Natacha S. Hogan

REPRODUCTIVE DEVELOPMENT OF YELLOW PERCH (PERCA FLAVESCENS) EXPOSED TO OIL SANDS-AFFECTED WATERS

In similar experiments conducted in 1996 and 2009, yellow perch (Perca flavescens) were stocked into two experimental systems: a demonstration lake where oil sands fine tailings were capped with natural water and a lake in a watershed containing bitumen-bearing sodic clays. In both experiments, yellow perch were captured in May from a nearby reservoir and released into the experimental ponds. Perch were recaptured in the experimental systems, the source lake, and two reference lakes in late September and lethally sampled to examine reproductive parameters. In the 1996 experiment, gonad size and steroid hormones were not affected in either pond environment. In the 2009 experiment, male perch in the water-capped tailings pond showed a significant reduction in the testicular development and reductions in circulating testosterone and 11-ketotestosterone, while no reductions were seen in the second experimental pond. No changes were observed in ovarian size or circulating steroid levels in female perch. In the pond containing tailings, the release of water from underlying tailings caused approximately a twofold increase in salinity, alkalinity, and naphthenic acids, and a pH increase from 8.4 to 9.4 over the 13-year period of the study. In the pond influenced by unextracted oil sands materials, total dissolved solids, major ions, and pH did not change substantially. However, naphthenic acids in this system dropped more than twofold post-watershed reclamation. Because the selective reproductive effect observed in male perch in the experimental end-pit lake were accompanied by increases in naphthenic acids, alkalinity, and pH, a specific cause cannot be determined. The present study adds to the evidence, suggesting the presence of endocrine-disrupting substances in oil sands.

Modulation of steroidogenesis and estrogen signalling in the estuarine killifish (Fundulus heteroclitus) exposed to ethinylestradiol.

Previous studies have shown that mummichog (Fundulus heteroclitus; a lunar, asynchronous-spawning killifish of the western Atlantic) exposed to 17alpha-ethynylestradiol (EE2) exhibit decreased plasma reproductive steroid levels, decreased gonadal steroid production, increased plasma vitellogenin, decreased fecundity and impaired fertilization. The objective of this study was to determine the potential mechanisms by which EE2 depresses gonadal steroidogenesis and influences estrogen signalling in the mummichog. Adult recrudesced fish were exposed to the potent synthetic estrogen, ethinylestradiol (EE2; 0-270ng/L) for 14 days. Following exposure, gonadal tissue was removed and incubated for 24h with stimulators of steroidogenesis, including forskolin; 25-OH cholesterol; or pregnenolone. Testosterone production was decreased in basal, forskolin-stimulated and pregnenolone-stimulated EE2-exposed males, indicating effects on the steroidogenic pathway both at and downstream of cholesterol mobilization to P450 side-chain cleavage (P450scc) and/or P450scc conversion of cholesterol to pregnenolone. Hepatic transcript levels of estrogen receptor alpha (ERalpha) and vitellogenin were increased in EE2-treated males compared to control recrudescing males and females confirming an estrogenic response. Hepatic heat shock protein 90 (Hsp90), a chaperoning molecule involved in estrogen signalling, was not affected by EE2 exposure at either the transcript or protein level. However, higher levels of Hsp90 observed in the membrane fractions of female fish raise interesting questions regarding the influence of gender on Hsp90s role in estrogen signalling. These results demonstrate that EE2 can alter steroid production at specific sites within the steroidogenic pathway and can stimulate hepatic estrogen signalling, providing important information regarding the molecular mechanisms underlying the endocrine response of the mummichog to exogenous estrogen.

