ARTICLE pubs.acs.org/est
Alkylnaphthalenes: Priority Pollutants or Minor Contributors to the Poor Health of Marine Mussels? Alan G. Scarlett,† Robert Clough,† Charles West,† C. Anthony Lewis,† Andrew M. Booth,† and Steven J. Rowland*,† †
Petroleum & Environmental Geochemistry Group, Biogeochemistry Research Centre, University of Plymouth, Plymouth PL4 8AA, Devon, U.K.
bS Supporting Information ABSTRACT: Alkylnaphthalenes (AN) are relatively water-soluble hydrocarbons which, following spillages of crude oils, have been widely reported in contaminated marine organisms such as mussels. In the present report we show, by tandem-gas chromatography-time-of-flight-mass spectrometry (GC GCToF-MS), that the range of AN in contaminated wild mussels from the UK extends beyond the previously GC resolved isomers to those with at least seven substituent carbon atoms. Since surprisingly little information on AN toxicity to such marine organisms has been reported we synthesized two C8 AN and measured the toxicity of C28 AN to mussels (clearance rate assay). C23 AN were appreciably toxic (concentration for 50% clearance rate inhibition, 48 h IC50 1.42.6 μmol g1 dry weight tissue), but several C4, 6 and C8 AN, including branched isomers expected to be resistant to biodegradation and more accumulative, were relatively nontoxic (48 h IC50 > 10 μmol g1) and longer term exposure (8 d) failed to elicit a greater toxic response. The accumulation profiles of AN in laboratory mussels exposed to oil were similar to those of the wild mussels. Moreover, laboratory oil-exposed mussels depurated toxic C23 AN within 5 days in clean water and clearance rates recovered. The latter might imply that, in contrast with branched alkyl benzenes tested previously, AN are of less toxic concern, but such a straightforward conclusion cannot necessarily be drawn; a synthetic branched C8 AN persisted following depuration and was as toxic to mussels as a C3 AN (IC50 1.3 μmol g1). This indicates that the structures of AN are also important.
’ INTRODUCTION Although oil spills from ships are in decline, pipeline bursts and sporadic blowouts continue to pose an environmental risk, as evidenced by the Deep Horizon Macondo well spill in the Gulf of Mexico in 2010.1 Even in the absence of accidents, produced waters from both onshore and offshore oil operations provide a significant source of hydrocarbon contaminants in the environment.2 Monitoring of hydrocarbons has largely been focused on the USEPA priority pollutant list of 16 polycyclic aromatic hydrocarbons (PAH), even though within crude oils, alkyl derivatives are much more prevalent than the parent compounds and environmental risk can be underestimated.3 Due to the increasing hydrophobicity with alkyl carbon number, the relative contribution toward toxic effects is potentially greatly increased. For example, it was reported that C4 isomers could contribute over 50 higher toxic units than the same concentration of naphthalene in sediment pore water.4 Naphthalene is the lowest molecular weight compound in the USEPA PAH list and due to its high volatility tends to be more rapidly lost when oil is spilled on the surface into aquatic and marine systems. However alkylnaphthalenes (AN) are less volatile and more commonly detected, and both naphthalene and AN may better survive underwater discharges such as that in the Gulf spill.1 AN are certainly major components of the water accommodated-fractions of r 2011 American Chemical Society
crude oils.5 Measured and estimated properties of selected AN are provided in Table 1. Studies by comprehensive gas chromatography gas chromatography time-of-flight mass spectrometry (GC GCToF-MS) of ‘unresolved’ complex mixtures (UCM) of aromatic hydrocarbons accumulated in health-impacted mussels collected from around the UK coastline revealed the presence of many isomers of branched alkylbenzenes (BABs), branched alkylindanes (BINs), and branched alkyl tetralins (BATs).6,7 Laboratory studies of a commercial mixture of BABs and of synthetic BINs and BATs suggested that large numbers of isomers of such compounds were toxic to mussels.68 Mussels are regarded as sentinel species that act as indicators for the general health of ecosystems throughout the world.9 Beds of health-impacted mussels with large accumulations of UCM hydrocarbons have been shown to contain communities with lower species diversity.10 Mussels are also important sources of food for many animals including humans and hence are of considerable commercial importance. Also detected in UK coastline mussels Received: April 12, 2011 Accepted: June 14, 2011 Revised: June 10, 2011 Published: June 14, 2011 6160
dx.doi.org/10.1021/es201234a | Environ. Sci. Technol. 2011, 45, 6160–6166
Environmental Science & Technology
ARTICLE
Table 1. Summary of Estimated and Measured Properties of Selected Alkylnaphthalenes Derived by Software Included within the EPISuite Estimation Programs Interface21 and References Thereina mol wt.
