Characterization of antimicrobial resistance of Pseudomonas aeruginosa isolated from canine infections

Authors

  • D. Lin,

    1.  Department of Small Animal Clinical Sciences, College of Veterinary Medicine, China Agricultural University, Beijing, China
    Search for more papers by this author
    • These authors contributed equally to this work.

  • S.L. Foley,

    1.  Division of Microbiology, National Center for Toxicological Research, Food and Drug Administration, Jefferson, AR, USA
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Y. Qi,

    1.  College of Animal Science, Henan Institute of Science and Technology, Xinxiang, China
    Search for more papers by this author
  • J. Han,

    1.  Division of Microbiology, National Center for Toxicological Research, Food and Drug Administration, Jefferson, AR, USA
    Search for more papers by this author
  • C. Ji,

    1.  Beijing Key Laboratory of Detection Technology for Animal-Derived Food Safety, College of Veterinary Medicine, China Agricultural University, Beijing, China
    Search for more papers by this author
  • R. Li,

    1.  Beijing Key Laboratory of Detection Technology for Animal-Derived Food Safety, College of Veterinary Medicine, China Agricultural University, Beijing, China
    Search for more papers by this author
  • C. Wu,

    1.  Beijing Key Laboratory of Detection Technology for Animal-Derived Food Safety, College of Veterinary Medicine, China Agricultural University, Beijing, China
    Search for more papers by this author
  • J. Shen,

    1.  Beijing Key Laboratory of Detection Technology for Animal-Derived Food Safety, College of Veterinary Medicine, China Agricultural University, Beijing, China
    Search for more papers by this author
  • Y. Wang

    1.  Beijing Key Laboratory of Detection Technology for Animal-Derived Food Safety, College of Veterinary Medicine, China Agricultural University, Beijing, China
    Search for more papers by this author

Yang Wang, Beijing Key Laboratory of Detection Technology for Animal-Derived Food Safety, College of Veterinary Medicine, China Agricultural University, Beijing 100193, China. E-mail: wangyang@cau.edu.cn

Abstract

Aims:  To determine the prevalence of Pseudomonas aeruginosa among dogs with suspected soft tissue infections and to characterize these isolates.

Methods and Results:  Swabs were taken from infected soft tissues of 402 dogs. Pseudomonas aeruginosa strains were confirmed phenotypically and tested for susceptibility to 11 antimicrobial agents and genotyped by SpeI pulsed-field gel electrophoresis (PFGE). The genetic basis of fluoroquinolone (FQ) resistance and the presence of integrons were also characterized. A total of 27 (6·7%) dogs tested positive for Ps. aeruginosa. Fourteen different SpeI patterns were observed in 25 typeable strains. Among the β-lactams, three isolates presented resistance to ticarcillin and carbenicillin, while only one isolate exhibited resistance to ceftazidime. Among the aminoglycosides (AGs), three strains showed resistance to amikacin, and four strains exhibited resistance to gentamicin and tobramycin. Four strains with mutations that led to the substitution of Thr at position 83 with Ile in GyrA and the exchange of Ser at position 87 with Leu in ParC displayed resistance to all tested FQs. These strains also carried class 1 integrons and showed resistance to between 6 and 10 antimicrobials. These integrons included four different gene cassettes (aacA4-aadA1, blaOXA-31-aadA2, aadA1-arr-3-catB3 and cmlA5-cmlA-aadA1).

Conclusions:  A small proportion of infected dogs treated in two animal hospitals in Beijing, China carried Ps. aeruginosa isolates. Low levels of resistance to anti-pseudomonal agents were observed in these strains.

Significance and Impact of the Study:  This study is the first report on the antimicrobial resistance profiles of Ps. aeruginosa isolated from infected canine origin in China. Additionally, this is the first report of the oxacillin resistance gene blaOXA-31 in a canine Ps. aeruginosa isolate.