Immunotoxic effects of oil sands-derived naphthenic acids to rainbow trout

Naphthenic acids are the major organic constituents in waters impacted by oil sands. To investigate their immunotoxicity, rainbow trout (Oncorhynchus mykiss) were injected with naphthenic acids extracted from aged oil sands tailings water. In two experiments, rainbow trout were injected intraperitoneally with 0, 10, or 100 mg/kg of naphthenic acids, and sampled after 5 or 21 d. Half of the fish from the 21 d exposure were co-exposed to inactivated Aeromonas salmonicida (A.s.) to induce an immune response. A positive control experiment was conducted using an intraperitoneal injection of 100 mg/kg of benzo[a]pyrene, a known immune suppressing compound. T-lymphocytes, B-lymphocytes, thrombocytes, and myeloid cells were counted in blood and lymphatic tissue using flow cytometry. In the 5d exposure, there was a reduction in blood leucocytes and spleen thrombocytes at the 100 mg/kg dose. However, at 21 d, leucocyte populations showed no effects of exposure with the exception that spleen thrombocyte populations increase at the 100 mg/kg dose. In the 21 d exposure, B- and T-lymphocytes in blood showed a significant Dose × A.s. interaction, indicating stimulated blood cell proliferation due to naphthenic acids alone as well as due to A.s. Naphthenic acid injections did not result in elevated bile fluorescent metabolites or elevated hepatic EROD activity. In contrast to naphthenic acids exposures, as similar dose of benzo[a]pyrene caused a significant decrease in B- and T-lymphocyte absolute counts in blood and relative B-lymphocyte counts in spleen. Results suggest that the naphthenic acids may act via a generally toxic mechanism rather than by specific toxic effects on immune cells.

Immunological impacts of oil sands-affected waters on rainbow trout evaluated using an in situ exposure

Rainbow trout were exposed in situ to oil sands-affected waters for 21 d, either with or without an immune stimulation using inactivated Aeromonas salmonicida. Three aquatic systems were utilized for the experiment: a pond containing oil sands tailings capped with approximately 3 m of natural surface water, a second pond where unextracted oil sands materials were deposited in the watershed, and a reservoir receiving Athabasca River water as a reference caging location. The three systems showed a gradient of oil sands-related compounds, most notably, total naphthenic acids were highest in the system containing tailings (13 mg/L), followed by the system influenced by unextracted oil sands (4 mg/L), followed by the reference cage location (1 mg/L). Biochemical and chemical measures of exposure in rainbow trout showed the same trend, with the tailings-influenced system having the highest hepatic EROD activity and elevated bile fluorescence measured at phenanthrene wavelengths. Trout caged in the tailings-influenced location had significantly fewer leukocytes and smaller spleens as compared to the reference fish, though liver size and condition factor were unaffected. Fish in the tailings-influenced waters also demonstrated increased fin erosion, indicative of opportunistic infection. The trout exposed to tailing-influenced waters also showed a significantly decreased ability to produce antibodies to the inactivated A. salmonicida. Given the complexity of the exposure conditions, exact causative agents could not be determined, however, naphthenic acids, polycyclic aromatic hydrocarbons and pH correlate with the immunotoxic effects while elevated salinity or metals seem unlikely causes.

Androgenic effects of a Canadian bleached kraft pulp and paper effluent as assessed using threespine stickleback (Gasterosteus aculeatus)

The presence of unidentified estrogens and androgens in effluents from pulp and paper mills is well documented. However, their role in effluent effects on fish reproduction remains unclear. The objective of this study was to investigate the hypothesis that reproductive impacts of a modern pulp mill effluent are mediated by androgens and/or estrogens in the effluent. Male and female threespine stickleback were exposed to biologically treated Canadian bleached kraft mill effluent under flow-through conditions in the laboratory at 0, 1, 10 and 100% (v/v) dilutions. After 7 and 21 d of exposure, steroidogenesis was assessed using in vitro incubations of gonadal tissue in both males and females. mRNA expression of the estrogen-regulated gene vitellogenin, and the androgen-responsive gene spiggin were assessed using quantitative RT-PCR in the livers of male and posterior kidneys of female stickleback, respectively. Hepatic 7-ethoxyresorufin-O-deethylase (EROD) activity was assessed in both sexes. Effluent extracts were examined for estrogenic and androgenic bioactivity using receptor binding bioassays, and were screened for pulp and paper related extractives and steroidal androgens using GC-MS. This effluent up-regulated spiggin mRNA in the kidney of female stickleback at 10% and 100% (v/v) effluent at 21 d, but not at 7 d of exposure. This change at the mRNA expression of the gene was associated with an increase in cell height in kidney proximal tubule epithelial cells at 100% effluent after both 7 and 21 d. Liver vitellogenin mRNA in male stickleback was not induced at either 7 or 21 d. EROD was induced at 10 and 100% after 21 d of exposure in both sexes, but not after 7 d of exposure. Despite evidence of exposure to androgens, there was no reduction in steroidogenic capacity at any effluent dilution. Effluent extracts were capable of eliciting the displacement of androgens and estrogens from receptors, but androgenic potency was 4-fold greater. A screen of more than 30 androgenic androstane steroids showed no detections. Hence, the androgenic constituents in this particular effluent remain unknown.

