To overcome this hurdle, freshly prepared B cells are preferred, and the controls need to be employed for each batch test.

In summary, B cell MARC 29f8 loaded with Ca2+ -indicator Fura-2 was successfully developed as a convenient and sensitive biosensor. By using E. coli O157:H7 as a model target pathogen, the target bacteria were detected by measuring the intracellular Ca2+ concentration caused by BCR induced Ca2+ flux. This B cell based biosensor combined the properties of the sensitivity to Ca2+ of Fura-2 and the rapid response to the antibody of B cells.

The results showed that the responses of the developed B cell biosensor were sensitive and specific with a detection limit as low as 102 CFU/mL in both pure culture and ground beef within 10 min. The combination of the fluorescently loaded B cells specifically against E. coli O157:H7 and a microplate reader will pave the way for designing B cell based biosensors with features of easy operation, low-cost and high throughput for food detection of pathogens in food safety, biosafety, clinical diagnostics and environmental monitoring.



Dulbecco’s modified Eagle’s medium (DMEM), Dulbecco’s modified Eagle’s medium without phenol red (DMEM without PR), heat-inactivated fetal bovine serum (FBS), Hank’s balanced salt solution (HBSS), Ca2+ and Mg2+-free HBSS, MEM non-essential amino acids (MEM NAA), 0.4% Trypan Blue Solution, Fura-2/AM and 0.5 M EDTA were purchased from Life Technologies (Carlsbad, CA, USA). All bacteria culture mediums were purchased from BD (Sparks, MD, USA). Goat antimouse IgM-HRP (horseradish peroxidase), Lipopolysaccharide (LPS) extraction kit and silver staining kit were obtained from AbD (AbD Serotec, USA), Intron Biotechnology (Gyeonggi-do, Korea) and Sangon Biotech (Shanghai, China), respectively. All other reagents were obtained from Sigma (St. Louis, MO, USA). All solutions and buffers used in this study were prepared in sterile de-ionized (DI) water from Millipore Direct-8 system (Billerica, MA, USA).

B cell lines and culture conditions

The B cell line MARC 29F8 was obtained from ATCC (American Type Culture Collection, Manassas, VA; ATCC number CRL-2508). The organism of this B cell line is mouse, cell type is hybridoma B lymphocyte, and isotype is IgM. It was routinely cultured in DMEM supplemented with 4 mM L-glutamine adjusted to contain 1.5 g/L sodium bicarbonate, 4.5 g/L glucose, 1 vol% MEM NAA and 10% heat-inactivated FBS. The cells were cultured at 37 °C in a humid atmosphere containing 7% CO2. Liquid-nitrogen-stored MARC 29F8 cells (passage numbers between 5 and 10) were suspended at a ratio of 1:10 in the complete medium and grown for 72 h. MARC 29F8 cells were passaged again in DMEM with or without phenol red supplemented with 10% heat-inactivated FBS in T-25 and T-75 flasks (Falcon, Oxnard, USA) until they reached to a logarithmic phase of growth (usually 72 h) after which they were used for experiments. The cell count was performed using a TC10 Automated Cell Counter (Bio-Rad, Hercules, CA, USA) according to the manufacturer’s instructions.

Bacterial strains and culture conditions

Bacterial strains used in this study were obtained from the American Type Culture collection (ATCC), China Center of Industrial Culture Collection (CICC), China National Center for Medical Culture Collections (CMCC), the Chinese Zhejiang Center for Disease Control and Prevention (Zhejiang Province CDC) and Zhejiang University (ZJU). The detailed informations are shown in Supplementary Table S1 and Table S2.

Lactobacillus plantarum and other strains were grown on MRS broth and BHI broth at 37 °C, respectively. Listeria monocytogenes and other strains were cultured 48 h and 24 h, respectively. The concentration of bacteria in this study was determined in triplicate by enumeration on Trypticase Soy Agar (TSA). The detailed information is described in the Supplementary Information.