Introduction

Pseudomonas aeruginosa is a rod-shaped, Gram-negative, glucose-nonfermenting aerobic bacterium. It is an important pathogen to both humans and other animals, but is rarely involved in primary disease. In humans, Ps. aeruginosa is an important opportunistic nosocomial pathogen, particularly evident in hospital-acquired pneumonia in immunocompromised patients (Sarlangue et al. 2006). In animals, especially in dogs, it has been assigned as the distinct cause of infections such as otitis externa, chronic deep pyoderma and wound/urinary tract infections (Hariharan et al. 2006; Hillier et al. 2006; Mekic et al. 2011). Usually, antibiotics are effective in treating bacterial infection; however, Ps. aeruginosa is notorious for its resistance to multiple antibiotics. Because of its low outer membrane permeability (Hancock 1998), intrinsic or induced expression of efflux pumps (Li et al. 2000), and β-lactamase production, Ps. aeruginosa is uniformly resistant to a wide range of antimicrobials including ampicillin, first- and second-generation cephalosporins, and erythromycin and is also often resistant to chloramphenicol and tetracycline (Nikaido 1994). In addition to these intrinsic resistances, Ps. aeruginosa can also acquire resistance through horizontal transfer of genetic elements that carry resistance genes (integrons, transposons, and plasmids) as well as mutational resistance (e.g. mutation in DNA gyrase and topoisomerase IV) (Livermore 2002; Linder et al. 2005). Owing to the highly variable resistance patterns, the empiric therapy of Ps. aeruginosa infection using penicillins, third- and fourth-generation cephalosporins, carbapenemes, aminoglycosides (AGs) and fluoroquinolone (FQs) may be ineffectual. Thus, antimicrobial susceptibility testing should be a crucial step in the selection of appropriate antimicrobial therapy in both human and veterinary medicine.

Until now, limited data have been available on the antimicrobial resistance profiles of Ps. aeruginosa of canine origin, with the available data limited primarily to the USA and European countries (Petersen et al. 2002; Teresa Tejedor et al. 2003; Ledbetter et al. 2007; Rubin et al. 2008). Currently, no epidemiological investigation data related to Ps. aeruginosa of canine origin in China are available. The objectives of this study were to determine the prevalence and the antimicrobial resistance profiles of Ps. aeruginosa isolates from samples of infected canines in two veterinary hospitals in Beijing, China. Furthermore, FQ resistance and integrons in these Ps. aeruginosa isolates were also characterized.

Materials and methods

Bacterial isolates

A total of 402 swabs from dogs with soft tissue infections, for example, otitis externa, pyoderma and wound discharge, were collected from two animal hospitals (372 and 30 specimens, respectively): the Veterinary Teaching Hospital of China Agricultural University (VTH-CAU) and Beijing Saijia Veterinarian Center (BSVC) during an 18-month period (from May 2009 to August 2010). One specimen per dog was collected in this study and the swabs were directly inoculated on Acetamide agar medium (Aoboxing, Beijing, China) and incubated for 24 h at 37°C. Colonies with purple-red colour were selected and subcultured onto trypticase soy agar (TSA) plates (Aoboxing) supplemented with 5% defibrinated sheep blood (Aoboxing). The Ps. aeruginosa isolates were further confirmed by ATB 32GN system (BioMerieux, Marcy-L’Etoile, France). A total of 27 confirmed Ps. aeruginosa strains were suspended in TSA containing 15% glycerol and stored at −80°C until use.

Antimicrobial susceptibility testing

Antimicrobial susceptibilities for all Ps. aeruginosa isolates were determined by microdilution broth methods according to Clinical and Laboratory Standards Institute documents M31-A3 (2008) and M100-S21 (2011). Eleven antimicrobial agents from three classes of antimicrobial agents including β-lactams (ceftazidime, ticarcillin, carbenicillin, imipenem), FQs (norfloxacin, levofloxacin, ciprofloxacin, lomefloxacin) and AGs (gentamicin, amikacin, tobramycin) were tested. Escherichia coli ATCC 25922 and Ps. aeruginosa ATCC 27853 were used as quality control strains.

