Increased macrolide resistance of Mycoplasma pneumoniae in France directly detected in clinical specimens by real-time PCR and melting curve analysis.
ABSTRACT Mycoplasma pneumoniae is a common aetiological agent of community-acquired respiratory tract infections for which macrolides are the treatment of choice. In France, only two macrolide-resistant isolates were reported in 1999. In contrast, several recent data reported that macrolide-resistant M. pneumoniae isolates have been spreading since 2000 in Japan. Mutations A2058G (Escherichia coli numbering), A2058C, A2059G, A2062G, C2611A and C2611G in domain V of the 23S rRNA gene were associated in vivo or in vitro with this resistance. The aim of this study was to determine whether macrolide resistance of M. pneumoniae is emerging in France.
We developed a duplex real-time PCR for the detection of the six 23S rRNA mutations associated with macrolide resistance in M. pneumoniae and a simplex real-time PCR for the identification of the A2058G mutation, the most common one. Both methods rely on fluorescence resonance energy transfer coupled to melting curve analysis and are directly applicable to clinical samples. The duplex real-time PCR assay, first validated on 40 genetically characterized M. pneumoniae strains, was then applied directly on 248 French respiratory tract clinical samples.
Among M. pneumoniae-positive specimens collected before 2005, no macrolide-resistant M. pneumoniae isolate was detected. In contrast, among 51 samples collected between 2005 and 2007, five (9.8%) yielded a resistant genotype, suggesting a recent increase in macrolide-resistant M. pneumoniae isolates in France.
The epidemiological monitoring of macrolide resistance in this species has become necessary in France and Europe, and will be made easier by using these PCR assays.
- SourceAvailable from: PubMed Central[Show abstract] [Hide abstract]
ABSTRACT: Throat swabs from children with suspected Mycoplasma pneumoniae (M. pneumoniae) infection were cultured for the presence of M. pneumoniae and its species specificity using the 16S rRNA gene. Seventy-six M. pneumoniae strains isolated from 580 swabs showed that 70 were erythromycin resistant with minimum inhibitory concentrations (MIC) around 32-512 mg/L. Fifty M. pneumoniae strains (46 resistant, 4 sensitive) were tested for sensitivity to tetracycline, ciprofloxacin, and gentamicin. Tetracycline and ciprofloxacin had some effect, and gentamicin had an effect on the majority of M. pneumoniae strains. Domains II and V of the 23S rRNA gene and the ribosomal protein L4 and L22 genes, both of which are considered to be associated with macrolide resistance, were sequenced and the sequences were compared with the corresponding sequences in M129 registered with NCBI and the FH strain. The 70 resistant strains all showed a 2063 or 2064 site mutation in domain V of the 23S rRNA but no mutations in domain II. Site mutations of L4 or L22 can be observed in either resistant or sensitive strains, although it is not known whether this is associated with drug resistance.BioMed research international. 01/2014; 2014:320801.
- [Show abstract] [Hide abstract]
ABSTRACT: The aim of this study was to develop a single-nucleotide polymorphism (SNP) PCR assay to be performed directly on respiratory samples for the simultaneous detection of Mycoplasma pneumoniae and its 23S rRNA gene mutations, which are responsible for macrolide resistance. For multiplex SNP PCR, two outer primers for amplification of the 23S rRNA gene and two mutant-specific primers for the discrimination of single base changes were designed. A total of 73M. pneumoniae-positive samples and 100M. pneumoniae-negative samples were analyzed using this assay. By SNP PCR, we detected two mutations conferring high-level macrolide resistance in 22 samples (A2063G from 20 and A2064G from 2 samples); these results are identical to those produced by the 23S rRNA gene sequencing of M. pneumoniae-positive samples. Thus, this assay can be used as a practical method for the simultaneous detection of M. pneumoniae and mutations associated with macrolide resistance directly from respiratory samples.Journal of microbiological methods 04/2014; · 2.43 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: National treatment/diagnosis guidelines for lower respiratory tract infections (LRTIs) are generally based on local epidemiological data. Etiology and drug-resistance patterns could differ between China and European/American countries, and simply following their respective guidelines might cause problems in clinical practice. Therefore, we need to summarize the microbiology and clinical manifestations of LRTIs in China and develop our own guidelines. Three major national multicenter epidemiology surveillance studies on LRTI were completed recently. The data were compared in detail with those from European/American studies. Clinical and microbiological differences were observed in community-acquired pneumonia (CAP), hospital-acquired pneumonia (HAP), and pulmonary mycosis between our country and European/American countries. The microbiological and clinical characteristics of the major LRTIs in China differ in many respects from those in European/American countries. Patients should have personal treatment plans instead of simply following the guidelines from foreign countries.Journal of thoracic disease. 02/2014; 6(2):134-42.
