Do Toxicity Identification and Evaluation ... - ACS Publications

Jul 24, 2009 - Administration, c/o Waste Management Division, U.S.. Environmental ... 30303, and Harvard School of Public Health, Boston,. Massachuset...
0 downloads 0 Views 667KB Size
Download PDF
Environ. Sci. Technol. 2009, 43, 6857–6863

Do Toxicity Identification and Evaluation Laboratory-Based Methods Reflect Causes of Field Impairment? K A Y T . H O , * ,† M I C H E L L . G I E L A Z Y N , ‡ MARGUERITE.C. PELLETIER,† ROBERT. M. BURGESS,† MARK C. CANTWELL,† MONIQUE M. PERRON,§ JONATHAN R. SERBST,† AND ROXANNE L. JOHNSON† Atlantic Ecology Division, U.S. Environmental Protection Agency, Narragansett, Rhode Island 02882, Office of Response and Restoration, National Oceanic and Atmospheric Administration, c/o Waste Management Division, U.S. Environmental Protection Agency, Region IV, Atlanta, Georgia 30303, and Harvard School of Public Health, Boston, Massachusetts 02115

Received January 21, 2009. Revised manuscript received July 2, 2009. Accepted July 9, 2009.

Sediment toxicity identification and evaluation (TIE) methods are relatively simple laboratory methods designed to identify specific toxicants or classes of toxicants in sediments; however, the question of whether the same toxicant identified in the laboratory is causing effects in the field remains unanswered. The objective of our study was to determine if laboratory TIE methods accurately reflect field effects. A TIE performed on sediments collected from the Elizabeth River (ER) in Virginia identified polycyclic aromatic hydrocarbons (PAHs) as the major toxicants. Several lines of evidence indicated PAHs were the major toxic agents in the field, including elevated PAH concentrations in ER sediments, comet assay results from in situ caged Merceneria merceneria, and chemical analyses of exposed M. merceneria, which indicated high PAH concentrations in the bivalve tissue. Our final evidence was the response from test organisms exposed to ER sediment extracts and then ultraviolet (UV) radiation. UV radiation caused a toxic diagnostic response unique to PAHs. The aggregation of these various lines of evidence supports the conclusion that PAHs were the likely cause of effects in laboratory- and field-exposed organisms, and that laboratory-based TIE findings reflect causes of field impairment.

Introduction Sediment toxicity identification and evaluation (TIE) methods have been developed for freshwater, marine interstitial waters, and whole sediments, and guidance for using these methods has recently been released by the U.S. Environmental Protection Agency (1). These methods are designed to identify specific toxicants or chemical classes of toxicants * Corresponding author phone: (401) 782-3196; e-mail: ho.kay@ epa.gov. † U.S. Environmental Protection Agency. ‡ National Oceanic and Atmospheric Administration. § Harvard School of Public Health. 10.1021/es900215x CCC: $40.75

Published on Web 07/24/2009

 2009 American Chemical Society

by combining toxicity testing with chemical manipulations to remove or alter bioavailability of specific chemical classes. By comparing the toxicity of samples before and after specific chemical classes have been altered, we can characterize the chemical class likely responsible for the observed toxicity (1, 2). TIE methods used in marine sediments have successfully characterized toxicity from organic contaminants (1, 3, 4); however, the question of whether the same toxicant identified in the laboratory causes effects in the field remains unanswered. The objective of this study was to determine if the same toxicant(s) identified in the laboratory cause the effects observed in the field. To achieve this end, the following steps were taken: (1) Identify a field site that had toxicological and benthic community impacts. Elizabeth River (ER), VA (Figure 1 of the Supporting Information), has a history of creosote inputs largely from companies that preserved southern pine wood products on the shores of the river. Creosote, used as a wood preservative, contains very high concentrations of polycyclic aromatic hydrocarbons (PAHs). These preservation activities resulted in high sediment PAH concentrations (5), external abnormalities (6), and a high incidence of fish cancers in bottom dwelling fish (7). In addition to the documented toxicological effects on bottom dwelling fish, we performed benthic community analyses at ER and at a nearby reference site, Thorntons Creek (TC), VA, to determine if benthic community effects also occurred. (2) Identify the cause of toxicity in laboratory exposures with ER sediments using TIE methods. We used phase 1 (identification phase) TIE methods with juvenile Merceneria merceneria, the hard clam, to characterize the cause of toxicity from ER sediments. M. merceneria is an ecological and economically important species in the area. (3) Determine potential causes of a field impact(s). We exposed M. merceneria for two weeks to ER and TC in situ. We then analyzed the exposed clams for DNA damage using the comet assay. The comet assay, also called single cell gel electrophoresis, was chosen as a simple and sensitive technique to measure cellular DNA damage (8, 9). In the comet assay, cells exhibiting DNA damage are comet shaped with a circular head and an elongated tail. The comet assay is sensitive to DNA damage that has occurred from PAHs and other toxicants (10). (4) Collect any additional evidence linking the toxicant and effect, including specific mode of action, characteristic special effects of the suspected toxicant, or evidence of bioaccumulation. Chemical analysis of clam tissue was performed to determine the organic compounds that had bioaccumulated in the clams. We analyzed digestive glands from clams exposed in situ to ER and TC sediments as well as the sediments themselves for PAHs, polychlorinated biphenyls (PCBs), and pesticides. Bioaccumulation demonstrates that the compound is available to organisms. On the basis of the results from the previous three steps, we performed a photoenhanced toxicity exposure with juvenile M. merceneria exposed to sediment extracts from ER. Photoenhanced toxicity is an unusual mode of action associated with PAHs (11). Phototoxic responses of organisms to ultraviolet excited PAHs are well-known (11–13). Ultraviolet (UV) light enhanced toxicity occurs when UV absorbed by the conjugated bonds in the PAHs excites them to a triplet state (11). The energy of the excited PAH can be transferred to a ground state oxygen molecule causing high energy singlet oxygen and other high energy intermediaries that can be damaging to surrounding cells (11). Occurrence of phototoxicity would be additional evidence that toxicity of ER sediments was caused by PAHs. VOL. 43, NO. 17, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6857

