Tracking antibiotic resistance genes and class 1 integrons in Escherichia coli isolates from wastewater and agricultural fields

Considering high concentrations of multidrug-resistant bacteria and antibiotic resistance genes (ARGs) in wastewater, agricultural reuse of treated wastewater may be a public health threat due to ARG dissemination in different environmental compartments, including soil and edible parts of crops. We investigated the presence of antibiotic-resistant Escherichia coli as an indicator bacterium from secondary treated wastewater (STWW), wateror wastewater-irrigated soil and crop samples. ARGs including blaCTX-m-32, blaOXA-23, tet-W, sul1, cml-A, erm-B, along with intI1 gene in E. coli isolates were detected via molecular methods. The most prevalent ARGs in 78 E. coli isolates were sul1 (42%), followed by blaCTX-m-32 (19%), and erm-B (17%). IntI1 as a class 1 integrons gene was detected in 46% of the isolates. Cml-A was detected in STWW isolates but no E. coli isolate from wastewater-irrigated soil and crop samples contained this gene. The results also showed no detection of E. coli in water-irrigated soil and crop samples. Statistical analysis showed a correlation between sul1 and cml-A with intI1. The results suggest that agricultural reuse of wastewater may contribute to the transmission of antibiotic-resistant bacteria to soil and crop. Further research is needed to determine the potential risk of ARB associated with the consumption of wastewater-irrigated crops.


INTRODUCTION
In recent decades, the excessive use of antibiotics as therapeutic drugs in humans, animals, and plants has caused selective pressure on bacterial populations, leading to the dissemination of antibiotic-resistant bacteria (ARB) and antibiotic resistance genes (ARGs) in the environment (Osinśka et al. 2017;Yuan et al. 2020). As a result, World Health Organization (WHO) established antibiotic resistance (AR) as a critical global health concern of the 21st century (Campo et al. 2020;Yuan et al. 2020).
ARB and ARGs are mainly spread by various environmental compartments such as wastewater treatment plant (WWTP) effluents (Osinśka et al. 2017), surface water (Koczura et al. 2012), livestock manure (Yuan et al. 2020), and agricultural environment (Bougnom et al. 2020). Wastewater as a nutrient-rich environment with a high microbial concentration and diversity provides favorite conditions for horizontal gene transfer (HGT) among bacteria (Osinśka et al. 2017;Campo et al. 2020).
As communities are increasingly facing with water crisis, agricultural irrigation becomes a key point challenge, especially in arid and semi-arid areas. Therefore, wastewater reuse for agricultural activities is considered worldwide (Farhadkhani et al. 2018). Considering high concentrations of ARB and ARGs in the effluent of WWTPs, agricultural reuse of secondary treated wastewater (STWW) could considerably spread antibiotics, ARB and ARGs into the environment (Cerqueira et al. 2019a;Kumar et al. 2020). This may lead to an elevated resistance level in microbiome of soil and crops (Cerqueiraet al. 2019a(Cerqueiraet al. , 2019b, thus threatening human and animal health. In other words, these environments turn into hotspots for ARB, ARGs, and mobile genetic elements (MGEs) (Yuan et al. 2020).
The increase in AR is facilitated by the association of ARGs with a variety of MGEs (Osinśka et al. 2017). Integrons, as one of the genetic elements involved in the environmental spread of AR are found in MGEs such as plasmids and transposons, which promote their spread within bacterial communities (Cerqueira et al. 2019a). The common types of integrons which confer resistance to antibiotics including β-lactam antibiotics, chloramphenicol, sulfonamides, aminoglycoside, spectinomycin and others (Su et al. 2012) are class 1 integrons. This type of integrons are widely distributed in ARB, especially in Gramnegative bacteria (Osinśka et al. 2017).
In this context, the presence of Gram-negative bacteria including Enterobacteriaceae members such as Escherichia coli in the environment as important vectors of ARGs, may pose a threat to human health (Cerqueira et al. 2019a). E. coli is a commensal bacterium which lives in the gut of humans and warm-blooded animals and could widely spread in different natural environments via feces or treated wastewater (Osinśka et al. 2017). Furthermore, some pathogenic strains of E. coli can cause infectious diseases in animals and humans, such as urinary tract infections, diarrhea, and septicemia (Araújo et al. 2017).
It seems that E. coli, as an important indicator of fecal contamination, has an important role in the transfer of ARGs to pathogens (Holvoet et al. 2013;Araújo et al. 2017;Osinśka et al. 2017). Therefore, the frequency of antibiotic-resistant E. coli in the environment can be used as a good estimate of the prevalence of other resistant pathogenic bacteria such as Salmonella (Holvoet et al. 2013). Conversely, determination of the resistance level of indicator bacteria may be useful to monitor the changes in the AR of the intestinal microbiota of human populations (Paulshus et al. 2019).
We surveyed the presence of antibiotic-resistant E. coli in STWW, water-or wastewater-irrigated soil and crop samples. Antibiotic resistance was determined by detecting six ARGs, including bla CTX-m-32 , bla OXA-23 , tet-W, sul1, cml-A, erm-B as well as intI1 gene (a key component of class 1 integrons) as a marker of anthropogenic pollution in E. coli isolates.

