Wastewater-based epidemiology methodology to investigate human exposure to bisphenol A, bisphenol F and bisphenol S.
Abstract
Wastewater-based epidemiology (WBE) has become an invaluable tool for tracking the evolution of use or exposure of/to numerous substances. Bisphenols, commonly utilized in manufacturing plastic goods, have been categorized as endocrine disrupting chemicals, underscoring the critical need for real-time data on their local-level exposure to safeguard public health. In this study, we have developed a novel analytical method and WBE framework for the assessment of population-level exposure to bisphenol A (BPA) and its most prominent substitutes, bisphenols F and S (BPF and BPS), through the determination their Phase II metabolites in wastewater by WBE. Stability and exclusivity tests denoted that glucuronides are not stable in sewage, whereas sulfate metabolites are good biomarkers. Therefore, a solid-phase extraction followed by liquid chromatography-tandem mass spectrometry method was developed for the bisphenols’ monosulfates and BPA bissulfate. The analytical method was validated with three different wastewater matrices, providing trueness (as recovery) in the 79–112 % range with relative standard deviations < 12 %, and method quantification limits below 2 ng L−1 for monosulfates, but higher (35 ng L−1) for BPA bissulfate. Subsequently, the method was applied to 24h-composite raw wastewater samples collected over a week in 4 different locations in Spain and Portugal. BPA bissulfate was not detected, but the three monosulfate metabolites of each bisphenol were positively detected in the samples, being the metabolite of BPA the most prevalent, followed by those of BPF and BPS. Community-wide BPA intake was then estimated to be higher than the European Food Safety Agency (EFSA) tolerable daily intake (TDI) of 2 × 10−4 µg kg−1day−1 in all locations. In the case of BPF and BPS, there is not enough metabolism data or even established limit, but they would also surpass safe levels in several locations if a similar metabolism and TDI would be assumed. This innovative method could be used to a larger set of wastewater-treatment plants as an early-warning approach on human exposure to bisphenols.
Full text
- 1 - SUPPORTING INFORMATION TO: Development And Pilot Application Of A Wastewater-Based Epidemiology Methodology To Investigate Human Exposure To Bisphenol A, Bisphenol F and Bisphenol S. Andrea Estévez-Danta a, Rosa Montesa, Ailette Prietob, Miguel M. Santosc, Gorka Orived, Unax Lertxundie, José Benito Quintanaa,*, Rosario Rodil a a Aquatic One Health Research Center (iARCUS). Department of Analytical Chemistry, Nutrition and Food Chemistry. R. Constantino Candeira S/N, IIAA building, Universidade de Santiago de Compostela, 15782 Santiago de Compostela, Spain b Department of Analytical Chemistry, Faculty of Science and Technology, University of the Basque Country (UPV/EHU), Leioa, Basque Country, Spain c CIMAR/CIIMAR, Interdisciplinary Centre of Marine and Environmental Research, Group of Endocrine Disrupter