Isolation, identification, molecular typing, and drug resistance of Escherichia coli from infected cattle and sheep in Xinjiang, China

Abstract Background Escherichia coli infections are common in Xinjiang, a major region of cattle and sheep breeding in China. Therefore, strategies are required to control E. coli. The aim of this study was to investigate the phylogenetic groups, virulence genes, and antibiotic resistance characteristics of E. coli isolates. Methods In this study, 116 tissue samples were collected from the organs of cattle and sheep that were suspected of having E. coli infections between 2015 and 2019. Bacteria in the samples were identified using a biochemical identification system and amplification of 16S rRNA, and the phylogenetic groupings of E. coli isolates were determined by multiplex polymerase chain reactions. In addition, PCR detection and analysis of virulence factors, antibiotic resistance genes, and drug‐resistant phenotypes of E. coli isolates were performed. Results A total of 116 pathogenic E. coli strains belonging to seven phylogenetic groups were isolated, with the majority of isolates in groups A and B1. Among the virulence genes, curli‐encoding crl had the highest detection rate of 97.4%, followed by hemolysin‐encoding hlyE with the detection rate of 94.82%. Antimicrobial susceptibility test results indicated that the isolates had the highest rates of resistance against streptomycin (81.9%). Conclusion These characteristics complicate the prevention and treatment of E. coli‐related diseases in Xinjiang.


INTRODUCTION
Escherichia coli isolates can be classified based on their virulence factors as intestinal pathogenic E. coli, extraintestinal pathogenic E. coli (ExPEC), or symbiotic E. coli Wirth et al., 2006).
Pathogenic E. coli, a key zoonotic pathogen, is responsible for more than 160 million cases of dysentery and 1 million deaths every year (Guan & Li, 2007). Currently, E. coli strains are assigned to eight phylogenetic groups of isolates with different pathogenicities based on genetic and evolutionary characteristics: A, B1, B2, C, D, E, F, and clade I (Bok et al., 2020;Clermont et al., 2013;Ksiezarek et al., 2021). The pathogenicity of E. coli is determined by regulatory and interactive effects among various virulence factors and is affected by host type, host health, and interactions with other bacteria. Excrements of cattle and sheep commonly contain both pathogenic and non-pathogenic E. coli, which enables the rapid transmission of E. coli across livestock populations (Chase-Topping et al., 2012;Busia et al., 2014;Reiss et al., 2006;Sargeant et al., 2004). Other studies have shown that pathogenic E. coli isolates originating from the intestinal tracts of cattle and sheep may infect humans (Irino et al., 2007). Furthermore, selective pressure from the widespread use of antibiotics in livestock breeding has resulted in an increase in bacterial resistance against antibacterial agents, as well as cross-resistance to other agents (Roth et al., 2019;Tasho et al., 2016). Multidrug antibiotic resistance in bacteria is a severe threat to human health (Bai et al., 2021). In 2019, antimicro-

Sample collection and E. coli isolation and identification
Over 4 years (2015-2019), lung, spleen, heart, and lymphoid tissue samples were collected from cattle and sheep that manifested the following symptoms: diarrhoea, elevated body temperature, listlessness, limb weakness, neurological symptoms, sepsis, and evidence of respiratory tract infection. In the sampling process, we collected the fresh tissue using sterile surgical instruments, and then placed them in 10 mL aseptic tubes. The samples were placed on ice packs and later transported to the laboratory within 24 h of collection for bacterial isolation.
The samples were inoculated into a nutrient broth and cultured at 37 • C and 180 rpm for 6-8 h. Subsequently, the cultures were inoculated onto eosin-methylene blue agar, MacConkey agar, and blood agar plates and cultured at 37 • C for 10-15 h. In addition, genes were purified and sequenced using 16S rRNA, verifying the possibility of certain bacterial species (Clarridge, 2004). Suspected E. coli colonies on each agar plate were purified and identified using an automated microbial identification system (VITEK 2 Compact; bioMérieux, France) (Gu et al., 2018

Gene sequencing
Selected PCR amplification products were sent to Rui Biotech Co. Ltd.

Statistical methods
We used the SPSS software for Windows, version 20.0 (SPSS Inc., Chicago, IL, USA) for statistical analysis (Islam et al., 2017). Comparisons among frequencies of occurrence of each phenotypic or genotypic feature in E. coli isolates from samples were carried out by contingency table χ 2 tests (at p < 0.05). The result was considered to be significant at p ≤ 0.05.

