Identification and fate of known and unknown transformation products of pharmaceuticals in the aquatic system

  1. Bozo Zonja
Dirixida por:
  1. Sandra Pérez Solsona Director
  2. Mira Petrovic Director
  3. María Teresa Galcerán Huguet Titora

Universidade de defensa: Universitat de Barcelona

Ano de defensa: 2017

  1. Juan Manuel Lema Rodicio Presidente
  2. Encarnación Moyano Morcillo Secretario/a
  3. Serge Chiron Vogal

Tipo: Tese

Teseo: 446759 DIALNET lock_openTDX editor


Pharmaceuticals which are used worldwide are designed to facilitate the life for the human society and have an important role in treatment and prevention of disease for both humans and animals. They are ubiquitous in the aquatic environment and are mainly derived from municipal wastewater treatment plants (WWTPs) due to their low removal rate. Therefore, their presence in the environment is directly linked to the human impact. Various biological and abiotic processes in the environment can transform them to transformation products (TPs). In many cases, transformation is already initiated in the human body by a variety of drug-metabolizing enzymes. The metabolites formed through human metabolism present some modifications in their chemical structures that can differ in physicochemical properties to their parent compound. Once they are excreted from the human body, both the unmetabolised parent drug and their metabolites enter WWTPs by means of the sewer system. Since the WWTPs are not designed to remove completely pharmaceutical residues, the fraction not removed after the treatment will eventually end up in the receiving water bodies. Consequently, due to pharmaceutical transformations in the human body, biotransformations in WWTPs and phototransformations in surface water, they can potentially produce a high number of TPs in real world samples which makes their identification a challenge. In this thesis, two different approaches (TPs profiling and suspect screening) based on high resolution mass spectrometry (HRMS) for the detection and identification of TPs of pharmaceuticals were investigated. TPs profiling approach was applied for the identification of phototransformation products of an antiviral zanamivir (ZAN) in batch reactors filled with surface water. On the other hand, suspect screening approach was applied for evaluation of transformation, prioritization and identification of photoTPs of six iodinated contrast media in surface water. Finally, a combination of suspect and TPs profiling approach was applied for the detection of TPs of an anticonvulsant lamotrigine and its main human metabolite lamotrigine N2-glucuronide which were formed as the result of their degradation in both activated sludge and pH dependent hydrolysis. The TPs profiling approach for evaluation of these transformations is illustrated in the example of photodegradation of an antiviral ZAN with identification of its TPs in surface water (Chapter 3.). Here a set of lab-scale experiments was performed in order to determine the susceptibility of ZAN towards photodegradation under simulated and natural sunlight. The identification of the TPs was performed using hydrophilic interaction liquid chromatography coupled to high resolution mass spectrometry (HILIC-HRMS) where four photoTPs were tentatively identified and their proposed structures were rationalized by photolysis mechanisms. Kinetic experiments showed that photodegradation kinetics of ZAN in surface waters would proceed with slow kinetics since upon exposure of aqueous solutions of surface water (20 μg L-1) to simulated sunlight, ZAN was degraded with t1/2 of 3.6 h. Under natural sunlight irradiating surface water, about 30 % of the initial concentration of the antiviral disappeared within 18 days. However, when ZAN and its TPs were retrospectively screened from surface water extracts, neither the parent nor the TPs were detected. The results of this TP profiling used for the identification of TPs of ZAN, although straightforward, suggests that it is not suitable when dealing with a considerably elevated number of TPs formed in batch experiments. However, time and effort needed to be optimised for the structure elucidation of 108 photoTPs of six iodinated contrast media (ICM) (Chapter 3.). Again, the photodegradation study was performed in surface water spiked with the ICMs using a sunlight lab-scale simulator. 