and Emerging Contaminants, University of Porto, Avenida General Norton de Matos s/n, 4450208 Matosinhos, Portugal d NanoBioCel Group, Laboratory of Pharmaceutics, School of Pharmacy, University of the Basque Country UPV/EHU, Paseo de la Universidad 7, Vitoria-Gasteiz 01006, Spain e Bioaraba Health Research Institute, Osakidetza Basque Health Service, Araba Mental Health Network, Araba Psychiatric Hospital, Pharmacy Service, c/Alava 43, 01006 Vitoria-Gasteiz, Alava, Spain
- 2 - INDEX Page Table S1. Chemical structures of target bisphenols and the corresponding human metabolites considered 3 Table S2. Details on WWTPs sampled. Sampling Mode: time proportional, followed by subsampling interval and volume taken per subsample. Dates presented as DD.MM.YYYY. 4 Table S3. Chromatographic conditions used with the EC 150/2 Nucleodur HILIC and Luna Omega Polar C18 columns 5 Table S4. Chemical formulae, retention time (RT), transitions (Q n ) used for quantification (Q1) and confirmation (Q2 and Q3), ratio between the transitions, optimal collision energy (CE) values and compounds used as internal standards (IS) for bisphenols and their glucuronides. 6 Table S5. Instrumental quantification and detection limits (IQL and IDL) obtained with the EC 150/2 Nucleodur HILIC and the Omega Polar C18 columns. 7 Figure S1. Chromatogram of a 500 ngmL-1 standard obtained on the EC 150/2 Nucleodur HILIC column. 8 Figure S2. Chromatograms showing the 3 SRM transitions of a wastewater extract spiked with 500 ng mL-1 of BPA-S obtained with the EC 150/2 Nucleodur HILIC column. 9 Figure S3. Absolute recoveries obtained with ultrapure water with the Oasis HLB and Oasis WAX protocols 10 Table S6. Concentration of bisphenols’ sulfates in the wastewater samples. BPA-DS was below the MDL in all samples. 11 Table S7. Population normalized daily loads (PNDLs) calculated from the concentrations measured in wastewater (Table S6) after considering WWTP flow data and population served (Table S2). 12 Table S8. Urinary concentration of bisphenol metabolites estimated from PNDLs (Table S7). 13 Table S9. Compilation of literature studies where the excretion factors of sulfate metabolites were investigated. 14 Table S10. Human exposure to bisphenols calculated from the PNDL (Table S7) after considering the correction factor (CF) and human intake for bisphenols calculated from human exposure ,after considering the average European population weight (70.8 kg). 16
- 3 - Table S1. Chemical structures of target bisphenols and the corresponding human metabolites considered. Z OHHO Z OR 2 R 1 O Parent plasticizer Z Metabolite R1 and R2 Bisphenol A (BPA) -C(CH3)2 Bisphenol A Monosulfate (BPA-S) R1 -H R 2 -SOH Bisphenol A Bissulfate (BPA-DS) R1 = R2 -SO3H Bisphenol A β-D-Glucuronide (BPA-G) R1 -H R2 O HO CO 2 H OH OH Bisphenol A Bis-(β-D-Glucuronide) (BPA-G) R1 = R2 O HO CO 2 H OH OH Bisphenol F (BPF) -CH2 Bisphenol F Monosulfate (BPF-S) R1 -H R 2 -SO3H Bisphenol F β-D-Glucuronide (BPF-G) R1 -H R2 O HO CO 2 H OH OH Bisphenol S (BPS) -SO2 Bisphenol S Monosulfate (BPS-S) R1 -H R 2 -SO3H Bisphenol S β-D-Glucuronide (BPS-G) R1 -H R2 O HO CO 2 H OH OH Parent compound Metabolite
