Enterobacter species may cause severe nosocomial infections, including bloodstream, respiratory tract, and central nervous system infections as well as endocarditis1,2. Nosocomial infections caused by these microorganisms have been associated with high rates of mortality and morbidity1. Enterobacter cloacae and Enterobacter aerogenes are the most common species isolated from clinical samples3. Several virulence genes are involved in the pathogenesis of these microorganisms4–7. Curli fimbria, encoded by csgBAC, is an important factor for cell adhesion, aggregation, and biofilm formation in many enterobacteria4. In addition, RpoS regulation is known to play an important role in multiple stress conditions such as acid, heat, and oxidative stress, starvation, high osmolarity, and near UV exposure5. Another important virulence factor is the type III secretion system encoded by FliI that delivers a variety of effectors directly into the cytosol of host as well as aerobactin, encoded by the iutA, described as a virulence factor related to iron acquisition from host-binding proteins6,7. β-lactam antibiotics, especially third-generation cephalosporins and carbapenems, are used to treat infections cause by several species of Enterobacter1–3. β-lactamase enzymes, including extended-spectrum β-lactamases (ESBLs) and AmpC, are involved in the mechanism underlying resistance to β-lactam antibiotics in Enterobacter spp1–3. ESBLs are often encoded by genes located on large plasmids that also carry genes for resistance to other antimicrobial agents such as aminoglycosides and fluoroquinolones1. ESBLs are capable of hydrolyzing penicillins, broad-spectrum cephalosporins, and aztreonam, but may not hydrolyze cephamycin, and are inhibited by clavulanic acid. AmpC β-lactamases are usually encoded on the bacterial chromosome and in some cases on the bacterial plasmid (plasmid-mediated AmpC)3. In Iran, ESBL production was recently reported in 44.28% of E. cloacae isolates1. Despite the high incidence of Enterobacter spp. infection among Iranian patients, very little is known about the antibiotic resistance pattern, virulence factors, and molecular characteristics of Enterobacterspp. isolates. In the current study, the genes encoding antibiotic resistance enzymes and virulence factors were determined and the genetic relationship between Enterobacter spp. isolated from different clinical samples was evaluated.
Bacterial isolates
A total of 57 isolates of Enterobacter spp. were obtained from different patients admitted to three teaching hospitals of the Tehran University of Medical Sciences between September 2013 and April 2014. The isolates were collected from various clinical samples, including urine, wounds, tracheal aspirate, and blood. No duplicate isolates from the same patient and no environmental strains were included in this study. All isolates of Enterobacter spp. were identified by standard biochemical tests8.
Susceptibility testing
Antibiotic-containing discs (Mast, UK) were used to determine the susceptibility of Enterobacter spp. using the disc diffusion method, as per the Clinical and Laboratory Standards Institute (CLSI) guidelines9. The antimicrobial agents used were as follows: aztreonam-amikacin (30µg), amoxicillin-clavulanic acid (20/10µg), cefpodoxime (10µg), cefotaxime (30µg), ceftazidime (30µg), imipenem (10µg), cefepime (30µg), gatifloxacin (5 mg), cefoxitin (30µg), gentamicin (30µg), ciprofloxacin (30µg), levofloxacin (5µg), ertapenem (10µg), and meropenem (10µg). Isolates that showed resistance to at least three classes of antibiotics were defined as multi-drug resistant (MDR) strains1. ESBL-producing strains were detected using the combined double-disc test1. In addition, organisms were screened for carbapenemase production with the modified Hodge test (MHT)9. The minimum inhibitory concentration (MIC) of imipenem was determined by the microbroth dilution method according to CLSI criteria9. AmpC overproduction was confirmed according to the method of Kalantar-Neystanaki et al10.