Increased macrolide resistance of Mycoplasma pneumoniae in France
directly detected in clinical specimens by real-time PCR and melting
O. Peuchant1, A. Me ´nard2, H. Renaudin1, M. Morozumi3, K. Ubukata3, C. M. Be ´be ´ar1
and S. Pereyre1*
1Laboratoire de Bacte ´riologie EA 3671, Universite ´ Victor Segalen Bordeaux 2 and CHU de Bordeaux,
33076 Bordeaux cedex, France;2INSERM U853, Universite ´ Victor Segalen Bordeaux 2, Laboratoire de
Bacte ´riologie, 33076 Bordeaux cedex, France;3Laboratory of Molecular Epidemiology for Infectious Agents,
Kitasato Institute for Life Sciences, Kitasato University, Tokyo 108-8641, Japan
Received 13 February 2009; returned 10 March 2009; revised 1 April 2009; accepted 9 April 2009
Objectives: Mycoplasma pneumoniae is a common aetiological agent of community-acquired respiratory
tract infections for which macrolides are the treatment of choice. In France, only two macrolide-resistant
isolates were reported in 1999. In contrast, several recent data reported that macrolide-resistant
M. pneumoniae isolates have been spreading since 2000 in Japan. Mutations A2058G (Escherichia coli
numbering), A2058C, A2059G, A2062G, C2611A and C2611G in domain V of the 23S rRNA gene were
associated in vivo or in vitro with this resistance. The aim of this study was to determine whether
macrolide resistance of M. pneumoniae is emerging in France.
Patients and methods: We developed a duplex real-time PCR for the detection of the six 23S rRNA
mutations associated with macrolide resistance in M. pneumoniae and a simplex real-time PCR for the
identification of the A2058G mutation, the most common one. Both methods rely on fluorescence reson-
ance energy transfer coupled to melting curve analysis and are directly applicable to clinical samples.
The duplex real-time PCR assay, first validated on 40 genetically characterized M. pneumoniae strains,
was then applied directly on 248 French respiratory tract clinical samples.
Results: Among M. pneumoniae-positive specimens collected before 2005, no macrolide-resistant
M. pneumoniae isolate was detected. In contrast, among 51 samples collected between 2005 and
2007, five (9.8%) yielded a resistant genotype, suggesting a recent increase in macrolide-resistant
M. pneumoniae isolates in France.
Conclusions: The epidemiological monitoring of macrolide resistance in this species has become
necessary in France and Europe, and will be made easier by using these PCR assays.
Keywords: target gene mutation, 23S rRNA, laboratory methods, antimicrobial resistance epidemiology, low
respiratory tract infections, LRTI
Mycoplasma pneumoniae causes respiratory tract infections and is
responsible for up to 20% of all cases of community-acquired
pneumonia, especially among school-aged children and young
adults.1,2Owing to the lack of sensitivity and to the prolonged
time needed for M. pneumoniae detection by culture, molecular
techniques are currently used for the diagnosis of M. pneumoniae
lincosamide, streptogramin and ketolide antibiotics (MLSKs) can
be used for the treatment of M. pneumoniae infections. Macrolides
are generally considered as the treatment of choice in both children
Very few M. pneumoniae macrolide-resistant isolates have
been reported in the literature before 2000.3In Japan, several
recent data reported that macrolide-resistant M. pneumoniae iso-
lates have been spreading since 2000, with a prevalence increas-
ing up to 30.6% according to the studies.4–8In contrast, only
*Corresponding author. Tel: þ33-5-57-57-16-25; Fax: þ33-5-56-93-29-40; E-mail: email@example.com
Page 1 of 7
# The Author 2009. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved.
For Permissions, please e-mail: firstname.lastname@example.org
Journal of Antimicrobial Chemotherapy
Journal of Antimicrobial Chemotherapy Advance Access published May 9, 2009
two resistant isolates were reported in 1999 in France.9The
A2058G mutation in domain V of 23S rRNA (Escherichia coli
numbering) is the most frequent substitution associated with
macrolide resistance in clinical isolates, followed by the
A2059G mutation, while the A2058C and C2611G mutations
are rare.3Moreover, we previously showed that the C2611A and
A2062G mutations could be selected in vitro in the presence of
different MLSKs.10Current methods to detect macrolide resist-
ance in M. pneumoniae rely on time-consuming phenotypic or
genotypic methods, such as susceptibility testing, restriction
fragment length polymorphism (RFLP)4and 23S rRNA sequen-
cing analysis. Recently, a real-time PCR and high-resolution
melt (HRM) analysis-based technique has been developed to
detect only the A2058G or A2059G mutation.11
In order to determine whether macrolide resistance in
M. pneumoniae was emerging in France, we developed a rapid
method to detect all 23S rRNA mutations associated with macro-
lide resistance in M. pneumoniae, directly from clinical speci-
mens. Two real-time PCR assays using fluorescent resonance
energy transfer (FRET) with melting curve analysis were devel-
oped and applied on French clinical respiratory tract specimens.