- 4 - Table S2. Details on WWTPs sampled. Sampling Mode: time proportional, followed by subsampling interval and volume taken per subsample. Dates presented as DD.MM.YYYY. Code Location 1 Location 2 Location 3 Location 4 Country Spain Spain Spain Portugal Population served 860,237 136,500 255,052 150,000 Sampling date 1 20.10.2021 19.10.2021 19.10.2021 21.04.2021 Sampling date 2 21.10.2021 20.10.2021 20.10.2021 22.04.2021 Sampling date 3 22.10.2021 21.10.2021 21.10.2021 23.04.2021 Sampling date 4 23.10.2021 22.10.2021 22.10.2021 24.04.2021 Sampling date 5 24.10.2021 23.10.2021 23.10.2021 25.04.2021 Sampling date 6 25.10.2021 24.10.2021 24.10.2021 26.04.2021 Sampling date 7 26.10.2021 25.10.2021 25.10.2021 27.04.2021 Flow (m3day-1) day 1 205916 39866 70100 31670 Flow (m3day-1) day 2 203351 42948 70400 36500 Flow (m3day-1) day 3 231651 39397 67300 36170 Flow (m3day-1) day 4 191835 36981 68000 35110 Flow (m3day-1) day 5 191507 40300 66000 37780 Flow (m3day-1) day 6 194848 41486 61200 35370 Flow (m3day-1) day 7 200330 42081 66200 35460 Sampling mode 60 min, 100 mL 10 min, 150 mL 10 min, 150 mL 15 min, 125 mL
- 5 - Table S3. Chromatographic conditions used with the EC 150/2 Nucleodur HILIC and Luna Omega Polar C18 columns. Mobile Phase A 5 mM of ammonium formiate in ultrapure water: acetonitrile 95: 5 at pH 4 5 mM of ammonium acetate in ultrapure water Mobile Phase B 5 mM of ammonium formiate in ultrapure water: acetonitrile 5: 95 at pH 4 5 mM of ammonium acetate in MeOH Injection Volume 2 µL 2 µL Flow rate 0.3 mL min-1 0.4 mL min-1 Chromatographic gradient Target analytes BPA-S, BPA-DS, BPA-G, BPA-DG BPS-S, BPS-G BPF-S, BPF-G BPA, BPS and BPF 0 20 40 60 80 100 0 5 10 15 20 % B Time / min 0 20 40 60 80 100 0 5 10 15 % B Time / min
- 6 - Table S4. Chemical formulae, retention time (RT), transitions (Qn) used for quantification (Q1) and confirmation (Q2 and Q3), ratio between the transitions, optimal collision energy (CE) values and compounds used as internal standards (IS) for bisphenols and their glucuronides. Compound Chemical formulae RT (min) Q1 (m/z) CE (eV) Q2 (m/z) CE (eV) Q3 (m/z) CE (eV) IS BPA C 15 H 16 O 2 5.05 a 227 > 211 37 227 > 133 25 227 > 93 57 BPA-d6 BPF C 13 H 12 O 2 4.27 a 199 > 93 21 199 > 105 21 199 > 77 29 BPF-d10 BPS C 12 H 10 O 4 S 3.13 a 249 > 108 29 249 > 156 21 249 > 92 37 BPS-d8 BPA-d6 C 15 H 10 2H 6 O 2 5.05 a 233 > 138 25 - - - - - BPS-d8 C 12 H 2 2H 8 O 4 S 3.13 a 257 > 112 29 - - - - - BPF-d10 C 13 H 2 2H 10 O 2 4.27 a 209 > 97 21 - - - - - BPA-G C 21 H 24 O 8 7.35 b 403 > 227 29 403 > 175 13 403 > 113 17 13C 12 BPA-G BPA-DG C 27 H 32 O 14 10.1 b 579 > 403 21 579 > 113 33 579 > 227 49 13C 12 BPA-G BPS-G C 18 H 18 O 10 S 7.90 b 425 > 249 25 425 > 156 49 425 > 113 17 13C 12 BPA-G BPF-G C 19 H 20 O 8 7.65 b 375 > 199 33 375 > 175 9 375 > 113 13 13C 12 BPA-G 13C 12 BPA-G C 9 13C 12 H 24 O 8 7.35 b 415 > 113 17 - - - - - a: referred to Omega Polar C18 column. b: referred to EC 150/2 Nucleodur HILIC column .