Scanning electron microscopy (SEM) imaging SEM samples were prepared (see the Supplementary Information) and imaged using a Hitachi TM1000 scanning electron microscope (SEM) (Hitachi, Japan).

Preparation of a B cell biosensor

The B cell biosensor was constructed under sterile conditions by preparing B cells loaded with Ca2+ -indicator Fura-2, with a slight modification to the measurement of [Ca2+]i in whole cell suspensions using Fura-233. In brief, for fluorescent Ca2+ flux experiment, undifferentiated MARC 29F8 cells were harvested in DMEM without PR and incubated at 37 °C for 5 min. B cells were washed three times in HBSS after DMEM (without PR) was removed. Then the cell suspensions were incubated with 4.5 nmol of Fura-2/AM per 106 cells for 30 min at 37 °C in the dark and the cells sediment were resuspended in 30 mL HBSS buffer. After incubating for an additional 15 min at room temperature in the dark to allow for the de-esterification of the Fura-2/AM, the cells were sedimented, and then split and resuspended in 30 mL of HBSS buffer and Ca2+ and Mg2+-free HBSS buffer twice respectively.

Detection of E. coli O157:H7 in pure culture

Briefly, a portion of 30 μl Fura-2 loaded MARC 29F8 B cells were added into each well of a 96-well plate, and then 30 μl analytes were loaded into the wells. To assess dynamic ranges for the microplate reader, maximum and minimum fluorescence values (at 340 and 380 nm excitation wavelengths) were determined in separate experiments in which Fura-2 loaded MARC 29F8 cells were incubated with 0.1% Triton X-100 and 4.5 mM ethylenediamine-tetraacetic acid (EDTA), respectively. Fluorescence was evoked by 340- and 380-nm excitation wavelengths (F340 and F380) and collected at 510 nm using Synergy H1 Hybrid Multi-Mode Microplate reader (BioTek, Winooski, VT, USA). Data was collected every 120 s in the plate reader. Data from all measurements were presented as 340/380 fluorescence ratios directly representative of changes in intracellular Ca2+30. E. coli O157:H7 ATCC 43888 was serially diluted in HBSS from 6.2×101 to 6.2×109 and from 5.9×101 to 5.9×105 in the optimization test and the accuracy test, respectively.

Equivalent amounts of extracted LPS from E. coli O157:H7 ATCC 43888 were used to verify the linear relation between the bacterial concentration and the response of the B cell biosensor. The detailed information is described in the Supplementary Information.

Totally, 8 strains and 7 strains at the bacterial concentration of 103 CFU/mL were used for inclusivity tests and exclusivity tests, respectively (Supplementary Table S1 and Table S2). E. coli O157:H7 ATCC 43888 and HBSS without any bacteria were added into Fura-2 loaded B cells as the positive control and negative control, respectively. The sensor response to different concentrations of bacteria was calculated as the mean value of ratios measured in three independent tests.

Detection of E. coli O157:H7 in ground beef

To validate the developed B cell biosensor, experiments were conducted using ground beef extract. The ground beef were purchased from a local grocery store, and contaminated with three different concentrations of E. coli O157:H7 to mimic bacterial contamination. Twenty-five grams of ground beef was mixed with 225 ml of HBSS in Filtra bags (Labplas Inc., Quebec, Canada) and stomached with Stomacher 400 (Seward, Norfolk, UK) for 1 min. Then the washing solution was collected. One milliliter of E. coli O157:H7 cells at dilutions of 103, 104, and 105 CFU/mL was added to 9 ml of washing solution of the ground beef sample, to obtain the ground beef samples contaminated with E. coli O157:H7 at 102, 103, and 104 CFU/mL. Then the same procedure for detection of E. coli O157:H7 in HBSS was followed using the prepared ground beef extract samples. The Fura-2 loaded B cells in the ground beef extract sample were used as a control for the test.