Detection of β-lactamases and virulence genes
Genomic DNA was extracted by the boiling method2. The genes encoding ESBLs (bla TEM, bla SHV, bla CTX-M, and bla PER), AmpC (bla ACC, bla FOX, bla MOX, bla DHA, bla CIT, and bla EBC), and carbapenemase (bla IMP, bla VIM, blaNDM, bla KPC, bla GIM, and bla OXA-48) were targeted by polymerase chain reaction (PCR) using specific primers10,11. The detection of seven different virulence genes (csgA, csgB, csgD, rpos, FliI, fepA, and iutA) was performed with PCR using the oligonucleotide primers listed in Table 1.
TABLE 1: The oligonucleotide primers used in this study for the amplification of virulence genes.
| Gene | Primer sequence (5′ to 3′) | Annealing temperature (°C) | Product size (bp) | Reference |
|---|---|---|---|---|
| csgA | F- TTCAAAGTGGCAGTTATTGCAG | 56 | 276 | [4] |
| R- TTTTTGCAGCAGATCGATAGAA | ||||
| csgD | F- GAAATTGCATAATATTCAACGTTC R- TTTGTTCAGGATCTCTTTTTCAC | 54 | 385 | |
| csgB | F- TCCTGGGAAACGATGGACAA | 54 | 193 | this study |
| R- TTACATTACTGGGAGCGCCT | ||||
| fliI | F- ATACGGCGCAGTGCGTTAC | 54 | 154 | this study |
| R- ACCAAAGAGAGGACACAATGC | ||||
| rpoS | F- CACTTCACGCTGTTTGGCG | 56 | 273 | this study |
| R- CGCGAGTTGTCCCATAAACTG | ||||
| fepA | F- TCTTTT TTCACCGGCATGGA | 57 | 572 | this study |
| R- CGTGCGGTGGTCAATATCT | ||||
| iutA | F- TGAAACGTTCTCATCTTTGGGTT | 56 | 1117 | this study |
| R- TCG AAGGTTTCATGGTCGGC |
Random amplified polymorphic DNA-PCR
For molecular analysis of isolates, random amplified polymorphic DNA (RAPD)-PCR was performed as previously described12. In brief, PCR protocol comprised a pre-denaturation step at 95 °C for 5 min, followed by 30 cycles of 60 s at 95 °C, 60 s at 33 °C, and 60 s at 72 °C. A final extension step was performed at 72 °C for 10 min. PCR products were separated by electrophoresis on 1% agarose gels with 0.5× Tris-borate-ethylenediaminetetraacetic acid (EDTA) buffer (TBE buffer). Gels were stained with ethidium bromide and the images were captured using a gel documentation system. Isolates that differed by more than two prominent bands were assigned to different types.
Of 57 isolates, 44 (77.1%) were E. cloacae and 13 (22.8%) were E. aerogenes. These were cultured from wounds (n = 26), urine (n = 15), blood (n = 8), and other sources (n = 8). Resistance to cefoxitin (84.3%), cefotaxime (49.1%), cefpodoxime (36.8%), and ceftazidime (36.8%) was more prevalent, but only eight (14.1%), seven (12.3%), and six (10.5%) isolates were resistant to imipenem, levofloxacin, and gatifloxacin, respectively. Microbroth dilution method showed that 20 (35.1%) strains were resistant to imipenem. Ten (17.5%) isolates were defined as MDR. The phenotypic test for ESBL, AmpC β-lactamase, and carbapenemase production showed that 30 isolates (22 E. cloacae and 8 E. aerogenes) produced ESBL, 21 isolates (16 E. cloacaeand 5 E. aerogenes) produced AmpC β-lactamases, and 8 isolates (6 E. cloacae and 2 E. aerogenes) produced carbapenemases. The phenotypic and genotypic characteristics of ESBL and AmpC-producing isolates of E. cloacae and E. aerogenes are shown in Table 2 and Table 3, respectively. The genes encoding ESBL, bla TEM, blaCTX-M, and bla SHV, were detected in 19 (63.3%), 19 (63.3%), and 8 (26.6%) isolates, respectively, making them the most prevalent ESBL genes in these isolates. We failed to detect bla PER .
