Liquid Chromatography Quadrupole Time-of-flight Mass Spectrometry Determination of Six Pharmaceuticals in Vegetal Biota. Uptake Study in Lavandula Dentata

1 LIQUID CHROMATOGRAPHY QUADRUPOLE TIME-OF-FLIGHT MASS 1 SPECTROMETRY DETERMINATION OF SIX PHARMACEUTICALS IN 2 VEGETAL BIOTA. UPTAKE STUDY IN LAVANDULA DENTATA 3 Sofía Barreales-Suárez (1), Manuel Callejón-Mochón(1), Stéphane Azoulay(3), Miguel Ángel 4 Bello-López(1), R. Fernández-Torres(1),(2)* 5 (1) Departamento Química Analítica, Facultad Química, Universidad Sevilla. 6 (2) Centro de Investigación en Salud y Medio Ambiente (CYSMA).Universidad Huelva. 7 (3) Université Côte d’Azur, CNRS, Institut de Chimie de Nice, France. 8 9 ABSTRACT 10 A procedure based on microwave assisted extraction for the determination of 6 11 pharmaceuticals in samples of Lavandula dentata, Salicornia ramosissima and Juncus 12 sp. by liquid chromatography-quadrupole time of flight mass spectrometry (LC- 13 QTOF/MS) was optimized and validated. Best results were obtained using microwave 14 assisted extraction of 1.0 g of homogeneous lyophilized samples and 5 mL of a mixture 15 ACN:H2O (1:1 v/v) as extracting solvent. Analytical recoveries ranged from 60 to 107 % 16 with relative standard deviation (RSD) lower than 15 %. Limits of quantitation (LOQ) 17 for the 6 pharmaceuticals flumequine (FLM), carbamazepine (CBZ), ciprofloxacin 18 (CPR), enrofloxacin (ENR), diclofenac (DCL), and ibuprofen (IBU) were in the range 19 20.8-125 ng g-1. The method was satisfactory applied for an uptake study in Lavandula 20 dentata samples finding quantifying concentrations of FLM and CBZ in roots, leaf and 21 stem. 22 23 Keywords: liquid chromatography quadrupole time-of-flight, pharmaceuticals, 24 microwave extraction, lavandula dentata, Juncus, Salicornia 25 ------------ 26 *Corresponding author: Rut Fernandez Torres, address: Facultad Química, Profesor 27 Garcia González s/n, 41012, Sevilla, Spain. E-mail: [email protected], +34954556442. 28 29 30 31 32 33 34 2 1. Introduction 35 It is well known that the processes used in wastewater treatment plants (WWTPs) do not 36 eliminate many substances including those namely as emerging pollutants that include, 37 among others, pharmaceuticals products (Tauxe-Wuersch et al, 2005; Hijosa-Valesco et 38 al., 2011; Al-Qaim et al., 2013; Al-Qaim et al., 2015). The use of treated wastewater to 39 irrigate many arid and semi-arid areas has become a usual practice years ago. 40 Additionally, since 1991 the European Union Regulations established that all its member 41 states must treat urban wastewaters prior to their discharge into riverbanks, lakes and seas 42 (EC council directive, 1991). 43 Surface and groundwater can contain variable amounts of active emerging pollutants even 44 at points far from urban and industrial effluents. Thus pharmaceuticals have been found 45 in all kind of waters and sediments around the world (Tiedekena et al. 2017), even in 46 those from protected areas such as Natural Parks (Cesário et al. 2016, Hrabovsky et al. 47 2004, Carmona et al. 2014). 48 In the last decades of the 20th century, environmental interest has focused on some kinds 49 of substances that until then had not been considered contaminants: the emerging 50 pollutants. It is also well known that several emerging pollutants can be easily 51 incorporated to the biota from the aquatic media. Thus, for example, some antibiotics 52 have been found in sea fish samples from the southern of Spain (Fernandez-Torres et al. 53 2011), green algal species have been proposed for depuration of pharmaceutical 54 pollutants in urban wastewater (Gentili et al. 2016) and diverse pharmaceuticals 55 (antibiotics, non-steroidal anti-inflammatory drugs, lipid regulators, steroid estrogens, β- 56 adrenoceptor blocker, antiepilectic drug, antifungal and antidepressant) have been found 57 in 14 aquatic species from a Chinese lake (Zhengxin et al. 2017). 58 The plants can incorporate, through the roots essentially, compounds dissolved in the 59 water that irrigates them. The incorporation of heavy metals to all kind of plants and the 60 analytical strategies for their analysis have been widely studied for many years 61 (Arasimowicz et al. 2013, Martínez-Alcalá et al. 2017). On the contrary, less attention 62 has been paid to how pharmaceuticals incorporate to plants. However, these processes 63 have been studied in detail in the case of vegetables intended for human consumption 64 (Riemenschneider et al. 2017a, Riemenschneider et al. 2017 b). Thus, Boxall et al. (2006) 65 and Marsoni et al. (2014), among others, showed that pharmaceuticals were accumulated 66 3 by plants when soil was irrigated at environmentally-realistic concentrations. Some 67 authors have measured variable amounts of antibiotics (Bao et al. 2010, Hu et al. 2010, 68 Yao et al. 2010) or antiepileptic drugs (Shernker et al. 2011) in vegetables and several 69 pharmaceuticals of diverse nature have been measured in root vegetables (Malchi et al. 70 2014). Other studies have shown that organic compounds from the soil and water can be 71 accumulated in different parts of the plants at different concentrations and rates (Al Nasir 72 et al. 2008, Redshaw et al. 2008, Shernker et al. 2011). 