Lethal and sublethal effects of simvastatin, irgarol, and PBDE-47 on the estuarine fish, Fundulus heteroclitus

Abstract

This study investigated the effects of simvastatin, a lipid-regulating drug; irgarol, an antifouling biocide; and PBDE-47, a brominated flame retardant, on the estuarine fish, Fundulus heteroclitus. Sublethal effects (changes in glutathione (GSH), lipid peroxidation (LPx), acetylcholinesterase (AChE), and cholesterol (CHL) levels) and lethal effects (survival) were determined after individual exposure to the three compounds. There were no significant differences in GSH or CHL levels in fish exposed to any of the test compounds. LPx levels significantly decreased with increasing irgarol concentrations. AChE levels were significantly lower in fish exposed to simvastatin at the 1.25 mg/L concentration and significantly higher at the PBDE-47 concentration of 0.0125 mg/L. The LC50 values were 2.68, 3.22, and > 0.1 mg/L for simvastatin, irgarol and PBDE-47, respectively.

Keywords:

• Simvastatin
• irgarol
• PBDE-47
• Fundulus heteroclitus

Introduction

Increased urbanization of southeastern US coastal areas has led to contamination of adjacent sediments and waters with a variety of compounds, including pharmaceuticals, pesticides, and industrial by-products.^{[ 1 ]} These compounds may have deleterious effects on estuarine biota.

Simvastatin, a lipid regulating drug sold under the trade name Zocor®, inhibits 3-hydroxy-3-methylglutaryl coenzyme A which lowers total and low-density lipoprotein cholesterol production.^{[ 2 ]} It has also been shown to decrease ubiquinone, which could result in impaired antioxidant protection.^{[ 3 ]} Irgarol 1051 is an herbicide ingredient in many marine antifouling paints used worldwide. It causes photosystem II inhibition like other triazine herbicides. PBDE-47 is one of a class of flame retardants (polybrominated diethyl ethers) found in many consumer products. It has been shown to accumulate in crustaceans, fishes and marine mammals and may interfere with endocrine system function.^{[ 4 ]}

The mummichog, Fundulus heteroclitus, inhabits estuaries from the Gulf of St. Lawrence south to northeastern Florida. These fish are quite abundant and density can range up to 6 fish/m² in tidal creeks. They are ecologically significant as they are prey for commercially and recreationally important fishes in addition to aquatic birds. Mummichogs also have an important role in mosquito control.^{[ 5 ]}

The purpose of this research was to determine the lethal effects (survival) on Fundulus heteroclitus using the three contaminants simvastatin, irgarol, and PBDE-47; along with measuring sublethal effects using the biomarkers glutathione (GSH), lipid peroxidation (LPx), acetylcholinesterase (AChE), and cholesterol (CHL).

Materials and methods

Collection

Mummichogs (F. heteroclitus), at 45 – 75 mm in length, were collected from Cherry Point Creek (N 32°36′ 04.29^″; W 080°11′07.01^″), a tidal tributary of the North Edisto River Estuary, SC, USA. Fish were acclimated at least seven days in 76-L tanks at 25°C, 20 ppt salinity and 16-h light:8-h dark cycle and fed Tetramin® Fish Flakes.

Toxicity tests

To assess acute toxicity 96-h static renewal toxicity tests were performed with each compound. Tests were conducted in 4-L glass jars to which 3.5-L of test media was added per jar. There were three replicates per concentration with five fish in each jar. The nominal test concentrations of simvastatin were 0, 0.625, 1.25, 2.5 and 5.0 mg/L. The nominal test concentrations of irgarol were 0, 0.625, 1.25, 2.5 and 5.0 mg/L. The nominal test concentrations of PBDE-47 were 0, 0.0125, 0.025, 0.05 and 0.10 mg/L. PBDE-47 concentrations higher than 0.10 mg/L were not attempted due to its low water solubility ranging from 0.00146 to 0.0153 mg/L.^{[ 6 ]} The jars were placed in an environmental chamber set at 25°C with a 16-h light:8-h dark cycle and aerated. Fish were not fed during the tests. At each 24 h media change, mortality and water quality parameters [temperature (°C), pH, salinity (ppt), and dissolved oxygen (mg/L)] were recorded.

Biomarker tests

For the biomarker tests, separate 96-h exposures were performed at levels based on previous testing and under conditions as described for toxicity testing. The nominal test concentrations of simvastatin were control, 0.625, 1.25, 2.5 and 5.0 mg/L. The nominal test concentrations of irgarol were control, 0.3125, 0.625, 1.25, and 2.5 mg/L. The nominal test concentrations of PBDE-47 were control, 0.0125, 0.025, 0.05 and 0.10 mg/L. These tests were performed as described previously. At the end of each test, surviving fish were dissected and livers removed for CHL, GSH and LPx analysis and brains removed for AChE analysis. All samples were frozen after collection at −70°C until further analysis.

