Extended-Spectrum Beta-Lactamase Producing Strains of Escherichia Coli Isolated from Avian Cellulitis Lesions

Avian cellulitis causes significant losses to the poultry industry. Avian-pathogenic Escherichia coli (APEC) is the etiological agent of that disease. This microorganism has zoonotic potential and may act as reservoir of antimicrobial-resistance genes. In this context, the production of extended-spectrum B-lactamase (ESBL) is one of the main antimicrobial resistance mechanisms. The objective of this study was to determine the production of ESBL in an Escherichia coli (E. coli) strain isolated from avian cellulitis lesions. Twenty-two E. Coli isolates were harvested from cellulitis lesions in chicken carcasses in a commercial processing plant. Isolates were then submitted to virulence genotypic profile (iutA, hlyF, iss, ironN, ompT) analysis, antimicrobial susceptibility test, and detection of ESBL production. The results showed that 22.7% of the isolates presented five virulence genes, 9.1% four genes, 36.4% three genes, 13.6% two genes, and 18.2% one gene. The tested isolates showed resistance to ampicillin (90.9%), ceftiofur (54.5%), gentamicin (45.5%), tetracycline (72.1%), sulfamethoxazole/ trimethoprim (54.5%), and enrofloxacin (54.5%). Furthermore, 77.3% of the isolates presented multidrug resistance (MDR) profile and 72.7% were positive for ESBL production. This study is the first description of ESBL-producing APEC isolated from avian cellulitis lesions, which suggests the need to establish efficient APEC control measures and programs to prevent flock productivity losses due to colibacillosis and public health risks.


INTRODUCTION
Avian cellulitis is an emerging disease that has caused significant economic losses to the poultry industry (Quel et al.,2013). It is a multifactorial disease resulting from management failures, immunosuppressive diseases, and skin lesions. The lesions, often referred to as "plaques," are typically located in the skin on the abdomen or between the thigh and the midline, and therefore, are identified only after slaughter (Barnes et al., 2008).
Avian pathogenic Escherichia coli (APEC) belongs to the pathotype extraintestinal pathogenic Escherichia coli (ExPEC) (Lestrange et al., 2017). Studies indicate that APEC has zoonotic potential, considering its profile of virulence genes and its capacity of acting as a reservoir of antimicrobial resistance genes, which can be transmitted to humans through the food chain (Aslam et al., 2014;Cunha et al., 2014).

eRBCA-2019-0981
One the hypothesis for the selection of resistant microorganisms is the extensive use of antimicrobials in animal production (Cunha et al., 2014, Hoelzer et al., 2017. The phenomenon of antimicrobial resistance poses a serious problem to human as well as veterinary medicine. The production of extendedspectrum beta-lactamase (ESBL) is one of the main resistance mechanisms of bacteria to antibiotics. This enzyme hydrolyzes the b-lactam ring of several antibiotics, including cephalosporin of third and fourth generations, penicillin, cephalosporin (ceftazidime, cefotaxime, ceftriaxone) and monobactam (aztreonam), inactivating them and, consequently, reducing treatment options (Casella et al., 2017;Shaikh et al., 2015).
Determining the virulence profile is essential to differentiate APEC isolates from non-pathogenic E. coli residing in the natural microbiota. In addition, the detection of virulence factors and antimicrobial resistance helps understanding the epidemiology and allows designing measures for the prevention and control of the disease. This study aimed at establishing the profile of virulence genes, phenotypic resistance to antimicrobials, and ESBL-producing capacity of APEC isolates obtained from avian cellulitis lesions.

Bacterial isolates
Swabs of lesions [n=42] compatible with cellulitis were collected from broiler carcasses in the slaughter line of a broiler processing plant in northern Paraná State, Brazil. Samples were collected at three different times at 7-d intervals in order to obtain isolates from flocks from different sources. Sterile swabs were rubbed on the lesions, refrigerated (2-8°C) in Cary Blair medium and submitted, under refrigeration, to the Laboratory of Avian Medicine, State University of Londrina, for processing.
Swabs were incubated in BHI broth (Brain Heart Infusion, Difco ® , France) at 37°C for 18-24 h, and streaked on MacConkey agar (Acumedia ® , USA) plates and incubated at 37°C for 18-24 h. The isolated colonies were subjected to biochemical tests, and the colonies compatible with E. coli (positive for indole, negative for H 2 S, negative for citrate and urease, gas production in TSI (triple sugar iron) agar with acid slant and acid butt, sorbitol-fermenting) were stored at -20°C in BHI broth supplemented with glycerol (20%) until processing, totaling 22 E. coli isolates.

Phenotypic profile of ESBL production
The production of extended-spectrum betalactamase (ESBL) was determined by the disk approximation test (D-test), according to the Clinical and Laboratory Standards Institute (CLSI, 2015). Four antimicrobial agents were tested using the double-disk synergy test: amoxicillin +clavulanic acid (AMX/C-20/10 µg), ceftazidime (CAZ-30 µg), ceftriaxone (CRO-30 µg), cefotaxime (CTX-30 µg). The AMX/C disc was applied at the center of the plate containing Mueller Hinton (MH; Difco ® ) agar and the others at a distance of 20 mm from the edge of the central disk. Isolates in plates showing any distortion or increase in the area of the amoxicillin-clavulanate disk was considered positive for ESBL production.

