**3. Results and Discussion**

#### *3.1. Distribution of Integrons in ESBL-Producing E. coli*

The occurrence and types of integrons, according to the isolate origin, is presented in Figure 1. As expected, class 1 showed the highest dissemination, being present in 92% of the isolates (*n* = 138) and in all environments, without significant differences among them (*p* < 0.05). Class 1 integrons have been reported as the most ubiquitous type among enteric bacteria [34–36]. In a similar way, Solberg et al. [37] reported the presence of class 1 integron in 70% of *E. coli* causing community-acquired infections. According to Roe et al. [38], the occurrence of class 1 integrons in healthy people suggests a possible acquisition of resistance genes circulating in different environments by a constant horizontal exchange of these genes. By contrast, class 2 integron was found in only 13 strains (8.5%), in accordance with the study of Ozgumus et al. [39], who found this class of integron in pathogenic, environmental, and commensal *E. coli* with a lower frequency than class 1. Finally, *intI3* was not detected, similarly to the report by Vinué et al. [40]. In fact, limited studies describe the presence of class 3 integron in *E. coli* [14,41] and, to date, there are no published data reporting the presence of this integron in ESBL-producing *E. coli* isolated from Spain.

**Figure 1.** Prevalence (percentages) and distribution of (**A**)class 1 (*intI1*) and (**B**) class 2 (*intI2*) integrons in ESBL-producing *E. coli* according to their origin. ESBL: Extended spectrum beta-lactamases.

It should be noted that *intI2* was mainly detected in food isolates (18.4%), but not in farming environments (*p* = 0.044). This situation seems a little bit contradictory, but it could be due to the low number of isolates coming from farms and feed (*n* = 20), compared with food (*n* = 48). Probably, if we extended the study by increasing *n*, we would find positive results for this type of integron, as shown in the literature. In any case, our results are comparable to those obtained by Goldstein et al. [42], who demonstrated the presence of class 1 and class 2 integrons in food, livestock, and water contaminated with farm animal feces. In a similar way, *intI2* has been detected in poultry products [38].

In addition, it is remarkable that *intI1* and *intI2* coexist in 8% of the isolates (92.1% of those carrying *intI2*). Rizk et al. [20] reported the co-existence of more than one type of integron in 36.9% of isolates, and a prevalence of 38% was reported by Kargar et al. [41] in a study performed in 69 multidrug-resistant (MDR) *E. coli*. By contrast, Kor et al. [43] found only one isolate carrying both integrons among clinical isolates, and Odetoyin et al. [16] reported a prevalence of 2.4% in fecal *E. coli* isolated from mother–child pairs in Nigeria. The simultaneous existence of multiple integrons represents a great threat for the dissemination of antimicrobial resistance genes among Enterobacteriaceae [31].