Detection of antibiotic resistance genes and antimicrobial susceptibility testing
Among the various aminoglycoside resistance genes detected, aph(3′)-Ia was the most common at 63.8% (74/116) and aadB was the least common at 6.0% (7/116). Among the various β-lactam resistance TA B L E 4 Frequency of different virulence genes among the phylogenetic groups of E. coli isolated from cattle and sheep.   Table 5).
The average carrying rate of drug resistance genes in bovinederived strains was 6.83%, and the average carrying rate of drugresistant genes in sheep-derived strains was 5.5%. The two strains seem to have little difference, the average carrying rate of drug resistance genes in bovine-derived strains was 6.83%, and the average carrying rate of drug-resistant genes in sheep-derived strains was 5.5%. However there is a large difference in the carrying situation between bovine-derived strains, and the maximum carrying rate of 15 drug resistance genes, and a minimum of 1. Among them, tet(C), qnrA and aac(6')-Ib, are unique to bovine-derived strains, but not detected in sheep-derived strains ( Table 6).
All of the 116 E. coli isolates exhibited MDR (defined as resistance towards three or more antibiotics) (Figure 1). In particular, 41.4% of strains were resistant to more than 10 antibiotics. The rates of resistance of the isolates to antibiotics were as follows: streptomycin:

DISCUSSION
Escherichia coli infections are among the three major bacterial infections that are detrimental to livestock breeding around the world.
Escherichia coli can be either symbiotic or pathogenic, and the phylogenetic positions of these bacteria are determined by different gene cluster combinations that are critical for understanding the pathogenesis of E. coli and host-E. coli interactions (Irino et al., 2007). Research on E. coli genomes has indicated that phylogenetic groupings of strains may be related to isolation sources (Clermont et al., 2013). Previous studies have shown that pathogenic E. coli from cattle, non-cattle (Karczmarczyk et al., 2011), and birds (Cunha et al., 2014) mainly belong to groups A and B1. In this study, the majority of the 116 pathogenic E. coli strains isolated from diseased cattle and sheep in Xinjiang also belonged to groups A and B1, and the minority belonged to groups B2, C, D, E, and F.
Escherichia coli is a pathogen of worldwide significance, resulting in diseases such as dysentery (e.g., shigellosis), neonatal meningitis (associated with the presence of a K1 capsular polysaccharide), and haemolytic-uremic syndrome (associated with E. coli O157:H7).
Escherichia coli is regarded as a reservoir of antibiotic resistance determinants. This bacterium readily develops antibiotic resistance and is capable of transferring antibiotic resistance factors to other pathogenic microbes in gastrointestinal tracts (Blake et al., 2003).
Antibiotic resistance rates of E. coli have risen steadily over the years.
In a study by Massot et al. (2016) (Koczura et al., 2013). As shown in Table 6, the number of virulence genes carried by isolates in groups C and F are similar, and isolates in group F carried a larger number of antibiotic resistance genes.
Several combinations of virulence and antibiotic resistance genes were observed and lacked significant patterns, partly limited by the number of strains analyzed. In addition, the number of B2 and D groups in the strains was small and could not be statistically analyzed. The emergence of relationships between antibiotic resistance and virulence genes suggests that resistant isolates are more virulent than susceptible isolates. Because of the resistance, bacteria can persist longer in the host than susceptible bacteria and may therefore be difficult to treat. According to these results, cattle isolates are more virulent than sheep isolates, possibly due to excessive use of anti-microbials without proper diagnosis. These strains were found to be more toxic and resistant to multiple drugs, leading to initiation of several tests related to the development of animal husbandry in Xinjiang.

CONCLUSION
This study revealed that the prevalence of multi-antibiotic-resistance,

CONFLICT OF INTEREST STATEMENT
The authors declare no conflict of interest.

DATA AVAILABILITY STATEMENT
Data sharing not applicable as no new data were generated, and the article describes entirely theoretical research.

ETHICS STATEMENT
All animal experiments were reviewed and approved by the ethics committee of the First Affiliated Hospital of Medical College, Shihezi University, China (approval number: A2019-149-01).