Pharmaceuticals in domestic and agricultural waste waters – problem and its solution

In recent times, the global production of human and veterinary pharmaceuticals has strongly risen, together with their affordability not only in developed countries but due to their availability and decreasing prices also in Third World countries. Moreover, several frequently used pharmaceuticals occur in rather high concentrations in ecosystems (Jux et al. 2002). Despite the global availability and use of synthetic drugs, the most polluted areas are particularly developed and industrial countries with a high living standard and population density (Kotowska et al. 2014). One of the most significant problems is persistence of the pharmaceutical residues and their metabolites in the environment. Many of them do not exhibit acute toxicity for water ecosystems, but have a cumulative effect on non-target organisms (Han et al. 2010; Saravanan et al. 2012).

Non-steroidal anti-inflammatory drugs (NSAIDs) are one of the most frequently used groups of remedies in the world. They inhibit production of inflammation mediators (prostaglandins) (Martin & Leibovich 2005). NSAIDs and their metabolites have been found in almost all water sources (rivers, ponds, municipal water, etc.) (Wang et al. 2011). Groundwater analysis showed that NSAIDs (e.g., ibuprofen, diclofenac, naproxen, and ketoprofen, Figure 1) together with paracetamol and antibiotics were the most abundant pharmaceuticals in the samples collected in Czech test localities as well as in European water systems (Helenkár et al. 2010; Kotowska et al. 2014).

There are several studies describing the degradation and uptake of pharmaceuticals in model plant systems under in vitro as well as hydroponic conditions (Kotyza et al. 2010; Zhang et al. 2013).

The aim of this work was to prove the possibility of utilization of constructed wetlands for elimination of selected pharmaceuticals and their degradation products from waste waters.

Our data presented here suggest that there are plants available to metabolize commonly used pharmaceuticals occurring in water environments. Their metabolic potential was proved by experiments with cell cultures of Melilotus officinalis and Rheum palmatum in vitro. These results were verified in reed (Phragmites australis L.) cells, tissues, and whole plants, and finally tested in real scale in constructed wetland with combined vertical and horizontal flow designed for agriculture waste water cleaning.

The seeds of reed (P. australis (Cav.) Trin. ex Steud.) were obtained from wild plants on a river bank in the area of central Bohemia. The seeds were cleaned in 70% ethanol for 1 minute and sterilized by 1% sodium hypochlorite supplemented with 10 mL L⁻¹ of detergent TWEEN 20 for a period of 10 minutes. Then they were rinsed three times in sterile water and placed on hormone-free solid RH medium (Hoagland medium modified for reed cultivation; Hoagland 1920) with 40 g L⁻¹ sucrose. Germination started after 7 days. Callus culture was achieved from primary callus rising on stem segments of reed seedlings after replanting them on solid RH medium supplemented by 2,4-dichlorophenoxyacetic acid (2,4-D) and kinetin in concentrations (0.225 mg L⁻¹ and 0.215 mg L⁻¹, respectively) used previously for callus induction of other species (Podlipná et al. 2008). Suspension cultures derived from this callus were grown in the dark at 24 °C, on a rotary shaker at 100 rpm in 250 mL flasks containing 100 mL RH medium supplemented with growth regulators 0.225 mg L⁻¹ 2,4-D, and 0.215 mg L⁻¹ kinetin.

The Melilotus officinalis and Rheum palmatum seeds were obtained from Seed Service s.r.o., Czech Republic. Callus cultures were initiated from the surface sterilized seeds, using the basal MS medium supplemented with 10 mg L⁻¹ 2,4-D, 1 mg L⁻¹ BAP, and 1 mg L⁻¹ kinetin and solidified with agar (0.7%). The callus culture was transferred into the liquid medium and cultivated in Erlenmeyer flasks on horizontal shaker in the dark at 25 °C to obtain suspension cultures. The suspension culture was subcultured at 2-week intervals. Ten grams of aseptically filtered cell mass was inoculated to 150 mL of fresh medium in every flask.

At the beginning of incubation, each flask containing 50 mL of medium with cell suspension (10 g FW) was supplemented with pharmaceuticals pre-dissolved in dimethyl sulfoxide (DMSO).