Materials and Methods Step 1. Identification of a Field Site: Elizabeth River, VA. To ensure that ER had benthic community structure impacts as well as toxicologically impacted benthos, we looked at the macroinvertebrate benthic community. In 2002, Van veen sediment grab samples (0.045 m2) were collected from ER and TC (control site). TC was chosen as a field reference because it has no known point sources, with very little development (a few single family homes) and is well buffered by marshland and pine forest. Sediment samples were sieved through a 0.5 mm screen, the organisms identified, and data analyzed for benthic community differences between sites. Benthic community analyses were performed on species lists from the two sites using a similarity index (Primer, Plymouth Marine Laboratory, Plymouth, U.K.) and Fisher’s R to measure diversity. At the same time, surface sediments were collected (top 2-3 cm) for toxicity testing and chemical analyses. Sediment samples for toxicity testing and chemical analysis were respectively stored in the dark at 4 °C or frozen until analysis. Step 2. Identify the Toxic Agent(s): Toxicity Identification and Evaluation. Toxicity tests using M. merceneria and Americamysis bahia (mysids) were performed with TIEs on ER and the control sediment Long Island Sound (LIS). LIS is a chemically and geologically well-characterized control sediment used many times in our laboratory. LIS was used as the TIE control sediment rather than the TC field reference because of its availability. The M. merceneria growth assays were based on the methods of Ringwood and Keppler (14). Clams for all assays were obtained from Atlantic Littleneck Clam Farms (Charleston, South Carolina). Briefly, juvenile (1.2-1.4 mm) clams were shipped overnight and upon arrival held at 20 °C in 30 ppt seawater, fed Tetraselmis sp. for at least 24 h before use, and used within 72 h of arrival. Mysid toxicity tests were conducted according to standard methods using mysids cultured onsite (1, 15). Exposure chambers were 100 mL glass containers with 20 g of sediment or sediment with TIE additions (see below) and 60 mL of hypersaline brine diluted to 30 ppt reconstituted seawater (SW) (1, 15, 16). Ten clams or mysids were placed into each triplicate exposure chamber. An additional 40 clams were used to determine the initial weight for the growth end point. Testing was conducted under static conditions at 20 ( 2 °C with aeration, and mysids were fed newly hatched Artemia daily ad libidum. After 7 days, mysids and clams were sieved from the sediments. Mortality was recorded, and live clams were rinsed, dried, and weighed to determine growth. The TIE was performed on ER sediments using additions of (1) coconut charcoal to assess organic toxicity, (2) cation resin to assess metal toxicity, and (3) zeolite to assess ammonia toxicity (1) (see Supporting Information for more information on TIE methods). Results were analyzed using a one-tailed t test (R ) 0.05) (Microsoft Office Excel, 2003) after normalizing (dividing) for the TIE manipulation controls. Guided by TIE results, we performed chemical analysis of ER sediments (see Chemical Analysis). Step 3. Demonstrate and Determine Potential Causes of Field Impact(s). The comet assay was performed on M. mercenaria exposed in situ to ER and TC sediments. Twenty similar sized clams (average ) 3.7 cm height, 4.0 cm length) per site were exposed in mesh bags suspended at the sediment surface for two weeks. After deployment, clams were shipped (with shipping control clams) to the University of South Carolina Natural Resources Laboratories (Charleston, SC). Twenty clams from each site were processed for the comet assay. Hemolymph was removed from the adductor muscle and kept at 4 °C in the dark in 1.5 mL microcentrifuge tubes until processed (less than 1 h). The digestive gland was removed and stored at -80 °C for later chemical analyses. The comet assay was 6858

