Warning This Web page has been archived on the Web.

Archived Content

Information identified as archived on the Web is for reference, research or recordkeeping purposes. It has not been altered or updated after the date of archiving. Web pages that are archived on the Web are not subject to the Government of Canada Web Standards. As per the Communications Policy of the Government of Canada, you can request alternate formats on the Contact Us page.

Skip booklet index and go to page content

Pharmaceuticals and Personal Care Products in the Canadian Environment: Research and Policy Directions

3.2 Effects of PPCPs on Aquatic Ecosystems

The primary driver for the issue of PPCPs in the Canadian environment is the concern that long-term exposure to low levels of PPCPs residues could have adverse effects on aquatic and terrestrial ecosystems and/or human health. Effects research spans a wide range of activities, from observations at the molecular level about up- and down-regulation of genes to observations on growth and reproduction of a fish species as a result of exposure to a compound under controlled conditions. The presentations about the current state of PPCP effects research provided background information to guide discussions on future needs and priorities in this area.

Dr. Karl Fent, a professor at the University of Applied Sciences in Muttenz, Switzerland, provided the keynote address on the subject of ”Pharmaceuticals and ultraviolet absorbing compounds (UV filters) effects in the aquatic environment”

The chronic toxicity and potential subtle effects of pharmaceuticals are only marginally known. In my presentation, I critically review the current knowledge about human pharmaceuticals in the environment and address several key questions (Fent et al. 2006a). What kind of pharmaceuticals and what concentrations occur in the aquatic environment? What are the modes of action of these compounds in humans and are there similar targets in lower animals? What acute and chronic ecotoxicological effects are there elicited by pharmaceuticals and by mixtures of pharmaceuticals? What are the effect concentrations and how do they relate to environmental levels? I show that only very little is known about long-term effects of pharmaceuticals to aquatic organisms, in particular with respect to biological targets. For most human medicines analyzed, acute effects to aquatic organisms are unlikely to occur, except for spills. For investigated pharmaceuticals chronic lowest observed effect concentrations (LOEC) in standard laboratory organisms are about two orders of magnitude higher than maximal concentrations in WWTP effluents. However, for diclofenac, the LOEC for fish toxicity was in the range of wastewater concentrations, whereas the LOEC of propranolol and fluoxetine for zooplankton and benthic organisms were near to maximal measured WWTP effluent concentrations. Targeted ecotoxicological studies are lacking almost entirely and such investigations are needed for better and comprehensive risk assessments of pharmaceuticals. Using recombinant yeast systems, we show that some pharmaceuticals show estrogenic activity in vitro (Fent et al. 2006b). We evaluate the cytotoxicity in vitro of 34 pharmaceuticals from different classes and modes of action. Cytotoxicity was found for 21 pharmaceuticals with EC50-values ranging from 2.1 µM (doxorubicin) to 8.66 mM (salicylic acid). A good correlation with log Dow and between the in vitro data and in vivo data was found for Daphnia. We show that in vitro assays using cell lines are well suited for the first screening of the acute in vivo toxicity of pharmaceuticals (Caminada et al. 2006). In addition, we show that lipid lowering agents do not reduce Daphnia reproduction at reasonable concentrations. Currently, we analyze biochemical effect parameters in fish cell systems and in fish besides classical toxicity endpoints.

