Genetic Linkage of the Antibiotic Resistance Ability in the Escherichia coli UR4 Strain Isolated from Urine

INTRODUCTION

Escherichia coli (E. coli) is a Gram-negative, rod-shaped, motile, facultatively anaerobic and non-sporulating bacterium. It belongs to family enterobacteriaceae in which most bacterial genera are human pathogens (Vogt and Dippold, 2005). It is part of the microflora of human colon and can benefit their hosts by producing vitamin K and by neutralization of pathogenic bacteria within the intestine and can be changed to pathogenic bacteria under certain conditions such as suppression of immune system (Reid et al., 2001).

Transmission of E. coli is commonly by fecal/oral route with contaminated foods. At least five types of intestinal infections that differ in pathogenic mechanisms have been identified for E. coli viz., enterotoxigenic, enteropathogenic, enterohaemorrhagic, enteroinvasive and enteroaggregative (Harvey and Champe, 2012). E. coli was isolated from many pathogenic cases such as gastroenteritis, neonatal meningitis, haemolytic-uremic syndrome, peritonitis, mastitis, cystitis, pyelonephrities, urethrities, vaginities, endometrities and prostatitis (Tauschek et al., 2002). Unfortunately, many authors described the existence of E. coli variants resistant to many antibiotics (Saffar et al., 2008). The antibiotic resistance ability in E. coli is due to secretion of enzymes by target strains, exopolyrsacharide production, modification of cell surface, drug efflux system and existence of genes associated to certain plasmids (Fange et al., 2009). This clearly showed that there is a need to continue research on E. coli strains to able to characterize the multidrug resistant strains and to understand the actual reasons of the antibiotic resistance phenomena which is many cases even at modification of cell membrane or secretion of enzymes and modifications of plasmid associated genes (Harvey and Champe, 2012).

The present study was undertaken to isolate and characterize the multidrug resistant strains of E. coli from Egyptian patients by possible biochemical and molecular methods. The study was also aimed to study the genetic determinants of the antibiotic resistance ability in E. coli UR4.

MATERIALS AND METHODS

Isolation of bacterial cultures and antibiotic sensitivity test: About 50 clinical samples of urine, blood and stool were taken from patients who did not respond to treatment by antibiotics Egyptian patients and were transferred asceptically to the Microbiology Lab. Bacterial cultures were isolated onto MacConkey agar media (Oxoid) and Hichrome E. coli selective media (HiCrome,^TM). Cultures grown on both media were predicted as Escherichia bacteria and were consequently, chosen for study. The antibiotic bioassay were studied using disc diffusion assay for 17 bacterial isolates grown on Hichrome media against 14 types of antibiotics listed in Table 1 (Ehinmidu, 2003). Results were taken according the manufacturer’s instructions of antibiotic discs (Oxoid).

Identification of multidrug resistant bacteria: Gram staining, cell morphology, motility, catalase reaction, oxidase test, production of H₂S and growth of bacteria on both MacConkey agar and Hichrome agar confirmed that the 17 bacterial isolates belonged to genus Escherichia. The species was determined by tests listed in Table 2 (Wolfgang et al., 1998). Identification was carried out according to diagnostic key of Holt et al. (1994) and Garrity et al. (2005).

The E. coli UR4 strain which was resistant to all antibiotics used were subjected to molecular characterization by elucidation of 16 S rRNA cataloging analysis.

Table 1:	Antibiotic sensitivity test of 17 bacterial isolates from many patients at two hospitals in Cairo

R: Resistant bacteria, I: Intermediate, S: Sensitive

Table 2:	Biochemical characteristics of the antibiotics resistant bacteria

DNA was extracted from UR4 strain according to protocol of Liu et al. (2002). PCR mediated amplification of the 16 S r RNA gene and its purification were done using PrepMan Ultra (Applied Biosystem), Microseq PCR and Miroseq Cycle Sequencing Applied Biosystem. The 16 S rRNA gene was amplified with primer mixture viz., 5'AGAGTTT GATCMTGGCTCAG '3 forward primer and 3' TACAG CATTGTTCCATTGGCAT '5 reverse primer.

The PCR mixture consisted of 30 picomoles of each primer, 10 ng of chromosomal DNA, 200 μM dNTPs and 2.5 Units of Taq polymerase in 50 μL of polymerase buffer. The PCR was carried out for 30 cycles in 94°C for 1 min., 55°C for 1 min and 72°C for 2 min. After completion, a fraction of the PCR mixture was examined using agarose gel electrophoresis (Ausubel et al., 1999) and the remnant was purified using QIA quick PCR purification reagents (Qiagen). DNA sequences were obtained using a 3130 X DNA sequencer (Genetic Analyzer, Applied Biosystems, Hitachi, Japan), BigDye Terminator Cycle Sequencing (see details below). The PCR product was sequenced using the same PCR primers. Blast program was used to assess the DNA similarities and multiple sequence alignment and molecular phylogeny were performed using Bio Edit software (Hall, 1999).