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 17, 2009

conducted as described in Gielazyn (17) but without enzyme treatment. Briefly, cells were embedded in layers of agarose on a microscope slide, and then lysed using a strong salt solution, leaving only DNA embedded in the agarose. The slides were placed in an electrophoresis solution and damaged DNA (i.e., DNA with strand breaks) migrated toward the anode. Cells exhibiting DNA damage appeared comet shaped with a circular head and an elongated tail; the longer the tail, the more DNA damage that had occurred. After electrophoresis, slides were neutralized (3 × 2 min rinses in 0.4 M Tris, pH 7.5), rinsed in ethanol, and dried for later analysis. For the analysis, DNA was stained with ethidium bromide (20 µg/mL). Digital analysis of the comet slides was conducted using the KOMET 4.0 image analysis software (Kinetic Imaging, Ltd., Liverpool, U.K.) in Dr. R. Lee’s laboratory (Skidaway Institute of Oceanography). Two slides (5000 to 10000 cells) were prepared for each site, and 50 cells per slide were quantified. Data analyses for the comet assay were conducted using standard methods of analysis of variance (ANOVA) (18). All analyses were conducted using SAS statistical software (SAS Institute, Inc., Cary, NC). Komet 4.0 software produces a number of indices, including comet tail length (micrometers), amount of DNA in the tail (% based on tail intensity), and tail moment (product of tail length and amount of tail DNA). Because tail moment is a product of two different measurements, its units are not meaningful. All three measures were recorded for 50 cells per slide. Observations for analysis were based on per clam averages. Adjustments for outliers were made by removing any observations that exceeded their upper quartile by more than 1.5 times the interquartile range (IQR) or dropped below their lower quartile by more than 1.5 times the IQR. To identify differences between field sites, we used a Tukey-Kramer analysis of all possible pairwise treatment differences (19). Step 4. Additional Information: Specific Mode of Action, Bioaccumulation Chemical Analyses- Sediment and Tissue. Chemical Analyses: Sediment and Tissue. Chemical analyses for PCBs, PAHs, and pesticides were performed on ER and TC sediments and clams exposed in situ to ER and TC sites. All samples were spiked with internal standards d-10 phenanthrene, d-12 benz[a]anthracene, and d-12 perylene for PAHs and congener 198 (IUPAC number 198) for PCBs. The clam digestive gland samples were homogenized with a Brinkman Polytron and extracted with an acetone:hexane 50:50 solvent mixture 3 times. Sediments were thoroughly mixed, dried, and extracted with an acetone:hexane 50:50 solvent mixture 3 times. Extracts were volume reduced to 1 mL and exchanged to hexane. Total PAHs (∑PAH) were quantified as the sum of 13 compounds: fluorene, phenanthrene, anthracene, fluoranthene, pyrene, benz[a]anthracene, chrysene, benzo[e]pyrene, benzo[a]pyrene, perylene, indeno[1,2,3,cd]pyrene, dibenz[ah]anthracene, and benzo[ghi]perylene. PAHs were analyzed using an Agilent 6890 gas chromatograph equipped with a 5973 mass selective detector and a DB-5 MS 60 m capillary column (Agilent Technologies, Wilmington, DE). Limits of detection for individual PAHs were 65 ng/g in sediments and 20 ng/g for clams. For PCB and pesticide analysis, an aliquot of sediment or tissue was treated with concentrated sulfuric acid, followed by addition of activated copper to remove sulfur. The chlorinated compounds were quantified using a Hewlett-Packard series II 5890 gas chromatograph (Hewlett-Packard, Avondale, PA) equipped with an electron capture detector and a 30 m DB-5 capillary column (Agilent Technologies, Wilmington, DE). Both instruments were calibrated with NIST traceable calibration standards and multipoint calibrations. Twentythree PCB congeners (IUPAC numbers 8, 18, 28, 52, 44, 66, 101, 99, 110, 151, 118, 153, 105, 138, 187, 183, 128, 180, 170, 194, 195/208, 206, and 209) and five pesticides (Aldrin and

FIGURE 1. Species abundance in Elizabeth River and Thorntons Creek. Abundance is the sum of triplicate 0.5 m2 samples at each site. p,p′-DDE, heptachlor, hexaclorobenzene, and mirex) were measured. The detection limit for PCBs and pesticide compounds in clams and sediments was 0.5 ng/g. All concentrations are reported in dry weight unless otherwise stated. Total organic carbon was measured in sediments after acidification with 1 M HCl using a ThermoFinnigan FlashEA Series 1112 automated elemental analyzer (ThermoQuest Italia SpA, Milan, Italy). Photoenhanced Toxicity. When testing for photoenhanced toxicity, we felt it was important to test in an aqueous system that had toxicant concentrations and partitioning behavior similar to interstitial water in contact with whole sediment but did not have a whole sediment matrix that could block the UV light. We used semipermeable membrane devices (SPMDs) that contained sediment extracts from ER or LIS. Placing SPMDs containing organic extracts in water is a method used to create and maintain solutions with organic toxicant concentrations similar to interstitial water (20, 21). Extracts from ER or LIS sediments were prepared using reverse SPMD methods (20). Briefly, 40-100 g of air-dried, ground sediment was mixed with an equal amount of anhydrous sodium sulfate and extracted 3 times with 75 mL of hexane:acetone:dichloromethane (60:20:20, v/v/v) in a sonic bath at 35 °C. The extracts were combined, dried with anhydrous sodium sulfate, and volume-reduced to 40 mL. The extracts were then solvent exchanged to triolein and placed into SPMDs [50 cm of cleaned low-density polyethylene (LDPE) tubing, 2.5 cm wide with 100 µm thick walls (EST Laboratories, St. Joseph, MO) containing 0.5 g of triolein extract (20)]. The SPMDs were placed into 150 mL of 30‰ SW and allowed to equilibrate for 2 days before adding organisms. Test organisms were either juvenile (1.2-1.4 mm) M. mercenaria or 48 h old mysids. Organisms were exposed to the reverse SPMD containing either ER or LIS control sediment extracts for 7 days and then exposed to UV light for 2 days. The UV exposures were conducted in an environmental chamber with half illuminated with fluorescent lights (TL841, Philips, Somerset, NJ) and the other half with UVA 340 bulbs that simulate natural sunlight (Q-Panel, Cleveland, OH). The UV area and fluorescent light area were separated by black plastic. Ultraviolet light was measured daily using an International light meter (International Light Technologies, Peabody MA) positioned 26 cm from each of the light sources (the same distance as from the water surface in the test chambers). UVA and UVB were measured at 365

( 36 and 310 ( 34 nm, respectively. Fluorescent and UV light regimes were conducted under a 16:8 light:dark photoperiod. Temperature was held constant at 21.5 ( 0.7 °C. Survival of clams and mysids was assessed as described previously in the TIE section.