- 7 - Table S5. Instrumental quantification and detection limits (IQL and IDL) obtained with the EC 150/2 Nucleodur HILIC and the Omega Polar C18 columns. Compounds IDL (ng mL-1) IQL (ng mL-1) EC 150/2 Nucleodur HILIC BPA-S 0.41 1.39 BPA-DS 0.53 1.77 BPA-G 0.32 1.08 BPA-DG 0.69 2.31 BPS-S 0.06 0.20 BPS-G 0.18 0.61 BPF-S 0.05 0.17 BPF-G 0.23 0.76 Omega Polar C18 BPA 12 40 BPS 0.13 0.43 BPF 7.4 24.8
- 8 - Figure S1. Chromatogram of a 500 ng mL-1 standard obtained on the EC 150/2 Nucleodur HILIC column. 2 x10 2 BPA-DG 10.137 4 x10 0 5 BPA-DS 1.568 3 x10 0 2 BPA-G 7.345 5 x10 0 BPA-S 1.182 3 x10 0 1 BPF-G 7.645 5 x10 0 5 BPF-S 1.236 3 x10 0 1 BPS-G 7.901 5 x10 0 5 BPS-S 1.127 Time (min) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 403 > 227 387 > 307 579 > 403 307 > 227 375 > 199 279 > 199 425 > 249 329 > 249 Signal Intensity
- 9 - Figure S2. Chromatograms showing the 3 SRM transitions of a wastewater extract spiked with 500 ng mL-1 of BPA-S obtained with the EC 150/2 Nucleodur HILIC column. 3 x10 0 0.2 0.4 0.6 0.8 1 1.2 * 1.220 3 x10 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5* 1.212 4 x10 0 0.5 1 1.5 2 2.5 3 * 1.205 Time (min) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Signal Intensity BPA-S 307 > 133 BPA-S 307 > 212 BPA-S 307 > 227
- 16 - Table S10. Human exposure to bisphenols calculated from the PNDL (Table S7) after considering the correction factor (CF) and human intake for bisphenols calculated from human exposure ,after considering the average European population weight (70.8 kg). Location Date (DD.MM.YYYY) Human exposure (µg day-1 inhabitant-1) Human intake (µg kg-1 day-1) BPA from BPA-S BPS from BPS-S BPF from BPF-S BPA from BPA-S BPS from BPS-S BPF from BPF-S Location 1 20.10.2021 113 92 140 1.6 2.0 1.3 21.10.2021 160 100 140 2.3 2.0 1.4 22.10.2021 565 44 387 8.0 5.5 0.62 23.10.2021 288 41 235 4.1 3.3 0.59 24.10.2021 86 42 56 1.2 0.79 0.60 25.10.2021 1213 43 333 17 4.7 0.60 26.10.2021 430 45 153 6.1 2.2 0.64 Average ± SD 408 ± 396 58 ± 26 206 ± 118 5.8 ± 5.6 2.9 ± 1.7 0.82 ± 0.37 Location 2 19.10.2021 4.8 11 <MDL 0.068 0.16 <MDL 20.10.2021 13 8.2 <MDL 0.18 0.12 <MDL 21.10.2021 19 <MDL <MDL 0.27 <MDL <MDL 22.10.2021 16 <MDL <MDL 0.23 <MDL <MDL 23.10.2021 13 6.6 <MDL 0.18 0.093 <MDL 24.10.2021 <MQL <MDL <MDL <MQL <MDL <MDL 25.10.2021 18 23 <MDL 0.26 0.32 <MDL Average ± SD 12 ± 6 a 7.3 ± 8.1 a <MDL 0.17 ± 0.09 a 0.10 ± 0.11 a <MDL Location 3 19.10.2021 455 <MDL 37 6.3 <MDL 0.52 20.10.2021 283 <MDL 22 4.0 <MDL 0.30 21.10.2021 307 <MDL 14 4.3 <MDL 0.20 22.10.2021 486 <MDL 53 6.9 <MDL 0.76 23.10.2021 733 <MDL 54 10 <MDL 0.76 24.10.2021 147 <MDL 45 2.1 <MDL 0.64 25.10.2021 334 <MDL 22 4.7 <MDL 0.32 Average ± SD 391± 187 <MDL 35 ± 16 5.5 ± 2.6 <MDL 0.50 ± 0.23 Location 4 21.04.2021 37 12 <MDL 0.52 0.17 <MDL 22.04.2021 22 8.5 <MDL 0.31 0.12 <MDL 23.04.2021 75 <MDL <MDL 1.1 <MDL <MDL 24.04.2021 17 12 <MDL 0.24 0.17 <MDL 25.04.2021 24 <MDL <MDL 0.34 <MDL <MDL 26.04.2021 29 7.3 <MDL 0.40 0.11 <MDL 27.04.2021 24 <MDL <MDL 0.34 <MDL <MDL Average ± SD 33 ± 20 6.0 ± 5.4 a <MDL 0.46 ± 0.28 0.084 ± 0.076 a <MDL a Average and standard deviation calculated by substituting the values <MQL and <MDL by MQL/2 and MDL/2, respectively. MDL/MQL: method detection/quantification limit.