A real-time PCR antibiogram for drug-resistant sepsis

Abstract

Current molecular diagnostic techniques for susceptibility testing of septicemia rely on genotyping for the presence of known resistance cassettes. This technique is intrinsically vulnerable due to the inability to detect newly emergent resistance genes. Traditional phenotypic susceptibility testing has always been a superior method to assay for resistance; however, relying on the multi-day growth period to determine which antimicrobial to administer jeopardizes patient survival. These factors have resulted in the widespread and deleterious use of broad-spectrum antimicrobials. The real-time PCR antibiogram, described herein, combines universal phenotypic susceptibility testing with the rapid diagnostic capabilities of PCR. We have developed a procedure that determines susceptibility by monitoring pathogenic load with the highly conserved 16S rRNA gene in blood samples exposed to different antimicrobial drugs. The optimized protocol removes heme and human background DNA from blood, which allows standard real-time PCR detection systems to be employed with high sensitivity (<100 CFU/mL). Three strains of E. coli, two of which were antimicrobial resistant, were spiked into whole blood and exposed to three different antibiotics. After real-time PCR-based determination of pathogenic load, a ΔC(t)<3.0 between untreated and treated samples was found to indicate antimicrobial resistance (P<0.01). Minimum inhibitory concentration was determined for susceptible bacteria and pan-bacterial detection was demonstrated with 3 gram-negative and 2 gram-positive bacteria. Species identification was performed via analysis of the hypervariable amplicons. In summary, we have developed a universal diagnostic phenotyping technique that assays for the susceptibility of drug-resistant septicemia with the speed of PCR. The real-time PCR antibiogram achieves detection, susceptibility testing, minimum inhibitory concentration determination, and identification in less than 24 hours.

Figure 1. The real-time PCR antibiogram achieves rapid susceptibility testing of septicemia in less than 24 hours.

Schematic that illustrates how phenotypic response can be monitored with the diagnostic speed of PCR to yield drastically decreased detection times. The real-time PCR antibiogram (middle) is determined by measuring the change in ΔC_t values for different antibiotics against a control. Parameters for the simulation (right) were optimized using experimental data and incubation times greater than 8+ hours allowed our detection system to assay cultures at ∼100 CFU/mL. The three-hour horizontal dashes before amplification depict manual sample preparation time.

Figure 2. The real-time PCR antibiogram for aiding antimicrobial selection and administration.

a-c) Real-time PCR amplification curves for measuring the susceptibility of susceptible, spectinomycin resistant, and the minimum inhibitory concentration of E. coli spiked in blood. Amplification curves were run in triplicate. d-f) The corresponding ΔC_t values for the real-time amplification curves. ΔC_t values >3.0 were assigned to indicate susceptibility, while ΔC_t values <3.0 designate resistance. *P<0.01 for multiple comparisons by the Holm t Test. t = 0 signifies initial bacterial levels with no incubation and negative curves denote sample preparation with no bacteria. Susceptible E. coli was used for minimum inhibitory concentration determination and 1x designates a spectinomycin concentration of 50 µg/mL.

Figure 3. The real-time PCR antibiogram for pan-bacterial detection and identification.

a) Pan-bacterial amplification illustrates universal binding of the primers and b) melt curve analysis provides evidence that consistent, species-specific amplicons were generated. c) Amplicons run on gel electrophoresis exhibit similar lengths of 200-300 base pairs and d) sequences submitted to the BLAST are readily identifiable.