Molecular Characterization of Virulent Genes and Antimicrobial Resistance Patterns of Escherichia coli Isolated From Calf Scours in Western Iran

ABSTRACT

Objective

This study was designed to determine the virulence factors and antimicrobial resistance profiles of Escherichia coli pathotypes isolated from neonatal calf diarrhoea in Western Iran.

Methods

A total of 350 diarrhoeic faecal samples were collected. The samples were assessed for the presence of E. coli phenotypically. The molecularly confirmed isolates were further classified as different pathotypes on the basis of the genotypic traits. Additionally, the distribution of diverse serotypes and the antimicrobial resistance profile of the isolates were determined.

Results

Among the 307 E. coli isolates, 235 isolates could be categorized as the following pathotypes: enterohaemorrhagic E. coli (EHEC), 36.59%; enteropathogenic E. coli (EPEC), 20.42%; enterotoxigenic E. coli (ETEC), 14.46%; Shiga‐toxigenic E. coli (STEC), 12.34% and necrotoxigenic E. coli (NTEC), 11.48%. STEC O157:H7 was detected in four isolates. O103 and O15 serotypes were the most prevalent. Totally, high rates of antibiotic resistance were observed, with the most common resistance for penicillin (99.14%) and tetracycline (94.46%), and the least common for nitrofurantoin (62.16%) and neomycin (65.10%).

Conclusion

The results substantiated the potential reservoirs of calves for various E. coli pathotypes in the region. With an emphasis on the identification of important O‐serogroups and the high rates of antimicrobial resistance among the isolates, the need for continuous monitoring, strict biosecurity measures in cattle farms and proper antibiotic stewardship are underscored to prevent the transmission of pathogenic and resistant strains to humans.

This study in Western Iran analysed E. coli in diarrhoeic neonatal calves and identified 307 isolates from 350 faecal samples. The pathotypes included EHEC (36.59%), EPEC (20.42%), ETEC (14.46%), STEC (12.34%) and NTEC (11.48%), with four STEC O157:H7 isolates. Serotypes O103 and O15 were the most common. High antibiotic resistance was observed, especially against penicillin (99.14%) and tetracycline (94.46%). Calves may be reservoirs of E. coli, highlighting the need for monitoring, biosecurity and antibiotic stewardship.

1. Introduction

Neonatal calf diarrhoea (NCD) is a pervasive veterinary concern that causes economic losses in the livestock industry worldwide. The adverse outcomes of the disease are not only due to the high mortality rate but also due to impaired growth performance in calves (Cho et al. 2014). It is a multi‐factorial disease where, besides the causative pathogenic agent, calf age, management and environmental factors may influence the clinical outcome of the disease. NCD, defined as diarrhoea occurring during the first month of life, is the most common disease and cause of death in calves worldwide during this period (Kumar et al. 2022). Although it is a multi‐factorial disorder, Escherichia coli is intensively attributed as the main causative agent of the disease, especially in un‐weaned calves (Meganck et al. 2014). Likewise, predisposing factors, including management and nutritional, environmental and physiological influences, may interplay with the prevalence of this bacterium (Cho et al. 2014).

Considering the pathogenesis, virulence factors and clinical manifestations, E. coli is segregated into pathotypes nominated as enterotoxigenic (ETEC), enteropathogenic (EPEC), enterohaemorrhagic (EHEC), necrotoxigenic (NTEC) and enteroaggregative (EAEC) E. coli (Mainil 2013). ETEC is the main cause of diarrhoea in the first week of newborn calf life, and it produces a watery white scour. The pathogenesis of ETEC in NCD is associated with two enterotoxins, heat‐labile (LT) and heat‐stable (ST), and adhesion antigens F5 (K99), F41 and F17. ST and LT enterotoxins, which are encoded by the est and elt genes, respectively, are divided into two subtypes: STa/STb and LT‐I/LT‐II (Kolenda et al. 2015). EHEC pathotype includes Shiga‐toxigenic (STEC) strains that produce two cytotoxins, Stx1 and/or Stx2. The devastating effects of these toxins on intestinal epithelial cells lead to attaching/effacing (A/E) lesions (Werber et al. 2007). EHEC strains harbouring hly and saa genes may possess additional virulence factors, including enterohaemolysin and autoagglutination adhesion (Kang et al. 2004). The EPEC pathotype excretes no enterotoxins. Instead, this pathotype produces an outer membrane protein, intimin, which moderates the typical A/E lesions in the gut mucosa. The ability to produce intimin may be a common feature of both EPEC and EHEC pathotypes (DebRoy et al. 2001).

Despite the reduced involvement of EPEC and EHEC in NCD, these two pathotypes are well substantiated in human diseases. The latter comprises O157:H7 STEC, which causes life‐threatening illnesses in humans, such as haemorrhagic colitis (HC) and haemolytic uraemia syndrome (HUS). Additionally, STEC is a reservoir of cattle. At the same time, STEC causes calf diarrhoea depending on the age, immune system and gastrointestinal system conditions of the calves. In addition to the presence of Shiga toxin, EHEC causes bloody diarrhoea in calves due to its attaching and effacing effects, similar to EPEC (Bielaszewska et al. 2000; Coskun and Sahin 2023). In addition, non‐O157 STEC strains have been identified in severe human diseases. Ruminants are known to be the mere reservoirs of STEC, particularly O157:H7, in their intestinal tract. Hence, the consumption of contaminated domestic animal‐originated food and direct or indirect contact with infected animals is assumed to be a pivotal public health hazard (Bolton et al. 2011). Virulence of the NTEC pathotype correlates with the dispersion of two cytotoxic necrotizing factors, CNF1 and CNF2. CNF1‐positive strains trigger enteritis in ruminants and extra‐intestinal infections in humans, whereas CNF2‐harbouring strains are mainly responsible for intestinal infections or septicaemia in ruminants (Blanco et al. 1996). Antibiotic resistance is a major global health issue.

