Virulence determinant and antimicrobial resistance traits of Emerging MDR Shiga toxigenic E.coli in diarrheic dogs

Shiga-toxigenic Escherichia coli (STEC) is incriminated in severe hemorrhagic enteritis in dogs, which is considered a veterinary and public health alarm. To investigate the prevalence, antimicrobial resistance patterns, virulence determinants, and distribution of antimicrobial resistance genes in STEC strains isolated from dogs: 80 fecal samples were obtained from diseased dogs suffering from hemorrhagic diarrhea from pet animal clinics in Ismailia governorate, Egypt. The obtained samples were examined bacteriologically. Moreover, the retrieved isolates were tested for serogrouping, Congo-red binding, antimicrobial resistance, and PCR-based determination of virulence and antimicrobial resistance genes. The prevalence of E. coli in the examined diseased dogs was 23.75% (19/80). The serogrouping of the recovered isolates revealed that 84.2% of the tested isolates were distributed into three serogroups: O146 (36.8%), O111 (31.5%), and O26 (15.7%). Meanwhile, three isolates were untypable (15.8%). Moreover, all the tested E. coli serovars were positive for CR-binding. PCR revealed that the prevalence of stx1, eaeA, hlyA, and stx2 virulence genes was 100%, 100%, 100%, and 47.3%, respectively. Our findings revealed that 31.5% of the recovered isolates showed MDR to five antimicrobial classes and harbored blaTEM, blaCTX-M, tetA, tetB, and sul1 genes. Alarmingly, three isolates were carbapenem-resistant. Two strains harbored the blaKPC gene, while one strain carried the blaNDM-1 gene. Concisely, as far as we know, this is the first study that reported the existence of MDR-STEC in dogs in Egypt. The stx1 gene is the most predominant Shiga toxin gene that accompanied the STEC isolated from hemorrhagic enteritis in dogs. The emerging MDR-STEC in dogs commonly harbors blaTEM, blaCTX-M, sul1, tetA, tetB, and qnrA resistance genes. Meropenem, levofloxacin, and tigecycline exhibited talented antimicrobial activity against MDR-STEC isolated from dogs.


Introduction
Acute diarrhea in canine puppies is one of the most common and serious gastrointestinal problems, potentially leading to severe dehydration and death (Duijvestijn et al. 2016). Escherichia coli (E. coli) is found frequently in the gastrointestinal tract of both humans and animals as natural commensals (Cho et al. 2006). Based on the mechanism of disease occurrence, the pathogenic E. coli is classified into Enterohemorrhagic E. coli (EHEC), Enterotoxigenic E. coli (ETEC), Enteropathogenic E. coli Shiga toxigenic E. coli (STEC) is a common zoonotic diarrheagenic pathotype that can produce Shiga-toxins family (Stxs) cytotoxins (Melton-Celsa 2014). STEC possesses over 400 serotypes, some of which are associated with severe illnesses in animals and humans (Torres et al. 2018). The most predominant STEC serovars are O157:H7 and non-O157 serogroups (Such as O45, O121, O111, O26, O145, and O103) (Miko et al. 2014;Conrad et al. 2014). The emergence of STEC in dogs was reported with a prevalence ranging from 4 to 15.5% (Bentancor et al. 2007;Galarce 2019). The serotype O157: H7 is incriminated in most food-borne outbreaks in humans. In contrast, the non-O157 serovars are commonly associated with severe illness in animals (Rangel et al. 2005;Hedican et al. 2009).
The most important virulent determinants associated with STEC include 1-Shiga-toxins (stx1 and stx2) encoded by Shiga-toxins genes stx1 and stx2, respectively, 2-Intimin encoded by E. coli attaching and effacing gene (eae), and 3-Hemolysin that is necessary for STEC pathogenicity, especially hlyA (Puño- Salvadori et al. 2003;Puño-Sarmiento et al. 2013). The polymerase chain reaction is an accurate, sensitive, and consistent diagnostic assay used for bacterial identification and the determination of virulence determinants and antimicrobial resistance genes (Carvalho et al. 2021).
The multidrug resistance (MDR) patterns have increased worldwide, especially in the last decade. Several recent epidemiological studies have revealed the emergence of MDR bacterial pathogens from different origins that reflect a public health threat (Vega-Manriquez et al. 2020; Dazio et al. 2021). Multidrug resistance is common in E. coli and is primarily caused by the presence of antimicrobial resistance genes such as; bla TEM (responsible for penicillin-resistance), bla CTX-M (cephalosporins-resistance), tetA and tetB (tetracyclineresistance), qnrA, qnrB, and qnrS (Plasmid-mediated quinolone resistance), and sul1 (sulfonamide resistance). Although the emergence of carbapenem-resistant bacterial pathogens is low, it has increased recently worldwide. The carbapenem-resistance is mediated mainly by bla KPC and bla NDM-1 genes (Wedley et al. 2017;Peterhans et al. 2018;Alba et al. 2021).
The present study aimed to elucidate the clinical findings in naturally infected puppies with STEC. Also, to detect the rate, Cong-red binding, antimicrobial resistance patterns, virulence and antimicrobial resistance genes in STEC isolated from the examined puppies.

