Phylogenetic analysis of Escherichia coli isolated from broilers with colibacillosis based on gyrA gene sequences

Abstract

Escherichia coli isolates from chickens with colibacillosis were assigned to phylogenetic groups based on multiplex polymerase chain reaction (PCR) and antibacterial resistance of E. coli belonging to these groups was examined. Furthermore, the gyrA gene of isolates was sequenced and a phylogenetic tree was generated. A total of 84 E. coli isolates were grouped using multiplex PCR of TSPE4.C2, chuA, yjaA, and gadA molecular markers. Four phylogenetic groups were identified with strains divided as follows: 16 in group A (19.05%), 17 in group B1 (20.24%), 23 in group B2 (27.38%), and 28 in group D (33.33%). Escherichia coli isolates belonging to phylogenetic groups B2 and D were resistant to Soltrim and Flumequine unlike the majority of E. coli isolates that belonged to groups A and B1, and which were susceptible to these antibiotics. The phylogenetic results based on gyrA gene sequences from multiplex PCR revealed that E. coli phylogenetic grouping was in accordance with the clusters obtained in the phylogenetic tree. In conclusion, the comparative sequence analysis of gyrA sequences provides a firm framework for an accurate classification of E. coli and related taxa and may constitute a pertinent phylogenetic marker for E. coli.

Introduction

Avian colibacillosis is caused by avian-pathogenic Escherichia coli (APEC), which is the etiological agent of extraintestinal infections including respiratory infections, pericarditis, and septicemia in poultry (1). Avian colibacillosis is the most common infectious bacterial disease in poultry and is responsible for major economic loss in the poultry industry worldwide (2).

Molecular characterization of infectious agents is useful for epidemiological surveillance and public health activities. Based on an improved multiplex polymerase chain reaction (PCR) using the genetic markers chuA, yjaA, and gadA, and the DNA fragment TspE4.C2, E. coli strains have been phylogenetically classified into 4 main groups: A, B1, B2, and D (3,4). It has been reported that strains from these 4 groups vary in terms of their characteristics, including antibiotic-resistance level (5) and virulence. Virulent extraintestinal strains mainly belong to group B2 and to a lesser extent, group D (6,7).

Multiplex PCR using genetic markers chuA, yjaA, gadA, and TspE4. C2, is an effective tool for the speciation of E. coli isolated from different sources and many researchers have successfully applied this method (3,4,8). However, a high rate of incorrect classifications using multiplex PCR protocol has been reported in previous work (4,9).

DNA gyrase enzyme is composed of 2 subunits, gyrA and gyrB, which introduce negative supercoils into DNA. This essential bacterial enzyme is found in all bacteria and is conserved among species, making it an attractive target for phylogenetic studies (10). The presence of highly conserved motifs in gyrA gene sequences provides a useful tool when designing universal primers for the study of bacterial identity and diversity (11).

Menard et al (12), showed that a phylogenetic tree generated based on gyrA gene sequencing enabled reliable clustering of Helicobacter cinaedi and Flexispira strains. In the present study, a 1.8 kb fragment of the gyrA gene from bacterial strains isolated from clinical specimens of poultry with collibacilosis was amplified. This region was sequenced and the results were used to compile phylogenetic relationships for the isolated bacteria. We presented a new phylogenetic analysis of E. coli based on the gyrA gene region and compared the results with those of the phylogenetic analysis by multiplex PCR.

Materials and methods

Bacterial isolates

Eighty-four E. coli isolates obtained from broiler chickens with colibacillosis from private veterinary laboratories across northern Iran between 2014 and 2015 were used herein. All isolates were identified as E. coli by their morphology, growth characteristics, Gram staining, and biochemical methods [indole, methyl red, Voges-Proskauer, citrate (IMViC) tests and urease production, H₂S production, and various sugar fermentation tests] as described by Nolan et al (1).

DNA extraction and multiplex PCR

Bacterial genomic DNA was extracted using the rapid boiling method described by Wang et al (13). A single colony of grown E. coli isolate was boiled at 100°C for 5 min in 100 mL of distilled water. After cooling to room temperature, the suspension was centrifuged for 3 min at maximum speed to remove cell debris. Assignment of E. coli isolates into phylogenetic groups was performed using the multiplex PCR procedure described by Doumith et al (3).

