Antimicrobial susceptibility and molecular characteristics of Mycoplasma pneumoniae isolates across different regions of China

Background In China mainland, most Mycoplasma pneumoniae related studies are carried out in Beijing and Shanghai, while rare studies are performed in the other regions. In this study, we analyzed the molecular biology characteristics and antimicrobial susceptibility of clinical isolates of M. pneumoniae from 5 regions between January 2017 and December 2018. Methods Genotyping was performed to 154 M. pneumoniae isolates from 5 cities using PCR and multiple-locus variable-number tandem repeat analysis (MLVA) method. Antimicrobial susceptibility test was performed to all the isolates against 4 antibiotics. Sequencing was performed to the amplification products of the 23S rRNA drug resistant gene. Results Genotype I was detected in 118 M. pneumoniae isolates (76.6%), and genotype II was identified in 36 isolates (23.4%). The majority (92.2%) of the MLVA genotypes were 4–5–7-2 and 3–5–6-2, which represented the genotype I and II, respectively. The total macrolide (ML) resistance rate was 79.7%. The minimum inhibitory concentration (MIC) of the erythromycin was in a range of 128- > 256 μg/ml, while that for the azithromycin was 2-32 μg/ml. There were mutations in the 23S rRNA in each ML resistance isolate. Jilin city showed the highest prevalence of genotype I (100%) and ML resistance rate (100%), while Jinan showed the lowest prevalence of genotype I (45.5%) and ML resistance rate (54.5%). Conclusions A large variance was identified in the M. pneumoniae genotype and ML resistance among the 5 cities. The proportion of M. pneumoniae with a genotype II genotype (3–5–6-2) showed an increased trend. Electronic supplementary material The online version of this article (10.1186/s13756-019-0576-5) contains supplementary material, which is available to authorized users.

virulence factors. This leads to urgent demands on the epidemiological and molecular analysis.
In China mainland, most of the M. pneumoniae related studies are conducted in Beijing and Shanghai [15][16][17][18][19], while extremely rare studies are performed in the other regions. In a recent study, Xue et al. [20] firstly reported the M. pneumoniae genotype and antimicrobial resistance in 6 regions simultaneously. However, the M. pneumoniae genotype and antimicrobial resistance were derived from M. pneumoniae DNA detection, not from the clinical isolates. In this study, we analyzed the genotype and antimicrobial susceptibility of clinical isolates of M. pneumoniae from 5 regions between January 2017 and December 2018.

M. pneumoniae strains
M. pneumoniae strains were provided by the National Institute for Communicable Disease Control and Prevention, Chinese Center for Disease Control and Prevention. A total of 154 strains were collected from 267 pediatric patients preliminarily confirmed with M. pneumoniae pneumonia in local hospitals in 5 cities from north to south of China, including Jilin (Jilin Province, China), Beijing, Jinan (Shandong Province, China), Fuyang (Anhui Province, China) and Soochow (Jiangsu Province, China) (Fig. 1). Each throat swab specimen was cultured in 2 ml Mycoplasma selective liquid media (Bioreal Coming) at 37°C. In cases of media color changing from red to yellow, 0.1 ml suspension was transferred onto agar to subculture. Then M. pneumoniae isolates were purified using dilution technique. Nucleotide identification from each purified isolate was verified by Real-Time PCR as previously described [21].

Real-time PCR-based genotyping
Genotyping of M. pneumoniae was performed by Real-Time PCR. Amplification of the specific regions of the genotype from all isolates was performed as previously described [22]. Real-time PCR was performed on the   [23].

Antimicrobial susceptibility testing
Minimum inhibitory concentrations (MICs) against 4 antibiotics (i.e. erythromycin, azithromycin, levofloxacin and tetracycline) (Sigma-Aldrich, CA, USA) were determined using SP4 broth (Remel) based on the micro-dilution methods. M. pneumoniae M129 (ATCC 29342) and ICDC P028 (clinical isolate) served as macrolide-susceptible and -resistant controls, respectively. MIC was defined as the lowest concentration of antimicrobial agents that induced a color change in the control media. The MIC test was conducted based on the latest version of CSLI M43-A (2017 version). Sterilized liquid paraffin oil was utilized to seal the micro-plates, and then was incubated at 37°C for 5 days. All the susceptibility tests were performed at least in triplicate.