Clinical examination and sampling
The clinical examination was conducted in small animal clinics in Ismailia province, Egypt, from March 2020 to June 2020. A total of 80 puppies (n = 80; 45 male and 35 female German Shepherd dogs) aged between 2 and 3 months old with an average weight of 5-10 kg were involved in the present study. The examined puppies suffered from bloody diarrhea. The clinical examination was performed according to the procedures described previously (Cote and Cohn 2019). Eighty fecal swabs were collected from the examined dogs under aseptic conditions into tryptic soy broth (Oxoid, UK), then transported to the microbiology laboratory as soon as possible for further bacteriological identification.

Isolation and identification of E. coli
The collected fecal swabs were inoculated in MacConkey broth (Difco, USA) and left incubated at 37 °C for 24 h. A loopful of the incubated broth was streaked on MacConkey agar and Eosin Methylene Blue agar (EMB) plates (Difco, USA) and left incubated at 37 °C for 24 h. The purified colonies were identified according to their cultural characteristics, Grams's staining, hemolysis on blood agar, motility test, and biochemical tests (lactose fermentation, oxidase, Voges-Proskauer, indole, methylred, citrate-utilization, catalase, urease, and H 2 S production) as previously described by MacFaddin (MacFaddin 1985).

Serogrouping of the retrieved isolates
The serogrouping of the recovered isolates was carried out at the Serology Division of the AHRI, Giza, Egypt, using the slide agglutination test for detection of O antigen, where polyvalent and specific monovalent E. coli antisera were involved (Mast Group Ltd.-Co., Merseyside, UK) (Starr 1986).

Congo-red binding test
To evaluate the in-vitro invasiveness and virulence of the retrieved isolates, the Congo-red binding assay was performed using tryptic soy agar (containing 0.03% Congo-red dye) (Difco, USA). The retrieved isolates were streaked on tryptic soy agar and left incubated at 37 °C for 24 h. The positive result is indicated by the appearance of red colonies, as previously reported by Panigrahy and Yushen (1990).

Screening of virulence and antimicrobial resistance genes using PCR
The bacterial DNA was extracted consistently with the manufacturer's guidelines of the GenElute DNA Kit (Sigma Aldrich, USA). Briefly, the DNA Kit uses spin column technology. An optimized combination of mechanical and chemical lysis is used to disrupt bacterial cells subsequently, and carefully designed buffer chemistry promotes the binding of the released DNA to the silica membrane. Finally, the bacterial DNA is eluted from the column after washing away unwanted components. Nanodrop (Nanodrop 1000, Thermo Scientific, UK) was used to quantify genomic DNA templates, which were then adjusted to 100 ng l −1 and stored at − 20 °C until used for PCR. The PCR reactions were carried out in a volume of 25 μl (12.5 μl of Taq Green Master Mix 2X (Promega, Wisconsin, USA), one μl of each primer, 5 μl of DNA template, and PCR grade water up to 25 μL. Positive controls were involved in all reactions (E. coli strains previously isolated and kindly supported by the Animal Health Research Institute, Egypt). Besides, a reaction with no DNA template was used as a negative control. The oligonucleotide sequences (ThermoFisher Scientific, USA), specific amplicon size, and cycling conditions are shown in Table 1. Amplified PCR products were screened by 1.5% (w/v) agarose gel electrophoresis (Applichem GmbH, Darmstadt, Germany) for 45 min at 100 V in 1X TAE buffer (0.04 M Tris, 0.0001 M EDTA, pH 8.0), visualized using 15 µL of ethidium bromide (Sigma Aldrich, USA) and photographed under a UV transilluminator.

Statistical analyses
The Chi-square was carried out to analyze the resistance phenotype using the statistical analysis software (SAS, software version 9.4), and significance was determined at a p value < 0.05. The results of antimicrobial susceptibility testing were graphed by a heatmap through GraphPad Software (version 8.0.1, GraphPad Software Inc., La Jolla, CA, USA). A heatmap with hierarchical clustering was performed to visualize the overall distribution of virulence genes among recovered isolates using the "heatmap" package in R software (version 4.0.2; https:// www.r-proje ct. org/).
The association among various variables (resistance phenotypes and antimicrobial resistance genes) was carried out using Spearman correlations. The correlation coefficients and their p-values were visualized using a correlation plot.

