Fluorimetric assays facilitate antibiotic screening

David L. Shenkenberg

When penicillin was discovered, it seemed to be a panacea. However, some bacteria have developed a resistance to penicillin and other antibiotics, and new antibiotics must be developed to kill these resistant strains. Although several classical microbiology methods can assess the efficacy of antibiotics, they require a second step because they are unable to determine in a single step whether bacteria have died or have merely stopped reproducing.

Fluorimetric methods can avoid the second step. Furthermore, they are quantitative and can be automated, whereas classical antibiotic susceptibility tests are entirely manual, and thus are time-consuming and prone to human error. Researchers from the University of Würzburg in Germany and from Queen’s University Belfast in the UK have developed a fluorimetric antibiotic screening assay based on a Beckman Coulter Inc. capillary electrophoresis system that includes equipment for fluorescence excitation and detection, and on Live/Dead BacLight fluorescence-labeling kits from Invitrogen Corp. of Carlsbad, Calif.

Fluorescent labeling of bacterial cells with Syto9 (green) and propidium iodide (red)enabled researchers to determine in a single step whether bacteria were alive by using a combination of capillary electrophoresis and fluorescence measurement.

The kits contain the fluorophore Syto9, which labels all of the cells green, and the fluorophore propidium iodide, which labels dead cells that have damaged membranes red. The red dye replaces the green in the dead cells, allowing investigators to measure each group. The researchers used the 488-nm line of an argon-ion laser to excite the fluorophores, and a corresponding detector equipped with the appropriate filters to detect emissions at 520 nm for live cells and at 655 nm for dead ones.

Fluorescence measurement with capillary electrophoresis yielded graphs such as this one, which showed the quantity of bacterial cells that were alive depending on the antibiotic concentration. Reprinted with permission of Analytical Chemistry.

Fluorescence measurement with capillary electrophoresis not only can distinguish between living and dead cells but also between single cells and longer chains, said Ulrike Holzgrabe, the principal investigator in Würzburg.

In capillary electrophoresis, cells or molecules of interest are driven by an electric field through tubing that is usually made of a soft polymer.

To prove that the capillary electrophoresis method works, the researchers measured the efficacy of various antibiotics on two strains of bacteria and compared the results to those measured with a Varian fluorescence microplate reader and with a classical microbiology assay. The graph resulting from the investigators’ technique and that obtained with the microplate reader give the ED50 measurement, the dose of the antibiotic effective for half of the bacterial population, whereas the classical assay gives the minimum inhibitory concentration, the minimum concentration of antibiotic needed to inhibit bacterial growth.

The researchers used the facultative anaerobe Streptococcus vestibularis and the aerobe Pseudomonas fluorescens bacterial strains, neither of which is pathogenic. They used ofloxacin, ciprofloxacin, ampicillin and vancomycin antibiotics, all of which are used to treat serious infections.

As reported in the Oct. 1 issue of Analytical Chemistry, the peaks from the capillary electrophoresis fluorescence measurements plotted with the percentage of live cells yielded a graph that showed in a single step the quantity of cells that were alive or dead from the antibiotic treatment. The data were in good agreement with the results of the microplate reader and classical microbiology measurements, demonstrating the reliability of the capillary electrophoresis method.

The researchers also found that injecting unlabeled bacteria into the capillary electrophoresis system yielded sharper peaks because it enabled the machine to compare the background signal from the unlabeled bacteria to the desired signal from the labeled bacteria.

In the future, the scientists will attempt to automate more of the assay and to develop new antibiotics, Holzgrabe noted.