Amplification and sequencing domains V of the 23S rRNA gene
Domain V of the 23S rRNA gene was amplified using the primers according to the previous study [24]. Each amplification was performed in a total volume of 20 μl, containing 2 μl 10× Ex Taq Buffer (Mg 2+ plus), 0.4 μM primers, 1.6 μl dNTP mixture, 0.2 μl TaKaRa Ex Taq (5 U/μl), and 1 μl of DNA template. Amplification was performed using the following conditions: 95°C for 3 min; followed by 30 cycles of 95°C for 30 s, 55°C for 30 s, and 72°C for 60 s; and 72°C for 5 min. After purification, the obtained products were sequenced using Sanger technique (Sangon Biotech, Beijing, China).

Statistical analysis
The data were entered into Excel 2007 sheet. Then SPSS 17.0 software was utilized for the statistical analysis. Chi square test was performed for the analysis of genotyping and drug resistance of strains isolated from different regions. P value of less than 0.05 was considered to be statistically significant.

Real-time PCR genotyping
Genotype I was detected in 118 M. pneumoniae isolates (76.6%), and genotype II was identified in 36 isolates (23.4%, Table 1). The proportion of genotype I and II showed differences across various cities. The proportion of genotype I and II in Beijing was 60.8% (31/51) and 39.2%, respectively. The proportion of genotype I and II in Soochow was 91.7% (22/24) and 8.3% (2/24), respectively. In Fuyang, the proportions for genotype I and II were 75% (6/8) and 25% (2/8), respectively. In Jinan, the proportions were 45.4% (10/22) and 54.5% (12/22), respectively. In Jilin, the proportion for genotype I was 100% (49/49), and no genotype II was identified. Statistical analysis indicated that there were significant differences in the genotype of the M. pneumoniae in different regions (χ 2 = 37.068, P < 0.05). Taken together, the distribution of the two genotypes was different in these cities ( Fig. 1).