Glutathione (GSH) concentrations were assessed using the 5,5′-dithiosbis(2-nitrobenzoic acid)-oxidized glutathione (DTNB-GSSG) reductase recycling assay described in Ringwood et al.^{[ 7 ]} Data are expressed as GSH (nmol/g wet weight). Lipid peroxidation (LPx) was assessed using the thiobarbituric acid assay, as described in Ringwood et al.,^{[ 7 ]} to quantify malondialdehyde (MDA), a by-product of LPx. Data are expressed as MDA (nmol/g wet weight). Acetylcholinesterase (AChE) concentrations were assessed as described in Key and Fulton.^{[ 8 ]} Data are expressed as AChE (nmol/mg protein/min). Cholesterol (CHL) concentrations were assessed using the Amplex Red Cholesterol Kit (Molecular Probes A12216) as based on Hoguet and Key.^{[ 9 ]} Data are expressed as CHL (mg/mg tissue).

Chemical analysis

For all tests with the three compounds, the stock solutions were made in 100% pesticide grade acetone which was used as a carrier (0.1%). Acetone was added to the control groups equal to the amount of carrier solvent used for the tests.

Technical grade simvastatin (97% purity) was obtained from Sigma–Aldrich (St. Louis, MO). Analytical grade simvastatin (99.4% purity) was obtained from Fisher Scientific (Fairlawn, NJ). Stock solutions were quantified to be 101.8% of nominal for a 10,000 mg/L stock solution and 100.4% of nominal for a 100 mg/L stock solution.^{[ 10 ]}All nominal treatment concentrations were made from these stocks.

For all irgarol tests, technical grade Irgarol 1051 (98.8% purity) was obtained from Sigma-Aldrich (St. Louis, MO, USA). For chemical analysis, Irgarol 1051 standard was obtained from Ciba Specialty Chemicals Inc. (Tarrytown, NY, USA). The stock solution was quantified to be 92.6% of a nominal 10000 mg/L stock solution.^{[ 11 ]} All nominal treatment concentrations were made from this stock.

For all PBDE-47 tests, technical grade (99% purity) was obtained from Chem Service Inc. (West Chester, PA). Analytical grade PBDE-47 in isooctane (100 mg/L) was obtained from o2si smart solutions, LLC (Charleston, SC). The stock solution was quantified to be 103.8% of nominal for a 100 mg/L stock solution.^{[ 12 ]} All nominal treatment concentrations were made from this stock.

Statistics and analysis

Median lethal concentrations (LC50s) with 95% confidence intervals were determined using the trimmed Spearman-Karber method.^{[ 13 ]} For the mortality data, analysis of variance (ANOVA) with Dunnett's procedure for comparison to determine significant differences from the control response was performed to determine the Lowest Observed Effect Concentration (LOEC) and the No Observed Effect Concentration (NOEC).^{[ 10 , 11 , 14 ]} For comparisons within each biomarker assay, the data were analyzed with ANOVA. For the biomarker tests, a one-way ANOVA with Dunnett's (PROC GLM, SAS Institute Cary, NC, USA) procedure for multiple comparisons versus the control was used to detected significant differences between treatment groups. If ANOVA showed no significance, a monotonic trend analysis (isotonic regression) was performed on this data using Williams' test as described by Newman.^{[ 15 ]} For all statistical tests, alpha was set to 0.05 a priori.

Results and discussion

Average water quality parameters for the toxicity tests are shown in Table 1. Dissolved oxygen levels ranged from 5.91 to 6.57 mg/L, pH ranged from 7.44 to 7.67, temperature ranged from 22.02 to 24.08°C and salinity ranged from 19.62 to 21.39 ppt. These ranges were all within the acclimated conditions described previously and within natural mummichog field conditions.^{[ 5 ]}

Table 1 Average water quality parameters with standard deviations for the simvastatin, irgarol and PBDE-47 96-h toxicity tests.

Compound	Dissolved oxygen (mg/L)	pH	Temperature (°C)	Salinity (ppt)
Simvastatin	5.91 ± 0.64	7.58 ± 0.15	24.08 ± 0.31	19.88 ± 0.09
Irgarol	6.57 ± 1.35	7.44 ± 0.31	22.29 ± 0.36	19.62 ± 0.385
PBDE-47	6.52 ± 1.17	7.67 ± 0.19	22.02 ± 0.27	21.39 ± 0.39