Statistical analysis

Data represents mean±s.e.m. (standard error of the means) of three experiments. The data were analyzed by GraphPad Prism software (GraphPad, San Diego, CA). Paired t-test was used in section Optimization of the B cell biosensor to analyze the differences between two groups which were both added with nine different concentrations of E. coli O157: H7 respectively. Unpaired t-test was used in the rest of experiments to analyze the differences between two groups which were added with the same concentration of bacterial pathogens. p < 0.05 were considered significant.

Additional Information

How to cite this article: Wang, L. et al. B cells Using Calcium Signaling for Specific and Rapid Detection of Escherichia coli O157:H7. Sci. Rep. 5, 10598; doi: 10.1038/srep10598 (2015).


  1. 1.

    & Foodborne Diseases Active Surveillance Network (FoodNet) in 2012: a foundation for food safety in the United States. Clin. Infect. Dis. 54 Suppl 5, S381–384 (2012).

  2. 2.

    , , , & Foodborne illness acquired in the United States—unspecified agents. Emerg. Infect. Dis. 17, 16–22 (2011).

  3. 3.

    et al. Foodborne illness acquired in the United States—major pathogens. Emerg. Infect. Dis. 17, 7–15 (2011).

  4. 4.

    Economic burden from health losses due to foodborne illness in the United States. J. Food Prot. 75, 123–131 (2012).

  5. 5.

    et al. Detection of food-borne pathogens with DNA arrays on disk. Talanta 101, 405–412 (2012).

  6. 6.

    et al. Incidence and trends of infection with pathogens transmitted commonly through food - Foodborne diseases active surveillance network, 10 U.S. Sites, 2006–2013. Morb. Mortal. Weekly Rep. 63, 328–332 (2014).

  7. 7.

    , , & Biosensors as innovative tools for the detection of food borne pathogens. Biosens. Bioelectron. 28, 1–12 (2011).

  8. 8.

    , & Whole-cell biosensors for environmental monitoring. Biosensors 4, 299–311 (1989).

  9. 9.

    et al. Genetically engineered pfabA pfabR bacteria: an electrochemical whole cell biosensor for detection of water toxicity. Anal. Chem. 78, 4952–4956 (2006).

  10. 10.

    et al. Luminescent yeast cells entrapped in hydrogels for estrogenic endocrine disrupting chemical biodetection. Biosens. Bioelectron. 21, 2263–2269 (2006).

  11. 11.

    et al. A B cell-based sensor for rapid identification of pathogens. Science 301, 213–215 (2003).

  12. 12.

    , , & Development of a whole-cell-based biosensor for detecting histamine as a model toxin. Anal. Chem. 76, 4156–4161 (2004).

  13. 13.

    et al. Continuous, noninvasive monitoring of local microscopic inflammation using a genetically engineered cell-based biosensor. Lab. Invest. 85, 1429–1439 (2005).

  14. 14.

    et al. Detection of heavy metal toxicity using cardiac cell-based biosensor. Biosens. Bioelectron. 22, 3224–3229 (2007).

  15. 15.

    , , , & A novel and simple cell-based detection system with a collagen-encapsulated B-lymphocyte cell line as a biosensor for rapid detection of pathogens and toxins. Lab. Invest. 88, 196–206 (2008).

  16. 16.

    & Mammalian cell-based biosensors for pathogens and toxins. Trends Biotechnol. 27, 179–188 (2009).

  17. 17.

    , , & Probing the bioelectrochemistry of living cells. Electrochim. Acta 53, 6690–6697 (2008).

  18. 18.

    , & Biosensors for ligand detection. Adv. Appl. Microbiol. 64, 137–166 (2008).

  19. 19.

    Cellular biosensing: Chemical and genetic approaches. Anal. Chim. Acta 568, 211–216 (2006).

  20. 20.

    Cell-based biosensors in proteomic analysis. Front. Drug Des. Discovery 2, 225–239 (2005).

