Introduction

Mycoplasma pneumoniae (M. pneumoniae) is a major pathogen responsible for community-acquired pneumonia, with a genome size of approximately 800 kb has remained largely stable over time and across geographic regions1. Epidemiological data show that epidemic peaks of M. pneumoniae occur at intervals of 3–7 years and can persist for 1–2 years, as seen in outbreaks in Japan and Europe in 2011–2012, 2014–2015, and 2015–20162. From 2017 to 2020, the global incidence of M. pneumoniae, confirmed through direct testing methods, was reported at 8.61% across all age groups3. However, during the COVID-19 pandemic (2020–2021), the incidence decreased to 1.69% due to non-pharmaceutical interventions. Despite the lifting of COVID-19 restrictions, a continued reduction in incidence was observed across 20 countries in Europe, Asia, the Americas, and Oceania, lasting into 20224. More recently, since mid-October 2023, M. pneumoniae infections have sharply increased in China5,6,7, as well as in Spain and Denmark8,9. Several studies have investigated the genotypic variations and macrolide resistance of M. pneumoniae strains in southern China; however, further research is necessary to understand the underlying causes of this sudden surge.

M. pneumoniae strains are typically classified into two major genetic groups, P1 type 1 (P1-1) and P1 type 2 (P1-2), based on nucleotide differences in the repetitive elements RepMP2/3 and RepMP4 within the MPN141 gene10. Moreover, P1-1 linage have P1-1 and P1-1a variants and the P1-2 lineage can be further divided into P1-2, P1-2a, P1-2b, P1-2bv, P1-2c, P1-2d, P1-2e, P1-2f, P1-2 g, P1-2 h, P1-2i11. The P1 protein in the polarized attachment organelle serves as the primary adhesin and is crucial for the pathogen’s virulence. Studies on M. pneumoniae typing and antibiotic susceptibility have shown that different P1 gene types may be linked to varying degrees of macrolide resistance. Specifically, P1-2 strains may be more susceptible to macrolides compared to P1-1 strains, making it important to monitor the molecular epidemiology of M. pneumoniae infections. Genotypic variations may influence macrolide resistance, disease severity, and the pathogen’s periodic outbreaks and epidemics. Macrolides remain the first-line treatment for M. pneumoniae infections in children in China12.

The emergence of macrolide-resistant M. pneumoniae in Japan in the early 2000s, with resistance rates exceeding 90% within a decade, led to its spread across Asia, Europe, and North America13. The primary mechanism of macrolide resistance involves mutations at positions 2063 and 2064 in domain V of the 23 S rRNA gene. PCR-based surveillance data has also reported a significant increase in infections caused by macrolide-resistant P1-2 strains in northern China from 2021 to 202214. Without the restrictions imposed during the COVID-19 pandemic, P1-2 macrolide-resistant clones could likely have triggered nationwide outbreaks as early as 2020. Additionally, the epidemic cycle of M. pneumoniae may be influenced by herd immunity, which tends to last around four years before populations become susceptible to reinfection.

In the present study, we sequenced the genomes of 421 M. pneumoniae strains isolated from various regions in China and compared them to 147 publicly available genomes. This multicenter, retrospective study aimed to analyze the genotypes and macrolide-resistance mutations of M. pneumoniae across four hospitals in Liaoning, Zhejiang, Wuhan, and Guizhou. Infection rates in 2023 ranged from 19.4% to 35.9%, reflecting a significant surge in cases. Genomic sequencing of the 421 M. pneumoniae-positive samples revealed that the P1-1 genotype was the predominant type (87.3% to 94.9%), with the detection of rare variant P1-2 g. Macrolide resistance was detected in 97.1% of the samples, primarily due to the A2063G mutation in the 23 S rRNA gene. Notably, 80% of the P1-2 subtype strains did not exhibit any known resistance mutations. Phylogenetic analysis, which included sequences from Beijing, Shenzhen, Suzhou, and Taiwan, showed alignment with other regions, indicating the widespread nature of the current epidemic. This study underscores the high prevalence of macrolide-resistant M. pneumoniae and highlights the critical need for ongoing surveillance and targeted therapeutic strategies.

Materials and methods

Sample collection

In this work, 421 children (age range from 1 to 16) with M. pneumoniae were collected as the research objects. The study was conducted under the Declaration of Helsinki and the study was approved by the Medical Ethics Committee of Shengjing Hospital of China Medical University, Zhongnan Hospital of Wuhan University, People’s Hospital of Bozhou District, and the Affiliated Hospital of Hangzhou Normal University.