The 96-h LC50 test results for all compounds are shown in Table 2. A 96-h LC50 could not be obtained for PBDE-47 as there was only 13.4% mortality in the highest exposure of 0.1 mg/L. As stated previously, further testing to obtain a 96-h LC50 in mummichogs was not attempted due to the compound's low water solubility. In other estuarine species, a grass shrimp (Palaemonetes pugio) 96-h LC50 was obtained at 0.078 mg/L.^{[ 12 ]} However, 50% mortality was not obtained with other estuarine crustaceans tested.^{[ 16 , 17 ]} PBDE-47 is an important contaminant in terms of bioaccumulation in aquatic animals and has been found in surface waters at concentrations much lower than that of simvastatin and irgarol. In North America, PentaBDEs (of which PBDE-47 is a major constituent), have been found in surface waters at concentrations up to 0.00016 μ g/L.^{[ 4 ]} The main concern with PBDE exposure in aquatic animals is bioaccumulation.^{[ 12 ]}

Table 2 Toxicity values (mg/L) for mummichogs exposed to simvastatin, irgarol and PBDE-47 for 96-h.

Compound	LC50 (95 % CI)	NOEC	LOEC
Simvastatin	2.68 (2.21–3.24)	1.25	2.50
Irgarol	3.22 (2.85–3.64)	1.25	2.50
PBDE-47	> 0.10	0.05	0.10
NOEC: No observed effect concentration. LOEC: Lowest observed effect concentration.

Simvastatin has been measured in sewage treatment plant effluent and influent at levels of 0.001 and 0.004 μ g/L, respectively.^{[ 18 ]} Clearly, this is well below the LC50 and LOEC values for mummichogs (Table 2). Other published toxicity values in fish could not be found. When toxicity is compared to other organisms, the adult grass shrimp (Palaemonetes pugio) 96-h LC50 was above 10 mg/L, while the larval grass shrimp 96-h LC50 was 1.18 mg/L^{[ 10 ]} which was comparable to the mummichog value of 2.68 mg/L.

Irgarol has been measured in seawater worldwide at levels up to 4.2 μ g/L in Singapore^{[ 19 ]} to 1.03 μ g/L in the US.^{[ 20 ]} As with simvastatin, these levels are well below the LC50 and LOEC values for mummichogs (Table 2). The 96-h LC50 of 3.22 mg/L for mummichogs is comparable to other estuarine fish values of 1.58 mg/L for Menidia beryllina and 3.5 mg/L for Cyprinodon variegatus.^{[ 20 ]} Crustaceans, such as grass shrimp, had a similar 96-h LC50 value of 2.46 mg/L.^{[ 11 ]}

Biomarker results for simvastatin only yielded significant changes for AChE. All concentrations measured had AChE levels lower than the control, but only the lowest concentration of 1.25 mg/L was significantly reduced (p = 0.0193; Figure 1A). While the NOEC for mortality was 1.25 mg/L, clearly sublethal effects were occurring. However, this was still well above published environmental concentrations (see above). While simvastatin can reduce the production of CHL, it did not occur in mummichogs at the levels used in this research. In fact, there was a slight increase from 0.005674 mg/mg tissue for the controls up to 0.007074 mg/mg tissue for the 5.0 mg/L exposure.

Fig. 1 Acetylcholinesterase (AChE) levels in mummichogs exposed to simvastatin (A.) and PBDE-47 (B.) for 96-h. ∗Significantly different from control value.

Irgarol biomarker results only showed significant changes with LPx levels (Figure 2). After a peak in the lowest concentration of 0.313 mg/L, LPx levels then showed a significant downward trend (p = 0.0059). Sublethal effects were therefore occurring at levels 4 times lower than the NOEC of 1.25 mg/L but still well above reported environmental levels. Other biomarker levels were not significantly affected by irgarol exposures.

Fig. 2 Lipid peroxidase levels expressed as malondialdehyde (MDA) in mummichogs exposed to irgarol for 96-h. Williams' test indicated a downward significant trend from control to the highest concentration.

As with simvastatin, PBDE-47 biomarker results were only significant in the AChE assay. AChE levels in the lowest concentration of 0.0125 mg/L were significantly higher than controls (p = 0.0239; Figure 1B). Significant sublethal effects occurred below the PBDE-47 NOEC value but were still well above environmental levels. For all compounds tested, biomarker data with no significant effects were not shown.

No one biomarker of the four tested emerged as a useful measure of simvastatin, irgarol or PBDE-47 exposure in mummichogs. Both the LC50s and the sublethal effects detected in simvastatin and irgarol exposures occurred at levels many times higher than that found in the environment. For PBDE-47, a better assay of exposure in mummichogs may be a measure of thyroid levels since this compound bioaccumulates. Contamination of the estuarine environment is an evolving occurrence due to the vast array of manmade chemical products available. While individually these three compounds may not appear to be toxic compared to environmental levels, it is the mixture of these contaminants with other anthropogenic compounds that should be investigated.

Acknowledgments

NOAA's National Ocean Service (NOS) does not approve, recommend, or endorse any proprietary product or material mentioned in this publication.