Different persistence among strains of E. faecalis and E. faecium in sterile treated wastewater microcosms: Effects of temperature and ciprofloxacin

Enterococci are Gram-positive facultative anaerobic bacteria that occur as single cells or in short chains. They are ubiquitous in nature as well as in the intestines of humans, mammals, and birds, and can be found in nearly all types of environments. The most often encountered intestinal enterococcal species are Enterococcus faecalis and Enterococcus faecium followed by Enterococcus durans and Enterococcus hirae (Gilmore et al. 2002). Enterococci are problematic due to their intrinsic and acquired resistance to a wide range of antibiotics. They may also contribute to the transmission of virulent and/or multi-drug resistant elements to other bacteria species.

Enterococci have the ability to survive adverse environmental conditions, such as a wide range of temperatures, pH, and salinities (Fisher & Phillips 2009), and are thus commonly used as indicators of faecal contamination in aquatic environments (USEPA 2009). Large amounts of enterococci reach water systems through wastewater treatment plants (WWTPs) (Blanch et al. 2003; Kühn et al. 2003). Release from WWTP is also suggested to be the main source of antibiotic resistant bacteria (ARB) of human origin into the environment as well as of antibiotics, thus promoting selection of ARB and/or antibiotic resistance genes (Michael et al. 2013; Rizzo et al. 2013).

Fluoroquinolones including ciprofloxacin (CIP) are a highly prescribed group of antibiotics used worldwide mostly due to its broad spectrum effect, quick absorption and penetration into tissues, and few treatment side effects (De Lastours & Fantin 2015). Fluoroquinolones are also excreted more or less intact in urine and bile. Indeed, as much as 50–80% of the fluoroquinolone CIP is reported to be excreted in an unmetabolized form (Schlender et al. 2018). These antibiotics are also very persistent, which leads to substantial concentrations in aquatic environments (Sodhi & Singh 2021).

The environmental stress that enterococci experience outside the intestine can generate different responses contributing to survival. Survival is promoted by transformation of the cells to a viable but non-culturable (VBNC) state, or a persister state. VBNC is often described as a subpopulation of bacteria that have lost their ability to grow under normal culturing conditions in the laboratory. Persisters are defined as ‘dormant’ cells that remain viable and culturable (Heim et al. 2002). Harsh conditions allow bacteria to become tolerant to antibiotics due to stringent response, as often seen in persisters (Gutierrez et al. 2017). The ability to survive antibiotics is thus not always an inherited trait (Ayrapetyan et al. 2018).

The aim of this study was twofold: (i) to further increase the knowledge of long-term effects of temperature on the persistence of different environmental isolates of E. faecalis and E. faecium in nutrient-poor treated wastewater, and (ii) investigate whether the presence of CIP alters the persistence of different sensitive and resistant strains of E. faecalis and E. faecium in treated wastewater.

In this study, 12 environmental enterococcal isolates were collected in a previous study (Ehn Börjesson et al. 2013) belonging to the species E. faecium (7 strains) and E. faecalis (5 strains, previously isolated from raw wastewater, treated wastewater, or duck faeces, and 3 clinical reference strains, E. faecalis CCUG 48513A, E. faecalis CCUG 41843, and E. faecium CCUG 45138) (Table 1). The non-clinical strains were selected on the basis of their susceptibility or resistance to CIP (R ≥ 4 mg L⁻¹) according to EUCAST (EUCAST 2019). Strains were confirmed to species level with MALDI-ToF MS (Bruker Diagnostics, Germany). Antimicrobial susceptibility to CIP (BioChemica, Germany) was verified and measured through antimicrobial susceptibility testing of minimum inhibitory concentration (MIC) in broth microdilution as described in ISO 20776-1 (ISO 2006). The temperatures used in this study correlate with water temperatures in WWTPs and recipients during summer (20 °C) and spring/autumn (≤10 °C) in a temperate climate (Annadotter & Forsblad 2012).

The strains were grown in 10 ml of Brain Heart Infusion Broth (Oxoid, UK) at 37 °C in 20 ml sterile Falcon tubes to stationary phase (OD₆₄₀ = 1.0–1.3), which corresponds to 10⁸–10⁹ CFU ml⁻¹ (ISO 2006). In total, 14 sterile plastic tubes (50 ml Falcon) were used as microcosms for every enterococcus strain, with two replicas for each sampling occasion. Treated wastewater was collected from the outlet to the receiving wastewater wetland at Hässleholm WWTP, southern Sweden. The water was filtered through 0.45 μm and thereafter through 0.22 μm cellulose nitrate filters (Whatman, UK), and autoclaved for 10 min at 121 °C. This additional sterilization of the effluent WWTP water was needed to eliminate natural biota. The sterilized wastewater was weighed (25 ± 0.1 g) to the microcosm tubes. The different Enterococcus strains were inoculated into each separate microcosm at an initial concentration of 10⁶–10⁷ cell ml⁻¹. For the temperature experiment, microcosms were incubated in darkness at 10 or 20 °C for 108 days. For the CIP experiment, microcosms were incubated at 20 °C with the addition of 0, 0.5, or 8 mg L⁻¹ CIP. The antibiotic was added to the microcosms 1 h prior to the addition of the bacterial strains. The mesocosms were incubated in darkness at 20 °C for 108 days.

