Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
Environmental Origins of Methylmercury Accumulated in Subarctic Estuarine Fish Indicated by Mercury Stable Isotopes Miling Li, Amina Traore Schartup, Amelia P. Valberg, Jessica D. Ewald, David P. Krabbenhoft, Runsheng Yin, Prentiss H. Balcom, and Elsie M. Sunderland Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b03206 • Publication Date (Web): 02 Oct 2016 Downloaded from http://pubs.acs.org on October 4, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 35
Environmental Science & Technology
1
Environmental Origins of Methylmercury Accumulated in Subarctic Estuarine Fish
2
Indicated by Mercury Stable Isotopes
3 4
Miling Li1,2*, Amina T. Schartup1,2, Amelia P. Valberg1,2, Jessica D. Ewald2, David P.
5
Krabbenhoft3, Runsheng Yin4, 5, Prentiss H. Balcom2, Elsie M. Sunderland1,2
6 7
1
8
Boston, MA, USA
9
2
Department of Environmental Health, Harvard T.H. Chan School of Public Health,
Harvard John A. Paulson School of Engineering and Applied Sciences, Harvard
10
University, Cambridge MA, USA
11
3
U.S. Geological Survey, Middleton, WI, USA
12
4
Department of Civil and Environmental Engineering, University of Wisconsin-Madison,
13
WI, USA
14
5
15
Academy of Sciences, Guiyang 550002, China
State Key Laboratory of Ore Deposit Geochemistry, Institute of Geochemistry, Chinese
16 17
*Corresponding author. E-mail:
[email protected]. Phone: 617-496-5492. Fax:
18
617-496-4551. Address: 29 Oxford Street, Cambridge, MA 02138, USA.
ACS Paragon Plus Environment
1
Environmental Science & Technology
Page 2 of 35
19
Abstract
20
Methylmercury (MeHg) exposure can cause adverse reproductive and
21
neurodevelopmental health effects. Estuarine fish may be exposed to MeHg produced in
22
the terrestrial environment, benthic sediment and the marine water column but the
23
relative importance of each source is poorly understood. We measured stable isotopes of
24
mercury (δ202Hg, ∆199Hg, and ∆201Hg), carbon (δ13C), and nitrogen (δ15N) in fish with
25
contrasting habitats from a large subarctic coastal ecosystem to better understand MeHg
26
exposure sources. We identify two distinct food chains exposed to predominantly
27
freshwater and marine MeHg sources but do not find evidence for a benthic marine
28
MeHg signature. This is consistent with our previous research showing benthic sediment
29
is a net sink for MeHg in the estuary. Marine fish display lower and less variable ∆199Hg
30
values (0.78‰ to 1.77‰) than freshwater fish (0.72‰ to 3.14‰) and higher δ202Hg
31
values (marine: 0.1‰ to 0.57‰; freshwater: -0.76‰ to 0.15‰). We observe a shift in the
32
Hg isotopic composition of juvenile and adult rainbow smelt (Osmerus mordax) when
33
they transition between the freshwater and marine environment as their dominant
34
foraging territory. The Hg isotopic composition of Atlantic salmon (Salmo salar)
35
indicates they receive most of their MeHg from the marine environment despite a similar
36
or longer duration spent in freshwater regions. We conclude that stable Hg isotopes
37
effectively track fish MeHg exposure sources across different ontogenic stages.
38 39
ACS Paragon Plus Environment
2
Page 3 of 35
40 41
Environmental Science & Technology
Introduction Methylmercury (MeHg) is a bioaccumulative neurotoxicant produced from
42
divalent inorganic mercury (HgII) in aquatic ecosystems that adversely affects the health
43
of humans and wildlife.1-3 Benthic sediment, rivers, wetlands, and the marine water
44
column are all potential locations for MeHg formation and uptake into estuarine
45
foodwebs.4-8 The origin of environmental MeHg sources accumulated by fish affects their
46
responses to changes in HgII inputs. For example, there may be a substantial temporal lag
47
between atmospheric Hg inputs and MeHg levels in benthic sediment but a more rapid
48
response in surface seawater.9 Tracing the environmental origins of fish MeHg is
49
challenging due to their diverse foraging preferences and migration patterns. Stable Hg
50
isotopes show promise as a new empirical tracer for fish foraging behavior and
51
corresponding MeHg exposure sources.10-13 Here we use stable isotopes of Hg, carbon
52
and nitrogen in subarctic estuarine fish with diverse habitats to characterize their
53
predominant environmental MeHg exposure sources.