73 Additionally to uptake studies, the possible toxicity of certain pharmaceuticals for the 74 plants has been also demonstrated. Thus, the toxicity of ENR in four agricultural crops 75 was evaluated and it was found that water levels in the range 50-5000 µg L-1 produced 76 toxic effects. Likewise, it was demonstrated that the plants metabolized enrofloxacin to 77 ciprofloxacin, like in animals (Migliore et al. 2003). Kong et al. (2003) evaluated the 78 uptake and phytotoxicity of oxytetracycline to alfalfa. 79 The analysis of organic compounds in vegetables and plants frequently include long and 80 tedious extraction procedures followed by a clean-up step in order to eliminate interfering 81 substances co-extracted with the interesting analytes. In general, the usual extraction 82 procedures are ultrasounds assisted solid-liquid extraction or accelerated solvent 83 extraction on dried or lyophilized samples. On the other hand, solid phase extraction 84 (SPE) using different stationary phases is the most usual clean-up procedure. (Vaclaik et 85 al. 2014, Emhofer et al. 2017). 86 In this work, our main objectives were (i) to develop and validate an analytical method 87 based on high performance liquid chromatography coupled to quadrupole time of flight 88 (UHPLC-Q-TOF) mass spectrometry detection to determine FLM, ENR, CPR, CBZ, 89 DCL and IBU (see table 1 for physicochemical characteristics) in vegetal samples. 90 Previous studies have reported the presence of these compounds, among others, in 91 environmental samples (mainly water and sediments) due to a possible inefficient 92 removal in wastewater treatment plants or even by direct discharge of untreated 93 wastewaters (Gros et al. 2010, Camacho-Muñoz et al. 2010, Camacho-Muñoz et al. 2012, 94 Verlicchi et al. 2012). Great interest show their presence in protected areas like Doñana 95 National Park among others (Camacho-Muñoz et al. 2013, Carmona et al. 2014, Rivera- 96 Jaimes et al. 2018) where their presence might suppose a potential risk for the 97 surrounding ecosystem. The vegetal species selected for this study were Lavandula 98 4 dentata, Salicornia ramosissima and Juncus sp. These plants were selected for several 99 reasons, they are much extended in the Mediterranean area ecosystems where several 100 protected areas are located, they grow in different habitats which is important to evaluate 101 the global risk on wild species, and lavandula was easily accessible to overcome an uptake 102 study in the laboratory. Juncus sp. is an aquatic specie that usually grow in river banks 103 and wetlands margings, Salicornia is a plant that grows in coastal salt marshes and inland 104 salty habitats and Lavandula is a non-aquatic specie very extended through Mediterranean 105 ecosystems. (ii) After validation, the optimized procedure was applied to overcome a 106 study to evaluate the uptake of the selected drugs in plants irrigated with a mixture of 107 them at three different levels of concentration for a period of one month. The analysis of 108 the drugs was realized in leaves, stems and roots of the plant at different times in order to 109 evaluate different behaviours in the ability of the plant to uptake the analytes through the 110 time. Usually, these studies are carried out in short or long periods of irrigation, in this 111 paper we have performed the uptake study of the selected drugs at several irrigation times 112 in order to evaluate a possible change in the plant uptake behaviour throughout the time. 113 Table 1. Physicochemical properties of selected pharmaceutical chemicals. 114 Compound Structure Therapeutic class log Pa pKaa log Kow Ciprofloxacin Antibiotic 0.65 6.43 0.28a Carbamazepine Antiepileptic agent 2.67 13.9 2.45b Diclofenac Non steroidal antinflamatory 4.02 4.15 4.51c Enrofloxacin Antibiotic 1.88 6.43 0.70a Flumequine Antibiotic 2.41 5.70 2.60a Ibuprofen Non steroidal antinflamatory 3.72 4.41 3.97c a Predicted values from database of Royal Society of Chemistry: http://www.chemspider.com. 115 b Dal Pozzo et al. 1989. 116 c Avdeef et al. 1997 117 118 119 The procedure substitutes the traditional solid phase extraction clean up by a dilution of 120 the extract from a microwave assisted extraction to minimize the matrix effect obtaining 121 good results in terms of recovery for the selected analytes. The simple procedure of 122 5 extraction allow the sample processing in a very short time as well as not loosing of 123 information of the processed sample which would be compatible with the searching of 124 metabolites of the selected compounds in future works. 125 2. Experimental 126 2.1 Chemicals and reagents 127 Solutions and dilutions were prepared using ultrapure water from a Milli-Q Plus 128 (Millipore, Billerica, MA, USA) water purification system. Formic acid 98-100% 129 Suprapur was supplied by Merck Darmstadt (Germany), acetonitrile and water LC-MS 130 Ultra Chromasolv® were from Sigma-Aldrich (Madrid, Spain), sodium hydroxide in 2- 131 propanol:water and Leucine Enkephalin were purchased from Waters (Barcelona, Spain). 132 2.2 Standards and stocks solutions 133 FLM, ENR, CPR, CBZ, DCL and IBU were purchased from Sigma–Aldrich (Madrid, 134 Spain). The purity of all the standards was higher than 98% in all cases. 135 Methanolic stock solutions 100 µg mL-1 were prepared for all the pharmaceuticals except 136 CPR that was prepared by diluting 2.5 mg in 5 mL ultrapure water to a final volume of 137 25 mL with methanol. Aqueous working solutions of the studied compounds were 138 prepared daily by adequate dilutions from 100 µg mL-1 stock solutions. 