UV filters have been detected in surface water, wastewater and fish, and some of them are estrogenic in fish. At present, little is known about their additional hormonal activities in different hormonal receptor systems. We systematically analyzed the estrogenic, antiestrogenic, androgenic, and antiandrogenic activity of 18 UV filters and one metabolite in vitro at non-cytotoxic concentrations with recombinant yeast systems carrying either a human estrogen (hER or androgen receptor (hAR). All 19 compounds elicited hormonal activities, surprisingly most of them multiple activities. We found 10 UV-filters having agonistic effects towards the hER. Surprisingly, 6 UV filters showed androgenic activities and many of them having pronounced antiestrogenic and antiandrogenic activities. (Kunz and Fent, 2006). Mixtures of UV filters show in most cases synergistic interactions (Kunz and Fent 2006b). The UV filter 3-benzylidene camphor (3BC) has been shown in vitro and in vivo in fish to be estrogenic (Kunz et al. 2006a). After a pre-exposure period of 21 days, reproductively mature fathead minnows were exposed to increasing concentrations of 3BC for 21 days in a staticrenewal procedure. 3BC affected reproduction in a dose-dependent manner with weak effects on fecundity at 3 µg/L, a significant decrease at 74 µg/L and a cessation of reproduction at 285 µg/L. 3BC was accumulated in fish with an average bioconcentration factor of 313. Dose-dependent demasculinisation in secondary sex characteristics of male fish and dose-dependent induction of plasma vitellogenin occurred, which was significant at 74 µg/L. 3BC had a profound and dosedependent effect on the histology of gonads of males and female fish at 3 µg/L and higher. At 74 and 285 µg/L oocyte and spermatocyte development was inhibited in male and female gonads. Our studies show estrogenic activity of UV filters and significant effects of the common UV filter 3BC on fertility, gonadal development and reproduction of fish after short-term exposure that may have negative consequences on the population level (Kunz et al. 2006b).

There have been some measurements of UV filter chemicals, such as benzophenone, in treated wastewater in Switzerland and California. A study on municipal sewage sludge showed concentrations of some UV filters occurring in ranges of 1 to 6 mg/kg of dry matter. Concentrations have also been found in fish, including some bioconcentration factors in the range of 300 - 500 and other compounds concentrated higher.

Dr. François Gagné of the Aquatic Ecosystem Protection Research Division of Environment Canada provided a review of current and new strategies to assess the toxicological effects of PPCPs on aquatic species.

Pharmaceutical, personal care and veterinary products, which have been found in wastewater and surface water, are likely to contaminate the aquatic environment, including groundwater. Aquatic sentinel species that bioaccumulate some of these drugs remain to be identified, but studies with mussels and plants have shown that some antibiotics significantly accumulate in tissues. Laboratory tests have been conducted with some success on several aquatic species, including bacteria, plants, invertebrates (molluscs and arthropods) and fish, with commonly found drugs both individually and in mixtures. These toxicity tests generally indicate that acute lethal effectsMale fish showing female characteristics | Photo: Joanne Parrott are not likely to occur in the environment but that chronic or long-term effects are possible. In an attempt to measure the effects of pharmaceuticals and personal care products, two types of biomarkers are proposed. The first class, known as integrative biomarkers, consists of measuring ecologically relevant biomarkers that encompass the effects of drugs, such as oxidative stress or DNA damage. Biomarkers that have been shown to predict changes at both the individual and population levels, and that respond to these products, are particularly useful for taking into account the final effects of pollution on feral aquatic organisms. The second class of biomarkers, known as drug target-specific biomarkers, measures the state/integrity of drug targets likely to impede the organism’s health and reproduction. For example, prostaglandin synthase produces prostaglandins necessary to assist spawning in bivalves, and its activity could be blocked by nonsteroidal anti-inflammatory drugs such as acetylsalicylate and ibuprofen.

Finally, two case studies are presented to exemplify the use of biomarkers to assess drug target specific interactions and tissue damage in aquatic species. In the first case study, primary cultures of rainbow trout hepatocytes were used to assess the cytotoxicity of carbamazepine, a drug commonly found in municipal wastewater, at µg/L range, after exposure for 48 h at 18oC. Results showed that carbamazepine induced the activity of cytochromes P4503A4 and 2B6 (benzyloxyresorufin as the substrate), known biotransformation enzymes for this drug class (iminostillbene), and was highly correlated with lipid peroxidation and cell viability at environmentally relevant concentrations. Lipid peroxidation and cell viability are considered integrative biomarkers, while cytochrome P4503A4/2B6 activity is a target-specific biomarker. The second case study concerns feral carp that had survived for four years in an aerated lagoon that treats domestic municipal effluent. Results showed that cytochrome P3A4 activity, as determined by dibenzyloxyfluorescein (another substrate specific for cytochromes P450 3A4, 3A5 and 2C9), was readily induced in the post-mitochondrial supernatant of liver homogenates. ATP-dependent dopamine transport activity in synaptosome preparations of brain tissues was shown to be significantly reduced. Increased cytochrome P450-related activities and reduced dopamine uptake suggest the pharmacological effects of opiate-like substances. Preliminary findings thus indicate that some aquatic species are likely to accumulate some drugs and that they are likely to produce harmful effects on fish. Further research is needed to validate such biomarkers and to relate changes in drug targets with their residual levels in tissues.