Antimicrobial resistance is rapidly spreading among E. coli strains. Antibiotic resistance is increasing due to the overuse and misuse of antibiotics in cattle breeding. Although most antibiotics are consumed by livestock (Adesoji and Liadi 2020), there is poor antibiotic diversity. The increase in antibiotic resistance in E. coli could cause difficulties in treatment and increase the mortality rate of NCDs. Additionally, it may result in economic losses due to reduced animal weight gain, increased treatment costs and diminished development in affected individuals (da Costa Custodio et al. 2024). Considering the inevitable antibiotic therapy in some cases of NCD and the potential of E. coli to acquire resistance genes, it is crucial to define the antimicrobial profile of the isolates (van den Bogaard et al. 2000).

Due to scarce information regarding the role of diarrhoeagenic E. coli in NCD and the antibiotic resistance patterns of the isolates in Iran, an attempt was made to evaluate the prevalence, virulence and antimicrobial resistance properties of E. coli pathotypes isolated from diarrhoeic calves in Iran.

2. Materials and Methods

2.1. Sample Collection and E. coli Isolation

In a cross‐sectional study from October 2018 to September 2019, 350 rectal swabs were purposively collected from diarrhoeic, up to 70‐day‐old calves in the four seasons (Table 1). The sampling zone included 34 industrial dairy farms from Kurdistan province, in the west of Iran. Neither dams nor newborns were vaccinated to prevent calf scours. No antibiotics were received 3–5 days preceding the sampling time. The swabs were placed into sterile tubes containing Stuart medium (Merck, Germany) and transported to microbiology laboratory in a portable insulated cold box. The samples were immediately streaked onto MacConkey agar (MAC, Merck, Germany) and incubated for 18–24 h. One lactose‐positive colony on each MAC plate was further subcultured on eosin‐methylene‐blue (EMB, Merck, Germany) and incubated overnight at 37°C. Cultures with a distinctive metallic‐green sheen were ascertained as E. coli on the basis of Indole, Methylene Blue, Voges Proskauer, Simmons Citrate and Triple Sugar Iron reactions.

TABLE 1.

The frequency of collected samples regarding age, sex and season.

	Spring		Summer		Fall		Winter
Age group (weeks)	Heifer a calves n (%)	Bull b calves n (%)	Heifer calves n (%)	Bull calves n (%)	Heifer calves n (%)	Bull calves n (%)	Heifer calves n (%)	Bull calves n (%)	Total n (%)
1	9 (2.57)	7 (2)	2 (0.57)	3 (0.85)	7 (2)	6 (1.71)	12 (3.42)	8 (2.28)	54 (15.42)
2	4 (1.14)	4 (1.14)	0 (0)	9 (2.57)	13 (3.71)	3 (0.85)	8 (2.28)	5 (1.42)	46 (13.14)
3	4 (1.14)	6 (1.71)	4 (1.14)	4 (1.14)	2 (0.57)	5 (1.42)	7 (2)	6 (1.71)	38 (10.58)
4	5 (1.42)	13 (3.71)	1 (0.28)	3 (0.85)	6 (1.71)	8 (2.28)	11 (3.14)	3 (0.85)	50 (14.28)
5	7 (2)	7 (2)	3 (0.85)	7 (2)	6 (1.71)	4 (1.14)	3 (0.85)	4 (1.14)	41 (11.71)
6	10 (2.85)	8 (2.28)	5 (1.42)	1 (0.28)	2 (0.57)	4 (1.14)	8 (2.28)	0 (0)	38 (10.58)
7	1 (0.28)	5 (1.42)	1 (0.28)	3 (0.85)	0 (0)	5 (1.42)	6 (1.71)	5 (1.42)	26 (7.42)
8	10 (2.85)	4 (1.14)	3 (0.85)	6 (1.71)	6 (1.71)	4 (1.14)	5 (1.42)	2 (0.57)	40 (11.42)
9	3 (0.85)	4 (1.14)	2 (0.57)	2 (0.57)	0 (0)	0 (0)	3 (0.85)	3 (0.85)	17 (4.85)
Total	53 (15.14)	58 (16.57)	21 (6)	38 (10.85)	42 (12)	39 (11.14)	63 (18)	36 (10.28)	350 (100)
	111 (31.71)		59 (16.85)		81 (23.14)		99 (28.28)

^a A female cow that has not yet given birth to a calf.

^b A young intact male.

2.2. DNA Extraction and Genotypic Characterization of Virulence Factors of E. coli Isolates

DNA of the E. coli isolates was extracted from overnight cultures of the isolates in Luria‐Bertani broth (Merck, Germany) using the boiling method (Reischl et al. 2002). Molecular identification of the isolates as E. coli was performed using universal Eco2083 and Eco2745 primers (Riffon et al. 2001). To determine the pathotypes, specific virulence genes were detected using conventional single and multiplex PCR assays as follows: ETEC was identified by detecting the presence of sta, f5 and f41 genes (Franck et al. 1998), as well as the stII gene (Vidal et al. 2005). EHEC isolates were screened for stx1, stx2 and eae genes (Vidal et al. 2005; Islam et al. 2007). Additionally, the saa gene was targeted for further confirmation (Jenkins et al. 2003). EPEC was identified based on the presence of the eae gene (Islam et al. 2007) and further subtyped into typical or atypical strains by detection of the bfpA (bundle‐forming pili) gene (Gunzburg et al. 1995). EAEC was initially identified by detecting the astA gene (Yamamoto et al. 1997). In astA‐positive isolates, additional EAEC‐associated genes, including aafA (aggregative adherence fimbriae‐II), pet (plasmid‐encoded toxin) and aggR (transcriptional regulator), were screened (Tokuda et al. 2010). NTEC detection was based on the presence of cnf1/cnf2 genes (Tóth et al. 2003), along with screening for the f17 adhesin gene, which is commonly associated with pathogenic strains (Bertin et al. 1996). Furthermore, putative virulence factors, including iha (IrgA homologue adhesion), efa1 (E. coli factor for adherence) and toxB, were investigated in EHEC, STEC and EPEC isolates (Tatarczak et al. 2005). The used primers’ details and their thermal conditions are summarized in Table 2.

TABLE 2.

Details of primer sequences and thermal conditions used in the present study.