The immunological effects of oil sands surface waters and naphthenic acids on rainbow trout (Oncorhynchus mykiss)

There is concern surrounding the immunotoxic potential of naphthenic acids (NAs), a major organic constituent in waters influenced by oil sands contamination. To assess the immunological response to NAs, rainbow trout (Oncorhynchus mykiss) waterborne exposures were conducted with oil sands-influenced waters, NAs extracted and purified from oil sands tailings waters, and benzo[a]pyrene (BaP) as a positive control. After a 7d exposure, blood, spleen, head kidney, and gill samples were removed from a subset of fish in order to evaluate the distribution of thrombocytes, B-lymphocytes, myeloid cells, and T-lymphocytes using fluorescent antibodies specific for those cell types coupled with flow cytometry. The remaining trout in each experimental tank were injected with inactivated Aeromonas salmonicida and held in laboratory water for 21 d and subjected to similar lymphatic cell evaluation in addition to evaluation of antibody production. Fluorescent metabolites in bile as well as liver CYP1A induction were also determined after the 7 and 21 d exposure. Oil sands waters and extracted NAs exposures resulted in an increase in bile fluorescence at phenanthrene wavelengths, though liver CYP1A was not induced in those treatments as it was with the BaP positive control. Trout in the oil sands-influenced water exposure showed a decrease in B- and T-lymphocytes in blood as well as B-lymphocytes and myeloid cells in spleen and an increase in B-lymphocytes in head kidney. The extracted NAs exposure showed a decrease in thrombocytes in spleen at 8 mg/L and an increase in T-lymphocytes at 1mg/L in head kidney after 7d. There was a significant decrease in antibody production against A. salmonicida in both oil sands-influenced water exposures. Because oil sands-influenced waters affected multiple immune parameters, while extracted NAs impacts were limited, the NAs tested here are likely not the cause of immunotoxicity found in the oil sands-influenced water.

Simultaneous determination of androgenic and estrogenic endpoints in the threespine stickleback (Gasterosteus aculeatus) using quantitative RT-PCR

A method to evaluate the expression of three hormone responsive genes, vitellogenin (estrogens), spiggin (androgens), and an androgen receptor (ARbeta) using real-time PCR in threespine stickleback is presented. Primers were designed from previously characterised spiggin and ARbeta sequences, while a homology cloning strategy was used to isolate a partial gene sequence for stickleback vitellogenin (Vtg). Spiggin mRNA was significantly higher in kidneys of field-caught males compared to females by greater than five orders of magnitude while ARbeta levels were only 1.4-fold higher in males. Female fish had four order of magnitude higher liver Vtg expression than wild-captured males. To determine the sensitivity of these genes to induction by hormones, male and female sticklebacks were exposed to 1, 10 and 100 ng/L of methyltestosterone (MT) or estradiol (E2) in a flow-through exposure system for 7 days. Spiggin induction in females, and Vtg induction in males were both detectable at 10 ng/L of MT and E2, respectively. MT exposure did not induce ARbeta expression in the kidneys of female stickleback. In vitro gonadal steroid hormones production was measured in testes and ovaries of exposed stickleback to compare gene expression endpoints to an endpoint of hormonal reproductive alteration. Reduction in testosterone production in ovaries at all three MT exposure concentrations, and ovarian estradiol synthesis at the 100 ng/L exposure were the only effects observed in the in vitro steroidogenesis for either hormone exposure. Application of these methods to assess both androgenic, estrogenic, and anti-steroidogenic properties of environmental contaminants in a single fish species will be a valuable tool for identifying compounds causing reproductive dysfunction in fishes.

The effects of the urea-based herbicide linuron on reproductive endpoints in the fathead minnow (Pimephales promelas).

