Detection of Mycoplasma pneumoniae in Spiked Clinical Samples by ...

Download PDF
2 downloads 0 Views 418KB Size Report
Comment
Oct 11, 2001 - Isothermal nucleic acid sequence-based amplification (NASBA) was applied to the detection of Mycoplasma pneumoniae. M. pneumoniae ...
JOURNAL OF CLINICAL MICROBIOLOGY, Apr. 2002, p. 1339–1345 0095-1137/02/$04.00⫹0 DOI: 10.1128/JCM.40.4.1339–1345.2002 Copyright © 2002, American Society for Microbiology. All Rights Reserved.

Vol. 40, No. 4

Detection of Mycoplasma pneumoniae in Spiked Clinical Samples by Nucleic Acid Sequence-Based Amplification K. Loens,1* D. Ursi,1 M. Ieven,1 P. van Aarle,2 P. Sillekens,2 P. Oudshoorn,2 and H. Goossens1 Department of Microbiology, University of Antwerp UIA, Antwerp, Belgium,1 and Organon Teknika BV, Boxtel, The Netherlands2 Received 20 August 2001/Returned for modification 11 October 2001/Accepted 24 January 2002

Isothermal nucleic acid sequence-based amplification (NASBA) was applied to the detection of Mycoplasma pneumoniae. M. pneumoniae RNA prepared from a plasmid construct was used to assess the sensitivity of the assay, and an internal control for the detection of inhibitors was constructed. The sensitivity of the NASBA assay was 10 molecules of wild-type M. pneumoniae RNA generated in vitro and 5 color-changing units (CCU) of M. pneumoniae. An appropriate specimen preparation procedure was developed: after protease treatment of the respiratory specimens, guanidine thiocyanate lysis solution (4.7 M guanidine thiocyanate [Sigma-Aldrich NV], 46 mM Tris-HCl [Merck, Darmstadt, Germany], 20 mM EDTA [Sigma-Aldrich NV], 1.2% [wt/vol] Triton X-100 [Sigma-Aldrich NV], pH 6.2.) was added. With spiked throats, nasopharyngeal aspirates, bronchoalveolar lavage specimens, and sputum specimens, the sensitivity of the NASBA assay in the presence of the internal control was 2 ⴛ 104 molecules of in vitro-generated RNA or 5 CCU of M. pneumoniae. The sensitivity of the NASBA assay was comparable to that of a PCR targeted to the P1 adhesin gene. Fifteen clinical specimens positive for M. pneumoniae by PCR were also positive by NASBA. These results indicate that the sensitivity of detection of M. pneumoniae in spiked respiratory samples by NASBA is high. Together with the use of the internal control, the assay merits evaluation as a diagnostic tool. Mycoplasma pneumoniae is a common etiologic agent of respiratory tract infections in humans and is responsible for 15 to 20% of all cases of pneumonia (12) and a wide range of mild to serious extrapulmonary complications (8, 18). In the past, diagnosis of infection with this organism was usually based on serology because culture is slow and insensitive (14, 22). Therefore, nucleic acid amplification techniques have been introduced. PCR of fragments of the P1 gene or the 16S rRNA gene was shown to be considerably more sensitive than culture for the detection of M. pneumoniae (9, 17, 20, 39). Amplification methods often lack appropriate controls. A human ␤-globin-specific amplification may be used to assess the presence of nucleic acids in the processed sample (1, 24, 31). For the detection of inhibitors, the use of an internal control (IC) to be amplified with the same primer set as the target sequence is straightforward since it avoids the use of different primer sets. ICs are now gradually being more widely used (10, 16, 19, 30, 41). Nucleic acid sequence-based amplification (NASBA; Organon Teknika, Boxtel, The Netherlands) is targeted at RNA. It makes use of the simultaneous enzymatic activities of avian myeloblastosis virus reverse transcriptase (AMV-RT), RNase H, and T7 RNA polymerase under isothermal conditions. One advantage of NASBA compared with PCR is that it is a continuous, isothermal process which does not require a thermocycler. The constant temperature maintained throughout the amplification reaction allows each step of the reaction to proceed as soon as an amplification intermediate becomes available. Thus, the exponential kinetics of the NASBA process,

which are caused by multiple transcription of RNA copies from a given DNA product, are intrinsically more efficient than DNA amplification methods, which are limited to binary increases per cycle (38). The products of NASBA are single stranded and thus can be applied to detection formats that use probe hybridization without any denaturation step. Furthermore, the detection of microorganisms by an rRNA-based amplification technique might be more sensitive than PCR because of the presence of multiple RNA copies, and it also implies biological activity. It may be a useful complement to culture in order to establish if the M. pneumoniae infection is productive or to follow an antibiotic therapy. NASBA also has some disadvantages. NASBA is an RNA amplification procedure. RNA integrity and amplification inhibitors are the main causes of concern for NASBA, RT-PCR, and other RNA amplification procedures as well. The stability of the RNA may be affected during collection, processing, and storage of specimens prior to isolation. The addition of RNase inhibitors to the clinical specimens, such as guanidine thiocyanate (GuSCN), is required to preserve RNA integrity. The specificity of the reactions might be lower. The enzymes used are not thermostable, and the reaction temperature may not exceed 42°C without compromising the reaction. However, the specificity is increased by additional hybridization with targetspecific probes by enzyme-linked gel assay (ELGA), electrochemiluminescence detection, or even real-time detection. Furthermore, the length of the amplified RNA target sequence should be in the range of 120 to 250 nucleotides. Shorter and longer sequences will be amplified less efficiently. This might be more important for RNA amplification assays. The NASBA technique has already been successfully applied for the detection of human immunodeficiency virus type 1 (HIV-1) (21), human cytomegalovirus (13), citrus tristeza virus (23), human papillomavirus (36), human hepatitis C

* Corresponding author. Mailing address: Department of Medical Microbiology, University of Antwerp, Universiteitsplein 1 S3, B-2610 Wilrijk, Belgium. Phone: 32-3-820-25-51. Fax: 32-3-820-26-63. E-mail: [email protected]. 1339

1340

LOENS ET AL.