139 2.3 Sample collection and preparation. 140 Blank samples of Lavandula dentata, Salicornia ramosissima and Juncus sp for 141 optimization purposes were collected from Doñana´s Natural Park area. Samples for 142 method development and validation were cleaned of remnants of soil with tap water first, 143 subsequently with deionized water and finally dried with a paper towel. After that, the 144 samples were lyophilized before being analysed by a freeze dry system (FreZone 2.5, 145 Labconco, Mo, USA). Then they were homogenized using a grinder and kept at −80 ◦C 146 until analysis. 147 To optimize the method, lyophilized samples were spiked with 5 mL aqueous solution, 148 containing the analytes at the adequate concentration, added drop by drop using stir bar 149 magnetic agitation to assure that the compounds spread throughout the sample and were 150 in sufficient contact with the matrix. The spiked samples were left in contact for all night 151 6 at 4ºC and then freeze for 2 hours at -80ºC. The freeze samples were lyophilized again to 152 eliminate the solvent and then submitted to the extraction procedure. 153 Samples for the uptake study were purchased in a local plant nursery and kept in our 154 laboratory, under adequate light and water conditions for two weeks until the assay. 155 Before extraction, plant leaves, stern and roots were washed with tap water to remove soil 156 residues, then rinse with deionized water and dried with a paper towel. Then, each part 157 of the plant was freeze at -80 ºC for 24 h and lyophilized. Dried samples were ground to 158 powder and stored at -80°C until extraction. 159 2.4 Sample extraction 160 5 mL of a mixture ACN:H2O (1:1 v/v) were added to 1.0 g of homogeneous lyophilized 161 samples. Microwave energy was applied at 50 watts for 5 minutes to achieve the 162 extraction. After that, the samples were centrifuged during 15 min at 6000 rpm and 1 mL 163 of supernatant was taken and diluted 1:5 (v/v) with a mixture ACN:H2O (1:1 v/v). Finally, 164 the extracts were microfiltered through 0.22 µm PTFE filters (VWR, Spain) before 165 injection into the UPLC system. 166 2.5 Liquid chromatographymass spectrometry analysis 167 The analysis of the selected pharmaceuticals was performance in an Acquity ultra- 168 performance liquid chromatography (UPLC) system (Waters, Milford, MA, USA) 169 coupled to a Xevo G2S QTOF mass spectrometer (Waters, Micromass, Manchester, UK) 170 with an orthogonal Z-spray lockspray and dual electrospray interface (ESI). Detection 171 was achieved operating in positive and negative ionization modes. The LC separation was 172 performed using a Zorbax Eclipse XDB-C18 analytical column (100 mm × 3.0 mm i.d., 173 3.5 µm particle size) preceded by a guard column Zorbax Eclipse XDB C-8 (12.5 mm × 174 2.1 mm i.d., 5 µm particle size) (Agilent Technologies Spain, Madrid, Spain). The column 175 temperature was maintained at 25ºC. The mobile phase used for MS and MS/MS 176 experiments consisted of aqueous formic acid solution 0.05% (solvent A) and acetonitrile 177 (solvent B). Two different gradient elution programs at a flow rate of 0.5 mL min-1 were 178 used for both positive and negative ionization modes. The elution in positive ionization 179 mode starts at 90% A and 10% B increasing % B from 10% to 90 % in 8.0 min, returning 180 then to initial conditions in one minute. Three minutes were waited between injections 181 which allowed re-equilibration of the column to initial conditions. Negative ionization 182 7 elution programme starts at 70% A increasing the B percentage until 100% in 7.0 min 183 returning to initial conditions in one minute more. An equilibration time of 2 min was 184 needed before next injection. The injection volume was 5 µL in both cases. 185 2.6 Mass detection conditions 186 The nebulisation gas (nitrogen, supplied by a high purity nitrogen generator Nitrogen 187 Zefiro 35 LC-MS, Cinel-gas, Italy) was set at 600 L/h and the cone gas (nitrogen) was set 188 to 30 L h-1, desolvation temperature was set to 400ºC and source temperature to 100ºC 189 for positive polarity and 120ºC for negative polarity. MS and MS/MS experiments were 190 operated in both (negative and positive) ionization modes. The resolution of TOF mass 191 spectrometer was approximately 30.000 at full width half maximum at m/z 556 and 192 22.000 at m/z 554. The capillary voltage was set to 1.5 kV for positive ionization and - 193 2.00 kV for negative, and sample cone voltage at 40 V. For MS spectra data being 194 collected in continuum between m/z 50 and 1000 in a scan time of 1.0 seconds. For 195 MS/MS experiments the following conditions were used: cone voltage 40, collision 196 energy (see table 2), collision gas was argon (99.995%, Praxair). The precursor ion was 197 selected using the quadrupole analyser, fragmented in the collision cell and subsequently 198 product ion spectra was recorded in the TOF analyser from m/z 50 to 800. All the obtained 199 spectra were lock mass corrected. An independent reference spray (lockSpray) to ensure 200 accuracy and reproducibility was used for all analyses. Leucine Enkephalin (200 pg µL-1 201 in ACN/H2O 1:1 v/v at 0.1% HCOOH) was used as lock mass (m/z [M+H]+ 556.268 and 202 m/z [M-H]- 554.2615) at a flow rate of 10 L min-1. The lockSpray frequency was set at 203 30 s, meaning that every 30 s the flow from the lockSpray needle was introduced into the 204 mass spectrometer for 0.3 s, thus giving the software the possibility to perform on going 205 correction of the exact mass of the analyte. The lockSpray capillary was set at 2.54 KV. 206 Data were averaged over 3 spectra min-1. Xevo G2S Q-TOF mass spectrometer was 207 calibrated in mass range of m/z 50-1200 with a mixture of NaOH 0,1M and formic acid 208 of 10% in ACN:H2O (80:20 v/v) daily. Data were acquired and processed by MassLynx 209 v4.1 software. 