A Novel PMA-Based PCR Assay Can Measure Viable Bacteria in Urine
Albert S T Lee, DO, PharmD, Olivia Lamanna, BS, Kenji Ishida, PhD, Michael Hsieh, MD, PhD.
Children's National Hospital, Washington, DC, USA.
BACKGROUND: Polymerase chain reaction (PCR) has been used in studies of the urine microbiome and the identification of pathogens causing urinary tract infections. However, PCR is limited by its inability to differentiate DNA originating from live and metabolically active versus dead, inactive bacteria. This has led to concern that urobiome studies and PCR-based diagnosis of UTI are confounded by the presence of relic DNA from dead bacteria. Propidium monoazide (PMA) dye can penetrate cells with compromised cell membranes and covalently bind to DNA, rendering it inaccessible to amplification by PCR. Although PMA has been shown to differentiate between dead and live bacteria in various media, its effectiveness in urine has not been previously studied. We sought to investigate the ability of PMA to differentiate between live and dead bacteria in urine.
METHODS: Varying amounts of viable or non-viable uropathogenic E. coli (UTI89) or PBS control were mixed with mouse urine. The samples were centrifuged to collect urine sediment or not centrifuged. Urine samples were incubated with PMA and DNA cross-linked using blue LED light. DNA was isolated and uidA gene-specific PCR was performed. For in vivo studies, mice were inoculated with UTI89, followed by ciprofloxacin treatment or no treatment. After the completion of ciprofloxacin treatment, an aliquot of urine was plated on non-selective LB agar and another aliquot was treated with PMA and subjected to uidA-specific PCR.
RESULTS: PMA’s efficiency in eliminating non-viable DNA signal was significantly better in PBS (dCT=13.69) versus unspun urine (dCT=1.58). This discrepancy was diminished by spinning down samples to collect sediment and resuspending it in PBS prior to PMA treatment. In 3 of 5 replicate groups of UTI89-infected mice, no bacteria grew in culture; however, there was PCR amplification of E. coli after PMA treatment in 2 of those 3 groups.
CONCLUSIONS: We have successfully developed PMA-based PCR methods for amplifying DNA from live bacteria in urine. Our results suggest that non-PMA bound bacteria DNA can be present in urine, even after antibiotic treatment. This indicates that viable but nonculturable E. coli can be present following therapy for UTI, and may explain why some patients have persistent symptoms but negative urine cultures after treatment of their infections.
Back to 2021 Posters