		PCR thermal condition
Gene	Primer Sequence (5′ → 3′)	Initial denaturation	Denaturation	Annealing	Extension	Final Extension	Product size (bp)	Concentration in PCR (µM)	Reference
E. coli universal	Eco2083: GCT TGA CAC TGA ACA TTG AG Eco2745: GCA CTT ATC TCT TCC GCA TT	94°C for 7 min	94°C for 45 s	57°C for 50 s	72°C for 2 min	72°C for 10 min	662	0.5	Riffon et al. (2001)
Sta	GCTAATGTTGGCAATTTTTATTTCTGTA AGGATTACAACAAAGTTCACAGCAGTAA		94°C for 30 s	50°C for 45 s	70°C for 90 s	70°C for 10 min	190	0.2	Franck et al. (1998)
			25 cycles
f5	TATTATCTTAGGTGGTATGG GGTATCCTTTAGCAGCAGTATTTC				314	0.1
f41	GCATCAGCGGCAGTATCT GTCCCTAGCTCAGTATTATCACCT				380	0.1
stII	AAA GGA GAG CTT CGT CAC ATT TT AAT GTC CGT CTT GCG TTA GGA C		94°C for 90 s	60°C for 30 s	72°C for 90 s	70°C for 10 min	129	0.2	Vidal et al. (2005)
			35 cycles
stx1	CAG TTA ATG TGG TGG CGA AGG CAC CAG ACA ATG TAA CCG CTG				348	0.5
stx2	ATC CTA TTC CCG GGA GTT TAC G GCG TCA TCG TAT ACA CAG GAG C				584	0.5
eae	TGCGGCACAACAGGCGGCGA CGGTCGCCGCACCAGGATTC		94°C for 90 s	58°C for 55 s	72°C for 90 s	72°C for 10 min	629	0.1	Islam et al. (2007)
			30 cycles
saa	CGTGATGAACAGGCTATTGC ATGGACATGCCTGTGGCAAC		94°C for 45 s	53°C for 50 s	72°C for 90 s	72°C for 7 s	119	0.1	Jenkins et al. (2003)
			30 cycles
bfpA	AATGGTGCTTGCGCTTGCTGC GCCGCTTTATCCAACCTGGTA		94°C for 45 s	56°C for 60 s	72°C for 120 s	72°C for 10 min	326	0.2	Gunzburg et al. (1995)
			29 cycles
astA	CCATCAACACAGTATATCCGA GGTCGCGAGTGACGGCT TTGT		95°C for 30 s	55°C for 120 s	72°C for 120 s	72°C for 5 min	111	0.2	Yamamoto et al. (1997)
			30 cycles
aafA	TGCGATTGCTACTTTATTAT ATTGACCGTGATTGGCTTCC		94°C for 30 s	54°C for 40 s	72°C for 60 s	72°C for 10 min	242	0.05	Tokuda et al. (2010)
			30 cycles
pet	ACTGGCGGACTCATTGCTGT GCGTTTTTCCGTTCCCTATT				832	0.2
aggR	TTAGTCTTCTATCTAGGG AAATTAATTCCGGCATGG				457	0.1
cnf1	GGGGGAAGTACAGAAGAATTA TTGCCGTCCACTCTCACCAGT		94°C for 60 s	55°C for 60 s	72°C for 60 s	72°C for 10 min	1111	0.1	Tóth et al. (2003)
cnf2	TATCATACGGCAGGAGGAAGCACC GTCACAATAGACAATAATTTTCCG		30 cycles				1240	0.5
F17	GCAGAAAATTCAATTTATCCTTGG CTGATAAGCGATGGTGTAATTAAC		94°C for 120 s	59°C for 60 s	72°C for 60 s	72°C for 4 min	537	0.2	Bertin et al. (1996)
			35 cycles
iha	CAGTTCAGTTTCGCATTCACC GTATGGCTCTGATGCGATG		94°C for 30 s	56°C for 60 s	72°C for 60 s	72°C for 5 min	1305	0.1	Tatarczak et al. (2005)
efa1	GAGACTGCCAGAGAAAG GGTATTGTTGCATGTTCAG						479	0.1
toxB	ATACCTACCTGCTCTGGATTGA TTCTTACCTGATCTGATGCAGC		30 cycles				602	0.1

All reactions were performed using a Bio‐Rad thermocycler (T100, USA). DNA‐free distilled water was used as the negative control for all reactions. Positive controls included CAPM6006, CAPM5933, O159:H20, O157:K88ac:H19, EAEC O42, E2348/69 and EDL 933 E. coli strains, purchased from Iranian Biological Resource Center, Iranian Research Organization for Science and Technology and Faculty of Veterinary Medicine in Tehran University. All amplicons were resolved on a 1.5% agarose gel stained with SYBR Safe (Invitrogen).

2.3. Serotyping

The isolates positive for virulence genes were serogrouped for the O antigen using a slide agglutination test (Thermo Scientific Remel Agglutination Sera, USA). O157 serogroup was further screened for the H7 antigen (Thermo Scientific Remel E. coli O157:H7 Latex test, USA).

2.4. Antibiotic Susceptibility Testing

Pure cultures of the isolates, harbouring virulence genes, on Muller Hinton agar (Merck, Germany) were evaluated for antimicrobial susceptibility using the standard agar disk diffusion method (Clinical and Laboratory Standards Institute [CLSI] 2018). Antibiotic disks (Padtanteb, Iran) included tetracycline (TE) (30 µg), erythromycin (15 µg), penicillin (10 µg), streptomycin (10 µg), gentamicin (10 µg), chloramphenicol (30 µg), trimethoprim–sulphamethoxazole (SXT) (1.25 + 23.75 µg), neomycin (30 µg), lincomycin (2 µg), nitrofurantoin (300 µg), ciprofloxacin (5 µg) and cefotaxime (30 µg). The antibiotics were chosen on the basis of the most common ones used in public and veterinary sectors in the study region. E. coli ATCC 25922 was used as the quality control organism in the antibiogram test.

2.5. Statistical Analysis

Statistical relationships between the prevalence of E. coli with variables (age, sex and season) and between the frequency of pathotypes with the same items (in addition to antimicrobial resistance) were evaluated using the chi‐square and Fisher's exact tests (SPSS software, version 21). A p ≤ 0.05 was considered statistically significant.

