Culture-based study on the development of antibiotic resistance in a biological wastewater system treating stepwise increasing doses of streptomycin

The broad use of antibiotics in the fields of medical therapy, animal husbandry, and plant disease control during the past 65 years has resulted in the rapid growth and global prevalence of antibiotic resistance. In recent years, municipal wastewater treatment plants (WWTPs) have been considered as the main source for the development of antibiotic resistant bacteria (ARB) and antibiotic resistance genes (ARGs), as they consist of a mixed bacterial community maintained by rich nutrient environments, and experience repeated contamination by resistant bacteria and exposure to various antibiotics (Gallert et al. 2005; Novo and Manaia 2010; Rizzo et al. 2013; Michael et al. 2013; Guo et al. 2017; Mao et al. 2015; Rodriguez-Mozaz et al. 2015; Aubertheau et al. 2017). In comparison with municipal WWTPs, however, antibiotic production WWTPs could become a hotspot for the development of antibiotic resistance due to the presence of much higher concentrations of antibiotics (Pruden et al. 2013; Ashbolt et al. 2013). It is important, therefore, to know the threshold concentration of antibiotics that leads to antibiotic resistance.

Our previous study showed that the presence of high concentrations of antibiotics (oxytetracycline, > 19.5 mg L⁻¹) in a biological WWTP could lead to the occurrence of tetracycline resistance and MDR in treated wastewater and downstream rivers (Li et al. 2010). Similar results have also been observed in antibiotic production WWTPs accepting penicillin and spiramycin production wastewater (Li et al. 2009, 2011; Liu et al. 2014). However, most of these studies have been derived from investigative results of full-scale real WWTPs, and do not reveal the influence of antibiotic concentrations. Recently some effort has been made to investigate the influence of minimum antibiotic concentrations on the transfer of MDR under controlled experimental conditions. Under such studies, concentrations of tetracycline and sulfamethoxazole, as low as 10–100 µg L⁻¹ were found to promote MDR through the horizontal dissemination of mobile resistance elements (Jutkina et al. 2016; Kim et al. 2014). However, pure cultures and overnight culture times have often been adopted for these experiments, which may not be able to reveal the impact of antibiotic concentrations on the wastewater treatment bacteria.

In the current new approach, a biofilm-type wastewater treatment reactor was constructed to treat synthetic wastewater under stepwise increasing doses of STM (0, 0.1, 1, 5, 25 and 50 mg L⁻¹) over a period of 618 days, with a control system (no antibiotics) operated in parallel. Streptomycin is one of the first broad-spectrum antibiotics to be utilized for the control of infectious human and plant diseases. Although STM resistance mechanisms have been proposed in many clinical bacteria, plant and animal pathogens (De Leon Door et al. 2013; Sundin et al. 1995; Sundin 2002; Um et al. 2016), the resistance spectra in wastewater communities are still poorly understood. In additions, the culture-based wastewater bacterial responses against stepwise increasing doses of STM have not been studied. In this study, wastewater bacterial strains were isolated from six different STM exposures, and the impacts of antibiotic concentrations on wastewater bacterial diversity and the characteristics of antibiotic resistance were assessed using bacterial enrichment method, sequencing and, MIC assay. The presence and distribution of 11 antibiotic genetic determinants from individual wastewater bacterial isolates were determined by the conventional polymerase chain reaction (PCR). The results of this new culture-based stepwise increasing STM approach will enable us to better understand the risks accompanying antibiotic wastewater treatment.

Although culture-independent approaches (meta-genomic) are widely used in recent times, bacterial antibiotic prevalence and resistant patterns are more accurate and evident only by the culture-dependent approaches (Czekalski et al. 2012). In the present culture-based long-term study, bacterial classes including Gammaproteobacteria, Bacilli and Betaproteobacteria were dominant in the reactor regardless of the presence or absence of STM (Fig. 2). In addition, major genera included Aeromonas, Pseudomonas, Comamonas, and Bacillus was dominant in all six STM exposures (Table 1; Additional file 1: Table S2). In our previous culture-independent study, bacterial classes including Gammaproteobacteria, Betaproteobacteria and Bacteroidetes dominated in the aerobic-biofilm reactors mainly receiving the wastewater containing STM, whereas the major genera included Dokdonella, Pseudomonas, Desulfocapsa and Geobacter (Deng et al. 2012). There are three possible reasons for this difference in bacterial compositions. The mixture of carbon sources used in the study, including glucose, tryptone, starch and sodium-carboxymethyl cellulose might have favored the different composition of genera. The culture-based approach might also produce favorable conditions for some bacterial genera in the WWTPs (Bramucci et al. 2003). On the other hand, anaerobic genera including Desulfocapsa and Geobacter might have come from the two anaerobic reactors used in the previous study, which received beta-lactam antibiotics at a minimal level (1.7–2.1 µg L⁻¹). In this study, major bacterial genera and their populations were not disturbed even at the higher STM dose (25 and 50 mg L⁻¹). It reveals that increasing STM doses did not affect the major bacterial diversity of the wastewater treatment system. Therefore, it is possible that they maintained their presence in the reactor by adopting antibiotic resistance in stepwise increasing doses of STM.