Determination of virulence factors
E. coli isolates were grown in BHI broth, and then subjected to DNA extraction. The boiling technique was applied, consisting of centrifugation of 1.5 mL bacterial growth for 1 min at 7500 x g. The precipitate was suspended in 200 μL of ultrapure sterile water, incubated at 100°C for 10 min, and centrifuged again for 5 min at 2800 x g. Next, 150 μL of the supernatant were removed and stored at -20°C until use.
The virulence genes iutA, hlyF, iss, ironN and ompT were studied in the isolates following the protocol described by Johnson et al. (2008).

Virulence genes
Twenty-two E. coli isolates were obtained from cellulitis lesions in chicken carcasses, out of which 22.7% (5/22) presented five virulence genes (iss, ompT, iroN, iutA, hlyF), 9.1% (2/22) four genes, 36.4% (8/22) three genes, 13.6% (3/22) two genes, and 18.2% (4/22) only one gene (Table 2). The genes iss, ompT and iroN were not detected in any of the isolates evaluated. In the present study, 27.3% (6/22) of isolates showed the combination of iroN, iutA and hlyF genes. Conversely, the presence of individual genes was observed in 18.2% of the isolates, with 9.1% prevalence of iutA and hlyF. Figure 1 shows the distribution of virulence genes in the E. coli isolates evaluated. The most prevalent genes were iutA and hlyF, in 77.3% and 72.7% of the isolates, respectively, iroN was observed in 63.6%, while the ompT and iss genes were detected in only 45.4% of the isolates.

DISCUSSION
Escherichia coli is a natural resident of the intestinal microbiota of birds and mammals; however, some pathotypes designated avian pathogenic E. coli (APEC) may invade different organs and cause systemic disease or colonize individual tissues, such as the case of cellulitis (Li et al., 2015).
The present study determined the virulence gene profile, phenotypic resistance to antimicrobials, and extended-spectrum beta-lactamase (ESBL) production of APEC isolates collected from avian cellulitis lesions. The objective was to demonstrate that the antimicrobial resistance profile of a microorganism may cause disease in broilers and pose a public health risk through the food chain, transferring resistance genes to potential human pathogens.
To the best of our knowledge, this is the first study to describe ESBL-producing APEC isolates in avian cellulitis lesions, although other studies have isolated ESBL-positive E. coli strains in live chickens and poultry products (Lim et al., 2015;Hussain et al., 2017;Silva et al., 2017).
Antimicrobial therapy is one of the primary control measures to reduce morbidity and mortality caused by APEC (Barbieri et al., 2013). However, the frequent use of antimicrobials allows the selection of resistant isolates, which is a global health concern (Nhung et al., 2017).
In the present study,77.3% of the isolates evaluated presented the MDR phenotype, which presented high resistance to ampicillin (90.9%) and tetracycline (72.1%). Similarly, Barbieri et al. (2013) found a high percentage of resistance to tetracycline (69.4%) in E. coli isolated from avian cellulitis in broilers in southern Brazil. Wu et al. (2014) found 84.4 and 71.1% of resistance to tetracycline and ampicillin, respectively, in E. coli isolates from poultry in China.
In our study, 54.5% of the isolates showed resistance to sulfamethoxazole/trimethoprim. This profile is alarming because these drugs are the first choice for the treatment of human patients in intensive care units (Brown & Foxman, 2002). Chen et al. (2014) observed that 90.6% APEC isolate were resistant to tetracycline, and suggest a co-resistance relationship among tetracycline, trimethoprim-sulfamethoxazole, and ampicillin.
In addition to public health aspects, the presence of antimicrobial-resistant pathogenic microorganisms directly affects livestock production, since it undermines treatments and prolong the disease in animals, reducing therapeutic alternatives, increasing production costs, and reducing profitability (El-Shazly et al., 2017).
ESBL production is an important resistance mechanism to antimicrobials (Daehre et al., 2017). In our study, the ESBL phenotype was present in 72.1% of the isolates, which is significant. This result is consistent with those reported by Casella et al. (2017), who observed 96.1% (74/77) positivity for ESBL in E. coli isolates from poultry in France. Our study is the first description of ESBL-producing APEC isolated from avian cellulitis lesions.
According to Johnson et al. (2008), the iutA, hlyF, iss, iroN and ompT genes are associated with plasmids and are considered good APEC markers, although other markers have been described. Our findings are consistent with those of Hiki et al. (2014), who determined that iutA and hlyF as the first and second most prevalent virulence genes in Escherichia coli isolates from healthy broilers in Japan. Ozaki et al. (2017) observed that iss gene was present in 93% of APEC isolates from perihepatite and pericarditis lesions in broilers. These data differ from our results, since iss was observed in 45% of the isolates. This difference may be attributed to the fact that we analyzed isolates from cellulitis lesions, whereas Ozaki et al. (2017) worked with isolates causing septicemia. According to Silveira et al. (2016), the presence of the ompT, iroN, and iss genes is correlated with disease severity, while the hlyF and iutA genes are involved with the capacity of causing disease. Therefore, we can infer that 27% of the isolates in the present study had capacity of causing severe disease.

CONCLUSIONS
The identification of APEC isolates from avian cellulitis lesions containing virulence genes with capacity of producing ESBL and MDR suggests the need to establish efficient APEC control measures and programs in order to reduce flock productivity losses due to colibacillosis and public health risks.