Isolation, Antimicrobial Resistance Phenotypes, and Virulence Genes of Bordetella bronchiseptica From Pigs in China, 2018–2020

Bordetella bronchiseptica is a leading cause of respiratory diseases in pigs. However, epidemiological data of B. bronchiseptica in pigs particularly in China, the largest pig rearing country in the world is still limited. We isolated 181 B. bronchiseptica strains from 4259 lung samples of dead pigs with respiratory diseases in 14 provinces in China from 2018 to 2020. The average isolation rate of this 3-year period was 4.25% (181/4259). Antimicrobial susceptibility testing performed by disc diffusion method revealed that most of the B. bronchiseptica isolates in this study were resistant to ampicillin (83.98%), while a proportion of isolates were resistant to cefotaxime (30.39%%), chloramphenicol (12.71%), gentamicin (11.60%), florfenicol (11.60%), tetracycline (8.84%), amoxicillin (8.29%), tobramycin (6.63%), ceftriaxone (4.97%), and cefepime (0.55%). There were no isolates with resistant phenotypes to imipenem, meropenem, polymyxin B, ciprofloxacin, enrofloxacin, and amikacin. In addition, ~13.18% of the isolates showed phenotypes of multidrug resistance. Detection of antimicrobial resistance genes (ARGs) by PCR showed that 16.57% of the B. bronchiseptica isolates in this study was positive to aac(3)-IV, while 3.87%, 2.21%, 1.10%, 0.55%, 0.55%, and 0.55% of the isolates were positive to aac6'-Ib, rmtA, blaTEM, blaSHV, oqxB, and tetA, respectively. Detection of virulence factors encoding genes (VFGs) by conventional PCR showed that over 90% of the pig B. bronchiseptica isolates in this study were positive to the five VFGs examined (fhaB, 97.24%; prn, 91.16%; cyaA, 98.34%; dnt, 98.34%; betA, 92.82%). These results demonstrate B. bronchiseptica as an important pathogen associated with pig respiratory disorders in China. The present work contributes to the current understanding of the prevalence, antimicrobial resistance and virulence genes of B. bronchiseptica in pigs.

Bordetella bronchiseptica is a leading cause of respiratory diseases in pigs. However, epidemiological data of B. bronchiseptica in pigs particularly in China, the largest pig rearing country in the world is still limited. We isolated 181 B. bronchiseptica strains from 4259 lung samples of dead pigs with respiratory diseases in 14 provinces in China from 2018 to 2020. The average isolation rate of this 3-year period was 4.25% (181/4259). Antimicrobial susceptibility testing performed by disc diffusion method revealed that most of the B. bronchiseptica isolates in this study were resistant to ampicillin (83.98%), while a proportion of isolates were resistant to cefotaxime (30.39%%), chloramphenicol (12.71%), gentamicin (11.60%), florfenicol (11.60%), tetracycline (8.84%), amoxicillin (8.29%), tobramycin (6.63%), ceftriaxone (4.97%), and cefepime (0.55%). There were no isolates with resistant phenotypes to imipenem, meropenem, polymyxin B, ciprofloxacin, enrofloxacin, and amikacin. In addition, ∼13.18% of the isolates showed phenotypes of multidrug resistance. Detection of antimicrobial resistance genes (ARGs) by PCR showed that 16.57% of the B. bronchiseptica isolates in this study was positive to aac(3)-IV, while 3.87%, 2.21%, 1.10%, 0.55%, 0.55%, and 0.55% of the isolates were positive to aac6'-Ib, rmtA, bla TEM , bla SHV , oqxB, and tetA, respectively. Detection of virulence factors encoding genes (VFGs) by conventional PCR showed that over 90% of the pig B. bronchiseptica isolates in this study were positive to the five VFGs examined (fhaB, 97.24%; prn, 91.16%; cyaA, 98.34%; dnt, 98.34%; betA, 92.82%). These results demonstrate B. bronchiseptica as an important pathogen associated with pig respiratory disorders in China. The present work contributes to the current understanding of the prevalence, antimicrobial resistance and virulence genes of B. bronchiseptica in pigs.
Keywords: Bordetella bronchiseptica, isolation, antimicrobial resistance, virulence factors encoding genes, pigs INTRODUCTION Bordetella bronchiseptica is an aerobic, motile, gram-negative rod, or coccobacillus belonging to genus Bordetella. It is an important pathogenic bacterium in agriculture and in veterinary medicine (1). In veterinary medicine, B. bronchiseptica is a leading cause of many respiratory infections including rhinitis, tracheitis, bronchitis, and pneumonia in a wide spectrum of animals (2). It can also enhance respiratory colonization of Streptococcus suis and Haemophilus parasuis, promote disease caused by S. suis, and interact with porcine reproductive and respiratory syndrome virus (PRRSV) and swine influenza virus (SIV) to increase severity of respiratory disease (3). While rarely to be reported, B. bronchiseptica is also potentially involved in infections in humans, and human cases are frequently associated with direct contact with infected animals such as swine, dog, rabbit and/or cat (4)(5)(6). Similar to the other members belonging to genus Bordetella, many B. bronchiseptica produces several important virulence factors, including filamentous hemagglutinin, and protein toxins, adenylate cyclase toxin, pertussis toxin, dermonecrotic toxin as well as type III secretion system (T3SS) and effector proteins, contributing to its pathogenesis (7,8).
In swine, B. bronchiseptica is proposed as a main causative agent of porcine respiratory disease complex (PRDC) and atrophic rhinitis; both of which are economically-important diseases in pig industry (9,10). Continuously monitoring the prevalence, antimicrobial resistance (AMR) and virulence profiles of B. bronchiseptica in pigs are beneficial for the prevention and control of swine bordetellosis. However, the relevant data are still limited. China is the largest pig-farming and pork consuming country in the world. Although the outbreak of African Swine Fever in August 2018 caused a huge loss of pigs in China, there are still more than 406 million pigs rearing in China in 2020 (11). To understand the current epidemiological and microbiological characteristics such as the antimicrobial resistance profiles of B. bronchiseptica isolates from pigs in China, we performed bacterial isolation of B. bronchiseptica strains from lung samples of dead pigs with a history of respiratory disorders in China from 2018 to 2020 in this study. These isolates were characterized by testing the antimicrobial susceptibility and detecting the antimicrobial resistance genes (ARGs) as well as virulence encoding genes (VFGs).