Two microcosm replicates for each strain were harvested days 0, 3, 10, 24, 49, 75, and 108. Colony forming ability was determined in duplicates from each microcosm by plating 0.1 ml onto Brain Heart Infusion Agar plates, and incubation at 37 °C for 48 h. When needed, serial dilutions (1:10) in sterile 0.85% NaCl were performed prior to plating. Microcosm experiments incubated at 10 and 20 °C were repeated twice, and the impact of CIP at 20 °C was repeated three times (replicated experiments).

The antimicrobial activity of CIP in the antibiotic treatment experiments were controlled at the end of the experiment (day 108). This control followed the RAFs Disc diffusion assay test (EUCAST 2019) with the major modification that microcosm water replaced the antibiotic discs. Wells were made in the agar with the back of a sterile 1 ml pipette and filled with 0.1 ml of autoclaved microcosm water (121 °C for 5 min). E. coli CCUG19400 (Culture Collection University of Gothenburg, Sweden), sensitive to CIP, was used as a control. If a zone larger than 4 mm was observed, the antimicrobial activity of CIP was confirmed. In addition, all strains were also re-tested for CIP susceptibility at day 108.

All data were normalized by setting the start value (CFU ml⁻¹) to zero, and the maximum value to one. Hypothesis testing was performed with the statistical software SPSS v. 25.0. Generalized linear models (GLMs) were used to explore the effect of temperature (10 or 20 °C), and on the dependent variables Log CFU ml⁻¹, with the covariate factor time (days after inoculation of microcosms). Controlling factors were microcosm replicates and species. All factors were treated equally without assigning differential weights. In addition, GLMs were used to investigate the effect of CIP on the dependent variables Log CFU ml⁻¹. Time (days after inoculation) was included as a covariate factor and controlling factors were microcosm replicates and strains. We included biologically relevant two-way interactions in all GLMs, and we used a backward selection procedure in which non-significant interactions were excluded stepwise, to yield a simplified final model. The fit of the most general model was considered acceptable for all GLMs (lack-of-fit test: p > 0.05).

The susceptibility to CIP varied among strains included. Strains with an MIC lower than 4 mg L⁻¹ was regarded as susceptible according to EUCAST (2019). All clinical strains included were resistant and showed a high level of CIP-tolerance (>16 mg L⁻¹). Most of the environmental isolates were susceptible with three exceptions, see Table 1. The susceptibility or tolerance of the strains did not change over time. All strains showed the same breakpoint value for CIP (Table 1), irrespective of level of exposure to CIP (0, 0.5, and 8 mg L⁻¹) during the 108 days of the experiments.

Culturability was shown to be species-dependent. E. faecium strains were more persistent than E. faecalis, as measured by difference in culturability (p < 0.001). Both species showed higher culturability at 10 °C then at 20 °C over time (p < 0.001, Figures 1 and 2). One log reduction for E. faecalis at 10 °C was calculated to 92 days and to 43 days at 20 °C. The corresponding data for E. faecium was 333 days at 10 °C and 68 days at 20 °C.

The study examined also the culturability of the two species and 15 strains at zero, low (0.5 mg L⁻¹), and high (8 mg L⁻¹) concentrations of CIP over a period of 108 days. Contrasting interactions between CIP and strains within a species, both in the absence of CIP and at concentrations of 0.5 and 8 mg L⁻¹, revealed that the strain significantly influences culturability. Specifically, for E. faecium, comparisons between 0 mg L⁻¹ CIP and 0.5 mg L⁻¹ CIP (p < 0.001) and 0 mg L⁻¹ CIP and 8 mg L⁻¹ CIP (p < 0.025) demonstrated significance. Similarly, for E. faecalis, comparisons between 0 mg L⁻¹ CIP and 0.5 mg L⁻¹ CIP (p < 0.017) and 0 mg L⁻¹ CIP and 8 mg L⁻¹ CIP (p < 0.015) indicated significant differences.

No significant difference was seen for E. faecalis incubated with the low concentration of CIP (p = 0.680, data not shown) or the high concentration of CIP compared with microcosm with or without CIP addition (p = 0.616, Figure 3).

Five out of eight E. faecium strains showed the same culturability irrespective of whether CIP was added to the microcosm or not, i.e., no significant difference was observed either at 0.5 mg L⁻¹ CIP (p = 0.27, data not shown) or at 8 mg L⁻¹ CIP (p = 0.655; Figure 4).

Three E. faecium strains (DF19, RW118, and RW125) showed another pattern over time. These strains were negatively affected by 8 mg L⁻¹ CIP, when compared with microcosms without added CIP (p < 0.001, Figure 5). Strain RW125 was also negatively affected at 0.5 mg L⁻¹ CIP despite an MIC value of 2 mg L⁻¹ (p < 0.001, data not shown).

