Pollution-Induced Tolerance of Soil Bacterial Communities to Oxytetracycline-Spiked Manure

The use of manure as a fertilizer is a common agricultural practice that can improve soil physicochemical and biological properties. However, antibiotics and their metabolites are often present, leading to the adaptation of soil bacterial communities to their presence. The aim of this study was to assess the effects of the extensively used, broad-spectrum antibiotic oxytetracycline on soil microbial community adaptation using a pollution-induced community tolerance assay. Manure-amended soil was spiked with oxytetracycline (0, 2, 20, 60, 150, and 500 mg kg −1 ) three times every ten days in the selection phase. The detection phase was conducted in Biolog EcoPlates with a second oxytetracycline exposure (0, 5, 20, 40, 60, and 100 mg L −1 ). All treatments demonstrated decreased metabolic activity after exposure to ≥ 5 mg L −1 oxytetracycline during the detection phase. Meanwhile, a signicant increase in tolerance was observed following exposure to ≥ 20 mg oxytetracycline per kg soil during the selection phase. Therefore, the pollution-induced community tolerance approach with Biolog EcoPlates was a useful system for the detection of antibiotic selection pressures on soil bacterial communities. It is important to properly manage animal waste before their application to the soil to reduce the occurrence of antibiotic-resistance in the environment.


Introduction
Livestock production has increased to meet the rising demand for food, with animal biomass for feed far Assessment of the potential risks of pollutants released into the environment is of utmost importance, particularly emerging contaminants such as antibiotics. Pollution-induced community tolerance (PICT) is an ecotoxicological tool that is used to determine the selective pressure of a pollutant and exerts a direct effect on the community (Schmitt et al. 2005). A PICT assay comprises of the following two phases: (i) a selection phase where the soil bacterial community is exposed to a concentration gradient of the studied pollutant, and (ii) a detection phase where the soils are subjected to a second pollutant gradient for the evaluation of tolerance levels. The increase in tolerance could be due to: (i) the replacement of sensitive species by more tolerant species, (ii) physiological changes that make the organisms less sensitive, or (iii) genetic changes through horizontal gene transfer (HGT) to acquire mobile genetic elements (MGEs) encoding for enhanced resistance (Schmitt et al. 2005). Cause-effect relationships between several pollutants and microbial communities using PICT are established for antibiotics (Schmitt et  The main objective of this study was to evaluate the effects of OTC exposure on soil bacterial communities in a PICT assay using Biolog EcoPlates. The effects of OTC exposure on soil bacterial metabolic activity, the number of utilized substrates, and the Shannon diversity index were analyzed. The OTC concentrations at which the metabolic activity decreased by half (EC 50 ) and the tolerance of soil bacterial communities was also assessed. The novel aspect of this study could be highlighted by the fact that we performed simulation of a repeated application of antibiotics into soils through manure application under controlled conditions. It is hypothesized that soils exposed to higher OTC concentrations demonstrate increased tolerance to OTC.

Collection and characterization of soil and manure
Soil sample for the experiment was collected from the upper 30-cm layer of a semi-natural grassland eld located in Derio, northern Spain, which has never been amended with inorganic or organic fertilizer to our knowledge. The soil was air-dried at 30°C for 48 h, following which it was sieved to < 4 mm and was subjected to physicochemical characterization according to standard methods (MAPA 1994). It was characterized as a clay loam with a pH of 6.2, 6.3% organic matter (OM), 0.32% total N content, with Olsen P and K + content at 3.4 and 395 mg kg −1 dry weight (DW) soil, respectively.
The aged manure (~two years old) from heifers was kindly provided by an organic dairy farm located in the province of Biscay, Spain, without a known history of antibiotic treatment. It was collected in polyethylene bags, following which it was immediately transferred to the laboratory and stored at 4°C; then, it was sieved to < 4 mm and subjected to physicochemical characterization (MAPA 1994). It had a pH of 8.7, 66% OM, 0.77% total N content, with Olsen P and K + content of 9.9 and 25 mg kg −1 DW soil, respectively. The following metal concentrations were determined following aqua regia digestion

Experimental design of the selection phase
Aged manure was manually incorporated into pots with one kg of soil and was thoroughly mixed to obtain an equivalent of 100 kg N ha −1 except for the control treatment (OTC0-M, no OTC and no manure). The pots were incubated at room temperature in the dark and allowed to exchange air for 55 days. During incubation, distilled water was added every week to maintain constant soil moisture.
Soil communities were exposed to a gradient concentration of OTC (CAS 2058-46-0, ≥95% purity, Merck) in the selection phase of the PICT assay. Each pot was contaminated by spiking with 60 mL OTC solution to obtain nal concentrations of 2, 20, 60, 150, and 500 mg OTC per kg soil (OTC2, OTC20, OTC60, OTC150, and OTC500, respectively). This spiking procedure was repeated three times on days 0, 10, and 20. OTC0-M and the other control pot containing manure with no OTC (OTC0) received an equivalent amount of distilled water. Each treatment was performed in triplicate.