Results and Discussion Benthic Community. Benthic communities in ER were significantly different from benthic communities at the TC reference site using a similarity index (ANOSIM, Bray-Curtis (22)), (3 replicates, 2 sites; significant at p ) 0.1). Diversity and evenness metrics were also calculated for both sites. Overall, there were half as many species in ER (S ) 9) as there were in TC (S ) 18). In contrast, the overall number of organisms was higher in ER (N ) 105) than in TC (N ) 41) (Figure 1). Diversity, measured using Fisher’s R (23), indicated that TC was more diverse (R ) 12.36) than ER (R ) 2.35). Both communities were relatively species poor: this may have been influenced by a severe drought preceding and during the sampling period that increased the salinity in both areas. TC was dominated by worms such as Streblospio benedicti and other small deposit feeding spionid and capitellid polychaetes. This site also contained mollusks and crustaceans. The feeding strategies of the animals at this site included suspension feeding (clams and amphipods), surfacedeposit feeding (spionid and capitellid polychaetes), deepdeposit feeding (maldanid polychaete), and predation. In contrast, ER was dominated by smaller surface-deposit feeders and suspension feeders. Contaminated areas are expected to be dominated by smaller, surface-dwelling organisms (24). While ER was dominated by smaller organisms (Laeonereis culveri, a small herbivorous polychaete classified as pollution tolerant by Borja et al. (25) and Gemma gemma, a small deposit feeding clam), Eteone heteropoda, a phyllodocid polychaete predator was also present. Streblospio benedicti was absent at this site, which was surprising knowning this species has an association with organically enriched sediment (26, 27). Elizabeth River had higher sediment organic carbon (8.63%) than TC (3.48%). However, this species is also sensitive to salinity fluctuations (28). Although two point measures taken during this study indicated similar salinities (19-26 ppt at TC and 17-26 ppt at ER), the evidence suggests that ER may experience more salinity fluctuations due to runoff. First, the species composition suggests salinity fluctuations; in addition to the absence of S. benedicti, G. gemma, which is known to be VOL. 43, NO. 17, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6859

FIGURE 2. Chemical concentrations of PAHs in sediments and in situ clams exposed to Elizabeth River and Thorntons Creek sediments. All clam concentrations have been normalized by the shipping control. euryhaline (13-28 ppt) (29), was present in large numbers. Second, the surrounding land use suggests that ER may experience more salinity fluctuations due to runoff than TC. The ER site is highly developed, while TC has little development and a large Spartina alterniflora marsh, which would probably act as a buffer to intercept runoff. These data indicate the benthic community in TC was more diverse with more species and a wider variety of feeding strategies than the benthic community in the ER. While this was not diagnostic as many factors including salinity, temperature, grain size, larval supply, and contaminants can influence benthic community abundance and distribution, it did show that the ER benthic communities appeared to be impacted. Toxicity Identification and Evaluation Results. Mysids exhibited no toxicity in the initial TIE toxicity screening, so no further TIE testing was performed with that species. Clams exposed to ER sediment resulted in significantly lower growth than clams exposed to LIS sediment (p ) 0.005). Reduced growth rates have been linked to population level effects in a number of studies (30, 31). Clam TIE manipulation controls indicated no difference between LIS and LIS + amendment exposures (Figure 2 of the Supporting Information). Clams exposed to ER sediment amended with coconut charcoal (CC) had a significantly higher growth (p ) 0.01) than those exposed to unamended ER sediment (Figure 3 of the Supporting Information). This implies that an organic toxicant which adsorbs onto CC was responsible for the toxicity. Clams exposed to ER sediment with cation resin or zeolite amendments have an apparent increase in toxicity, which may be a result of the organisms responding to ER stress compounded by cation and zeolite resin stress. While the results of the cation and zeolite resin amendments were unusual, the LIS manipulation blanks (reported above) did not indicate differences between amended and unamended sediments. These results indicate neither metals or ammonia caused toxicity to the organisms and characterized the toxicity of the sediment to be organic in nature; however, they do not identify a specific organic toxicant. Follow-up chemical analyses were performed as part of the TIE (see Chemical Analysis). Chemical Analysis. Chemical analysis of sediments indicated high concentrations of PAHs in ER sediment (∑13 PAHs ) 1090000 ng/g dry weight) and general background concentrations of PAHs in TC (∑13 PAH ) 523 ng/g dry weight) (Figure 2). Organic carbon normalized PAH concentrations in ER (ER ) 12700 µg/g organic carbon, TC ) 15.0 µg/g organic carbon) are a few orders of magnitude 6860