log
log
water solubility
water solubility
log BCF log BCF HC biodeg. half-life
(Dalton) KOW est. KOW exp. est. at 25 °C (mg L1) exp. at 25 °C (mg L1) est. low est. upp. naphthalene 2-methylnaphthalene
128.2 142.2
3.17 3.72
3.30 3.86
142.10 41.42
31.0 24.6
1.84 2.21
2.25 2.81
est. (days) 5.6 8.9
2-ethylnaphthalene
156.2
4.21
4.38
12.94
8.0
2.56
3.02
6.1
2,6-dimethylnaphthalene
156.2
4.26
4.31
14.85
2.0
2.51
2.87
14.2 13.0
2-iso-propylnaphthalene
170.3
4.63
-
6.89
-
2.72
3.31
2-ethyl-6methylnaphthalene
170.3
4.75
-
5.34
-
2.80
3.06
9.7
2,3,5-trimethylnaphthalene
170.3
4.81
-
4.78
-
2.84
2.94
22.7
2,6-diethylnaphthalene
184.3
5.25
-
1.74
-
3.13
3.20
6.6
2,6-diisospropylnaphthalene 2,7-ditert-butylnaphthalene
212.3 240.4
6.08 6.99
-
0.24 0.03
-
3.68 4.28
2.84 3.67
30.5 26.8
2-(1,5-dimethylhexyl)naphthalene
240.4
7.01
-
0.03
-
4.12
3.48
32.2
a
est. = estimated; exp. = experimental; HC = hydrocarbon; biodeg. = biodegradation (derived using BioHCwin v1.01. Bioconcentration Factor (BCF) derived from BCFBAF v3.00. Low refers to lower trophic level BCF from regression model, and upp. refers to upper trophic level from Arnot-Gobas method.
examined previously 6 were mixtures of AN, but the toxic effects of these, if any, were not studied. In fact, rather surprisingly, given the widespread environmental occurrence of AN, of which about 1400 isomers are theoretically possible even for the C16 isomers, little information on AN toxicity to marine organisms appears to have been reported at all. This possibly partly reflects the difficulties of working with these semivolatile hydrocarbons and the costs of, and somewhat restricted range of, isomers available from commercial sources. We therefore now report the toxicity of numerous commercial and two synthetic C28 substituted nonbranched and branched AN to the mussel, Mytilus sp. Using GC GC-TOF-MS analyses of tissue extracts, we compare the bioaccumulation of extended ranges of AN in mussels exposed in the laboratory to a weathered crude oil with those accumulated by wild mussels collected from the UK coast.
’ MATERIALS AND METHODS Oils and Fractionation. Fresh Alaskan North Slope crude oil was evaporatively weathered to achieve a 36% loss of mass, using a method described previously.11 The weathered oil was fractionated by open column chromatography, also as described previously.12 In brief, aliquots of oil were adsorbed onto deactivated alumina (4.5% Milli-Q water w/w). The column was packed with the deactivated alumina over activated silica (ratio 1:1 w/w) and eluted with increasingly polar solvents. The aromatic fraction (90:10 (v/v) hexane:dichloromethane) was used for toxicity testing. Chemicals. 2-methyl-, 2-ethyl-, 2,6-dimethylnaphthalenes were purchased from Sigma-Aldrich (Poole, UK). 2-ethyl-6methyl-, 2,3,5-trimethyl-, 2-isopropyl-, 2,6-diethyl-, 2,6-di-isopropyl-, and 2,7-di-tertbutylnaphthalenes were purchased from Chiron (Trondheim, Norway). 1- and 2-(10 ,50 -dimethylhexyl)naphthalene were synthesized as detailed below. Synthesis. 2- or 1-naphthyloctylmagnesium bromide solution (1.0 M in diethyl ether) and cerium III chloride heptahydrate (99.999%) were supplied by Sigma-Aldrich (Poole, UK). The 6-methylheptan-2-one was available from a previous study.13 All solvents were supplied by Rathburns (Walkerburn, UK). Palladium on carbon catalyst (5%) was supplied by BDH
(Poole, UK). The synthetic method used was based upon a cerium chloride promoted Grignard addition. The resultant alcohol and alkenes were purified from the other products in the crude mixture including binaphthyl and naphthalene by column chromatography or preparative high performance liquid chromatography. The alcohol and alkenes were hydrogenolyzed or hydrogenated to the hydrocarbon which was assigned by GCMS and 1H and 13C NMR spectroscopy (Figure S1). Purities of >99% were established by GC-MS for both compounds. Toxicology. The preparation of alkylnaphthalene solutions, exposure conditions, and measurement of mussel clearance rates was similar to the methods described previously;6 oil exposure test conditions were as reported in detail by Frenzel et al.14 for the aromatic fraction of Tia Juana Pesado crude oil (further details in the Supporting Information). The nominal concentration of the aromatic fraction of Alaskan North Slope oil in seawater was 3.9 mg L1. Collection and Maintenance of Mussels (Mytilus spp.). Mussels with mean shell length of 20.5 mm (standard error 0.07 mm), used for oil exposure tests, were collected from Port Quin, Cornwall, UK (50° 35.220 N, 004° 52.300 W). Alkylnaphthalene exposure test mussels, with mean shell length 38.6 mm (standard error = 0.08 