In this work, a method was developed to establish Staphylococcus aureus biofilms on 96-well plates and automatically quantify viable cells within these biofilms by ATP-bioluminescence. Different strains were compared for biofilm formation. Cells from slime producing (SP) strain variants were more adherent (p < 0.001) and therefore more suitable for biofilm formation than non-slime producing original isolates. To compare biofilm support surfaces, SP biofilms were formed for 6, 24 and 48 h on 96-well polystyrene plates, containing wells coated with gelatin, poly-L-lysine or pre-treated for tissue culture and uncoated wells. Tissue culture-treated wells enhanced biofilm formation, allowing the highest growth (p < 0.001) in well-established biofilms (24 or 48 h old). For ATP quantification, the efficacy of different ATP extractants was compared: dimethyl sulphoxide (DMSO), trichloroacetic acid (TCA), a commercially available releasing reagent(R) (RR) and lysostaphin. A greater inhibitory effect on the ATP detection (p < 0.01), a more variable light emission (variation coefficient >/=50% vs. <19%, respectively) and a lower extraction efficiency (p < 0.05) were found in the case of TCA or lysostaphin in relation to RR or DMSO. DMSO was found preferable in relation to RR (upper detection limits 2.3 x 10(9) and 2 x 10(8) CFU/mL respectively) for bacterial ATP extraction from biofilms with high bacterial density. DMSO extracted ATP within seconds, light emission being stable for 6 h. The method developed allows automated viability determination of biofilm cells using bioluminescence and simultaneous study of factors affecting this viability (culture media, antibiotic types, antimicrobial concentrations, support surfaces and biofilm ages). It may be of use in bacteriological and antimicrobial research.
Copyright 1999 John Wiley & Sons, Ltd.