A general method for rapid determination of antibiotic susceptibility and species in bacterial infections

Abstract

To ensure correct antibiotic treatment and reduce the unnecessary use of antibiotics, there is an urgent need for new rapid methods for species identification and determination of antibiotic susceptibility in infectious pathogenic bacteria. We have developed a general method for the rapid identification of the bacterial species causing an infection and the determination of their antibiotic susceptibility profiles. An initial short cultivation step in the absence and presence of different antibiotics was combined with sensitive species-specific padlock probe detection of the bacterial target DNA to allow a determination of growth (i.e., resistance) and no growth (i.e., susceptibility). A proof-of-concept was established for urinary tract infections in which we applied the method to determine the antibiotic susceptibility profiles of Escherichia coli for two drugs with 100% accuracy in 3.5 h. The short assay time from sample to readout enables fast appropriate treatment with effective drugs and minimizes the need to prescribe broad-spectrum antibiotics due to unknown resistance profiles of the treated infection.

FIG 1

Schematic illustration of the method. (A) The procedure starts with growth of a sample in antibiotic-free LB (−AB) and in LB supplemented with different antibiotics (we used ciprofloxacin and trimethoprim). Thereafter, the bacterial cells are treated to isolate the total DNA. (B) NaOH was added, in addition to heat treatment to lyse the bacteria, denature proteins, and fragment the DNA. (C) Magnetic bead particles with linked oligonucleotides were added to capture the relevant rRNA gene sequences. (D) To target the bacterial DNA, padlock probes were added. (E) Correctly hybridized padlock probes were circularized by a thermostable DNA ligase and amplified by RCA. RCPs were monomerized (F) and further amplified by a second cycle of RCA and simultaneous fluorescence labeling by hybridizing Cy3-tagged oligonucleotides to the generated RCPs (G). The fluorescence-labeled RCPs were analyzed using a high-throughput reader that digitally enumerates the RCPs, providing quantitation with high precision.

FIG 2

Growth curve of E. coli. E. coli (10⁴ CFU/ml) was inoculated into LB and cultured for 2 h at 37°C. Measurements were made every 30 min. The negative control (NC) did not contain any bacteria and was incubated for 90 min. The number of measured RCPs (●) is plotted on the left y axis, the viable count data (○) is plotted on the right y axis, and time is plotted on the x axis. The error bars indicate ±1 SD; n = 3.

FIG 3

Specificity of the assay. Approximately 10⁵ cells of three species (E. coli, P. aeruginosa, and P. mirabilis) were inoculated into LB and exposed to padlock probes (PLP) designed to specifically target each of these three species. The y axis displays the number of RCPs counted in a 100-s measurement. The error bars indicate ±1 SD; n = 3.

FIG 4

Antibiotic resistance patterns of E. coli causing UTI. Wild-type (WT), ciprofloxacin-resistant (Cip^r), and trimethoprim-resistant (Tmp^r) strains were cultured for 2 h either in LB without antibiotic (−AB) or LB with ciprofloxacin or trimethoprim. (A) Schematic illustration of the calculations made in order to establish the ASP. Growth is calculated as follows: X = x₂ × x₁ (absence of antibiotics), Y = y₂ × x₁ (ciprofloxacin-supplemented LB), and Z = z₂ × x₁ (trimethoprim-supplemented LB). Y/X is the relative growth in the presence of ciprofloxacin. Z/X describes the relative growth in the presence of trimethoprim. (B) Relative growth in the presence of these two antibiotics is plotted. The mean values of the results from two separate culture occasions are shown. The error bars indicate ±1 SD; n = 2.

FIG 5

Multiplex pathogen detection and ASP determination. Three different bacterial species with different resistances were mixed together (10⁴ CFU/ml of each species) and cultured in LB for 120 min in the presence or absence of ciprofloxacin and trimethoprim. (A) E. coli (Cip^s Tmp^r), P. aeruginosa (Cipⁱ Tmp^r), and P. mirabilis (Cip^s Tmp^s). (B) E. coli (Cip^s Tmp^s), P. aeruginosa (Cip^r Tmp^r), and P. mirabilis (Cip^s Tmp^s). (C) E. coli (Cip^s Tmp^r) and P. mirabilis (Cip^s Tmp^s) (D) E. coli (Cip^s Tmp^s) and P. mirabilis (Cip^s Tmp^s). The y axis represents relative growth, and the padlock probe used is displayed on the x axis.

FIG 6

Classification scheme of patient samples. All patient samples were classified according to the scheme shown in the flow chart.

FIG 7

Analysis of patient samples. (A) To develop the diagnostic algorithm, a training set of 32 patient samples (U1 to U34) was analyzed after 0 and 120 min of culture time. The samples included 14 E. coli-negative samples and 3 Cip^r, 15 Cip^s, 8 Tmp^r, and 10 Tmp^r E. coli-positive samples. The graph shows the percentage of growth in the presence of antibiotics. (B) Results of the validation study comprising 11 E. coli-positive samples out of 56 prospectively collected clinical samples. The percentage of growth in the presence of antibiotics is plotted on the y axis. The horizontal line displays the mean for the analyzed samples. The cutoff for resistance was set to 30% relative growth (horizontal dashed line) for both trimethoprim and ciprofloxacin.