3. Results

3.1. Isolation of the Bacteria in Relation to the Demographic Information

Of the 350 diarrhoeal faecal samples, 307 (87.71%) E. coli isolates were recovered, of which 235 (68.14%) harboured at least one of the studied virulence genes. The association of E. coli with NCD in heifers was slightly higher than that in bull calves (163 vs. 144), but that is not statistically significant (p = 0.07). The distribution of E. coli in the categorized age groups was different, with p < 0.001, dispersing the most in the first, second and fourth weeks of life, respectively. Table 3 presents the seasonal distribution of five types of E. coli (ETEC, EPEC, EHEC, STEC and NTEC) in bulls and heifers across the four seasons (spring, summer, fall and winter). ETEC was found in all seasons, with the highest counts observed in winter (seven bulls and seven heifers). The lowest was in summer (two bulls and none in heifers). EPEC was found in all seasons, peaking in winter heifers (14) and spring bulls (11); lowest in summer heifers (0). EHEC has a very high prevalence in spring bulls (17) and winter heifers (15); lower counts in fall (6 in bulls and 10 in heifers). STEC shows seasonal variations; high in summer bulls (9) and winter heifers (9). It was completely absent in spring and winter bulls (0). NTEC is highest in fall bulls (10); very low or absent in heifers, except in winter (3) and spring (2) (Figure 1 and Table 3). Considering season as an effective criterion for E. coli distribution in NCD (p = 0.009), spring and winter were denoted as the most involved bacteria (Figure 2 and Tables 4a, 4b, 4c).

TABLE 3.

Chi‐square test results for the prevalence of Escherichia coli with variables (age, sex and season).

Test	χ 2	p
Overall distribution	97.291	0.001
Age groups	174.372	0.001
Gender	11.6623	0.07
Season	35.174	0.009

FIGURE 1.

The frequency of Escherichia coli pathotypes regarding sex within each season. EHEC, enterohaemorrhagic Escherichia coli; EPEC, enteropathogenic Escherichia coli; ETEC, enterotoxigenic Escherichia coli; NTEC, necrotoxigenic Escherichia coli; STEC, Shiga‐toxigenic Escherichia coli.

FIGURE 2.

Weekly distribution of Escherichia coli pathotypes. EHEC, enterohaemorrhagic Escherichia coli; EPEC, enteropathogenic Escherichia coli; ETEC, enterotoxigenic Escherichia coli; NTEC, necrotoxigenic Escherichia coli; STEC, Shiga‐toxigenic Escherichia coli.

TABLE 4a.

The prevalence of the frequency of Escherichia coli with variable season and its 95% confidence intervals.

Confidence interval 95%
Pathotype	Spring	Summer	Fall	Winter	Spring	Summer	Fall	Winter
ETEC	9	2	9	14	4–15	0–7	4–15	8–20
EPEC	21	3	7	17	14–28	01‐Sep	3–14	11–24
EHEC	31	16	16	23	22–41	10–25	10–25	15–32
STEC	3	9	8	9	01‐Aug	5–15	4–14	5–15
NTEC	6	5	11	5	02‐Dec	02‐Oct	6–17	02‐Oct
EAST‐1	5	1	1	4	02‐Aug	0–5	0–5	01‐Aug
ntEC	18	20	17	17	11–26	13–29	11–25	11–25

Abbreviations: EHEC, enterohaemorrhagic Escherichia coli; EPEC, enteropathogenic Escherichia coli; ETEC, enterotoxigenic Escherichia coli; NTEC, necrotoxigenic Escherichia coli; STEC, Shiga‐toxigenic Escherichia coli.

TABLE 4b.

The prevalence of the frequency of Escherichia coli with variable sex and its 95% confidence intervals.

Confidence interval 95%
Pathotype	Bull	Heifer	Bull	Heifer
ETEC	17	17	11–23	11–23
EPEC	18	30	12–25	23–36
EHEC	42	44	33–51	35–53
STEC	13	16	8–19	10–21
NTEC	19	8	13–23	4–14
EAST‐1	2	9	0–6	05‐Nov
ntEC	33	39	25–42	30–47

Abbreviations: EHEC, enterohaemorrhagic Escherichia coli; EPEC, enteropathogenic Escherichia coli; ETEC, enterotoxigenic Escherichia coli; NTEC, necrotoxigenic Escherichia coli; STEC, Shiga‐toxigenic Escherichia coli.

TABLE 4c.

The prevalence of the frequency of Escherichia coli with variable age and its 95% confidence intervals.

Confidence interval 95%
Age	ETEC	EPEC	EHEC	STEC	NTEC	EAST‐1	ntEC	ETEC	EPEC	EHEC	STEC	NTEC	EAST‐1	ntEC
1	27	4	5	1	3	0	8	20–34	01‐Oct	0–10	0–6	01‐Sep	0–4	2–14
2	5	8	5	3	7	2	12	0–10	3–13	0–10	01‐Aug	02‐Dec	0–7	6–18
3	2	5	6	4	3	0	11	0–7	0–10	01‐Nov	01‐Sep	01‐Aug	0–4	5–17
4	0	8	17	6	3	3	5	0–4	3–13	10–24	01‐Nov	01‐Aug	01‐Aug	0–10
5	0	4	17	3	3	0	6	0–4	01‐Oct	11–23	01‐Aug	01‐Aug	0–4	01‐Nov
6	0	7	14	1	2	0	11	0–4	02‐Dec	8–20	0–6	0–7	0–4	5–17
7	0	4	6	3	2	3	6	0–4	01‐Sep	01‐Nov	01‐Aug	0–5	01‐Aug	01‐Nov
8	0	6	10	7	5	1	9	0–4	01‐Nov	4–16	02‐Dec	0–10	0–6	3–15
9	0	2	6	1	0	2	4	0–4	0–6	02‐Oct	0–5	0–4	0–6	01‐Aug

Abbreviations: EHEC, enterohaemorrhagic Escherichia coli; EPEC, enteropathogenic Escherichia coli; ETEC, enterotoxigenic Escherichia coli; NTEC, necrotoxigenic Escherichia coli; STEC, Shiga‐toxigenic Escherichia coli.

