|Home | About | Journals | Submit | Contact Us | Français|
We compared the performances of three recently optimized real-time PCR assays derived from distinct genomic regions of Mycoplasma pneumoniae during an outbreak. Comprehensive evaluation established that a newly described toxin gene represents a superior target for detecting M. pneumoniae DNA in clinical specimens, although use of multiple targets may increase testing confidence.
Mycoplasma pneumoniae accounts for approximately 15% to 20% of all community-acquired pneumonia cases and is a common cause of outbreaks (10, 16, 18). Outbreaks have been reported to occur in 3- to 7-year intervals, and although all age groups are susceptible, incidence rates vary with age and may occur more frequently in certain settings (10, 16, 17). M. pneumoniae infection spreads efficiently within households and close living quarters, with incubation periods as long as 3 weeks (17). The insidious nature of this infection and its protracted disease course make this agent a predominant cause of “walking pneumonia” that can persist within the population and cause community or institutional outbreaks. Upper- and lower-respiratory-tract symptoms are often mild, resulting in tracheobronchitis, headache, and cough. Occasionally, severe cases with extrapulmonary involvement can result in hospitalization and death due to neurological disease, such as encephalitis (1, 2, 4, 15, 17). The high rate of morbidity and the occasional mortality reinforces the need for timely diagnosis for administering proper antibiotic treatment (7, 9).
Conventional tests for detecting M. pneumoniae are fraught with limitations (3). M. pneumoniae culture can often take several weeks, requires special media and expertise, and is insensitive and prone to contaminants and inhibitors. Serological assays such as complement fixation and commercially available immunoglobulin detection kits are by nature retrospective, requiring paired serum samples from both acute and convalescent phases, and provide questionable specificity and sensitivity results. In sum, these approaches are impractical for a rapid diagnosis. A variety of nucleic acid-based tests based upon PCR have been developed for the rapid and sensitive detection of M. pneumoniae (5, 11, 13, 14, 16). The range of variables within each PCR study (specimen type, nucleic acid extraction and amplification procedures, target selection, definitions used in calculating data, etc.) makes it difficult to compare results and draw a single, comprehensive approach for reliable detection.
Recent community outbreaks of M. pneumoniae infection underscore a need among public health departments and local hospitals for a rapid and reliable diagnostic assay (1, 10, 12, 17, 18). Moreover, this test should be highly specific and sensitive and should be evaluated in an outbreak setting. The aim of the current study was to evaluate the use of three recently optimized real-time PCR assays for the detection of M. pneumoniae in respiratory samples from a recent outbreak. To our knowledge, this is the first prospective and comparative study of real-time PCR targets used to identify cases during an outbreak investigation of M. pneumoniae and the first report of a study targeting the recently identified ADP-ribosylating toxin gene encoding the CARDS (community-acquired respiratory distress syndrome) toxin for real-time PCR detection.
Multiple TaqMan primer-probe sets targeting the ATPase gene (GenBank accession no. U43738) and the CARDS toxin gene (GenBank accession no. DQ447750) of M. pneumoniae were designed using Primer Express version 3.0 (Applied Biosystems, Foster City, CA) (8). The real-time PCR mixture was prepared in a total volume of 25 μl. Each PCR mixture contained the following per reaction: 12.5 μl of Platinum quantitative PCR SuperMix-UDG (catalog no. 11730-025; Invitrogen), 1.5 μl of 50 mM MgCl2, 0.5 uM final concentrations of each primer, a 0.1 uM final concentration of the probe, 1.25 U of Platinum Taq DNA polymerase (catalog no. 10966-034; Invitrogen) (5 U/μl), 1 μl of 10 mM PCR nucleotide mix (catalog no. C1141; Promega), 5 μl of extracted nucleic acid from each specimen, and nuclease-free water (catalog no. P1193; Promega) to achieve a 25-μl final volume. Real-time PCR for each target was performed using an ABI 7500 system (Applied Biosystems) under the following conditions: initial activation of 95°C for 2 min, followed by 45 cycles of 95°C for 10 s and 60°C for 30 s. After thorough analysis and evaluation of the primer-probe sets, which involved BLAST searches, one toxin (Mp181) and two ATPase (Mp3 and Mp7) gene targets were selected based upon specificity and sensitivity performance and optimized to achieve maximum efficiency (Table (Table1).1). Each assay demonstrated >99% efficiency, as calculated using a standardized dilution series of quantitated DNA samples of M. pneumoniae tested using six replicates over six logs (100 pg to 1 fg). The average from these data is reported as the square of the coefficient of regression values (efficiency) in Table Table1.1. The sensitivity of each assay was determined by extracting a series of dilutions from quantitated