Pulsed-field gel electrophoresis

PFGE was performed on 27 Ps. aeruginosa isolates. SpeI-restricted DNA fragments were separated using a Chef Mapper XA apparatus (Bio-Rad Laboratories, Hercules, CA, USA). The electrophoresis conditions were 6·0 V cm−1 at 14°C for 20 h in 1·2% Seakem Gold Agarose in 0·5× TBE buffer. The XbaI-restricted DNA of Salmonella enterica serovar Braenderup H9812 was used as a size marker. Gel images in TIFF format were analysed using InfoQuest FP software (Bio-Rad Laboratories). The cluster analyses of restriction patterns of Ps. aeruginosa isolates were performed using the band-based Dice correlation coefficient with a 1·5% position tolerance, and dendrograms were generated with the unweighted pair group method using arithmetic averages (UPGMA). Isolates were considered genetically related if the Dice correlation coefficient was 80% or greater, according to the criterion of van Mansfeld et al. (2010).

DNA sequencing of quinolone resistance determining region (QRDR) in gyrA and parC genes

Chromosomal DNA from four ciprofloxacin-resistant Ps. aeruginosa isolates was extracted using Chromosomal DNA Extraction Kit (TianGen Biotech, Beijing, China). The QRDR regions of gyrA and parC were amplified using PCR primers (Table 1) and amplification conditions described previously (Mouneimne et al. 1999; Akasaka et al. 2001). The ciprofloxacin-susceptible Ps. aeruginosa strain ATCC27853 served as comparative control. The PCR amplicons were purified and sequenced. The DNA sequences were compared with published DNA sequences in the GenBank database (http://blast.ncbi.nlm.nih.gov/).

Table 1.   Primer sequences used to amplify the QRDR of gyrA, parC and classes 1, 2 and 3 integrons
Primer pairSequence DNA targetProduct sizeReferences
intI-for5′-GTTCGGTCAAGGTTCTG-3′intI 1 gene923 bpZhang et al. (2004)
intI-rev5′-GCCAACTTTCAGCACATG-3′
intII-F5′-CACGGATATGCGACAAAAAGGT-3′intI 2 gene788 bpMazel et al. (2000)
intII-R5′-GTAGCAAACGAGTGACGAAATG-3′
intIII-F5′-ATCTGCCAAACCTGACTG-3′intI 3 gene1041 bpPellegrini et al. (2009)
intIII-R5′-CGAATGCCCCAACAACTC-3′
5′-CS5′-GGCATCCAAGCAGCAAG-3′Conserved region of class 1 integronVariableSeverino and Magalhaes (2002)
3′-CS5′-AAGCAGACTTGACCTGA-3′
gyrA-F5′-AGTCCTATCTCGACTACGCGAT-3′QRDR region of gyrA gene330 bpAkasaka et al. (2001)
gyrA-R5′-AGTCGACGGTTTCCTTTTCCAG-3′
parC-F5′-CTGGATGCCGATTCCAAGCAC-3′QRDR region of parC gene186 bpMouneimne et al. (1999)
parC-R5′-GAAGGACTTGGGATCGTCCGG-3′

Screening and DNA sequence analysis of integrons

The presence of classes 1, 2 and 3 integrase genes (int1, int2, int3) was investigated using previously described PCR primers presented in Table 1. For each set of PCRs, laboratory E. coli strains known to harbour integron-associated genes were used as positive controls (Lu et al. 2010). The amplified PCR products were purified and sequenced. The variable gene cassettes in integron-harbouring strains were also amplified by primers 5′-CS and 3′-CS (Table 1) and then cloned into pEasy-T vector (TaKaRa, Dalian, China) for sequencing. The DNA sequences of amplified integrons and gene cassettes were compared with published sequences in the GenBank database (http://blast.ncbi.nlm.nih.gov/).