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 17, 2009

above the consensus threshold effects concentrations (TEC ) 290 µg/g organic carbon), and above the extreme effects concentrations (EEC ) 10000 µg/g organic carbon) for total PAH mixtures (32). This implies that the observed PAH levels could cause toxic and community level effects for organisms in these sediments. Chemical analysis of the clam digestive glands exposed in situ to ER indicated that PAHs were found in the same general pattern in the digestive gland and sediment (r ) 0.855), but concentrations in the clam (49990 ng/g) were several orders of magnitude lower than that in the sediments (Figure 2). The accumulation factor (AF) (concentration in the dry weight clam/concentration in the dry weight sediment) for 13 PAHs in ER had a mean of 0.036 (range ) 0.009-0.082). For TC, the clam digestive gland (4262 ng/g) and sediment had similar PAH patterns (r ) 0.905), but the clams accumulated PAHs at slightly higher concentrations than found in the sediment (Figure 2) (mean AF ) 7.07, range ) 5.21-11.3). Within this study, clams exposed to ER sediments accumulated lower concentrations of PAHs relative to the sediment than did clams exposed to TC sediments. Nasci et al. (33) reports PAH values in M. mercenaria digestive gland of 2822.7 ng/g wet weight (15162 ng/g dry weight) in contaminated sediment (13510 ng PAH/g dry weight sediment) and residues of 25.5 ng/g wet weight (137.1 ng/g dry weight) in clean sediments (34.6 ng PAH/g dry weight sediment). The AFs for their contaminated site was 0.89, an order of magnitude higher than ours, and 0.252 for the clean site, an order of magnitude lower than ours. Reasons for the lower concentrations and AFs in the ER clams may include the higher TOC content of ER (8.63%) relative to TC (3.48%). Higher levels of TOC can bind and decrease PAH bioavailability (34). Also, the very high PAH concentrations in ER sediment may have caused the ERexposed clams to stop feeding, thereby limiting the amount of exposure that occurred at ER relative to TC and other sediments that are not as heavily contaminated. The lower AF at the clean site also may be a reflection that our organisms were not fully equilibrated during their two week exposure. Analysis of PCBs in ER and TC clam digestive glands indicated concentrations generally below detection limits (0.5 ng/g). For PCBs, when they were detected, congener concentrations were less than 2.3 ng/g, and the sum of the congeners in ER was less than 40 ng/g. Clams from TC had a slightly higher PCB congener sum (80 ng/g); however, these concentrations are generally considered low background levels (35). In addition, developmental and biochemical

effects do not seem to occur at PCB concentrations below 114 and 360 ng/g, respectively (36, 37). Chemistry results also indicated that hexachlorobenzene, heptachlor, p,p′-DDE, aldrin, and mirex concentrations were at or below detection limits of 0.5 ng/g. Comet Assay and Toxicological Effects. Clams deployed in ER for two weeks exhibited significantly greater levels of DNA damage in hemocytes than did clams similarly deployed in TC or the shipping controls (p ) 0.05) (Figure 4 of the Supporting Information). ER clams had higher damage as detected by all three comet indices: tail moment, tail length, and percent DNA in comet tail (p ) 0.03, 0.02, and 0.04, respectively). These results indicate that a bioavailable contaminant in ER sediments induced DNA damage in the hemocytes of exposed clams. DNA or genetic damage can lead to necrosis, apoptosis, or heritable mutations and can impact populations as well as individuals (38–40). The comet assay is sensitive to a number of compounds that induce DNA damage such as PAHs, PCBs, and some metals. Together with the TIE results and chemical analyses, these results indicate that PAHs are bioavailable and toxic in field exposures at ER. Nasci et al. (33) found that PAH concentrations of approximately 15161 ng/g dry weight in the digestive gland of M. mercenaria increased levels of benzo(a)pyrene hydroxylase, which indicates the presence of PAHs, and catalase, which protects the cell from harmful effects of oxidants. In addition, they observed a decrease in latency, a measure of cell permeability, and increased lysosome fragility. On the basis of concentrations in our clams, these effects would have occurred in the ER exposed clams (49990 ng/g) but not the TC exposed clams (4262 ng/g). Unlike this study, residue data for other bivalves is often reported as whole body concentrations. Ahn et al. (41) found that the bivalve digestive gland concentrated metals approximately 2.3 times more than muscle tissues. If the digestive gland:muscle homogenate ratio is applied to PAH concentrations in bivalve muscle homogenates, the literature levels for effects ranged from 2300-36000 ng/g dry weight in digestive gland homogenates. Effects from these levels included increased hemocycte count and hemocyte killing index (42), stress response (43), decreased growth (44),and delay in gametogenesis (45). These levels are lower than the residues observed in ER clam digestive gland homogenates (49990 ng/g) and are more evidence that the bioaccumulated PAH concentrations we observed from our field-exposed organisms are sufficiently high to exert toxicity. Photoenhanced Toxicity. The photoenhanced toxicity experiments were performed to demonstrate that the PAHs accumulated in the organisms exposed to ER sediments were capable of causing a biological effect. Results from these experiments indicate clams and mysid shrimp exposed to ER sediments and UV lights had a photoenhanced toxicity response (Figure 3). Mysid shrimp survival dropped from 70% in ER: no UV treatment to 0% survival in ER with UV light treatment (p ) 0.02 normalized for LIS control). Clams survival dropped from 65% survival in an ER: no UV treatment to 25% survival in the ER-UV treatment (p ) 0.02 normalized for LIS control). While the clam response was significant, it was less dramatic than the mysid response. This was most likely due to the protective shell on the clams shielding them from direct UV light tissue exposure. UVA and UVB exposures for treatments under laboratory fluorescent lights were 6.69 ( 1.06 and 0.002 ( 0.0006 µW/cm2, respectively. UVA and UVB exposures for treatments under UV lights were 293 ( 27.3 and 0.069 ( 0.006 µW/cm2, respectively. While we did not measure the UV spectra at the sediment-water interface, NOAA charts indicate the average depth of the two estuaries at our stations is approximately 0.3 m. The UV spectra penetrates estuarine waters (46, 47), and estimating that the