54
Bioaccumulation modeling for MeHg suggests dietary ingestion accounts for
55
>90% of the body burden in fish.14, 15 Stable isotopes of carbon (reported as δ13C) and
56
nitrogen (reported as δ15N) provide information on fish habitat type and trophic
57
position.16 However, δ15N and δ13C change more rapidly in fish than MeHg because their
58
half lives (5-173 days) are generally much shorter than persistent and bioaccumulative
59
contaminants like MeHg (~1-4 years).17-22 Understanding long-term foraging patterns is
60
particularly important for anticipating MeHg levels in species such as salmon and trout
61
that shift their diet and/or foraging territory over different ontogenic stages. Thus,
ACS Paragon Plus Environment
3
Environmental Science & Technology
62
naturally occuring Hg stable isotopes in fish can potentially provide complementary
63
information to δ15N and δ13C.10, 12, 23
64
All seven Hg stable isotopes undergo mass-dependent fractionation (MDF,
65
reported as δ202Hg) as the result of many physical, chemical, and biological processes.24-
66
30
67
predominantly during photochemical reactions, wherein the odd-mass-number isotopes
68
are either enriched or depleted in reaction products relative to the even-mass-number
69
isotopes.26, 29, 31 MIF is widely observed in aquatic organisms.26, 32, 33 However, no
70
substantial isotopic fractionation has been observed between MeHg in plankton and
71
higher trophic level fish.11, 34, 35 This allows MIF signatures to serve as a conservative
72
tracer for environmental MeHg sources accumulated by fish.
73
Page 4 of 35
Mass-independent fractionation (MIF, reported as ∆199Hg and ∆201Hg) occurs
Prior work suggests different sources and/or reaction pathways of Hg species
74
drive the variable Hg isotopic composition measured in terrestrial, coastal, and oceanic
75
biota. Tsui et al.13 found riverine and terrestrial food webs have distinct δ202Hg values,
76
which they attributed to different MeHg sources. In aquatic ecosystems, MIF is primarily
77
a function of photochemistry and varies with the extent of light penetration in different
78
foraging regions. Blum et al.12 showed pelagic marine fish foraging in deeper waters with
79
limited light penetration had lower ∆199Hg values relative to those feeding in surface
80
waters. Senn et al.10 showed the ∆199Hg values in pelagic fish were greater than coastal
81
fish from the Gulf of Mexico, presumably due to reduced water turbidity, enhanced
82
photochemistry and faster demethylation in the open ocean water column.
83 84
The main objective of this paper is to investigate whether stable Hg isotopes can elucidate the environmental origins of MeHg accumulated in fish from coastal
ACS Paragon Plus Environment
4
Page 5 of 35
Environmental Science & Technology
85
ecosystems with multiple potential exposure sources (rivers, benthic sediment, marine
86
water column). Prior work has focused on mid-latitude ecosystems and demonstrated
87
differences in Hg isotopic signatures of biota from two contrasting food webs (oceanic
88
vs. coastal, terrestrial vs. riverine)10, 13 but has not extended these findings to the diversity
89
of MeHg exposure sources and contrasting food-webs within a single coastal ecosystem.
90
Here we report new measurements of Hg, C, and N stable isotopes and tissue Hg and
91
MeHg burdens in fish and shellfish with contrasting benthic, benthopelagic, and pelagic
92
habitats from the region within and surrounding Lake Melville, a large subarctic fjord in
93
Labrador, Canada (Supporting Information (SI), Figure S1).
