Background
Mycoplasma pneumoniae (M. pneumoniae) can cause atypical pneumoniae and many respiratory illnesses, particularly in childhood and specify young adults, its infection rate range from 10% to.
80% [
1‐
3]. Macrolides and related antibiotics have been generally considered to be the first-choice antibiotic for the treatment of
M. pneumoniae infection [
4]. However, since the first isolates macrolide-resistant
M. pneumoniae (MRMP) in 2001 [
5], MRMP has been spreading globally for about seventeen years, with prevalences ranging from below 10% in Europe [
1,
6‐
10], approximately 30% in Israel [
11], and up to 90% in Asia [
12‐
14].
The commonly used phenotypic methods for determining susceptibility to macrolides is to measure the minimum inhibitory concentration (MIC), and this process is rather time-consuming.
In previous studies, point mutations in several positions have been reported to be related to development of macrolide resistance in
M. pneumoniae [
15]. Among them, the 2063 and 2064 point mutations in the peptidyl-transferase loop of domain V of 23S rRNA, which interferes with the binding of macrolides to rRNA [
1,
5,
16,
17], are the major mutations responsible for different grade macrolide-resistance of
M. pneumoniae [
5,
17]. For example, one study reported that 90.90% had an A2063G transition and 9.10% had an A2064G transition in domain V of the 23S rRNA gene among 55 macrolide-resistant strains [
18]. Other mutations, such as those at positions 2617, 2067 and 2611 in domain V of the single-copy 23S rRNA gene, and mutations in the ribosomal proteins L4 and L22, are very rare [
19,
20]. Several molecular detection methods for identification of these mutations include sequencing of PCR products, real-time PCR, pyrosequencing, restriction fragment length polymorphism (RFLP) analysis, high-resolution melting curve analysis and allele-specific PCR [
21‐
26]. However, most methods have limitations as following: sequencing of PCR products and RFLP analysis are time-consuming and expensive, and have no value in clinical practice due to only used in research; real-time PCR may be much easier, quicker, and simpler to perform than other methods, but may not be able to distinguish between the 2063 and the 2064 mutation without additional simplex real-time PCR, and cannot detect unknown mutations [
8,
19]. Therefore, a rapid, sensitive, and specific laboratory test is vital for rapid detection of
M. pneumoniae infections.
Allele-specific real-time PCR (ASPCR) applies the real-time PCR to allele-specific PCR and integrates the advantages of these two systems. ASPCR has increased the sensitivity of the allele-specific PCR several-fold and quantified the PCR products [
27,
28]. Therefore, ASPCR is a highly sensitive and time-saving method for detection of point mutations and is significantly labor-intensive and reproducible. In this study, we developed ASPCR assays for detecting 23S rRNA gene of
M. pneumoniae and determine the macrolide resistance-associated mutations at 2063 (A2063G) and 2064 (A2064G) sites. In addition, we detected 178 pharyngeal swab specimens using ASPCR to reveal the prevalence of macrolide-resistant and sensitive
M. pneumoniae quasispecies in clinical specimens.
Methods
Specimens and reference strains
A total of 178 pharyngeal swab specimens were tested using ASPCR. All specimens were collected from pediatric patients (aged, 2–12 years) with a clinical diagnosis of M. pneumoniae infection from August 2013 to March 2015. The reference strain M129 (ATCC 29342) used in this study was preserved in our laboratory. The specificity of the ASPCR was tested using DNA of the following reference strains: Neisseria mucosa, Klebsiella pneumoniae, Escherichia coli, Staphylococcus aureus, Streptococcus pneumoniae, Pseudomonas aeruginosa, Mycoplasma hominis, Mycoplasma fermentans, Mycoplasma pyriformis and Ureaplasma urealyticum. All reference strains were preserved in our laboratory.
Primer design
All the primers were designed based on the sequence of
M. pneumoniae reference strain M129 (GenBank accession no. U00089). The mutant-specific primers (Msp) and non-specific primers (Np) were designed to determine the macrolide resistance-associated mutations at 2063 (A2063G) and 2064 (A2064G) site. The mutant-specific primers incorporate the target mutation in their 3′-end and plus three intentional mismatch bases (hypoxanthylic acid) near the end to enhance the specificity of mutant-specific amplification (Table
1). The non-specific primer was identical to the corresponding mutant-specific primer, except that the sequence ends right before the mutation position and without the substitutions of hypoxanthylic acid. The same reverse primer was used in the mutant-specific and the non-specific reactions performed in separate wells.
