Background: Antibiotic susceptibility testing (AST) using fluorogenic and colorimetric reporters are well-established methods. Despite the superior sensitivity of luminescence, AST based on bioluminescence remains unexploited due to the need for addition of reagents and its limitations with respect to end-point analysis. Chemiluminescence detection is limited by the lack of stable luminogenic substrates compatible with microorganisms in an aqueous medium.
The AquaSpark™ substrates recently commercialized by BIOSYNTH are claimed to fulfill these criteria. We evaluated a luminogenic substrate of this range for AST. The study was focused on the AST of Gram positive bacteria using the broad range phosphatase substrate.
Materials/methods: 40 strains (30 staphylococci and 10 enterococci) were tested by broth microdilution with 10µM of broad range phosphatase substrate (BIOSYNTH). Cefoxitin, ciprofloxacin, erythromycin, gentamicin, levofloxacin, tetracycline and vancomycin were tested with appropriate strains. Medium, inoculum and antibiotic dilutions were based on CLSI standard M07 method. Incubation at 35°C, absorbance (OD660nm) and luminescence (LUM) measurements were performed using a micro-plate reader (TECAN). Kinetics of OD660nm and LUM signals were compared to the MIC obtained visually.
Results: LUM and OD660nm signals were correlated at 96% (16/16 antibiotic/strain tests for enterococci; 57/60 for staphylococci). All discordances were observed in presence of ciprofloxacin: no luminescence was detected while growth of resistant strains was obvious at high concentrations (above 8 mg/L).
For rapid MIC determination, analysis criteria were calculated for both OD660nm and LUM. Average time gain using LUM was 415 minutes for enterococci and 290 minutes for staphylococci. Application of these criteria led to 3 major errors for LUM (previous discordances with ciprofloxacin) and 4 very major errors (VME) for both OD660nm and LUM when compared to the MIC obtained visually. VME were obtained for vancomycin and could be resolved by modifying the interpretation criteria for this drug.
Conclusions: AquaSpark™ broad range phosphatase substrate can be used for chemiluminescence detection in kinetics in the presence of antibiotics and bacterial strains. LUM signal correlates to growth detection and can predict MIC with confidence. Its optimized use for AST requires definition of interpretation criteria which is challenging due to the diversity of resistance mechanisms.