210 2.7 Method Validation 211 The proposed method was validated checking the quality parameters of the analytical 212 method. We have mainly considered the items related to the estimation of the well-known 213 performance characteristics parameters. The following criteria were considered: linearity, 214 8 sensitivity (limit of detection and quantitation), matrix effect, precision and recovery 215 (trueness). 216 Linearity was evaluated by matrix-matched calibration curves by spiking blank extracts 217 at ten concentration levels in the range 0.5 to 500 ng mL-1. 218 Instrumental detection and quantitation limits (ILOD, ILOQ) and method detection and 219 quantitation limits (LOD, LOQ) were calculated as the minimum concentration of analyte 220 giving peaks whose signal to noise ratios are 3 and 10, respectively, for each vegetal 221 matrix in spiked samples (n=3). 222 To evaluate the precision of the method, the repeatability (in one single day) and 223 intermediate precision (in different days) were studied. Precision study was performed 224 using the prediction of actual concentrations from the validation standards selected for 225 the analytical assay and calculating a RSD. 226 Matrix effect was evaluated by comparing the slopes of the matrix-matched calibration 227 for each part (leaf, stem and root) of the plant for the three selected plants, and a 228 calibration prepared in solvent (external calibration). A matrix effect percentage (ME%) 229 was calculated by ME (%) = [(Sm/Se)-1]x100 where Sm is the slope of the matrix- 230 matched calibration and Se the slope of the external calibration in solvent, considering 231 that 0% indicates no matrix effect, and values above indicates enhancement effect and 232 values below indicates ion suppression. 233 The trueness was evaluated by means of recovery assays on spiked samples (see section 234 2.3 for details). The recoveries were calculated as percentage of extracted compound at 235 low and high level of concentration for each kind of matrix and compound. Spiked 236 samples at 25 (125 for IBU) and 500 ng·g-1 final concentration (in triplicate) were 237 submitted to the extraction procedure. 238 2.8 Uptake study design 239 Lavandula dentata plants were purchased in a local plant nursery in order to ensure the 240 homogeneity of the sample as much as possible and then they were subjected to an uptake 241 study of a mixture of the six drugs for 30 days. Before beginning the experiment, the 242 plants were left to achieve acclimatization to the new light and temperature conditions for 243 9 two weeks and those that were not healthy or seemed to be affected by the change of 244 environment were discarded. 245 Eight plants of 30-40 cm length were used for the experiment at three concentration 246 levels: 10 ng·mL-1 (low level of exposition); 700 ng·mL-1 (medium level of exposition) 247 and 10 µg·mL-1 mL (high level of exposition). Two plants per level were irrigated with 248 20 mL of solution, containing all the analytes in the concentration described, every 2 or 249 3 days for one month by pouring the solution directly onto the soil. The low volume of 250 water used assured that all of the solution containing the analytes remained in the soil 251 until the next water addition. Additionally, two plants were selected as blanks (controls) 252 and they were irrigated with tap water following the same irrigation pattern as uptake 253 study plants. 254 Cuts of leaves and stems were made every 7 days in order to evaluate the accumulation 255 of the compounds through time. Once the assay was finished, roots were also collected. 256 3. Results and discussion 257 258 3.1 UPLC-QTOF/MS method development 259 Initially, mass spectra for each target compound individually was analyzed by direct 260 infusion of 0.25 µg· mL-1 aqueous standard solutions into the mass detector, looking for 261 the optimum conditions for all compounds that produce less fragmentation. Several 262 ionization conditions were studied including the use of formic acid, ammonium formiate, 263 acetonitrile and methanol as modifiers to evaluate the enhancement or suppression effects 264 on the ionization produced by their presence. Both negative and positive ionization modes 265 were evaluated finding that DLC was also detected in both modes. Additionally, adducts 266 formation ([M+H]+; [M+Na+] and [M+NH4+]) was also evaluated finding that for all 267 compounds under the optimum conditions the [M+H]+ and [M-H]- were the most 268 abundant ions. Desolvation and interface temperatures were high to produce efficient 269 ionization due to the high percentage of water in the mobile phase. Table 2 shows the 270 retention times as well as the elemental compositions for all the analysed compounds. 