stocks of both prototypical strains of M. pneumoniae (M129 [type I] and FH [type 2]) by use of a ChargeSwitch gDNA Mini Bacteria kit (catalog no. CS11301; Invitrogen) following the manufacturer's instructions. The CFU value per milliliter of each dilution was determined using a hemolytic plaque formation procedure as previously described (6). Each dilution was then tested with each assay in six replicates; values are reported in Table Table11 as total CFU detected within each real-time PCR. Use of Mp181 consistently detected between 1 and 5 CFU, while assays using both Mp3 and Mp7 were slightly less sensitive, detecting 5 to 50 CFU. Each assay was also tested for specificity with at least 15 ng of nucleic acid and showed no cross-reactivity with an extensive bacterial and viral pathogen panel consisting of the following targets: Mycoplasma faecium, Mycoplasma lipophilum, Mycoplasma salivarium, Mycoplasma pirum, Mycoplasma orale, Mycoplasma penetrans, Mycoplasma genitalium, Mycoplasma hominis, Mycoplasma fermentans, Mycoplasma buccale, Mycoplasma arginini, Mycoplasma hyorhinis, Mycoplasma amphoriforme, Lactobacillus planitarium, Staphylococcus epidermidis, Coxiella burnetii, Streptococcus salivarius, Bordetella pertussis, Legionella pneumophila, Legionella longbeachae, Streptococcus pneumoniae, Ureaplasma urealyticum, Neisseria meningitidis, Chlamydia trachomatis, Chlamydophila psittaci, Chlamydophila pneumoniae, Streptococcus pyogenes, Haemophilus influenzae, Neisseria elongata, Pseudomonas aeruginosa, Moraxella catarrhalis, Mycobacterium tuberculosis, Candida albicans, Escherichia coli, Staphylococcus aureus, Ureaplasma parvum, human DNA, human coronavirus, human rhinovirus, human parainfluenza virus 2, human parainfluenza virus 3, human adenovirus, influenza virus A, influenza virus B, respiratory syncytial virus A, and respiratory syncytial virus B.
These assays were used to identify a recent M. pneumoniae outbreak within a college setting. A total of 54 respiratory samples (oropharyngeal and nasopharyngeal swabs) from patients (n = 35) and negative controls (n = 19) 18 to 35 years of age (mean, 21.7) were tested in triplicate with each M. pneumoniae-specific assay (Mp3, Mp7, and Mp181) along with an RNase P internal control to ensure proper nucleic acid extraction and integrity. The data in Table Table22 demonstrate that 18 of 35 pneumonia cases (~51%), where a pneumonia case is defined as exhibiting fever (≥100.4°F) and cough or pneumonia diagnosed by chest X-ray or clinical examination during a 12-week outbreak period, were positive with all three signature sequences. Of note, the Mp181 assay routinely exhibited crossing-threshold values earlier than both the Mp7 and the Mp3 assays (P < 0.0001 and P = 0.06, respectively, following Student's t test), which is concordant with the sensitivity data of Table Table1.1. The crossing-threshold values ranged from ~26 to ~39, depending on the target, and displayed a typical sigmoidal curve similar to that seen with the positive controls, as did all RNase P assays (data not shown). All negative controls (defined as age-matched asymptomatic subjects within the same population) and samples from 16 patients demonstrated no reactivity with any of the M. pneumoniae-specific markers but gave positive RNase P signals (data not shown). The lack of PCR reactivity in the 16 cases may reflect the presence of an infection with a different pathogen or poor sample quality. Serum samples were collected from a limited number of subjects and proved to be of little value for diagnosis. Serological assays are often unreliable due to specificity and sensitivity limitations and the documented persistence of antibodies in patients (3).
This study evaluated three real-time PCR assays targeting the ATPase and newly described CARDS toxin genes during a recent outbreak for the purpose of assessing its utility in such instances. Although other real-time PCR assays for the detection of M. pneumoniae have been reported, none have been applied to an investigation involving an outbreak. Interestingly, the assay targeting the CARDS toxin gene (Mp181) proved to be the most sensitive in identifying positive specimens during this outbreak and has been subsequently used to positively identify M. pneumoniae DNA in other specimens (respiratory and cerebrospinal fluid) in sporadic cases. These data support the use of the Mp181 assay as an initial screening marker for detecting the presence of M. pneumoniae DNA in respiratory clinical specimens, although the inclusion of Mp3 and Mp7 may provide an increased level of confidence for the reporting of results. The use of these assays may allow the rapid identification of an M. pneumoniae outbreak at the local and state levels when testing is implemented in a timely manner. Further investigation of each assay may be warranted for possible use in clinical practice.
We thank Lauri Hicks, Nick Walter, Gavin Grant, and Utpala Bandy for providing specimens and helpful discussions surrounding this report. We also thank Lauri Hicks for critical review of the manuscript and Bernard J. Wolff for assistance with statistical analysis.
Published ahead of print on 9 July 2008.