Clinical manifestations of hemorrhagic diarrhea in the examined dogs
The prominent recorded clinical signs included anorexia, lethargy, depression, vomiting, foul-smelling bloody watery diarrhea, and severe dehydration manifested by retardation of skin turgor tests for up to 10 s. Besides, the affected dogs showed an elevation in body temperature (≥ 39.5 °C) that gradually decreased with the progression of the disease.

The phenotypic characteristics and the prevalence of E. coli in the examined dogs
Morphologically, the recovered isolates were moderatesized Gram-negative, motile, and non-spore-forming bacilli. On MacCkoney agar, the obtained isolates were lactose fermenters (pink colonies). Moreover, the colonies were metallic sheen colonies on EMB and were hemolytic on Blood agar. Besides, all the retrieved isolates were negative for oxidase, H 2 S production, citrateutilization, urease production, and Voges-Proskauer tests. Concurrently, they were positive for lactose fermentation, methyl-red, catalase, indole, and nitrate reduction tests. The prevalence of E. coli in the examined fecal swabs collected from diseased dogs was 23.75% (19/80).

Antimicrobial classes
The distribution of phenotypic resistance patterns, virulence, and antimicrobial resistance genes between the recovered E.coli serotypes (n = 19) is illustrated at the serovar level in Table 4 and Fig. 4.