Materials and methods
Bacterial strains and susceptibility testing
Two M. pneumoniae reference strains [M129 (ATCC 29342) and
FH (ATCC 15531)] and 14 susceptible clinical isolates obtained
from patients hospitalized at Pellegrin Hospital (Bordeaux, France)
were used. Two French9and nine Japanese4erythromycin-resistant
clinical isolates with characterized A2058G, A2059G, A2058C or
C2611G mutations in the 23S rRNA gene were used. Moreover,
eight Japanese in vitro-selected mutants4,8harbouring the A2058G
or the A2059G transition and five French in vitro-selected mutants10
with the A2062G or the C2611A substitution were tested.
MICs of erythromycin and azithromycin were determined in
Hayflick modified medium by an agar dilution method as previously
Twenty-one DNA extracts from Japanese respiratory tract clinical
specimens infected with M. pneumoniae macrolide-resistant isolates
harbouring the A2058G (n¼19) or the A2059G (n¼2) mutation
were used to validate the real-time PCR assays.6
We tested 142 respiratory tract samples, received between March
2007 and February 2008 at Pellegrin Hospital, for which the culture
and in-house diagnostic real-time PCR targeting the M. pneumoniae
adhesin P1 gene were negative. These samples were collected from
patients with a negative M. pneumoniae serology. Moreover, 106
respiratory tract clinical samples, collected in the same hospital,
with a positive M. pneumoniae detection by PCR, were tested.
Among them, 39 samples, received between January 2005 and
August 2008, yielded a M. pneumoniae isolate, for which erythro-
mycin and azithromycin MICs were determined. The 67 other speci-
mens, only M. pneumoniae PCR-positive, were systematically
collected between July 1998 and August 2008.
DNA was extracted from clinical specimens or from bacterial
cultures with the MagNA Pure LC kit (Roche, Meylan, France)
according to the manufacturer’s instructions.
Duplex real-time PCR
A duplex real-time PCR assay was performed on DNA extracted from
M. pneumoniae strains and specimens to detect point mutations
conferring resistance to macrolides. The method included duplex
amplification of two fragments of the M. pneumoniae 23S rRNA gene
with simultaneous hybridization of four probes and analysis of
melting curves. Primers and probes specific for the M. pneumoniae
23S rRNA gene were designed after alignment of published 23S
rRNA sequences from Mycoplasma genitalium, Mycoplasma hominis,
urealyticum. A basic local alignment search tool (BLAST) analysis of
each primer and probe was performed and all organisms that had a
significant BLAST hit were included in the analytical specificity
assays (see below). Primers F1-Mpn and R1-Mpn (Table 1) amplified
a 262 bp fragment, encompassing nucleotides at positions 2058, 2059
and 2062. Simultaneously, a 495 bp fragment was amplified with
primers F3-Mpn and R4-Mpn (Table 1), encompassing the 2611
Table 1. Primers and probes used for real-time PCR assays
Primer and probe designationSequence (50!30) Amplicon size (bp)
Duplex real-time PCR
Simplex real-time PCR
Underlined bases emphasize the nucleotides at position 2058, 2059, 2062 or 2611.
Peuchant et al.
Page 2 of 7
nucleotide. Two hybridization probe sets were designed for duplex
analysis of the two mutation regions in the 23S rRNA gene. Each
couple of probes included a sensor probe, 50-labelled with LC-Red
640 or LC-Red 705, which hybridized to the region containing the
mutation sites, and an anchor probe, fluorescein 30-labelled, which
hybridized to the three or four bases upstream from the former probe
(Table 1). Probes were obtained from Sigma-Aldrich (St-Quentin
The PCR and hybridization reactions were carried out in glass
capillaries by using the LightCycler 1.5 thermocycler (Roche).
