Impacts of the human pharmaceutical diclofenac in the aquatic environment
Mehinto, Alvine Coralie
Thesis or dissertation
University of Exeter
Reason for embargo
The data from two chapters are being used for publications.
An increasing number of pharmaceuticals have been found in the aquatic environment and the issue has become a human and environmental health concern. Many pharmaceuticals are not fully degraded in wastewater treatment plants (WWTPs) and are continuously released in the aquatic environment resulting in concentrations in the low µg/l range in the receiving waters. Diclofenac is a widely used non-steroidal anti-inflammatory drug (NSAID) and is persistent in the aquatic environment. This pharmaceutical has been frequently reported in wastewater effluents, surface waters, groundwaters and even drinking water. NSAIDs are known to inhibit the cyclooxygenase activity, an enzyme present in many species of the animal kingdom responsible for the synthesis of prostanoids, and chronic exposure to environmental diclofenac may have detrimental effects on metabolism of non-target organisms including microbes and fish. In this thesis, microbiology, genomics and metabolomics approaches were used to investigate the effects of diclofenac on aquatic microbes and fish. In the first study of the thesis (chapter 3), the biodegradation of selected NSAIDs was investigated, together with their potential toxicity to aquatic microbes. Aerobic biodegradation experiments were conducted using activated sludge and wastewater effluents as microbial inocula and diclofenac, ketoprofen or naproxen as sole carbon source (1-10 mg/l) in order to isolate and identify the bacterial degraders. Changes in the bacterial populations were monitored by optical density and PCR-DGGE. The analytical techniques solid phase extraction (SPE) and ultraperformance liquid chromatography-mass spectrometry (UPLC-TOF-MS) were optimised to quantify the pharmaceuticals in environmental samples. High recovery rates were obtained with 94% for diclofenac; 92% for ketoprofen and 85% for naproxen and with detection capabilities down to 3-7 ng/l. Results from the biodegradation experiments showed that ketoprofen and naproxen were eliminated at up to 99 and 55% respectively over a 40 days period. Consistently with previous studies, diclofenac showed no significant degradation. In all the enrichments, a significant decrease in the bacterial abundance was observed as a consequence of NSAIDs exposure and attempts to isolate the bacterial degrading populations were unsuccessful. Given the apparent micro-toxicity of these NSAIDs, the standardised test Microtox@ was carried out with Vibrio fischeri. The EC50 (15 min) estimated ranged from 13.5 mg/l + 2.3 for diclofenac to 42.1 mg/l + 3.9 for naproxen. Further toxicological tests were performed with diclofenac on bacterial strains isolated from activated sludge. Growth inhibitory effects were observed from 50-70 mg/l for Micrococcus luteus, Zoogloea ramigera and Comamonas denitrificans. Pseudomonas putida seemed more tolerant to diclofenac exposure and toxic effects were observed from 90 mg/l. These studies showed that diclofenac was the most toxic NSAID but toxicological effects in bacteria only occurred at concentrations at least 1,000 times higher than those found in the environment. However, chronic exposure to lower concentrations may cause similar interferences and affect the degradation potential of naturally occurring microbial populations. The second study (chapter 4) investigated the biological effects of sub-chronic exposure to waterborne diclofenac (0.5, 1, 5 and 25 µg/l) in female juvenile rainbow trout Oncorhynchus mykiss. After 21-day exposure, mRNA expression levels of cytochrome p450 1a1 (cyp1a1), cyclooxygenase (cox) 1 and 2, and p53 were investigated in the liver, kidney and gills using RT-PCR and QPCR. These genes were selected as they are likely targets for diclofenac in mammals. Histopathological investigations were carried out in the small intestine, liver and kidney because diclofenac has been reported to induce toxicity responses in these tissues. Fish bile was also analysed by SPE and UPLC-TOF-MS to evaluate the bioconcentration potential of diclofenac and look for evidences of diclofenac metabolism. Results showed a significant reduction of both cox1 and cox2 expression in the liver, gills and kidney from 1 μg diclofenac/l. In contrast diclofenac induced an increase in mRNA levels for cyp1a1 in the liver and gills but a significant reduction of cyp1a1 expression in the kidney from 1 µg/l. There were no clear effects of diclofenac on the mRNA levels of p53. Diclofenac exposure caused tissue damages at exposure concentrations as low as 1 µg/l. Histopathological injuries included inflammation, hyperplasia and fusion of the villi in the small intestine and tubule necrosis in the kidney. There were no obvious changes in the liver of diclofenac-exposed fish. The analysis of bile revealed a bioconcentration potential between 509 + 27 and 657 + 25. A reactive metabolite of diclofenac was also detected at the highest exposure concentration which may be responsible for the severe injuries found in those fish. Sub-chronic exposure to environmental concentrations of diclofenac altered gene expression and it is possible that long term exposure to environmental diclofenac lead to significant impacts on fish health. In the final part of this thesis (chapters 5 and 6) effects on the metabolite composition of biofluids were analysed in diclofenac-exposed fish. This work entailed developing and validating appropriate methodologies to analyse fish bile and blood plasma. Methanol extraction and UPLC-TOF-MS were optimised to analyse the plasma metabolome but the methodologies were not suitable to detect low abundance molecules such as eicosanoids due to the interferences (ion suppression) in the samples matrix. Multivariate data analysis failed to detect the endogenous metabolites of the plasma affected by the chemical exposure. The only discriminating metabolite was found after analysis of the plasma samples from control vs. 25 µg/l treatment groups and identified as the exogenous compound diclofenac. To analyse the bile, the developed SPE methodology was carried out in order to separate the metabolites between a free steroids (fatty acids, eicosanoids, etc.) fraction and a conjugated steroids (bile salts) fraction. Due to high levels of taurocholic acid masking other metabolites in the conjugated fraction, some bile samples were hydrolysed to deconjugate these metabolites. The non-hydrolysed and hydrolysed bile fractions were analysed by UPLC-TOF-MS in positive and negative ionization. Multivariate data analysis using principal component analysis (PCA) and partial least square discriminant analysis (PLS-DA) revealed significant perturbations in the bile metabolite profile of diclofenac-exposed rainbow from the lowest exposure concentration (0.5 µg/l). Over 50 metabolites were elevated or reduced as a result of the 21-day exposure, suggesting that diclofenac affected several metabolic pathways. One metabolite was identified as a lipooxygenase product. This suggests that the inhibition of prostanoids synthesis can cause a shift in the arachidonic cascade and increase the synthesis of other eicosanoids. Most of the other discriminative metabolites remain unidentified and FT-MS analysis will be performed to obtain a structural identity. The metabolomics study further highlights the concern of environmental diclofenac in non-target organisms and the need to investigate the metabolic pathways affected.
Tyler, Charles R.
PhD in Biological Sciences