3.2. Distribution of Pathotypes and Virulence Genes

The overall prevalence of the pathotypes was as follows: EHEC, 86 (36.59%); EPEC, 48 (20.42%); ETEC, 34 (14.46%); STEC, 29 (12.34%) and NTEC, 27 (11.48%). Because of the lack of any additional virulence genes, including aaf1, aggR and pet, 11 (4.67%) astA‐positive isolates were not classified as EAEC. Instead, these isolates were identified as EAST‐1‐positive (Coura et al. 2017). Moreover, five isolates carrying F5 and stx genes and one isolate harbouring stx1 and sta genes, simultaneously, were detected as ETEC/EHEC hybrids. Additionally, no virulence genes were detected in the 72 E. coli isolates (Table 5).

TABLE 5.

The frequency of virulence genes in Escherichia coli pathotypes isolated in the present study.

Pathotype n (%)	Genotype	No. of isolates
ETEC	sta, F5	5
34 (14.46%)	sta, F41	7
	sta, F5, F41	11
	stII, F5	4
	stII, F5, F41	7
EPEC	Eae	11
48 (20.42%)	eae, iha	6
	eae, efa1	2
	eae, iha, efa1	6
	eae, iha, toxB	4
	eae, efa1, toxB	2
	eae, iha, efa1, toxB	2
	eae, iha, bfpA	7
	eae, efa1, toxB, bfpA	5
	eae, iha, toxB, astA	2
	eae, toxB, astA, pet, aggR	1
EHEC	stx1, eae	3
86 (36.59%)	stx1, hly	8
	stx1, sta	1
	stx1, eae, hly	2
	stx1, eae, efa1	1
	stx1, eae, iha	7
	stx1, eae, efa1, toxB	4
	stx1, eae, iha, toxB	1
	stx1, eae, efa1, iha, toxB	2
	stx1, eae, astA, aggR	7
	stx1, eae, hly, efa1, iha	3
	stx1, eae, hly, iha, toxB	1
	stx1, eae, hly, efa1, iha, toxB	8
	stx2, eae	6
	stx2, eae, hly	2
	stx2, eae, astA, aggR, pet	4
	stx2, eae, efa1	1
	stx2, eae, iha	6
	stx2, eae, efa1, iha, toxB	11
	stx1, stx2, eae	3
	stx1, stx2, eae, hly	1
	stx1, stx2, eae, astA	2
	stx1, stx2, eae, efa1, iha	2
	stx1, stx2, eae, efa1, iha, toxB	1
STEC	stx1, hly	5
29 (12.34%)	stx1, hly, saa	3
	stx1, hly, saa, F5	1
	stx1, hly, efa1	2
	stx1, efa1, F17, F5	1
	stx1, efa1, iha	2
	stx2, saa	4
	stx2, hly, efa1, toxB	1
	stx1, iha, astA, pet	1
	stx1, stx2, hly	4
	stx1, stx2, hly, F5	3
	stx1, stx2, saa, efa1	1
	stx1, stx2, saa, efa1, iha	1
NTEC	cnf1	4
27 (11.48%)	cnf1, F17	1
	cnf1, α‐hly, F17	2
	cnf2	12
	cnf2, F17	1
	cnf1, cnf2, F17	5
	cnf1, cnf2, α‐hly, F17	2
EAST‐1	astA	1
11 (4.68%)	astA, F5	3
	astA, F17	7

Abbreviations: EHEC, enterohaemorrhagic Escherichia coli; EPEC, enteropathogenic Escherichia coli; ETEC, enterotoxigenic Escherichia coli; NTEC, necrotoxigenic Escherichia coli; STEC, Shiga‐toxigenic Escherichia coli.

The frequencies of virulence genes included 126 eae, 81 stx1, 58 efa1, 62 iha, 53 stx2, 45 toxB, 43 hly, 35 F5, 28 astA, 25 F41, 24 sta, 20 cnf2, 19 F17, 14 cnf1, 12 aggR, 11 stII, 10 saa, 6 pet and 4 α‐hly. The total number of isolated E. coli pathotypes based on their virulence genes is presented in detail in Table 5.

3.3. Distribution of O‐Serogroups

Serotyping of the E. coli isolates carrying virulence factors segregated them into 14 serotypes. O103 (53, 22.55%) and O15 (36, 15.31%) were the most prevalent serotypes, followed by O86 (32, 13.61%), O128 (24, 10.21%) and O119 (18, 7.65%). Other identified serotypes were represented as O121 (14, 5.95%), O127 (14, 5.95%), O145 (11, 4.68%), O26 (9, 3.82%), O157 (7, 2.97%), O45 (7, 2.957%), O113 (6, 2.55%), O111 (3, 1.27%) and O141 (1, 0.42%). Among the seven O157 strains, four isolates (57.14%) belonged to serotype O157:H7, whereas three (42.85%) isolates were non‐H7. The distribution of serotypes among the pathotypes is displayed in Table 6 and 7).

TABLE 6.

Distribution of Escherichia coli serotypes among the pathotypes.

	ETEC	EPEC	EHEC	STEC	NTEC	EAST‐1
O103	8	17	12	14	—	2
O15	10	13	6	—	7	—
O86	—	11	21	—	—	—
O128	2	—	17	—	5	—
O119	4	—	9	—	5	—
O121	—	—	9	—	5	—
O127	6	3	3	2	—	—
O145	—	—	2	—	5	4
O26	3	—	3	—	—	3
O157	—	—	—	7	—	—
O45	—	3	2	—	—	2
O113	—	—	—	6	—	—
O111	1	—	2	—	—	—
O141	—	1	—	—	—	—

Abbreviations: EHEC, enterohaemorrhagic Escherichia coli; EPEC, enteropathogenic Escherichia coli; ETEC, enterotoxigenic Escherichia coli; NTEC, necrotoxigenic Escherichia coli; STEC, Shiga‐toxigenic Escherichia coli.

TABLE 7.

Antimicrobial resistance profile of the Escherichia coli pathotypes isolated from diarrhoeic calves.