271 Full MS acquisition mode was used for quantification purposes using the extracted ion 272 chromatograms of [M+H]+ and [M-H]-. MS/MS experiments for confirmation purposes 273 were realized for each compound evaluating several collision energies in the range 6-20 274 16 concentrations of 50 and 500 ng·g-1 for positive ionization compounds and 150 and 500 ng·g-1 were submitted to the extraction procedure described above. Table 4 depicts the results, and as it can be seen, the overall recoveries ranges from 60 to 107 %. Higher recoveries were obtained for CBZ in all the matrices evaluated which is in accordance with results obtained by other authors previously for vegetal matrices (Marsoni et al. 2014, Riemenschneider C et al. 2017b). In general, recoveries for ENR and CPR were 70% approximately. This results are quite similar to those reported by Hu et al. 2014 in vegetal matrices. IBU showed 70% recovery for most matrices and DCL showed recoveries ranged between 80- 90 % as average. Recoveries reported by Montemurro et al. (2017) in lettuce for IBU and DCL (in the order 90-120%) were higher compared to our data. FLM was the analyte that showed more dependence with the matrix, thus results for stem are in all cases nearly 90-100% but low recoveries around 60% were obtained with root of salicornia and lavandula. The results were corrected according to the obtained recoveries percentages. Figure 2 shows a typical chromatogram of Juncus stem extract spiked at 50 (150 for IBU) ng·g-1 of each compound and figure 3 the corresponding full mass spectra obtained at the retention time of each analyte. 17 Table 4. Recoveries (average % ±SD, n=3) for each matrix at low and high spiked concentration level. FLM CBZ CPR ENR DCL IBU 50 ng·g-1 500 ng·g-1 50 ng·g-1 500 ng·g-1 50 ng·g-1 500 ng·g-1 50 ng·g-1 500 ng·g-1 50 ng·g-1 500 ng·g-1 150 ng·g-1 500 ng·g-1 Salicor. Stem 100 ± 15 105 ± 10 101 ± 7.6 90.3 ± 5.7 75.7 ± 2.8 69.2 ± 5.6 62.5 ± 1.5 71.2 ± 3.6 95.6 ± 3.4 99.8 ± 2.4 75.2 ± 19 69.8 ± 7.6 Root 67.1 ± 6.4 87.1 ± 10 66.5 ± 0.6 67.1 ± 6.5 75.7 ± 0.2 73.2 ± 6.5 74.1 ± 0.2 75.6 ± 6.9 61.5 ± 3.8 89.2 ± 7.6 82.7 ± 1.1 75.4 ± 4.6 Juncus Stem 89.6 ± 8.4 77.3 ± 9.0 99.5 ± 1.3 89.9 ± 9.4 72.1 ± 0.5 65.3 ± 7.9 72.1 ± 0.1 75.0 ± 5.0 101 ± 13 95.1 ± 5.1 69.3 ± 9.4 74.1 ± 7.7 Root 97.7 ± 5.5 87.6 ± 5.9 100 ± 8.7 97.7 ± 3.4 71.2 ± 3.6 81.6 ± 7.3 74.5 ± 0.6 78.5 ± 6.4 78.1 ± 1.6 88.8 ± 2.9 65.7 ± 1.1 70.0 ± 6.6 Lavandula Leaf 100 ± 8.9 90.1 ± 11 88.4 ± 5.4 90.6 ± 9.9 70.4 ± 2.2 77.0 ± 8.4 71.2 ± 5.2 71.3 ± 4.3 100 ± 6.8 89.7 ± 9.6 81.9 ± 5.1 67.9 ± 2.3 Stem 87.8 ± 7.8 67.8 ± 12 77.5 ± 0.2 77.8 ± 7.2 78.2 ± 2.3 82.3 ± 4.9 78.3 ± 8.4 77.9 ± 2.3 68.9 ± 3.4 96.3 ± 6.9 70.5 ± 1.5 70.7 ± 4.8 Root 60.1 ± 2.2 70.3 ± 5.2 107 ± 8.9 80.1 ± 6.2 72.5 ± 0.3 66.8 ± 2.6 66.9 ± 0.8 74.1 ± 9.3 72.1 ± 1.1 92.8 ± 12 79.4 ± 5.6 73.2 ± 0.9 18 3.4 Uptake study in Lavandula dentata. 336 Lavandula dentata plants were submitted to the uptake study described in section 2.8. 337 The plants that were irrigated with 700 ng·mL-1 and 10 µg·mL-1 solutions, showed a sick 338 appearance “nearly dead” after 21 days exposition, whereas the rest of the plants remained 339 alive for all the assay. The plants were submitted to the procedure described in the above 340 sections and finally analyzed (a triplicate of analysis was performed at each cut) in the 341 UPLC-QTOF/MS instrument. The results are detailed in table 5. Only FLM and CBZ 342 were detected in the plants. This fact, previously reported by other authors, is due to 343 chemicals uptake depends on great extent on the hydrophobic nature (log P) and 344 ionization state of the molecules which highly depends on the pKa and the pH of the soil 345 (Marsoni et al 2014, Wu et al. 2013). Thus, in our study, drugs with different octanol- 346 water partition coefficients and pKa values were selected. FLM and CBZ were quantified 347 in all samples nearly at all concentration levels, finding, as expected, higher 348 concentrations in the plants at high irrigation concentration. As it can be seen, data show 349 a clear trend of accumulation in root samples for both drugs, reaching very high 350 concentrations in the case of high level exposure concentration for CBZ. However in leaf 351 samples, it can be seen how initially the plant accumulates the highest concentration of 352 drug and later these values gradually decrease until almost disappears, this results do not 353 agree with those previously reported by Tanoue et al. (2012) and Wu et al. (2014) who 354 found higher concentrations for CBZ in leaves than in roots. This could be due to the 355 damage suffered by the plant throughout the exposure, which would cause the leaf to stop 356 absorbing and what it is observed is the degradation of the drug until the plant ends up 357 dying. The same behavior was followed by FLM. Nevertheless, a different behavior was 358 observed in stem samples, FLU concentration remained stable until 21 days and then a 359 considerable decrease in concentration was observed probably due to the damage suffered 360 by the plant. A quite similar behavior was followed by CBZ at low concentration level of 361 irrigation but a surprising increase in concentration was observed at high concentration 362 levels of exposure which would disagree with the above suggested damage in the plant. 