Antimicrobial susceptibility test

An antibiotic sensitivity test was performed using the Kirby-Bauer method on Müeller-Hinton agar (disc diffusion technique) as described by the Clinical Laboratory Standards Institute (CLSI) (14). Antimicrobial disks are used frequently in the poultry industry to detect antibiotic sensitivity in chickens diagnosed with colibacillosis. Antimicrobial susceptibility of E. coli isolates to the following antibiotics was tested: Cefazolin (KZ; 30 μg), Imipenem (IPM; 10 μg), Trimetoprim-sulfametoxazol (SXT; 23.1 + 75.25 μg), Doxycycline (DO; 30 μg), Colistin (CT; 10 μg), Soltrim (ST; 23.75 μg), Flumequine (FLM; 30 μg), Linco-spectin (LP; 15 + 200 μg), Florfenicol (FFC; 30 μg), and Enrofloxacin (ENR; 5 μg).

Amplification of gyrA gene

A pair of primers [gyrA-F (5′-TTACGGCCCACGATGCTGATTT-3′) and gyrA-R (5′-CATCAACGGTTGTCTGGCGTAT-3′)] was designed based on E. coli gyrA nucleotide sequences retrieved from GenBank. These primers were used to amplify a 1.8 bp fragment of gyrA between positions 3062739 and 3064622 of the E. coli O157:H7 Sakai genome (accession no. BA000007.2). The PCR reaction was prepared in a 25 μL mixture containing 2.5 μL of 10× PCR buffer (SinaClon, Iran), 2 mM of MgCl₂, 50 picomole of each primer, and 50 μM of each ATP, CTP, GTP, and TTP, 2.5 units of TaqDNA polymerase (SinaClon, Ekbatan, Tehran, Iran) and 5 μL of extracted DNA. The thermal profile was initiated with a denaturation step at 94°C for 5 min followed by 35 cycles at 94°C for 45 s, then the annealing step at 60°C for 45 s, and lastly, the extension step at 72°C for 130 s and again for 5 min. Polymerase chain reaction cycles were performed using a QB-96 gradient thermal cycler (Quanta Biotech, Surrey, England). The PCR products were electrophoresed on 1.5% (w/v) agarose gel containing 2.5 μL per 50 mL of SimplySafe (EURx, Gdańsk, Poland) for 1 h at 75 V and visualized under a UV transilluminator.

Purification of amplified gyrA gene and nucleotide sequencing

The amplified gyrA gene was purified from the agarose gel using an AmbiClean gel kit (Vivantis, Subang Jaya, Malaysia) according to the manufacturer’s instructions. Purified gyrA from 6 E. coli isolates of different phylogenetic groups were chosen for nucleotide sequencing and 15 μL of purified PCR products from E. coli isolates belonging to each phylogenetic group (A, B1, B2, and D) were sent to SinaClon for sequencing.

gyrA gene nucleotide sequence analysis

Nucleotide sequences were searched in GenBank using the advanced BLAST similarity search option. Nucleotide sequences were aligned and compared with other gyrA gene nucleotide sequences from GenBank (Table I) using Clustal W and a phylogenetic tree was generated using the neighbor-joining (NJ) method in MEGA version 6.0 software (15,16).

Table I.

Names and accession number of E. coli strains used for generating phylogenetic tree

	E. coli strain	Accession number
1	94-3024	CP009106.2
2	E24377A	CP000800.1
3	ECC-1470	NZ_CP010344.1
4	O26:H11 str. 11368	NC_013361.1
5	APEC O78	NC_020163.1
6	ACN001	NZ_CP007442.1
7	BW2952	NC_012759.1
8	ER2796	NZ_CP009644.1
9	P12b	NC_017663.1
10	KLY	NZ_CP008801.1
11	O157:H7 str. SS52	NZ_CP010304.1
12	SMS-3-5	NC_010498.1
13	IAI39	CU928164.2
14	UMN026	CU928163.2
15	NA114	NC_017644.1
16	CFT073	NC_004431.1
17	ED1a	CU928162
18	BL21 (DE3)	CP001665.1
19	B str. REL606	CP000819.1

Statistical analysis

Statistical analysis was performed with SPSS version 22 (SPSS IBM; Armonk, New York, USA).

