Overestimation of the Legionella spp. load in environmental samples by quantitative real-time PCR: pretreatment with propidium monoazide as a tool for the assessment of an association between Legionella concentration and sanitary risk

Abstract

Quantitative polymerase chain reaction (qPCR) offers rapid, sensitive, and specific detection of Legionella in environmental water samples. In this study, qPCR and qPCR combined with propidium monoazide (PMA-qPCR) were both applied to hot-water system samples and compared to traditional culture techniques. In addition, we evaluated the ability of PMA-qPCR to monitor the efficacy of different disinfection strategies. Comparison between the quantification obtained by culture and by qPCR or PMA-qPCR on environmental water samples confirms that the concentration of Legionella estimated by GU/L is generally higher than that estimated in CFU/L. Our results on 57 hot-water-system samples collected from 3 different sites show that: i) qPCR results were on average 178-fold higher than the culture results (Δ log10=2.25), ii) PMA-qPCR results were on average 27-fold higher than the culture results (Δ log10=1.43), iii) propidium monoazide-induced signal reduction in qPCR were nearly 10-fold (Δ log10=0.95), and that iv) different degrees of correlations between the 3 methods might be explained by different matrix properties, but also by different disinfection methods affecting cultivability of Legionella. In our study, we calculated the logarithmic differences between the results obtained by PMA-qPCR and those obtained by culture, and we suggested an algorithm for the interpretation of PMA-qPCR results for the routine monitoring of healthcare water systems using a commercial qPCR system (iQ-check real-time PCR kit; Bio-Rad, Marnes-la-Coquette, France).

Keywords: Colony count; Legionella spp; PCR (polymerase chain reaction); Propidium monoazide; Sensitivity and specificity; Water microbiology.