Dr. Thomas W. Moon, a professor in the Department of Biology and Centre for Advanced Research in Environmental Genomics at the University of Ottawa, discussed environmental effects in freshwater ecosystems.

This presentation addresses changes in the knowledge base and methods used for effects assessment of pharmaceuticals in the environment over the past five years. Although a number of PPCPs are studied by academic researchers, this paper deals primarily with the fibrates and the selectiveserotonin re-uptake inhibitor (SSRI) fluoxetine, compounds commonly found in waste-water effluents and surface waters. Studies have indicated both groups act in fish as endocrine disruptors affecting the production of both sex (testosterone, estradiol) and stress (cortisol) steroids. Given that steroid synthesis shares common pathways but generally different tissues, tools for their study are similar. Fibrate drugs are peroxisomal proliferators (PPs) and as such act through a nuclear receptor, the peroxisomal-proliferator activated receptor (PPAR) which heterodimerizes with the RXR (retinoic acid X receptor, also a nuclear receptor) to bind to response elements in the promoter regions of PP-sensitive genes.

It has been demonstrated in goldfish that a fibrate drug, gemfibrozil, bioconcentrates in blood by more than 400-fold at environmental concentrations and reduces testosterone levels by 50%. To understand the mechanism(s) responsible for such dramatic changes, PPAR transcript levels and other components within the steroidogenic pathway were examined. Strong seasonal effects were found in all parameters studied which is probably (although not conclusively) related to the seasonal reproductive behavior of this species. We have also examined ACTH-induced cortisol release using a dispersed rainbow trout head-kidney cell preparation as a model that may be less seasonally-dependent. Using this system we have isolated the effects of fibrate drugs to specific enzymes within the cortisol synthetic pathway. What is intriguing about these results is that using a similar preparation, others have shown that salicylate and ibuprofen do not act on enzymes of the pathway but rather at the level of cholesterol entry.

We have also developed a very sensitive rainbow trout gill cell reporter-gene assay that responds to environmental concentrations of fibrate drugs. And finally, using a goldfish-carp cDNA array, it is demonstrated that fluoxetine not only up- and down-regulates a number of critical genes, but specifically down-regulates isotocin, a neuropeptide that modulates spawning and social behaviors in fish. Using similar designs, fluoxetine is shown to reduce the number of eggs produced and aromatase expression in zebrafish. These results demonstrate that we know much more about the mechanisms by which PPCPs act on fish, but at the same time we have huge gaps in our understanding of differential sensitivities of species, the seasonal responses that are seen and whether these can be linked specifically to seasonal reproductive cycles, and the role of mixtures of chemicals including PPCPs and their impacts. Clearly more studies using environmentally-relevant PPCPs concentrations, a toxicokinetic approach with parent compounds and metabolites, and a multi-generational approach but at defined time periods and chronic exposure, are required before we are able to demonstrate the full impact of PPCPs on aquatic organisms, including fish.

Dr Katsuji Haya, from the Marine Environmental Sciences Division of Fisheries and Oceans Canada, discussed the environmental effects of PPCPs on marine ecosystems.

Research in Canada concerned with the biological effects of PPCPs on marine organisms is limited. Overviews of active research on the exposure to municipal and industrial effluents during seaward migration of salmon; laboratory and field exposure to municipal waste water effluents; and environmental effects of salmon aquaculture wastes is presented.