363 At medium exposure level, both plants showed the same anomalous behavior, thus a high 364 accumulation was suffered by both plants at 21 days and then decrease, however 365 concentrations at 7 and 14 days were very low compared with the rest of results obtained 366 by the other samples which would suggest a problem that could alter the absorption of 367 CBZ by the plant at the beginning of the assay. 368 19 CPR, ENR were not found in any sample, however they were expected to be found as 369 their absorption have been previously reported by Lillenberg et al. 2010 and they low 370 values for log P and log Kow. This could be due to the plant physiology itself or 371 compound degradation in the soil. 372 DCL and IBU were not identified in any samples probably due to their low pKa value the 373 were mainly present as anions which implies low absorption by the roots (Marsoni et 374 al.2014). This results agree with some published previous papers (Tanoue et al. 2012, 375 Montemurro et al. 2017), however Emhofer at al. (2017) found unaltered DCL in cress 376 roots cultivated under hydroponic conditions after seven days irrigation, this discrepancy 377 might be attributed to the low potential for uptake of these pharmaceuticals (see table 1, 378 log P and log Kow show high values) due to its low lipophicity or even due to the 379 experimental conditions (Marsoni et al. 2014). 380 From the data depicted in Table 5 some other conclusions can be drawn. As it can be seen 381 in the minigraphs included in the table, the accumulation trends in leaves for CBZ and 382 FLM do not depend on the irrigated concentration, showing quite the same behavior for 383 the three concentrations studied. Initially a strong adsorption occurs that subsequently 384 decreases with time as already mentioned above. However, this behavior is different in 385 stem samples that show different trends at the three concentrations studied. Thus, CBZ as 386 of FLM at low and medium concentration, show a slight or constant accumulation during 387 the first three weeks and a decrease is observed in the concentration in the last week. 388 However, at high concentrations, the accumulation of FLM occurs more rapidly in the 389 first or second week and subsequently a decrease is observed. In contrast, CBZ exhibits 390 a completely opposite behavior to that of FLM at this concentration level. To the best of 391 our knowledge, the founded concentration in stem dependence on the concentration level 392 of irrigation have not been previously reported. 393 Additionally, as can be seen, concentration average values obtained remain in the same 394 order between the two plants used for each assay which means that plants suffers the same 395 accumulation when the uptake study is performed under quite similar conditions. 396 397 398 399 20 Table 5. Results obtained in Lavandula dentata after uptake assay. 400 Matrix level Cuts Found compounds concentrations Flumequinea P1 Flumequinea P2 Carbamazepinea P1 Carbamazepinea P2 Averageb SD Averageb SD Averageb SD Averageb SD Leaf 10 ng mL-1 7 days 111 23 220 14 141 22 145 18 14 days 58 0.2 70 8 69 10 132 8 21 days <MLOQ 20 4 36 8 83 12 30 days <MLOQ <MLOQ 24 3 21 10 700 ng mL-1 7 days 1554 68 1326 49 1148 68 1754 24 14 days 449 20 363 23 312 57 413 11 21 days 187 6 420 28 223 22 676 18 30 days 206 19 346 22 163 53 353 2 10 µg mL-1 7 days 1156 67 2076 43 2498 38 1390 17 14 days 631 68 468 82 1672 37 545 31 21 days 242 34 361 35 245 13 368 61 30 days 150 20 62 13 150 27 207 4 Stem 10 ng mL-1 7 days 172 14 209 37 109 25 84 39 14 days 151 3 184 15 101 5 156 25 21 days 646 19 79 20 129 8 48 18 30 days 335 25 57 9 54 15 49 4 700 ng mL-1 7 days 502 21 559 4 102 6 167 16 14 days 666 5 565 14 101 9 103 14 21 days 417 14 561 8 647 11 717 33 30 days 160 1 331 12 102 12 103 28 10 µg mL-1 7 days 764 17 804 47 464 9 633 7 14 days 1020 45 686 16 3837 37 1862 57 21 days 875 14 490 35 3074 20 2318 37 30 days 494 27 308 14 15635 272 4630 113 Root 10 ng mL-1 127 4 92 3 832 9 443 43 700 ng mL-1 195 7 241 9 1354 32 722 72 10 µg mL-1 1904 16 1924 16 15035 110 14076 41 aconcentration expressed in ng g-1 baverage concentration (n=3) 21 4. Conclusions 401 The developed analytical method allows the efficient extraction of the 6 selected 402 pharmaceuticals from different parts of the selected plants in five minutes by means of 403 microwave energy with no additional clean-up step. Regarding the extraction procedure, 404 several procedures, solvents and treatments were evaluated looking for the best results in 405 terms of recoveries and sensitivity, obtaining results for quantification limits in the order 406 of ng·g-1, with good accuracy and precision in the order of the recommended values for 407 pharmaceutical determinations when 5 mL of a mixture ACN:H2O (1:1 v/v) was used as 408 extracting solvent. 409 The method has been applied for the identification and quantification of the analytes in 410 samples of Lavandula dentata after irrigation the soil of the plant with a mixture of six 411 drugs. The uptake assay developed lasted 30 days. From this assay can be stated that FLM 412 and CBZ were the compounds more accumulated in leaf and stem of the plant, showing 413 similar trends at low concentration levels of exposure. Contrary to what it was initially 414 expected at high and medium level of exposure the plants showed a higher accumulation 415 at 21 days and then a high decrease at 30 days, behavior that could be related to the 416 damage of the plant. However additional studies should be done to assure this supposition. 