J. CLIN. MICROBIOL. TABLE 1. Bacterial species and strains

Isolate no.

Species

Strain or source

Origina

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

Mycoplasma pneumoniae type 1 Mycoplasma pneumoniae type 2 Mycoplasma fermentans Mycoplasma hominis Mycoplasma genitalium Mycoplasma orale Mycoplasma buccale Mycoplasma salivarium Mycoplasma pirum Ureaplasma urealyticum Legionella pneumophila Chlamydia pneumoniae Moraxella catarrhalis Haemophilus influenzae Streptococcus pneumoniae Streptococcus pyogenes Viridans group streptococci Staphylococcus aureus Klebsiella pneumoniae Escherichia coli Neisseria meningitidis Pseudomonas aeruginosa

ATCC 29085 (PI 1428) ATCC 15492 (MAC) NC 10117 NC10111 ATCC 33530 (G-37) NC10112 NC10136 NC10113 NC11702 Clinical isolate Clinical isolate TW-183 Clinical isolate Clinical isolate Clinical isolate Clinical isolate Clinical isolate Clinical isolate Clinical isolate Clinical isolate Clinical isolate Clinical isolate

ATCC ATCC NTCC NCTC ATCC NCTC NCTC NCTC NCTC UZA UZA WRF UZA UZA UZA UZA UZA UZA UZA UZA UZA UZA

a ATCC, American Type Culture Collection, Manassas, Va.; NCTC, National Collection of Type Cultures, Central Public Health Laboratory, London, United Kingdom; UZA, Universitair Ziekenhuis Antwerpen, Edegem, Belgium; WRF, Washington Research Foundation, Seattle, Wash.

virus (34), malaria parasites (37), Chlamydia trachomatis (25), Campylobacter jejuni (42), and Mycobacterium leprae (44) and for the detection and identification of Mycobacterium avium and Mycobacterium tuberculosis (43). We previously described the use of NASBA for the typing of M. pneumoniae strains and isolates (27). In the study described here we used the NASBA technique for the detection of M. pneumoniae RNA, constructed an IC for the assay, optimized the sample preparation procedure for detection of M. pneumoniae RNA in clinical specimens, and compared its performance with that of PCR on a number of clinical samples. MATERIALS AND METHODS Bacterial strains. The bacterial strains used to test the specificity of the NASBA primers are presented in Table 1. Mycoplasma strains were cultured in spiroplasma (SP4) medium (40) without thallium acetate supplemented with amphotericin B (0.5 mg/ml), polymyxin B (500 U/ml), glucose (0.5%), and arginine (0.25%) or urea (0.5%), depending on the nutritional needs of the species. Legionella pneumophila was cultured on buffered charcoal-yeast extract; H. influenzae was cultured on a lysed blood agar; Moraxella catarrhalis, Streptococcus pneumoniae, Streptococcus pyogenes, viridans group streptococci, Staphylococcus aureus, Klebsiella pneumoniae, Escherichia coli, Neisseria meningitidis, and Pseudomonas aeruginosa were cultured on blood plates. Chlamydia pneumoniae was cultured on HEP cells. Suspensions of these organisms were made in lysis buffer. M. pneumoniae strain PI 1428 was quantitated by incubation of 10-fold dilutions in SP4 medium at 37°C. The cultures were monitored for a change in color for 2 months. The titer was expressed in color-changing units (CCU) per ml. One CCU corresponds to 10 to 100 cells (3). Respiratory specimens. Throat swabs, bronchoalveolar lavages (BAL), nasopharyngeal aspirates (NPAs), sputum specimens, and bronchial aspirates (BAs) were obtained from the University Hospital Microbiology Laboratory, University of Antwerp, Antwerp, Belgium, and tested either as individual specimens or as pools of 10 specimens. All were previously tested and were found to be negative for M. pneumoniae by PCR (17). Fifteen specimens (2 BAs, 11 NPAs, 1 BAL specimen, 1 sputum specimen) from pediatric hospitalized patients who had acute lower respiratory tract infections and who were previously found to be positive for M. pneumoniae by PCR (17) were also tested.

