Spring of Warta River as a Source of Antibiotic-Resistant Coliform Bacteria

Abstract

The aim of the study was to evaluate the occurrence of integron-carrying coliform bacteria in the spring water of Warta river, a major river in Poland. We isolated 18 strains (12 Escherichia coli, three Pantoea sp., 2 Citrobacter freundii, and 1 Klebsiella oxytoca) harbouring class 1 and/or class 2 integrons. Class 1 integrons contained aadA1, dfrA7, dfrA17, dfrA1-aadA1, and dfrA17-aadA5 gene cassette arrays, whereas the genetic content of class 2 integrons was stable and consisted of sat2-aad1 gene cassette array. The strains carrying integrons were resistant to 4–11 antimicrobials, most frequently to sulfamethoxazole, ampicillin, piperacillin, trimethoprim, and trimethoprim/sulfamethoxazole.

1. Introduction

Integrons are genetic elements capable of integration and rearrangements of gene cassettes that are often responsible for antibiotic resistance. An integron consists of an intI integrase gene, an attI recombination site, and a PC promoter that directs transcription of the integrated genes. Several classes of integrons have been distinguished upon the amino acid sequence of integrases. The most frequent are class 1 integrons, which are considered to play the main role in spread of resistance genes, because they are usually located within mobile genetic elements such as plasmids and transposons [1,2].

Integrons are widely dispersed in Gram-negative bacteria and occasionally detected in Gram-positive ones [3]. They have been discovered in clinical settings, but later also detected in bacterial strains inhabiting diverse environments, including those under low anthropogenic pressure, such as national parks or remote areas [2,4]. Integron presence has been proposed to serve as a marker of antibiotic resistance level [5] and anthropogenic pollution [6]. Bacteria carrying integrons have been isolated from various river waters, e.g., [7,8,9,10,11]; however, most of the studies have concerned anthropogenically impacted areas and/or the impact of the discharge of effluent from wastewater treatment plants, without taking spring waters into account. Hence, the aim of this study was to determine the presence of integron-harbouring coliform bacteria in spring water of the Warta river, the third longest river in Poland.

2. Materials and Methods

2.1. Sampling and Isolation of Bacteria

The Warta river runs in central-western Poland and has a length of 808 km. The river spring is localized in the suburbs of Zawiercie (GPS coordinates: 50.492° N; 19.4939° E), a town with a population of 50,000. Water samples from the spring of Warta river were collected three times: in March, May, and December. The samples were transported to laboratory in a TB50A (POL-EKO-APARATURA) temperature-controlled box at 4 °C and processed within 8 h. The total number of culturable heterotrophs was determined by plate counts on Brain Heart Infusion agar (bioMérieux, Marcy-l’Étoile, France). Coliform bacteria were isolated and counted on Brilliance™ E. coli/Coliform Agar (Oxoid, Basingstoke, UK). The isolates were identified with API 20E (bioMérieux). To avoid analysis of redundant clones, genetic relatedness of the isolates was determined by ERIC-PCR fingerprinting [12], followed by the analysis of band patterns by using Gel-Compar II v.3.5 software (Applied Maths, Sint-Martens-Latem, Belgium) with Dice similarity coefficient and unweighted pair-group method with average linkages (UPGMA) clustering method. Only genetically unrelated isolates with a DNA fingerprinting pattern identity below 90% were included in further analyses.

2.2. Detection of Integron Integrase Genes

The presence of class 1, 2 and 3 integron integrase genes (intI1, intI2, and intI3, respectively) in the genomes of coliform strains was determined by using multiplex PCR method according to Dillon et al. [13]. The amplification involved an initial denaturation (94 °C; 5 min), followed by 30 cycles of denaturation (94 °C; 1 min), annealing (59 °C; 1 min) and extension (72 °C; 1 min), with a final extension step at 72 °C for 5 min.

2.3. Analysis of the Variable Regions of Class 1 and Class 2 Integrons

Variable parts of the class 1 and class 2 integrons were amplified by using primers recommended by Lévesque et al. [14] and White et al. [15], respectively. In both reactions, the PCR amplification was conducted as follows: initial denaturation at 94 °C for 5 min, and 30 cycles of 94 °C for 1 min, 55 °C for 1 min, 72 °C for 5 min, and final elongation at 72 °C for 8 min. The amplicons were sequenced in a 3130xl Genetic Analyzer (Applied Biosystems, Waltham, MA, USA). Sequence data were assembled with DNA Baser (HeracleSoftware, Arges, Romania) and aligned with the available GenBank data using Nucleotide BLAST (Basic Local Alignment Search Tool). A gene cassette was identified if the identity on a nucleotide level was ≥95%.