The final concentration of DMSO in medium was 0.1% (v/v) to prevent its harmful impact on cells. The suspensions were incubated for 4, 8, 24, 48, and 96 hours. Three flasks for each time point and drug were used. After the incubation, medium was taken up, placed into plastic tubes, frozen, and stored at −80 °C. The cell suspensions were repeatedly washed, transferred into tubes, and lyophilized. In chemical blank samples, medium containing the pharmaceuticals but not the cell suspension was incubated. In biological blank samples, the cell suspensions were incubated in a drug-free medium.

For high-performance liquid chromatography (HPLC) analysis, the samples (1 mL) were filtrated through nylon membrane filter (0.45 μm), collected in 1.5-mL Eppendorf centrifuge tubes and immediately frozen in a freezer (−20 °C) to inhibit further degradation. After thawing, samples were directly injected into the HPLC.

Samples of waste water and cultivation liquid media for UPLC/MS/MS were acidified with acetic acid at pH 2.5, then were filtered through nylon membrane filter (0.45 μm) as previously and 25 mL of sample was applied on SPE columns (Strata C8 (55 μm, 70 A), 500 mg/3 mL, Phenomenex, USA) preconditioned with 5 mL of methanol and 10 mL of deionized water. The columns were then washed with acidified water (acetic acid, pH 2) and eluted with 5 mL of methanol. Then, the samples were evaporated under a stream of nitrogen to dryness and stored at −80 °C. Before the analysis, the samples were dissolved in 1 mL of methanol and then dissolved 100× for analysis.

The plant tissues (1 mg) were ground to a fine powder with a mortar and pestle in liquid nitrogen and twice extracted with 100% methanol (4 mL) for 5 minutes in an ultrasonic bath followed by 30 minutes on a laboratory shaker. After centrifugation (10 minutes at 12,000 rpm) (centrifuge Hettich, Germany) the supernatant (2 mL) was moved into plastic test tubes and evaporated by a stream of N₂. Then, the samples were dissolved in water (5 mL), acidified to pH 2.5, and applied on SPE columns and then purified following the waste water sample method.

HPLC with UV detection was used for determination of the concentrations of selected pharmaceuticals. The media (direct injection of 100 μL of media) were analyzed with HPLC using UV detection (SM 5000, LDC Analytical, USA). A stainless steel column 4.6 × 250 mm platinum phenyl 5 μm (Alltech USA) using acetonitrile–water (83:17) as the mobile phase under isocratic conditions at a flow rate of 0.8 mL min⁻¹ was used. Measurements were repeated three times. Results were recalculated using calibration curves constructed for each compound using standards. The curve was linear in the range of 0.65–6.5 μg for each compound. The Clarity Chromatography Data Station with Clarity software (DataApex, Czech Republic) was used for evaluation of calibration curve.

For quantitative analysis, the system Q-Trap 4000 (AB Sciex, USA) was used. The stainless steel column Kinetex Phenyl-Hexyl 100 × 2.1 mm with particle size 1.7 μm was utilized. As a mobile phase, a water–methanol mixture under gradient conditions was used using flow rate 250 μL min⁻¹. Ibuprofen-d₃ was used as an internal standard. Compounds of interest were determined using specific MRM transitions (Yu et al. 2006).

The differences among treatments were tested by one-way ANOVA followed with Tukey HSD multiple comparison test. Significance level P ≤ 0.05 was used for both analyses. Each treatment was represented by four biological replicates. STATISTICA 8 (StatSoft, Tulsa, OK, USA) software was used for all the computations.

An experimental constructed wetland was built during 2012–2013. The wetland has been used since autumn 2013.

The system is located at an altitude of 355 m and the area is characterized by low precipitation (ca. 500 mm per year) and high average annual air temperature (8.4–8.8 °C).

Hybrid constructed wetland consists of pretreatment, a series of three constructed wetlands (HF1, VF, HF2; Table 1), and tertiary treatment is achieved in three small ponds with littoral zones connected by a meandering stream. The treated water is discharged into the existing pond (Figure 2).

Pretreatment consists of two accumulation tanks (10 and 16 m³) and a three-chamber septic tank (effective volume 14.1 m³). Outflow from the pretreatment stage feeds HF1 CW where reduction of TSS, BOD5, COD, and nitrate is achieved. The next stage, VF CW, is designed to achieve nitrification and the final stage HF2 CW is designed to denitrify. The discharged water flows into a sealed meandering stream overgrown by wetland plants to a series of three shallow ponds.