	TET30	ERY15	PEN10	STR10	GEN10	CHL10	SXT25	NEO30	LIN2	NIT300	CIP5	CTX30
ETEC (34)	33	31	34	29	23	25	23	21	25	18	29	22
EPEC (48)	45	46	48	46	39	39	34	32	33	25	33	37
EHEC (86)	82	80	85	81	75	73	66	57	73	61	76	72
STEC (29)	26	26	29	26	23	19	18	19	20	22	19	21
NTEC (27)	27	24	27	25	19	21	18	17	18	14	21	18
EAST‐1 (11)	9	9	10	11	10	11	7	7	6	6	10	11
Total (235)	222 (94.46%)	216 (91.91%)	233 (99.14%)	220 (93.61%)	189 (80.42%)	188 (80.00%)	166 (70.63%)	153 (65.10%)	175 (74.46%)	146 (62.12%)	188 (80.00%)	181 (77.02%)

Note: TET30: tetracycline, ERY15: erythromycin, PEN10: penicillin, STR10: streptomycin, GEN10: gentamicin, CHL10: chloramphenicol, SXT25: trimethoprim–sulphamethoxazole, NEO30: neomycin, LIN2: lincomycin, NIT300: nitrofurantoin, CIP5: ciprofloxacin, CTX30: cefotaxime.

Abbreviations: EHEC, enterohaemorrhagic Escherichia coli; EPEC, enteropathogenic Escherichia coli; ETEC, enterotoxigenic Escherichia coli; NTEC, necrotoxigenic Escherichia coli; STEC, Shiga‐toxigenic Escherichia coli.

3.4. Antimicrobial Resistance Patterns

The antimicrobial patterns of E. coli pathotypes isolated from diarrhoeic calves in the present study are presented in Table 7. Multidrug resistance (MDR), resistance to at least three antibiotic classes, was observed in 207 (88.08%) out of 235 E. coli isolates. In general, 18 distinct antibiotic resistance patterns were perceived, among which 15 were MDR (Table 8). Diverse degrees of resistance and sensitivity were observed against the tested antibiotics in the different pathotypes. Overall, the highest rates of resistance were observed against penicillin (99.14%), TE (94.46%), streptomycin (93.61%) and erythromycin (91.91%). The most frequent sensitivity rates were observed for nitrofurantoin (37.88%), neomycin (34.90%) and trimethoprim–SXT (29.37%).

TABLE 8.

The antimicrobial resistance patterns of Escherichia coli isolated from diarrhoeic calves.

Antimicrobial resistance pattern	Multidrug resistance pattern
PEN10/TET30	−
PEN10/ERY15	−
PEN10/STR10	−
PEN10/STR10/TET30	+
TET30/ERY15/STR10	+
PEN10/ERY15/CTX30	+
PEN10/TET30/GEN10/SXT25	+
NEO30/NIT300/CTX30/SXT25	+
PEN10/TET30/STR10/GEN10	+
PEN10/ERY15/CTX30/LIN2	+
PEN10/TET30/STR10/ERY15/CXT30	+
PEN10/TET30/GEN10/CHL10/CTX30	+
PEN10/TET30/STR10/ERY15/CHL10	+
PEN10/ERY15/CHL10/NEO30/CIP5/NIT300	+
PEN10/TET30/ERY15/STR10/NEO30/GEN10/CHL10	+
PEN10/TET30/ERY15/STR10/GEN10/LIN2/CIP5	+
PEN10/TET30/STR10/SXT25/LIN2/CHL10/CTX30	+
PEN10/ERY15/STR10/NEO30/CHL10/GEN10/CTX30	+

3.5. Correlation Among Different Variables

Furthermore, no statistical association was observed between diarrhoeagenic E. coli and gender (p = 0.051), diarrhoeagenic E. coli and age (p = 0.511) and diarrhoeagenic E. coli and season (p = 0.148). No sex factors associated with the presence of ETEC, EPEC, EHEC and STEC were detected (p > 0.05), whereas there was a significant relationship between sex and NTEC (p = 0.013) and sex and EAST‐1 (p = 0.023). A statistically significant correlation was observed between age groups and ETEC and EHEC (p = 0.001). Statistical analysis revealed no association between the season and ETEC, EHEC, NTEC and EAST‐1 (p > 0.05). Similarly, a significant association was found between the incidence of season and EPEC (p = 0.036) and STEC (p = 0.017). Finally, the incidence of pathotypes resistant to penicillin (p = 0.004), lincomycin (p = 0.037) and ciprofloxacin (p = 0.026) was statistically significant.

4. Discussion

NCD is a devastating disease that causes profound economic losses in veterinary medicine. This study presents the frequency, putative genotypes and antimicrobial resistance patterns of E. coli pathotypes involved in NCD in Iran. There was nearly concordance between the overall prevalence of E. coli in NCD in the current study and elsewhere (Osman et al. 2013). Higher and lower prevalence rates have been reported in Iran (76.45%) (Shahrani et al. 2014) and Ethiopia (36.8%) (Gebregiorgis and Tessema 2016), respectively. Geographical, management and hygienic conditions may influence the final distribution of E. coli in NCD (Cho et al. 2014). Although it is stated that high temperature is the most effective meteorological condition affecting E. coli‐originating NCD (Gautam et al. 2011), the lowest prevalence of E. coli isolates in the present study was detected in summer. This is consistent with results reported in Iran (Shahrani et al. 2014). The high prevalence of E. coli involved in NCD in cold seasons can be attributed to climate and atmospheric fluctuations, which may affect the immune system and therefore calves’ sensitivity to infections (Shahrani et al. 2014). Notably, Naderi et al. (2024) did not report seasonal variations in pathotype prevalence, suggesting contextual differences in environmental influences. In addition, the higher prevalence of E. coli in NCD in winter may be attributed to the lower mean serum IgG1 concentrations in winter‐borne calves (Norheim et al. 1985). Hence, complementary research is needed to gauge seasonal shedding of pathogenic E. coli. Regarding sex variability, heifer calves yielded a subtler distribution of E. coli in the NCD. Although a higher prevalence rate of E. coli in NCD in female calves has been reported in Ethiopia (Gebregiorgis and Tessema 2016), the difference in frequency rates in both sexes was high, which indicates a discrepancy with the results of the present study. Male calves are explicitly conniving to get enough attention and management care because of their irrelevant role, especially in replacement stock (Gebregiorgis and Tessema 2016).

