Previous Article | Next Article ![]()
Journal of Clinical Microbiology, June 2006, p. 2078-2083, Vol. 44, No. 6
0095-1137/06/$08.00+0 doi:10.1128/JCM.00003-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Mycoplasma Laboratory, Statens Serum Institut, DK-2300 Copenhagen S, Denmark,1 Huddinge University Hospital, Karolinska Institutet, Department of Dermatovenereology, S-14 186 Huddinge, Sweden,2 Department of Dermatology and Venereology, Örebro University Hospital, SE -701 85 Örebro, Sweden,3 Section of Infectious Diseases, Department of Medicine, Louisiana State University Health Sciences Center, New Orleans, Louisiana4
Received 2 January 2006/ Returned for modification 9 February 2006/ Accepted 22 March 2006
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
At present, the detection of M. genitalium is dependent mainly on PCR methods. The PCR can be based on the MgPa-1/MgPa-3 primer set located in conserved regions of the mgpB gene and producing an amplified fragment of 281 bp, as previously described (14, 15). This primer set has a very low limit of detection in clinical specimens and is consequently suited for diagnostic use. Furthermore, based on restriction enzyme analysis, it was found in 1991 that the region amplified with these primers was somewhat heterogeneous (15). Since then, sequencing has shown the presence of several mutations in the few available clinical strains compared to the sequence of the type strain M. genitalium G37 (13).
The aim of the present study was to document, by DNA-based typing, that M. genitalium is transmissible through sexual contact. We determined the discriminatory index (DI), i.e., the likelihood that unrelated strains sampled from the test population would be placed into different typing groups (10), by sequencing specimens from unrelated patients. The stability of the typing system was determined by analyzing patients with consecutive positive specimens.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Genetic diversity as expressed by the number of established sequence types was determined by comparing all 267 specimens sequenced, regardless of the availability of information regarding possible contact between the patients.
Reproducibility of the typing system was not evaluated in the classical sense with repeated typing of the same specimen but by selecting 2 male patients with positive M. genitalium PCR results for both urethral swab and FVU specimens at 7 time points and 12 female patients with positive specimens from more than one anatomical site, i.e., urethra, cervix, or FVU, at 17 time points.
Stability was analyzed by comparing 156 consecutive M. genitalium-positive specimens collected from 36 patients seen in Norway and Sweden as well as 7 specimens collected from patients seen in New Orleans, La., and three consecutive M. genitalium isolates (M6090, M6151, and M6312) obtained by culture from a French male patient.
The DI (10) was determined by comparing the number of different sequence types in specimens obtained from 52 unrelated patients from whom information about partners was available. The analysis was extended to include patients from whom information about partners was not available but excluded repeated samples and samples from reported partners. In this extended analysis, 144 patients were included.
Concordance of sequence types between couples was determined by comparing 79 M. genitalium-positive specimens from 19 patients with reported sexual relationships. Patients having repeatedly positive specimens were included in the stability study, and those with more than one positive specimen from any anatomical site were included in the reproducibility study. The first male FVU specimen from each couple with a sexual relationship was included in the determination of the discriminatory index.
DNA sequencing. PCR products were obtained with the MgPa-1/MgPa-3 primer set (15) using sample preparation and optimized reaction conditions as previously described (12, 14). The PCR products were purified using a QiaQuick PCR purification kit (QIAGEN, Hilden, Germany). Sequencing was performed with the ABI Prism Big Dye terminator reaction kit v. 2.0 (Applied Biosystems, Foster City, CA) and read on an ABI 377 or ABI 3100 genetic analyzer. Both strands of the amplified fragments were sequenced.
The sequences were aligned and compared with the 281-bp sequence located between MgPa-1 and MgPa-3 in the genome of M. genitalium strain G37T (GenBank accession number NC 000908) using BioNumerics 4.0 (Applied Maths, Sint-Martens-Latem, Belgium).
| RESULTS |
|---|
|
|
|---|
An alignment of the 56 sequences with the M. genitalium G37 sequence showing the differences is also available for download in the supplemental material.
Discriminatory index. The sequences from the 52 unrelated patients were divided into 29 different sequence types, and the resulting DI was 0.95 (Fig. 1). Expanding the analysis to include 144 patients from whom partner information was unavailable but excluding repeat specimens and specimens from known partners revealed 54 different sequence types and resulted in a DI of 0.93.
|
M. genitalium-positive specimens were collected from eight women on two to four visits over a median observation period of 75 days (range, 37 to 201), and seven of the specimens had stable sequence types. The specimens from the female partner of couple A described above, collected over a 40-day interval, had different sequences.
Reproducibility. The 24 specimen sets obtained from two anatomical sites at the same visit yielded the same sequence types except for the urethral and cervical swabs from the female patient in couple A described above; the urethral swab specimen had a mixed sequence, with dual peaks in the chromatogram at the variant positions clearly reflecting the sequences of the old and the newly introduced strains.
Sexual transmission. A total of 79 M. genitalium-positive specimens from 19 couples were investigated. For 18 of the couples, nine different sequence types were identified, and the partners of each of the 18 couples had the same sequence type. Five sequence types included only one couple, three sequence types three couples, and one sequence type four couples (Fig. 2). Couple A described above had a new sequence type introduced apart from their initial unique sequence type but had initially had the same sequence types.
|
| DISCUSSION |
|---|
|
|
|---|
Sexual transmission of M. genitalium has previously been strongly suggested by epidemiological studies which have found a concordance rate of infection between sexual partners ranging from 46 to 63% for female partners of M. genitalium-infected male index patients (1, 2, 7, 16).