Discussion
We analyzed the molecular biological features and antimicrobial susceptibility of M. pneumoniae isolates in 5 cities in China mainland between January 2017 and December 2018. As is known to all, M. pneumoniae is classified into genotype I and genotype II according to the genome and proteomics data [25][26][27].  (Table 3). Recently, Xue et al. [20] investigated the MLVA genotypes of M. pneumoniae in Kunming City, which indicated that there was a base deletion (7 bp) in the repeats of Mpn15 in two patients. This imply that there are indeed some limitations in the current MLVA genotyping loci. In this study, the most common M. pneumoniae genotype was genotype I (4-5-7-2), while the rest was genotype II (3-5-6-2). There were significant differences in the genotyping among the five cities. The proportion of genotype I in Jilin city was 100%, while that in Jinan was merely 45.5%. According to the previous description [20], in geographic locations, Jilin and Harbin localizing in the Northeast of China with a linear distance of about 200 km showed a genotype I prevalence of 100 and 71.4%, respectively. In addition, the variations of genotypes in Nanjing and Shanghai with a linear distance of about 280 km were extremely large. Soochow, localizing at the middle of Nanjing and Shanghai in geography, showed a closer genotype I prevalence to Shanghai (91.7% vs. 80%). The most high prevalence of M. pneumoniae genotype II in the five cities was Nanjing, with a prevalence of up to 80%. In this study, the prevalence of genotype II in Beijing and Jinan, localizing at the north of Nanjing, was 54.5 and 33.3%, respectively (Fig. 2). This implied that the distance may not be an important cause for variations in the M. pneumoniae genotypes. The highest ML resistance proportion of isolates from Jilin, with the highest prevalence of genotype I, was 100%. Jinan, with the lowest prevalence of genotype I, showed the lowest ML resistance proportion of 54.5%. It seemed that ML resistance was correlated with the M. pneumoniae genotype (Tables 1 and 3, Fig. 1). Such phenomenon had been reported by other studies in China mainland [15,17,37,38] rather than the other countries [24,39,40]. To date, the ML resistance of M. pneumoniae has been acknowledged to be associated with the mutations in 2617, 2063 and 2064 loci in 23 s rRNA [1,41]. In this study, mutations on 2063 and 2064 loci were identified in the ML resistant strains, and no mutations were found in the ML susceptible strains. In a previous study, Okaziki et al. reported transition of ML susceptible to ML resistant M. pneumoniae strains in the presence of ML exposure. Meanwhile, there were point mutations at 2,063 and 2,064 of the 23 s rRNA gene, which resulted in stable phenotype-related antimicrobial resistance [42]. Therefore, the ML resistance of M. pneumoniae may be related to the antibiotics pressure to some extent, which was a gradual process. Nowadays, ML antibiotics have been commonly utilized in Chinese population, especially children. What's more, abuse of ML antibiotics is also reported in some regions. In this study, the time span was short and was similar with the cross-sectional studies. Our data might not reflect the correlation between ML resistance and genotype of M. pneumoniae. However, our team performed 10-year cohort studies on the molecular characteristics and antimicrobial susceptibility of M. pneumoniae in Beijing. Between 2008 and 2012, the predominant genotype of M. pneumoniae was genotype I [15,43]. In 2013, the proportion of genotype II M. pneumoniae showed a trend of increase [44]. Between 2014 and 2016, the proportion of genotype II M. pneumoniae was in a range of 30 to 40% [45]. In this study, the proportion of genotype II M. pneumoniae in Beijing between 2017 and 2018 was 39.3%. The prevalence of antimicrobially resistant genotype II M. pneumoniae in Beijing increased from 0 to 20% over the past decade [44,45], which indicated that the prevalence of genotype II M. pneumoniae showed a gradual trend in the presence of antibiotics exposure. It seemed that the total antimicrobial resistance rates showed decline in Beijing [20,45], but in fact, the antimicrobial resistance proportion of genotype I strains was high with no trend of decline, while the proportion of antimicrobial resistance in genotype II showed gradual increase. The decline of total antimicrobial resistance rates may be related to the rapid increase of genotype II M. pneumoniae in Beijing within these years compared to the increase of its antimicrobial resistance. In 2016, Xue et al. [20] reported the proportion of genotype II M. The MIC of each agent was defined as the lowest concentration of each antibiotic that prevented the color change ERY erythromycin, AZM azithromycin, LVX levofloxacin, TET tetracycline pneumoniae in Beijing was lower than the reported data (10.9% vs. 39.3%), which may be related to the variations induced by population and determination methods. In future, more persistent studies are required in order to obtain more accurate data. Although the total ML resistance rate of M. pneumoniae among the 5 cities were in a range of 54.4 to 100%, the resistance rate of genotype I M. pneumoniae after genotyping was higher than 90% in all these regions ( Table 4). The major causes for variations of the ML resistance were proportion and antimicrobial resistance rate of genotype II strains. There were only 2 genotype II strains isolated from Fuyang and Soochow, and then data analysis was not performed. The antimicrobial resistance rate of the genotype II strains (MLVA, 3-5-6-2 genotype) in Beijing and Jinan were all higher than 20%. This indicated that more studies are required to support the correlation between genotype II (MLVA, 3-5-6-2 genotype) and ML resistance. The proportion of genotype I and II strains in Jinan was similar. However, the rate of antimicrobial resistance of genotype I was 90%, while that of the genotype II was merely 25%. On this basis, we speculated that there might be a transmission between genotype I and II in Jinan, rather than the transmission from genotype II to genotype I. Our future studies will focus on it. Our study contributes to the understanding on the molecular epidemics of M. pneumoniae in China mainland, the drug resistance and clinical medication in different regions. In future, more studies involving a large sample size and more monitoring sites are required.

Bias and limitations
There are some limitations in our study. Firstly, the sample size of strains in some regions was small. Secondly, the study duration was relatively short. It was similar to the cross-sectional study lacking of cohort studies. Thirdly, we only included five cities in this study, and our data could not represent the whole situation of M. pneumoniae infection in China mainland.  Fig. 2 Locality map of ten cities included in this study(•) and newly publication(▲ Ref [21]). The genotyping and ML resistance were nearly the same in the cities in the cycle

Conclusions
In summary, we investigated the molecular characteristics of M. pneumoniae in 5 cities between January 2017 and December 2018. A large variance was identified in the M. pneumoniae genotype and ML resistance among the 5 cities. The proportion of M. pneumoniae with a genotype II (3-5-6-2) showed an increased trend, with an antimicrobial resistance of 0-25%. In future, more studies are required to confirm the correlation between specific M. pneumoniae genotype and ML resistance.

Additional file
Additional