Twenty microlitres of PCR mixture containing 2 mL of template
DNA, 1.6 mL of 25 mM MgCl2, 1 mL of the four primers (5 mM
each), 2 mL of the four probes (20 mM each) and 2 mL of FastStart
DNA Master Hybridization Probes (Roche) was prepared. The
cycling conditions consisted of an initial denaturation cycle at 958C
for 10 min, followed by 50 amplification cycles (with a transition
rate of 208C/s) consisting of 958C for 10 s, annealing at 588C for
20 s and extension at 728C for 20 s. After amplification a melting
step was performed, consisting of 958C for 1 s, 358C for 40 s and fol-
lowed by a slow rise in the temperature to 858C at a rate of 0.18C/s
with continuous acquisition of fluorescence, with a final cooling step
for 30 s at 408C. Data were analysed with the LightCycler software
version 1.5 (Roche). The option ‘polynomial’ was selected for the
calculation method for melting curve analysis and melting tempera-
ture (Tm) values were determined using the ‘manual Tm’.
Identification of the A2058G mutation by simplex
An additional simplex real-time PCR assay was used to identify the
A2058G mutation. Primers F1-Mpn and R1-Mpn were used with
two hybridization probes, designed on the reverse strand of the
M. pneumoniae 23S rRNA gene. The sensor probe, 30-labelled with
fluorescein, perfectly matched the 23S rRNA gene sequence har-
bouring the A-to-G transition at position 2058. The anchor probe
was 50-labelled with LC-Red 640 and 30-phosphorylated (Table 1).
PCR mixture and amplification were performed as described above.
Domain V of the 23S rRNA gene was amplified with primers
MP23S-17b and MP23S-23 from strain or specimen DNA extracts.10
Two fragments of interest in this domain were sequenced, one with
primers MP23S-11 and MP23S-22 encompassing positions 2058,
2059 and 2062, and one with primers MP23S-9 and MP23S-23
encompassing position 2611.10
Specificity and limit of detection of the assays
To assess the specificity of the assays, DNA extracts from
several Mycoplasma strains (M. genitalium, M. hominis, Mycoplasma
fermentans, M. penetrans, Mycoplasma pulmonis, Mycoplasma
faucium), Ureaplasma strains (U. urealyticum and U. parvum) and
bacterial strains from the respiratory tract (E. coli, Staphylococcus
Streptococcus anginosus, Streptococcus pneumoniae, Haemophilus
influenzae, Enterococcus faecalis, Corynebacterium urealyticum,
Neisseria perflava, Pseudomonas aeruginosa and Chlamydia pneumo-
niae) were used. In order to validate the efficiency of the extraction
step, PCRs targeting the 16S rRNA genes were performed on each
extract using primers F1-16S (CGTGTCGTGAGATGTTGGGTTA)
and R1-16S (GACGTCATCCCCACCTTCCT) for non-mycoplasmal
strains, and primers GPO3W (GGGAGCAAAYAGGATTAGATACC
mycoplasmal strains. To evaluate the detection limit of both assays,
M. pneumoniae M129 DNA was extracted and amplified with each of
the two primer sets F1-Mpn/R1-Mpn and F3-Mpn/R4-Mpn, respect-
ively. Each amplified product was then purified using the Wizardw
PCR Preps DNA purification system (Promega, Charbonnie `res,
France) and subsequently cloned in the plasmid vector pGEMw-T
Easy (Promega), according to the manufacturer’s instructions.
Plasmids pGEM-T MpnF1–R1 and pGEM-T MpnF3–R4 were trans-
formed into E. coli and purified from the E. coli transformants using
(Promega). Purity and concentration of each plasmid DNA were
determined by optical density measurements (NanoDrop 1000,
Wilmington, USA). The detection limit of both PCRs assays was
assessed by using 10-fold serial dilutions of plasmids pGEM-T
MpnF1–R1 and MpnF3–R4.
Detection of point mutations associated with macrolide
resistance by duplex real-time PCR
A duplex real-time PCR assay was developed to detect
point mutations conferring resistance to macrolides in the
M. pneumoniae 23S rRNA gene. This assay, first evaluated on
several genetically characterized M. pneumoniae strains, rapidly
distinguished the 16 wild-type strains from those with a resistant
genotype (24 strains). In the latter group, the existence of a
nucleotide mismatch between the gene sequence and the hybrid-
ization probe produced a Tmlower than the Tmof the wild-type
sequence. On the channel detecting the fluorescence emitted by
LC-Red 640, melting curve analysis of DNA from control strains
produced four different curves with Tms of 58.28C for the wild-
type strains, 54.18C for mutants harbouring the A2058C or the
A2062G substitution, 49.18C for mutants harbouring A2059G
and 48.28C for mutants harbouring A2058G (Figure 1a). Strains
harbouring the C2611G or C2611A substitutions produced the
same Tmas the wild-type strains and were not detected here. On
the second channel of the LightCycler, the LC-Red 705 fluor-
escence was measured. Two different melting curves were
obtained, with Tms of 64.58C for the wild-type strains and
60.58C for mutants C2611G or C2611A (Figure 1b). As
expected, strains harbouring mutations at positions 2058, 2059 or
2062 had the same Tmas the wild-type strains on this channel. It
should be noted that an additional peak, weakly intense, which
had a lower Tm of ?478C (Figure 1b), was observed in the
melting curve analysis detecting substitutions at position 2611,
but had no consequence on the result interpretation.