417 As it was expected the concentrations of the compounds in root increased with the higher 418 concentration of the mixture used in the assay. 419 On the other hand, CPR, ENR, DCL and IBU were not identified in the samples as 420 unaltered form of the drug. 421 Acknowledgements 422 This work is supported by the Project CTM2015-67902-C-1-P from the “Dirección 423 General de Investigación y Gestión del Plan Nacional de I+D+i (Ministerio de Educación 424 y Ciencia, Spain)”. Authors also want to thank FEDER Project UNSE10-1E-429 and 425 Servicio de Microanálisis de la Universidad de Sevilla (CITIUS, Celestino Mutis) for the 426 support offered with UPLC-QTOF/MS instrument. 427 428 429 430 22 References 431 Al Nasir, F., Batarseh, M.I. 2008, Agricultural reuse of reclaimed water and uptake of 432 organic compounds: Pilot study at Mutah University water treatment plant, Jordan. 433 Chemosphere 72, 1203-1214. 434 Al-Qaim, F. F., Abdullah, M. P., Othman, M. R., Mussa, Z. H., Zakaria, Z., Latip, J., 435 Afiq, W. M. 2015. Investigation of the environmental transport of human pharmaceuticals 436 to surface water: A case study of persistence of pharmaceuticals in effluent of sewage 437 treatment plants and hospitals in Malaysia. J. Braz. Chem. Soc. 26(6), 1124-1135. 438 Al-Qaim, F. F., Abdullah, P., Othman, M. R., Latip, J., & Afiq, W. M. 2013. Development 439 of analytical method for detection of some pharmaceuticals in surface water. Trop. J. 440 Pharm. Res. 12(4), 609-616. 441 Arasimowicz M., Wisniowska-Kielian B., Niemiec M. 2013, Efficiency of antioxidative 442 system in spinach plants growing in soil contaminated with nickel. Ecol. Chem. Eng. A. 443 20(9), 987-997. 444 Avdeef A., Box K.J., Comer J.E.A., Hibbert C., Tam K.Y. 1998. pH-Metric logP 10. 445 Determination of Liposomal Membrane-Water Partition Coefficients of lonizable Drugs. 446 Pharm. Res. 15, 209-215. 447 Bao Y.P., Li Y.W., Mo C.H., Yao Y., Tai Y.P., Wu X.L., Zhang Y. 2010, Determination 448 of six sulfonamide antibiotics in vegetables by solid phase extraction and high 449 performance liquid chromatography. Environ. Chem. 29(3), 513–518. 450 Boxall, A.B.A., Johnson, P., Smith, E.J., Sinclair, C.J., Stutt, E., Levy, L. 2006. Uptake 451 of veterinary medicines from soils into plants. J. Agric. Food Chem. 54(6), 2288-2297. 452 Camacho-Muñoz, D., Martín, J., Santos, J. L., Aparicio, I., Alonso, E. 2012. Effectiveness 453 of conventional and lowcost wastewater treatments in the removal of pharmaceutically 454 active compounds. Water Air Soil Pollut., 223, 2611–2621. 455 Camacho-Muñoz, D., Martín, J., Santos, J. L., Aparicio, I., Alonso, E. 2013. Distribution 456 and risk assessment of pharmaceutical compounds in river sediments from Doñana Park 457 (Spain). Water Air Soil Pollut.. 224, 1665-1680. 458 Camacho-Muñoz, M. D., Santos, J. L., Aparicio, I., Alonso, E. 2010. Presence of 459 pharmaceutically active compounds in Doñana Park (Spain) main watersheds. J. Hazard. 460 Mater., 177, 1159–1162. 461 Carmona E., Andreu V., Picó Y. 2014. Occurrence of acidic pharmaceuticals and personal 462 care products in Turia River Basin: From waste to drinking water. Sci. total Environ. 484, 463 53-63. 464 Cesário R., Monteiro C.E., Noguira M., O’Driscoll J.N., Caetano M., Hintelmann H., 465 Mota A.M., Canário J. 2016. Mercury and methylmercury dynamics in sediments on a 466 protected area of Tagus estuary (Portugal). Water Air Soil Pollut. 227, 475-XX. 467 Dal Pozzo A., Donzelli G., Rodriquez L., Tajana L. 1989. “In vitro” model for the 468 evaluation of drug distribution and plasma protein-binding relationships. Inter. J. Pharm. 469 50, 97-101. 470 23 EC COUNCIL DIRECTIVE of 21 May 1991 concerning urban wastewater treatment 471 (91/271/EEC). Official Journal of the European Communities. 30/5/91. No L 135/41. 472 Emhofer L., Himmelsbach M., Buchberger W., Klampfl C.W. 2017, High-performance 473 liquid chromatography-mass spectrometry analysis of the parent drugs and their 474 metabolites in extracts from cress (Lepidium sativum) grown hydroponically in water 475 containing four non-steroidal anti-inflammatory drugs. J. Chrom. A 1491, 137-144. 476 Fernández-Torres R., Bello-López M.A., Olias-Consentino M., Callejon-Mochon M., 477 Ramos-Payan M. 2011. Enzymatic-microwave assisted extraction and high-performance 478 liquid chromatography–mass spectrometry for the determination of selected veterinary 479 antibiotics in fish and mussel samples. J of Pharm. Biomed. Anal. 54, 1146–1156. 480 Gentili, F.G.; Fick, J. 2016. Algal cultivation in urban wastewater: an efficient way to 481 reduce pharmaceutical pollutants. J. Appl. Phycol. 29(1), 255-262. 482 Gros, M., Petrovic, M., Ginebreda, A., & Barceló, D. 2010. Removal of pharmaceuticals 483 during wastewater treatment and environmental risk assessment using hazard indexes. 484 Environ. Intern., 36, 15–26; 485 Hijosa-Valesco M., Fink G., Schlüsener P., Sidrach-Cardona R., Martín-Villacorta J., 486 Temes T., Bécares E. 2011. Removal of antibiotics from urban wastewater by 487 constructed wetland optimization. Chemosphere. 