Abstract
Background
Pertussis (whooping cough), a highly contagious respiratory disease caused by Bordetella pertussis, has resurged worldwide and requires increased attention. This study aimed to characterize the molecular epidemiology and antimicrobial susceptibility profiles of B. pertussis isolates circulating in Guangzhou, China.
Methods
A total of 91 culture-confirmed pertussis cases in Guangzhou between January 2020 and August 2024 were enrolled and studied. B. pertussis isolates (from January 2020 to May 2024) were recovered from 62 cases. All isolates underwent antigenic genotyping and phylogenetic analysis based on whole-genome sequencing. Antimicrobial susceptibility testing using E-test was performed on 12 representative isolates.
Results
The majority of culture-confirmed cases occurred in children under 1 year of age who were unvaccinated or partially vaccinated. Genotypic analysis revealed a significant shift: only 3 isolates (all from 2022) harbored the ptxP1 allele, while the remaining 59 exhibited ptxP3. Three types of pertactin (prn) allele were identified: prn1 (4.84%, 3/62), prn2 (11.29%, 7/62), prn150 (80.65%, 50/62), and 2 untyped prn allele. By 2024, prn150 became the predominant allele. Phylogenetic analysis revealed distinct branches separating ptxP1 and ptxP3 lineages. E-test results demonstrated that all macrolide-resistant isolates (exhibiting MICs > 256 mg/L for erythromycin, azithromycin, and clarithromycin) carried the A2047G mutation in 23S rRNA gene. The proportion of isolates harboring this mutation increased significantly after 2022 and dominated by 2024. All tested isolates displayed low MICs to alternative agents: trimethoprim/sulfamethoxazole (MICs ≤ 0.5/9.5 mg/L), cefoperazone/sulbactam (MICs ≤ 0.064/0.032 mg/L), and piperacillin/tazobactam (MICs ≤ 0.064/4 mg/L). Notably, macrolide-resistant isolates harboring ptxA1-ptxP3-prn150 genotype formed a distinct sub-clone within the ptxP3 clade.
Conclusions
Our findings demonstrate the current dominance of macrolide-resistant B. pertussis strain harboring ptxP3 and prn150 in Guangzhou since 2024. This study provides epidemiological and microbiological insights to guide local pertussis control strategies and to offer antimicrobial resistance monitoring efforts.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12879-025-11577-z.
Keywords: Bordetella pertussis, Genetic analysis, Macrolide-resistant
Introduction
Bordetella pertussis, a Gram-negative coccobacillus, is the causative agent of whooping cough (pertussis) in humans, with infants constituting the most vulnerable population [1–3]. Pertussis manifests as a highly contagious acute respiratory infection characterized by three distinct clinical phases: the catarrhal phase (initiating with cold-like symptoms), the paroxysmal phase (marked by severe coughing paroxysms often accompanied by an inspiratory whoop or post-tussive vomiting), and finally a gradual recovery phase [1]. Despite widespread vaccination programs, pertussis remains a significant global health burden, with an estimated 24.1 million cases and 160,700 deaths occurring among children under 5 years of age in recent years. Notably, a resurgence in pertussis incidence has been documented globally since 2000, with a pronounced increase among adolescents and adults [4].
The pathogenicity of B. pertussis is mediated by an array of virulence factors, including toxins, surface adhesins, autotransporter proteins, lipo-oligosaccharides, Bordetella polysaccharide, and siderophores [1, 5]. Among these, pertussis toxin (PTx), an AB5 toxin that elevates intracellular cyclic adenosine monophosphate (cAMP) levels in host cells, has been widely investigated for its critical role in disrupting cellular signaling pathways [6]. A notable evolutionary development is the emergence of the ptxP3 allele in the 1980 s, which enhances PTx production in vitro [7]. The rapid global dissemination of ptxP3 lineage suggests adaptive evolution within the bacterial population in response to vaccine selection pressure [8]. These phenotypic and genotypic shifts are likely contributors to the observed resurgence of pertussis. For instance, Li et al. reported that the predominant genotype in mainland China (2014–2016) was ptxA1-ptxP1-prn1-fim2-1-fim3A [9], while by 2021, strains carrying ptxP3 and prn150 had become increasingly dominant in Shanghai [10]. Similarly, Michael et al. documented ptxA1-ptxP3-prn2 as the prevalent genotype in the United States between 2000 and 2013 [11]. Collectively, the evidence above highlights the importance of sustained molecular epidemiological surveillance to trace these genotypic and phenotypic variations.
Furthermore, the emergence of macrolide-resistant B. pertussis strains presents a significant public health concern in China, where erythromycin and azithromycin serve as first-line antimicrobials for B. pertussis. In Shanghai, the prevalence of macrolide-resistant isolates surged dramatically from 60 to 98% between 2016 and 2024 [12]. Alarmingly, recent reports confirm the presence of such resistant strains in Japan and Vietnam [13, 14], suggesting potential clonal expansion across Asia. The primary mechanism underpinning macrolide resistance is an A-to-G mutation at position 2047 (A2047G) in the domain V region of the 23S rRNA gene. However, standardized interpretive criteria for antimicrobial susceptibility testing specific to B. pertussis remain undefined. Therefore, detection of the A2047G mutation may serve as a rapid method for predicting macrolide resistance in clinical isolates.
To elucidate the evolving molecular epidemiology and antimicrobial resistance patterns of pertussis, this study characterized pediatric pertussis cases and investigated the dominant genotypes and macrolide-resistant phenotypes among B. pertussis isolates circulating in Guangzhou, China. Our findings may provide evidence to guide public health intervention and optimize therapeutic strategies for B. pertussis infection.