94 95
Methods
96
Sample collection
97
Table 1 lists the fish and shellfish species analyzed in this study and their
98
predominant habitats. We reviewed the literature on the indigenous composition of
99
benthic, benthopelagic, and pelagic marine and freshwater fish commonly found in the
100
Lake Melville region to identify species that represent each habitat type. Inuit community
101
members harvested 15 types of fish and shellfish (n=202) from the main freshwater
102
tributary to Lake Melville (lower Churchill River), the estuary (Lake Melville), and the
103
Labrador Sea between June and August of 2014 and 2015 (Table 1). Smaller fish were
104
captured by minnow traps and larger fish were obtained by rod or directly from
105
commercial fishing vessels in Lake Melville. Standard length and weight of each whole
106
fish was measured in the field or lab. Each fish was frozen after collection and shipped to
107
Harvard University for analysis.
ACS Paragon Plus Environment
5
Environmental Science & Technology
108
Page 6 of 35
In the laboratory, we collected subsamples of axial muscle tissue and freeze-dried
109
and homogenized samples prior to analysis for Hg, MeHg and stable isotopes. Minnow
110
species (stickleback, juvenile smelt, and chub) were freeze dried and whole fish samples
111
were homogenized using a blender because they were too small to fillet muscle tissue.
112
Surface sediment samples (0-3 cm) were taken from a box corer or stainless steel gravity
113
corer in Goose Bay, Groswater Bay, and the Churchill river (September 2012 and June
114
2013) (SI, Figure S1), as described in Schartup et al.5, and analyzed for Hg, MeHg and
115
Hg isotopes in this work.
116
Fish species age, habitat and diet classification
117
We harvested available fish scales and otoliths for age determination at NOAA’s
118
Northeast Fisheries Science Center in Woods Hole, MA. We determined the age of
119
Atlantic salmon and brook trout by analyzing their scales, following established
120
methods.36, 37 Otoliths were used to age other species following the “thin-section and
121
bake” technique.38 Stomach contents of all fish were dissected and examined prior to
122
chemical analysis. Qualitative diet composition from this analysis was combined with a
123
literature synthesis of the feeding ecology and life history for all fish to determine the a
124
priori habitat of all species (Table 1), which we compared to information from Hg
125
isotopes in this analysis.
126
Total Hg and MeHg analysis
127
We measured total Hg concentrations in all fish samples by thermal
128
decomposition, amalgamation, and atomic absorption spectrophotometry (EPA method
129
7473) using a Nippon MA-3000 Mercury Analyzer. At least one method blank and one
130
certified fish tissue reference material (CRM: TORT 3) was tested every 10 samples.
ACS Paragon Plus Environment
6
Page 7 of 35
131
Environmental Science & Technology
The average Hg recovery for the TORT 3 CRM was 100.7±2.2% (n=22).
132
We measured MeHg concentrations in all minnows, mussels and benthic fish
133
following a modified EPA 1630 method established in prior work.40-42 The tissue Hg
134
burden of predatory fish (cod, salmon, trout, pike, and whitefish) is widely assumed to be
135
close to 100% MeHg43 but lower trophic level organisms have more variable MeHg
136
fractions.44, 45 Samples were spiked with 1 ml enriched Me201Hg (2 ng/ml) then digested
137
with 5N HNO3 solution at 70 ºC overnight prior to MeHg analysis. Two CRMs (TORT 3
138
and DORM 4) were included in each digestion cycle. Acid was neutralized with 8 N
139
KOH, buffered with a 2M acetate buffer. Aqueous MeHg was ethylated using sodium
140
tetraethylborate (NaBEt4). Ethylated MeHg was purged onto a Tenax packed column and
141
separated by gas chromatography using a Tekran 2700 MeHg auto-analyzer coupled to a
142
Thermo iCAP-Q ICP-MS with Teflon tubing for MeHg detection.8, 46 We analyzed
143
ongoing precision and recovery (OPR) standards with different concentrations every 10
144
samples. Mean recovery was 105.5±5.7% (SD; n=8). The average recoveries for TORT 3
145
and DORM 4 were 92.4±5.0% (SD; n=8) and 99.7±9.8% (SD; n=8). Precision, estimated
146
by replicate analysis of CRMs and duplicate fish samples, was better than 7% and 8%,
147
respectively.