Detection phase with Biolog EcoPlates
A second OTC gradient was established in the detection phase to reveal differences in community tolerance using 96-well Biolog EcoPlates. The plates contained a triplicate set of 31 relevant carbon sources for environmental samples (Insam 1997). Fresh soil equivalent to 5 g DW was added to 50 mL of autoclaved Milli-Q water, following which the mixture was agitated for 1 h in an orbital shaker (220 rpm) and then allowed to settle for 5 min. Subsequently, 450 µL of the liquid was mixed with 30 mL of Milli-Q water. This solution (100 µL) was mixed with OTC (20 µL), and then the mixture was aliquoted into each well to obtain the following six nal OTC concentrations: 0, 5, 20, 40, 60, and 100 mg L −1 . The plates were incubated at 30°C for 14 days (336 h), followed by color development and absorbance measurement at 595 nm using a microplate reader (Anthos Zenyth 3100, Anthos Labtec Instruments GmbH, Salzburg, Austria).

Data processing
Average well color development (AWCD) was determined by calculating the mean absorbance values of each treatment at each time point. The absorbance value for each well was corrected by subtracting the zero hour time point and the blank control for each reading time (Epelde et al. 2008). The values corresponding to the incubation time of the midpoint of the exponential portion of the curve representing the highest microbial growth rate were selected for further calculations. The number of utilized substrates (NUS) was calculated when the absorbance value was > 0.1 (Epelde et al. 2008). Similarly, Shannon's diversity (H') index was determined by considering the absorbance values at each well as equivalent to species abundance. Nonlinear curve tting based on the Gompertz function was performed to yield the kinetic parameters (Lindstrom et al. 1998): where K represents the asymptote that the absorbance curve approaches, r represents the exponential rate of absorbance changes, t represents the time following microplate inoculation, and s represents the time to the midpoint of the exponential portion of the curve when y = K/2. OTC concentrations that reduced the color formation to 50% of the maximum (EC 50 values) were determined. Bacterial community tolerance was quanti ed using an adapted tolerance index (TI) (Brandt et al.

Results
Dose-response curves showed that the application of Dose-response curves of each treatment with respect to the Shannon diversity index against different antibiotic concentrations in the detection phase are shown in Figure 3. When both OTC0-M and OTC0 treatments were exposed to 20 mg OTC L −1 , a notable decrease in H' was observed. The H' curve was similar in the OTC20, OTC60, OTC150, and OTC500 treatments, with each of these treatments having a signi cantly higher H' than the OTC0-M and OTC0 treatments. In summary, OTC exerted an inhibitory effect on H' and decreased as the antibiotic concentration in the detection phase increased.
The following average EC 50 values represent the OTC concentrations required to reduce the microbial activity by half in the tested treatments: 15.6, 17.1, 24.5, 30.2, 32.1, 38.9 and 49.0 mg L −1 for OTC0-M, OTC0, OTC2, OTC20, OTC60, OTC150, and OTC500, respectively (Figure 4). OTC150 and OTC500 treatments presented signi cantly higher EC 50 values than the OTC0-M treatment. In summary, higher OTC concentration in the selection phase resulted in the requirement of greater OTC concentrations (higher EC 50 ) in the detection phase to reduce the activity of soil bacterial communities by half.

Discussion
Soil microorganisms play essential roles in soil function, such as organic matter decomposition, nutrient cycling, and formation of soil structure ( (Epelde et al. 2008). In this study, the PICT assay and EcoBiolog approach were considered to assess whether OTC addition to soils exerted impact on microbial communities. In fact, the assessment of tolerance to a chemical helps infer causal relationships between exposure and effects (Tlili et al. 2015).
Antibiotics may be subjected to various processes in the soil environment, including sorption by soil components, transformation, photodegradation, and plant uptake and transport (Kumar et  The EC 50 helps estimate the concentration of a toxicant that causes a 50% reduction of test populations against a speci c endpoint under particular conditions (Rozman et al. 2010). An EC 50 of 79 mg sulfachloropyridazine kg −1 soil amended with pig slurry was found in a PICT assay with Biolog EcoPlates (Schmitt et al. 2005). The EC 50 values in this study ranged from 15.6 to 49.0 mg L −1 for the non-OTCexposed soil in the selection phase to the soil treated with the highest OTC concentration (500 mg kg −1 ), respectively. This indicated that soil exposed to higher OTC concentration in the selection phase required higher OTC concentrations in the detection phase to reduce their microbial metabolic activity by half.
In this study, the tolerance of soil microbial communities to OTC increased with higher OTC In another respect, differences between the treatment amended with aged manure and the unamended treatment were expected since the presence of metals in aged manure might promote the spread of antibiotic resistance. This occurs via the following co-selection mechanisms: (i) co-resistance, when different resistance systems are present in the same genetic element, and (ii) cross-resistance, when one resistance system confers resistance to both a metal and an antibiotic (Baker-Austin et al. 2006). However, statistically signi cant differences were not found between the treatments.

Conclusions
Repeated short-term contamination of soils by manure amended with OTC promoted antibiotic tolerance in microbial communities with possible adverse effects exerted on the environment. The repeated application of organic amendments containing antibiotics to soils is a common agricultural practice. Its use should be limited well below 20 mg OTC per kg soil to avoid the dissemination of antibioticresistance. To this end, it is important to effectively manage animal waste and to decrease the amount of antibiotics present in manure before applying them to the soil.
Although PICT is a sensitive method used for detecting changes in soil microbial communities due to OTC exposure, it is di cult to determine the cause of this tolerance. A metagenomic analysis may provide insights into the structure of microbial communities and their adaptation processes along the OTC gradient.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests. Editing.    OTC concentrations that reduced the AWCD to 50% of the maximum AWCD (EC50 values) have been tted with the Gompertz function(n =3). Different letters represent signi cant differences (p < 0.05) according to Tukey's post hoc test.