Methods and materials
Sampling and population
A total of 91 culture-confirmed pertussis cases from Guangdong Women and Children Hospital between January 2020 and August 2024 were enrolled and studied, among which 62 isolates (from January 2020 to May 2024) were recovered and recognized. For phylogenetic analysis, genomes of global isolates were downloaded from National Center of Biotechnology Information (NCBI) database [15]. Information of selected strains was Listed in Supplementary Table 1. All data of medical records including clinical information and laboratory testing results was anonymous. This study was approved by the ethics committee of this hospital.
Bacterial isolation and identification
Nasopharyngeal specimens were collected from patients with suspected whooping cough. Nasopharyngeal swabs were cultured on fresh charcoal agar (Oxoid, Basingstoke, UK) supplemented with 10% defibrinated sheep blood and cephalexin (40 µg/ml) to inhibit the growth of normal flora. Plates were incubated for at least five days, aerobically at 35–36℃. Suspected B. pertussis isolated were identified using matrix-assisted laser desorption ionization-time of flight mass spectrometry (Bruker, Germany).
Antimicrobial susceptibility testing (AST)
The minimum inhibitory concentrations (MICs) of erythromycin (ERY), azithromycin (AZM), clindamycin (CLI), clarithromycin (CLR), cefoperazone/sulbactam (CSL), piperacillin/tazobactam (TZP) and trimethoprim/sulfamethoxazole (SXT) were detected using E-test strips (Bio-Kont, China). According to the manufacturer’s instruction of E-test strips (Bio-Kont, China), Staphylococcus aureus ATCC29213 and Escherichia coli ATCC25922 were used as quality control strains in the AST. The concentration range of antimicrobials and acceptable MICs of quality control for reliable results are available in Supplementary Table 2. The Clinical and Laboratory Standard Institute (CLSI) and the European Committee on Antimicrobial Susceptibility Testing (EUCAST) have not yet provided the breakpoint for AST of B. pertussis. Herein the B. pertussis isolates with MICs < 0.25 mg/L (ERY, AZM and CLR) were considered to be macrolide-susceptible, and MICs > 256 mg/L (ERY, AZM and CLR) indicates macrolide-resistant, referred to Laboratory Diagnosis and Molecular Surveillance of Bordetella pertussis (ECDC) [16].
Whole-genome sequencing
Chromosomal DNA of 62 B. pertussis isolates was extracted from fresh bacteria using DNA extraction kit (Qiagen) according to manufacturer’s instruction. Purified DNA was examined for integrality and adequate concentration, followed by generated multiplexed libraries using VAHTS®Universal Plus DNA Library Prep Kit for IIIumina. Paired-end 150-bp indexed reads were generated on the Illumina HiSeq platform, and the average sequencing depth of the genomes was about 300-fold. Proportion of sequences qualified as Q30 was larger than 85%. Subsequently, the genome sequences were assembled by Megahit version 1.2.9. Genomic sequencing data have been deposited in the National Center of Biotechnology Information (NCBI) database with the accession code PRJNA1193776.
Analysis based on whole-genome sequencing
BIGSdb-Pasteur genomic platform (https://bigsdb.pasteur.fr/bordetella/) for Bordetella was utilized for identification of B. pertussis sequence types. Pertussis antigen genes including ptxP, ptxABCDE, fim2, fim3, and fhaB2400_5550 were characterized by BPagST scheme, while prn and 23S rRNA mutation were identified within Auto-transporters and 23S rRNA loci scheme, respectively. Phylogenetic analysis was conducted using MGAP version 2.0.0 and phylogenetic relation with maximum likelihood algorithm based on the multiple SNP (Single nucleotide polymorphism) sites sequence alignment was provided. The strain Tohama Ⅰ (GenBank accession number: NC_002929) was the reference strain for identification. Phylogenetical trees were annotated using iTOL.
Results
Epidemiological characteristics of pertussis in Guangzhou, China
From January 2020 to August 2024, 91 pediatric patients were culture-confirmed with B. pertussis infection. The cohort comprised 48.4% (44/91) males and 51.6% (47/91) females, aged 1 month to 11 years. Infants under 3 months represented 50.0% (1/2) of cases in 2020, 42.9% (3/7) in 2022, 44.4% (4/9) in 2023, and 35.6% (26/73) in 2024, respectively (Table 1).
Table 1.
Hospitalization, gender and age distribution among pertussis cases
| Year | 2020 | 2022 | 2023 | 2024 |
|---|---|---|---|---|
| Number of cases | 2 | 8 | 9 | 73 |
| Culture-confirmed | 2 | 8 | 9 | 73 |
| Hospitalization | 2 | 5 | 8 | 39 |
| Gender, n (%) | ||||
| Male | 2 (100.00) | 7 (87.50) | 3 (33.33) | 32 (43.84) |
| Female | 0 (0.00) | 0 (0.00) | 6 (66.67) | 41 (56.16) |
| Age, n (%) | ||||
| ≤ 3 months | 1 (50.00) | 3 (42.86) | 4 (44.44) | 26 (35.62) |
| 3–6 months | 1 (50.00) | 2 (28.57) | 3 (33.33) | 26 (35.62) |
| 7–11 months | 0 (0.00) | 0 (0.00) | 0 (0.00) | 2 (2.74) |
| 1–2 years | 0 (0.00) | 1 (14.29) | 1 (11.11) | 5 (6.85) |
| 3–5 years | 0 (0.00) | 1 (14.29) | 1 (11.11) | 8 (10.96) |
| ≥ 6 years | 0 (0.00) | 0 (0.00) | 0 (0.00) | 6 (8.22) |
For detailed clinical characterization, 51 cases with available demographic, clinical, and laboratory records were analyzed (Table 2). Manifestations included paroxysmal cough (82.4%, 42/51), spasmodic cough (29.4%, 15/51), and inspiratory whoop (35.3%, 18/51). Leukocytosis (WBC 10–60 × 10⁹ cells/L) was observed in 88.2% (45/51) of all. Vaccination status revealed 58.8% (30/51) unvaccinated, 23.5% (12/51) partially vaccinated (≥ 1 DTaP dose), and 3.9% (2/51) Fully vaccinated; status was undocumented for 7 cases (13.7%). Patients originated from 12 distinct geographic regions within Guangdong Province.