Linuron is a widely used urea-based herbicide that has anti-androgenic activity in both fish and rodents. To further elucidate the potential mode of action (MOA) of linuron on the vertebrate endocrine system, adult male and female fathead minnows were exposed for 21 days to dechlorinated water, a solvent control, 17β-estradiol (E2; 0.1 μg/L), dihydrotestosterone (DHT; 100 μg/L), linuron (1, 10, 100 μg/L) and one co-treatment of DHT (100 μg/L) and linuron (100 μg/L). There were no effects of linuron on egg hatching, 7 day egg survival, nuptial tubercle formation or gonadal histopathology. Administration of DHT and 1 and 100 μg/L linuron reduced plasma vitellogenin in females, while male plasma vitellogenin were induced after E2 exposure and co-exposure of DHT and linuron. Ovarian mRNA levels were examined for several genes involved in steroidogenesis (e.g. p450scc, cyp19a, star, tspo, hsd17b and hsd11b) and estrogen-mediated responses (esr1, esr2b, esr2a). Only p450scc mRNA was significantly decreased with DHT+linuron co-treatment. Clustering of steroidogenic mRNA transcript expression patterns revealed that patterns for linuron were more similar to E2 compared to DHT. Collectively, this study supports the hypothesis that linuron may not be a pure anti-androgen and may have multiple MOAs that affect vertebrate reproduction.

Phylogenetic analysis and molecular methods for the detection of lymphocystis disease virus from yellow perch, Perca flavescens (Mitchell).

Lymphocystis disease is a prevalent, non-fatal disease that affects many teleost fish and is caused by the DNA virus lymphocystis disease virus (LCDV). Lymphocystis-like lesions have been observed in yellow perch, Perca flavescens (Mitchell), in lakes in northern Alberta, Canada. In an effort to confirm the identity of the virus causing these lesions, DNA was extracted from these lesions and PCR with genotype generic LCDV primers specific to the major capsid protein (MCP) gene was performed. A 1357-base pair nucleotide sequence corresponding to a peptide length of 452 amino acids of the MCP gene was sequenced, confirming the lesions as being lymphocystis disease lesions. Phylogenetic analysis of the generated amino acid sequence revealed the perch LCDV isolate to be a distinct and novel genotype. From the obtained sequence, a real-time PCR identification method was developed using fluorgenic LUX primers. The identification method was used to detect the presence/absence of LCDV in yellow perch from two lakes, one where lymphocystis disease was observed to occur and the other where the disease had not been observed. All samples of fin, spleen and liver tested negative for LCDV in the lake where lymphocystis disease had not been observed. The second lake had a 2.6% incidence of LCD, and virus was detected in tissue samples from all individuals tested regardless of whether they were expressing the disease or not. However, estimated viral copy number in spleen and liver of symptomatic perch was four orders of magnitude higher than that in asymptomatic perch.

The effects of the alkyl polycyclic aromatic hydrocarbon retene on rainbow trout (Oncorhynchus mykiss) immune response

The objective of this study was to examine the immune toxicity of the alkyl polycyclic aromatic hydrocarbon (PAH) retene (7-isopropyl-1-methyl phenanthrene) in rainbow trout. Retene is a common alkyl PAH associated with combustion of terrestrial plants or industrial effluents. Rainbow trout were injected intraperitoneally with a single dose of 0, 1, or 10mg/kg of retene and sampled at 21d. Within each retene treatment, co-injection of formalin-killed Aeromonas salmonicida or a phosphate buffered saline sham was used to stimulate the trout immune system. At the highest retene dose (10mg/kg) there was an increase in total blood leukocyte counts but only in the A. salmonicida-injected group. This result was paralleled by an increase in leukocytes in differential blood cell counts. There was an overall increase in A. salmonicida-specific antibody titre due to the antigen injection and there was also a significant stimulation of antibody production response at the 10mg/kg retene dose. At the tissue level, immunohistochemistry revealed a greater density of B-lymphocytes at the highest retene dose in spleen and head kidney. A number of immune specific transcripts including the Th2 marker CD4, TCR, MHCII TNFα, and the Th1 marker CD8, INFγ were examined by quantitative RT-PCR in spleen and head kidney. There was no influence of the A. salmonicida antigen on the expression of Th1 related genes in either tissue but increased production of Th2 related CD4 co-receptor transcripts was observed in spleen at both retene concentrations. Retene appeared to be mildly immunostimulatory, either on its own, or in combination with an inactivated A. salmonicida immune challenge and these data suggest that exposures of fishes to retene may not pose significant risk of eliciting immunotoxic effects.