DNA sequencing: Automated DNA sequencing based on enzymatic chain terminator technique, developed by Sanger et al. (1977) was done using 3130 X DNA Sequencer (Genetic Analyzer, Applied Biosystems, Hitachi, Japan). The sequencing reaction was performed with four different fluorescent labels identifying the ddNTPs, instead of the radioactive labels. These flurophores were excited with two argon lasers at 488 and 514 nm, respectively when the respective bands passed the lasers during the electrophoresis. The specific emissions were detected and the data were collected for analysis (Prober et al., 1987 and Freeman et al., 1990). The thermal cycling mixture was as follows: 8 μL of BigDye terminator mix, 6 μL of the sequencing primer (10 pmol) and 6 μL of the PCR product, then the reaction was run in the thermal cycler. The cyclic reaction composed of 1 min at 95°C, then 49 cycles of 30 sec at 95°C, 10 sec at 52°C and 4 min at 60°C. The products were purified using special column according to the instruction of the manufacturer. The eluates were taken and add high dye formamide with (1:1)/volume ratio, run at 95°C for 5 min for denaturation, shock on ice; then the sample become ready for sequencing in 3130 X DNA sequencer and analysis. The sequence of 16 S gene was sent to glue bank under accession number JQ388912.

Selection of E. coli UR4 mutants: Elimination of the antibiotic resistance ability was carried out according to Enan et al. (1996). A series of test tubes, each containing 9 mL nutrient broth with or without 20 mg mL^-1 of each of the curing agents or antibiotics listed in Table 3, were inoculated with log phase cells of the E. coli UR4 wild strain to give 2x10⁷ CFU mL^-1. Tubes supplemented with curing agents; tubes without curing agents were incubated at 30 and 42°C, respectively for 48 h. Tubes containing visible bacterial growth were transferred another time in another tubes like the first rounds of the experiment. After successive transfer, bacterial suspensions were plated onto nutrient agar plates and were counted. Number of colonies sensitive to the cold solution I [50 mM glucose, 25 mM Tris buffer (pH antibiotics listed in Table 4 were counted after their antibiotic bioassay onto separate nutrient agar plates (Enan et al., 1996).

Plasmid isolation and agarose gel electrophoresis: Plasmids were extracted from UR4 wild strain and its mutants according to Sambrook et al. (1989):

•	Multiple colonies were collected using a sterile loop and added to individual tubes containing 2 mL of nutrient broth, incubated overnight at 37°C
•	One and half mililitter of the culture was poured into microcentrifuge tube and centrifuged at 12,000 rpm for 30 sec at 4°C
•	The supernatant was removed, leaving the bacterial pellet as dry as possible
•	The bacterial pellet was resuspnded in 100 μL of ice 8.0), 10 mM EDTA (pH 8.0)]. Vigorous vortexing was done to ensure that the pellet was completely dispersed in solution 1
•	Two hundred microlitter of freshly prepared solution II [0.2 NaOH in 1% SDS) were added and the tube was closed tightly, inverted rapidly five times and stored on ice for 5 min. At this stage a viscous bacterial lysate was formed
•	One hundred and fifty μL of ice cold solution III [3M potassium acetate in 11.5% glacial acetic acid] were added to neutralize the reaction and the tubes were closed and vortexed for 10 sec to disperse solution III through the viscous bacterial lysate then stored on ice for 3-5 min
•	The tubes were centrifuged at 12,000 rpm for 5 min at 4°C then the supernatant was transferred to a fresh tube
•	Two volumes of ethanol (double the volume of the supernatant) were added (to precipitate the double stranded DNA) and mixed by votexing. The mixture was allowed to stand at room temperature for 2 min
•	The tubes were centrifuged at 12,000 rpm for 5 min at 4°C
•	The supernatant was removed and the tubes were inverted on a paper towel to drain away all fluids
•	The pellets were rinsed with 1 mL of 70% ethanol and allowed to dry in the air for 10 min
•	The pellets were redissolved in 50 μL of tris EDTA (TE) buffer (10 mM Tris HCl pH 8.0 in 1 mM EDTA), vortexed briefly and stored at 20°C

Table 3:	Effect of curing agents on the antibiotic resistance ability of E. coli UR4

Table 4:	Biochemical characteristics of the wild E. coli UR4 strain and its mutants

Agarose gel electrophoresis was carried out according to Davis et al. (1986). Twenty five microliter of tracking dye were added to 100 μL of plasmid DNA preparation. The sample was loaded on 0.7% agarose gel in Tris-EDTA buffer pH 8.2 and electrophoresis was conducted at 100 volt for 1-2 h. The gel was stained is EtBr (0.5 g mL^-1) for 15 min., distained in 1 mM MgSO₄ for 30 min, photographed by gel documentation system.

