INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the United States alone, approximately 12 million people have asthma, which results in health care costs of approximately 4.6 billion dollars annually (1). Yet the etiology and pathogenesis of this important disease remain poorly defined. Mycoplasma pneumoniae is a common cause of both upper and lower respiratory infection in humans, with tracheobronchitis being the most common clinical manifestation (2). The effects of infection with this organism can persist for months, resulting in decreased expiratory flow rates and increased airway hyperresponsiveness in nonasthmatic individuals (3).

Epidemiologic evidence exists linking Mycoplasma infection with asthma exacerbation, and possibly as a factor in the pathogenesis of asthma (4). These studies have demonstrated such an association by isolation of the organism in throat swabs and by serologic documentation, but not by lower airway analysis. If M. pneumoniae is a causal factor in chronic asthma, then this organism should be present in the lungs of certain individuals with stable, chronic asthma. The present study was undertaken to determine if M. pneumoniae can be detected in the upper and lower airways of adults with stable, chronic asthma as compared with nonasthmatic control subjects. Also evaluated was the presence of Chlamydia pneumoniae and seven common respiratory viruses, as they have also been implicated in the exacerbation of asthma (5, 6).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Thirty subjects were recruited via newspaper and radio advertisements from the general Denver, Colorado community over a period from June 1994 to April 1996, with subject testing among asthmatics and control subjects distributed relatively equally over the time period. The asthmatics fulfilled criteria for asthma exhibiting a provocative concentration of methacholine causing a 20% reduction in FEV₁ (PC₂₀) of < 8 mg/ml and reversibility of lung function by at least 15% with bronchodilator. Control subjects exhibited no evidence of atopy or bronchial hyperresponsiveness (PC₂₀ > 16 mg/ml). Exclusion criteria included inpatient status, upper or lower respiratory tract infection within 3 mo of study, use of macrolides, tetracyclines, or quinolones within 3 mo of study, smoking history greater than 5 pack-yr or any cigarettes within the previous 2 yr, significant nonasthma pulmonary disease, or other medical problems. No subject was a hospital worker to ensure that contact with potential patients infected with M. pneumoniae was minimized. Subject characteristics are shown in Table 1.

Study Protocol

Physiologic and radiographic evaluation. After informed consent of this institutional review board-approved protocol, subjects underwent spirometric evaluation and methacholine challenge. A chest radiograph was performed to rule out infiltrates. Measurement of total serum immunoglobulin levels (IgG and IgM) was also performed to ensure subjects were not hypogammaglobulinemic and were thus capable of mounting an appropriate serologic response to infection.

Specimen collectionupper airway, lower airway, and serology. Nasal epithelial cells were obtained by placing a Rhinoprobe (ASI, Arlington, TX) into the anterior nares. Oropharyngeal specimens were obtained from the posterior oropharynx with a wire-shafted dacron swab (Dacroswab; Spectrum Laboratories, Houston, TX). Specimens from each site were placed and agitated in tubes of SP4 Mycoplasma media for Mycoplasma culture and polymerase chain reaction (PCR). Separate specimens were placed into chlamydia transport media (1.5 ml of 2-sucrose phosphate buffer containing 20% fetal calf serum) for C. pneumoniae culture and in 1× PCR transport buffer for PCR. All specimens were frozen at 70° C until analysis.

Both asthmatics and control subjects underwent bronchoscopy as previously described (7) with endobronchial biopsy (BX), endobronchial protected brushing, and bronchoalveolar lavage (BAL) to obtain specimens for lower airway evaluation. The location of the biopsies was randomized to the right or left lower lobe followed in the opposite lung by brushing of the lower lobe and BAL in the right middle lobe or lingula to ensure the right and left lungs were biopsied, brushed, and lavaged equally in the group as a whole. Four BX were performed under direct visualization from the fourth- or fifth-generation airways. Three BX specimens were inoculated directly into the Mycoplasma and Chlamydia medias to be evaluated by culture and PCR (referred to as "media" specimens), and the fourth was embedded in paraffin for Mycoplasma PCR only (referred to as the "paraffin" specimen). The brushing of the proximal airways was performed under direct visualization using a separate protected cytologic brush for each pass. Cell count and differential were performed as previously described (7). BAL fluid was then concentrated, and the pellet was resuspended in 5 to 10 ml of sterile saline and distributed equally among the three vials. Blood for Mycoplasma and Chlamydia serology was sampled at the time of bronchoscopy.

