Bacteriology
Design and implementation of a protocol for the detection of Legionella in clinical and environmental samples

https://doi.org/10.1016/j.diagmicrobio.2008.05.004Get rights and content

Abstract

Our laboratory has developed a novel real-time polymerase chain reaction (PCR) assay for the detection of Legionella pneumophila and differentiation from other Legionella spp. in clinical and environmental samples. The 23S rRNA gene was used as a target to detect all Legionella spp., and the mip gene was targeted for the specific detection of L. pneumophila in this multiplex Taqman® real-time PCR assay. The 23S rRNA gene is a novel target for Legionella testing; it detects all species and serogroups of Legionella without the contamination issues that accompany the use of the 16S rRNA gene as a target. This assay provides an analytical sensitivity of <1 colony-forming unit and a specificity of 100%. Because culture is important and provides a means for molecular typing via pulsed-field gel electrophoresis (PFGE), we developed a testing algorithm that includes both the new real-time PCR assay and culture for clinical and environmental samples and applied this algorithm during a period of 3 years. Of the 64 clinical samples received by our laboratory for Legionella testing during this period, PCR was found to be an essential diagnostic tool because only 13.3 % (2/15) clinical samples that were determined to be L. pneumophila were detected by culture during this period. Of the 276 environmental samples received for Legionella testing during this period, 140 were found to be positive for L. pneumophila. Of these 140 samples, 69.3% were detected by both PCR and culture methods, 29.3% were positive by PCR alone, and 1.4% were positive by culture methods alone. We feel these results indicate that our algorithm, including both PCR and culture, should be used for environmental samples. Among both the clinical and environmental Legionella samples identified by PCR, a subset was not suitable for culture because of issues of lengthy transport, antimicrobial treatment, or bacterial overgrowth. Samples like these are commonly submitted to our laboratory, so the use of our testing algorithm combining these methods is critical. We conclude that molecular and culture methods must be used in combination to provide the best and most comprehensive approach to laboratory detection and investigation of legionellosis.

Introduction

Whereas the bacteria of the genus Legionella are responsible for an acute self-limiting nonpneumonic illness known as Pontiac fever, a far more significant public health issue is the potentially fatal pneumonic form of legionellosis, Legionnaires' disease. Legionella pneumophila is the most common pathogenic species within the genus Legionella, accounting for 90% of all reported cases of legionellosis in the United States (Fields et al., 2002). L. pneumophila serogroup 1 accounts for the majority of these (Rantakokko-Jalava and Jalava, 2001). Other species of Legionella that have been implicated in disease include Legionella dumoffii, Legionella micdadei, Legionella bozemanii, and Legionella longbeachae; however, they occur at much lower frequencies. As natural inhabitants of aquatic environments, legionellae are ubiquitous. However, their presence can be particularly problematic when found in the water distribution systems of healthcare facilities. Of particular concern is the risk of nosocomial and community-acquired infection of the immune compromised, the elderly, and individuals with underlying respiratory conditions via aerosolization of water from contaminated water supplies.

Currently, culture is considered to be the standard technique for the identification of Legionella. However, a single ideal test for Legionella does not exist. The fastidious nature of the organism and the prolonged incubation periods necessary for growth can make isolation and identification challenging (Ng et al., 1997) for the diagnostic laboratory. In its clinical manifestations, a pneumonia caused by Legionella is not always distinguishable from pneumonias of other bacterial etiologies (Mulazimoglu and Yu, 2001); this lack of specific symptoms can contribute to delay in isolation and identification of the causal agent. Rapid diagnosis is important because legionellosis can result in case fatality rates exceeding 15% to 20%, and patients with severe underlying disease are at particular risk (Edelstein, 1995). When environmental water samples are analyzed, legionellae can be difficult to isolate from samples that are grossly contaminated with other organisms. In such a situation, the source of the nosocomial or community-acquired infection could go unidentified as increasing numbers of individuals fall sick.

Rapid methodologies used for the identification of Legionella include urine antigen detection, direct fluorescent antibody (DFA), and polymerase chain reaction (PCR). The most widely used urine antigen detection test, which is considered to be specific for L. pneumophila serogroup 1, can miss as many as 40% of legionellosis cases. As when DFA testing is used, the sensitivity from respiratory samples for the diagnosis of Legionnaires' disease has ranged from 25% to 75% (Fields et al., 2002), and specificity can be an issue.

A rapid, sensitive, and specific method for the detection of Legionella is clearly important to identify infected individuals and to expedite cleanup of contaminated water systems so as to prevent additional cases of infection. Our laboratory performs Legionella testing for numerous hospitals and facilities in New York State each year in the context of outbreak and remediation investigations. To assist in this testing, we have designed a novel multiplex Taqman® real-time PCR assay to detect both the Legionella genus and, specifically, L. pneumophila serogroups 1 to 16. Our multiplex assay is the 1st to use the 23S rRNA gene as a target to detect 42 species and 19 serogroups of Legionella; this design enables rapid, sensitive, and specific detection of Legionella spp. and L. pneumophila in a single reaction from both environmental and clinical samples. In addition, we describe and assess a testing algorithm that combines this assay with culture and 16S rRNA gene sequence analysis. Use of this algorithm resulted in markedly improved sensitivity and specificity for the clinical and environmental Legionella samples received by our laboratory.