FIGURE 3. Survival of mysids and clams exposed to Elizabeth River and control sediment (LIS) extracts in a semipermeable membrane exposure under laboratory conditions and ultraviolet light exposure. * indicate a significant difference between sediment extract exposure with and without ultraviolet light treatment. UV spectrum is 10% of incident light (48) for incident light values measured at 0.5 m depth in the ER (Chesapeake Bay program http://www.chesapeakebay.net/data_waterquality. aspx), we can estimate UVA and UVB levels to be 87 and 102 µW/cm2, respectively, at the sediment interface. These UVA levels are lower than our laboratory levels; however, UVB levels are higher indicating a possibility of UV effects on field organisms. Overall, these results complement the tissue analyses showing evidence that PAHs can accumulate and have the potential to cause toxicity in field-exposed organisms. Overall Discussion and Conclusion. The evidence presented in this research agrees with recent evidence in the scientific literature pointing to PAHs as the cause of toxicity in the ER. Wassenberg et al. (49, 50) showed that Fundulus heteroclitus embryos exposed to ER sediment extracts responded similarly to fish exposed to PAH model compounds and differently from fish exposed to PCB model compounds. The objective of this research was to determine if the toxicants identified in the laboratory by TIE methods were also causing toxicity to organisms in the field. We demonstrated that ER had an impaired benthic community relative to the reference area TC. A TIE performed on ER sediments and M. mercenaria indicated organic toxicants were the major cause of toxicity. Subsequent chemical analysis guided by the TIE indicated high levels of PAHs in the sediment, while other measured organic contaminants including PCBs and several pesticides were not present at elevated levels. Comet assay results from M. merceneira exposed in situ to ER sediments indicated ER contained bioavailable compounds that increase DNA damage. The comet assay responds to a number of organic toxicants, including PAHs and PCBs; however, chemical analysis performed on the same exposed M. mercenaria with DNA damage indicate high concentrations of PAHs in the clam digestive gland tissues but no detectable or low levels of PCBs and pesticides. Exposure of clams and mysids to an extract of ER sediment in a system designed to mimic an interstitial water exposure and low levels of UV light resulted in mortality due to phototoxicity, which is characteristic of PAH compounds but not PCBs. In addition, it is probable that field levels of UVA and UVB could also cause effects. While it is difficult to definitively “prove” the cause of toxicity in the field, this body of evidence is consistent with the results of the laboratory TIEs and identified PAHs as the class of compounds causing toxicity in the field. When planning a TIE, the choice of toxicity test organism is important. Under ideal conditions, the toxicity test organism is a surrogate for field organisms. If the response of the toxicity test organisms is not linked to the field response, the TIE chemical manipulations may correctly identify the VOL. 43, NO. 17, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6861

source of toxicity in the toxicity test, but it may not be the source of toxicity in the field. Ultimately, the TIE process will fail, despite the fact that chemical manipulations were properly performed. This demonstration project linked laboratory TIE identified toxicants to field effects. In this case, a TIE has correctly identified the active toxicant in the field; however, these results cannot be extrapolated to all toxicants under all field conditions. We found that in situ exposures were key to linking field and laboratory findings, and the steps used in this research generate a large enough body of evidence to determine if the TIE has identified the correct stressor in the field.

Acknowledgments We gratefully acknowledge the assistance of Drs. Amy Ringwood, Walter Piegorsch, Richard Lee, Sharon Hook, Deena Wassenberg, and Bruce Voght. This paper is contribution number AED-08-073 of the U.S. Environmental Protection Agency (U.S. EPA), Atlantic Ecology Division (AED) and has been technically reviewed by AED; however, it does not necessarily represent the views of the U.S. EPA. No official endorsement of any aforementioned products should be inferred.

Supporting Information Available Details of the TIE methods, toxicity test results, and a location map. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Sediment Toxicity Identification Evaluation (TIE) Phases I, II, and III Guidance Document; EPA/600/R-07/080; Office of Research and Development, U.S. Environmental Protection Agency: Washington, DC, 2007. (2) Marine Toxicity Identification Evaluation (TIE) Procedures Manual: Phase I Guidance Document; EPA 600/R-96/054; Office of Research and Development, U.S. Environmental Protection Agency: Washington, DC, 1996. (3) Anderson, B.; Hunt, J.; Phillips, B.; Tjeerdema, R. Navigating the TMDL Process: Sediment Toxicity; 02-WSM-2 Final Report; Water Environment Research Foundation: Alexandria, VA, 2006; p 181. (4) Ho, K. T.; McKinney, R. A.; Kuhn, A.; Pelletier, M. C.; Burgess, R. M. Identification of acute toxicants in New Bedford Harbor sediments. Environ. Toxicol. Chem. 1997, 16 (3), 551–558. (5) Huggett, R. J.; Bender, M. E.; Unger, M. A. Polynuclear aromatic hydrocarbons in the Elizabeth River, Virginia. In Fate and Effects of Sediment-Bound Chemicals in Aquatic Systems: A Special Publication of SETAC; Dickson, K. L., Maki, A. W., Brungs, W. A., Eds.; Pergamon Press: New York, 1987; pp 327-352. (6) Hargis, W.; Colvocoresses, J.; Williams, C.; Zwerner, D.; Thoney, D.; Faisal, M. Associations of several externally visible lesions in finfishes with contaminated sediments in the lower Chesapeake Bay. Mar. Environ. Res. 2000, 50 (1-5), 309. (7) Fournie, J. W.; Vogelbein, W. K. Exocrine pancreatic neoplasms in the mummichog (Fundulus heteroclitus) from a creosotecontaminated site. Toxicol. Pathol. 1994, 22 (3), 237–247. (8) Nacci, D. E.; Cayula, S.; Jackim, E. Detection of DNA damage in individual cells from marine organisms using the single cell gel assay. Aquat. Toxicol. 1996, 35, 197–210. (9) Steinert, S. A.; Streib-Montee, R.; Leather, J. M.; Chadwick, D. B. DNA damage in mussels at sites in San Diego Bay. Mutat. Res. 1998, 399, 65–85. (10) Woo, S.; Kim, S.; Yum, S.; Yim, U. H.; Lee, T. K. Comet assay for the detection of genotoxicity in blood cells of flounder (Paralichthys olivaceus) exposed to sediments and polycyclic aromatic hydrocarbons. Mar. Pollut. Bull. 2006, 52 (12), 1768–1775. (11) Newsted, J. L.; Giesy, J. P. Predictive models for photoinduced acute toxicity of polycyclic aromatic hydrocarbons to Daphnia magna, Strauss (Cladocera, Crustacea). Environ. Toxicol. Chem. 1987, 6, 445–461. (12) Pelletier, M. C.; Burgess, R. M.; Ho, K. T.; Kuhn., A.; Mc Kinney, R. A.; Ryba, S. A. Phototoxicity of individual polycyclic aromatic hydrocarbons and petroleum to marine invertebrate larvae and juveniles. Envioron. Toxicol. Chem. 1997, 16 (10), 2190–2199. 6862