Sample collection and preparation
This study was carried out in Isfahan, in the central part of Iran. In total, 51 samples including wastewater (18 samples); agricultural soil (18 samples) and crop (15 samples) were collected and analyzed for the presence of antibiotic-resistant E. coli.
Wastewater samples were collected from STWW of two WWTPs. Water-or wastewater-irrigated soil and crop samples were taken from agricultural fields. Soil samples were collected by an auger from the 0-20 cm topsoil layer. From each agricultural field a composite soil or crop sample consisting of three subsamples were analyzed. The edible parts of crop samples washed with tap water and 10 À1 dilutions were prepared in sterile peptone water by homogenization for 3 minutes in a stomacher. A certain amount of each soil sample was also mixed with sterile peptone water (1:10 w/v) and then homogenized in a shaker incubator for 1 h.
All samples were taken in sterilized glass bottles or bags and were immediately transported to the laboratory for further processing.

E. coli isolation
For detection of E. coli, ten-fold serial dilutions (10 À1 -10 À6 ) of samples were prepared. From all diluted samples, aliquots of 100 μL were plated onto lauryl sulphate MUG X-gal (LMX) agar in duplicate and incubated at 37°C for 18-24 h. After incubation time the presence of E. coli in samples was determined by a light blue fluorescence of colonies under UV light (Farhadkhani et al. 2020). Finally, E. coli colonies were isolated using eosin methylene blue agar and incubated at 37°C for 18-24 h.

Molecular detection of antibiotic resistance genes and class 1 integrons (intI1)
E. coli isolates were screened by PCR for detection of genes conferring resistance to β-lactam (bla CTX-m-32 , bla OXA-23 ), tetracycline (tet-W), sulfonamide (sul1), chloramphenicol (cml-A), and macrolide (erm-B). The PCR amplification of intI1, was performed to evaluate the presence of class 1 integrons. PCR conditions and primer sets used for amplification of the genes are presented in Table 1.
For PCR detection of the genes, genomic DNA was extracted from the E. coli isolates using a boiling technique. The PCR assay was performed in a 25 μL volume, as described previously (Farhadkhani et al. 2019). The PCR cycling conditions consisted of 5 minutes denaturation at 95°C followed by 35 cycles of 94°C for 45 seconds, annealing at varied temperatures (Table 1) for 45 seconds, polymerization at 72°C for 45 seconds, and a final extension at 72°C for 5 minutes. Positive and negative controls were included in all PCR assays. Analysis of the PCR products was performed by electrophoresis in a 1.5% (w/v) agarose gel and visualizing of gels by ultraviolet (UV) transilluminator (UV Tech, France).

Statistical analysis
Spearman's correlation was performed to assess the relationship between the presence of ARGs at a significance level (P-value) of ,0.01. Statistical analyses were carried out using the statistical package SPSS software and Microsoft Excel (version 2016).

RESULTS AND DISCUSSION
Detection of E. coli AS a part of the normal gut flora in warm-blooded animals, E. coli is a good indicator of the fecal pollution of environmental samples (Osinśka et al. 2017). E. coli was detected in almost all STWW samples and 20% (3 of 15) and 13% (2 of 15) of soil and crop samples, respectively. E. coli was not detected in soil or crop samples irrigated with other types of water (well or channel water) which indicates no fecal pollution of the samples. Low detection frequency of E. coli in STWW-irrigated soil and crop samples, could be related to the environmental condition of the region. Environmental conditions of arid and semi-arid areas including high temperature, sunlight intensity and low humidity affect the survival of microorganisms in soil and on crop samples (Farhadkhani et al. 2018). In consistent with our results, E. coli was detected with a low frequency in wastewater-irrigated soil and crop samples despite relatively high levels in wastewater (Gatta et al. 2016;Farhadkhani et al. 2018).
In this work, 78 E. coli isolates were identified, including 62 isolates from STWW, 12 from the soil, and four from the crop samples. Detection of E. coli as a fecal indicator bacterium in wastewater-irrigated soil and crop samples indicates that agricultural reuse of wastewater may be a potential route for the transmission of pathogenic microorganisms and ARB to humans through the food chain.

