Clinical Application of a Multiplex Real-Time PCR Assay for Simultaneous Detection of Legionella Species, Legionella pneumophila, and Legionella pneumophila Serogroup 1

Abstract

We developed a single-tube multiplex real-time PCR assay capable of simultaneously detecting and discriminating Legionella spp., Legionella pneumophila, and Legionella pneumophila serogroup 1 in primary specimens. Evaluation of 21 clinical specimens and 115 clinical isolates demonstrated this assay to be a rapid, high-throughput diagnostic test with 100% specificity that may aid during legionellosis outbreaks and epidemiologic investigations.

TEXT

Legionellae are facultative intracellular Gram-negative bacteria found in soil and water environments, where they parasitize and proliferate within free-living protozoa (1). Legionellae possess the ability to replicate in environmental protozoa and within mammalian alveolar macrophages and epithelial cells (2–4). Legionellae are common contaminates of artificial water systems, including air-conditioning systems, cooling towers, and jacuzzis, where conditions may be optimal for growth and proliferation (5, 6). Once aerosolized, the bacteria can enter the human respiratory tract and cause disease manifesting as Legionnaires' disease, a severe form of pneumonia, or Pontiac fever, a self-limiting flu-like illness (1).

More than 50 species comprising 70 distinct serogroups have been identified to date (http://www.bacterio.cict.fr/l/legionella.html). Although more than 90% of isolates associated with Legionnaires' disease are Legionella pneumophila, with 84% being L. pneumophila serogroup 1 (sg1), nearly one-half of Legionella species have been associated with human disease (1, 7, 8). Given this epidemiological background, an ideal diagnostic assay would rapidly detect the presence of Legionella and simultaneously identify the species and, if applicable, the serogroup. Although bacterial culture remains the gold standard for the diagnosis of Legionnaires' disease, it requires specialized media and expertise and is time-consuming, requiring several days to obtain definitive and reliable results. The urine antigen test is a common and convenient test used for diagnosis but detects only L. pneumophila sg1. Serology-based tests, although widely used, suffer from specificity limitations (1). Currently, molecular diagnosis is based largely on detection of the 16S rRNA gene for identification of the Legionella genus and the mip gene, used for identification to the species level, along with sequence based-typing (SBT) for L. pneumophila characterization (9–11). These tests require post-PCR analysis and/or sequencing to distinguish L. pneumophila sg1 from other serogroups and/or non-pneumophila species of Legionella, thus making them less ideal (12, 13). The lack of available diagnostic tests for species other than L. pneumophila may lead to underreporting and/or unrecognized cases of legionellosis. Since the worldwide burden of disease caused by non-sg1 L. pneumophila and other species is largely unknown but is suspected to be significant (14, 15), an improved detection method would ideally integrate a more inclusive approach for identification and characterization.

This article reports the development and evaluation of a single-tube multiplex real-time PCR assay that allows simultaneous detection and differentiation of Legionella spp., Legionella pneumophila, and Legionella pneumophila sg1 for clinical isolates and in primary specimens. The assay targets the ssrA (Legionella species) gene (16), the mip (L. pneumophila) gene, and the recently identified wzm gene (17), specific for L. pneumophila sg1.

Multiple TaqMan primer-probe sets were designed targeting the ssrA gene (GenBank accession no. AE017354 [bp 172917 to 173015]), the mip gene (GenBank accession no. AJ810179 [bp 76 to 190]), and the wzm gene (GenBank accession no. AM778127 [complement bp 6098 to 6167]) either manually (ssrA) or using the software program Primer Express 3.0 (Applied Biosystems, Foster City, CA). The RNase P assay designed to detect human DNA was described previously (18). Primers and probe sets were initially tested and optimized in singleplex format (data not shown). Primer and probe sequences, their final concentrations, and the distinct fluorophores for each probe used in the multiplex assay are listed in Table S1 in the supplemental material. Purified total nucleic acid from L. pneumophila sg1 strain Philadelphia was used as a positive control. The multiplex real-time PCR mixture was prepared in a total volume of 25 μl containing the following, per reaction: 12.5 μl of PerfeCTa Multiplex qPCR SuperMix (Quanta Biosciences, Gaithersburg, MD), the appropriate volume of each primer and probe (see Table S1), 5 μl of template, and nuclease-free water (Promega), to achieve a 25-μl final volume. The assay was carried out using the ABI 7500 real-time PCR system (Applied Biosystems) under the following conditions: 95°C for 5 min, followed by 45 cycles of 95°C for 15 s and 60°C for 1 min.

Analytical specificity of the multiplex assay was 100% (data not shown), as verified using a comprehensive panel of 215 clinical and environmental isolates representing 52 Legionella species and 44 non-Legionella strains (see Table S2 in the supplemental material). Analytical sensitivity was established by testing L. pneumophila sg1 nucleic acid diluted from 1 ng/μl, followed by 10-fold dilutions down to 1 fg/μl. Each dilution was tested in 10 replicates, and limits of detection (LOD) were established for each assay, defined as the lowest dilution in which ≥50% of replicates had positive crossing threshold (C_T) values. An LOD of 25 fg was established for each assay, except for the RNase P assay, which had an LOD of 25 pg. Figure 1 presents real-time PCR analysis curves of samples spiked with 500 pg of Legionella anisa (Fig. 1A), L. pneumophila sg2 (Fig. 1B), L. pneumophila sg1 (Fig. 1C), and human nucleic acid. Samples were run in duplicate with all channels shown to display assay specificity and the lack of cross-channel signal bleed.

Fig 1.