Results

Prevalence of Pseudomonas aeruginosa

Overall, 402 clinical specimens from infected dogs collected in two animal hospitals were examined for the occurrence of Ps. aeruginosa, of which 27 (6·7%; 24 from VTH-CAU and three from BSVC) isolates were identified as Ps. aeruginosa. The infection sites where these strains were collected were as follows: 12 isolates collected from skin and mucosa infections, seven from wounds, four from the auditory canal, three from eyes and conjunctiva and one from the urogenital tract.

Antimicrobial susceptibility testing

A total of 27 clinical isolates of Ps. aeruginosa were tested for their susceptibility to 11 antimicrobial agents that are used in human and veterinary medicine. The results of the susceptibility testing are shown in Table 2. Among the β-lactam antimicrobials, three isolates demonstrated resistance to ticarcillin and carbenicillin, and one isolate exhibited resistance to ceftazidime. Imipenem exhibited the greatest anti-Pseudomonas activity with only three isolates being intermediate susceptible to this drug (MIC = 8 μg ml−1). For the FQs, four strains (236, 286, 374 and 391) showed resistance to each of the four FQs examined. Among the AGs, three strains showed resistance to amikacin, and four strains exhibited resistance to gentamicin and tobramycin. Overall, susceptibility patterns varied greatly. Twenty-three strains were susceptible to all tested 11 antimicrobials. The remaining four strains were resistant to multiple antimicrobials tested, ranging from 6 to 10 antimicrobials.

Table 2.   Antimicrobial resistance phenotypes of Pseudomonas aeruginosa isolated from dogs (= 27)
Class and/or antimicrobialMIC range (μg ml−1)Resistant breakpoint (μg ml−1)Resistant strains (n)Intermediate strains (n)Susceptible strains (n)Resistant strains (%)MIC50 (μg ml−1)MIC90 (μg ml−1)
β-Lactams
 Ceftazidime1–128≥3212243·728
 Ticarcillin8–32≥128302411·11664
 Carbenicillin8–1024≥512322211·164256
 Imipenem0·125–16≥160324044
Fluoroquinolones
 Norfloxacin1–128≥16412214·8<1128
 Levofloxacin1–128≥8422114·8<1128
 Ciprofloxacin0·5–64≥4432014·8<0·2532
 Lomefloxacin1–128≥8432014·8<1128
Aminoglycosides
 Gentamicin1–128≥16412214·84>128
 Amikacin4–512≥64302411·1<48
 Tobramycin1–128≥16412214·8<116

PFGE

To assess whether the infected dogs carrying Ps. aeruginosa strains were caused by the spread of a specific clone, the 27 isolates were analysed by PFGE profile. Twenty-five strains were typeable, resulting in the detection of 14 PFGE clusters (denoted as clusters A–N, Fig. 1). Two strains derived from skin and mucosa infections from BSVC were untypeable. Interestingly, strains derived from same infection sites usually belong to the same PFGE clusters, even though they were from different dogs. Five strains were assigned to cluster G (subtypes G1–G5), three of which derived from wounds; both cluster A and M contained three strains, two strains from each cluster derived from infection sites of skin and mucosa and eye and conjunctive, respectively; cluster C, I and K contained two strains, respectively, in which the strains from cluster C and K were isolated from same infection site. The other clusters only contained a single isolate.

Figure 1.

 PFGE of SpeI-digested chromosomal DNA of Pseudomonas aeruginosa. Isolate Y2 was originated from BSVC, while the other 24 were collected from VTH-CAU.

QRDR mutations

The four FQ-resistant Ps. aeruginosa isolates were selected for sequence analysis of QRDR mutations. All of the strains had two QRDR mutations; one encoded the exchange of Thr at position 83 to Ile in GyrA and the other encoded the exchange of Ser at position 87 to Leu in ParC. No other mutations in QRDR of gyrA and parC were detected in these strains.