Under laboratory conditions juvenile Atlantic salmon were exposed to sublethal concentrations of nonylphenol, estradiol, or methyl testosterone in fresh water for one to two weeks in May. The salmon were then acclimated to ambient sea water and held until October. Salmon exposed to the chemicals had a greater proportion of small fish compared to the controls. Similarly juvenile Biological effects monitoring using mussels | Photo: Photos.comsalmon caged in the Miramichi Esturary, New Brunswick, for one week had a greater proportion of small fish compared to those held in the less impacted Tabusintac Estuary. Juvenile salmon were exposed to nonylphenol, tagged and released in the Burrishoole River, Ireland in 2003 and 2004, and returning adults are being monitored.

Early life stages and underyearlings of rainbow trout, Coho salmon and Chinook salmon were exposed to both end of pipe and environmentally relevant concentrations of municipal wastewater in fresh water and saltwater. Chemical analyses for “emerging chemicals” were tailored to those previously identified as being of paramount concern with respect to biological effects. Gene array analysis of chemicals indicated significant up and down regulation of key genes; changes in immune-related and metabolism-based genes predominated the response to wastewater effluent; and individual analytical chemistry results did not correlate with biological effect(s).

Blue mussels were caged in various locations in Pictou Harbour, New Brunswick and Burrard Inlet, British Columbia and the potential toxicity of wastewater effluents on the immune system and disease resistance were determined. Results suggest that treated and untreated wastewater can modulate the immune system of blue mussels and that most immunological endpoints measured were sensitive to the exposure and not different than results observed from experiments under laboratory conditions.

Antibiotics are used in salmon aquaculture operations and administration is by feed fortified with the antibiotic. Oxytetracycline has been the antibiotic of choice for salmon aquaculture in Atlantic Canada. The presence of oxytetracycline has been observed in sediment samples collected from under salmon cages and along a transect 100 m from under the cage presumably due to excess feed wastes and excretory products from cultured salmon. Also antibiotic resistant bacteria have been observed in the sediments. There was no correlation between oxytetracycline concentrations and the occurrence of antibiotic resistant bacteria.

Dr. Tom Edge, of the Aquatic Ecosystem Protection Research Division of Environment Canada, provided an overview of antibiotic resistance as an environmental effect of PPCPs.

The spread of enteric bacteria with antibiotic resistance is a growing public health concern. While hospital settings and the retail food supply are increasingly recognized as important sources of these bacteria, the significance of waterborne sources is less understood. Large quantities of enteric bacteria from human and animal fecal wastes can be released into rivers and lakes that serve as sources of water for drinking, recreation or irrigation. A better understanding is needed of the prevalence of antibiotic resistance in these enteric bacteria, and the significance of their occurrence in aquatic ecosystems. The potential of antibiotic resistance analyses for microbial source tracking of fecal pollution also needs further investigation. Escherichia coli is a useful enteric bacterium for studying antibiotic resistance because it is adapted to diverse human and warm-blooded animal gastrointestinal tracts, and is readily exposed to a variety of medical and veterinary antibiotic treatments. It is also the basis of many microbial water quality decisions across Canada. We have found that E. coli from municipal wastewater sources generally have higher levels of antibiotic resistance than those from pet or wildlife fecal droppings. In some cases, these results can be used for microbial source tracking purposes to determine the source of fecal pollution in aquatic ecosystems. Antibiotic resistance analysis for source tracking seems to work because the effects of antibiotics occur largely at the point of use in intestinal tracts rather than from exposure to low levels of antibiotics in aquatic ecosystems. We have detected E. coli in Lake Ontario recreational waters that are resistant to antibiotics at clinical breakpoints, although they do not appear to be common from preliminary observations. Further research will be required to ascertain the significance of antibiotic resistant bacteria in untreated waters used for drinking (e.g. wellwater), recreation or irrigation of food crops.