Section snippets

Bacterial strains

The bacterial strains were supplied by the American Type Tissue Culture Collection (ATCC), the Centers for Disease Control and Prevention (CDC), and the New York State Department of Health Wadsworth Center Bacteriology Laboratory Culture Collection. All Legionella strains tested (Table 3) were grown on buffered charcoal yeast extract (BCYE) agar and incubated at 37 °C in 5% CO2. A 1.0-McFarland suspension was made of each isolate, and each was heat treated at 95 °C for 20 min. The additional

Optimization of multiplex real-time PCR assay

Our multiplex assay was initially evaluated as 2 separate assays, 1 for the mip gene and 1 for the 23S rRNA gene. Primer, probe, and MgCl2 titrations were performed. Optimal concentrations for each component were determined by testing replicate samples of a 10−4 dilution of L. pneumophila serogroup 1 and plotting the concentration of each component against the Ct value. Final concentrations of 4 mmol/L MgCl2, 900 nmol/L of each primer, and 250 nmol/L of each probe were established via this

Discussion

Legionella bacteria are associated with potentially fatal pneumonias and can be found as natural inhabitants of aquatic environments such as water distribution systems, cooling towers, and fountains. Investigation and remediation of Legionella outbreaks are essential to ensure the rapid identification of a new or recurring source, as well as to assess whether the problem has been resolved, so that patients or the community are no longer at risk of acquiring a Legionella infection.

To improve the

Acknowledgments

The authors acknowledge the Wadsworth Center Molecular Genetics Core Facility for the synthesis of oligonucleotides and Dr William Wolfgang and Dr Adriana Verschoor for critical readings of this manuscript.

References (17)

  • MulazimogluL. et al.

    Can Legionnaires disease be diagnosed by clinical criteria? A critical review

    Chest.

    (2001)
  • AkaneA. et al.

    Identification of the heme compound copurified with deoxyribonucleic acid (DNA) from bloodstains, a major inhibitor of polymerase chain reaction (PCR) amplification

    J. Forensic. Sci.

    (1994)
  • CloudJ.L. et al.

    Detection of Legionella species in respiratory specimens using PCR with sequencing confirmation

    J. Clin. Microbiol.

    (2000)
  • CorlessC.E. et al.

    Contamination and sensitivity issues with a real-time universal 16S rRNA PCR

    J. Clin. Microbiol.

    (2000)
  • EdelsteinP.H.

    Improved semiselective medium for isolation of Legionella pneumophila from contaminated clinical and environmental specimens

    J. Clin. Microbiol.

    (1981)
  • EdelsteinP.H.

    Antimicrobial chemotherapy for legionnaires' disease: a review

    Clin. Infect. Dis.

    (1995)
  • FieldsB.S. et al.

    Legionella and Legionnaires' disease: 25 years of investigation

    Clin. Microbiol. Rev.

    (2002)
  • HarrisonT.G. et al.

    A Laboratory Manual for Legionella

    (1988)
There are more references available in the full text version of this article.

Cited by (84)

  • Occurrence of opportunistic pathogens in private wells after major flooding events: A four state molecular survey

    2022, Science of the Total Environment
    Citation Excerpt :

    Detection of shiga toxin-producing E. coli (stx1 and stx2 genes) were determined using PCR. Primers, reagents, standards, and thermocycling settings are described in detail in the supplementary information (Table SI-1) (Nazarian et al., 2008; Radomski et al., 2010; Wilton and Cousins, 1992; Mull et al., 2013; Fagan et al., 1999; Suzuki et al., 2000). Serially-diluted genomic DNA standards (from 108 to 102 gene copies (gc) per reaction for 16S rRNA and from 106 to 5 gc per reaction for OPs) were included in each qPCR run.

  • Residential water heater cleaning and occurrence of Legionella in Flint, MI

    2020, Water Research
    Citation Excerpt :

    Environmental DNA was extracted from filter concentrated (as before) samples using a commercially-available DNA extraction kit (FastDNA Spin Kit, MP Biomedicals). Legionella spp. (23S rRNA) and L. pneumophila (mip) were enumerated by quantitative PCR (qPCR) using previously published methods (Nazarian et al., 2008; Wang et al., 2012). Though qPCR detects and quantifies both live and dead cells, causing potential false positive results, relative comparisons between levels occurring at the point of entry to a home and at individual outlets can be used as an indicator for potential regrowth (e.g., Rhoads et al., 2016).

View all citing articles on Scopus
View full text