Results

Phylogenetic group determination by multiplex PCR

Phylogenetic groups of isolated E. coli were determined according to the multiplex PCR method described by Doumith et al (3). Figure 1 shows the amplified gadA, chuA, yjaA, and TSPE4.C fragments which were used for phylogenetic grouping. A larger proportion of isolates were classified as group D, comprising 33.33% of all isolates, compared with groups B2, B1, and A, which accounted for 27.38%, 20.24%, and 19.05% of isolates respectively.

Figure 1.

Multiplex PCR profiles showing phylogenetic groups of E. coli examined in the present study. Lanes 1 — 100bp molecular marker (Bioflux, South Korea). Lane 2 — Phylogenetic group A. Lane 3 — Phylogenetic group B1. Lane 4 — Phylogenetic group B2. Lane 5 — Phylogenetic group D.

Antimicrobial susceptibility

All E. coli isolates from the different phylogenetic groups were susceptible to Imipenem and Linco-spectin antibiotics. All E. coli isolates showed resistance against Cefazolin, Doxycycline, and Florfenicol, except for 6 isolates from phylogenetic group B1, which were sensitive to the antibiotics. All E. coli isolates belonging to phylogenetic groups B2 and D were resistant to Soltrim and Flumequine, while most E. coli isolates from phylogenetic groups A and B1 were susceptible to them (Table II).

Table II.

Antimicrobial resistance of E. coli from broilers with colibacillosis

	Number (%) of resistant E. coli isolates
Phylogenetic groups Antibiotics	A (n = 16)	B1 (n = 17)	B2 (n = 23)	D (n = 28)	Total (n = 84)
Cefazolin	16 (100.0)	11 (64.7)	23 (100.0)	19 (67.8)	84 (100.0)
Imipenem	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
Trimetoprim-sulfametoxazol	0 (0.0)	5 (29.4)	22 (95.6)	28 (100.0)	55 (65.5)
Doxycycline	16 (100.0)	16 (94.1)	23 (100.0)	28 (100.0)	83 (98.8)
Colistin	10 (62.5)	0 (0.0)	0 (0.0)	10 (35.7)	20 (23.8)
Soltrim	0 (0.0)	12 (70.5)	23 (100.0)	28 (100.0)	63 (75.0)
Flumequine	5 (31.3)	0 (0.0)	23 (100.0)	28 (100.0)	56 (66.6)
Linco-spectin	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
Florfenicol	16 (100.0)	12 (70.6)	23 (100.0)	28 (100.0)	79 (94.0)
Enrofloxacin	6 (37.5)	0 (0.0)	0 (0.0)	20 (71.4)	26 (30.9)

Phylogenetic analysis of isolated E. coli based on nucleotide sequences of gyrA gene

Nucleotide sequences obtained from 6 E. coli isolates from different phylogenetic groups were used to generate a phylogenetic tree. Nineteen gyrA gene sequences from different strains of E. coli retrieved from GenBank were used for the phylogenetic analysis. Figure 2 shows the phylogenetic tree generated by the MEGA 6.0 program. Based on this phylogenetic tree, E. coli isolates were grouped in 5 clusters. Interestingly, E. coli strains from phylogenetic groups based on multiplex PCR were clustered within the same phylogenetic groups determined by gyrA sequencing, confirming that phylogenetic grouping based on multiplex PCR is in accordance with phylogenetic grouping performed using gyrA gene sequence analysis (Table III). Escherichia coli strains from phylogenetic group A were divided in 2 different clusters in the constructed tree.

Figure 2.

Phylogenetic tree of gyrA gene of E. coli isolates from chicken with colisepicemia and E. coli strains retrieved from GenBank generated using neighbor-joining method in MEGA 6.0.

Table III.