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 17, 2009

(13) Kosian, P. A.; Makynen, E. A.; Monson, P. D.; Mount, D. R.; Spacie, A.; Mekenyan, O. G.; Ankley, G. T. Application of toxicitybased fractionation techniques and structure-activity relationship models for the identificaton of phototoxic polycyclic aromatic hydrocarbons in sediment pore water. Environ. Toxicol. Chem. 1998, 17, 1021–1033. (14) Ringwood, A. H.; Keppler, C. J. Seed Clam Growth: An alternative sediment bioassay developed during EMAP in the Carolinian Province. Environ. Monit. Assess. 1998, 51, 247–257. (15) Ho, K. T.; Kuhn, A.; Pelletier, M.; Mc Gee, F.; Burgess, R. M.; Serbst, J. Sediment toxicity assessment: comparison of standard and new testing designs. Arch. Environ. Contam. Toxicol. 2000, 39, 462–468. (16) Ho, K. T.; Mitchell, K.; Zappala, M.; Burgess, R. M. Effects of brine addition on effluent toxicity and marine toxicity identification evaluation (TIE) manipulations. Environ. Toxicol. Chem. 1995, 14 (2), 244–249. (17) Gielazyn, M. L.; Ringwood, A. H.; Piegorsch, W. W.; Stancyk, S. E. Detection of oxidative DNA damage in isolated marine bivalve hemocytes using the comet assay and formamidopyrimidine glycosylase (Fpg). Mut. Res. 2003, 542 (1-2), 15–22. (18) Lovell, D. P.; Thomas, G.; Dubow, R. Issues related to the experimental design and subsequent statistical analysis of in vivo and in vitro comet studies. Teratog. Carcinog. Mutagen. 1999, 19, 109–119. (19) Neter, J.; Kutner, M. H.; Nachtsheim, C. J.; Wasserman, W. Applied Linear Statistical Models; Irwin, R. D., Ed.; McGrawHill: Chicago, 1996. (20) Heinis, L. J.; Highland, T. L.; Mount, D. R. Method for testing the aquatic toxicity of sediment extracts for use in identifying organic toxicants in sediments. Environ. Sci. Technol. 2004, 38 (23), 6256–6262. (21) Perron, M. M.; Burgess, R. M.; Ho, K. T.; Pelletier, M. C.; Friedman, C. L.; Cantwell, M. G.; Shine, J. P. Development and evaluation of reverse polyethylene samplers for marine phase II wholesediment toxicity identification evaluations. Environ. Toxicol. Chem. 2009, 28 (4), 749–758. (22) Clarke, K. R.; Warwick, R. M. Change in Marine Communities: An Approach to Statistical Analysis and Interpretation, 2nd ed.; PRIMER-E: Plymouth, U.K., 2001. (23) Fisher, R. A.; Corbet, A. S.; Williams, C. B. The relation between the number of species and the number of individuals in a random sample of animal population. J. Anim. Ecol. 1943, 12, 42–58. (24) Pearson, T. H.; Rosenberg, R. Macrobenthic succession in relation to organic enrichment and pollution of the marine environment. Ocean Mar. Biol. Ann. Rev. 1978, 16, 229–311. (25) Borja, A.; Franco, J.; Perez, V. A marine biotic index to establish the ecological quality of soft-bottom benthos within European estuarine and coastal environments. Mar. Pollut. Bull. 2000, 40 (12), 1100–1114. (26) Rakocinski, C. F.; Brown, S. S.; Gaston, G. R.; Heard, R. W.; Walker, W. W.; Summers, J. K. Macroresponses to natural and contaminant-related gradients in northern Gulf of Mexico estuaries. Ecol. Appl. 1997, 7, 1278–1298. (27) Weisberg, S. B.; Ranasighe, J. A.; Schaffner, L. C.; Diaz, R. J.; Dauer, D. M.; Frithsen, J. B. An estuarine index of biological integrity (BIBI) for Chesapeake Bay. Estuaries. 1997, 20, 149–158. (28) Reish, D. J. Bristle Worms (Annelida: Polychaeta). In Pollution Ecology of Estuarine Invertebrates Hart, C. W., Fuller, S. L. H., Eds.; Academic Press: New York, 1979; pp 77-125. (29) Sellmer, G. P. Functional morphology and ecological life history of the gem clam, Gemma gemma. Malacologia 1967, 5, 137–223. (30) Klok, C.; de Roos, A. M. Population level consequences of toxicological influences on individual growth and reproduction in Lumbricus rubellus (Lumbricidae, Oligochaeta). Ecotoxicol. Environ. Saf. 1996, 33 (2), 118–127. (31) Rinke, K.; Petzoldt, T. Modeling the effects of temperature and food on individual growth and reproduction of Daphnia and their consequences on the population level. Limnologica 2003, 33 (4), 293–304. (32) Swartz, R. C. Consensus sediment quality guidelines for polycyclic aromatic hydrocarbon mixtures. Environ. Toxicol. Chem. 1999, 18 (4), 780–787. (33) Nasci, C.; Da Ros, L.; Campesan, G.; Van Vleet, E. S.; Salizzato, M.; Sperni, L.; Pavoni, B. Clam transplantation and stress-related biomarkers as useful tools for assessing water quality in coastal environments. Mar. Pollut. Bull. 1999, 39 (1-12), 255–260. (34) Lake, J. L.; Rubinstein, N. I.; Lee II, H.; Lake, C. A.; Heltshe, J.; Pavignano, S. Equilibrium partitioning and bioaccumulation of sediment-associated contaminants by infaunal organisms. Environ. Toxicol. Chem. 1990, 9, 1096–1106.