Identification of antibiotic resistance genes
E. coli has great genomic plasticity in the nutritional aspects as well as the acquisition of MGEs, which highlights its importance as a vector for ARGs dissemination in the environment (Aristizábal-Hoyos et al. 2019). Analysis of ARGs showed the presence of sul1, erm-B, bla CTX-m-32 , cml-A, and intI1 genes in E. coli isolates with the highest frequency of detection for sul1. None of the isolates harbored tet-W and bla OXA-23 resistance genes (Table 2).
Detection of the sul1 gene as the most frequent ARG in E. coli isolates consistent with the results of another study which reported that the presence of the gene was correlated to the presence of Beta and Gammaproteobacteria phylum's (e.g. Escherichia). These bacterial phylum's with a relatively high AR are generally found in wastewater (Cerqueira et al. 2019b). Besides, a higher frequency of sul1 gene in STWW may be related to long-time use of sulfonamides in medicine in the past years (Kumar et al. 2020). Other researches also addressed the detection of sulfonamide ARGs at higher concentrations than quinolone (qnr), macrolide (erm-B), and tetracycline (tet-W) resistance genes in environmental samples (Dungan et al. 2018;Kumar et al. 2020).
The high frequency of detection of bla CTX-m-32 gene in E. coli isolates is also consistent with other studies that found a great number of CTX-m-group resistance genes in environmental samples As STWW isolates, the sul1 gene was detected with the highest frequency in soil and crop isolates, whereas the erm-B gene was not detected in crop samples. Cml-A with low frequency (10%) in STWW isolates was not detected in soil and crop samples (Table 2). This trend may be related to the relative abundance of genes in STWW. Conversely, it is most likely a consequence of ARB die-off when transferred from STWW to the soil and crop (Bougnom et al. 2020). However, other factors such as the presence of antibiotics, heavy metals, and antimicrobial disinfectant residues in STWW may influence ARGs frequency in the irrigated soils and crops (Dungan et al. 2018;Yuan et al. 2020).
Another interesting observation related to gene occurrence was that all detected ARGs in the soil and crop isolates were found in the STWW isolates. Some studies indicated the impact of WWTPs effluents on ARB dissemination in the receiving environments (Broszat et al. 2014;Chen et al. 2014). Although WWTPs play an important role to remove or minimize a wide range of pollutants, they may act as the main entry pathways of ARGs and ARB into the environment (Kumar et al. 2020). In addition, the chlorine-based disinfection in STWW could induce the discharge of ARGs from damaged microbial cells (Campo et al. 2020). Based on the results, using STWW for agricultural irrigation might contribute to the spread of ARG and ARB to the environment. A previous study conducted in Beijing and Tianjin, China indicated that soils receiving municipal wastewater in agricultural settings are enriched in ARGs (Chen et al. 2014), which supports the results of our study. For example, comparing ARG levels in the soil samples in China revealed that the relative abundance of sul1, sul 3, tet-A, tet-C, tet-E, tet-G, and tet-S were significantly higher in wastewater-irrigated soils than in non-irrigated soils (Chen et al. 2014). Detection of antibiotic-resistant E. coli in agricultural soils indicates that consumption of wastewater-irrigated crops may be a concern from a public health point of view (Bougnom et al. 2020).