Preparation of WT RNA and IC RNA. Constructs for the production of both wild-type (WT) and IC RNA were made in transcription vector pG3O, a pGEM 3 derivative with an altered multiple-cloning site and a unique BamHI site at position 2095 (25). For the generation of WT RNA, cDNA from part of the 16S rRNA from M. pneumoniae strain PI 1428, obtained by RT-PCR with adapted versions of NASBA primers OT2156 and OT2157 (27), which contain an EcoRI site and a Csp45I site, respectively, was inserted into plasmid pG3O, resulting in plasmid pG3O Mp 16S ribosomal DNA (rDNA). For the production of the IC RNA, M. pneumoniae 16S rDNA in pG3O Mp was modified by insertion of a 134-bp fragment of HIV-1 pv22 (26) comprising nucleotides 1015 to 1146 from the 5⬘ noncoding region (25). Both plasmids were transformed in E. coli DH5␣. Nucleotide sequence analysis did not reveal any mutations in the primer or probe annealing sites. These plasmids were used for large-scale generation of runoff transcripts after linearization with BamHI (Pharmacia Biotech). In vitro RNA was generated from these two constructs with T7 RNA polymerase (Pharmacia Biotech), as described previously (32). Plasmid DNA was removed by treatment with DNase I (Pharmacia Biotech). The RNA was quantitated spectrophotometrically and was stored in lysis buffer at ⫺80°C. Optimization of specimen treatment procedure. Twenty aliquots of 100 ␮l each from a sputum specimen pool were divided into four groups. Sample group 1 was treated with 66 U of protease (Sigma-Aldrich NV, Bornem, Belgium), suspended in RNase- and DNase-free H2O, at 37°C for 30 min, sample group 2 was treated with 33 U of protease at 37°C for 30 min, and sample group 3 was treated with 30 U of proteinase K (Boehringer Mannheim, Brussels, Belgium) at 37°C for 30 min. N-Acetyl-L-cysteine was added to a final concentration of 2.5 g/liter to the samples in sample group 4. The samples were incubated at 37°C for 15 min and vortexed at 5-min intervals. The first sample of each group served as a negative control. To the second, third, fourth, and fifth samples of each group, 2 ⫻ 104, 2 ⫻ 105, 2 ⫻ 106, and 2 ⫻ 107 molecules of in vitro-generated WT RNA were added, respectively. Nucleic acids were extracted, and NASBA was performed. Optimal pH of lysis buffer. Twelve aliquots of 100 ␮l of a sputum specimen pool were treated with 66 U of protease, suspended in RNase- and DNase-free H2O at 37°C for 30 min, and divided into six groups. A total of 900 ␮l of GuSCN lysis buffer at pH 7.2, 6.8, 6.4, 6.2, 6.0, and 5.6 was added to each of these groups, respectively. A total of 2 ⫻ 104 molecules was added to the first tube of each group, and, 2 ⫻ 105 molecules of in vitro-generated WT RNA was added to the second tube of each group. Nucleic acids were extracted, and NASBA was performed.

VOL. 40, 2002

DETECTION OF M. PNEUMONIAE BY NASBA

1341

FIG. 1. Sensitivity of NASBA assay. Lane 1, free probe; lanes 2 to 5, 10-fold serial dilutions of target RNA with 10, 100, 1,000, and 10,000 molecules, respectively; lanes 6 to 9, the same 10-fold serial dilutions used in lanes 2 to 5, respectively, with IC RNA; lanes 10 to 13, 10-fold serial dilutions of target RNA after nucleic acid extraction with 2 ⫻ 102, 2 ⫻ 103, 2 ⫻ 104, and 2 ⫻ 105 molecules, respectively; lanes 14 to 17, the same 10-fold serial dilutions used in lanes 10 to 13, respectively, with IC RNA after nucleic acid extraction; lanes 18 to 21, 10-fold dilutions of M. pneumoniae PI 1428 at 0.5, 5, 50, and 500 CCU, respectively. FP, free probe.

NASBA inhibitors in clinical specimens. Eight 100-␮l aliquots were prepared from each of 10 sputum specimens. Five CCU of M. pneumoniae was added to samples 1 to 4, and 50 CCU was added to samples 5 to 8. Subsequently, 66 U of protease suspended in RNase- and DNase-free H2O was added to each sample. Samples 1 and 5 were incubated for 30 min at 25°C, samples 2 and 6 were incubated for 60 min at 25°C, samples 3 and 7 were incubated for 30 min at 37°C, and samples 4 and 8 were incubated for 60 min at 37°C. Nucleic acids were extracted, and NASBA was performed. Three aliquots of 100 ␮l each were prepared from each of 10 throat swab specimens, 10 NPAs, 10 BAL specimens, and 10 BAs. The first aliquot served as a negative control, 5 CCU of M. pneumoniae was added to the second aliquot, and 50 CCU of M. pneumoniae was added to the third aliquot. The samples were treated with 66 U of protease and suspended in RNase- and DNase-free H2O for 30 min at 25°C. Nucleic acids were extracted, and NASBA was performed. Nucleic acid extraction. Nucleic acids were extracted as described by Boom et al. (4). Briefly, 100 ␮l of a clinical specimen, 100 ␮l of a protease-treated specimen, or 100-␮l aliquots of a bacterial culture was added to 900 ␮l of GuSCN lysis solution (4.7 M GuSCN [pH 6.2; Sigma-Aldrich NV], 46 mM Tris-HCl [Merck, Darmstadt, Germany], 20 mM EDTA [pH 6.2; Sigma-Aldrich NV], 1.2% [wt/vol] Triton X-100 [Sigma-Aldrich NV]), and the contents were mixed vigorously for rapid lysis. Seventy microliters of activated silica (1 g/ml of suspension in 0.1 M HCl; Organon Teknika) was added. The nucleic acid-silica complex was washed twice with GuSCN washing solution (5.25 M GuSCN, 50 mM Tris-HCl [pH 6.2]), twice with 70% (vol/vol) ethanol, and once with acetone. After the complex was dried at 56°C, the nucleic acids were eluted from the silica with 100 ␮l RNase- and DNase-free H2O and were stored at ⫺80°C. Primers and probes. The 16S rRNA-directed primers OT2156 (Mycoplasma 16S rRNA upstream primer; 5⬘ GATCCTGGCTCAGGATTAA 3⬘) and OT2157 (Mycoplasma 16S rRNA downstream primer; 5⬘ AATTCTAATACGACTCAC TATAGGGAGGTCCTTTCAACTTTGATTCA 3⬘) were used (27). Horseradish peroxidase-labeled oligonucleotide probe pd96 (Mycoplasma 16S rRNA probe; 5⬘ CGGGTGAGTAACACGTATCC 3⬘) was constructed, and the probe hybridized with both M. pneumoniae and Mycoplasma genitalium amplification products. NASBA. NASBA reactions were performed as described by Ovyn et al. (27). In negative control reactions, target nucleic acid was replaced by RNase- and DNase-free water. The amplification products were identified by a rapid nonradioactive in-solution hybridization assay (ELGA), as described by Ovyn et al. (27). Analytical sensitivity study. The analytical sensitivity of the NASBA-based M. pneumoniae 16S amplification assay was studied with serial 10-fold dilutions of suspensions of M. pneumoniae PI 1428 in SP4 medium, dilutions of in vitrogenerated WT 16S rRNA in water, and dilutions of M. pneumoniae PI 1428 and in vitro-generated WT 16S rRNA added to untreated and protease-treated samples of pools of respiratory specimens. The nucleic acid extracts of the