In the present study, bacterial isolates from the STM exposure reactor exhibited multi-resistance to other unrelated classes of antibiotics including macrolides, β-lactams, sulfonamides and tetracycline (Table 2; Additional file 1: Fig. S3). In our previous studies, similar multi-resistance observations were also described in wastewater systems accepting penicillin and oxytetracycline (Li et al. 2009, 2010). These related effects have also been observed in other sources such as clinical pathogens, chicken farms and aquatic systems (Tacao et al. 2014; Wong et al. 2014; Levy et al. 1976). According to previous other study on the overnight culture-based activated sludge samples exposed to minimal doses of STM (> 0.1 mg L⁻¹) showed decreased bacterial multi-resistance by controlling the transfer of mobile genetic elements (Kim et al. 2014). In the current long-term study, minimum doses of STM (> 0.1 mg L⁻¹) exhibited increased multi resistance (Fig. 3), whereas the higher doses of STM (> 25 mg L⁻¹) exhibited stable multi-resistance. Similarly, on other hand, STM resistance also increased at minimum doses of STM (> 0.1 mg L⁻¹) and was maintained at the higher doses of STM (> 25 mg L⁻¹), whereas negligible STM resistance and multi-resistance were noted under no STM dose (0 mg L⁻¹). We considered, therefore, that minimum doses of long-term exposure STM might induce significant MDR and STM resistance simultaneously, and the resistance could be able to persist even under higher doses of STM among the wastewater bacterial community.

It is interesting that the responses of every culturable wastewater bacterial strain to the stepwise increasing doses of STM were variable and unambiguous. Chryseobacterium lactis and Comamonas testosteroni naturally (0 mg L⁻¹) possessed STM resistance genes (strA and strB) and a class I integron gene (3′-CS) (Table 3; Additional file 1: Table S7), and showed high MIC values against STM (> 256 µg mL⁻¹) in the exposure without STM (Additional file 1: Table S5). Meanwhile, Lactococcus chungangensis, Pseudoxanthomonas mexicana and Lactococcus raffinolactis also naturally (0 mg L⁻¹) possessed STM resistance genes (strA and strB) and an integron gene (3′-CS), but showed low MIC values against STM (< 6 µg mL⁻¹) under the exposure without STM. Subsequently, however the above isolates obtained STM resistance under the lower doses of STM (0.1, 1, or 5 mg L⁻¹). A similar previous observation suggested that the distribution of streptomycin MICs was influenced and controlled by STM resistance genes in E. coli (Sunde and Norstrom 2005). Furthermore, parameters such as the presence of selected gene cassettes and integron genes located on low copy number plasmids and their placement in 3′-conserved segments might lead to variable STM MIC results (Bryan 1984; Levesque 1994). We believe that the above parameters might be controlled by the threshold concentrations of selected antibiotics, which mostly fall in minimal dose ranges.