RESULTS

Isolation, characterization and identification of multidrug resistant bacteria: Urine, blood and stool samples were taken from patients who did not respond to medications and were analysed microbiologically onto both MacConkey agar and HiCrome^TM agar media. About 17 culture grown well were chosen and were assayed against 14 types of antibiotics listed in Table 1. All the 17 bacterial cultures tested were showed to be multidrug resistant bacterial cultures.

All the 17 bacterial cultures were rod shaped, motile, Gram negative and catalase positive cells. Many biochemical tests used for bacterial identification were studied. Results are given in Table 2. All isolates showed positive results with regard to indole test, methyl red test, liquefaction of gelation, utilization of glucose, lactose, maltose, mannitol, L. arabinose, D-sorbitol and blood hemolysis; but showed negative results with other biochemical tests listed in Table 2. Following diagnostic key of Holt et al. (1994) and Garrity et al. (2005), all the 17 bacterial isolates were identified as belonging to E. coli.

Since, UR4 strain was resistant to all the 14 antibiotics used, it was important to characterize it by 16 S rRNA cataloging analysis for more description of this interesting multidrug resistant bacterium. DNA was isolated from this strain and the 16 S rRNA gene was amplified by PCR technique. Amplified 16 S gene was electrophoresed using agarose gel and the clear band of 16 S gene was taken and sequenced (Fig. 1). The sequence of 16 S gene was submitted to gene bank under accession number JQ388918. It was showed 99.5% similarity to E. coli category.

Selection of antibiotic resistance negative clones: To produce an antibiotic resistance negative clones, the UR4 strain was subjected to 20 mg mL^-1 of antibiotics, SDS, sodium azide and ethidium bromide and was grown at elevated temperature (42°C). The results are presented in Table 3. In population treated with antibiotic, no negative clones (sensitive colonies to antibiotic used) were observed. There was no dependence of the number of either antibiotic resistance negative clones or antibiotic resistance positive clones on the applied dose (20 mg mL^-1) of the antibiotic used.

However, the occurrence of an antibiotic resistance negative colonies (sensitive colonies to the antibiotic) was found in populations either treated with curing agents or grown at elevated temperature. The mutation percentage was 0.61; 1.24; 0.74 and 14.85% in the experimental antibiotic resistant cells of E. coli UR4 after using the curing agents viz SDS; elevated temperature sodium azide; ethidium bromide, respectively. The produced mutants of the antibiotic resistance negative colones became sensitive to many antibiotics (Fig. 2). The mutants produced by SDS; elevated temperature; sodium azide; ethidium bromide were termed SDS; ET; SAZ; ETH and were used for further experiments.


Fig. 1:	Sequence of 16S rDNA gene of the wild strain E. coli UR4


Fig. 2(a-d):	Antibiotic sensitivity test of the E. coli UR4 mutants produced after curing with, (a) SDS, (b) Elevated temperature, (c) Sodium azide and (d) Ethidium bromide

Stability of mutation in the antibiotic resistance negative clones: SDS, ET, SAZ and ETH, were subcultured ten times in nutrient broth. After each subculture, antibiotic sensitively test was carried out. None of them was antibiotic resistant again but all colonies were sensitive to some antibiotic used (Fig. 2).

Comparison of the E. coli UR4 wild strain and its mutants: The antibiotic resistant wild strain E. coli UR4 and its mutants SDS, ET, SAZ and ETH which became sensitive to the antibiotic used were grown in nutrient broth. Growth values were compared. No differences in the rate of growth were observed.

Comparison of biochemical characteristics of the E. coli UR4 wild strain and its mutants: E. coli UR4 wild strain and its mutants SDS, ET, SAZ and ETH which became sensitive clones were subjected to many biochemical tests used for bacterial identification Table 4. Results were compared. There was a quite similar results with regard to biochemical tests listed in Table 4.

Plasmid profile: To study the genetic linkage of E. coli UR4, it mandatory to compare between the plasmid content of the wild strain of E. coli UR4 and its mutants SDS, ET, SAZ and ETH to elucidate whether antibiotic resistance ability is plasmid encoded or is linked to gene located on the chromosome. If the antibiotic resistance ability is plasmid encoded, one plasmid could be deficient in the antibiotic resistance negative variant. However, in case of the antibiotic resistance is chromosomal encoded, the plasmid profile of the wild strain and each mutant should be similar to each other.