83, 713-719. 488 Hrabovsky V., Dzendzelyuk O.,Katerynchuk I., Furgala Y. 2004. Monitoring of 489 radionuclides contamination of soils in Shatsk National Natural Park (Volyn Region, 490 Ukraine) during 1994-2001. J. Environ. Radioac. 72, 25-33. 491 Hu F., Bian K., Liu Y., Su Y., Zhou T., Song X., He L. 2014. Development of a modified 492 Quick, Easy, CHeap, Effective, Rugged and Safe method for the determination of multi- 493 class antimicrobials in vegetables by liquid chromatography tándem mass spectrometry. 494 J. Chrom. A. 1368, 52-63. 495 Hu X.G., Zhou Q.X., Luo Y. 2010, Occurrence and source analysis of typical veterinary 496 antibiotics in manure, soil, vegetables and groundwater from organic vegetable bases, 497 northern China. Environ. Pollut. 158(9), 2992–2998. 498 Kong, W.D., Zhu, Y.G., Liang Y.C., Zhang J., Smith F.A., Yang M. 2007, Uptake of 499 oxytetracycline and its phytotoxicity to alfalfa (medicago sativa L.). Environ. Pollut., 500 147(1), 187- 193. 501 Lillenberg M., Litvin S.V., Nei L., Roasto M., Sepp K. 2010. Enrofloxacin and 502 Ciprofloxacin Uptake by Plants from Soil. Agron. Res. 8, 807–814. 503 Malchi T., Maor Y., Tadmor G., Shenker M., Chefetz B. 2014, Irrigation of Root 504 Vegetables with Treated Wastewater: Evaluating Uptake of Pharmaceuticals and the 505 Associated Human Health Risks. Environ Sci. Technol. 48(16), 9325-9333. 506 Marsoni M, De Mattia F, Labra M, Bruno A, Bracale M, Vannini C. 2014. Uptake and 507 effects of a mixture of widely used therapeutic drugs in Eruca sativa L. and Zea mays L. 508 plants. Ecotox. Environ. Saf. 108, 52-57. 509 Martínez-Alcalá I., Guillén-Navarro J.M., Fernández-López C. 2017, Pharmaceutical 510 biological degradation, sorption and mass balance determination in a conventional 511 24 activated-sludge wastewater treatment plant from Murcia, Spain. Chem. Eng. J. 316, 332- 512 340. 513 Migliore L., Cozzolino S., Fiori M. 2003. Phytotoxicity and uptake of enrofloxacin in 514 crop plants. Chemosphere 52, 1233-1244. 515 Redshaw, C.H., Wootton, V.G., Rowland, S.J. 2008. Uptake of the pharmaceutical 516 Fluoxetine hydrochloride from growth medium by brassicaceae. Phytochem. 69(25), 517 2510-2516. 518 Riemenschneider C, Seiwert B, Goldstein M, Al-Raggad M, Salameh E, Chefetz B. 519 2017a. An LC-MS/MS method for the determination of 28 polar environmental 520 contaminants and metabolites in vegetables irrigated with treated municipal wastewater. 521 Anal. Meth. 9, 1273-1281. 522 Riemenschneider C, Seiwert B, Moeder M, Schwarz D, Reemtsma T. 2017b. Extensive 523 Transformation of the Pharmaceutical Carbamazepine Following Uptake into Intact 524 Tomato Plants. Environ. Sci. Tech. 51, 6100-6109. 525 Rivera-Jaimes J.A.,Postigo C., Melgoza-Alemán R.M., Aceña J., Barceló D., López de 526 Alda M. 2018. Study of pharmaceuticals in surface and wastewater from Cuernavaca, 527 Morelos, Mexico: Occurrence and environmental risk assessment. Sci. Total Environ. 528 613-614, 1623-1274. 529 Shernker, M., Harush, D., Ben-Ari, J., Chefetz, B. 2011. Uptake of carbamazepine by 530 cucumber plants- A case study related to irrigation with reclaimed wastewater. 531 Chemosphere 125, 963-988. 532 Tanoue R., Sato Y., Motoyama M., Nakagawa S., Shinohara R., Nomiyama K. 2012. 533 Plant Uptake of Pharmaceutical Chemicals Detected in Recycled Organic Manure and 534 Reclaimed Wastewater. J. Agric. Food Chem. 60, 10203−10211. 535 Tauxe-Wuersch A., De Alencastro L.F., Grandjean D., Tarredellas J. 2005. Occurrence 536 of several acidic drugs in sewage treatment plants in Switzerland and risk assessment. 537 Water Res. 39, 1761-1772. 538 Tiedekena E.J., Tahara A., McHughb B., Rowan N.J. 2017. Monitoring, sources, 539 receptors, and control measures for three European Union watch list substances of 540 emerging concern in receiving waters – A 20 year systematic review. Sci. Total Environ. 541 574, 1140–1163. 542 Vaclaik L., Krynitsky A.J., Rader J.I. 2014. Targeted analysis of multiple 543 pharmaceuticals, plant toxins and other secondary metabolites in herbal dietary 544 supplements by ultra-high performance liquid chromatography-quadrupole-orbital ion 545 trap mass spectrometry. Anal. Chim. Acta 810, 45-60. 546 Verlicchi, P., Al Aukidy, M., & Zambello, E. 2012. Occurrence of pharmaceutical 547 compounds in urban wastewater: removal, mass load and environmental risk after a 548 secondary treatment: a review. Sci. Total Environ. 429, 123–155. 549 Wu X., Conkle J.L., Ernst F., Gan J. 2014. Treated wastewater irrigation: Uptake of 550 pharmaceutical and personal care products by common vegetables under field conditions. 551 Environ Sci Technol. 48, 11286–93. 552 25 Wu X., Ernst F., Conkle J.L., Gan J. 2013. Comparative uptake and translocation of 553 pharmaceutical and personal care products (PPCPs) by common vegetables. Environ. Int. 554 60, 15-22. 555 Yao J.H., Niu D.K., Li Z.J., Liang Y.C., Zhang S.Q. 2010, Effects of antibiotics 556 oxytetracycline on soil enzyme activities and microbial microbial biomass in wheat 557 rhizosphere. Scientia Agricultura Sinica 43(4), 721–728. 558 Zhengxin X., Guanghua L., Zhenhua Y., Jianchao L., Peifang W., Yonghua W., 2017. 559 Bioaccumulation and trophic transfer of pharmaceuticals in food webs from a large 560 freshwater lake. Environ. Pollut. 222, 356-366. 561 562