TABLE 2: Characteristics of Enterobacter cloacae isolates.
| Isolate ID | Date | Source | Resistance pattern | MDR | ESBL gene | MIC of IMI | AmpC gene | RAPD type |
|---|---|---|---|---|---|---|---|---|
| 1 | 11/11/2013 | Burn | CTX, CAZ, CPM, CPD, FOX, AUG, IMI, MEM, ETP, AK, GM, CIP, LEV, GAT | + | bla TEM, bla CTX-M | 1 | bla EBC | D |
| 2 | 11/11/2013 | Burn | CTX, CAZ, CPM, CPD, FOX, AUG, AK, GM | − | bla TEM, bla CTX-M | 2 | bla EBC | E |
| 3 | 11/25/2013 | Burn | CTX, CAZ, CPM, CPD, FOX, AUG, AK, GM, CIP | + | bla TEM, bla SHV | 2 | bla EBC | F |
| 4 | 12/21/2013 | Eye | CTX, CAZ, CPM, CPD, FOX, AUG, AK, GM | − | bla TEM | 4 | bla ACC, blaEBC | G |
| 5 | 12/29/2013 | Respiratory | CTX, CAZ, CPM, CPD, FOX, AUG, AK | − | bla TEM | 4 | bla ACC, blaDHA | H |
| 6 | 12/28/2011 | Urine | CTX, CAZ, CPM, CPD, FOX, AUG, GM, CIP | + | bla TEM, bla CTX-M, bla SHV | 0.25 | bla DHA, blaEBC | B |
| 7 | 12/28/2011 | Wound | CTX, FOX, AUG | − | bla TEM, bla CTX-M | 2 | bla ACC, blaEBC | C |
| 8 | 1/12/2011 | Wound | CTX, FOX, AUG | − | bla TEM | 2 | bla EBC | I |
| 9 | 6/1/2012 | Wound | CTX, CAZ, CPM, CPD, FOX, AUG, GM | − | bla TEM, bla CTX-M | 2 | bla EBC | C |
| 10 | 12/13/2013 | Wound | CTX, CAZ, CPM, CPD, FOX, AUG, AK, GM, CIP, LEV | + | bla TEM, bla CTX-M | 4 | bla EBC, blaCIT | J |
| 11 | 12/13/2013 | Urine | CTX, FOX, AUG | − | bla TEM, bla CTX-M | 4 | − | K |
| 12 | 2/25/2014 | Urine | CTX, CAZ, CPM, CPD, FOX, AUG, AK, GM | − | bla TEM, bla CTX-M | 1 | bla ACC, blaEBC | L |
| 13 | 3/11/2014 | Burn | CTX, CAZ, CPM, CPD, FOX, AUG, IMI, MEM, ETP, AK, GM, CIP, LEV, GAT | + | bla TEM, bla CTX-M | 64 | bla EBC, blaCIT | A |
| 14 | 3/11/2014 | Burn | CTX, CAZ, CPM, CPD, FOX, AUG, IMI, MEM, ETP, AK, GM, CIP, LEV, GAT | + | bla TEM, bla CTX-M | 64 | bla EBC | A |
| 15 | 4/22/2014 | Urine | CTX, FOX, AUG | − | bla CTX-M | 4 | bla EBC | M |
| 16 | 4/25/2014 | Respiratory | CTX, CAZ, CPM, CPD, AUG, GM | − | − | 4 | bla EBC | B |
| 17 | 4/29/2014 | Respiratory | CTX, CAZ, CPM, CPD, FOX, AUG, AK, GM | − | bla TEM, bla CTX-M | 4 | bla EBC | N |
| 18 | 5/5/2014 | Blood | CTX, CAZ, CPM, CPD, FOX, AUG, IMI, MEM, ETP, AK, GM, CIP, GAT | + | − | 16 | − | O |
| 19 | 5/6/2014 | Blood | CTX, CAZ, CPM, CPD, FOX, AUG, IMI | − | bla TEM, bla CTX-M, bla SHV | 16 | bla EBC | P |
| 20 | 5/7/2014 | Urine | FOX, AUG | − | bla TEM | 2 | − | Q |
| 21 | 5/7/2014 | Urine | FOX, AUG | − | − | 8 | bla EBC | R |
| 22 | 5/15/2014 | Wound | CTX, CAZ, CPM, CPD, FOX, AUG, IMI, MEM, ETP, GM, CIP, LEV, GAT | + | bla SHV | 4 | S |
CTX: cefotaxime; CAZ: ceftazidime; CPM: cefepime; CPD: cefpodoxime; FOX: cefoxitin; AUG: amoxicillin-clavulanate; IMI: imipenem; MEM: meropenem; ETP: ertapenem; AK: amikacin; GM: gentamicin; CIP: ciprofloxacin; LEV: levofloxacin; GAT: gatifloxacin; MDR: multi-drug resistant; ESBL: extended-spectrum β-lactamase; MIC: minimum inhibitory concentration; RAPD: random amplified polymorphic DNA.