protease-treated samples spiked with 10-fold dilutions of M. pneumoniae PI 1428 were also used to perform PCR, as described by Ieven et al. (17). Determination of the optimal IC concentration. Serial 10-fold dilutions of IC RNA were added to M. pneumoniae-negative specimens to define the minimal number of molecules that could be detected. A 5-fold larger number was then added to serial 10-fold dilutions of in vitro-generated WT RNA to test for competition between the IC and WT RNA. RNA degradation. The RNA degradation caused by lysis buffer produced with GuSCN from two different sources was monitored. Lysis buffers were prepared by using GuSCN either from Sigma-Aldrich NV or from ICN Biomedicals NV, Asse, Belgium. M. pneumoniae PI 1428 cells were added to both versions of the lysis buffer, and the resulting specimens were stored at ⫺80°C or at room temperature for different periods of time prior to extraction.

RESULTS Specificity of the M. pneumoniae 16S rRNA NASBA primers. By use of M. pneumoniae 16S rRNA NASBA primers OT2156 and OT2157 and probe pd96, positive results were obtained with nucleic acid extracts from M. pneumoniae types 1 and 2 and M. genitalium but with none of the other organisms listed in Table 1. Analytical sensitivity of the M. pneumoniae 16S rRNA NASBA primers compared with that of PCR primers. The analytical sensitivity of the M. pneumoniae 16S rRNA NASBA primers tested with dilutions of in vitro-generated WT and IC RNA were 10 and 102 molecules, respectively (Fig. 1). To test the analytical sensitivity of the M. pneumoniae NASBA assay, nucleic acid extraction was performed with 10fold dilutions of the in vitro-generated WT and IC RNA. The procedure reduced the sensitivity to 2 ⫻ 104 molecules for both sources of RNA, if it is assumed that 100% of the nucleic acids were recovered during the extraction procedure. The analytical sensitivity of the assay was 5 CCU when the assay was applied to dilutions of nucleic acids extracted from a culture of M. pneumoniae (Fig. 1). When the assay was applied to clinical specimens spiked with either M. pneumoniae or in vitro-generated WT 16S rRNA generated in vitro, the analytical sensitivity was 5 ⫻ 103 CCU or 2 ⫻ 107 molecules, respectively.

1342

LOENS ET AL.

J. CLIN. MICROBIOL.

TABLE 2. Detection limits of NASBA in spiked clinical specimens Specimen

Water SP4 Sputumb Protease-treated BALb Protease-treated Throat swabb Protease-treated NPAb Protease-treated BAb Protease-treated

Detection limit WT RNA

BALc throat swabc NPAc BAc

IC RNAa

4

2⫻10

7

2⫻107 2⫻105 2⫻106 2⫻105 2⫻105 2⫻104 2⫻105 2⫻105 2⫻105 2⫻104

2⫻10 sputumc

a

2⫻10 2⫻104 2⫻104 2⫻104 2⫻105 2⫻104 2⫻104 2⫻104 2⫻105 2⫻104

PI 1428

4

5 5,000 5 5 5 50 5 50 5 5,000 50

a

For WT and IC RNAs, the detection limit is the number of molecules obtained by nucleic acid extraction; for PI 1428 RNA, the detection limit is the number of CCU obtained by nucleic acid extraction. b PH 7.2 lysis buffer. c PH 6.2 lysis buffer.

However, the analytical sensitivity of the assay could be remarkably improved by appropriate pretreatment of the respiratory specimens. All respiratory specimens became more liquid by all methods tested, but only the treatment with 66 U of protease at 25°C for 30 min followed by extraction with lysis buffer at pH 6.2 improved the sensitivity to 5 CCU or 2 ⫻ 104 molecules for M. pneumoniae or in vitro-generated WT RNA, respectively, in vitro and added to different clinical specimens (Table 2). The only exception was BA, for which protease treatment resulted in a sensitivity of 50 CCU. When PCR was applied to the nucleic acid extracts of the protease-treated samples spiked with M. pneumoniae PI 1428, there was no difference in sensitivity between the two assays. When throat swab specimens, BAL specimens, NAs, BAs, and sputum specimens were spiked with 50 CCU of M. pneumoniae, all throat swab and BAL specimens, 9 of 10 NAs, 9 of 10 BAs, and 8 of 10 sputum specimens were NASBA positive. When the specimens were spiked with 5 CCU of M. pneumoniae, 6 of 10 throat swab and BAL specimens, 5 of 10 NAs, and 4 of 10 BAs produced a positive result. Thus, 1 of 10 NAs, 1 of 10 BAs, and 2 of 10 sputum specimens contained inhibitors. The intrarun coefficients of variation for the detection of 50, 500, and 5,000 CCU were 26.4, 8.1, and 4.6, respectively; the interrun coefficients of variation for the same inputs were 34.8, 10.7, and 13.9, respectively. The limits of detection of M. pneumoniae RNA and M. pneumoniae whole cells were not influenced by the addition of 106 molecules of the IC (Fig. 2). The 15 clinical specimens positive for M. pneumoniae by PCR were also positive by the NASBA assay. For 14 of these specimens, there was no amplification of the IC due to competition with the large load of M. pneumoniae present. None of 100 PCR-negative clinical specimens was NASBA positive, while 3 specimens contained inhibitors and the IC produced no signal. RNA degradation. M. pneumoniae stored at room temperature for 60 min in lysis buffer prepared with GuSCN from ICN Biomedicals NV was not detectable by NASBA, whereas for