Inclusion and exclusion criteria

Inclusion criteria: (1) aged 1–18 y; (2) presence of respiratory symptoms or fever; (3) M. pneumoniae-DNA positive. Exclusion criteria: (1) presence of immunodeficiency, congenital heart disease, or heredity neurological disorders; (2) incomplete clinical data.

Laboratory data

The laboratory data collected in this study included PCR tests for pathogens such as Mycoplasma pneumoniae (MP), Chlamydia pneumoniae (CP), Respiratory Syncytial Virus (RSV), Adenovirus (ADV), and influenza viruses A and B (FluA and FluB), as well as blood tests measuring the percentage and absolute count of white blood cells (WBC), neutrophils (NEU), lymphocytes (LYM), monocytes (MON), eosinophils (EO), and basophils (BA) were also collected on samples from patients at the hospitals. Patient disease status, drug treatment and length of hospital stay information were also collected.

PCR amplification and sequencing analysis

Pharyngeal swabs were collected from patients, and DNA was extracted using an automated nucleic acid extraction instrument. Mycoplasma pneumoniae was detected using the M. pneumoniae real-time PCR kit (Liferiver, Shanghai, China) in clinical laboratories. The remaining DNA was used as a template for amplification of the repMp4, repMp2/3, and 23 S rRNA gene fragments. The PCR primers and conditions were as previously described11,15. The primers used for PCR and Sanger sequencing in this study are shown in Table S1. PrimeSTAR Max DNA Polymerase (Takara, Cat. #R045A), a high-fidelity enzyme, was used for all PCR reactions. Sanger sequencing was analyzed by Sangon Biotech Co. (Shanghai, China). Briefly, PCR was performed to amplify positions 2063 and 2064 in domain V of the 23 S rRNA gene as follows: for the first round, an initial denaturation at 98 °C for 1 min was followed by 5 cycles of 98 °C for 10 s, 60 °C for 15 s, and 72 °C for 15 s. This was followed by 30 cycles of 98 °C for 10 s, 55 °C for 15 s, and 72 °C for 15 s, with a final extension at 72 °C for 5 min. For the second round, an initial denaturation at 98 °C for 1 min was followed by 35 cycles of 98 °C for 10 s, 54 °C for 15 s, and 72 °C for 10 s, with a final extension at 72 °C for 5 min.

For the amplification of the RepMp2/3 gene, the first round included an initial denaturation at 98 °C for 1 min, followed by 5 cycles of denaturation at 98 °C for 10 s, annealing at 60 °C for 15 s, and extension at 72 °C for 15 s. This was followed by 30 cycles of denaturation at 98 °C for 10 s, annealing at 55 °C for 15 s, and extension at 72 °C for 15 s, with a final extension at 72 °C for 5 min. For the second round, the reaction included an initial denaturation at 98 °C for 1 min, followed by 35 cycles of denaturation at 98 °C for 10 s, annealing at 55 °C for 15 s, and extension at 72 °C for 15 s, with a final extension at 72 °C for 5 min. The reaction was held at 4 °C after completion.

For RepMp4-c gene amplification, the first round included an initial denaturation at 98 °C for 1 min, followed by 5 cycles of denaturation at 98 °C for 10 s, annealing at 62 °C for 15 s, and extension at 72 °C for 15 s. This was followed by 30 cycles of denaturation at 98 °C for 10 s, annealing at 58 °C for 15 s, and extension at 72 °C for 15 s, with a final extension at 72 °C for 5 min. For the second round, the reaction included an initial denaturation at 98 °C for 1 min, followed by 35 cycles of denaturation at 98 °C for 10 s, annealing at 69 °C for 15 s, and extension at 72 °C for 15 s, with a final extension at 72 °C for 5 min. The reaction was held at 4 °C after completion.

P1 typing and detection of MR mutations

The repMp4, repMp2/3 regions were amplified, and sequences were aligned with Reference sequences of P1 subtypes and variants were downloaded from NCBI: P1 subtype 1 (P1-1, U00089.2:180858.185741, strain M129), P1 subtype 2 (P1-2, CP002077.1:179293–184197, strain FH), variant 1 (AF290000.1), variant 2a (AP012303.1:179359–184257), variant 2b (AP017318.1:179335–184254), variant 2bv (MK330954.1), variant 2c (AP017319.1:179294–184195), variant 2c2 (JN048894.1), variant 2d (EF656612.1), variant 2f (LC311244.1), and variant 2g (LC385984.1). The 23S rRNA gene, GenBank no. X68422 was used for the detection of drug-resistant mutations.

NGS metadata and phylogenetic tree

The NGS data for M. pneumoniae were extracted from two studies5,16 and the analysis was described as before16. The Sanger sequence results and extracted NGS results were used for phylogenetic tree construction.