148
Carbon and nitrogen isotope analysis
149
Freeze dried and homogenized fish tissue samples were analyzed for stable
150
isotopes of carbon and nitrogen at Boston University’s Stable Isotope Laboratory and the
151
University of Hawaii at Manoa’s Isotope Biogeochemistry Laboratory.47 Isotopic values
152
are reported in conventional δ-notation relative to international standards (C: Peedee
153
Belemnite; N: atmospheric nitrogen). A secondary standard (glycine) was included to
ACS Paragon Plus Environment
7
Environmental Science & Technology
Page 8 of 35
154
ensure instrumental accuracy and precision in both labs. The standard deviation around
155
the expected values for the secondary standards was within 0.3% and 0.2% for δ13C and
156
δ15N, respectively. As lipids have more depleted δ13C values relative to other tissues,
157
aquatic samples with high lipid content (i.e., C:N ratio >3.5) are generally reported as a
158
normalized δ13C value by either lipid extraction or mathematical methods.48, 49 We used
159
the C:N ratio as a proxy for lipid content to estimate δ13C values of lipid-free muscle and
160
whole-body tissue, following the relationships derived by Logan et al.49
161
Hg stable isotope analysis
162
Approximately 0.1 to 0.2 g of freeze-dried and homogenized fish (n=135) and
163
sediment samples (n=7) were digested at 120°C for 6 hours using a 2 mL acid mixture
164
(HCl:HNO3 = 1:3, v:v) for stable Hg isotope analysis, following the protocol established
165
by others.33, 50 The digest solutions were diluted to 1 ng mL-1 for most samples. For
166
samples with low available mass or low tissue total Hg concentrations, digest solutions
167
were diluted to 0.4 – 0.5 ng mL-1. Samples were analysed using a Neptune Plus
168
multicollector inductively coupled plasma mass spectrometry (MC-ICP-MS) housed at
169
the Wisconsin State Laboratory of Hygiene.
170
The sample introduction system and analytical methods for stable Hg isotope
171
analysis follow previous studies.51, 52 Briefly, samples containing Hg were continuously
172
mixed and reduced with 3% SnCl2. The cold vapor generator was custom designed, as
173
described in Yin et al.52 Volatile Hg(0) was separated by a frosted glass phase separator
174
and introduced to the MC-ICP-MS with Argon gas. Instrumental mass bias was corrected
175
using an internal Tl standard (NIST SRM 997) and sample-standard bracketing. The Hg
176
concentrations and acid matrices of the bracketing standard (NIST SRM 3133) were
ACS Paragon Plus Environment
8
Page 9 of 35
Environmental Science & Technology
177
systematically matched to the neighbouring samples. Data nominations follow the
178
protocols suggested by Blum and Bergquist.53 MDF is expressed using δ202Hg notation
179
(Equation 1). MIF is expressed as the difference between the measured δxxxHg value, the
180
value predicted based on MDF and the δ202 Hg value (Equation 2-3).53
181
δ202 Hg = [202/198Hgsample /202/198 HgNIST3133 -1]×103 ‰
(1)
182
∆199Hg ≈ δ199Hg - (δ202Hg * 0.2520)
(2)
183
∆201Hg ≈ δ201Hg - (δ202Hg * 0.7520)
(3)
184
All CRMs (TORT-2, TORT-3, DORM-2, DORM-4, MESS-1) were prepared and
185
analysed in the same way as the samples. Total Hg in all solutions was monitored by
186
MC-ICP-MS using 202Hg signals, which yielded mean recoveries of 94.0% for fish
187
samples and CRMs. The UM-Almadén standard solution (0.5~1.0 ng mL-1, diluted in
188
10% aqua regia) was measured once every 10 samples. We found the overall average and
189
uncertainty of UM-Almadén and CRMs agreed well with prior work (SI, Section 1).13, 51,
190
54
191
Data analysis Differences in isotopic composition (δ13C, δ15N, ∆199Hg, ∆201Hg, and δ202 Hg)
192 193