Microbiological analysisculture, PCR, and serology: Mycoplasma. Culture for Mycoplasma was performed as previously described (8); growth was monitored daily for approximately 8 wk. For PCR analyses, we used multiple primer sets directed against organism-specific target sequences of either the P1 adhesion gene or the 16S ribosomal ribonucleic acid (rRNA) gene of M. pneumoniae which can detect approximately 100 genomes (9, 10). Detection of the P1 target may be more sensitive for fixed and embedded tissue (9, 10) so our embedded samples were analyzed using only the P1-specific primers. This primer set generates a small 103 bp product and was specifically designed to detect (possibly) fragmented DNA. The outer primers of the 16S rRNA PCR were those described by Williamson and coworkers (9). The primers for a semi-nested PCR consisted of the same sense primer from the outer set and an additional 27 bp reverse primer positioned four nucleotides internally to the 5' end of the outer reverse primer. Conditions for use of these latter, internal primers were as described (9) with the following modifications: Primer concentrations were 0.5 mm and cycling conditions consisted of an initial incubation at 94° C for 60 s, 60° C for 60 s, and 72° C for 90 s. All primers and PCR conditions used for detection of M. pneumoniae in these studies have been shown to detect 30 different clinical isolates of M. pneumoniae collected over a period of 10 yr from diverse international regions. Under the conditions utilized, none of the primers reacted with any other Mycoplasmas of human origin.

A primer pair that detects Mycoplasma species was also employed as previously described (11). Reaction conditions and the detection of amplified products were as described (9). Experimental positive controls (per 15 patient specimens) included the purified genomic DNA of the M. pneumoniae EATON reference stain as well as the deproteinized reference strain and a plasmid insert (see TA cloning in next paragraph) of the target of interest. Two negative controls that were carried through all steps were included for every 20 samples. Internal controls were employed to determine the presence of inhibitory factors. All patient samples were spiked with M. pneumoniae target sequence, then PCR was performed. If no signal was detected then the sample DNA was purified to remove inhibitors before analyzing without internal control added.

To verify that PCR positivity was a true indication of the presence of M. pneumoniae in the lower respiratory tract of patients, representative PCR products were directly ligated into the TA cloning vector (Invitrogen, San Diego, CA). The cloned plasmid was isolated (Mini-Prep plasmid preparation kits; Promega Corp., Madison, WI) and was sequenced using the Sanger dideoxy chain termination method (Sequenase Version 2.0 DNA sequencing kit; United States Biochemical, Cleveland, OH). Sequence analysis was performed using the GCG (Genetics Computer Group) program (University of Wisconsin, Madison).

Respiratory specimens were also evaluated by an indirect antigen capture assay for M. pneumoniae. The validation of the EIA for detection of M. pneumoniae has been reported previously (12). Finally, the IgG and IgM serologic response to M. pneumoniae was determined using an enzyme-linked immunosorbent assay (ELISA) which has previously been described (13).

Microbiological analysisChlamydia. Chlamydia cultures and PCR were performed as previously described with a sensitivity of 76.5% as compared with culture and direct fluorescent antigen (14). IgG and IgM antibodies to C. pneumoniae were detected by the microimmunofluorescent (MIF) test (14). Serologic evidence of infection with C. pneumoniae was defined as an IgG titer 1:512 or a fourfold rise and an IgM titer of  1:16.

Microbiological analysisrespiratory viruses. A composite enzyme immunoassay (EIA), using 8-well microstrips, was used for the detection of seven respiratory viruses (influenza A and B; parainfluenza 1, 2, and 3; adenovirus; and respiratory syncytial virus) as previously described (15).