Table 2.
Demographic, clinical and laboratory characteristics of pediatric pertussis
| Year | Total | 2022 | 2023 | 2024 |
|---|---|---|---|---|
| Cases of patients with available information | 51 | 6 | 9 | 36 |
| Patient characteristics, n (%) | ||||
| Fever | 14 (27.45) | 1 (16.67) | 2 (22.22) | 11 (30.56) |
| Rhinorrhea | 15 (29.41) | 3 (50.00) | 2 (22.22) | 10 (27.78) |
| Nasal congestion | 17 (33.33) | 3 (50.00) | 2 (22.22) | 12 (33.33) |
| Cough, n (%) | ||||
| Paroxysmal | 42 (82.35) | 6 (100.00) | 6 (100.00) | 27 (75.00) |
| Spasmodic | 15 (29.41) | 0 (0.00) | 2 (22.22) | 13 (36.11) |
| Whooping | 18 (35.29) | 2 (33.33) | 3 (33.33) | 13 (36.11) |
| Sputum | 16 (31.37) | 2 (33.33) | 2 (22.22) | 12 (33.33) |
| Vomiting | 13 (25.49) | 1 (16.67) | 4 (44.44) | 8 (22.22) |
| Cyanosis | 21 (41.18) | 3 (50) | 3 (33.33) | 15 (41.67) |
| Wheeze | 13 (25.49) | 2 (33.33) | 3 (33.33) | 8 (22.22) |
| Dyspnea | 18 (35.29) | 2 (33.33) | 4 (44.44) | 12 (33.33) |
| Convulsion | 2 (3.92) | 0 (0.00) | 0 (0.00) | 2 (5.56) |
| Days to onset, n (%) | ||||
| < 1 week | 8 (15.69) | 1 (16.67) | 2 (22.22) | 5 (13.89) |
| 1–2 weeks | 20 (39.22) | 3 (50.00) | 2 (22.22) | 15 (41.67) |
| ≥ 2 weeks | 23 (45.1) | 2 (33.33) | 5 (55.56) | 16 (44.44) |
| Leukocytosis, n (%) | ||||
| (10–60) × 109 cells/L | 45 (88.24) | 4 (66.67) | 6 (66.67) | 35 (97.22) |
| ≥ 60 × 109 cells/L | 2 (3.92) | 0 (0.00) | 2 (22.22) | 0 (0.00) |
| Vaccination, n (%) | ||||
| Full | 2 (3.92) | 1 (16.67) | 0 (0.00) | 1 (2.78) |
| Partial | 12 (23.53) | 2 (33.33) | 2 (22.22) | 8 (22.22) |
| None | 30 (58.82) | 2 (33.33) | 5 (55.56) | 23 (63.89) |
| Unknown | 7 (13.73) | 1 (16.67) | 2 (22.22) | 4 (11.11) |
| Originated address, n (%) | ||||
| Guangdong Province | 49 (96.08) | 6 (100.00) | 8 (88.89) | 35 (97.22) |
| Guangzhou | 25 (49.02) | 4 (66.67) | 2 (22.22) | 19 (52.78) |
| Qingyuan | 4 (7.84) | 1 (16.67) | 0 (0.00) | 3 (8.33) |
| Foshan | 2 (3.92) | 1 (16.67) | 0 (0.00) | 1 (2.78) |
| Huizhou | 1 (1.96) | 0 (0.00) | 1 (11.11) | 0 (0.00) |
| Yunfu | 3 (5.88) | 0 (0.00) | 1 (11.11) | 2 (5.56) |
| Jieyang | 2 (3.92) | 0 (0.00) | 2 (22.22) | 0 (0.00) |
| Heyuan | 2 (3.92) | 0 (0.00) | 1 (11.11) | 1 (2.78) |
| Jiangmen | 1 (1.96) | 0 (0.00) | 1 (11.11) | 0 (0.00) |
| Maoming | 3 (5.88) | 0 (0.00) | 0 (0.00) | 3 (8.33) |
| Zhanjiang | 1 (1.96) | 0 (0.00) | 0 (0.00) | 1 (2.78) |
| Zhaoqing | 2 (3.92) | 0 (0.00) | 0 (0.00) | 2 (5.56) |
| Yangjiang | 3 (5.88) | 0 (0.00) | 0 (0.00) | 3 (8.33) |
| Guangxi Province | 1 (1.96) | 0 (0.00) | 1 (11.11) | 0 (0.00) |
| Unknown | 1 (1.96) | 0 (0.00) | 0 (0.00) | 1 (2.78) |
| Suspected household exposure, n (%) | ||||
| Yes | 28 (54.9) | 4 (66.67) | 4 (44.44) | 20 (55.56) |
| No | 18 (35.29) | 0 (0.00) | 3 (33.33) | 15 (41.67) |
| Unknown | 5 (9.80) | 2 (33.33) | 2 (22.22) | 1 (2.78) |
Among all culture-confirmed cases, 63 isolates obtained before May 2024 were recovered for analysis. Whole-genome sequencing confirmed 62 isolates as B. pertussis. The remaining isolate (BP03) was identified as B. parapertussis and excluded from further analysis.