This study investigated various strains of E. faecium and E. faecalis, previously isolated from raw wastewater, treated wastewater, or duck faeces, to assess their capacity to endure prolonged periods in treated wastewater at two distinct temperatures or two different concentrations of CIP. The finding reveals that enterococci possess a noteworthy capability to survive and maintain culturability during extended periods, i.e. more than 108 days, in treated wastewater without any addition of nutrients. Although enterococci represent less than 1% of the gut microbiota in humans (Tendolkar et al. 2003), the content of enterococci in effluent wastewater can be as high as 10–10⁶ CFU 100 ml⁻¹, including antibiotic resistant strains (Vilanova et al. 2004; Gallert et al. 2005; Da Silva et al. 2006). Thus, one of the most important sources of enterococci of human origin in the environment is the discharge of wastewater to recipients. The higher the proportion of viable cells passing the wastewater treatment, the greater the risk that human enterococci or their resistance genes reach susceptible humans or animals.

However, lack of nutrients and other environmental stressors influence their culturability. In untreated wastewater, E. faecalis is reported to be culturable for extended periods compared with oligotrophic lake water, probably due to availability of carbon sources (Sinclair & Alexander 1984; Lleò et al. 2005). In the present study, the carbon content in the microcosm water was estimated to be around 29 mg chemical oxygen demand (COD) L⁻¹, of which about 14% bioavailable, data received from Hässleholm's municipal WWTP (Hässleholms Vatten 2012). Thus, the carbon content in the microcosms was too low to support enterococci growth but may have contributed to the prolonged culturability of persister cells.

Culturability of all strains was in this experiment directly influenced by temperature and significantly lower at 20 °C compared with 10 °C, irrespective of strain or species. Similar results have been reported previously, where both E. faecium and E. faecalis stayed culturable for longer periods at low temperatures (Lleò et al. 2005). In our study, E. faecium was found to be the most persistent of the two species (p < 0.001). However, this might be a strain-specific rather than species-specific behaviour, since other studies that used single strains have shown that E. faecium only remained culturable for up to 4 weeks (Lleò et al. 2001). E. faecalis is known to not only form persister cells that remains culturable but can also transform into the VBNC state (Lleò et al. 2001; Ayrapetyan et al. 2015). This means that the difference between survival of the species may be smaller than detected in this study. During the 3-month temperature experiments, the culturable fraction was not lower than Log₁₀ 3.5 for any of the tested strains, irrespective of species. The reduction of E. faecium at 10 °C was even lower, with less than 1 log reduction. This indicates the extended survival of E. faecium and E. faecalis in recipients, especially during periods of low temperatures. However, it is important to consider that the reduction of enterococci in natural systems also is influenced by other extrinsic factors, not included in this study, such as sunlight and predation (Byappanahalli et al. 2012).

Fluroquinolones interfere with bacterial replication and induce DNA strand breaks leading to an SOS response. In a recent study, Le Pont et al. (2024) showed that E. faecium persister cells, though genetically identical to logarithmically growing cells, exhibited higher levels of expression of genes related to stress response, DNA repair and DNA protection. The present study was designed to determine if enterococcal strains in treated wastewater were affected by CIP over an extended time period. The dominant part of the enterococci strains analysed were not affected by CIP, irrespective of strain-specific MIC values or CIP concentrations used, shown as equal culturability. This is in concordance with the physiological state reported for persister cells, namely a reversible shut down of metabolism and SOS-related response, leading to protection of the cell that otherwise would have been killed by bactericidal antibiotics (Ayrapetyan et al. 2018; Le Pont et al. 2024). However, three out of eight E. faecium strains were negatively affected by CIP. Two of these strains were affected at the low CIP concentration (0.5 mg L⁻¹), despite MIC values of 0.5 or 2 mg CIP L⁻¹. Persisters have been shown to be affected by antibiotics, yet the mechanisms remain unclear (Vogwill et al. 2016). Although there are indications that variants of quinolones may interfere with bacterial cell membranes (Gao et al. 2024). These findings further stress the need to include several different strains from the same species when studying antibiotic persisters.

This study confirmed the ability of E. faecalis and E. faecium strains to survive at nutrient-poor conditions such as treated wastewater at relatively high temperatures, but that lower temperatures favour culturability and survival. We could also show that the different isolates survived for as long as 3 months at CIP concentrations of 8 mg CIP L⁻¹, irrespective of strain-specific MIC values. Furthermore, the study shows that the culturability, here regarded as persister cells, is related to the species, but also to different strains within the same species. This became particularly evident in the presence of CIP. Thus, it is important to consider responses among the disparity of enterococci species, as well as intra-species variation in future studies regarding survival and selection through WWTPs or in other environments where enterococci occur as persistent cells.

We wish to acknowledge the following support: Kristianstad University for funding, Andreas Håkansson and Gunnar Gunnarsson for statistical advice. We are highly appreciative of these contributions.

This research received no external funding.

S-M.E.B. conceptualized the study, performed all experiments, wrote, reviewed, and edited the article. A-S.R.-H. conceptualized the study, wrote, reviewed, and edited the article.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.