FIG. 2. Suitability of the assay for biological samples. Lane 1, free probe; lanes 2 to 5, 10-fold serial dilutions of target RNA were added to untreated sputum and nucleic acid was extracted with lysis buffer (pH 7.2) and 2 ⫻ 104, 2 ⫻ 105, 2 ⫻ 106, and 2 ⫻ 107 molecules, respectively; lanes 6 to 9, the same 10-fold serial dilutions used in lanes 2 to 5, respectively, were added to treated sputum and nucleic acid was extracted with lysis buffer (pH 6.2); lanes 10 to 13, 10-fold serial dilutions of M. pneumoniae PI 1428 were added to sputum and the optimized treatment procedure was used. A total of 106 molecules of IC RNA was added to these last four samples: 0.5, 5, 50, and 500 CCU, respectively. FP, free probe.

suspensions prepared with GuSCN from Sigma-Aldrich NV, the reaction was positive even after storage for up to 240 min at room temperature (Fig. 3). DISCUSSION By use of M. pneumoniae 16S rRNA-specific NASBA primers OT2156 and OT2157 and probe pd96, positive results were

FIG. 3. M. pneumoniae RNA degradation. ■, detection level with GuSCN from Sigma-Aldrich NV; Œ, detection level with GuSCN from ICN Biomedicals NV.

VOL. 40, 2002

obtained with nucleic acids extracted from M. pneumoniae types 1 and 2 and M. genitalium. The cross-reactivity with M. genitalium does not cause problems because the organism is seldom found in the respiratory tract. Baseman et al. (2) reported that M. genitalium may be isolated from the respiratory tract and that its possible presence should always be considered when trying to diagnose a potential case of M. pneumoniae pneumonia, but this has been confirmed only by De Barbeyrac et al. (9), who applied an M. pneumoniae-specific PCR and an M. genitalium-specific PCR to respiratory specimens, and by Williamson et al. (45), who also applied two organism-specific PCR assays to respiratory specimens. The cross-reactivity does not limit the ability to diagnose a Mycoplasma infection and has no influence on the antibiotic therapy; it may only influence epidemiological data. This may be avoided by confirmation of a positive result by a second hybridization with an M. pneumoniae-specific probe (27) and an M. genitalium-specific probe. Furthermore, confirmation of all positive findings by reanalysis by reextraction of the original specimen and an independent amplification reaction and detection is always recommended. The second analysis may specifically be targeted to M. pneumoniae and M. genitalium. Some commercial DNA hybridization kits for M. pneumoniae also show cross hybridization with M. genitalium (11). By use of primers OT2156 and OT2157 (Fig. 1), the detection limits for M. pneumoniae were 10 molecules of in vitrogenerated WT RNA and 100 molecules of IC RNA. Although the guanidine extraction procedure is known to be efficient, it resulted in a loss of 2 to 3 logs when applied to RNA in water and to RNA or M. pneumoniae cells added to clinical specimens. To increase the sensitivity of detection, several specimen treatment procedures were tested. Treatment of clinical specimens with 66 U of protease for 30 min at 25°C and the use of GuSCN lysis solution at pH 6.2 rather than pH 7.2 was found to be optimal. The lower pH during nucleic acid extraction possibly allows better adhesion of the RNA to the activated silica. Thus, the detection limit for M. pneumoniae cells in clinical specimens was reduced to 5 CCU, or 50 to 500 cells. A similar sensitivity was found when the results obtained by the 16S rRNA NASBA and the P1 PCR with protease-treated samples were compared. While E. coli and Bacillus subtilis have 7 and 10 rRNA operons, respectively, mollicutes have only 1 or 2 (5). More operons may be responsible for more transcription and thus more 16S rRNA molecules. Furthermore, fast-growing organisms may contain 104 rRNA molecules per cell, but it is generally accepted that slowly growing organisms have fewer rRNA molecules. For M. pneumoniae, the amount of rRNA molecules would be comparable to that for M. tuberculosis, which is about 10 molecules per cell. The P1 adhesin gene is an intriguing target for PCR because of its repetitive nature within the M. pneumoniae genome. Approximately 8% of this genome is composed of repetitive DNA elements with regions homologous to the P1 adhesin gene (15). For molecular biological detection of M. pneumoniae, both targets may have the same advantage. The detection limits of the various M. pneumoniae molecular amplification assays described in the literature are the smallest numbers of organisms measured according to the number of CFU (1 CFU has been shown to correspond to 160 organisms