Antigen gene profiling and phylogenetic analysis of B. pertussis isolates
To characterize molecular shifts in circulating B. pertussis in Guangzhou, we analyzed key antigen genes (ptxA, ptxP, prn) and the 23S rRNA A2047G mutation using whole-genome sequencing data.
Phylogenetic analysis based on core-genome SNPs (Fig. 1) resolved two major clades corresponding to ptxP1 and ptxP3 lineages. The ptxP1 clade exclusively exhibited the genotype ptxP1-ptxA1-ptxB1-ptxC1-ptxD1-ptxE4-fim2-1-fim3-1-fhaB3. The predominant ptxP3 clade harbored ptxP3-ptxA1-ptxB1-ptxC4-ptxD1-ptxE4-fim2-1-fim3-1-fhaB1. Notably, within the ptxP3 lineage, isolates carrying prn150 formed a distinct subclade that emerged in 2022 and became dominant by 2024. Additionally, ptxP3 isolates with wild-type 23S rRNA and prn2 clustered into separate subclades. These phylogenetic relationships were further supported by SNP variation heatmap analysis (Supplementary Fig. 1).
Fig. 1.
Phylogenetic analysis of 62 Bordetella pertussis from 2020 to 2024 in Guangzhou, China
A significant Shift in antigen genotypes and macrolide resistance-related genes was observed from 2020 to 2024 (Fig. 2). The frequency of the 23S rRNA A2047G mutation increased sharply, rising from 75.0% (6/8) in 2022 to 97.7% (42/43) in 2024, and both two 2020 isolates lacked this mutation. Genotypic analysis revealed a significant shift: only 3 isolates (all from 2022) harbored the ptxP1 allele, while the remaining 59 exhibited ptxP3. All isolates harbored ptxA1. The prn gene exhibited three alleles: prn1 (4.84%, 3/62), prn2 (11.29%, 7/62), prn150 (80.65%, 50/62), with 2 isolates non-typable for prn allele. By 2024, prn150 became the predominant allele. Concurrently, a rapid increase in ptxA1-ptxp3-prn150 isolates from 2022 to 2024 coincided with a pronounced decline in ptxA1-ptxP3-prn2 prevalence.
Fig. 2.
Characterization of vaccine antigen type from Bordetella pertussis isolates. A Proportion of A2047G mutation detected in 23S rRNA in 2020 (n = 2), 2022 (n = 8), 2023 (n = 9), 2024 (n = 43). B Number of different genotypes of strains from 2020 to 2024
Macrolide resistance profiles of B. pertussis isolates
Given the established association between the A2047G mutation and macrolide resistance, we determined the antimicrobial susceptibility profiles of representative isolates using E-test methodology. All five isolates lacking the A2047G mutation in 23S rRNA gene exhibited susceptibility to macrolides, with low MIC values (MICs ≤ 0.25 mg/L) for erythromycin, azithromycin, clarithromycin, clindamycin (Table 3). Conversely, isolates harboring the A2047G mutation exhibited high-level resistance (MICs > 256 mg/L) to all tested macrolides (Table 3). Importantly, all isolates remained relative low MIC values to trimethoprim/sulfamethoxazole (MICs ≤ 0.5/9.5 mg/L), piperacillin/tazobactam (MICs ≤ 0.064/4 mg/L), and cefoperazone/sulbactam (MICs ≤ 0.064/0.032 mg/L).
Table 3.