Table 1
Oligonucleotide sequences and locations on the M129 genome
23S rRNA-F | CTTTCTAATGGAGTTTTTTACTT | 119,806—119,828 |
23S rRNA-R | GCTTGGTGCTTTCCTATTCT | 123,068—123,087 |
23SA2063G-F | GGACGGGAAGACCCCGTGAAGCTTTACT | 122,094—122,121 |
23SA2064G-F | GGACGGAGAGACCCCGTGAAGCTTTACT | 122,094—122,121 |
23S2063/2064-R | CGTTGCGCCTAACGGGTGTCTTCAC | 122,069—122,093 |
NU-F (Np) | TTAGGCGCAACGGGACGG | 122,082—122,099 |
2063MU-F (Msp) | TTAGGCGCAACGGGAIIIG | 122,082—122,100 |
2064MU-F (Msp) | TTAGGCGCAACGGGAIIIAG | 122,082—122,101 |
23S2063/4D-R | CTGGATAACAGTTACCAATTAGAACAGC | 122,233—122,260 |
Construction of standards and ASPCR amplification
The standards of the ASPCR assays for A2063G and A2064G were then constructed. Plasmids containing a wild-type fragment of M. pneumoniae 23S rRNA full-length sequence were obtained by cloning the PCR products amplified by the primers 23SrRNA-F and 23SrRNA-R into pMD18-T (pMD™18-T Vector Cloning Kit, TaKaRa, Japan) vectors. The mutations of A2063G and A2064G were introduced into wild-type plasmids by sit-directed mutagenesis (TaKaRa MutanBEST Kit, TaKaRa, Japan) using primers 23SA2063G-F, 23SA2064G-F, and 23S2063/2064-R. The plasmids containing mutations were confirmed by sequencing and quantified by spectrophotometer. Serial 10-folds dilutions of plasmids ranging from 10 to 106 copies/μL were made as standards for ASPCR.
The mutant-specific and nonspecific standard curves of A2063G and A2064G were obtained by amplifying the mutant standards with the corresponding primer sets. The ASPCR mix contained 12.5 μL power SYBR Green PCR Master 2 × Mix, 2 μL mutant-specific/non-specific upstream primer, 2 μL corresponding downstream-primer, 2 μL DNA templates, 6.5 μL ddH2O. The reaction conditions were 50 °C for 2 min, followed by a denaturation step at 95 °C for 10 min, 40 cycles of real-time PCR amplification (95 °C for 15 s, 60 °C for 1 min) with a final dissociation stage (95 °C for 15 s, 60 °C for 30 s, 95 °C for 15 s). Each reaction was conducted in triplicate.
Evaluation of ASPCR
The threshold cycle (Ct) values of 107 copies/μL of mutant and wild-type plasmids with the specific primer were tested and calculated the ΔCt. The mixture templates were generated by adding 105 copies of wild-type DNA to serial dilutions (106–101 copies) of mutant DNA. We compared the Ct values of the mutant template and mixture of mutant and wild-type plasmids to evaluate the discrimination ability of ASPCR.
The mixture templates in which the proportion of mutant plasmids ranged from 0.01 to 100% were generated by adding 105 copies of wild-type DNA into the serial dilutions (106–101 copies) of mutant DNA. The cut-off value was defined as the mean Ct value plus three standard deviations (SD) of 12 independent determinations of 105 copies wild-type template with mutant-specific primer. Mutants were inferred to be present in the mixing templates when the Ct value was less than the cutoff value. The measured proportions and nominal proportions of the mixtures with mutant template ranging from 0.01 to 100% were compared to evaluate the accuracy of ASPCR. The sensitivity of ASPCR was determined by the cutoff value and accuracy. The coefficient of variations (CVs) of intra-assay and inter-assay were calculated to evaluate the reproducibility of ASPCR.
Detection of clinical specimens
Genomic DNA was extracted from M129 strain-enriched culture solution and pharyngeal swab specimens using the QIAamp® DNA Mini Kit (QIAGEN, Shanghai, Germany), according to the manufacturer’s protocol. The DNA fragment in domain V of
M. pneumoniae 23S rRNA region were detected using nested PCR method as previously reported [
29,
30] and ASPCR as above. Each sample was amplified by mutant-specific and non-specific primers of each mutation separately and the amplifications ran in duplicate. The corresponding standards of the mutations were tested in each plate, and the standard cure was drawn every time for quantification. The Ct values of the amplifications of clinical samples using specific and non-specific primer were interpolated into the corresponding standard curves to get the numbers of mutant DNA and the total DNA. Then the proportions of A2063G and A2064G MRMP quasispecies populations were calculated.
Statistical analysis
All statistical analysis was performed using IBM SPSS Statistics 19. Continuous variables were compared with the Student’s t test or Analysis of Variance, and categorical variables were compared with the Chi-square test. To evaluate the coincidence ratio between the ASPCR and nested PCR assay, the kappa coefficient was calculated using Chi-square test. A P-value of < 0.05 were considered to statistical significance.