A total of 72 E. coli isolates did not possess any of the virulence factors studied. A plausible explanation for this is that non‐pathogenic E. coli strains are part of the normal gut microbiota. Therefore, the role of other pathogens in NCD should not be elucidated. Bartels et al. (2010) assessed enteropathogens in the faeces of young Dutch dairy calves. The proportions of identified pathogens were 2.6% and 3.1% for coronavirus and 17.7%, 27.8% and 54% for E. coli, coronavirus, rotavirus, Cryptosporidium parvum and Clostridium perfringens, respectively.

The highest pathogenic E. coli distribution among the age groups was in 1–7‐day‐old diarrhoeic calves (15.63%), and the most common pathotype was ETEC, harbouring the sta, F5 and F41 genes as the predominant genotype. Calves are consistently documented to be susceptible to ETEC in this age group (Ok et al. 2009; Younis et al. 2009; Nguyen et al. 2011). This contrasts with Naderi et al. (2024), who reported ETEC as a minor pathotype in diarrhoeic calves but noted its prominence in hybrid strains (e.g., STEC/ETEC). Because of the lack of vaccination programmes to prevent NCD in Iran, optimized reception of colostrum and navel disinfection trials must be emphasized. Interestingly, the simultaneous existence of F5 and stx genes was detected in five isolates as a newly emerging phenotype causing NCD, which has been previously reported in Vietnam (Nguyen et al. 2011) and Brazil (Andrade et al. 2012). This aligns with Naderi et al. (2024), who identified stx2/ST1 (STEC/ETEC) as a prevalent hybrid profile. Despite the frequent association of the F41 fimbrial antigen with the F5 antigen in ETEC, 11 isolates were identified as carrying F41 alone. Although this is an occasional finding, these isolates can cause NCD (Andrade et al. 2012).

STEC is regularly reported as an etiological agent of NCD (Coura et al. 2017; Shahrani et al. 2014; Nguyen et al. 2011; Andrade et al. 2012), and these animals may act as reservoirs for humans (Akiyama et al. 2017). The overall prevalence of STEC was 12.34%, with stx1 + hly being the predominant genotype. Naderi et al. (2024) similarly identified STEC as dominant but reported higher prevalence in healthy calves, contrasting our diarrhoeic‐focused findings. This is significantly lower than the results reported for Vietnam (51%) (Nguyen et al. 2011) and Brazil (66.7%) (Andrade et al. 2012). The close frequency of stx1 and stx2 in our study is inconsistent with the higher prevalence of stx1 than stx2 previously reported in Austria (Herrera‐Luna et al. 2009), Iran (Badouei et al. 2010) and Brazil (Andrade et al. 2012). Conversely, a higher occurrence of stx2 has been observed in other studies conducted in Brazil (Farah et al. 2007; Timm et al. 2007). In addition, atypical coexistence of stx1 + sta was detected in one isolate. A possible explanation is the mobile genetic elements caused by the transmission of plasmid‐harbouring enterotoxin genes or bacteriophage‐encoding STX genes (Kaper et al. 2004). An adhesin factor, encoded by the saa gene, was detected in 10 STEC isolates. Bardiau et al. (2010) reported the sole existence of this gene in eae‐negative E. coli strains. Although it has been documented that the saa gene is associated with hly (Paton et al. 2001; Paton and Paton. 2002), in the present study, six isolates carried saa in the absence of hly. This is in agreement with results obtained in Vietnam (Nguyen et al. 2011). Hly coded enterohaemolysin is frequently detected in STEC isolated from calve (Coura et al. 2017; Shahrani et al. 2014; Salvadori et al. 2003), although its role in E. coli pathogenesis in NCD is unknown (Nguyen et al. 2011). In our study, the frequency of hly‐positive STEC isolates was 65.51%, which was higher than the prevalence reported in Brazil (30.30%) (Salvadori et al. 2003) and Vietnam (51%) (Nguyen et al. 2011). In contrast, the presence of the hly gene in STEC isolated from both diarrhoeic and healthy calves in Brazil is nil (Coura et al. 2017). The distribution of the O157:H7 STEC strain was 1.70% (4 235 E. coli isolates). This strain is of concern to human welfare because it is the causative agent of intense diseases such as HC and haemolytic uremic disorder (HUS) (Thomas et al. 2017). Other STEC serotypes were O103 (14 isolates), O113 (six isolates), O127 (two isolates) and O157: non‐H7 (three isolates). All these are among the serotypes involved in severe human illnesses (Espie et al. 2008). Other adhesion genes, including efa1 and iha, were detected in seven and four STEC isolates, respectively. Although it has been suggested that toxB is found in eae‐positive E. coli strains, we could detect this gene in one STEC isolate with a stx2, hly, efa1 and toxB profile. stx1 and hly were the predominant STEC virulence profiles in the present study, whereas other genotypes, including adhesion genes, were demonstrated to be the most common genotypes in STEC strains of bovine origin. Example reported from Brazil included stx2, iha and stx, as well as hly, iha, and saa (Coura et al. 2017; Oliveira et al. 2008). F5, an important colonization antigen found primarily in ETEC, was detected in the five STEC strains. This was comparable to the frequency reported by Salvadori et al. (2003) in Brazil.

F5, an important colonization antigen found primarily in ETEC, was detected in 5 STEC strains. This is comparable to the frequency reported by Salvadori et al. (2003) from Brazil.