Phylogenetic grouping of E. coli isolates from chickens with colisepticemia using multiplex PCR and gyrA gene nucleotide sequences

	E. coli isolate	Phylogenetic grouping based on multiplex PCR	Phylogenetic grouping based on gyrA gene sequences
1	E. coli isolate 7-IR	B2	B2
2	E. coli isolate 11-IR	D	D
3	E. coli isolate 37-IR	D	D
4	E. coli isolate 52-IR	B1	B1
5	E. coli isolate 72-IR	A	A
6	E. coli isolate 73-IR	A	A

Discussion

Avian colibacillosis has been known for more than a century; however, it still is an endemic disease worldwide (1). In order to understand colibacillosis outbreaks, characterization of E. coli population structure based on phylogenetic analysis plays an important role (9). Escherichia coli can be classified by a number of distinct phylogenetic groups and it is now obvious that strains in these different phylogenetic groups vary in terms of their ecological niches, life-history characteristics, and propensity to cause disease. Therefore, assigning E. coli isolates to one of the recognized phylogenetic groups allows for better understanding of the epidemiology of the diseases caused by this bacteria (4).

Different techniques including multiplex PCR, single strand conformation polymorphism (SSCP) (17), denaturing gradient gel electrophoresis (DGGE) (18), restriction fragment length polymorphism (RFLP) (19), whole genome sequencing, and multilocus sequence typing (MLST) have been employed for the assignment of particular strains to phylogenetic lineages (4,20,21). Sequencing may still be time-consuming and require expertise for interpretation of results, however, it has higher discriminatory power than other rapid genotyping techniques such as SSCP and DGGE which both cover less than 400 bp nucleotide sequences, and RFLP which is based on the variations associated with restriction sites only (22). Multilocus phylogenic lineages will be a more powerful technique for strain identification of E. coli isolates (23), however, gyrA sequencing may be used as an initial screening tool.

In the present study, E. coli isolates from chickens with colibacillosis were assigned to phylogenetic groups based on multiplex PCR and gyrA gene sequence analysis. Phylogenetic grouping of E. coli isolates based on multiplex PCR and nucleotide sequence analysis of gyrA showed that both methods grouped E. coli isolates in the same groups.

Escherichia coli isolates belonging to phylogenetic groups A and B1 showed different susceptibility levels to Soltrim and Flumequine compared to E. coli isolates from B2 and D phylogenetic groups. This finding may elicit further research on antimicrobial susceptibility differences among different E. coli phylogenetic groups. Among the 10 antibiotics used for the antibiogram, E. coli isolates showed moderate to high levels of resistance to 8 antibiotics. These results confirm the necessity of antimicrobial susceptibility tests for E. coli isolates in order to select appropriate antibiotics for treatment of disease.

The majority of E. coli isolates (61.71%) examined in this study belonged to phylogenetic groups B2 and D which supports previous findings that most virulent strains of E. coli are in B2 and D phylogenetic groups (24,25). Previous studies in Japan and the United States have shown that phylogenetic groups A and D are predominant among APEC isolates (24,26). In a study by Hiki et al (8), phylogenetic groups A and B1 comprised more that 80% of E. coli isolated from healthy broilers, while in the present study these groups accounted for only 39.29% of E. coli isolates from APEC. There was a significant difference (P < 0.01) between frequencies of phylogenetic groups of E. coli isolates from healthy and diseased broilers.

As previously reported, phylogenetic groups A and B1 are recognized as sister groups and in the NJ trees depicted, strains assigned to phylogenetic group A made up a relatively homogeneous group among all the strains assigned to groups A and B1 (4). This finding was confirmed by the NJ tree constructed based on gyrA nucleotide sequences in the present study.

In conclusion, E. coli isolates from chickens with colibacillosis were distributed in different phylogenetic groups. Moreover, our study suggests that the gyrA gene can be used as a candidate marker for phylogenetic grouping of E. coli. It also revealed that E. coli isolates from chickens with colibacillosis have moderate to high levels of resistance to many antibiotics. Further studies on gyrA in E. coli in other animals may better clarify the importance of the gyrA gene for genotyping E. coli.