DETECTION OF M. PNEUMONIAE BY NASBA

1343

[14] and 10 to 1,000 organisms [28]), the number of CCU (1 CCU corresponds to 10 to 100 organisms [3]), the number of cells, or the quantity of DNA, which makes comparison of the analytical sensitivities of the different assays very difficult. Abele-Horn et al. (1), for example, reported that the detection limit of their assay was 3,000 genome copies, 30 pg of DNA, 19 CFU, or 1.9 ⫻ 103 organisms. Since the number of rRNA molecules present per M. pneumoniae cell is unknown and we wanted to calculate the analytical sensitivity of our assay, we decided to use in vitro-generated WT RNA, which might be a more objective method for calculation of the sensitivity of our assay. The use of ICs is indispensable for the reliable detection of microorganisms in clinical specimens by PCR, RT-PCR, or NASBA since inhibitors may still be present in some clinical specimens and give rise to false-negative or invalid results (7, 17, 29, 35). Therefore, IC RNA was constructed for the M. pneumoniae NASBA assay. The IC is a modified amplicon that has been made longer and that is added to each reaction tube. The primer binding sites are identical to those of the target, and therefore, they are amplified by the same reagents as the real target, but they are easily differentiated from it because they are longer. By adding the control at the very start of the process, the efficacy of the sample processing procedure can be assessed. During the amplification procedure there is competition between the IC RNA and the WT RNA, and the amount of IC to be added to the specimens should be carefully titrated. When the IC is added to negative specimens, only the IC amplicon will be visible on the gel; when the IC is added to weakly positive specimens, both the amplicons from the IC RNA and the WT RNA will be visible; and when the IC is added to strongly positive specimens, only the amplicon of the WT will be visible. In the case of inhibition, no amplicon will be visible at all. The IC used in our NASBA was detected in 97 of 100 M. pneumoniae-negative samples but in none of the M. pneumoniae-positive samples. This results from the minimal amount of IC added and from the fact that WT RNA is preferentially amplified because of its shorter length. In the present study, false-negative (invalid) results were observed with very viscous samples. This could result from the glycoproteins that adhere to the activated silica, producing clumps and making the silica inaccessible for RNA. The method studied detected M. pneumoniae in the 15 respiratory specimens that previously tested positive by PCR. In our hands, the GuSCN from ICN Biomedicals NV was found to be unsuitable for RNA extraction. The lysis buffer used mainly contains GuSCN, a chaotropic reagent, which inhibits all nuclease activity and which destroys (sub)cellular components. The differences seen with the GuSCNs from the different companies may be caused by inefficient inhibition of RNA-degrading enzymes, by a lower yield of nucleic acid after isolation, or by the presence of an unknown inhibitor in the eluate. The inhibitor may be introduced by the purification procedure itself: the inhibitor may be present in the GuSCN, may be bound to the silica and eluted with the nucleic acids, and may inhibit the enzymes during the amplification procedure. Samuelson et al. (33) reported problems with an unexplained inhibitory factor with the RNAce Purification system (Bioline), a commercially available extraction kit based on the same chemistry as the procedure described by Boom et al. (4),

1344

LOENS ET AL.

J. CLIN. MICROBIOL.

when low dilutions of the eluate were processed. The present study and other studies (6, 33) suggest that quality control of the buffers used for storage of clinical material is critical, particularly when RNA is to be analyzed. Most RNA degradation is likely to occur during transportation of the specimens. Bruisten et al. (6) studied the stability of HIV-1 RNA in whole blood, plasma, and serum before and after the addition of lysis buffer and concluded that the specimens could be kept at 4°C, provided that the transportation time was as short as possible. Furthermore, those investigators suggested that specimens should be processed on the day of sampling and stored at ⫺70°C, thus stabilizing the RNA for at least 6 months. Despite the limited number of known M. pneumoniae-positive samples available for investigation, the NASBA assay described here is a promising assay for the detection of M. pneumoniae in respiratory specimens. In summary, it was shown that the NASBA-based 16S rRNA assay, combined with a nonradioactive hybridization procedure (ELGA), provides a sensitive and specific method for the detection of M. pneumoniae. The complete procedure, including protease pretreatment of the specimens, applied to a group of 12 samples takes less than 7 h to perform and can detect between 50 and 500 cells in a 100-␮l specimen.

15. 16. 17.

18. 19. 20.

21.

22.

23.

ACKNOWLEDGMENT We thank S. R. Pattyn for critically reviewing the manuscript.

24.