Discussion
The current study was directed to investigate the rate, antimicrobial resistance patterns, screening of virulence and antimicrobial resistance genes in STEC isolated from dogs suffering from hemorrhagic diarrhea. The clinically examined diseased dogs suffered from fever, vomiting, dehydration, and foul-smelling bloody diarrhea. Our findings agreed with those reported by (Nayel et al. 2019). Bloody diarrhea is attributed mainly to the destruction of the epithelium of the intestinal crypts and subsequent villous atrophy (Unterer and Busch 2020). Moreover, dehydration was also a characteristic clinical sequel caused by losing fluids due to vomiting and diarrhea (Sykes and Greene 2013).
In the present study, the phenotypic characteristics of the recovered E. coli isolates are similar to those reported by (Sengupta et al. 2011). Besides, the prevalence of E. coli is nearly similar to that reported by a previous study in urban dogs in France (24.5%) (Haenni et al. 2014). Moreover, a lower prevalence was reported in Tunisia (17.5%) (Sallem et al. 2013). Furthermore, a higher prevalence of E. coli was reported in Brazil (31.3%) (Carvalho Virulence determinant genes Antimicrobial resistance genes Fig. 3 The dissemination of virulence determinants and antimicrobial resistance genes between the retrieved E. coli isolates Table 3 The distribution of multidrug-resistance patterns and the corresponding-resistance genes among the tested E. coli isolates (n = 19)  (Carvalho et al. 2021). Regarding E. coli serogrouping, the retrieved serovars are similar to those reported by Jay-Russell et al. (2014). The determination of E. coli serotypes reflects epidemiological and public health importance. All the examined E. coli serotypes (O111, O146, and O26) were positive for Congo-red binding that emphasized the virulence and invasiveness of the tested serovars. Furthermore, all the non-typeable isolates were negative. These results are consistent with those reported by (Algammal et al. 2020b).
In the present work, the obtained isolates displayed a remarkable resistance to trimethoprim-sulfamethoxazole, tetracycline, amoxicillin, amoxicillin-clavulanic acid, ceftazidime, and cefotaxime, which is greatly threatening public health. Our findings agreed with those recorded by Jay- Russell et al. (2014) and Gupta et al. (2017). Recently, antimicrobial resistance has increased globally due to the widespread unregulated application of antimicrobial agents in the health and veterinary sectors and the dissemination of antibiotic-resistant genes carried on the bacterial chromosome or plasmid (Yamamoto et al. 2013).  Fig. 4 The heat map shows the correlation between virulence genes, the antimicrobial-resistance genes, and the retrieved STEC serovars In the present study, the PCR revealed that the stx1 gene is the most predominant virulence gene associated with STEC isolated from diseased dogs. Our findings emphasized the presence of the stx1 gene in combination with the eaeA and hlyA genes and/or the stx2 gene. On the other hand, a previous study reported that the stx2 gene was the most predominant virulence gene (23%) in STEC strains isolated from diarrheic dogs in the USA (Staats et al. 2003). The pathogenicity of Shiga toxin-producing E.coli is attributed mainly to a variety of virulence determinants, including 1-Shiga toxins (encoded by stx1 and stx2 genes), which regularly destruct the eukaryotic ribosomes and prevent protein-biosynthesis in the host, 2-Intimin (encoded by the eae gene) that is responsible for E. coli attachment to the intestinal mucosa, and 3-Hemolysin-A (plasmid gene EHEC-hlyA) is an exotoxin that lyses the erythrocytes in the intestinal mucosa (El-Baky et al. 2020). The virulence genes detected in E. coli of animal origin showed more toxin expression than those obtained from human strains (Keen and Elder 2002).
Regarding the multidrug resistance (MDR) profiles and the dissemination of antimicrobial resistance genes, our findings emphasized that most obtained isolates express MDR to 4 or more different antimicrobial classes and harbored sul1, tetA, tetB, bla TEM , bla CTX.M-1 , and qnrA genes. The existence of MDR-E. coli reflects a public health alarm and suggests a poor prognosis of diseases induced by these isolates. Pet animals, especially dogs, are known as potential reservoirs of antimicrobial resistance. However, the available data are limited. The extensive application of antibiotics in both veterinary practice and the health sector predispose to the emergence of new superbugs. A recent study by Joosten et al. (2020) reported that 16% of the recovered E. coli isolates from dogs and cats from three European countries were The heat map shows the correlation between the phenotypic antibiotic-resistant and corresponding genes recognized as multidrug-resistant strains (resistant to 6 antibiotics). One E. coli strain from Italy was resistant to 10 different antibiotics. Moreover, three E. coli strains (2 from Italy and one from Belgium) were multidrugresistant to ampicillin, cefotaxime, and ceftazidime. The Extended-spectrum β-lactamases (ESBLs) are responsible for the hydrolysis of Broad-spectrum β-lactam antibiotics such as penicillins and cephalosporins. ESBLS are frequently detected in E. coli (Schultz et al. 2015). The high prevalence of the bla TEM and bla CTX-M genes among the retrieved E. coli strains enables them to resist penicillins and cephalosporins.
Tetracycline resistance is mainly attributed to tetracycline resistance genes. In the current study, the tetA gene was detected frequently in the recovered MDR E. coli strains, either alone or combined with the tetB gene. These findings agreed with a previous study performed on E. coli isolated from pigs in Spain (Jurado-Rabadán et al. 2014). Plasmid-mediated quinolone resistance (PMQR) was reported firstly in K. pneumoniae in the USA in 1998s. The qnrA gene codes for a 218 amino acid protein (QnrA protein) that prevents the binding of quinolones to DNA gyrase. Moreover, two other PMQR genes, including qnrB and qnrS genes, have been recognized to code for the same protein family and share 41% and 60% amino acid identity with QnrA, respectively (Robicsek et al. 2006). Alarmingly, in the present study, three E.coli strains were carbapenem-resistant. Worldwide, the existence of carbapenem-resistant strains is somewhat low; however, it has increased over time. The resistance to carbapenems is endorsed by bla KPC and bla NDM-1 resistance genes (Cole et al. 2020). Moreover, a lower resistance rate against amikacin was recorded in the current study. These findings agreed with those reported by Qekwana et al. (2018), who stated that the aminoglycoside resistance in E. coli strains isolated from urinary tract infections in dogs was relatively low (22%). This lower resistance rate could be attributed to the less frequent use of amikacin in treating bacterial infections in dogs in Egypt. The resistance mechanisms in E. coli include: a-A common resistance mechanism that occurs in antibiotics of the same class due to the expression of extended-spectrum β-lactamases or mutations in the penicillin-binding proteins, b-The repeated combined antimicrobial therapies, and c-The association between the antimicrobial resistance genes; this type exhibits a vital role in the occurrence of resistance to various antimicrobial agents (Algammal et al. 2020a(Algammal et al. , 2020b. As far as the authors know, this is the first study that reported the existence of MDR-Shiga toxigenic E. coli (STEC) in dogs in Egypt. The stx1 gene is the most predominant Shiga toxin gene, found mainly in combination with the eaeA and hlyA genes and/or the stx2 gene. The emerging MDR-STEC in dogs commonly harbors bla TEM , bla CTX-M , sul1, tetA, tetB, and qnrA resistance genes. Meropenem, levofloxacin, and tigecycline exhibited talented antimicrobial activity against MDR-STEC isolated from dogs. The regular conduction of antimicrobial susceptibility testing is necessary to detect the most effective antibiotic and determine the emergence of new MDR strains. The emergence of virulent MDR-STEC strains comprises a vital threat to pet animals and human health, especially carbapenem-resistant strains harboring bla KPC or bla NDM-1 resistance genes.