  21. 21.

    Electronic sensors with living cellular components. P. IEEE 91, 915–929 (2003).

  22. 22.

    et al. Assessment of the virulence of Listeria monocytogenes: agreement between a plaque-forming assay with HT-29 cells and infection of immunocompetent mice. Int. J. Food Microbiol. 68, 33–44 (2001).

  23. 23.

    , , , & Development of a mast cell-based biosensor. Biosens. Bioelectron. 23, 1024–1031 (2008).

  24. 24.

    , & A label-free biosensor-based cell attachment assay for characterization of cell surface molecules. Sensor. Actuat. B-Chem. 114, 559–564 (2006).

  25. 25.

    et al. Lipid rafts unite signaling cascades with clathrin to regulate BCR internalization. Immunity 17, 451–462 (2002).

  26. 26.

    & Alkaline phosphatase release assay to determine cytotoxicity for Listeria species. Lett. Appl. Microbiol. 26, 305–310 (1998).

  27. 27.

    CANARY sensor for rapid, sensitive identification of pathogens. Bio. Micro. and Nanosystems Conference: BMN’ 06, San Francisco. New York, USA: IEEE. (2006, Jan. 15-18)

  28. 28.

    et al. Rapid sensors for biological-agent identification. Lincoln Lab. J. 17, 63–84 (2007).

  29. 29.

    & B-lymphocyte calcium influx. Immunol. Rev. 231, 265–277 (2009).

  30. 30.

    , & A new generation of Ca2+ indicators with greatly improved fluorescence properties. J. Biol. Chem. 260, 3440–3450 (1985).

  31. 31.

    The use of fura-2 for estimating Ca buffers and Ca fluxes. Neuropharmacology 34, 1423–1442 (1995).

  32. 32.

    , , & Fura-2 measurement of cytosolic free Ca2+ in monolayers and suspensions of various types of animal cells. J. Cell Biol. 105, 2145–2155 (1987).

  33. 33.

    , , & Calcium Signaling Protocols 2nd edn, Vol. 312 (ed. ) Ch. 2, 37–45 (Humana Press, 2006).

  34. 34.

    & Store-operated calcium channels. Physiol. Rev. 85, 757–810 (2005).

  35. 35.

    & Cyclic nucleotide-gated ion channels. Physiol. Rev. 82, 769–824 (2002).

  36. 36.

    , & Principles and properties of ion flow in P2X receptors. Front. Cell. Neurosci. 8, 1–18 (2014).

  37. 37.

    , & Ca2+ buffering and action potential-evoked Ca2+ signaling in dendrites of pyramidal neurons. Biophys. J. 70, 1069–1081 (1996).

  38. 38.

    , , & Spine-neck geometry determines NMDA receptor-dependent Ca2+ signaling in dendrites. Neuron 46, 609–622 (2005).

  39. 39.

    & Imaging calcium in neurons. Neuron 73, 862–885 (2012).

  40. 40.

    , , , & Molecular components of the B-cell antigen receptor complex of the IgM class. Nature 343, 760–762 (1990).

  41. 41.

    et al. BANK regulates BCR-induced calcium mobilization by promoting tyrosine phosphorylation of IP3 receptor. EMBO J. 21, 83–92 (2002).

  42. 42.

    et al. Orai1 is an essential pore subunit of the CRAC channel. Nature 443, 230–233 (2006).

  43. 43.

    & Conformational dynamics of STIM1 activation. Nat. Struct. Mol. Biol. 20, 918–919 (2013).

  44. 44.

    & Calcium signaling in lymphocytes. Curr. Opin. Immunol. 20, 250–258 (2008).

  45. 45.

    Calcium signalling in lymphocyte activation and disease. Nat. Rev. Immunol. 7, 690–702 (2007).

  46. 46.

    & Regulation of phosphoinositide-specific phospholipase C isozymes. J. Biol. Chem. 272, 15045–15048 (1997).