Antimicrobial susceptibility tests of representative B. pertussis isolate in this study
| Year | BP isolate | Sequence type | Antimicrobial susceptibility (mg/L) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 23 S rRNA | ptxP | prn | ERY | AZM | CLR | CLI | SXT | TZP | CSL | |||
| 2020 | BP01 | 1 | 3 | 2 | 0.125 | ≤ 0.064 | 0.25 | 0.125 | 0.125/2.375 | ≤ 0.064/4 | ≤ 0.064/0.032 | |
| 2020 | BP58 | 1 | 3 | 2 | ≤ 0.064 | ≤ 0.064 | 0.25 | 0.25 | 0.125/2.375 | ≤ 0.064/4 | ≤ 0.064/0.032 | |
| 2022 | BP02 | 13 | 1 | 1 | > 256 | > 256 | > 256 | > 256 | 0.125/2.375 | ≤ 0.064/4 | ≤ 0.064/0.032 | |
| 2022 | BP04 | 1 | 3 | 2 | ≤ 0.064 | ≤ 0.064 | 0.125 | 0.125 | 0.032/0.608 | ≤ 0.064/4 | ≤ 0.064/0.032 | |
| 2022 | BP05 | 1 | 3 | 2 | ≤ 0.064 | ≤ 0.064 | 0.125 | 0.25 | 0.016/0.304 | ≤ 0.064/4 | ≤ 0.064/0.032 | |
| 2022 | BP06 | 13 | 1 | 1 | > 256 | > 256 | > 256 | > 256 | 0.032/0.608 | ≤ 0.064/4 | ≤ 0.064/0.032 | |
| 2022 | BP11 | 13 | 1 | 1 | > 256 | > 256 | > 256 | > 256 | 0.064/1.216 | ≤ 0.064/4 | ≤ 0.064/0.032 | |
| 2023 | BP12 | 13 | 3 | 150 | > 256 | > 256 | > 256 | > 256 | 0.5/9.5 | ≤ 0.064/4 | ≤ 0.064/0.032 | |
| 2023 | BP13 | 13 | 3 | 150 | > 256 | > 256 | > 256 | > 256 | 0.5/9.5 | ≤ 0.064/4 | ≤ 0.064/0.032 | |
| 2024 | BP46 | 13 | 3 | 150 | > 256 | > 256 | > 256 | > 256 | 0.016/0.304 | ≤ 0.064/4 | ≤ 0.064/0.032 | |
| 2024 | BP47 | 1 | 3 | 2 | ≤ 0.064 | ≤ 0.064 | 0.25 | 0.125 | 0.125/2.375 | ≤ 0.064/4 | ≤ 0.064/0.032 | |
| 2024 | BP48 | 13 | 3 | 150 | > 256 | > 256 | > 256 | > 256 | 0.125/2.375 | ≤ 0.064/4 | ≤ 0.064/0.032 | |
erythromycin ERY, azithromycin AZM, clindamycin CLI, clarithromycin CLR, cefoperazone/sulbactam CSL, piperacillin/tazobactam TZP and trimethoprim/sulfamethoxazole SXT
23S rRNA typed 1 represents wildtype strain, and 23S rRNA typed 13 represents strain with A2047G mutation
Phylogenetic analysis within the global epidemiology context
To elucidate the phylogenetic features of the ptxA1-ptxP3-prn150 clone, we compared representative isolates from this study, including three macrolide-resistant ptxA1-ptxP1-prn1 (MR-ptxP1), seven macrolide-susceptible ptxA1-ptxP3-prn2 (MS-ptxP3), and six macrolide-resistant ptxA1-ptxP3-prn150 (MR-prn150) isolates, with 58 globally sourced lineages from public database. From phylogenetic analysis, the MR-ptxP1 clones clustered closely with strains circulating in mainland China between 2012 and 2019 (Fig. 3), indicating regional genetic continuity. Most MS-ptxP3 isolates exhibited homologous to Shanghai-derived macrolide-susceptible ptxP3-prn2 isolates. Notably, the MR-prn150 clone demonstrated significant phylogenetic relatedness with the MT28-Shanghai clone (defined by ptxP3, erythromycin resistance, MT28 profile, and prn150 by Cai et al. [10]).
Fig. 3.
Phylogenetic analysis of B. pertussis genomes from Guangzhou within worldwide lineages
Discussion
Our analysis revealed a significant antigenic genotype shift from ptxP1 to ptxP3 among B. pertussis isolates in Guangzhou, China, between 2022 and 2024. The ptxP3 Lineage appears to have been established in Guangzhou prior to 2020, as both isolates recovered that year harbored this genotype, although the 2020 sample size was limited. This study aligns with global trends: ptxP3 strains have predominated in Europe, the United States and Spain since the 1990 s [11, 17, 18], whereas ptxP1 strains remained dominant in mainland China until 2016 [9, 19]. Subsequent surveillance data indicate an increasing prevalence of ptxP3 strains in China between 2018 and 2019 [10, 12, 20]. Collectively, these observations demonstrate that the expansion of ptxP3 lineage in China reflects the ongoing global epidemic of B. pertussis.
Crucially, the majority of ptxP3 isolates in Guangzhou carried the A2047G mutation in the 23S rRNA gene, suggesting potential expansion of erythromycin-resistant strains across southern China. Yang et al. reported that 91.9% (91/99) of isolates collected in the north part of China during 2013–2014 exhibited high-level macrolide resistance [21], while Wang et al. identified erythromycin resistance in 87.5% (14/16) of isolates from Xi’an between 2012 and 2013 [22]. Alarmingly, resistance prevalence surged from 46.5% in 2019 to 97.3% in 2024 in southern China [23]. Although other putative antimicrobial resistance genes (e.g., erm(C), estT genes and even tet(X4)) have been reported in some B. pertussis strains [24], the possibility of contaminant-derived misinterpretations need to be classified [25]. Given the high prevalence of macrolide resistance, alternative antimicrobial agents are essential. Sulfamethoxazole-trimethoprim (SXT) is recommended for pediatric pertussis treatment [26, 27], and low resistance rates to SXT were also observed in this study. Furthermore, all representative isolates exhibited significant susceptibility to cefoperazone/sulbactam (CSL) and piperacillin/tazobactam (TZP), identifying them as potential therapeutic antibiotics.
Herein we showed that isolates harboring the pertactin allele prn150 became predominant among all collected strains. This variant possesses one synonymous mutation (C531T) compared to prn2. The dominance of the macrolide-resistant ptxA1-ptxP3-prn150 strain may challenge pertussis vaccination efficacy. The Chinese pertussis vaccine strain CS (ptxA2-ptxP1-prn1-tcfA2-fim2-1-fim3-1) exhibits significant antigenic divergence from this emerging strain, particularly in ptxA1, ptxP3 and prn150 [28]. Notably, PRN2 (encoded by prn2) contains seven distinct amino acids in repeat region 1 compared to PRN1 (encoded by prn1), a region implicated in eliciting protective antibodies in murine models [29, 30]. Furthermore, the SNP distinguishing ptxP1 and ptxP3 resides within the BvgA binding site (a key virulence gene regulator), with ptxP3 strains demonstrating enhanced toxin production relative to ptxP1 [7]. This macrolide-resistant clone shares key features with the MT28-Shanghai strain (ptxP3, erythromycin resistance, prn150), first reported in Shanghai by Cai et al. [10]. and subsequently documented across five Chinese cities (Erdos, Anhui, Xiamen, Haikou, Shanghai) during the 2024 pertussis outbreak [31]. The high prevalence of macrolide-resistant ptxA1-ptxP3-prn150 strain likely stems from complex causes including insufficient vaccination, immune evasion and macrolide overuse; however, infections with MR-MT28 strains appear associated with milder symptoms and reduced hospitalization rates [12]. Concurrently, increasing global attention focuses on PRN-deficient strains, which pose significant challenges for both vaccination and antimicrobial therapy [31], although no such isolates were confirmed in this study.