Statistical Analysis

The asthmatics and control groups were compared for the presence or absence of Mycoplasma or Chlamydia infection and the presence of positive serology (dichotomous variable) using Fisher exact test (16). Continuous variables were compared using the t test or Wilcoxon test, depending on distribution of the data. Data are presented as means ± SEM or medians with interquartile range (IQ). For all analyses, all tests were two-sided with the level of significance defined as p  0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Upon entry into the study, no chest radiographs revealed infiltrates. IgG and IgM concentrations were within normal limits without significant differences between asthmatics and normals (p > 0.05). IgE was elevated in the asthmatics (66 IU/ml, [26, 202 IQ]) as compared with normals (6 IU/ml, [2, 6 IQ]; p = 0.01).

The BAL fluid return was higher in the control subjects, with a mean recovery of 62.8 ± 4.1% in controls and 46.1 ± 4.2% in the asthmatics (p = 0.02). This increased return in control subjects is seen in all studies. The median total cell counts were 10.0 × 10⁶ [7.8-19.5, IQ] in the control subjects and 9.5 × 10⁶ [6.3-13.2, IQ] in the asthmatics. When corrected for BAL fluid return, the control cell count/ml was 5.4 × 104/ ml [4.2-10.5, IQ] and the asthmatic cell count/ml was 6.8 × 104/ml [4.6-9.6, IQ] (p = NS).

Mycoplasma

M. pneumoniae detected by PCR is presented in two ways. A conservative approach is to demand positivity at two sites or with two primers at one site, and a more general approach is to present all data regarding PCR positivity. Conservatively, 10 of 18 asthmatics and 1 of 11 control subjects were PCR-positive for M. pneumoniae (p = 0.02). The more liberal approach increased the positivity to 11 of 18 asthmatics and in 2 of 11 control subjects (p = 0.02). Importantly, 9 of 10 subjects positive for M. pneumoniae by the conservative approach were positive in the lower respiratory tract and BAL or biopsy were positive in nine of these patients (Figure 1). The distribution of the PCR-positive patients was as follows: 3 of 11 nasal, 4 of 11 throat, 8 of 11 BAL, 4 of 11 BX media specimens, and 4 of 11 paraffin-embedded biopsy specimens. A summary of the PCR results in the asthmatics and control subjects by site and primer is presented in Table 2. Seven of 18 asthmatics and 1 of 11 control subjects were also positive for M. fermentans and M. genitalium by PCR. Of the seven, five were also positive for M. pneumoniae. In contrast, nonpathogenic oral commensal Mycoplasma species were recovered by culture only from the throats of 7 of 11 normal and 11 of 18 asthmatic patients. Nonpathogenic mycoplasmas were recovered by culture only from the lower respiratory tract of 1 of 11 normal and 5 of 18 asthmatic patients. All cultures and EIAs for M. pneumoniae were negative in both asthmatic and control subjects. No subjects were antibody-positive.

View larger version (18K): [in this window] [in a new window]   Figure 1.   The number of asthmatic subjects positive for M. pneumoniae, M. fermentans, and M. genitalium by PCR in the lower airway (Lower Aw), upper airway (Upper Aw), lower airway only (Only Lower Aw), and upper airway only (Only Upper Aw). Black bars represent the number of subjects who were PCR-positive for M. pneumoniae; white bars represent the number of subjects positive for M. fermentans and M. genitalium. BAL = bronchoalveolar lavage; BX = biopsy.

Chlamydia

Culture and PCR for C. pneumoniae were negative for all asthmatic and control subjects. Nine of the 18 asthmatics and 1 of 11 control subjects were serologically positive for C. pneumoniae (p = 0.05). Seven of the asthmatics were IgM-positive, and seven were IgG-positive (not the same group). The one control subject was IgM-positive.