Wetland beds are planted with reed canarygrass (Phalaris arundinacea) and common reed (Phragmites australis). Meandering stream and littoral zones are planted with sweet mannagrass (Glyceria maxima), purple loosestrife (Lythrum salicaria), yellow iris (Iris pseudacorus), Siberian iris (Iris sibirica), sweet flag (Acorus calamus), beaked sedge (Carex rostrata), and broadleaf cattail (Typha latifolia). The design flow is 5 m³ d⁻¹ and theoretical retention time is 15.7 days. The tracer tests with KBr (Kadlec & Wallace 2009) indicated a retention time of 10 days.

Concentrations of four of the most used antiphlogistics were determined in the environment (cultivation media) of model plants and plant cells using an in vitro system. For the first screening, HPLC coupled with PDA detector was used, then the most sensitive and specific LC/MS/MS analysis was applied. The developed analytical methods were then used for assessment of concentrations of the NSAIDs in constructed wetland and real waste water (see Figure 3).

Accumulation/degradation of ibuprofen by Rheum palmatum and Melilotus albus is presented as an example of utilization of plants in a model system of suspension culture (Figure 4).

The above-mentioned results (Figure 3) show that nearly one-third of starting concentration of ibuprofen was taken up from liquid medium during the first hour after application. The proportional decrease of the ibuprofen concentration in tested cell suspension cultures was found to be 26% and 30% in M. albus and R. palmatum, respectively.

The same results were achieved using intact plants of Phragmites australis under in vitro conditions (Figure 5). In this case, degradation products (1-, 2-, and 3-hydroxyibuprofen, carboxyibuprofen) were also identified in the medium. Content of 2-hydroxyibuprofen, which is a major first-step degradation product of ibuprofen observed in the studied system, gradually increased during the monitored period.

The laboratory experiments with in vitro cultures of Arabidopsis thaliana, Melilotus officinalis, and Rheum palmatum proved the presence of several common metabolites of oxidation in cultivation medium containing common NSAIDs (Kotyza et al. 2010). In comparison with published data (Huber et al. 2009, Huber et al. 2012), the achieved results indicate the ability of the plant cells to biotransform human and veterinary pharmaceuticals, performing at least several detoxification steps.

The start of the construction of the hybrid constructed wetland is shown in Figure 6, and the final installation in Figure 7.

The experiment was focused on assessment of four selected NSAID concentrations in the system after application of doses of real waste water to the first sedimentation tank.

Sampling points (1–10) were selected in the outflow of sedimentation tank, outflow of each horizontal and vertical units, as well as in outflow of final cleaning ponds.

The analysis of ibuprofen, diclofenac, ketoprofen, and naproxen (Figure 8(a)–8(d)) show that starting compounds were degraded/accumulated mostly after passing horizontal flow units and only traces were found in vertical units. Concentration of all analyzed pharmaceuticals at the outflow from the constructed wetland system was close to and/or below the detection limits.

These results together with previous laboratory tests demonstrate the possible role of plant component of wetlands in removing pharmaceuticals from water environments, which has also been described in other model studies for naproxen (Matamoros et al. 2013; Zhang et al. 2013) and ketoprofen (Matamoros & Salvado 2012).

The detailed fate of selected pharmaceuticals is now under investigation.

The results of the present study show that widely used human pharmaceuticals such as NSAIDs are taken up and metabolized by higher plants. Experiments in laboratory in vitro conditions show that the concentration of model NSAID ibuprofen was significantly decreased by cell suspension cultures of two tested species, Melilotus albus and Rheum palmatum. The following findings obtained on intact plants of common reed (Phragmites australis), which is the dominant species of many wetland ecosystems around the world, growing in sterile conditions confirmed the absorption as well as metabolic degradation of added ibuprofen. Results achieved in laboratory conditions both during in vitro and hydroponic experiments were verified in real conditions in constructed wetland for ibuprofen and also for other widely spread acidic NSAIDs naproxen, ketoprofen, and diclofenac.

This work was supported by the Grant Agency of the Czech Republic project No. 14-22593S, by the Technology Agency of the Czech Republic project No. TA01020573 and OPPK project No. CZ.2.16/3.1.00/24014.