Phylogenetic analysis revealed global clustering that MR-prn150 isolates in this study and MT28-Shanghai strain formed a significant clade phylogenetically proximate to ptxA1-ptxP3-prn2 strains from Shanghai and Guangzhou. Given their close phylogenetic relationship and identical profiles (ptxA1, ptxP3, prn150 and A2047G mutation), the MR-prn150 strain in this study likely evolved from the MT28-Shanghai lineage (discovered previous to this study by Cai et al.), although their separation into distinct subclusters suggests additional genomic divergence. As supportive evidence, Cai et al. reported that MT28-Shanghai lineage has been dominant in Haikou (a provincial capital city of southern China) by 2024 [31]. Moreover, macrolide-resistant strains harboring ptxP3 and prn150 remain unreported outside China. Collectively, these findings indicate the MR-prn150 strain most likely evolved from the circulating strains in China, especially from MT28-Shanghai lineage, rather than originating from external sources.
Analysis of available clinical data revealed that most pertussis-confirmed patients were unvaccinated (58.82%) or partially vaccinated (23.53%) infants, while only 3.92% were fully vaccinated. This distribution approximates to data reported by Pan et al. and Cai et al. [10, 12]. However, the proportion of infected children aged over 1 year in differed from Cai et al. [31], a discrepancy Likely attributable to the small sample size, single-center design and exclusion of PCR-confirmed cases in this study, but a marked increase in infections among children aged over 3 years was observed in 2024. Increased white blood cells (over 10 × 109 cells/L) were observed in 88.24% of all pertussis patients, serving as predictable but nonspecific indexes for pertussis infection during screening [4]. Notably, China’s DTaP (Diphtheria, Tetanus, acelluar Pertussis vaccine) immunization schedule prior to 2025 consisted of primary series doses at 3, 4, 5, and 18 months of age, followed by a DT booster at the age of 6 years. In response to the evolving pertussis epidemiology, the National Disease Control and Prevention Administration implemented an updated schedule effective on January 1, 2025, which introduces earlier initiation (2 months of age) and adds a pertussis-containing booster at 6 years of age. Our findings strongly support the strategic Shift toward earlier vaccination initiation at 2 months. Nevertheless, ongoing surveillance remains essential to evaluate the effectiveness of these updated immunization strategies.
The diagnosis of pertussis typically relies on B. pertussis nucleic acids detection, culture from nasopharyngeal specimens, and serological recognition testing [32, 33]. Whole-genome sequencing (WGS) provides feasibility and efficacy for genotyping and phylogenetical analysis. Also, multi-locus variable-number tandem-repeat analysis (MLVA) [34] and pulsed-field gel electrophoresis (PFGE) [35] are effective and feasible approaches for strain genotyping. However, the primary B. pertussis MLVA reference database (http://www.mlva.net/) is no longer maintained. Consequently, advancing higher-resolution typing methodologies for B. pertussis isolates is essential to accurately characterize emerging subclones and novel genotypes.
This study documents the regional emergence and dominance of macrolide-resistant ptxP3 isolates in Guangzhou, China by 2024, providing critical molecular epidemiological insights for pertussis surveillance. There are some limitations in this study: (1) the Limited number of isolates recovered prior to 2023 severely impedes comprehensive analysis of early clonal dynamics and the complete timeline of strain replacement; (2) the single-center design restricts the generalizability of the findings regarding the prevalence of the dominant clone across the broader southern China regions; (3) unavailable clinical information for some patients may limit the representativeness of the reported prevalence; (4) the inability to determine MLVA types hinders investigation into the potential origins of the study isolates, as MLVA remains valuable for certain phylogenetic comparisons.
Conclusion
This study documents a significant genotypic shift among Bordetella pertussis isolates circulating in Guangzhou, China, between 2020 and 2024. We observed the emergence and dominance of macrolide-resistant B. pertussis strains harboring ptxP3 and prn150 by 2024. Phylogenetic analysis revealed that these resistant isolates formed a distinct sub-clade within the ptxP3 lineage. Given the high prevalence of macrolide resistance, antimicrobial agents such as trimethoprim/sulfamethoxazole (SXT) may serve as therapeutic alternative. Collectively, these findings provide molecular epidemiological and microbiological insights to guide pertussis control strategies and to offer antimicrobial resistance monitoring efforts in southern China.
Supplementary Information
Abbreviations
- PTx
Pertussis toxin
- Prn
Pertactin
- MICs
Minimum inhibitory concentrations
- ERY
Erythromycin
- AZM
Azithromycin
- CLI
Clindamycin
- CLR
Clarithromycin
- CSL
Cefoperazone/sulbactam
- TZP
Piperacillin/tazobactam
- SXT
Trimethoprim/sulfamethoxazole
Authors’ contributions
ML, QM, YL, and JG designed the experiment and revised the manuscript. JW and YQ analyzed sequences data. WL, WD, CJ, RM, MZ, and ZC performed laboratory testing and experiments. YC, CG, LG, HL participated in the experiments. YT and DH collected clinical information.