Identification of the A2058G mutation by simplex
Differentiating the A2058G and A2059G mutation-related Tms
could be difficult during the duplex real-time PCR assay.
Consequently, we developed an additional simplex real-time
PCR to detect specifically the A2058G mutation, which is the
most frequent one. The sensor-probe1-2058G was designed to
form a mismatch T:G with the wild-type sequence during
Detection of M. pneumoniae macrolide resistance by PCR
Page 3 of 7
hybridization. Therefore, the Tm for the wild-type strain was
lower than the one for the A2058G mutant, 60.28C versus
65.58C, respectively (Figure 1c). It should be noted that melting
peaks obtained with the A2058C, A2059G or A2062G substi-
tutions could not be differentiated from that of the wild-type
strain (Figure 1c).
Specificity and limit of detection of the real-time PCR assays
With both real-time PCR assays, no amplification was observed
with various mycoplasmal and non-mycoplasmal bacterial
species, although all DNA extracts were amplified with
the 16S rRNA PCRs. The detection limit of both the
simplex and duplex PCR assays was 10 copies/mL for each
Patient specimen testing
The duplex real-time PCR was first validated by testing 21 DNA
extracts from Japanese respiratory tract clinical specimens
infected with a macrolide-resistant M. pneumoniae isolate,
known to harbour the A2058G (n¼19) or the A2059G (n¼2)
mutation.6All DNA extracts were found to possess the expected
substitution using the duplex real-time PCR assay. Moreover,
the A2058G mutation was identified with the additional simplex
real-time PCR in all the 19 DNA extracts known to harbour it.
45 4060 65
Figure 1. Melting curve analysis obtained with the anchor-probe7/sensor-probe7 (a) and with the anchor-probe8/sensor-probe8 (b) by the duplex real-time
PCR, and with the anchor-probe1/sensor-probe1-2058G by the simplex real-time PCR (c) for the M129 (continuous line) and FH (small dashes) wild-type
strains and mutant isolates harbouring the A2058G (large and small dashes), A2058C (filled squares), A2059G (open circles) A2062G (crosses), C2611A
(filled diamonds) and C2611G (large dashes) mutations. Values on the y-axis represent the ratio of the first negative derivative of the change in fluorescence
(dF) to the variation in temperature.
Peuchant et al.
Page 4 of 7
Using the duplex real-time PCR, 142 clinical samples for
which no M. pneumoniae was detected by culture or by the
in-house diagnostic real-time PCR remained negative.
Among 39 specimens for which both the diagnostic real-time
PCR and culture were positive for M. pneumoniae, eight failed
to amplify and 28 produced a melting peak characteristic of the
wild-type genotype. These samples yielded M. pneumoniae iso-
lates susceptible to erythromycin and azithromycin, as deter-
mined by MIC studies (data not shown). Three clinical
specimens, named Mpn-3655, Mpn-3927 and Mpn-4276, col-
lected in 2005–06, were positive for resistance with the duplex
real-time PCR. An A-to-G substitution, at position 2058 or
2059, was detected for the Mpn-3655 and Mpn-3927 specimens,
according to the melting curve analysis (Table 2). Then, using
the simplex real-time PCR, the A2058G mutation was identified
in the Mpn-3655 clinical sample and confirmed by both
sequence analysis and macrolide MIC values. With the simplex
real-time PCR, the Mpn-3927 specimen produced a Tmlower
than that of the A2058G control strain, suggesting that the
A2058G transition was not present in this sample. The A2059G
transition was identified by sequencing analysis and confirmed
by MICstudies for the corresponding isolate MP-3996
(Table 2). Concerning the Mpn-4276 clinical sample, two
melting peaks were observed with the duplex real-time PCR,
one with a Tmcorresponding to a wild-type strain and one with
a Tm corresponding to an A2058G or A2059G mutation
(Table 2). With the simplex real-time PCR, two melting peaks
were also observed, one showing the presence of the A2058G
mutation and the other showing a lower Tm. Analysis of the 23S
rRNA sequence of the corresponding isolate MP-4391 showed a
mixture of bases A and G at position 2058, suggesting that wild-
type and mutated M. pneumoniae populations could be present
in the same sample. Eight subcultured clones obtained from the
MP-4391 isolate only harboured the A2058G mutation.
