Abstract
qnrD, unlike other qnr genes, is mainly located on small nonconjugative plasmids. We investigated the presence of qnrD among 1,373 Enterobacteriaceae isolates in China. Twelve qnrD-positive strains were detected, and all were nonsusceptible to fluoroquinolones. The complete sequence of plasmids showed that the qnrD determinants were located on two plasmids with a respective size of ∼4.2 and 2.7 k-bp. Interestingly, the identification of qnrD in this study revealed the highest prevalence of Proteeae among Enterobacteriaceae identified.
Introduction
Materials and Methods
Sampling and cultivation of bacteria
A total of 1,373 Enterobacteriaceae isolates from different host species (humans, n=412; companion animals, n=185; birds, n=472; pigs, n=304) were collected in South China during 2009 and 2010. The isolates were obtained from both nonclinical and clinical specimens, and each isolate was from a separate human or animal (feces and urine samples, n=597; liver, fallopian tube, and blood samples, n=451; vomitus and sneeze samples, n=185; pork and chicken meat samples, n=140). In total, 140 isolates were from nonclinical (healthy) subjects and 1233 from diseased specimens. All isolates were grown on tryptic soy agar (TSA) plates and stored at −80°C in a Luria-Bertani broth containing 30% glycerol.
Polymerase chain reaction amplification
The presence of qnrD was determined using the published polymerase chain reaction (PCR) method. 2 All qnrD-positive strains were also screened for other PMQR determinants, including qnrA, 29 qnrB, 29 qnrS, 29 aac(6′)-Ib, 29 and qepA, 29 and the quinolone resistance-determining regions of the gyrA and parC genes in qnrD-positive strains were sequenced to determine the presence of mutations as previously described.27,29
Antimicrobial susceptibility testing
The minimum inhibitory concentrations (MICs) of ciprofloxacin, ofloxacin, and nalidixic acid were determined by the agar dilution method. Antimicrobial susceptibility testing was conducted by the agar dilution method, according to the guidelines of the CLSI. 26
Assay of qnrD transfer
A conjugation experiment was carried out in mixed-broth cultures as previously described, and rifampin-resistant Escherichia coli C600 (lacZ− Nalr Rifr) was used as the recipient strain. 19 Transconjugants were selected on TSA plates containing 600 mg/L rifampin plus 0.06 mg/L ciprofloxacin.
For the transformation experiment, plasmids of qnrD-positive strains were extracted by the QIAGEN Plasmids Midi Kit, and then were transferred into E. coli TOP10 cells by electroporation. Transformants were selected on SOB agar plates containing 0.06 mg/L ciprofloxacin.
Plasmid analysis
The complete sequences of the plasmids harboring the qnrD gene were obtained by inverted PCR with the primer sets previously described, 11 namely, INVDF/R: 5-TATTCCCCGTAAATTGATCTCG-3 and INVDR/F: 5-CAGGCGCTTCAGCTTGTT-3, and then was sequenced by primer walking. PCR was conducted by the following thermal profile: one cycle at 95°C for 2 min, 30 cycles of 94°C for 30 sec, 53°C for 30 sec 72°C for 3 min with a final elongation of 72°C for 10 min, and 4°C to close the reaction. 11
Bioinformatic analysis
DNA sequence alignments were processed using the ClustalX programs. 23 Phylogeny of plasmids harboring the qnrD gene was constructed using MEGA4 software. 21
Nucleotide sequence accession numbers
The complete sequences of the plasmids harboring the qnrD gene were deposited in the GenBank database, and their accession numbers are JQ776501, JQ776502, JQ776503, JQ776504, JQ776505, JQ776506, JQ776507, JQ776508, JQ776509, JQ7765010, JX982605, and JX982606.
Molecular typing
Pulsed-field gel electrophoresis (PFGE) analysis of XbaI-digested (E. coli) or SfiI-digested (P. mirabilis) genomic DNA of all qnrD-positive strains was performed, using a CHEF Mapper system (Bio-Rad Laboratories, Hercules, CA). 22
Results and Discussion
PMQR has become a significant concern for clinical treatments, as these determinants reduce bacterial susceptibility to fluoroquinolone antibiotics and promote the emergence of resistant mutants. The PMQR determinants have been identified in a number of clinically important pathogens and environmental bacteria in different geographic regions of the world.10,17,18,20,24 qnrD, as a new qnr determinant, is increasingly reported. However, a large-scale investigation of qnrD distribution among different host species has not been reported. In this study, we screened qnrD-positive strains from 1,373 Enterobacteriaceae isolates from birds, swine, retail meat products, companion animals, and humans and characterized the qnrD-carrying plasmids. Our results identified 12 qnrD-positive strains (0.87%) among the 1,373 isolates (Table 1), with nalidixic acid MICs ranging from 8 to 512 (mg/L) and ciprofloxacin MICs from 0.25 to 2 (mg/L) (Table 2). The 12 strains positive for qnrD included five P. mirabilis, four E. coli, one Klebsiella pneumoniae, one Citrobacter freundii, and one M. morganii. PFGE analysis of four E. coli and five P. mirabilis revealed that these strains were divergent and not clone-related (Fig. 1), suggesting that the dissemination of qnrD was not due to clonal dissemination of qnrD-positive strains. These findings provide new data about PMQR distribution and improved understanding of prevalence and dissemination of qnrD.

PFGE-fingerprinting patterns of total DNA preparations. Lanes: M, PFGE marker; 1, Proteus mirabilis CGP180; 2, P. mirabilis CGP248; 3, P. mirabilis CGS49; 4, P. mirabilis CGH15; 5, P. mirabilis CGH40; 6, Escherichia coli CGP246; 7, E. coli CGP169; 8, E. coli CGB40; 9, E. coli CGS13; PFGE,pulsed-field gel electrophoresis.
♦, No qnrD-positive isolates were detected.
Number of the qnrD-positive isolates.
Total number of isolates.
PMQR, plasmid-mediated quinolone resistance; QRDR, quinolone resistance-determining regions
Nal, nalidixic acid; CIP, ciprofloxacin; OFL, ofloxacin; MIC, minimum inhibitory concentration.
Unlike other qnr determinants that are usually located on large and variable plasmids, the qnrD-carrying plasmids typically have a simple backbone and are stably maintained. As observed in previous studies, the qnrD gene was found on plasmids of 2.73,11 or 4.2 kbp2, respectively. Of the 12 qnrD-harboring plasmids identified in this study, the 2.7- and 4.2-kbp plasmids account for 41.7% and 58.3% of all plasmids, respectively. No relationship was seen between plasmid sizes and bacterial species (Table 1). The complete sequences of the qnrD-positive plasmids matched with p20070572 (4,270 bp) and pT8011 (2,687 bp) (accessions numbers FJ228229 and JN183060). The results showed that the qnrD genes in the 12 plasmids were identical. Among the seven 4.2k-bp plasmids harboring the qnrD gene, pCGP246, pCGP169, pCGP180, pCGH25, and pCGH40 exhibited 100% identity with p2007057, whereas pCGB40 and pCGF41 exhibited 99% identity with p2007057. Of the five qnrD-positive 2.7-kbp plasmids, pCGS13 exhibited 100% identity with pT80, while pCGP248, pCGS49, pCGH15, and pCGH69 were identical and exhibited 99% identity with pT80.
Although only two qnrD-positive-harboring plasmids (4.2 and 2.7 kbp) have been reported, the qnrD plasmids were detected in bacterial strains isolated from a variety of sources, including human-, animal-, and environmental origin,3,11,13,30,31 in which other qnr plasmids were also identified. Interestingly, qnrD plasmids seem to be more common in Proteeae. Zhao and Dang investigated the prevalence of qnr determinants in coastal seawater and found that qnrD was the predominantly prevalent qnr determinants (27%), which were all carried by P. vulgaris. 31 The present study revealed that six (50%) qnrD-positive strains were Proteeae, further supporting the notion that Proteeae are an essential source or carrier in the spreading of qnrD-harboring plasmids. In this study, five of the qnrD-positive strains also carried the gene aac(6′)-Ib-cr, and one was also positive for qnrS, while qnrA, qnrB, and qepA were absent in all of them (Table 1). The gyrA and parC mutations were detected in 3 and 1 strains, respectively (Table 1). In the study of Zhao and Dang, however, all 22 Proteus vulgaris isolates carrying qnrD had a single mutation in the gyrA gene, but not other PMQR determinants. 31
For cross-environmental transmission of different plasmids, mobility is considered one of the most important factors.1,9,15,20 Due to lack of their own set of mating pair formation (MPF) genes, the qnrD plasmids were nonconjugative. No transconjugants were obtained despite of several attempts in this study, which is in accordance with previous studies on the qnrD gene using the liquid- or filter-mating assays.3,31 Cavaco et al. have speculated that ORF4 in the 4.2-kbp qnrD plasmid might function as a mob-like gene, 2 and if it obtains other MPF genetic elements present in the same cell, it may gain the mobility. It is surprising that no mob-like genes or oriT sequence were identified in the 2.7-kbp plasmids. 3 Nevertheless, 12 qnrD-positive transformants were obtained on SOB agar plates containing 0.06 mg/L ciprofloxacin in the electroporation experiments conducted in this study. The qnrD-positive transformants showed 4-fold to 8-fold, 64-fold to 128-fold, and 16-fold increases in the MICs of nalidixic acid, ciprofloxacin, and ofloxacin, respectively, when compared with the recipient strain (Table 2). This result suggests that the qnrD-carrying plasmids can be transferred by nonconjugative means and may be spread by natural transformation or transduction. Under certain stressful environments, bacteria are able to establish naturally a physiological state to take-up DNA,12,14 which may be responsible for the spread of the qnrD-carrying plasmids.
Footnotes
Acknowledgments
We thank Dr. Qi-Jing Zhang (Iowa State University) for critical reading of the manuscript and for very helpful suggestions. This work was supported by the grants from the National Science Fund for Distinguished Young Scholars (no.31125026) and from the Special Fund for Agroscientific Research in the Public Interest (no.201203040). This study was also supported by the National Natural Science Foundation of China (no. U0631006 and U1031004).
Disclosure Statement
No competing financial interests exist.