Funding
This study was funded by the Research Project of the Guangdong Provincial Administration of Traditional Chinese Medicine (Grant no. 20251040 and 20251044), the Guangdong Foundation for Basic and Applied Basic Research (Grant no. 2022A1515012226), the Guangdong Medical Research Foundation (Grant no. B2024025 and B2025291).
Data availability
Genomic sequencing data have been deposited in the National Center of Biotechnology Information (NCBI) database with the accession code PRJNA1193776.
Declarations
Ethics approval and consent to participate
This study was approved by the Medical Ethics Committee of Guangdong Women and Children Hospital, China, and it was conducted according to the ethical guidelines of the Declaration of Helsinki. The informed consent to participants was waived by the Medical Ethics Committee of Guangdong Women and Children Hospital and the use of residual specimens was strictly anonymized in this study.
Consent for publication
All authors have read and approved the final manuscript for publication.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Qiongdan Mai, Jinzhou Wen, Yasha Luo and Junfei Guo contributed equally to this work.
References
- 1.Mohamed YF, Manivannan K, Fernandez RC. Bordetella pertussis. Trends Microbiol. 2023;31(11):1192–3. 10.1016/j.tim.2023.03.012. [DOI] [PubMed] [Google Scholar]
- 2.Domenech de Celles M, Rohani P. Pertussis vaccines, epidemiology and evolution. Nat Rev Microbiol Nov. 2024;22(11):722–35. 10.1038/s41579-024-01064-8. [DOI] [PubMed] [Google Scholar]
- 3.Cherry JD. Pertussis in Young Infants Throughout the World. Clin Infect Dis. Dec 1. 2016;63(suppl 4):S119-s122. 10.1093/cid/ciw550 [DOI] [PMC free article] [PubMed]
- 4.Liu Y, Yu D, Wang K, et al. Global resurgence of pertussis: A perspective from China. The J Infect Nov. 2024;89(5):106289. 10.1016/j.jinf.2024.106289. [DOI] [PubMed] [Google Scholar]
- 5.Melvin JA, Scheller EV, Miller JF, et al. Bordetella pertussis pathogenesis: current and future challenges. Nat Rev Microbiol Apr. 2014;12(4):274–88. 10.1038/nrmicro3235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Belcher T, Dubois V, Rivera-Millot A, et al. Pathogenicity and virulence of Bordetella pertussis and its adaptation to its strictly human host. Virulence. 2021;12(1):2608–32. 10.1080/21505594.2021.1980987. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Mooi FR, van Loo IH, van Gent M, et al. Bordetella pertussis strains with increased toxin production associated with pertussis resurgence. Emerging Infect Dis Aug. 2009;15(8):1206–13. 10.3201/eid1508.081511. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bart MJ, Harris SR, Advani A, et al. Global population structure and evolution of Bordetella pertussis and their relationship with vaccination. mBio. 2014. 10.1128/mBio.01074-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Li L, Deng J, Ma X, et al. High prevalence of macrolide-resistant Bordetella pertussis and ptxP1 genotype, Mainland China, 2014–2016. Emerg Infect Dis. 2019;25(12):2205–14. 10.3201/eid2512.181836. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Cai J, Chen M, Liu Q, et al. Domination of an emerging erythromycin-resistant ptxP3 Bordetella pertussis clone in Shanghai, China. Int J Antimicrob Agents. 2023;62(1):106835. 10.1016/j.ijantimicag.2023.106835. [DOI] [PubMed] [Google Scholar]
- 11.Weigand MR, Williams MM, Peng Y, et al. Genomic survey of Bordetella pertussis diversity, united states, 2000–2013. Emerging Infect Dis Apr. 2019;25(4):780–3. 10.3201/eid2504.180812. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Fu P, Yan G, Li Y, et al. Pertussis upsurge, age shift and vaccine escape post-COVID-19 caused by ptxP3 macrolide-resistant Bordetella pertussis MT28 clone in China. Clinical Microbiol Infection: Official Publication Eur Soc Clin Microbiol Infect Dis Nov. 2024;30(11):1439–46. 10.1016/j.cmi.2024.08.016. [DOI] [PubMed] [Google Scholar]
- 13.Kamachi K, Duong HT, Dang AD, et al. Macrolide-resistant Bordetella pertussis, Vietnam, 2016–2017. Emerg Infect Dis. 2020;26(10):2511–3. 10.3201/eid2610.201035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Yamaguchi T, Kawasaki Y, Katsukawa C, et al. The first report of Macrolide-Resistant Bordetella pertussis isolation in Japan. Japanese J Infect Dis Sep. 2020;24(5):361–2. 10.7883/yoken.JJID.2019.421. [DOI] [PubMed] [Google Scholar]
- 15.Yao K, Deng J, Ma X, et al. The epidemic of erythromycin-resistant Bordetella pertussis with limited genome variation associated with pertussis resurgence in China. Expert Rev Vaccines. 2020;19(11):1093–9. 10.1080/14760584.2020.1831916. [DOI] [PubMed] [Google Scholar]
- 16.Control ECfDPa. Laboratory diagnosis and molecular surveillance of Bordetella pertussis Stockholm: ECDC. 2022;2022.