Characterization of Integrons

PCR-based examination for the presence of the classes 1, 2 and 3 integrons was carried out on the entire collection. Of the 27 isolates, four strains (236, 286, 374 and 391) carried detectable class 1 integrons, and all of these strains were resistant to 6–10 tested antimicrobials, while classes 2 and 3 integrons were not detected. The class 1 integrons with variable gene cassettes in these four strains were also amplified and cloned for sequencing. Five different variable regions were found in these four strains (Table 3), strain 236 carried two variable regions, one 1816-bp region harbouring gene cassettes aadA4 and aadA1, which confer resistance to streptomycin (Kadlec and Schwarz 2008); another 2025-bp region harbouring gene cassettes blaOXA-31, which confers resistance to 2-amino-5-thiazolyl cephalosporins (cefpirome, cefepime and cefclidine) but not other cephalosporins (ceftazidime, cefotaxime) (Aubert et al. 2001), and aadA2, which confers resistance to streptomycin. Strain 286 also carried two variable regions, one 2344-bp region harbouring gene aadA1, arr-3 and catB3, as well as a 2570-bp region harbouring cmlA5, cmlA and aadA1. Genes catB3, cmlA5 and cmlA confer chloramphenicol resistance, while arr-3 confers rifampin resistance. Both strains 374 and 391 carried one 1822-bp variable region harbouring the gene cassette aac4 and aadA1, which were nearly identical to the genes from the variable region in strain 236.

Table 3.   Antimicrobial resistance phenotype, gene cassette arrays and PFGE clusters of integron-positive Pseudomonas aeruginosa isolates of canine origin
StrainsResistance phenotypeGene cassettes: amplicon size (bp)Gene cassette arrayPFGE clusters
  1. CFZ, ceftazidime; TIC, ticarcillin; CAB, carbenicillin; NOF, norfloxacin; LOM, lomefloxacin; CIP, ciprofloxacin; LEV, levofloxacin; GEN, gentamycin; AMI, amikacin; TOR, tobramycin.

236CFZ, TIC, CAB, NOF, LOM, CIP, LEV, GEN, AMI, TOR1816aacA4-aadA1M2
2025blaOXA-31-aadA2
286NOF, LOM, CIP, LEV, GEN, TOR2344aadA1-arr-3-catB3C1
2570cmlA5-cmlA-aadA1
374CFZ, TIC, CAB, NOF, LOM, CIP, LEV, GEN, AMI, TOR1822aacA4-aadA1M1-1
391CFZ, TIC, CAB, NOF, LOM, CIP, LEV, GEN, AMI, TOR1822aacA4-aadA1M1-2

Discussion

The rate of Ps. aeruginosa (6·7%, 27/402) isolation among infected dogs was slightly higher than that (5·75%) reported in the clinics of the veterinary faculty at the University of Zagreb, Croatia, in which 183 Ps. aeruginosa strains were isolated from 3184 pathological specimens originating from dogs (Seol et al. 2002). In the USA, Ps. aeruginosa was isolated from 7·5 to 27·8% of canine skin and ear samples, respectively, during the years 1992 through 1997 (Petersen et al. 2002), while in this study, Ps. aeruginosa strains were isolated from 12·2% (10/82) of canine skin and 11·1% (4/36) of ear samples, respectively. The PFGE analysis revealed a considerable heterogeneity among these Ps. aeruginosa isolates (Fig. 1), which suggested that these Ps. aeruginosa isolates were most likely acquired from different sources rather than originating from a single source and being disseminated among the canine population.

Pseudomonas aeruginosa is an opportunistic bacterial pathogen and is well known for its intrinsic and acquired resistance and the ability to cause serious infections in animals. In this study, 23 of 27 strains showed susceptibility to all 11 anti-pseudomonal drugs tested, including FQs, AGs, carboxypenicillins and third-generation cephalosporins. The resistance rate to the anti-pseudomonal drugs tested was lower in the isolates of infected canine origin in this study than in previous studies of human clinical isolates (Severino and Magalhaes 2002; Yoo et al. 2008; Chen et al. 2009), which may be due to lower levels of selective pressure, because 85·2% of the tested strains (23/27) were derived from dogs with no history of antibiotic usage. Moreover, these susceptibility results are important for veterinarians to know that many of the available anti-pseudomonal drugs are still efficacious for use in the treatment of canine infections caused by this Ps. aeruginosa in the Beijing area.

In this study, high-level MIC values to FQs were observed and hot spot mutations in both gyrA and parC were detected among the four FQ-resistant Ps. aeruginosa strains. These results were consistent with those reported by Rubin et al. (2008), which suggested a strong association between the mutations in gyrA/parC in Ps. aeruginosa strains and the observed high level of resistance to all FQs tested. Interestingly, the four FQ-resistant Ps. aeruginosa strains also showed further multi-drug resistance profiles, which indicated the presence of other antimicrobial resistance mechanisms that may be associated with plasmids and transposons, changes of outer membrane permeability, biofilm formation, and up-regulation of multi-drug efflux pumps, as well as integrons and gene cassette-mediated resistance (Tenover 2006).

Integrons, especially class 1 related integrons, have been commonly found in clinical isolates of Ps. aeruginosa (Severino and Magalhaes 2002; Fonseca et al. 2005; Wu et al. 2008; Sanchez-Martinez et al. 2010). For instance, 63·5% (n = 54) of the Ps. aeruginosa isolates derived from human clinical specimens from the intensive care unit of a Brazilian hospital carried class 1 integrons (Severino and Magalhaes 2002), and 38% (n = 27) of isolates collected from the Affiliated People’s Hospital of Jiangsu University, China contained class 1 integrons (Chen et al. 2009). The relative lack of integrons in Ps. aeruginosa isolated from canine infection has also been previously reported, in which only two of 106 isolates were found to carry the class 1 integrons, both of which contained the aadA gene (Rubin et al. 2008). In our study, only four Ps. aeruginosa isolates harbouring class 1 integrons (14·8%) were detected among 27 strains of dog origin. Interestingly, three of these four integron-carrying strains belonged to PFGE cluster M, potentially indicating that these related Ps. aeruginosa might more easily harbour integrons than other genotypes (Table 3). Integrons are known to be associated with multi-drug resistance, especially class 1 integrons, which are widely distributed among Gram-negative bacteria; our study also showed that these four integron-harbouring strains were multi-drug resistant (Table 3). Four different types of gene cassettes in these integrons were identified in this study. All of these cassettes contained the AG resistance gene aadA1, similar to the previous results of Rubin et al. (2008). These AG resistance genes were also commonly identified in integrons in Ps. aeruginosa isolated from human infections in China (Gu et al. 2007; Chen et al. 2009). Interestingly, one of the class 1 integrons in this study contained blaOXA-31 and aadA2 genes and displayed >99% similarity to the sequence of the class 1 integron, which carried the same blaOXA-31 and aadA2 genes, isolated from a Ps. aeruginosa clinical isolate collected from a 1-month-old child in France (Aubert et al. 2001). To the best of our knowledge, the oxacillin resistance gene blaOXA-31 has never been reported in Ps. aeruginosa isolated from dogs. The high similarity between these gene cassettes (blaOXA-31-aadA2) suggested this blaOXA-31 gene in Ps. aeruginosa of canine origin might have originated from humans. The OXA-31 β-lactamase encoded by blaOXA-31 conferred resistance to 2-amino-5-thiazolyl cephalosporins such as cefepime, but not to ceftazidime and cefotaxime in Ps. aeruginosa strain SOF-1 (Aubert et al. 2001).

In summary, the present study characterized the different levels of antimicrobial resistance of Ps. aeruginosa isolated from canine infections in two animal hospitals in Beijing, China. The findings stress the need for continued monitoring of antimicrobial resistance among animal bacterial pathogens and the value of laboratory antimicrobial susceptibility testing as the basis for clinical treatment decisions.

Acknowledgements

This study was supported by the programme for Chang Jiang Scholars and the Innovative Research Team at the University of China (no. IRT0866), and the grant from the National Natural Science Foundation of China (no. U0631006). Dr Jing Han is supported through the Oak Ridge Institute for Science and Education.

Transparency declarations

The views presented in this manuscript do not necessarily reflect those of the US Food and Drug Administration.

Ancillary