EHEC, the most common pathotype in the present study, rendered stx2, eae, efa1, iha and toxB the most prevalent genotypes. Coura et al. (2017) identified the genetic profiles of stx1, eae, e‐hly, efa‐1, iha and toxB as the predominant EHEC genotypes. All EHEC strains, except for eight isolates, possessed the eae gene. The association of eae with both stx genes in nearly equal proportions is in contrast with the results revealing the sole or higher frequency of eae/stx1 in diarrhoeic (Coura et al. 2017; Badouei et al. 2010; Salvadori et al. 2003) and non‐diarrhoeic (Coura et al. 2017; Andrade et al. 2012; Guler et al. 2008) calves. In the present study, the hly‐positive strains comprised 29.06% of the EHEC isolates, in comparison with the higher frequency of this gene observed in STEC isolates (65.51%). The overall frequency of the hly genes in stx‐harbouring strains (STEC and EHEC) was 32.7%. A higher association between hly and Stx‐producing strains in cattle has been reported in Vietnam (50.28%) (Nguyen et al. 2011) and Brazil (66.99%) (Coura et al. 2017). The astA gene, which encodes an additional virulence factor, was detected in 13 EHEC and one STEC strain. The accompaniment of accessory genes with astA was aggR and aggR/pet in seven and four EHEC isolates, respectively, and pets in one STEC isolate. Eleven E. coli isolates possessing astA in the absence of aggR, pet and aafA genes were categorized as none of the pathotypes (Coura et al. 2017). Coura et al. (2017) depicted three and one astA‐positive EHEC strains isolated from diarrhoeic and healthy calves, respectively. Cattle are regularly documented reservoirs of EHEC and STEC in humans (Akiyama et al. 2017; Venegas‐Vargas et al. 2016). Our results highlight the public threat imposed by cattle in this area.

The EPEC pathotype was detected at a distribution of 20.42%. This coincides with other studies conducted in Vietnam (Nguyen et al. 2011), Turkey (Guler et al. 2008) and Brazil (Coura et al. 2017; Andrade et al. 2012), which indicate a low frequency of EPEC in calves. Because EPEC isolates are either healthy or diarrhoeic calves, the magnitude of this pathogen in NCD is contestable (Foster and Smith 2009). Alternatively, this pathotype is deleterious to infantile diarrhoea in human (Croxen et al. 2013). Twelve EPEC isolates carried the bfpA gene and were classified as typical strains. Both typical and atypical strains are associated with diarrhoea in children (Croxen et al. 2013). All EPEC strains detected by Coura et al. (2017) were atypical.

The distribution of cnf1 and cnf2 among the NTEC strains was seven and 13, respectively. Seven NTEC isolates harboured both cnf1 and cnf2. Although CNF2‐positive E. coli is a causative agent of septicaemia and diarrhoea in calves (van Bost et al. 2001), no association was observed between CNF2 and diarrhoea syndrome in another study (Coura et al. 2017). There was concordance between the low association of α‐hly and cnf genes in our study and that reported by Salvadori et al. (2003).

There is disparity in the distribution of E. coli serotypes isolated from calf scours in different studies. Serotypes O103 and O15 represented the most frequent serotypes in our study, whereas O111 and O141 were the least common. The most and least prevalent serotypes involved in NCD in a previous Iranian study were O26/O157 and O113/O121, respectively (Nguyen et al. 2011). Nguyen et al. (2011) reported O15, O20, O103 and O157 as the most frequent serotypes isolated from calf diarrhoea in Vietnam, in comparison with the most frequent of O118, O11, O26 and O111 retrieved from Brazil (Salvadori et al. 2003).

Various studies have underscored the development of resistant bacteria in recent years (Nguyen et al. 2011; Herrera‐Luna et al. 2009; de Verdier et al. 2012). The misuse of antibiotics is discriminated against this trend in the dairy and public sectors. Surprisingly, a high distribution of antimicrobial resistance was observed in the present study. Naderi et al. (2024) similarly reported high resistance to trimethoprim–SXT and TE, with SXT resistance at 52.77% in healthy calves versus our 49.01% overall—highlighting consistent misuse patterns in Iran. Nonetheless, it is better to assess the antimicrobial resistance/susceptibility of intestinal bacteria in randomly selected cultures of intestinal contents (de Verdier et al. 2012). Despite previous studies that indicate a higher frequency of resistance patterns in commonly used antibiotics (Herrera‐Luna et al. 2009; Hariharan et al. 2004), approximately all of the antibiotic resistance rates were high. There is a dilemma as to whether to prescribe antibiotics in cases of NCD, as it is recommended not to administer antibiotics in NCD in the absence of a systemic malady (Constable 2004).

5. Conclusion

The results of the present study indicated that calves may serve as potential reservoirs of various E. coli pathotypes with diverse genotypic and phenotypic traits in terms of virulence factors and antibiotic resistance. The identification of important O‐serogroups and the high rates of antimicrobial resistance among the isolates pose a significant public health threat, as these traits may enhance pathogenicity, environmental persistence and resistance to standard treatments. Therefore, the results underscore the need for continuous monitoring, strict biosecurity measures in cattle farms and proper antibiotic stewardship to prevent the transmission of pathogenic and resistant strains to humans.

Author Contributions

The authors’ respective contributions to the article are as follows: Elham Ahmadi: study design, supervision and data collection. Mohammad Sadegh Safaee Firouzabadi: analysis, editing and interpretation of results. Seyed Ali Mousavi Rad: performed data analysis and formal analysis, writing – original draft. Samira Ghorbani: performed writing – reviews and editing. All authors have reviewed the results and approved the final version of the manuscript.

Funding

The authors have nothing to report.

Ethics Statement

The authors confirm that the ethical policies of the journal, as noted on the journal's author guidelines page, have been adhered to and the appropriate ethical review committee (State Committee on Animal Ethics, Ardakan University, Ardakan, Iran) approval has been received (IR.Ardakan.REC.1404.066). The authors confirm that they have followed EU standards for the protection of animals used for scientific purposes.

Conflicts of Interest

The authors declare no conflicts of interest.

Ahmadi, E. , Firouzabadi M. S. S., Rad S. A. M., and Ghorbani S.. 2026. “Molecular Characterization of Virulent Genes and Antimicrobial Resistance Patterns of Escherichia coli Isolated From Calf Scours in Western Iran.” Veterinary Medicine and Science 12, no. 1: e70747. 10.1002/vms3.70747

Data Availability Statement

The data sets generated during the present study are available from the corresponding author upon reasonable request.