Study Design, Sample Collection, and Ethic Statement
Study design was shown in Figure 1A. From 2018 to 2020, a total of 4259 lung samples (3022 samples in 2018, 841 samples in 2019, 396 samples in 2020) from 14 provinces (Guangdong, Henan, Hubei, Shandong, Fujian, Hebei, Zhejiang, Hunan, Anhui, Sichuan, Shanxi, Inner Mongolia, Xinjiang, Guizhou) in China were used for B. bronchiseptica isolation and identification ( Figure 1B). All of the clinical samples used in this study were submitted by veterinarians/or the farm owners to the Veterinary Diagnostic Laboratory of Huazhong Agricultural University (Wuhan, China) for routine testing.

Bacterial Isolation and Identification
Collected samples (∼10 grams per sample) were cut into pieces and lysed in sterile 0.9% normal saline by using a TissueLyser II (QIAGEN, Venlo, Netherlands). Thereafter, tissue homogenates of each sample were streak-plated onto one tryptic soy agar (TSA; Becton, Dickinson and Company, MD, USA) containing 10 µg/ml nicotinamide adenine dinucleotide (NAD; Sigma, St. Louis, MO) and 10% new-born bovine serum. The agar plates were incubated at 37 • C for 24∼48 h. Isolates growing on the plates were then purified and cultured following the standard methods used for bacterial identification (12). On each of the agar plates, five colonies with similar morphological characteristics to B. bronchiseptica [small circular glistening or rough colonies with 0.5 to 1.0 mm in diameter after 48 h of incubation in air at 37 • C (4)] were selected for biochemical test. Presumptive isolates of B. bronchiseptica were finally confirmed using polymerase chain reaction (PCR) assay amplifying the species-specific gene fla with the primers listed in Table 1 (26). Considering B. bronchiseptica possesses only one serotype (27), we therefore chose one colony confirmed by both PCR and biochemical tests (positive for fla and displaying similar biochemical characteristics to B. bronchiseptica) to represent B. bronchiseptica strain recovered for its corresponding sample.

Antimicrobial Susceptibility Testing
Antimicrobial susceptibility of the B. bronchiseptica isolates was tested by using Disk diffusion method following the Clinical and Laboratory Standards Institute (CLSI) antimicrobial susceptibility testing standards (28). Briefly, purified overnightcultured colonies of B. bronchiseptica were picked up from TSA plates and resuspended in sterile 0.9% normal saline to 0. 5 (28). As clinic breakpoints specific to B. bronchiseptica are limited available (2), we thereby used breakpoints to Enterobacteriaceae published in CLSI document M100 for result-interpretation in this study. Breakpoints used are listed in Table 2. Escherichia coli ATCC R * 25922 was used as quality control.