REFERENCES 1. Abele-Horn, M., U. Busch, H. Nitschko, E. Jacobs, R. Bax, F. Pfaff, B. Schaffer, and J. Heeseman. 1998. Molecular approaches to diagnosis of pulmonary diseases due to Mycoplasma pneumoniae. J. Clin. Microbiol. 36: 548–551. 2. Baseman, J. B., S. F. Dallo, J. G. Tully, and D. L. Rose. 1988. Isolation and characterization of Mycoplasma genitalium strains from the human respiratory tract. J. Clin. Microbiol. 26:2266–2269. 3. Bernet, C., M. Garret, B. de Barbeyrac, C. BéBéar, and J. Bonnet. 1989. Detection of Mycoplasma pneumoniae by using the polymerase chain reaction. J. Clin. Microbiol. 27:2492–2496. 4. Boom, R., C. J. A. Sol, M. M. M. Salimans, C. L. Jansen, P. M. E. Wertheimvan Dillen, and J. van der Noordaa. 1990. Rapid and simple method for purification of nucleic acids. J. Clin. Microbiol. 28:495–503. 5. Bové, J. M. 1993. Molecular features of Mollicutes. Clin. Infect. Dis. 17(Suppl. 1):S10–S31. 6. Bruisten, S. M., P. Oudshoorn, P. van Swieten, B. Boeser-Nunnink, P. van Aarle, S. P. Tondreau, and H. T. M. Cuypers. 1997. Stability of HIV-1 RNA in blood during specimen handling and storage prior to amplification by NASBA-QT. J. Virol. Methods 67:199–207. 7. Clewley, J. P. 1989. The polymerase chain reaction, a review of the practical limitations for human immunodeficiency virus diagnosis. J. Virol. Methods 25:179–188. 8. Clyde, W. A., Jr. 1993. Clinical overview of typical Mycoplasma pneumoniae infections. Clin. Infect. Dis. 17(Suppl. 1):S32–S36. 9. de Barbeyrac, B., C. Bernet-Poggi, F. Fébrer, H. Renaudin, M. Dupon, and C. Bébéar. 1993. Detection of Mycoplasma pneumoniae and Mycoplasma genitalium in clinical samples by polymerase chain reaction. Clin. Infect. Dis. 17(Suppl. 1):S83–S89. 10. De Wit, D., M. Wootton, B. Allen, and L. Steyn. 1993. Simple method for production of internal control DNA for Mycobacterium tuberculosis polymerase chain reaction assays. J. Clin. Microbiol. 31:2204–2207. 11. Dular, R., R. Kajioka, and S. Kasatiya. 1988. Comparison of Gen-Probe commercial kit and culture technique for the diagnosis of Mycoplasma pneumoniae infection. J. Clin. Microbiol. 26:1068–1069. 12. Foy, H. M. 1993. Infections caused by Mycoplasma pneumoniae and possible carrier state in different populations of patients. Clin. Infect. Dis. 17(Suppl. 1):S37–S46. 13. Gerna, G., F. Baldanti, J. M. Middeldorp, M. Furione, M. Zavattoni, D. Sillesi, and M. G. Revello. 1999. Clinical significance of expression of human cytomegalovirus pp 67 late transcript in heart, lung, and bone marrow transplant recipients, as determined by nucleic acid sequence-based amplification. J. Clin. Microbiol. 37:902–911. 14. Harris, R., B. P. Marmion, G. Varkanis, T. Kok, B. Lunn, and J. Martin. 1988. Laboratory diagnosis of Mycoplasma pneumoniae infection. 2. Com-

25.

26. 27. 28. 29.

30. 31. 32. 33. 34.

35.

36.

37.

parison of methods for the direct detection of specific antigen or nucleic acid sequences in respiratory exudates. Epidemiol. Infect. 101:685–694. Himmelreich, R., H. Hilbert, H. Plagens, E. Pirkl, B.-L. Li, and R. Herrmann. 1996. Complete sequence analysis of the genome of the bacterium Mycoplasma pneumoniae. Nucleic Acids Res. 24:4420–4429. Ieven, M., and H. Goossens. 1997. Relevance of nucleic acid amplification techniques for diagnosis of respiratory tract infections in the clinical laboratory. Clin. Microbiol. Rev. 10:242–256. Ieven, M., D. Ursi, H. Van Bever, W. Quint, H. G. M. Niesters, and H. Goossens. 1996. Detection of Mycoplasma pneumoniae by two polymerase chain reactions and role of M. pneumoniae in acute respiratory tract infections in pediatric patients. J. Infect. Dis. 173:1445–1452. Ieven, M., H. Demey, D. Ursi, G. van Goethem, P. Cras, and H. Goossens. 1998. Fatal encephalitis caused by Mycoplasma pneumoniae diagnosed by the polymerase chain reaction. Clin. Infect. Dis. 27:1552–1553. Kaneko, S., S. Murakami, M. Unoura, and K. Kobayashi. 1992. Quantitation of hepatitis C virus RNA by competitive polymerase chain reaction. J. Med. Virol. 37:278–282. Kessler, H. H., D. E. Dodge, K. Pierer, K. K. Y. Young, Y. Liao, B. I. Santner, E. Eber, M. G. Roeger, D. Stuenzer, B. Sixl-Voigt, and E. Marth. 1997. Rapid detection of Mycoplasma pneumoniae by an assay based on PCR and probe hybridization in a nonradioactive microwell plate format. J. Clin. Microbiol. 35:1592–1594. Kievits, T., B. van Gemen, D. van Strijp, R. Schukkink, M. Dircks, H. Adriaanse, L. Malek, R. Sooknanan, and P. Lens. 1991. NASBA™ isothermal enzymatic in vitro nucleic acid amplification optimized for the diagnosis of HIV-1 infection. J. Virol. Methods 35:273–286. Kok, T. W., G. Varkanis, B. P. Marmion, J. Martin, and A. Esterman. 1988. Laboratory diagnosis of Mycoplasma pneumoniae infection. 1. Direct detection of antigen in respiratory exudates by enzyme immunoassay. Epidemiol. Infect. 101:669–684. Lair, S. V., T. E. Mirkow, J. A. Dodds, and M. F. Murphy. 1993. A single temperature amplification technique applied to the detection of citrus tristeza viral RNA in plant nucleic acid extracts. J. Virol. Methods 47:141–152. Lan, J., A. J. C. van den Brule, D. J. Hemrika, E. K. Risse, J. M. M. Walboomers, M. E. I. Schipper, and C. J. M. Meijer. 1995. Chlamydia trachomatis and ectopic pregnancy: retrospective analysis of salpingectomy specimens, endometrial biopsies and cervical smears. J. Clin. Pathol. 48:815– 819. Morré, S. A., P. Sillekens, M. V. Jacobs, P. van Aarle, S. de Blok, B. van Gemen, J. M. M. Walboomers, C. J. L. M. Meijer, and A. J. C. van den Brule. 1996. RNA amplification by nucleic acid sequence-based amplification with an internal standard enables reliable detection of Chlamydia trachomatis in cervical scrapings and urine samples. J. Clin. Microbiol. 34:3108–3114. Muesing, M. A., D. H. Smith, C. D. Cabradilla, C. V. Benton, L. A. Lasky, and D. J. Capon. 1985. Nucleic acid structure and expression of the human AIDS/lymphadenopathy retrovirus. Nature 313:450–458. Ovyn, C., D. van Strijp, M. Ieven, D. Ursi, B. van Gemen, and H. Goossens. 1996. Typing of Mycoplasma pneumoniae by nucleic acid sequence-based amplification, NASBA. Mol. Cell. Probes 10:319–324. Razin, S. 1994. DNA probes and PCR in diagnosis of Mycoplasma infections. Mol. Cell. Probes 8:497–511. Reznikov, M., T. K. Blackmore, J. J. Finlay-Jones, and D. L. Gordon. 1995. Comparison of nasopharyngeal aspirates and throat swab specimens in a polymerase chain reaction based test for Mycoplasma pneumoniae. Eur. J. Clin. Microbiol. Infect. Dis. 14:58–61. Rosenstraus, M., Z. Wang, S.-Y. Chang, D. deBonville, and J. S. Spadoro. 1998. An internal control for routine diagnostic PCR: design, properties, and effect on clinical performance. J. Clin. Microbiol. 36:191–197. Saiki, R. K., D. H. Gelfand, S. Stoffel, S. J. Scharf, R. Higuchi, G. T. Horn, K. B. Mullis, and H. A. Erlich. 1988. Primer-dissected enzymatic amplification of DNA with a thermostable DNA-polymerase. Science 239:487–491. Sambrook, J., T. Manniatis, and E. Fritsch. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. Samuelson, A., D. Westmoreland, R. Eccles, and J. D. Fox. 1998. Development and application of a new method for amplification and detection of human rhinovirus RNA. J. Virol. Methods 71:197–209. Sillekens, P., W. Kok, B. van Gemen, P. Lens, H. Huisman, T. Cuypers, and T. Kievits. 1994. Specific detection of HCV RNA using NASBA™ as a diagnostic tool, p. 71–82. In Hepatitis C virus. John Libbey Eurotext, Paris, France. Skakni, L., A. Sardet, J. Just, J. Landman-Parker, J. Costil, N. Moniot-Ville, F. Bricout, and A. Garbarg-Chenon. 1992. Detection of Mycoplasma pneumoniae in clinical samples from pediatric patients by polymerase chain reaction. J. Clin. Microbiol. 30:2638–2643. Smits, H. L., B. van Gemen, R. Schukkink, J. van der Velden, S. P. TjongA-Hung, M. F. Jebbink, and J. ter Schegget. 1995. Application of the NASBA nucleic acid amplification method for the detection of human papillomavirus type 16 E6-E7 transcripts. J. Virol. Methods 54:75–81. Smits, H. L., G. C. Gussenhoven, W. Terpstra, R. A. F. Schukkink, B. van Gemen and T. van Gool. 1997. Detection, identification and semi-quantifi-