Respiratory Viruses

The EIA for seven respiratory viruses was negative in all asthmatic and control patients.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study illustrates that Mycoplasma pneumoniae is present in the airways by PCR in greater than 55% of the asthmatics studied, with seven asthmatic subjects and one control subject exhibiting PCR evidence for M. genitalium and M. fermentans. Previously, there has been no systematic evaluation of Mycoplasma presence in the lower airways in patients with chronic asthma. In the present study, we demonstrated M. pneumoniae by PCR but not culture in the respiratory tracts of adults with chronic asthma. Our failure to culture the organism might be explained by its extreme fastidiousness and/or its presence in low numbers. Culture is the least sensitive of the methods used in this study for detection of M. pneumoniae. However, the culture methods used in the present study have been utilized by us in the evaluation of > 2,000 respiratory specimens during the past 5 yr in patients with radiographically confirmed, community-acquired pneumonia and have resulted in recovery of M. pneumoniae by culture in up to 17% of patients (17). Failure to detect M. pneumoniae by culture is not explained by lidocaine used to anesthetize the airways as M. pneumoniae was not inhibited in vitro by lidocaine (18).

PCR positivity can be present for longer periods than culture or seropositivity (19). Guinea pigs experimentally infected with M. pneumoniae became chronically infected as detected by PCR for up to 200 d, but are culture-negative by 70 d. Also, by this time, antibody levels become negative (19). Thus, our asthmatic patients appear to be chronically infected or colonized with M. pneumoniae despite culture negativity, as they are PCR-positive. If their positivity was only due to persistence after an infection that was present in the community, we would have expected a positive antibody response, and a greater incidence of PCR positivity in control subjects. In addition, we know of no significant rise in the number of infections due to M. pneumoniae in the 12 mo prior to beginning the study. However, infection with M. pneumoniae is known to occur in 5-year epidemic cycles, thus the sampling was done throughout the study period equally in both asthmatics and control subjects to avoid this potential confounder. Furthermore, that positive PCR is truly reflective of the lower respiratory tract by M. pneumoniae is supported by the fact that nine of the 10 patients positive for M. pneumoniae were positive in the BAL and bronchial biopsies. Use of multiple primer pairs as well as confirmation of PCR findings in two different laboratories (9) attest to the validity of the PCR results as well as the presence of only one (conservative approach) or two positives among age-matched control patients. Finally, the data regarding M. pneumoniae culture positivity in the literature are from subjects experiencing acute Mycoplasma infection clinically. In that setting, culture positivity is only 17% as discussed earlier. Our subjects do not have evidence of acute infection, suggesting a very different setting where extrapolation to data on acute infection cannot be done.

The lack of antibody response to M. pneumoniae has been a response previously noted by us in both pediatric and adult populations with community-acquired pneumonia. In ambulatory children or adults, up to 6% with radiographically confirmed pneumonia in the absence of other etiologic agents are culture- and/or PCR-positive but seronegative (17). In two different studies of over 250 hospitalized adults, the number culture-positive and seronegative (28 of 258, 11%) was significantly higher than in nonhospitalized adults (2 of 54, 3.7%) (17). These results indicate a subset of infected individuals do not mount an antibody response perhaps owing to genetic differences (17). Lack of a response may in fact contribute to the organism's persistence (17).

For C. pneumoniae, the lack of correlation between serology results and culture and PCR was not an unexpected result. We have previously seen discordance between culture and serologic results in patients with community-acquired pneumonia, with several possibilities raised as explanations for the discrepancy (20). First, criteria posed for the serologic diagnosis of C. pneumoniae, particularly a single, high IgG titer, are arbitrary and may overestimate the incidence of C. pneumoniae. Limitations also exist with the MIF test as antibodies to the lipopolysaccharide antigen may cross react between species of the Chlamydia genus (20). Although prior studies have linked C. pneumoniae to acute asthma exacerbations (21), we could not find this organism in the nasopharynx or lung. Thus, the exact significance of the positive serology in nine asthmatics and one control subject is uncertain but needs further investigation.

We have demonstrated that asthmatic patients exhibit evidence of M. pneumoniae colonization/infection with significantly greater frequency than nonasthmatic subjects. Further study is needed to assess if the presence of M. pneumoniae is an epiphenomenon due to enhanced airway inflammation or as we hypothesize, a pathogenic mechanism in asthma. If the latter is correct, greater evaluation of the involved process is needed to further our understanding of the pathogenesis and treatment of asthma.