Among 67 clinical specimens, for which only the diagnostic
real-time PCR was positive for M. pneumoniae, 21 failed to
amplify with the duplex real-time PCR while 44 showed a sus-
ceptible genotype. Two clinical specimens, Mpn-4293 and
Mpn-4294, harboured a macrolide-resistant genotype. A mutation
at position 2611 was found for the Mpn-4293 bronchial aspirate
specimen (Table 2). The sequencing analysis from this sample
identified the C2611G transition. An A-to-G mutation, either at
position 2058 or 2059, was detected for the Mpn-4294 clinical
specimen with the duplex real-time PCR and the A2058G tran-
sition was identified using the simplex real-time PCR.
To summarize, among 106 M. pneumoniae PCR-positive
clinical samples collected over the last decade, 77 were ampli-
fied with the duplex real-time PCR, showing that the sensitivity
of our assay was 72.6% (77/106). Among them, a resistant
genotype was detected for five specimens, that is to say a fre-
quency of 6.5% (5/77). These five specimens were collected
between 2005 and 2007 and represented 9.8% (5/51) of the 51
M. pneumoniae-positive clinical samples amplified with our
duplex real-time PCR during this period.
Respiratory tract infections caused by M. pneumoniae are
empirically treated with macrolides. However, recent reports
indicate that macrolide-resistant M. pneumoniae isolates are
Table 2. Information about respiratory tract clinical samples with macrolide-resistant M. pneumoniae
Duplex real-time PCR result
Simplex real-time PCR result
A2058G or A2059G
A2058G or A2059G
no A2058G mutation
wild-type and A2058G or A2059G A2058G and non-A2058G genotypes
C2611G or C2611A
A2058G or A2059G
ERY, erythromycin; AZM, azithromycin; CLR, clarithromycin; JOS, josamycin; SPI, spiramycin; PRI, pristinamycin; TEL, telithromycin; AMX, amoxicillin; ND, not determined.
aErythromycin and azithromycin MICs were 0.006 and 0.0017 mg/L, respectively, for the M. pneumoniae M129 reference strain.
Detection of M. pneumoniae macrolide resistance by PCR
Page 5 of 7
spreading, which could modify the therapeutic management of
M. pneumoniae infections.6,13In M. pneumoniae, macrolide
resistance is associated in vivo with point mutations in domain
V of the 23S rRNA gene at positions 2058, 2059 and 2611.3
Currently, except sequencing analysis, few technologies have
Concerning M. pneumoniae, only a RFLP method4and, recently,
a real-time PCR followed by HRM curve analysis were reported
to detect A-to-G substitution at position 2058 or 2059.11
However, sequencing the 23S rRNA gene remained necessary to
identify the substitution and only those two mutations were
reported to be detected with this technology. In our study, we
describe a duplex real-time PCR for the detection of at least six
point mutations associated in vivo or in vitro with macrolide
resistance in M. pneumoniae. This method was applied directly
on clinical samples. Two probe sets were designed for multiplex
analysis of the two mutation regions in the M. pneumoniae 23S
rRNA gene. As the nucleotide environment of 2058, 2059 and
2062 positions was very rich in guanine and cytosine bases and
as mutations were situated in a region with many high-binding
secondary structures, it was difficult to design probes compatible
with a FRET technology. Consequently, the design of the probes
was handmade as software usually used to design real-time
assays produced no convincing results. Primers and probes were
chosen to be specific for the M. pneumoniae species. For this
purpose, the R4-Mpn primer was designed downstream of the
23S rRNA gene, in the 23S–5S intergenic space. Their speci-
ficity was confirmed since no positive reaction was obtained
with M. pneumoniae-negative samples or with a range of bac-
teria usually present in respiratory tract samples.
Our assay allowed the detection of the A2058G and A2059G
transitions, but their distinction was difficult because of the
close proximity of their respective Tms. As the A2058G mutation
is involved in ?90% of cases of macrolide-resistance in
M. pneumoniae,4–6we developed an additional simplex real-
time PCR, using the same technology, to accurately identify this
transition in clinical specimens.