Multiplex assay for detection of Legionella spp., Legionella pneumophila, Legionella pneumophila sg1, and human DNA. (A) L. anisa DNA/human DNA; (B) L. pneumophila sg2 DNA/human DNA; (C) L. pneumophila sg1 DNA/human DNA. Samples were run in duplicate. All channels are shown to display specificity. Delta Rn, baseline-corrected fluorescence.

Clinical specimens previously collected by the CDC Legionella Lab from individual cases and/or domestic outbreaks and identified as positive (n = 15) or negative (n = 6) for Legionella by culture and direct immunofluorescence assay (DFA) were tested with the current assay. Culture-positive clinical specimens were limited to L. pneumophila sg1 positives due to limited availability of other Legionella-positive specimens. Total nucleic acid from each clinical specimen or culture was extracted with the MagNA Pure compact instrument (Roche Applied Bioscience, Indianapolis, IN) using the total nucleic acid isolation kit, following the manufacturer's instructions. Clinical sensitivity was established by testing nucleic acid extracts in triplicate using the multiplex assay, which included an RNase P internal control to ensure proper nucleic acid extraction and integrity. Table 1 shows the average C_T values obtained from each clinical specimen extract and demonstrates 100% clinical sensitivity. All negative specimens demonstrated no reactivity with any of the Legionella-specific markers but gave positive RNase P signals (Table 1). Total nucleic acid extracted from clinical isolate cultures (n = 115) previously characterized in our lab by mip sequencing and/or DFA was normalized to 1 ng/μl and tested using the multiplex assay for specificity (Table 2). These data demonstrate assay specificity and sensitivity and provide an overall improvement in the diagnostic capability for detecting and characterizing causes of legionellosis by demonstrating a sensitive, high-throughput, and rapid testing procedure.

Table 1.

Multiplex real-time analysis of Legionella-positive and -negative clinical specimens

Sample no.	Sourcec	Culture result	DFA	Multiplex real-time PCR CT value
				Legionella spp. (ssrA)	L. pneumophila (mip)	L. pneumophila sg1 (wzm)	Human DNA (RNase P)
1	Sputum	+	sg1	27.63	26.84	27.74	25.96
2	Lung	+	sg1	17.15	16.85	17.11	17.66
3	Lung	+	sg1	15.14	14.97	15.19	16.56
4	Br. swab	+	sg1	25.53	24.47	24.53	23.91
5	Lung	+	sg1	26.89	26.18	25.95	20.00
6	Lung	+	sg1	28.55	28.07	26.98	17.66
7	Lung	+	sg1	25.19	24.37	24.17	19.59
8	Lung	+	sg1	25.93	25.30	24.72	19.05
9	Lung	+	sg1	29.36	28.74	28.27	20.50
10	Lung	+	sg1	26.04	25.18	25.15	17.50
11	Lung	+	sg1	26.54	25.97	25.81	21.24
12	Lung	+	sg1	26.72	27.07	26.52	19.83
13	BAL	+	sg1	33.17	32.13	31.97	24.70
14	Sputum	+	sg1	27.88	27.21	26.66	16.75
15	Lung	+	sg1	28.18	27.24	26.95	20.48
16	OP swab	−	NTb	NDc	ND	ND	24.11
17	Urine	−	NT	ND	ND	ND	23.07
18	OP swab	−	NT	ND	ND	ND	25.17
19	NP swab	−	NT	ND	ND	ND	24.56
20	Sputum	−	NT	ND	ND	ND	22.26
21	Sputum	−	NT	ND	ND	ND	23.98

^aBr. swab, OP swab, and NP swab, bronchial, oropharyngeal, and nasopharyngeal swabs, respectively; BAL, bronchoalveolar lavage specimen.

^bNT, not tested.

^cND, not detected.

Table 2.

Multiplex real-time analysis of Legionella clinical isolates

Species (na)	Multiplex real-time PCR CT value
	Legionella spp. (ssrA)		L. pneumophila (mip)		L. pneumophila sg1 (wzm)
	Avg (SD)	Range	Avg (SD)	Range	Avg (SD)	Range
Legionella species, non-pneumophila (13)	21.84 (0.85)	20.64–23.76	NDb		ND
L. pneumophila sg2 to sg14 (42)	20.59 (0.68)	18.11–21.70	19.19 (0.63)	17.01–20.36	ND
L. pneumophila sg1 (60)	19.07 (0.63)	18.04–20.25	17.97 (0.60)	17.04–19.13	19.11 (0.59)	17.95–20.10

^an, no. of isolates.

^bND, not detected.

To our knowledge, our study is the first report of a single-tube, 4-plex real-time PCR assay capable of simultaneous detection and differentiation of Legionella spp., Legionella pneumophila, and Legionella pneumophila sg1, along with an internal control (RNase P) in clinical specimens. The multiplex assay is considerably faster (4 h versus ≥8 days) and easier to interpret than culture. In addition, it allows direct quantification of Legionella. Although this study was limited to Legionella pneumophila sg1-positive clinical specimens, the assay affords the opportunity to expand the diagnostic capabilities to detect species other than L. pneumophila sg1, thereby yielding a better understanding of the causative agent of legionellosis. The assay does not require an isolate, thereby providing an advantage in cases where antibiotics may have been administered that may render the Legionella nonviable but allow for the potential to detect any remaining nucleic acid present. Although further testing would be required, this assay could potentially be used for surveillance purposes or to screen environmental samples to assist in identifying a source during an outbreak. Use of this test along with currently established diagnostic tests for legionellosis may lead to a more effective public health response and expedite appropriate treatment of patients, especially during outbreaks.