VOL. 40, 2002

38. 39.

40. 41. 42.

cation of malaria parasites by NASBA amplification of small subunit ribosomal RNA sequences. J. Microbiol. Methods 28:65–75. Sooknanan, R., and L. T. Malek. 1995. NASBA a detection and amplification system uniquely suited for RNA. Bio/Technology 13:563–564. Tjhie, J. H. T., F. J. M. van Kuppeveld, R. Roosendaal, W. J. G. Melchers, R. Gordijn, D. M. MacLaren, J. M. M. Walboomers, C. J. M. Meijer, and A. J. C. van den Brule. 1994. Direct PCR enables detection of Mycoplasma pneumoniae in patients with respiratory tract infections. J. Clin. Microbiol. 32:11–16. Tully, J. G., D. Taylor-Robinson, R. M. Cole, and D. L. Rose. 1981. A newly discovered Mycoplasma in the human urogenital tract. Lancet i:1288–1291. Ursi, J. P., D. Ursi, M. Ieven, and S. R. Pattyn. 1992. Utility of an internal control for the polymerase chain reaction: application to detection of Mycoplasma pneumoniae in clinical specimens. APMIS 100:635–639. Uyttendaele, M., R. Schukkink, B. van Gemen, and J. Debevere. 1994. Identification of Camplylobacter jejuni, Campylobacter coli and Camplylobacter

DETECTION OF M. PNEUMONIAE BY NASBA

1345

lari by the nucleic acid sequence based amplification system NASBA™. J. Appl. Bacteriol. 77:694–701. 43. van der Vliet, G. M. E., R. A. F. Schukkink, B. van Gemen, P. Schepers, and P. R. Klatser. 1993. Nucleic acid sequence based amplification (NASBA) for the identification of Mycobacteria. J. Gen. Microbiol. 139:2423–2429. 44. van der Vliet, G. M. E., S.-N. Cho, K. Kampirapap, J. van Leeuwen, R. A. F. Schukkink, B. van Gemen, P. K. Das, W. R. Faber, G. P. Walsh, and P. R. Klatser. 1996. Use of NASBA RNA amplification for detection of Mycobacterium leprae in skin biopsies from untreated and treated leprosy patients. Int. J. Leprosy 64:396–403. 45. Williamson, J., B. P. Mamion, D. A. Worswick, T. W. Kok, G. Tannock, R. Herd, and R. J. Harris. 1992. Laboratory diagnosis of Mycoplasma pneumoniae infection. 4. Antigen capture and PCR gene amplification for detection of the Mycoplasma: problems of clinical correlation. Epidemiol. Infect. 109:519–537.