At the same time, Aeromonas allosaccharophila and Aeromonas veronii were not resistant to STM (< 8 µg mL⁻¹) and did not contain any ARGs naturally (0 mg L⁻¹). Subsequently, however, they became resistant to STM and ARGs were found (aacA4, strA, strB, aadA, 3′-CS, and Int1) in the corresponding STM exposures. This result indicates that STM resistance was promoted by stepwise increased STM concentrations, which might have occurred by the horizontal transfer of plasmids or transposons containing ARGs (Lekunberri et al. 2017; Van Overbeek et al. 2002; Rizzo et al. 2013; Czekalski et al. 2012). Interestingly, Bacillus anthracis was present and dominant in all six STM exposures (0–50 mg L⁻¹), but the MIC results showed high sensitivity to STM (< 6 µg mL⁻¹) and they were not grown in the growth agar media with low added STM dose (0.1 mg L⁻¹). Surprisingly, this bacterial strain exhibited strA, strB, aadA, aacA4, Int1 and 3′-CS as confirmed by conventional PCR. Bacillus anthracis is a spore-forming bacteria and the causative agent of anthrax disease. In general, spores of Bacillus sp. are dormant and resistant to unfavorable conditions such as extreme temperature, radiation, and antibiotics (Tetz et al. 2016; Severson et al. 2009; Nicholson et al. 2000). Previous findings revealed that bacterial spores can maintain their complete genome and antibiotic resistance genes might be involved in the horizontal transfer of ARGs to other bacterial species (Tetz and Tetz 2017; Barra-Carrasco et al. 2014). Furthermore, some ARGs are not expressed in the bacterial spores, but will be part of the resistome. In our study, we believed that spores might be produced under low STM exposure (0.1 mg L⁻¹) and persist until the high STM exposure (> 25 mg L⁻¹). However, further studies are needed to determine, whether the spores of B. anthracis play an important role in delivering STM resistance.

Non-pathogenic environmental bacteria such as Lactococcus chungangensis, Lactococcus raffinolactis, Chryseobacterium lactis, Aeromonas allosaccharophila and Pseudoxanthomonas mexicana showed the prevalence of strA, strB, aacA4, IntI and 3′-CS genes in the various STM exposures (0–50 mg L⁻¹) and subsequently obtained STM resistance (> 256 µg mL⁻¹) in the corresponding STM exposures. These results reveal that long-term increasing doses of STM can lead to non-pathogenic bacteria becoming perfect reservoirs of ARGs, which they might transfer to another bacterium via gene transfer, resulting in an abundance of MDR bacteria in WWTPs (Rizzo et al. 2013; Balcazar et al. 2015; Wellington et al. 2013).

In this study, most of the STM resistant strains from the reactor treated with various doses of STM (0–50 mg L⁻¹) consisted of a 3′ conserved segment (3′-CS) of the clinically important class 1 integron gene and class I integrase (Int1) gene (85–92.5%). In addition, amino glycoside resistance genes aacA4, strA, strB and aadA (67.5–82.5%) were also distributed in most STM resistant strains from all STM exposures. These amino glycoside genes and integron genes were notably observed under the lowest STM dose (0.1 mg L⁻¹). Therefore, minimum doses of STM might strongly induce the gene cassette of amino glycoside resistance genes among the wastewater bacterial community related to the clinical class I integron and class I integrase enzymes. Subsequently, these class I integrons might induce the dissemination of STM resistance to other wastewater bacteria through horizontal mobile gene transfer, which might then be maintained even under higher doses of STM (Mokracka et al. 2012; Masarikova et al. 2016; Said et al. 2016; Stalder et al. 2013; Nemec et al. 2004).

We also observed high prevalence of a non-STM resistance gene aacA4 among the isolates harvested from the STM reactor exposure by involving stepwise increasing doses of STM (0–50 mg L⁻¹). According to previous research, aacA4 is an amino glycoside gene that encodes the enzyme, amino glycoside 6′-N-acetyl transferase (AAC(6′)-IbC) and confers resistance to antibiotics such as tobramycin, gentamycin and kanamycin (Nemec et al. 2004; Sacha et al. 2012). In the present study, however a prevalence of the aacA4 gene was observed among 82.5% of STM isolates, equal to the prevalence of STM resistance genes strA, strB, aadA and aadE (15–82.5%). It is unclear, whether increased doses of STM modified the aacA4 gene to encode STM-modifying enzymes or if it accelerated an unknown STM resistance mechanism. Further research is required to clarify this as well as antibiotic risk assessment in WWTPs. In conclusion, our present work reveals that the wide bacterial diversity among the indigenous bacteria from WWTPs exposed to long-term stepwise increasing doses of STM could become an enhanced and stable reservoir for the development of STM resistance. Here we concluded that stepwise increasing doses of antibiotics will expose a number of unanswered questions related to antibiotic resistance among the wastewater bacterial community. We therefore suggest that the current new approach could apply and extended in future studies to better understand the risk assessment of WWTPs.