Statistical analysis

Microsoft Excel was used to collect the data. Data was expressed as the median. A comparison between groups was performed using the chi-square test or the Mann–Whitney U-test. Differences were considered statistically significant at a p-value of < 0.05.

Accession numbers

The genotyping sequencing results were deposited in GenBank under accession numbers PV209202 to PV209665, and the macrolide resistance mutation sequences were assigned accession numbers PV110195 to PV110615.

Results

A resurgence of pediatric M. pneumoniae infections in China

We conducted a multicenter, retrospective study to analyze the genotypes and macrolide-resistance-associated mutations of M. pneumoniae across four provinces in China, representing different regions: Shenyang (northeast), Zhejiang (southeast), Hubei (central), and Guizhou (southwest) (Fig. 1A Left). Our analysis revealed a significant increase in infection rates, with positivity ranging from approximately 19.4% to 35.9% in 2023 (Fig. 1A right), indicating a notable surge in M. pneumoniae cases compared to the last four years. This increase is consistent with the resurgence of M. pneumoniae infections both within several locations in China, such as Henan17, Suzhou5, and Beijing18, as well as countries like France19 and Spain9, indicated in gray, shown in Fig. 1A. This study collected samples between November 1, 2023, and January 31, 2024, to further analyze M. pneumoniae genotypes and drug resistance. A total of 421 patients were included in the disease spectrum analysis, with a median age of 7 years (range 1–16 years). Of the participants, 207 (49.2%) were female and 214 (50.8%) were male. M. pneumoniae-positive patients were more frequent among school-aged children from 6 to 10 years of age than at other ages. Notably, we observed a low prevalence of co-infections in our dataset (Fig. 1B).

Fig. 1
figure 1

Sample regions and basic characteristics of Mycoplasma pneumoniae. (A) A map of China highlighting the four regions included in this study: Liaoning (red), Guizhou (blue), Zhejiang (green), and Wuhan (purple). Regions shown in brown (Beijing, Shenzhen, Suzhou, Taiwan) are included for reference. The base map was generated using DataV GeoAtlas V3.0, a tool provided by Alibaba Cloud. The source is available at: https://datav.aliyun.com/tools/atlas/. Positive rates of Mycoplasma pneumoniae from 2019 to 2023 across the different regions are displayed. The gray line represents the Mycoplasma pneumoniae positive rate from recently published literature as a control. (B) Basic characteristics of individuals infected with Mycoplasma pneumoniae, including gender, age, and co-infection status.

Full size image

P1-1 subtype predominates in the 2023 resurgence of Mycoplasma pneumoniae

Previous studies have demonstrated that dynamic changes in the proportion of P1 subtypes are likely associated with the periodic outbreaks and epidemics of M. pneumoniae20. To investigate the distribution of P1 genotypes in clinical isolates, we employed a well-established method for amplifying the RepMp regions of the P1 gene, followed by Sanger sequencing. A total of 421 M. pneumoniae-positive samples from four different regions were analyzed and aligned with 13 known variants. Our results showed that the P1-1 genotype was predominant, accounting for 87.3 to 94.9% of the isolates. The second most common subtype was P1-2c, which was observed in Liaoning (12.7%), Zhejiang (6.9%), and Guizhou (11.1%), but was absent in Hubei (0%). The P1-2 subtype was detected in Zhejiang (3.3%) and Hubei (2.6%). Notably, a single case of P1-2 g was identified in Wuhan (Fig. 2A, B). These data suggest that the P1-1 strain remains dominant in the 2023 resurgence, while diverse P1-2 lineage is detectable across different regions.

Fig. 2
figure 2

Genotype distribution of Mycoplasma pneumoniae. (A) The pie chart illustrates the proportion of genotypes across different categories: P1-1 (green), P1-2 (blue), P1-2c (orange), and P1-2 g (red). Each slice represents the percentage of each genotype among a total of 421 samples. The right panel displays the genotype distribution across four regions. (B) The absolute number of individuals with each genotype from four regions in China: Zhejiang (southeast), Liaoning (northeast), Guizhou (southwest), and Hubei (central). Frequencies of each genotype and associated resistance mutations are shown.

Full size image

P1-1 and P1-2 subtypes exhibit different patterns of macrolide resistance mutations

Macrolides are the first line of antibiotics used to treat M. pneumoniae infections. Macrolides are the first-line antibiotics used to treat M. pneumoniae infections. Recent genotyping studies have revealed a correlation between P1 gene types and macrolide resistance. Most type 1 strains isolated in Asian countries in recent years have been macrolide-resistant, whereas type 2 lineage strains are predominantly macrolide-susceptible M. pneumoniae21. To further investigate the epidemiological and evolutionary implications of these two lineages, we analyzed macrolide resistance among different genotypes of M. pneumoniae. Fragments of the 23 S rRNA gene, containing key mutations (A2063G/C, A2064G, A2617G, C2353T), were amplified and sequenced. Our analysis revealed a 97.1% mutation rate in macrolide resistance-associated genes in the 23 S rRNA, with the A2063G mutation being the most prevalent (n = 409), consistent with recent studies5. Notably, the P1-2c subtype exhibited 100% prevalence (n = 32) of the A2063G mutation. Recent studies suggest that the P1-2c macrolide-resistant variant may play a key role in the ongoing outbreaks5. Among the P1 type II lineage, P1-2 accounted for 2.4% of our samples, with only 20% of P1-2 strains harboring the A2063G macrolide resistance mutation (p < 0.0001) (Fig. 3A, B).

Fig. 3
figure 3

Macrolide Resistance of Mycoplasma pneumoniae in four regions. (A) The pie chart shows the proportion of macrolide resistance across two categories: A2063G (blue) and WT (orange). Each slice represents the percentage of variants among a total of 421 samples. The difference in distribution between wild-type and variants was assessed using a Chi-square test (****: p < 0.0001). (B) The absolute number of WT and variant individuals from four regions in China—Zhejiang (southeast), Liaoning (northeast), Guizhou (southwest), and Hubei (central).

Full size image

No significant differences in bacterial load or blood parameters were observed between genotypes or resistance profiles

To compare our data with recent studies and include additional regions for analysis, we collected 147 well-characterized M. pneumoniae-positive sequences with known genotypes and macrolide resistance information from Beijing, Shenzhen, Suzhou, and Taiwan, spanning from 2016 to 20235. These, along with our own 421 samples, were used for phylogenetic analysis. Macrolide-resistant strains were found in most of the clusters or subtypes (Fig. 4A), indicating that macrolide-resistant M. pneumoniae strains have emerged independently due to macrolide use across various genetic backgrounds, rather than through clonal spread of a single MR strain. Our data aligned well with sequences from other regions, showing that the P1-1 strain with the A2063G variant has been predominant since early 2016. Notably, the P1-1 subtypes from Beijing, Shenzhen, and Taiwan exhibited the greatest heterogeneity in terms of origin (Fig. 4A). In addition, we analyzed the M. pneumoniae bacterial load and blood parameters, including the percentage of neutrophils, lymphocytes, and monocytes. We found no differences were observed between mutant and WT M. pneumoniae as well as P1-1 and P1-2 (Fig. 4B and C).

Fig. 4
figure 4

Phylogenetic tree analysis of Mycoplasma pneumoniae in infected individuals. (A) Phylogenetic tree illustrating the evolutionary relationships of predominant epidemic M. pneumoniae clones from various regions in China. Genotypes, locations, and mutation variants are represented by different colors. (B) The MP Ct values, proportions of neutrophils, lymphocytes, and monocytes were compared between patients infected with mutant (n = 410) and WT (n = 11) M. pneumoniae or P1-1(n = 392) with P1-2(n = 29). Statistical comparisons between groups were conducted using the Mann–Whitney U-test. Differences were considered statistically significant at p < 0.05.*: p < 0.05.

Full size image

Clinical characteristics and treatment of patients infected with Mycoplasma pneumoniae

We further analyzed the clinical characteristics of patients. The treatment and management were according to guidelines for the diagnosis and treatment of M. pneumoniae pneumonia in Children (2023). We collected 258 individuals who had completed medical records. Our analysis indicated that patients infected with P1-1 or P1-2 subtypes have no significant impact on disease severity; of note, the P1-2 subtype showed a slightly higher rate of hospitalization (Fig. 5A), which is consistent with previous findings. Among in-patient individuals, the median hospitalization length was 7.0 days (IQR: 5.0–8.0), where no difference between P1-1 or P1-2 subtypes (Fig. 5B). We next analyed the treatment of individuals, antibiotics were administered based on the diagnosis of M. pneumoniae pneumonia and other possible bacterial co-infections. The antibiotic and hormone types were similar between the two groups, and the percentage of patients taking those antibiotics was no significantly lower in individuals infected with Wild-type or drug-resistant M. Pneumoniae (Fig. 5C). Compared to that in-patient group, the proportion of patients administered with methylprednisolone and dexamethasone was found to be much higher in the out-patient group, although the usage of Zithromax remained similar.

Fig. 5
figure 5

Clinical characteristics and treatment of patients infected with Mycoplasma pneumoniae subtypes P1-1 and P1-2. (A) Disease severity (outpatient, inpatient, or ICU admission) stratified by M. pneumoniae subtype (P1-1 vs. P1-2). (B) Hospital length of stay for hospitalized patients (inpatient) infected with P1-1 or P1-2 subtypes. (C) Antibiotic treatments administered, comparing drug resistance patterns and disease severity among all patients. Statistical significance was determined by chi-square test; a p-value < 0.05 was considered significant.

Full size image

Discussion

Previous studies have shown that lockdown measures and public health interventions during the COVID-19 pandemic significantly impacted the incidence of respiratory infections22,23. While some research has examined the effect of PHIs on Mycoplasma pneumoniae infections in China post-pandemic, additional data are needed to provide a more comprehensive understanding of the epidemic, particularly data from a broader range of cities6,7,14,24. Furthermore, previous studies have not delved into the detailed subtypes of M. pneumoniae using the P1 gene, especially P1-2 variants.

In this study, we analyzed the P1 gene subtypes and macrolide-resistant mutations in 421 M. pneumoniae isolates randomly collected from four regions of China in 2023 and early 2024. Our multicenter study findings indicate that both major types of M. pneumoniae circulated in the population, with a high frequency of macrolide resistance. Notably, we observed that the P1-1 subtype remains dominant, and we did not identify other subtypes such as P1-2j, which has been reported as the second most prevalent P1-2 subtype in Japan21. Of note, P1-2c only accounts for 7.6% in 2023 from our analysis. A previous study in Japan showed that type 2 lineage 2c strains were dominant among the M. pneumoniae isolates collected during the time period of 2019 and 2020 from patients with respiratory infections in Japan. Type 2c strains are successful variants that have spread worldwide and are frequently detected in many genotyping studies25,26,27. in contrast to type 2c, we reported only one isolated type 2 g, this variant was early reported in Japan28, but remain low prevalence in many countries11,29.

Macrolides are considered the preferred antibiotics for treating M. pneumoniae infections. Mutations at positions 2063, 2064, 2067, and 2617 of the 23 S rRNA gene are commonly associated with macrolide resistance. Mutations at positions 2063 and 2064 are linked to high-level resistance, while mutations at positions 2067 and 2617 result in low-level resistance. In our analysis, we found that 97.1% of samples harbored the A2063G mutation, which confers macrolide resistance. The incidence of macrolide-resistant M. pneumoniae (MRMP) across different regions of China has ranged from 84.72 to 97.1%, with the A2063G mutation accounting for more than 99% of cases30,31. These findings suggest that China has the highest global incidence of MRMP, with a significant increase in recent years, complicating treatment. In contrast, 80% of P1-2 strains lacked resistance mutations, which is consistent with previous studies showing that type 2 strains are more susceptible to macrolides32.

The periodic occurrence of epidemics is driven by multiple factors, including declining herd immunity and the introduction of new subtypes into the population. In our analysis, genotyping and drug-resistance testing indicated that no new subtypes or variants have emerged as dominant during the recent resurgence. However, due to a lack of data, we were unable to assess changes in herd immunity within our study population. The delayed re-emergence of Mycoplasma pneumoniae is atypical and likely unique to this pathogen. Therefore, the distinguishing characteristics of M. pneumoniae should be carefully considered. These include its slow generation time (approximately 6 h), extended incubation period (1–3 weeks), and relatively low transmission rate, all of which may contribute to the extended time required for M. pneumoniae infection to re-establish itself within a population33,34.

There are several limitations in our study. First, the sample size for genotypic analysis and macrolide resistance mutations was relatively small, and future studies should aim to expand the sample size, particularly by incorporating data from diverse regions across China. Secondly, as the study focused solely on pediatric patients, it did not address the dynamics of community transmission of infectious diseases, including among adults. Nonetheless, community health management remains crucial for pandemic control from a public health perspective. Thirdly, as with all observational studies, we cannot infer causality, and further studies are needed to confirm our findings. Finally, our study did not assess disease severity or conduct a detailed genotypic analysis.

In conclusion, our study provides additional insights into the genetic diversity and macrolide resistance of M. pneumoniae infections in 2023, highlighting the urgent need for ongoing surveillance and updated treatment guidelines. The continued prevalence of macrolide-resistant strains, particularly the P1-1 subtypes with the A2063G mutation, underscores the importance of developing alternative therapeutic strategies to combat M. pneumoniae infections. Furthermore, continuous monitoring of drug resistance and genotypes is recommended to guide treatment strategies and inform public health policies in children.