It should be noted that our assays were less sensitive than the
in-house diagnostic real-time PCR performed at the Pellegrin
Hospital laboratory for M. pneumoniae detection in clinical speci-
mens. Indeed, among the 106 samples with a positive in-house
diagnostic real-time PCR, 29 (27.4%) failed to amplify. This
could be due to a difference in the target gene between the two
techniques. The present methods target the 23S rRNA gene,
which occurs in only one copy in the M. pneumoniae genome,14
while the in-house diagnostic real-time PCR targets the P1
adhesin gene, which is present in 8–10 copies.9The HRM tech-
nology also reported a limited sensitivity to detect point mutations
associated with macrolide resistance in M. pneumoniae, as seven
patient specimens failed to amplify among 30 PCR- and serology-
confirmed M. pneumoniae cases.11Thus, our real-time PCR
assays should be used secondarily, on clinical samples for which
the diagnostic real-time PCR would be positive.
Before 2000, very few macrolide-resistant M. pneumoniae
isolates had been reported worldwide.3Resistance to MLSKs
was reported in 2000 and spread rapidly in Japan, with 30.6%
(37/121 strains) resistant isolates described in 2006.6In a recent
M. pneumoniae outbreak in the USA, 27% (3/11) of isolates
harboured a macrolide-resistant genotype.11In France, only two
macrolide-resistant isolates were reported in 1999 among 155
collected between 1994 and 2006.9In our study, among 51
specimens amplified with our duplex real-time PCR and
collected between 2005 and 2007, 9.8% (5/51) yielded a
M. pneumoniae resistant genotype, suggesting a recent increase
of macrolide-resistant M. pneumoniae. This could also be due to a
better sensitivity of our genotype-based technique over phenotype-
based methods, which require strain isolation by culture before
in vitro susceptibility testing. Indeed, a M. pneumoniae isolate
was obtained from three of the five macrolide-resistant clinical
samples (Table 2). MLSK susceptibility testing was more fasti-
dious for these three isolates than for the macrolide-susceptible
strains. In agar medium, these isolates grew more slowly than
the control strain, yielding tiny and uncharacteristic colonies (data
not shown). These isolates were wrongly considered as susceptible
to MLSKs when the routine susceptibility testing was first realized.
As suggested for other bacterial species, in M. pneumoniae a
fitness cost could be connected with macrolide resistance-
associated mutations in the 23S rRNA gene15and could explain
these data. Indeed, it was reported that clarithromycin-resistant
H. pylori isolates, harbouring the A2058G or the A2059G
mutation, had a decreased fitness compared with wild-type iso-
lates.16This emphasizes the importance of developing genotypic
techniques to identify M. pneumoniae resistant isolates.
In our study, among the five patients infected with a
macrolide-resistant M. pneumoniae isolate, four of them received
a macrolide as the first antimicrobial treatment (Table 2). For all
of them, the initially prescribed treatment was changed to
another MLSK antibiotic, spiramycin, telithromycin or pristina-
mycin. In agreement with these observations, it was reported
that patients infected with macrolide-resistant M. pneumoniae
isolates were shown to have the initially prescribed macrolide
treatment more frequently changed to another antimicrobial
agent, e.g. minocycline or levofloxacin.6,13In our study,
neither tetracyclines nor fluoroquinolones were prescribed when
antimicrobial change occurred as these antibiotics are not rec-
ommended in children. It should be noted that the five patients
In M. genitalium, an urogenital pathogen phylogenetically
close to M. pneumoniae, macrolide-resistant isolates harbouring
mutations at position 2058 or 2059 were associated with treat-
ment failure.17In the majority of cases, drug-resistant mutants
were selected during macrolide treatment. In our study, as clini-
cal samples collected before treatment were not available for the
five patients infected with macrolide-resistant M. pneumoniae, it
was not possible to assess this hypothesis.
M. pneumoniae infections commonly occur in children and
young adults. In our series, 17 clinical samples were collected
from adults (16%). Among them, one macrolide-resistant
M. pneumoniae was detected in a 33-year-old man. To our
knowledge this is the first description of a macrolide-resistant
M. pneumoniae in an adult. Although the majority of studies were
conducted in paediatric specimens,4,5it was reported that no
macrolide-resistant M. pneumoniae had been observed among 30
isolates from adult patients with community-acquired pneumo-
nia.6Therefore, it would be of interest to include adults in a
global surveillance for the occurrence of macrolide-resistant
In conclusion, we have developed two real-time PCR assays: a
duplex real-time PCR for the detection of the 23S rRNA
point mutations associated with clinical macrolide resistance in
M. pneumoniae and a simplex real-time PCR for the identification
of the A2058G mutation, the most common mutation, both directly
Peuchant et al.
Page 6 of 7
applicable to clinical samples. The system of anchor and sensor
probes used in these assays is able to discriminate all the genotypes
in one reaction, contributing to the rapidity of the method.