All PCR reactions were carried out in a C1000 thermal cycler (Bio-Rad, Hercules, CA, USA). The products were separated in 1.5% agarose gel. Molecular weight of the amplicons was determined with GelCompar II 3.5 (Applied Maths).

2.4. Antimicrobial Susceptibility Testing

Antimicrobial susceptibility was determined with the standard disk diffusion method according to The European Committee on Antimicrobial Susceptibility Testing guidelines, with the EUCAST 11.0 version of breakpoint tables for interpretation of zone diameters [16]. The following antimicrobials were used: amikacin (30 μg), gentamicin (10 μg), ampicillin (10 μg), piperacillin (30 μg), piperacillin/tazobactam (30/6 μg), amoxicillin/clavulanic acid (20/10 μg), cefepime (30 μg), cefotaxime (5 μg), ceftazidime (10 μg), imipenem (10 μg), aztreonam (30 μg), ciprofloxacin (5 μg), chloramphenicol (30 μg), streptomycin (10 μg), trimethoprim (5 μg), and trimethoprim/sulfomethoxazole (1.25/23.75 μg). Production of extended-spectrum β-lactamases (ESBL) was checked by double-disk synergy test with ceftazidime, cefotaxime and amoxicillin/clavulanic acid. All antibiotic disks were provided by Oxoid.

3. Results and Discussion

Class 1 integrons are regarded as indicators of anthropogenic pollution and antimicrobial resistance [6]. We have previously reported the presence on integron-bearing coliforms in the water of Warta river [9,10,17], but those studies were focused mostly on the impact of anthropogenic pressure on the occurrence of integrons in Enterobacterales. In this paper, we describe for the first time the occurrence of integron-bearing bacteria isolated from spring water.

The total number of culturable heterotrophic bacteria in the water of the Warta river spring ranged from 2.0 × 10² to 5.2 × 10² CFU/mL. The number of coliform bacteria ranged from 1.0 × 10¹ to 8.6 × 10¹ CFU/mL and was lower in comparison to the values recorded for the Warta river in its middle course [10], but comparable to those determined for unpolluted rivers in southern Poland [18]. During three sampling events, 18 isolates of coliform bacteria were labeled positive for the presence of integron integrase genes. The isolates were identified as Escherichia coli (12 isolates), Pantoea sp. (3), Citrobacter freundii (2), and Klebsiella oxytoca (1). ERIC-PCR fingerprinting showed they were genetically unrelated. Nine strains harboured class 1 integron integrase gene, eight carried intI2 gene, whereas one strain had both intI1 and intI2 genes. No isolate with class 3 integrase gene was detected.

The frequency of class 1 integron carriage among coliform bacteria ranged from 0.1% in March and May to 1.0% in December. This is lower compared to the frequencies reported in Warta river near Poznań, the biggest city along its course, where it reached 6% upstream a wastewater treatment plant, and 14% downstream the point of the discharge of the final effluent of the wastewater treatment plant [17]. We noted a significant increase in the frequency of intI1-harbouring coliforms in winter. This confirms our previous findings which indicated that there is a seasonal impact on the frequency of integron-harbouring coliforms in water habitats [10].

Class 2 integrons were detected only during December sampling event and with relatively high frequency that reached 3.1% and exceeded that of class 1 integrons. While class 1 integrons are often reported from various aquatic environments, class 2 integrons are either not investigated or detected with frequency lower than that of class 1 integrons [5], so it is difficult to explain the reason of such high frequency of intI2 carriage in our study. Since it was noted in a single sampling event, one may presume it was incidental. All class 2 integrons in coliforms isolated from spring water contained a sat-aadA1 gene cassette array. The genetic content of class 2 integrons is usually stable, because the intI2 contains an internal stop codon, resulting in a non-functional integrase [5].

The strains carrying class 1 and/or class 2 integrons were resistant to 4 to 11 antimicrobials (

Table 1. Integrons and antibiotic resistance profiles of bacteria isolated from the spring of Warta river.