(35) Rubinstein, N. I.; Pruell, R. J.; Taplin, B. K.; Livolsi, J. A.; Norwood, C. B. Bioavailability of 2,3,7,8-TCDD, 2,3,7,8-TCDF and PCBs to marine benthos from Passaic River sediments. Chemosphere 1990, 20 (7-9), 1097–1102. (36) Ferreira, A. M.; Vale, C. PCB accumulation and alterations of lipids in two length classes of the oyster Crassostrea angulata and of the clam Ruditapes decussatus. Mar. Environ. Res. 1998, 45 (3), 259–268. (37) Ryan, R. L.; Lachmayr, K. L.; Jay, J. A.; Ford, T. E. Developmental effects of PCBs on the hard clam (Merceneria merceneria). J. Environ. Sci. Health, Part A 2001, 36 (9), 1571–1578. (38) Theodorakis, C. W. Integration of genotoxic and population genetic endpoints in biomonitoring and risk assessment. Ecotoxicology 2001, 10 (4), 245–256. (39) Jha, A. N. Ecotoxicological applications and significance of the comet assay. Mutagenesis 2008, 23 (3), 207–221. (40) Gagne´, F.; Blaise, C.; Pellerin, J.; Fournier, M.; Durand, M. J.; Talbot, A. Relationships between intertidal clam population and health status of the soft-shell clam Mya arenaria in the St. Lawrence Estuary and Saguenay Fjord (Que´bec, Canada). Environ. Int. 2008, 34 (1), 30–43. (41) Ahn, I.-Y.; Lee, S. H.; Kim, K. T.; Shim, J. H.; Kim, D.-Y. Baseline heavy metal concentrations in the Antarctic clam, Laternula elliptica, in Maxwell Bay, King George Island, Antarctica. Mar. Pollut. Bull. 1996, 32 (8-9), 592–598. (42) Oliver, L. M.; Fisher, W. S.; Volety, A. K.; Malaeb, Z. Greater hemocyte bactericidal activity in oysters (Crassostrea virginica) from a relatively contaminated site in Pensacola Bay, Florida. Aquat. Toxicol. 2003, 64 (4), 363–373.

(43) Weinstein, J. E. Fluoranthene-induced histological alterations in oysters, Crassostrea virginica: Seasonal field and laboratory studies. Mar. Environ. Res. 1997, 43 (3), 201–218. (44) Fukuyama, A. K.; Shigenaka, G.; Hoff, R. Z. Effects of residual Exxon Valdez oil on intertidal Protothaca staminea: Mortality, growth, and bioaccumulation of hydrocarbons in transplanted clams. Mar. Pollut. Bull. 2000, 40 (11), 1042–1050. (45) Frouin, H.; Pellerin, J.; Fournier, M.; Pelletier, E.; Richard, P.; Pichaud, N.; Rouleau, C.; Garnerot, F. Physiological effects of polycyclic aromatic hydrocarbons on soft-shell clam Mya arenaria. Aquat. Toxicol. 2007, 82 (2), 120–134. (46) Barron, M. G.; Little, E. E.; Calfee, R.; Diamond, S. Quantifying solar spectral irradiance in aquatic habitats for the assessment of photoenhanced toxicity. Environ. Toxicol. Chem. 2000, 19 (4), 920–925. (47) Vo, M.; Porter, D.; Chandler, G.; H., K.; S., W.; Jones, B. E. Assessing photoinduced toxicity of polycyclic aromatic hydrocarbons in an urbanized estuary. Ecol. Soc. 2004, 9 (5), 3. (48) Lowe, N. J. Sunscreens: Development, Evaluation, and Regulatory Aspects; 2nd ed.; Marcel Dekker: New York, 1997. (49) Wassenberg, D. M.; Giulio, R. T. D. Teratogenesis in Fundulus heteroclitus embryos exposed to a creosote-contaminated sediment extract and CYP1A inhibitors. Mar. Environ. Res. 2004, 58 (2-5), 163–168. (50) Wassenberg, D. M.; Swails, E. E.; Di Giulio, R. T. Effects of single and combined exposures to benzo(a)pyrene and 3,3′4,4′5-pentachlorobiphenylonERODactivityanddevelopmentinFundulusheteroclitus. Mar. Environ. Res. 2002, 54 (3-5), 279–283.

ES900215X

VOL. 43, NO. 17, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6863