Detection of Virulence Factors Encoding Genes
The presence of five well-characterized VFGs, including the filamentous haemagglutinin encoding gene fhaB, the pertactin encoding gene prn, the adenylate cyclase-haemolysin toxin encoding gene cyaA, the dermonecrotic toxin encoding gene dnt, and the Bordetella type-III secretion system effector A encoding gene bteA in each of the isolates were examined by PCR with primers listed in Table 1, as described previously (25). PCR assays were performed in a 20-µL reaction mixture comprised of 2-µL bacterial DNA, each of the forward and reverse primers 1-µL, 2×Taq Master Mix (Dye Plus) 10-µL, DMSO 2-µL, and ddH 2 O 4-µL. The cycling conditions were 94 • C for 5 min, followed by 35 cycles of 94 • C for 30 s, 59 • C for 30 s and 72 • C for 30 s, and a final extension at 72 • C for 5 min. Our laboratory stored B. bronchiseptica strain HH0809 (31) and the sterile ddH 2 O were included as the positive and negative controls, respectively. PCR products were analyzed by electrophoresis on a 1% agarose gel.

Statistical Analysis
We used SAS version 9.0 (SAS Institute Inc.) software to perform statistical analyses in this study, as described previously (26). Univariate association between variables and isolation rates of B. bronchiseptica was determined by using univariate ordinary logistic regression analysis. P < 0.05 was considered to be significant.

B. bronchiseptica Isolation and Identification
From 2018 to 2020, we isolated a total of 181 B. bronchiseptica strains (4.25%) from 4259 lung samples of dead pigs with   Figures 1B,C). Biochemical tests revealed that B. bronchiseptica isolates could not ferment fructose, glucose, mannitol, maltose, rhamnose, and lactose; the methyl red (MR), voges-proskauer (VP), and indole reactions were negative. It is positive testes for oxidase and catalase.