mm), were collected from Trebarwith Sands, Cornwall, UK (N50° 38.720 , W004° 45.650 ) and maintained as previously reported.7 Most previous studies have referred to the species as M. edulis e.g. Donkin et al.,15 but metabolomic studies have since suggested the N. Cornwall coast population is mainly or entirely M. galloprovincialis.16 Exposure Tests. Acute (48 h) semistatic exposure tests were performed using test solutions of individual C28 alkylnaphthalenes (concentration range 0.05 to 2 mg L1). In addition, an equimolar solution comprising of 0.37 μmol L1 of the individually tested compounds, excluding the in-house synthesized C8 substituted isomers, i.e. a mixture of eight compounds (Table 2), was tested to check for additive effects. For both the equimolar and oil exposure tests, additional mussels were subject to identical exposure conditions followed by 5 d in clean seawater to test for depuration and recovery. A longer term trial (8 d with daily water exchange) was conducted using 2.5 μmol L1 2,6-dimethylnaphthalene to test the effects of prolonged exposure. 6161
dx.doi.org/10.1021/es201234a |Environ. Sci. Technol. 2011, 45, 6160–6166
Environmental Science & Technology
ARTICLE
Table 2. Tissue Concentrations of Alkylnaphthalenes Required To Inhibit Clearance Rates of Mussels by 20% (IC20) and 50% (IC50) with 95% Confidence Limitsa
a
Abbreviations used in Figure 2 and text provided in parentheses. nc: not calculated as 50% effect not achieved.
Test solutions were prepared by injecting 0.5 mL of an acetone solution of the test compound in 10 L of filtered seawater held at 15 °C in a 10 L glass aspirator (i.e., acetone concentration 0.005% v/v). Due to solubility issues, a 0.01% carrier solution was used for the oil exposure tests as described in detail previously.14 The test solutions were vortex mixed (velocity reduced to 0.05), which was similar to the maximum inhibition observed following 48 h exposure (29%). It is therefore unlikely that there was insufficient time for internal redistribution of contaminants and that sites of toxic action were not being reached. 6164
dx.doi.org/10.1021/es201234a |Environ. Sci. Technol. 2011, 45, 6160–6166
Environmental Science & Technology Following 5 d depuration in clean water, the mussels exposed to the equimolar mixture fully recovered (Figure 1), and clearance rates were indistinguishable from those of untreated control mussels, despite considerable tissue concentrations of gC6 AN isomers remaining. It may thus be concluded that wild mussels with high tissue burdens of higher molecular weight AN isomers may not suffer from impaired clearance rates as a result. However, in contrast, no recovery was evident for mussels exposed to 0.37 μmol L1 of 2-(10 ,50 -dimethylhexyl)naphthalene following depuration for five days in clean seawater (Figure S5). (The alpha (1-) isomer had the same effect (Table 2)). This compound was least depurated of all the AN tested (40%), and therefore such compounds are perhaps more likely to bioaccumulate, especially if mussels are only periodically exposed to clean seawater. Biodegradation tests were outside the remit of this study, but the BioHCwin v1.01 model21 predicts hydrocarbon biodegradation half-lives of 32 d for 2-(10 ,50 -dimethylhexyl)naphthalene, 27 d for 2,7-DtBN, and considerably shorter periods for the lower molecular compounds e.g. six d for 2-EN (Table 1). This suggests that the isomers with similar structure to 2-(1,50 -dimethylhexyl)naphthalene will persist longer in the environment. Such compounds should therefore be considered as bioaccumulative, relatively persistent, and toxic. The present study was confined to accumulation of alkylnaphthalenes and their effects on mussels. Bivalve mollusks have a very limited ability to metabolize naphthalenes, and therefore effects relating to toxicity of metabolites were unlikely to occur. Evidence suggests that large alkyl side-chains, and/or a high level of alkyl substitution of the aromatic ring, protect the compound from biotransformation into highly reactive and harmful intermediates as these substituents are preferentially oxidized over the aromatic ring.22 Thus, it is unlikely that the larger alkylnaphthalenes will be converted into toxic ring-oxidized metabolites which are known for naphthalene and its monomethylated derivatives. This study has shown that relatively high tissue concentrations were found to be required to reduce mussel clearance rates by 20%. Although it is clear that while some of the compounds are indeed toxic to Mytilus spp, there appears to be a toxicity cutoff for some naphthalenes substituted with C > 3 alkyl substituents. Thus