Prevalence of multidrug resistance E. coli
Multidrug resistance (MDR) is defined as resistance to at least three different categories of antimicrobials agents (Osinśka et al. 2017). High concentrations of MDR bacteria were detected in municipal wastewater, hospital wastewater, and livestock feeding drainage (Kumar et al. 2020). In our study, 9% of (7 of 78) E. coli isolates were multiresistant, 13% of the isolates harboring 2 ARGs, 33% of the isolates harboring 1 ARG, and 45% of them did not carry any resistance gene (Figure 1). The results of this study revealed a small percentage of MDR (9%) E. coli compared with other studies (Osinśka et al. 2017;Aristizábal-Hoyos et al. 2019). Osinśka et al. (2017) reported that nearly 38% of E. coli strains isolated from river water and wastewater in Poland were MDR. Furthermore, 35% of E. coli strains isolated from treated wastewater were MDR, and 10% of isolates were resistant to one antibiotic. In addition, in Colombia 63.6% of E. coli isolates in the final WWTP were MDR (Aristizábal-Hoyos et al. 2019).
It is important to note that all the MDR isolates harbored sul1, erm-B, and bla CTX-m-32 resistance genes. It has been reported that MDR genotypes of β-lactam-resistant Enterobacteriaceae which have mainly been identified for resistance to sulfamethoxazole, were shown to have resistance caused by bla CTX-m (Bajaj et al. 2016).
The highest percentage of MDR was detected in E. coli strains isolated from the soil (42%) samples, whereas MDR was not observed in E. coli isolates from the crop samples ( Figure 1). A previous study by Broszat et al. (2014) indicated that MDR presence is more noticeable in wastewater-irrigated soils (25%) than in rain-irrigated soils (6%). Water Science & Technology Vol 84 No 5, 1186 Frequency of integron harboring E. coli Integrons play a crucial role in the transfer and dissemination of antimicrobial resistance genes (Su et al. 2012). They can capture one or more gene cassettes and transfer them among different bacterial species, especially Gram-negative bacteria via the HGT mechanism (Su et al. 2012;Yuan et al. 2020). Our results showed the detection of intl1 in 46% (36 of 78) of E. coli isolates. The frequency of detection of intl1 in E. coli isolates is presented in Table 2. The highest level of detection was noted in STWW samples (35 of 62; 56%). A similar result was reported by Yuan et al. (2020) who found that 51.17% of E. coli strains isolated from municipal WWTPs harbored the intI1 gene. In contrast, Koczura et al. (2012) found that only 11% of E. coli strains which were recovered from WWTP harbored intl1.
Among the 36 isolates containing intI1, 75% (27 of 36) carried ARGs, indicating that 25% (9 of 36) of intl1 harboring isolates were free of ARG or carried other types of ARGs. Here, 25% of E. coli isolates recovered from the crop samples were integron positive whereas intl1 was completely absent in E. coli isolates of soil samples (Table 2). Furthermore, 6% (2 of 36) of integron harboring E. coli isolates were multiresistant. In the study carried out by Araújo et al. (2017) on irrigation of water and vegetables in household farms, 16% of E. coli isolates were found to be MDR among them, five isolates carried class 1 integrons.
As noted in previous studies, integrons are frequently found in livestock manure, urban wastewater, hospital wastewater (Yuan et al. 2020), agricultural ecosystems, and even in the water bodies not exposed to antibiotics (Araújo et al. 2017). The presence of intl1 imposes a higher potential for the dissemination of ARGs in the environment. Therefore, the consumption of crops contaminated with bacteria harboring intl1 may trigger a human health risk in terms of exposure to ARGs (Cerqueira et al. 2019a).
Spearman's correlation among ARGs and intI1 in the different sample types showed that the presence of erm-B gene was correlated with sul1 (r ¼ 0.52) and bla CTX-m-32 (r ¼ 0.39) ( Figure 2). As mentioned earlier, these three ARGs were detected in all MDR isolates which indicates that these ARGs may be present in the same MGE (Cerqueira et al. 2019b).
The results showed that sul1(r ¼ 0.30) and cml-A (r ¼ 0.31) positively correlated with intI1. The correlation of sul1 with intI1 is related to the location of sul1 gene on 30-CS of the classic class 1 integrons (Su et al. 2012). Cml-A is one of the ARGs which confers resistance to chloramphenicol. The positive correlation between other chloramphenicol resistance genes (like cat) and intI1 was noted previously (Wu et al. 2011). The lack of correlation between intI1 and other AGRs may be attributed to the following reasons. Although the intI1 is known as an effective marker of antimicrobial resistance (Osinśka et al. 2017), the occurrence of the gene is often associated with anthropogenic sources but not necessarily with the AR (Narciso-da-Rocha & Manaia 2017). Furthermore, integrons may be free of gene cassettes encoding AR. Recently, integrons without ARG cassettes were recovered from soils and sediments from the bacterial community (Zhang et al. 2009). However, the lack of significance may be related in part to the low numbers of E. coli isolates.

CONCLUSIONS
The results indicated the presence of E. coli isolates harboring ARGs and intI1 in the STWW, wastewater-irrigated soil and crop samples. Although the environmental conditions of arid and semi-arid areas may contribute to die-off of many ARB in Water Science & Technology Vol 84 No 5, 1187 soil and on crop surfaces, STWW could be a major source for ARB dissemination in the environment. Therefore, agricultural reuse of wastewater may be a risk factor for human health. Further studies are needed to estimate the risk of ARGs associated with the consumption of wastewater-irrigated crops for human health.