DISCUSSION
Although B. bronchiseptica is a well-known leading cause of pig respiratory disorders and an important causative agent of PRDC, there is not too much report on the epidemiology of B. bronchiseptica in pigs round the world, particularly in China, the largest pig rearing and production country. In this study, we described the isolation and characterization of B. bronchiseptica in pigs in China from 2018 to 2020. The average isolation rate of this 3-year period was 4.25% (181/4259), which is much lower than that reported in pigs with clinical respiratory disease in China from 2003 to 2008 (4.25 vs. 18.6%, P < 0.05) (26). The average isolation rates of B. bronchiseptica in pigs in different regions from 2018 to 2020 were also much lower than those reported in the same regions from 2003 to 2008 (Hubei: 3.48 vs. 18.0%, P < 0.05; Henan: 3.42 vs. 19.6%, P < 0.05; Fujian: 4.14 vs. 18.4%, P < 0.05; Hunan: 5.96 vs. 19.2%, P < 0.05; Anhui: 7.32 vs. 18.0%, P < 0.05; Shandong: 3.98 vs. 20.7%, P < 0.05) (26). The significant decreasing average isolation rate of B. bronchiseptica from 2018 to 2020 compared to that from 2003 to 2008 might be owing to China's continuously efforts to promote transformation and upgrading of pig industry as well as improve the level of disease prevention and control in pig farms. In addition, the outbreak of African Swine Fever in 2018 and its continuous circulation in pigs in China also accelerates the improvement and enhancement of biosecurity on pig farms in recent years (33), which may also be beneficial for the control of B. bronchiseptica and the other pathogens.
Administration of antimicrobials is still one of the most effective way to control B. bronchiseptica and the other bacteria, but the emergence of drug-resistant bacteria may lead to the failure of using antibiotics in clinic (34)(35)(36). Therefore, monitoring the drug resistance profile of clinical microbiology is an important aspect in many epidemiological studies (25,37,38). In this study, we characterized the resistance phenotypes of B. bronchiseptica from pigs in China from 2018 to 2020. The results revealed that all isolates were susceptible to imipenem (100%), meropenem (100%), and polymyxin B (100%). All of these three types of antibiotics are proposed to be the last-resort antibiotics for the treatment of infections caused by gram-negative bacteria (29), and they are not approved to be used in veterinary medicine in China. In addition, the majority of the isolates were sensitive to ciprofloxacin (99.45%), cefepime (97.79%), enrofloxacin (97.79%), tobramycin (92.27%), gentamicin (86.74%), florfenicol (86.74%), chloramphenicol (86.19%), tetracycline (85.08%), amikacin (83.43%), and amoxicillin (83.43%). These results are in agreement with the results of previous studies in China (25,39), as well as in other countries such as Germany and Korea (2,(40)(41)(42), suggesting these antibiotics might be suitable candidates for treating B. bronchiseptica infections when necessary. A high level of resistance was found for ampicillin (83.98%), followed by resistance for cefotaxime (30.39%). These findings are also in agreement with those from the other articles (2,25,39), and in particular, B. bronchiseptica is documented to be commonly resistant to ampicillin (2). Therefore, these drugs are not recommended to be used in clinic settings. It should be also reminded that several B. bronchiseptica isolates from pigs in China displayed a level of multidrug resistance, particularly co-resistance to aminoglycosides, broadspectrum-cephalosporins, and penicillins. Continues studies should be taken to monitor the prevalence and changetrend of these MDR-isolates in clinic, as some antibiotics belonging to aminoglycosides, broad-spectrum-cephalosporins, and penicillins are commonly used for treating B. bronchiseptica infections in veterinary medicine (2,35).
Virulence factors (VFs) play an important role in the pathogenesis of bacteria (43). For B. bronchiseptica, important VFs include filamentous haemagglutinin (FHA), pertactin (PRN), adenylate cyclase-haemolysin toxin, dermonecrotic toxin (DNT), and types III secretion system (44)(45)(46)(47)(48), and the expression of these VFs facilitates the invasion of B. bronchiseptica in hosts (49). In the present study, we examined five genes encoding these VFs, including fhaB which encodes filamentous haemagglutinin; prn which encodes pertactin; cyaA which encodes adenylate cyclase-haemolysin toxin; dnt which encodes DNT; and bteA which encodes the T3SS effector A. Surprisingly, over 90% of the pig B. bronchiseptica isolates in this study were positive to these five VFGs examined (fhaB, 97.24%; prn, 91.16%; cyaA, 98.34%; dnt, 98.34%; betA, 92.82%). Importantly, approximately 84.36% of the isolates contained these five kinds of VFGs simultaneously. These results are also in agreement with those reported in B. bronchiseptica isolates from rabbits in China (25), suggesting carrying of these VFGs are broad characteristics of B. bronchiseptica. Laboratory studies have shown that FHA, and PRN expressed in E. coli and Salmonella enterica, as well as adenylate cyclase-haemolysin toxin expressed in B. bronchiseptica provide protection against fatal infections with B. bronchiseptica in mouse models (5,50,51).
Despite the findings, this work has several limitations that should be noted. First, all samples used for bacterial isolation were submitted by pig farms from different provinces in China. This way of sample collection may have some influences on the isolation rate. However, the outbreak of African Swine Fever since 2018 and its continuous circulation in pigs in China, and more recently, the worldwide pandemic of the novel coronavirus disease since the late 2019 (COVID-19) made it very difficult for us to collect samples initiatively. Second, the results of antimicrobial susceptibility testing in this study were interpreted by using breakpoints to Enterobacteriaceae published in CLSI document M100, and this is because clinic breakpoints specific to B. bronchiseptica are limited available (2). Third, a very few published epidemiological studies of swine B. bronchiseptica in China are available to date [On March 18, 2021, we searched PubMed with key words "(((Bordetella bronchiseptica) AND (Prevalence)) AND (Pigs)) AND (China)" for reports published, with no language restrictions. Our search identified two articles (26,39) of relevance to this study. All of them were published by our group in 2011], therefore, we only compared the results we obtained from this study to those reported in our previously published two studies in 2011 (26,39). However, the results from this work could still help understand the current epidemiological and microbiological characteristics of B. bronchiseptica in pigs in China.
In summary, we reported the isolation, antimicrobial resistance phenotypes, the detection of ARGs and VFGs of B. bronchiseptica from pigs in China from 2018 to 2020 in this study. Our results showed that B. bronchiseptica remains an important pathogen associated with pig respiratory disorders in China. While most of the isolates were still susceptible to ciprofloxacin, cefepime, enrofloxacin, tobramycin, gentamicin, florfenicol, chloramphenicol, tetracycline, amikacin, and amoxicillin, MDR-isolates were still determined. These isolates should receive more attentions and further studies are necessary to monitor the prevalence of drug-resistant B. bronchiseptica. In addition, our results also revealed that several VFGs, including fhaB, prn, cyaA, dnt, and betA displayed a high level of detection rate.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

AUTHOR CONTRIBUTIONS
YZ, ZP, and BW delineated the study conception and design. ZP and BW supervised the study. YZ, HY, LG, MZ, FW, WS, LH, LW, WL, and XT collected the bacterial isolates and performed laboratory tests as well as analyzed the data. ZP and YZ wrote the manuscript and approved the final version for publication. ZP, BW, and WL participated in the manuscript discussion and revision. All authors have read and approved the final version of the manuscript.

FUNDING
This work was supported in part by the Agricultural Science and Technology Innovation Program of Hubei Province (Grant Number: 2018skjcx05). ZP acknowledges the financial support from China Postdoctoral Science Foundation (Grant Numbers: 2020T130232 and 2018M640719). The funders have no role in the study design, data collection and interpretation, or the decision to submit the work for publication.