- 17.Barkoff AM, Mertsola J, Pierard D, et al. Surveillance of circulating Bordetella pertussis strains in Europe during 1998 to 2015. J Clin Microbiol. 2018. 10.1128/jcm.01998-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Mir-Cros A, Moreno-Mingorance A, Martin-Gomez MT, et al. Population dynamics and antigenic drift of Bordetella pertussis following whole cell vaccine replacement, Barcelona, Spain, 1986–2015. Emerg Microbes Infect. 2019;8(1):1711–20. 10.1080/22221751.2019.1694395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Wang Z, Luan Y, Du Q, et al. The global prevalence ptxP3 lineage of Bordetella pertussis was rare in young children with the co-purified aPV vaccination: a 5 years retrospective study. BMC Infect Dis. 2020;(1):615. 10.1186/s12879-020-05332-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wu X, Du Q, Li D, et al. A cross-sectional study revealing the emergence of Erythromycin-resistant Bordetella pertussis carrying ptxP3 alleles in China. Front Microbiol. 2022;13:901617. 10.3389/fmicb.2022.901617. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Yang Y, Yao K, Ma X, et al. Variation in Bordetella pertussis susceptibility to erythromycin and virulence-related genotype changes in China (1970–2014). PLoS One. 2015;10(9):e0138941. 10.1371/journal.pone.0138941. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Wang Z, Cui Z, Li Y, et al. High prevalence of erythromycin-resistant Bordetella pertussis in xi’an, China. Clinical Microbiol Infection: Official Publication Eur Soc Clin Microbiol Infect Dis Nov. 2014;20(11):O825–830. 10.1111/1469-0691.12671. [DOI] [PubMed] [Google Scholar]
- 23.Tian S, Chen Y, Su Q, et al. Pertussis clinical profile shift, severity, prediction in a tertiary hospital: A comparative study before, during, and after COVID-19 in Southern China. J Infect Public Health Nov. 2024;29(1):102610. 10.1016/j.jiph.2024.102610. [DOI] [PubMed] [Google Scholar]
- 24.Tang B, Hu X, Wu B, et al. Global antimicrobial resistance threats: insights from the resurgence of whooping cough. J Infect. 2024;89(6):106337. 10.1016/j.jinf.2024.106337. [DOI] [PubMed] [Google Scholar]
- 25.Zhuang Y, Li S, Bao D, et al. Genomic integrity in Bordetella pertussis: avoiding contaminant-derived misinterpretations of acquired antimicrobial resistance. J Infect. 2024;89(6):106356. 10.1016/j.jinf.2024.106356. [DOI] [PubMed] [Google Scholar]
- 26.Shi W, Meng Q, Hu Y, et al. Modifying antibiotic treatment strategies in the face of pertussis surge associated to erythromycin resistance in China. J Infect. 2024;88(6):106174. 10.1016/j.jinf.2024.106174. [DOI] [PubMed] [Google Scholar]
- 27.Zhu X, Wang Z. Resurgence of pertussis in china: evaluating the efficacy of sulfamethoxazole-trimethoprim as an alternative treatment. The J Infect Dec. 2024;4(1):106373. 10.1016/j.jinf.2024.106373. [DOI] [PubMed] [Google Scholar]
- 28.Zhang L, Xu Y, Zhao J, et al. Effect of vaccination on Bordetella pertussis strains, China. Emerg Infect Dis. 2010;16(11):1695–701. 10.3201/eid1611.100401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Mooi FR. Bordetella pertussis and vaccination: the persistence of a genetically monomorphic pathogen. Infect Genet Evol. 2010;10(1):36–49. 10.1016/j.meegid.2009.10.007. [DOI] [PubMed] [Google Scholar]
- 30.Charles IG, Li JL, Roberts M, et al. Identification and characterization of a protective immunodominant B cell epitope of Pertactin (P.69) from Bordetella pertussis. Eur J Immunol. 1991;21(5):1147–53. 10.1002/eji.1830210509. [DOI] [PubMed] [Google Scholar]
- 31.Cai J, Liu Q, Chen B, et al. Waning immunity, prevailing non-vaccine type ptxP3 and macrolide-resistant strains in the 2024 pertussis outbreak in China: a multicentre cross-sectional descriptive study. Lancet Reg Health. 2025;60:101628. 10.1016/j.lanwpc.2025.101628. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Peng Y, Williams MM, Xiaoli L, et al. Strengthening Bordetella pertussis genomic surveillance by direct sequencing of residual positive specimens. Journal Clin Microbiol Apr. 2024;10(4):e0165323. 10.1128/jcm.01653-23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Toizumi M, Le LT, Nguyen HAT, et al. Sero and carriage epidemiology of pertussis in urban and rural regions in Vietnam. Vaccines Feb. 2024;23(3). 10.3390/vaccines12030225. [DOI] [PMC free article] [PubMed]
- 34.Schouls LM, van der Heide HG, Vauterin L, et al. Multiple-locus variable-number tandem repeat analysis of Dutch Bordetella pertussis strains reveals rapid genetic changes with clonal expansion during the late 1990s. J Bacteriol Aug. 2004;186(16):5496–505. 10.1128/JB.186.16.5496-5505.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Bouchez V, Guglielmini J, Dazas M, et al. Genomic sequencing of Bordetella pertussis for epidemiology and global surveillance of whooping cough. Emerging Infect Dis Jun. 2018;24(6):988–94. 10.3201/eid2406.171464. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Genomic sequencing data have been deposited in the National Center of Biotechnology Information (NCBI) database with the accession code PRJNA1193776.