The gene for AmpC, bla EBC, was detected in only 17 (57%) isolates. Another common AmpC-associated gene, bla ACC, was detected in 5 (16.6%) isolates. The genes bla CIT and bla DHA were detected in only 2 (6.6%) and 2 (6.6%) of E. cloacae isolates, respectively. The genes bla FOX and bla MOX were not detected. In addition, we failed to detect carbapenemase genes. The most prevalent genes were rpos and fliI reported in 50 (87.7%) isolates, followed by csgB, csgD, csgA, iutA, and fepA observed in 40 (70.2%), 39 (68.4%), 34 (59.6%), 31 (54.4%), and 29 (50.9%) isolates, respectively. E. cloacae isolates were grouped into 21 RAPD types, which were designated as type A (two isolates) to S (one isolate each) (Table 2). E. aerogenes isolates were grouped into seven RAPD types, which were designated as type A (two isolates) to G (one isolate each) (Table 3). In the present study, the most prevalent species was E. cloacae (77.1%) and its predominance was similar to that reported by Khari et al. and Kanamori et al2,3. In recent years, E. cloacae is the most common pathogen causing nosocomial infections1. In this study, 84.3% of isolates were resistant to cefoxitin. High level resistance to cefoxitin has been previously reported by other investigators2,3, suggesting that treatment with these drugs should be avoided in Enterobacter infections.
TABLE 3: Characteristics of Enterobacter aerogenes isolates
| Isolate ID | Date | Source | Resistance pattern | MDR | ESBL gene | MIC of IMI | AmpC gene | RAPD type |
|---|---|---|---|---|---|---|---|---|
| 1 | 2/19/2012 | Wound | CTX, AUG | − | bla TEM | 2 | B | |
| 2 | 2/6/2014 | Blood | CTX, CAZ, CPM, CPD, FOX, AUG, IMI, MEM, ETP, AK, CIP | + | bla TEM, blaCTX-M | 4 | C | |
| 3 | 2/25/2014 | Urine | CTX, CAZ, ETP, GM, CIP | − | bla SHV, blaCTX-M | 2 | D | |
| 4 | 3/3/2014 | Urine | CTX, CAZ, CPM, CPD | − | bla SHV, blaCTX-M | 0.625 | A | |
| 5 | 3/11/2014 | Burn | CTX, CAZ, CPD, FOX, GM | − | bla CTX-M | 0.5 | bla EBC | E |
| 6 | 4/22/2014 | Respiratory | – | − | bla CTX-M | 0.25 | bla ACC | F |
| 7 | 5/15/2014 | Wound | CTX, CAZ, CPM, CPD, FOX, AUG, IMI, MEM, ETP, GM, CIP, LEV, GAT | + | bla SHV | 0.625 | A | |
| 8 | 5/25/2014 | Urine | – | − | bla SHV , blaCTX-M | 2 | G |
CTX: cefotaxime; CAZ: ceftazidime; CPM: cefepime; CPD: cefpodoxime; FOX: cefoxitin; AUG: amoxicillin- clavulanate; IMI: imipenem; MEM: meropenem; ETP: ertapenem; AK: amikacin; GM: gentamicin; CIP: ciprofloxacin; LEV: levofloxacin; GAT: gatifloxacin; MDR: multi-drug resistant; ESBL: extended-spectrum β-lactamase; MIC: minimum inhibitory concentration; RAPD: random amplified polymorphic DNA.