According to the results of these assays on clinical specimens, we
have observed that macrolide resistance of M. pneumoniae is
increasing in France, as 9.8% of resistant genotypes were detected
between 2005 and 2007. Thus, epidemiological monitoring of M.
pneumoniae macrolide-resistant genotypes in clinical specimens
has become necessary in Europe, in children as well as in adults.
In addition, the rapid determination of the M. pneumoniae genoty-
pic resistance profile to MLSKs could allow the prompt prescrip-
tion of an alternativeantimicrobial
macrolide-resistant strain is detected.
We are greatly indebted to T. Sasaki for providing the
M. pneumoniae macrolide-resistant isolates from Japan and
thank D. Ayache from the Sigma-Aldrich company (Paris,
France) and D. Papi from the TibMolBiol company (Berlin,
Germany) for their assistance in designing probes.
This study was supported from internal funding.
None to declare.
1. Waites KB, Talkington DF. Mycoplasma pneumoniae and its role
as a human pathogen. Clin Microbiol Rev 2004; 17: 697–728.
2. Atkinson TP, Balish MF, Waites KB. Epidemiology, clinical mani-
festations, pathogenesis and laboratory detection of Mycoplasma
pneumoniae infections. FEMS Microbiol Rev 2008; 32: 956–73.
3. Be ´be ´ar CM, Pereyre S. Mechanisms of drug resistance in
Mycoplasma pneumoniae. Curr Drug Targets Infect Disord 2005; 5:
4. Matsuoka M, Narita M, Okazaki N et al. Characterization and
molecular analysis of macrolide-resistant Mycoplasma pneumoniae
clinical isolates obtained in Japan. Antimicrob Agents Chemother
2004; 48: 4624–30.
5. Morozumi M, Hasegawa K, Kobayashi R et al. Emergence of
macrolide-resistant Mycoplasma pneumoniae with a 23S rRNA gene
mutation. Antimicrob Agents Chemother 2005; 49: 2302–6.
6. Morozumi M, Iwata S, Hasegawa K et al. Increased macrolide
resistance of Mycoplasma pneumoniae in pediatric patients with
community-acquired pneumonia. Antimicrob Agents Chemother 2008;
7. Okazaki N, Ohya H, Sasaki T. Mycoplasma pneumoniae isolated
from patients with respiratory infection in Kanagawa Prefecture in
1976–2006: emergence of macrolide-resistant strains. Jpn J Infect Dis
2007; 60: 325–6.
8. Okazaki N, Narita M, Yamada S et al. Characteristics of
macrolide-resistant Mycoplasma pneumoniae strains isolated from
patients and induced with erythromycin in vitro. Microbiol Immunol
2001; 45: 617–20.
9. Pereyre S, Charron A, Renaudin H et al. First report of
macrolide-resistant strains and description of a novel nucleotide
sequence variation in the P1 adhesin gene in Mycoplasma pneumo-
niae clinical strains isolated in France over 12 years. J Clin Microbiol
2007; 45: 3534–9.
10. Pereyre S, Guyot C, Renaudin H et al. In vitro selection and
characterization of resistance to macrolides and related antibiotics in
Mycoplasma pneumoniae. Antimicrob Agents Chemother 2004; 48:
11. Wolff BJ, Thacker WL, Schwartz SB et al. Detection of macrolide
resistance in Mycoplasma pneumoniae by real-time PCR and high-
resolution melt analysis. Antimicrob Agents Chemother 2008; 52:
12. Waites KB, Be ´be ´ar CM, Robertson JA et al. In: Nolte FS, coordi-
nating ed. Cumitech 34, Laboratory Diagnosis of Mycoplasmal
Infections. Washington, DC: American Society for Microbiology, 2001.
13. Suzuki S, Yamazaki T, Narita M et al. Clinical evaluation of
Chemother 2006; 50: 709–12.
14. Himmelreich R, Hilbert H, Plagens H et al. Complete sequence
analysis of the genome of the bacterium Mycoplasma pneumoniae.
Nucleic Acids Res 1996; 24: 4420–49.
15. Andersson DI. The biological cost of mutational antibiotic resist-
ance: any practical conclusions? Curr Opin Microbiol 2006; 9:
16. Bjorkholm B, Sjolund M, Falk PG et al. Mutation frequency and
biological cost of antibiotic resistance in Helicobacter pylori. Proc Natl
Acad Sci USA 2001; 98: 14 607–12.
17. Jensen JS, Bradshaw CS, Tabrizi SN et al. Azithromycin treat-
ment failure in Mycoplasma genitalium-positive patients with nongono-
coccal urethritis is associated with induced macrolide resistance. Clin
Infect Dis 2008; 47: 1546–53.
Detection of M. pneumoniae macrolide resistance by PCR
Page 7 of 7