The present typing system was developed on the basis of a diagnostic PCR assay (15) which allowed for a very low limit of detection and hence a high typeability. The observation that the 281-bp amplicon produced by the MgPa-1/MgPa-3 primer set contained a certain degree of variability was made during the initial validation of the assay. It was noted that only the isolates available from ATCC contained the EcoRI restriction enzyme cleavage site expected from the published sequence. Furthermore, some of the amplicons from clinical specimens also lacked an SspI site found in the type strain (15). In a later study, during the development of a TaqMan quantitative PCR assay, sequencing revealed a much higher heterogeneity than expected, and it was suggested that the MgPa-1/MgPa-3 assay could be applied for molecular typing purposes (13).
Typeability was not systematically evaluated, since specimens producing only faint bands in the MgPa-1/MgPa-3 PCR were not included in the study. It should be possible, however, to reamplify faint bands and produce enough amplicon for sequencing. Except for the urethral specimen from the female partner of couple A which produced ambiguous sequences due to an infection with two different strains, all amplicons produced unambiguous sequences.
The typing system described here was shown to have the excellent discriminatory index of 0.95. In other words, if the M. genitalium strains from two patients were unrelated, they would fall within different sequence types in 95% of the cases. This is well over the 0.90 level suggested by Hunter and Gaston (10) for an efficient typing system. The discriminatory index was determined for a collection of 52 specimens from mostly Scandinavian patients where information about sexual partners was available. However, when the panel was expanded to include 144 patients who were expected not to have sexual relations, but where this information was not directly available, the discriminatory index remained as high as 0.93. This group of specimens represented a huge geographical diversity as well as a considerable temporal separation.
Two important aspects of a typing system are reproducibility and stability, i.e., the ability to assign a strain to the same cluster when the typing is repeated or when it is performed on samples taken at various time points. The reproducibility was not investigated by repeating the sequencing of individual specimens but was estimated by examining specimens from different anatomical sites from the same patient and collected at the same time. Only for the female patient from couple A, where a new strain was introduced, was a mixed sequence found in the urethral swab specimen, showing the high reproducibility of the method. Due to the poor treatment efficacy of doxycycline (6), a relatively high number of consecutive M. genitalium-positive specimens could be collected. Sequence conservation was documented even up to 895 days after collection of the first specimen, demonstrating the ability of M. genitalium to cause persistent infections and the apparent lack of spontaneous clearance of the infection. Only two patients apart from couple A presented with different sequence types, at intervals of 472 and 1,395 days, respectively. These patients, however, had received treatment and submitted negative specimens after treatment before they were reinfected with new strains. These reinfections demonstrate that the protective immunity after a natural infection appears to be limited.
The genetic heterogeneity suggests that M. genitalium is endemic in the areas studied here and that infections are not due to the dissemination of a single strain. On the other hand, a few big clusters containing a relatively high number of strains may suggest either a clonal spread of certain, more-virulent strains or a particularly common sequence type, reflecting the limitations of the typing system. One such cluster contained sequence types from specimens collected in Denmark, Norway, Russia, Germany, the United Kingdom, France, Japan, and several sites in Sweden, illustrating the widespread geographic representation of this sequence type. Further discrimination could probably be achieved by combining the MgPa-1/MgPa-3 typing system with elements of the variable-number tandem repeat typing system or of the rRNA sequence typing system, both described by Ma and Martin (18).
With a discriminatory index of 0.95, it was very surprising to find that all seven strains available from the ATCC shared the same sequence type. Although the two original ATCC isolates from the urogenital tract, G37T and M30, had the same sequence type, this could be explained by a common source yielding strains that were cultivable. However, when an early passage (passage 7) of the M30 strain was acquired from the Mollicutes Collection of Cultures and Antisera, Gainesville, Florida, it lacked the characteristic EcoRI site and clustered together with other strains having another relatively common sequence type. Moreover, the four respiratory tract isolates (3) and a strain isolated from synovial fluid (22) also had sequence types identical to that of the G37 strain. These five strains had all been isolated from cultures of M. pneumoniae. Among the other 260 sequences we studied, none had the EcoRI site found in the ATCC strains, and it therefore appears likely that the ATCC strains were contaminated and overgrown with the G37 strain or that the isolates from the extragenital sites were contaminated at some point. These findings are in good agreement with previous studies (17, 18) and may have serious implications in many respects. Except for two studies (8, 9), all antimicrobial susceptibility studies have relied on the ATCC strains and thus underestimate the variability in susceptibility among isolates. Studies of antigen variation and validations of diagnostic PCR assays may also be in danger of misinterpretation due to the lack of strains representing different sequence types.
The present typing system is simple and reproducible and has an excellent discriminatory capacity which might prove useful in studies of sexual networks and for evaluation of treatment failures in terms of whether they are related to reinfection with new strains or to persistence. In the laboratory, this system may also help to document the uniqueness of newly isolated M. genitalium strains.
| ACKNOWLEDGMENTS |
|---|
This work was supported in part by NIAID grant 1 U19 AI061972-01.
| FOOTNOTES |
|---|
Supplemental material for this article may be found at http://jcm.asm.org/. ![]()
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Antimicrob. Agents Chemother. | Clin. Microbiol. Rev. |
|---|---|
| Clin. Vaccine Immunol. | ALL ASM JOURNALS |
|---|