Since, the E. coli UR4 wild strain and its mutants SDS, ET SAZ and ETH which became sensitive to antibiotics used had a quite similar physiological properties, plasmids were isolated of the wild strain of E. coli UR4 and its mutants SDS, ET, SAZ and ETH. A preparations of Lumda DNA (Omega) were used as standards. Plasmids and standard DNA were electrophoresed using agrose gel.; plasmid profile is given in (Fig. 3). The plasmid profile of the UR4 wild strain is shown in lane 1 and appears to contain five plasmids with a molecular mass range from 80-1031 bp.


Fig. 3:	Agrose gel electrophoresis of plasmids extracted from E. coli UR4, Lane 1: Wild strain, Lane 2: Mutant strains ET, Lane 3: SDS, Lane 4: Ethidium bromide, Lane 5: SAZ

The SDS, ET, SAZ and ETH mutants appeared to contain only four plasmids. One plasmid from each mutant of a molecular mass of about 600 bp was lost or dissociated in lane 2, 3, 4 and 5. This clearly indicated that this plasmid is the responsible one on the antibiotic resistance ability in the experimental E. coli UR4 strain and this antibiotic resistance ability appeared herein as plasmid encoded.

DISCUSSION

E. coli is a part of the normal flora of the human colon but can be pathogenic both within and outside of the gastrointestinal tract. The difference of the degree of virulence of different E. coli strains are caused by individual plasmid associated with each strain. Many strains of E. coli are pathogenic such as enteropathogenic; enterotoxigenic; enterohemorrhagic; enteroinvasive; enteroadherant which cause watery diarrhea; blood diarrhea; colitis; persistant watery diarrhea in children, respectively (Harvey and Champe, 2012. This clearly showed that there is a need to continue research on E. coli to control its growth for healthy reasons. Unfortunately, the existence of the antibiotic resistance ability in E. coli was previously shown by many authors (Whittam et al., 1993; Todar, 2007). Resistance of E. coli to antibiotic was in many reports due to genetic reasons and this clearly showed that there is a need to study this phenomena to able be find out new control strategies on the antibiotic resistant strains of E. coli.

In this study, 17 bacterial cultures were isolated on specific Hichrome medium from Egyptian patient suffering from urinogential infection symptoms even after antibiotic treatment. It was of interest to study such bacterial cultures. The 17 bacterial cultures appeared to be multidrug resistant bacteria. This supported latter work in this respect(Ehinmidu, 2003). Since this phenomena is very necessary clinically, isolates were identified as E. coli UR4 (Holt et al., 1994; Garrity et al., 2005). For further characterization of the most antibiotic resistant bacterium (UR4 strain), the molecular identification by 16 S r RNA gene sequence analysis was made and UR4 strain appeared to be similar to E. coli sequence in gene bank (Pot and Janssen, 1993). Later on, it was of interest to study the genetic determinants of the antibiotic resistance ability in the UR4. This to concur with previous study in this respect ( Enan and Saad, 2000).

Growing of the experimental UR4 and treatment of its cells with some curing agents led to mutants lost the antibiotic resistance ability and similar work was published by Lichstein and Soule (1994). The mutation was stable character and the SDS, ET, SAZ and ETH mutants continued sensitive to most antibiotic used and no recovery was observed in any of the above mutants. This support latter findings in this respect. The wild E. coli strain contained five plasmids and this clearly indicated the genetic structure of this strain which expresses on high virulence and high ability of this strain to resistant all antibiotics used. This is in agreement with Schafer et al. (1999). The SDS, ET, SAZ and ETH mutants lacked one plasmid of molecular region of about 600 bp and this clearly indicated the role of this plasmid in the antibiotic resistance ability of E. coli UR4. Hence the antibiotic resistance ability in this strain was plasmid encoded.

Further study will be necessary to digest the plasmid containing the antibiotic resistance gene by restriction enzymes to understand its genetic map and to its modifications.

CONCLUSION

The 17 antibiotic bacterial isolates were isolated, characterized by phenotypic and biochemical criteria. They were identified as belonging to E .coli. One only of them was resistant to 14 antibiotics used; was further characterized by 16Sr RNA sequence analysis. It was identified and designated E .coli UR4. Mutants of such strain were obtained by some curing agents; and the plasmids of UR4 wild strain and its mutants were obtained. One plasmid of about 600 bp was showed to be the committed genetic factor for the antibiotic resistance ability in the E .coli UR4 strain.