Discussion
Comparing with the routine drug resistance detecting assays, ASPCR is a highly sensitive, accurate, time-saving and high throughput method for detecting resistant
M. pneumoniae and analyzing the mutation frequency. The sensitivity and accuracy of ASPCR were directly proportional to the discrimination ability of the specific primers, which was strongly influenced by the particular 3′ end base sequence. The bases near the 3’end of the specific primer were replaced by hypoxanthine in order to enhance the specificity of the ASPCR assays [
27]. The number and location of hypoxanthine were critical for the discrimination ability of the specific primer. In this study, each ASPCR assay could detect more than 10 copies/reaction of 23S rRNA gene and the presence of mutant species with the sensitivity down to 0.1%. Although the mismatch occurred at the 3’end of the specific primer, the measured proportions of detecting A2063G and A2064G assays were accurate down to 0.1%, and the intra-assay and inter-assay CVs of the ASPCR were below 0.18.
The main limitation of the ASPCR is that only one mutation can be detected in every reaction, while many mutations can be tested at once using nested PCR with sequencing. Prior study had demonstrated that the polymorphisms that occur in the specific primer binding sites can significantly impair the accuracy of ASPCR assays [
27]. The polymorphism must be taken into account when testing clinical samples and the prior sequencing is suggested to overcome this limitation, while ASPCR was applied in detecting the drug-resistance associated mutations of HIV [
31]. The polymorphism had little effect on the detection of A2063G and A2064G mutations of
M. pneumoniae, because of the lower variability of
M. pneumoniae compared with HIV and the high conserved sequence near these mutations. Considering this, it is more suitable to apply ASPCR detecting mutations of
M. pneumoniae than that of other hypervariable virus.
Previous, Chan et al. [
32] compared the detecting result of low-frequency MRMP quasispecies that obtained by pyrosequencing with those obtained by Sanger sequencing and SimpleProbe PCR coupled to melting curve analysis on respiratory specimens. The results indicated that pyrosequencing identified A2063G MRMP quasispecies populations in 78.8% (67/88) of the specimens, and only 38.8% (26/67) of these specimens with the A2063G quasispecies detected by pyrosequencing were found to be A2063G quasispecies by Sanger sequencing or SimpleProbe PCR. In this study, the results showed 96.64% of the resistant specimens had A2063G mutation, 35.57% (53/149) with both A2063G and A2064G mutations that was not tested by other method in previous study, such as Chan et al. [
32], Lin et al. [
19] and Ji et al. [
21], which might indicate ASPCR is a highly sensitive, accurate, time-saving and high throughput method for detecting resistant
M. pneumoniae. Of the 164
M. pneumoniae positive samples, 61.59% had the mixing of wild-type and drug resistant
M. pneumoniae, and 56.44% of the latter contained the drug resistance mutations at low frequency (≤50%). These results of ASPCR indicated that sensitive and resistant quasispecies coexisted in most of the
M. pneumoniae-positive samples, and the resistant mutations could be at a relative low frequency. These finds were important directive significance for the clinical management. Furthermore, compared to the nested PCR with sequencing, ASPCR testing has short the turnaround time, is highly sensitive for testing
M. pneumoniae and is able to discriminate the samples with resistant
M. pneumoniae at a very low frequency
. All these make the ASPCR an attractive method for the highly sensitive and rapid diagnosis of
M. pneumoniae. The study of minor resistant variants in
M. pneumoniae infection is relevant to understanding the mechanisms of the generation and development of drug resistance. It is also very important for clinical management to test and monitor the drug-resistance associated mutations.
There are several limitations to our study. Firstly, the sample size was small. Secondly, clinical sample information was incomplete collection, which may lead to inaccurate statistical analysis. Thirdly, the specificity of the ASPCR assay was only used to test on the strains that associated with respiratory infections, not on respiratory tract specimens. Therefore, in future, a large sample size with complete clinical data is needed to further study to confirm the study, and the related factors between M. pneumoniae resistance and infection are needed to analyze.
Conclusions
In generally, due to its highly sensitivity and accuracy, the ASPCR assays can be used as a particularly useful tool for resistance surveillance and studying the mechanism of the generation and development of drug resistance. In future study, the ASPCR assays can be used to characterize the dynamics of mutants in vivo by measuring the proportion of A2063G and A2064G variants in serial pharyngeal swabs specimensying and study the kinetics of selection and decay of point resistance mutations. For future application, ASPCR can also be easily developed for detecting other mutations associated with drug resistance of M. pneumoniae using proper standards and primers.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.