among fish with different habitats and ecological zones were examined using two-way
194
ANOVA and Tukey's Honest Significant Difference (HSD) test. The ∆199Hg /∆201Hg
195
ratio was calculated from the slope of York regression which accounts for error in both
196
the dependent and independent variables.55 All statistical analyses were performed using
197
R.56
ACS Paragon Plus Environment
9
Environmental Science & Technology
198
Results and Discussion
199
Fish habitats and hypothesized MeHg sources
200
Page 10 of 35
Fish species collected as part of this work span predominantly freshwater and
201
marine habitats. Within these environments, we collected species that represent benthic,
202
benthopelagic, and pelagic habitats. Many species are exposed to a mixture of MeHg
203
sources (benthopelagic and anadromous fish). We hypothesized that the predominant
204
habitat type of each fish would dictate their environmental MeHg source and, in turn,
205
would be recorded by distinguishable differences in the Hg isotopic composition of their
206
tissues.
207
Table 1 shows the literature and δ13C derived habitats (SI, Figure S2) and prey of
208
different fish and shellfish included in this study. Lake chub, northern pike, and juvenile
209
rainbow smelt have low tolerance for high salinity conditions57-59 and thus primarily
210
forage in the freshwater environment. Potential MeHg sources in the freshwater
211
environment include benthic sediment in rivers, terrestrial soils, and/or wetlands.13, 60 In
212
brackish sections of the Churchill River (the main freshwater tributary to Lake Melville
213
estuary), some MeHg may also be transported inland from the estuarine river mouth
214
during seasonal intrusion of tidal waters.61, 62
215
Herring, capelin, and Atlantic cod are generally considered offshore marine
216
species, although they may migrate into estuaries for food and/or spawning (Table 1).59
217
In the marine environment, MeHg is produced both in the water column and in benthic
218
sediment.5, 6, 8, 63-65 Species with a strong link to the benthic environment across different
219
habitats include shorthorn sculpin (marine), flatfish (anadromous) and longnose sucker
220
(freshwater).59 All other species are anadromous or benthopelagic and likely to be
ACS Paragon Plus Environment
10
Page 11 of 35
Environmental Science & Technology
221
exposed to MeHg from multiple sources (Table 1).
222
Distinct freshwater and marine food chains
223
Distinct freshwater and marine food webs are evident from the Hg isotopic
224
composition of fish and shellfish species measured in this study (Figure 1a). Marine
225
species from this study (capelin, herring and cod) cluster together based on their Hg
226
isotope composition and are enriched in δ202Hg relative to freshwater species (chub, pike,
227
and juvenile smelt). Anadromous fish generally fall between the marine and freshwater
228
food webs (Figure 1a). We expected the Hg isotopic composition of benthic species
229
(shorthorn sculpin, flatfish, longnose sucker) to also be distinct. Instead, benthic species
230
cluster with pelagic and benthopelagic species across marine, anadromous, and
231
freshwater food webs. The lack of a distinct benthic signature suggests a common MeHg
232
source across benthic and pelagic environments (Table 1, Figure 1).
233
Further evidence for a common MeHg source is provided in Figure 2, which
234
shows that mean ∆199Hg and δ202Hg values in species within each foraging region
235
(freshwater, mixed, marine) do not differ significantly among benthic, benthopelagic and
236
pelagic habitats. By contrast, mean δ13C values in benthic species are significantly higher
237
(Tukey’s HSD, p