Antibiotic resistant bacteria have become a threat to world health. An advanced method of detection, based on the matrix assisted laser desorption ionisation time-of-flight mass spectroscopy can identify bacteria relatively rapidly, but it is not suitable to measure bacterial antibiotic resistance. Biosensors may be able to detect resistance by monitoring growth after capture on sensor surfaces but this option has not been addressed adequately. We have evaluated the growth of Escherichia coli after capture in 96 well microplates and observed that growth/capture efficiency was relatively similar for antibody-based techniques, but non-specific capture varied considerably. We confirm that neutravidin binds E. coli non-specifically, which limited its use with biotinylated antibodies or aptamers. Centrifugation enhanced bacterial growth/capture considerably, indicating that procedures enhancing the interaction between bacteria and surface-bound antibody have the potential to improve growth efficiency. Capture and growth required larger numbers of bacteria than capture and detection on biosensor surfaces. Previously, we reported that the minimum concentration of live E. coli required for initiating growth on a GaAs/AlGaAs biosensor was ~ 10 5 CFU/mL [Nazemi et al., Talanta 178 (2018) 69-77], and we speculated that this could be related to the poisonous effect of Ga-and As-ions released during dark corrosion of the biosensor, however in the present report we observed that the same minimum concentration of E. coli was required for growth in an ELISA plate. Thus, we argue that this limitation was related rather to bacterial inhibition by the capture antibodies. Indeed, antibodies at titres designed to capture bacteria inhibited bacterial growth when the bacteria were added to growth medium at titres less than 10 5 CFU/mL, indicating that antibodies may be responsible for the higher limits of sensitivity due to their potential to restrict bacterial growth. However, we did not observe E. coli release after 6 h following the capture indicating that these bacteria did not degrade antibodies.

It is recognized that bacterial adhesion usually occurs on conditioning films made of organic macromolecules absorbed to abiotic surfaces. The objectives of this study were to determine the extent to which milk protein-coated polystyrene (PS) pegs interfere with biofilm formation and the synergistic effect of this conditioning and hypertonic growth media on the bacterial adhesion and biofilm formation of Listeria innocua, used as a nonpathogenic surrogate for Listeria monocytogenes. PS pegs were uncoated (bare PS) or individually coated with whey proteins isolate (WPI), β-lactoglobulin, bovine serum albumin, or tryptic soy broth (TSB) and were incubated in bacterial suspensions in modified Welshimer's broth. After 4 h, the number of adherent cells was dependent on the coating, as follows: TSB (10(7) CFU/ml) > bare PS > β-lactoglobulin > bovine serum albumin ∼ WPI (10(4) CFU/ml). The sessile cell counts increased up to 24 h, reaching > 10(7) CFU per peg for all surfaces (P > 0.1), except for WPI-coated PS; this indicates that the inhibitory effects of milk protein conditioning films are transient, slowing down the adhesion process. The 4-h bacterial adhesion on milk protein-coated PS in modified Welshimer's broth supplemented with salt (0 to 10% [wt/vol]) did not vary (P > 0.1), indicating that conditioning with milk proteins was the major determinant for inhibition of bacterial adhesion and that the synergetic effect of salt and milk proteins on adhesion was minimal. Moreover, the presence of 5 to 10% salt significantly inhibited 24-h biofilm formation on the TSB-coated and bare PS, with a decrease of >3 log at 10% (wt/vol) NaCl and almost completely depleted viable sessile bacteria on the milk protein-coated PS.