Strain	Isolation Month	Integrase Gene	Gene Cassette Array	Resistance Profile
E. coli 521/1	May	intI1	aadA1	GEN 1, PIP, TZP, AMP, FEP, CIP, SUL, STX, ATM, TMP, TET
E. coli 521/5	May	intI1	dfrA17	AMK, PIP, TZP, AMP, SUL, STX, TMP
E. coli 521/6	May	intI1	dfrA17	AMK, PIP, TZP, AMP, FEP, CIP, SUL, STX, TMP
K. oxytoca 522/1	May	intI1	dfrA17-aadA5	PIP, AMP, CAZ, FEP, SUL, STX, TMP, TET
E. coli 522/2	May	intI1	dfrA7	GEN, PIP, AMP, SUL, STX, TMP, TET
C. freundii 522/3	May	intI1	dfrA7	GEN, PIP, AMP, SUL, STX, TMP
E. coli 606/4	December	intI2	dfrA1-sat2-aadA1	PIP, TZP, AMC, AMP, FEP, SUL, STX, TMP, TET
E. coli 606/5	December	intI2	dfrA1-sat2-aadA1	AMK, PIP, TZP, AMC, AMP, FEP, SUL, STX, TMP, TET
E. coli 608/7	December	intI2	dfrA1-sat2-aadA1	AMK, PIP, TZP, AMC, AMP, FEP, SUL, STX, TMP, TET
E. coli 616/3	December	intI2	dfrA1-sat2-aadA1	AMK, PIP, TZP, AMC, AMP, CXM, FEP, SUL, STX, TMP, TET
Pantoea sp. 627/5	December	intI2	sat2-aadA1	AMK, CHL, FEP, SUL
Pantoea sp. 628/7	December	intI2	sat2-aadA1	AMK, CHL, PIP, FEP, SUL
E. coli 630/2	December	intI2	dfrA1-sat2-aadA1	AMK, PIP, TZP, AMC, AMP, FEP, SUL, SXT, TMP, TET
Pantoea sp. 631/3	December	intI2	dfrA1-sat2-aadA1	AMK, AMC, AMP, CXM, CTX, FEP, SUL, SXT, TMP
E. coli 635/3	December	intI1 + intI2	dfrA7 + sat2-aadA1	GEN, PIP, AMC, CAZ, SUL, SXT, TMP
E. coli 635/4	December	intI1	not determined	PIP, TZP, AMC, AMP, FEP, SUL, SXT, TMP, TET
C. freundii 636/3	December	intI1	dfrA1-aadA1	CHL, AMC, AMP, CAZ, FEP, SUL, SXT, TMP
E. coli 480/6	March	intI1	aadA1	PIP, AMP, CTX, ATM, SUL, SXT, TMP

¹ Antimicrobials: AMC—amoxicyllin + clavulanic acid; AMK—amikacin; AMP—ampicillin; ATM—aztreonam; CAZ—ceftazidime; CIP—ciprofloxacin; CHL—chloramphenicol; CTX—cefotaxime; CXM—cefuroxime; FEP—cefepime; GEN—gentamicin; PIP—piperacillin; SUL—sulfamethoxazole; SXT—trimethoprim + sulfamethoxazole; TET—tetracycline; TMP—trimethoprim; TZP—piperacillin + tazobactam.

) and were all multiresistant, i.e., resistant to antibiotics from at least three unrelated classes. The strains were most frequently resistant to sulfamethoxazole (100%) and ampicillin, piperacillin, trimethoprim and trimethoprim/sulfamethoxazole (88.9%). None of them was resistant to imipenem. Resistance to sulfamethoxazole is typical for class 1 integron-bearing strains due to the presence of sul1 gene in the 3′-conserved region of the integron [1].

Despite numerous reports on the integron-harboring bacteria in various aquatic habitats, there are few data regarding their presence in spring water. Ozgumus et al. [19] have reported an E. coli strain carrying a class 1 integron with dfrA17-aadA5 gene cassette array, isolated from spring water in Turkey.

The strain showed a resistance phenotype towards antimicrobials: ampicillin, gentamicin, netilmicin, tetracycline, chloramphenicol, and trimethoprim/sulfamethoxazole.

4. Conclusions

In conclusion, we report here the presence of integron-carrying coliform bacteria in spring water. The origin of these strains is not known. It is possible that they could have been transmitted from soil to ground water.