108 TPs were generated and each photoTP was characterised by its unique exact mass of the molecular ion and retention time and added to a suspect list. Once the suspect list was generated, the photoTPs were searched in thirteen surface water samples which were extracted using a generic solid-phase extraction method (four cartridges of different chemistries in order to retain ample number of compounds with different chemical properties). Based on their detection frequency (those TPs with the frequency higher than 50 % were deemed important), eleven TPs were prioritized and their structures elucidated by HRMS and NMR (when possible). Out of the eleven prioritised TPs, ten were formed as the result of deiodination (either by deiodination, oxidative deiodination or intramolecular elimination). In the real surface water samples, median concentration of parent compounds was 110 ngL-1 reaching up to 6 µgL-1 for iomeprol while TPs were found at median concentration of 8 ngL-1, reaching up to 0.4 µgL-1 for iomeprol TP651-B. Here detection-based prioritization served as a crucial step to reduce the number of TPs to be identified and thereby reducing costs and time for the subsequent target analysis. This time-effective approach not only guaranties that the degradation products elucidated would be found, but also that they are environmentally relevant. In summary, the proposed screening approach facilitates the evaluation of the degradation of polar compounds at a real scale with a fast detection of TPs without prior availability of the standards. Approach used for detection and identification of TPs of ICM in Chapter 3 was an example of suspect screening where the suspect list of TPs was generated at lab-scale, In Chapter 4, the work started with a suspect screening of lamotrigine (LMG) and related compounds (its human metabolites, synthetic impurities and photoTPs) which were listed from the literature and searched in wastewater and surface water samples. As the result of suspect screening, LMG, three human metabolites and a LMG synthetic impurity (OXO-LMG) were detected in the screened samples. Preliminary results showed significantly higher concentrations of OXO-LMG in wastewater effluent, suggesting its formation in the WWTPs. However, biodegradation reactors amended with mixed liquor at neutral pH showed that LMG is resistant to biodegradation with only about 5 % elimination after 6 days. Since LMG is extensively and predominantly metabolised by phase II metabolism to its N2-glucuronide, this metabolite (LMG-N2-G) was degraded following the same experimental setup. Results showed that this metabolite was the actual source of the TP detected. Additionally, in batch experiments, LMG-N2-G was transformed, following pseudo-first kinetics, to three TPs as a result of i) deconjugation (to LMG), ii) oxidation of the glucuronic acid (to LMG-N2-G-TP430) and iii) amidine hydrolysis in combination with deconjugation (to OXO-LMG). In order to further rationalize the formation of the TP OXO-LMG, the stability of LMG-N2-G and related compounds was studied as a function of pH in the range of 4 – 9. Same as during biodegradation, LMG was stable across the entire pH range tested. However, LMG-N2-G was transformed to three TPs at neutral – basic pH. They were identified as TPs formed after hydrolysis of amidine and guanidine moieties. The third TP detected was an intermediated in the guanidine hydrolysis reaction. Kinetic experiments in wastewater samples at different concentration (20 and 200 nM) and pH (pH 6.5, 7, 8, 8.5 and 9) demonstrated that while the degradation constants were concentration independent, at higher pH, LMG-N2-G degraded at higher rate. The pH-dependent stability experiments of related compounds with different nitrogen N2-substituents on the 1,2,4-triazine ring showed that reaction of amidine and guanidine hydrolysis depends on imine tautomer equilibrium whose formation depends directly on the N2-supsitutent. LMG-N2-G major abiotic TP (amidine hydrolysis TP) was detected in hospital effluent and WWTP influent samples. Having in mind the concentrations of both biotic and abiotic TPs detected, a total mass balance at two-concentration levels batch reactors was closed at 86% and 102%, respectively. In three WWTPs total mass balance of LMG-N2-G ranged from 71-102%. Finally, LMG-N2-G and its TPs were detected in surface water samples with median concentration ranges of 23–186 ngL-1. The work presented in this chapter gives a new insight into the behaviour of glucuronides of pharmaceuticals, suggesting that they might also be sources of yet undiscovered, but environmentally relevant TPs.