Abstract
This study aimed to determine the prevalence of extended-spectrum beta-lactamases (ESBLs) in Enterobacteriaceae and to characterize the genetic composition of ESBL determinants among Enterobacteriaceae isolates from healthy people in Guangzhou, China. A total of 200 rectal swab samples were collected from healthy asymptomatic individuals and tested for ESBL production using ChromID ESBL agar. Phenotypic ESBL producers were screened for blaCTX-M, blaTEM, and blaSHV genes using PCR and DNA sequencing. The prevalence of ESBL-producing Enterobacteriaceae among rectal swab samples was 69.5%. All ESBL-producing isolates harbored blaCTX-M genes (n=138) except for one isolate that harbored blaSHV-2a. Eleven CTX-M ESBL genes were detected. The most predominant CTX-M-type genes were blaCTX-M-14 (n=82), followed by blaCTX-M-55 (n=19), blaCTX-M-65 (n=10), and blaCTX-M-27 (n=9). Isolates carrying blaCTX-M-38,-3,-15,-14b,-98,-121 and -123 were also identified. Molecular homology analysis of the selected isolates was performed by phylogenetic grouping and multilocus sequence typing and indicated that the predominant clone belonged to A-CC10. This study showed a high rate of CTX-M-type ESBL genes among Enterobacteriaceae isolates from healthy individuals in China.
Introduction
I
CTX-M-type ESBLs are the most common ESBL type identified in different Enterobacteriaceae isolates in many countries.3,36 To date, over 160 CTX-M β-lactamases have been identified and classified into five main subgroups, including CTX-M-1, -2, -8, -9, and -25 groups. Among the CTX-M family, CTX-M-14 has been reported as the dominant CTX-M ESBL in human and animal isolates in China.18,28,37 Moreover, CTX-M-55, CTX-M-65, and CTX-M-27 have also been frequently found in farm animals in southern China, suggesting a potential risk of transmission of ESBL-carrying strains from food to humans in this area.28,35,37 However, the prevalence of ESBL-producing Enterobacteriaceae in healthy people in southern China is unclear. We therefore investigated the prevalence of ESBL-producing Enterobacteriaceae and characterized the genetic composition of the identified ESBL enzymes among Enterobacteriaceae isolates in healthy individuals in Guangzhou, China.
Materials and Methods
Sampling
During the months of May and July 2012, rectal swabs were collected from 200 healthy asymptomatic individuals (aged >18 years) who visited the Guangzhou Center for Disease Control and Prevention (CDC) for general examinations. In China, people who want to engage in food-related jobs must be physically examined to receive a health certificate that allows them to be employed in the food industry. The examinations are held at the CDC or authorized hospitals. Such medical examinations primarily focus on certain infectious diseases (e.g., tuberculosis and hepatitis B). Rectal swabs are tested for the intestinal carriage of Salmonella typhi or Salmonella paratyphi. All 200 subjects received health certificates after their health check. Two hundred rectal swab samples, one specimen from each subject, were tested in this study. The study was approved by the Medical Ethics Committee of Guangzhou CDC, China.
Bacterial isolates, ESBL screening, and antimicrobial susceptibility testing
Rectal swabs (Multifunction Sampling System; Mabsky Biotech, Shenzhen, China) were inoculated onto ESBL chromogenic agar (ChromID ESBL agar; bioMérieux, Marcy l'Etoile, France) for 18 hr at 37°C. The ESBL medium is a selective chromogenic agar medium that is supplemented with cefpodoxime. This enables the detection and presumptive identification of ESBL-producing Enterobacteriaceae directly from clinical specimens. Enterobacteriaceae isolates grown on the chromogenic ESBL medium produce expected colony colors (blue-green with or without a purple halo for Klebsiella spp., Enterobacter spp., Serratia marcescens, and Citrobacter spp., pink-burgundy for Escherichia coli, and beige to brown with a diffusible orange halo for the PMP group). Moreover, cefpodoxime has been shown to be the best general substrate to identify all types of ESBLs presently found in clinical specimens. 10 Identification of the isolates was confirmed with the Vitek-2 Compact System.
Antibiotic susceptibility testing was performed using the Vitek-2 Compact System with AST-GN31 cards. A panel of 18 antimicrobial agents was tested, including piperacillin, cefotaxime, ticarcillin/clavulanate, ceftriaxone, piperacillin/tazobactam, ceftazidime, cefepime, aztreonam, imipenem, meropenem, ertapenem, ciprofloxacin, amikacin, levofloxacin, gentamicin, tobramycin, tigecycline, and trimethoprim/sulfamethoxazole. E. coli ATCC 25922 and Klebsiella pneumoniae ATCC 700603 were used as control strains. The results were interpreted according to the criteria of the Clinical and Laboratory Standards Institute (CLSI). Presumptive ESBL producers were confirmed on Mueller–Hinton agar plates using the double-disc synergy test according to CLSI protocols.
PCR amplification and sequencing
All isolates were screened for the presence of blaCTX-M, blaTEM, and blaSHV genes as described previously.18,34 To identify group-specific blaCTX-M genes, including the blaCTX-M-1, blaCTX-M-2, blaCTX-M-8, and blaCTX-M-9 group genes, a multiplex PCR was performed as described previously. 2 After blaCTX-M group identification, specific primers were used to amplify all genes of each group. Purified PCR products were sequenced bidirectionally or were cloned in pGEM-T vectors and then sequenced. Sequence alignment and analysis were performed on the NCBI website using the BLAST program. The GenBank accession numbers for the CTX-M-38, CTX-M-98, CTX-M-121, and CTX-M-123 sequences are KF308562, KF022000, KF021999, and KF308563, respectively.
Phylogenetic group analysis
The E. coli isolates were assigned to the phylogenetic groups A, B1, B2, or D using an improved multiplex PCR method with specific primers for gadA, chuA, yjaA, and TspE4.C2 determinants. 7
Multilocus sequence typing
All blaCTX-M-positive E. coli isolates within group B2 were subjected to PCR screening methods for ST131, and those that were positive were further confirmed by multilocus sequence typing (MLST). 4 In addition, 40 E. coli isolates harboring the most frequently identified blaCTX-M genes, blaCTX-M-14, blaCTX-M-55, and blaCTX-M-65, which belong to A, B1, and D phylogenetic groups, and blaCTX-M-15, which is distributed worldwide, were selected to be analyzed by MLST. Gene amplification and sequencing were performed using specific primers for the seven standard housekeeping genes (adk, fumC, gyrB, icd, mdh, purA, and recA) and analyzed at http://mlst.warwick.ac.uk/mlst/dbs/Ecoli. Clonal complexes (CCs) were defined as a group of multilocus genotypes in which every genotype shares at least five loci in common with at least one other member of the group.
Results
Bacteria isolated from healthy humans
Of the 200 rectal swabs tested, 153 (76.5%) showed bacterial growth on ESBL agar. Of these 153 isolates, 139 (69.5%) were ESBL-producing organisms as determined by the double-disc synergy test. E. coli was the predominant ESBL-producing organism and was recovered from 117 (58.5%) subjects, followed by K. pneumoniae (19/200, 9.5%), Klebsiella oxytoca (1/200, 0.5%), Enterobacter aerogenes (1/200, 0.5%), and Citrobacter freundii (1/200, 0.5%).
Antibiotic susceptibility
The β-lactam susceptibility profiles and the associated antimicrobial resistance of the ESBL-producing organisms are shown in Table 1. All ESBL-producing organisms were resistant to piperacillin, cefotaxime, and ceftriaxone, but remained susceptible to carbapenem and tigecycline. Fewer than 16% and 7% of the ESBL-producing isolates were resistant to ceftazidime and cefepime, respectively. For all ESBL-producing isolates, susceptibility to ticarcillin/clavulanate and piperacillin/tazobactam was 87.8% and 97.9%, respectively. Ninety-eight (70.5%) ESBL-producing isolates were susceptible to the quinolones (ciprofloxacin and levofloxacin). In addition, 60.4% of the ESBL-producing isolates were susceptible to gentamicin and tobramycin, and 97.9% of the ESBL-producing isolates were susceptible to amikacin.
Imipenem, meropenem, and ertapenem.
Genotypes of ESBL-producing isolates
Molecular analysis of β-lactamase genes showed that all ESBL-producing strains harbored CTX-M-type genes except one K. pneumoniae isolate that harbored blaSHV-2a only (Table 2). CTX-M-9 group ESBL genes were positive in 108 isolates, and sequencing results revealed that blaCTX-M-14, blaCTX-M-14b, blaCTX-M-65, blaCTX-M-27, blaCTX-M-38, blaCTX-M-98, and blaCTX-M-121 were positive in 82, 1, 10, 9, 5, 1, and 1 isolates, respectively; CTX-M-1 group ESBL genes were positive in 29 isolates, and blaCTX-M-55, blaCTX-M-15, and blaCTX-M-3 were positive in 19, 5, and 5 isolates, respectively. One E. coli isolate carried both blaCTX-M-14 and blaCTX-M-27; moreover, two K. pneumoniae isolates harbored blaCTX-M-14+blaSHV-27 or blaCTX-M-38+blaSHV-27. One E. coli isolate carried a CTX-M-1/9 group hybrid gene, blaCTX-M-123, which has recently been identified in animals in Guangzhou, China. 12 Furthermore, blaTEM-1 and blaTEM-1b narrow-spectrum β-lactamase genes were found in 44 and 2 isolates, respectively. Fifteen isolates harbored SHV-type narrow-spectrum β-lactamase genes: blaSHV-1 (n=2), blaSHV-11 (n=9), blaSHV-33 (n=2), and blaSHV-108 (n=2). Fifty-one CTX-M ESBL genes were found in combination with the following β-lactamase genes: blaCTX-M+blaTEM (n=36), blaCTX-M+blaSHV (n=5), and blaCTX-M+blaTEM+blaSHV (n=10). CTX-M-2 group and CTX-M-8 group ESBL genes were not detected in these isolates.
ESBL, extended-spectrum beta-lactamase.
Phylogenetic group analysis and MLST
Most ESBL-producing E. coli isolates belonged to the phylogenetic group A (38/117, 34.2%) and D (32/117, 36.8%), followed by B1 (25/117, 23.9%) and B2 (19/117, 5.1%). Furthermore, blaCTX-M-14-harboring E. coli isolates were most prevalent among phylogenetic groups A, B2, and D, comprising 52.5%, 89.5%, and 72.8%, respectively. In contrast, the phylogenetic group B1 more commonly included the blaCTX-M-55-harboring isolates (Table 3).
Four blaCTX-M-14-harboring E. coli isolates within group B2 were identified as ST131 by the rapid PCR screening method. MLST determination was carried out in the 4 initially identified ST131 isolates and the 40 selected isolates. The MLST results showed heterogeneous patterns in these blaCTX-M-positive E. coli isolates. Overall, the 44 isolates were divided into 35 different sequence types (STs), including 9 CCs and 18 singletons. The frequent ST/CCs were CC10 (8/44, 18.2%), ST131 (4/44, 9.1%), CC155 (3/44, 6.8%), CC405 (2/44, 4.5%), CC38 (2/44, 4.5%), ST602 (2/44, 4.5%), ST165 (2/44, 4.5%), and ST4187 (2/44, 4.5%). The remaining isolates were each a single ST, including two novel STs, ST4470 and ST4471 (Table 4).
CC not defined in the MLST database.
CC, clonal complex; MLST, multilocus sequence typing; ST, sequence type.
Discussion
The prevalence of intestinal carriage of ESBL-producing Enterobacteriaceae was 69.5% (139/200). This value was similar to that reported in Thailand (69.3%), 20 but higher than that reported in China (50.5%), 17 Switzerland (5.8%), 9 Sweden (3.0%), 27 Korea (20.3%), 15 and Saudi Arabia (13.1%). 14 Woerther et al. reported that the high population density may be linked to ESBL-producing Enterobacteriaceae dissemination and estimated that the southeast Asia and western Pacific regions had the highest number of ESBL carriers. 32 Considering that Guangzhou is the third largest Chinese city and that the city's administrative area had a population of 12.78 million in 2010, it is not surprising that the ESBL carriage rate is high.
We did not collect detailed demographic data of the subjects; therefore, it is difficult to assess the risk factors for the high incidence of ESBL-producing Enterobacteriaceae. A previous study has reported that antibiotic treatment for 3 months and a history of hospitalization in the previous year were two risk factors for ESBL carriage in Thailand, where antibiotics can be purchased without prescription. 20 Very similar results have been found in China, where people prefer to buy over-the-counter antibiotics. 29 Furthermore, doctors in China have significantly over-prescribed antibiotics. The frequency and proportion of antibiotics prescribed in China are higher compared with developed countries. The number of antibiotics per 100 prescriptions was 54.62, and personal use of cephalosporins and fluoroquinolones is common. This may have contributed to the high incidence of ESBL-producing Enterobacteriaceae in the community. 33 Some studies have also reported the high prevalence of ESBL-producing E. coli both in waste water and environmental water in China, suggesting that water pollution is as an important factor in ESBL-producing E. coli dissemination.13,32
All ESBL-producing strains were resistant to cefotaxime and ceftriaxone; however, fewer than 16% of the ESBL-producing strains were resistant to ceftazidime. Indeed, similar results have been previously found in hospitalized patients and healthy subjects in China, which revealed that the production of cefotaximase was an important resistance mechanism in ESBL-producing isolates in China.17,34
Molecular characterization of the 139 ESBL-positive isolates showed that 138 isolates carried CTX-M-type ESBL genes, 3 K. pneumoniae isolates harbored SHV-type ESBL genes and none of the isolates carried TEM-type ESBL genes. The blaCTX-M genes were identified in various Enterobacteriaceae, including E. coli, K. pneumoniae, K. oxytoca, E. aerogenes, and C. freundii; however, a significant majority of ESBLs were present in E. coli isolates.
Sequence analysis revealed that blaCTX-M-14 was the predominant CTX-M-type ESBL gene (82/138, 59.4%), and similar results have been previously found in clinical patients, companion animals, and food-producing animals in China.28,34–37 In addition to blaCTX-M-14, blaCTX-M-55 (19/138, 13.8%), and blaCTX-M-65 (10/138, 7.2%) were also frequently identified. This finding is similar to a recent study of food-producing animals in China. 28 Recently, blaCTX-M-27 has been reported in farm animals and pets in China, but has rarely been identified in humans. 28 In our study, blaCTX-M-27 accounted for 5.8% of CTX-M-type ESBL genes and reflects the increasing prevalence of CTX-M-27 ESBL in human isolates in China. Of note, blaCTX-M-98, blaCTX-M-121, and blaCTX-M-123 were first identified from food animals in Guangzhou 37 and, recently, two studies reported the detection of blaCTX-M-123 in human isolates in China.13,33 Two reports have commented on the transmission of ESBL genes from poultry to humans in the Netherlands.16,23 Stokes et al. also reported that pCT-like plasmids were important vectors for the horizontal dissemination of CTX-M-14 genes among E. coli isolates from humans, cattle, and turkeys in the United Kingdom. 26 The rising prevalence of ESBL genes mentioned above, both in humans and in food-producing animals, suggests the transmission of ESBL genes from animals to humans through the food chain.10,13 Furthermore, household contact between infected patients and pets has been shown to increase the risk of dissemination of ESBL-producing organisms to healthy people.22,31
Among the phylogenetic groups, groups A and B1 accounted for 58.1% and are described as the most prevalent among animal or human commensal E. coli isolates, suggesting that animals are important reservoirs for some of the ESBL-producing E. coli isolates in this locale. We also observed CC10, belonging to group A, was the most common E. coli clone. Two previous studies in China have found that CC10 was frequently identified among ESBL-producing or non-ESBL-producing E. coli isolates from food animals.6,11 These studies, along with this study, indicate a correlation between ESBL-producing E. coli isolates in food animals and humans. Only four ESBL-producers were identified as the B2-ST131 clone harboring blaCTX-M-14, which is different from many studies in which the CTX-M-15 carrier predominated among B2-ST131. 24
In summary, we report a surprisingly high prevalence of intestinal carriage of ESBL-producing Enterobacteriaceae in healthy persons. A diverse mix of factors might contribute to this finding, such as over-the-counter antibiotics and colonization of ESBL-producing Enterobacteriaceae through food and contact with pets. Further detailed investigations should be conducted to determine the reasons of the high incidence of CTX-M ESBL-producing Enterobacteriaceae in healthy people in this area, including questionnaires for demographic information, epidemiological data, and the genotyping of ESBL strains and the blaESBL-carrying plasmids in human and food isolates.
Footnotes
Acknowledgments
This research was funded by the Project for Key Medicine Discipline Construction of Guangzhou Municipality (grant number 2013–2015-07), the Sci-Tech Research Project of Guangzhou Municipality, China (grant number: 2011J4300061), and the Medical Sci-Tech Research Projects of Guangzhou Municipality, China (grant numbers: 20121A021019 and 20121A011112).
Disclosure Statement
No competing financial interests exist.