Our study revealed that 35.1%, 12.3%, and 10.5% of isolates were resistant to imipenem, levofloxacin, and gatifloxacin, respectively. Previous reports from Iran have shown that the resistance rate of Enterobacter isolates to imipenem and gatifloxacin was 2% and 7%, respectively1. Our results indicated the significant increase in the resistance to carbapenem and ciprofloxacin, which may be attributed to the inappropriate and widespread use of antibiotics1. Of the 30 isolates that were recognized as phenotypically positive for ESBL production in this study, 27 were positive for ESBL genotypes. In the study conducted by Kanamori et al. from Japan, 22 of 364 Enterobacter spp. were identified phenotypically positive for ESBL production, but only 11 isolates harbored ESBL genes; ESBL genes were undetected in the remaining 11 isolates2. Discrepancy between disc tests and PCR detection results may be associated with the lack of any standardized method for the detection of ESBLs in Enterobacter spp2. In the present survey, 30 (52.6%) Enterobacter isolates were found to be ESBL producers. Kanamori et al. also reported that 6% Enterobacter spp. were ESBL producers2. The high prevalence of ESBL-positive isolates in our study may be associated with the extensive use of third-generation cephalosporins for the treatment of Enterobacter infections. It should be noted that 10% (3/30) isolates were ESBL negative and eight isolates that were recognized phenotypically positive for carbapenemase failed to show any carbapenemase-related genes, suggestive of the involvement of other resistance mechanisms. In our study, 26.7% (8/30) of ESBL-positive isolates were MDR. Peymani et al. reported that all ESBL-positive Enterobacter isolates were MDR1. In our study, bla TEM and bla CTX-M were the most common ESBL resistance genes, which were frequently reported in other countries2. In the present study, bla EBC (57.7%) was the most common type of AmpC β-lactamase, followed by bla ACC (16.6%). Miró et al. reported that the CMY (78.3%) and DHA (19.5%) families were the most prevalent type of AmpC β-lactamase in 35 hospitals in Spain13. However, the prevalence of ESBL and AmpC-producing Enterobacter spp. varied among different studies, which may be associated with the differences in the geographical area, type of infection, and settings (hospital or community). Similar to previous reports, we observed the coexistence of ESBL-encoding genes in clinical isolates1,2. Several virulence factors have been identified in the pathogenesis of Enterobacter spp4–7. The majority of isolates (87.7%) carried rposand fliI. The high frequency of these genes may indicate that these genes are essential for the development of disease. In contrast to the findings of our study, Krzyminska et al. observed that only 27% of isolates harbored fliI (TTSS gene)6. In the current study, the frequency of csgB, csgD, and csgA was 70.2%, 68.4%, and 59.6%, respectively, which is lower than that reported in the previous study by Akbari et al. These authors showed that csgD and csgA genes were present in 100% and 77.75% of isolates, respectively14. The genes iutA and fepAwere found in 54.4% and 50.9% of isolates in our study. Mokracka et al. reported that 49% of E. cloacae strains produced aerobactin15. However, differences were observed in the frequency of virulence genes reported in different studies; this difference may be associated with the variation in the geographical area, clinical samples, and other factors. RAPD-PCR analysis revealed the significant genetic heterogeneity. In addition, molecular analysis demonstrated that more than 90% (28/30) of ESBL-producing isolates were clonally unrelated, indicating that the reported infections had no relation with clonal outbreak. In conclusion, bla TEM, bla CTX-M, and bla EBC are the most common resistance gene types and more than 50% of isolates harbored virulence genes. RAPD-PCR analysis revealed high genetic diversity among isolates.
