Molecular Evolution and Increasing Macrolide Resistance of Bordetella pertussis, Shanghai, China, 2016-2022

Abstract

Resurgence and spread of macrolide-resistant Bordetella pertussis (MRBP) threaten global public health. We collected 283 B. pertussis isolates during 2016-2022 in Shanghai, China, and conducted 23S rRNA gene A2047G mutation detection, multilocus variable-number tandem-repeat analysis, and virulence genotyping analysis. We performed whole-genome sequencing on representative strains. We detected pertussis primarily in infants (0-1 years of age) before 2020 and older children (>5-10 years of age) after 2020. The major genotypes were ptxP1/prn1/fhaB3/ptxA1/ptxC1/fim2-1/fim3-1 (48.7%) and ptxP3/prn2/fhaB1/ptxA1/ptxC2/fim2-1/fim3-1 (47.7%). MRBP increased remarkably from 2016 (36.4%) to 2022 (97.2%). All MRBPs before 2020 harbored ptxP1, and 51.4% belonged to multilocus variable-number tandem-repeat analysis type (MT) 195, whereas ptxP3-MRBP increased from 0% before 2020 to 66.7% after 2020, and all belonged to MT28. MT28 ptxP3-MRBP emerged only after 2020 and replaced the resident MT195 ptxP1-MRBP, revealing that 2020 was a watershed in the transformation of MRBP.

Keywords: Bordetella pertussis; China; Shanghai; antimicrobial resistance; bacteria; genotype; macrolide; vaccine-preventable diseases.

Figure 1

Distributions of pertussis patients in different age groups, Shanghai, China, 2016–2022. Pertussis was primarily detected from infants (0–1 years of age) before 2020 but mostly from older children and adolescents (>5–10 years of age) after 2020.

Figure 2

Distributions of prevalent Bordetella pertussis subtypes over time (A) and by age group (B), Shanghai, China, 2016–2022. Fourteen MTs were identified in this study. MT27 was the major strain during 2016–2019, whereas MT28 isolates increased quickly during 2021–2022 (panel A). MT distributions in infants (0–1 years of age) and noninfants (>1–12 years of age) change substantially from 2016–2019 to 2021–2022 (panel B). MT, multilocus variable-number tandem-repeat analysis type.

Figure 3

Changing macrolide resistance of circulating Bordetella pertussis strains, Shanghai, China, 2016–2022. A) ptxP3-strains showed very high resistance to macrolides after 2020. Resistance to macrolides was different in non-MT28 (0%) and MT28 (100%) isolates. B) Percentages of macrolide-sensitive BP, ptxP1-MRBP, and ptxP3-MRBP before and after 2020 show that ptxP1-MRBP strain was prevalent before 2020 but predominately ptxP3-MRBP spread after 2020. MRBP, macrolide-resistant Bordetella pertussis; MT, multilocus variable-number tandem-repeat analysis type.

Figure 4

Minimum-spanning tree of 283 Bordetella pertussis MTs, Shanghai, China, 2016–2022. A) Virulence genotype profiles; B) A2047G mutations; C) pre-2020 versus post-2020. Circle sizes indicate the number of each MT. Differences in the length and thickness of the lines linking 2 circles indicate differences in the number of variable-number tandem repeats between the 2 linked MTs. MT, multilocus variable-number tandem repeat analysis type.

Figure 5

Maximum-likelihood phylogenetic tree of 4 Shanghai, China, and 1,491 global Bordetella pertussis strains, by geographic area, 2016–2022. Red lines indicate Shanghai strains; bold indicates 2 MT28 strains. Black dot indicates MT28-MRBP (B19005) strain from Anhui Province, China. Arrow indicates reference strain Tohama I. Shanghai strains associated phylogenetic subclades are enlarged for visualization. MSBP, macrolide-sensitive Bordetella pertussis; MT, multilocus variable-number tandem-repeat analysis type; NA, not applicable.