  47. 47.

    , , , & Production and characterization of monoclonal antibodies specific for the lipopolysaccharide of Escherichia coli O157. J. Clin. Microbiol. 35, 679–684 (1997).

  48. 48.

    , & Immunochemistry of O and R antigens of Salmonella and related Enterobacteriaceae. Bacteriol. Rev. 30, 192–255 (1966).

  49. 49.

    et al. Induction of cytosolic calcium flux by CD20 is dependent upon B Cell antigen receptor signaling. J. Biol. Chem. 283, 16971–16984 (2008).

  50. 50.

    et al. Cell activation and apoptosis by bacterial lipoproteins through toll-like receptor-2. Science 285, 736–739 (1999).

  51. 51.

    et al. B cell ligand discrimination through a spreading and contraction response. Science 312, 738–741 (2006).

  52. 52.

    , , , & Detection of Escherichia coli O157: H7 using immunosensor based on surface plasmon resonance. J. Microbiol. Biotechnol. 12, 780–786 (2002).

  53. 53.

    , & Novel indirect enzyme-linked immunosorbent assay (ELISA) method to detect Total E. coli in water environment. Anal. Chim. Acta 590, 224–231 (2007).

  54. 54.

    , , & Fiber-optic biosensor employing Alexa-Fluor conjugated antibody for detection of Escherichia coli O157: H7 from ground beef in four hours. Sensors 6, 796–807 (2006).

  55. 55.

    & A self-assembled monolayer-based piezoelectric immunosensor for rapid detection of Escherichia coli O157: H7. Biosens. Bioelectron. 19, 563–574 (2004).

  56. 56.

    & Rapid electrochemical detection of polyaniline-labeled Escherichia coli O157: H7. Biosens. Bioelectron. 26, 2208–2214 (2011).

  57. 57.

    , , & Long-range surface plasmon-enhanced fluorescence spectroscopy biosensor for ultrasensitive detection of E. coli O157: H7. Anal. Chem. 83, 674–677 (2011).

  58. 58.

    et al. Novel biosensor based on electrospun nanofiber and magnetic nanoparticles for the detection of E. coli O157: H7. IEEE Trans. Nanotechnol. 11, 676–681 (2012).

Download references


We are grateful to Robert L. Dienglewicz for cell culture training and Lisa Cooney Kelso for SEM sample preparation and imaging. This research was financially supported in part by Arkansas Biosciences Institute (ABI) and China Ministry of Science and Technology (Project No. 2013BAD19B02).

Author information


  1. College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China

    • Ling Wang
    •  & Yanbin Li
  2. Department of Biological and Agricultural Engineering, University of Arkansas, Fayetteville, AR 72701, USA

    • Ling Wang
    • , Ronghui Wang
    •  & Yanbin Li
  3. Department of Poultry Science, University of Arkansas, Fayetteville, AR 72701, USA

    • Byung-Whi Kong
  4. Department of Biomedical Engineering, Binghamton University-SUNY, Binghamton, NY 13902, USA

    • Sha Jin
    •  & Kaiming Ye
  5. College of Animal Sciences, Zhejiang University, Hangzhou 310058, China

    • Weihuan Fang


  1. Search for Ling Wang in:

  2. Search for Ronghui Wang in:

  3. Search for Byung-Whi Kong in:

  4. Search for Sha Jin in:

  5. Search for Kaiming Ye in:

  6. Search for Weihuan Fang in:

  7. Search for Yanbin Li in:


L. W. and R. W. conceived of the method. R. W. and Y. L. supervised the project. B.-W.K., S. J., K. Y. and W.F. co-supervised the project and provided guidance and assistance in cell culture and the manuscript writing. L.W. performed the experiments and wrote the main manuscript with the input from all other authors.

Competing interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to Yanbin Li.

Supplementary information


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Creative Commons BYThis work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit