Unravelling distinct patterns of metagenomic surveillance and respiratory microbiota between two P1 genotypes of Mycoplasma pneumoniae

Abstract

To unravel distinct patterns of metagenomic surveillance and respiratory microbiota between Mycoplasma pneumoniae (M. pneumoniae) P1-1 and P1-2 and to explore the impact of the COVID-19 pandemic on epidemiological features, we conducted a multicentre retrospective study which spanned 90,886 pneumonia patients, among which 3164 cases M. pneumoniae were identified. Our findings revealed a concurrent outbreak of M. pneumoniae, with the positivity rate rising sharply to 9.62% from July 2023, compared to the 0.16% to 4.06% positivity rate observed during the 2020-2022 COVID-19 pandemic. P1-1 had a higher odds ratio of co-detecting opportunistic pathogens. However, no significant differences were observed in the co-detection odds ratio between children and other age groups in P1-2. This study is the first to demonstrate differences in relative abundance, diversity of respiratory microbiota and co-detection rate of opportunistic pathogen between M. pneumoniae P1-1 and P1-2. Through bronchoalveolar lavage (BAL) metagenomic and host transcriptomic analyses, we identified variations in co-detection rates of M. pneumoniae P1-1 genotype with opportunistic pathogens like S. pneumoniae, alterations in respiratory microbiota composition, lung inflammation, and disruption of ciliary function. Consistent with the results of host transcriptome, we found that P1-1 infections were associated with significantly higher rates of requiring respiratory support and mechanical ventilation compared to P1-2 infections (Fisher's exact test, p-value = 0.035/0.004). Our study provides preliminary evidence of clinical severity between M. pneumoniae strains, underscoring the need for ongoing research and development of targeted therapeutic strategies.

Keywords: Mycoplasma pneumoniae; P1 genotype; clinical outcome; host immune response; respiratory microbiota.

Figure 1.

The epidemiology of M. pneumoniae and positive rate of 23s rRNA mutation in China. (A) The number and locations of bronchoscopy mNGS tests applied to patients with pneumonia from 2019 to 2023. Red represents M. pneumoniae positive and green represents M. pneumoniae negative in mNGS data. (B) The average number of mNGS tests per season for pneumonia patients and positive number and rate of M. pneumoniae infection during 2019-2023. (C) Phylogenetic tree revealed the presence of two primary epidemic clones of EC1 and EC2 in China. (D) The composition of M. pneumoniae changed dramatically with P1-2 Mut replacing non-mut strains after COVID-19 pandemic. Purple/Mut: mutations in the 23S rRNA gene (C2617T, A2063G, A2064G). Orange/Non-mut: there were no mutations in the 23S rRNA gene (C2617T, A2063G, A2064G).

Figure 2.

Genotype prevalence of M. pneumoniae and relative abundances of P1-1 and P1-2 among different periods and different ages. (A) Positive rate of M. pneumoniae P1 genotypes (P1-1 and P1-2) before, during, and after COVID-19 pandemic. (A) Positive rate of M. pneumoniae P1 genotypes (P1-1 and P1-2) of children, school-age, adults and older-people among different periods. (C) Reads per millions between two genotypes (P1-1 and P1-2) of M. pneumoniae across different age groups (children, school-age, adults and older-people). *p < 0.05, **p < 0.01, and ***p < 0.001. “NS” represents “no significant difference.”

Figure 3.

Changes in the composition of common respiratory pathogens in patients testing positive for M. pneumoniae before, during, and after COVID-19 pandemic. I, Detection rates of bacterial (a) and viral (b) pathogens. HAdv: human adenovirus, RSV: respiratory syncytial virus, HMPV: human metapneumovirus, HCoV: human coronavirus, IFV: influenza virus, HBoV: human bocavirus. II, Co-infection patterns in school-age group before (a), during (b), and after (c) COVID-19 pandemic. Bigger size and darker colour of the circles indicate higher co-infection rates between two pathogens.

Figure 4.

Odds ratio of co-detection rate between P1-1 and P1-2 among different age groups. (A) P1-1 of M. pneumoniae co-detection rate with important respiratory pathogen and odds ratio of comparing children to school-age, adults and older-people. (B) P1-2 of M. pneumoniae co-detection rate with important respiratory pathogen and odds ratio of comparing children to school-age, adults and older-people. (C) Odds ratio of co-detection rate between P1-1 and P1-2 among children, school-age, adults and older-people. HAdv: human adenovirus; HSV: human herpes simplex virus; HBoV human bocavirus.

Figure 5.

Odds ratio of clinical outcomes between P1-1 and P1-2.

Figure 6.

Host immune profiling between P1-1 and P1-2 BAL. (A) Heatmap of 62 significantly differential expressed genes between P1-1 and P1-2 involved in Interferon stimulated pathway, immune-related pathway, phagosome, proteasome and calcium signalling pathway. (B–G) GSEA of Macrophage M1, Monocytes, NK cell activated, phagosome, proteasome and calcium signalling pathway. Each line represents one particular gene set with unique colour, and up-regulated genes are located in the left approaching the origin of the coordinates, by contrast the down-regulated lay on the right of x-axis. Only gene sets with NOM p-value < 0.05 and FDR < 0.05 were considered significant. And only several leading gene sets were displayed in the plot. (H–L) Feature value of two P1 genotypes. Colour distinctions represent various groups associated with P1 type. The median is visually depicted by black lines. The y-axis in each panel was trimmed at the maximum value among all groups of 1.5*IQR above the third quartile, where IQR is the interquartile range. For each gene, we conducted formal comparisons among groups within the training cohort. Pairwise comparisons were performed with a Mann–Whitney test.