Trace element profiles of the sea anemone Anemonia viridis living nearby a natural CO₂ vent

Study sites

This study was conducted along the sub-littoral in Levante Bay, Vulcano Island (38°25′N, 14°57′E), part of the Aeolian Island chain, NE Sicily (Fig. 1A) in May 2012. Shallow-water CO₂ vents create a natural CO₂/pH gradient along the north-easterly side of the bay ranging from pH 6.05 to 8.29 at >350 m from the vent site (Johnson et al., 2011; Boatta et al., 2013). The vents also create a gradient of trace elements in the seawater and the sediments, where elevated trace elements are recorded up to 300 m away from the vent (Boatta et al., 2013; Kadar et al., 2012; Vizzini et al., 2013). Two sites were selected for sampling (Fig. 1A); Site 1 (control) was an ambient seawater reference station, located outside the vent area influence (>400 m), whereas site 2 (high pCO₂) was in proximity to the CO₂ vents (∼260 m) in accordance with previous studies (Johnson et al., 2011; Arnold et al., 2012; Suggett et al., 2012; Borell et al., 2014).

Sample collection

Physicochemical properties of the seawater

At both sites seawater pH (NBS scale) and temperature were measured several times a day for 4 days using a pH meter (YSI Professional Plus, Handheld Multiparameter Instrument). Water samples for total alkalinity (TA) analyses were collected from each site, cooled and stored in the dark until analysis. TA was quantified with a Metrohm 862 compact titrosampler (Cohen, 2011). The pCO₂ levels were calculated from pH_NBS measurements, TA and salinity (Johnson, 2012), using the program CO2 SYS (Pierrot, Lewis & Wallace, 2006), selecting the constants of Mehrbach et al. (1973). Light intensity was measured hourly for 3 consecutive days close to the seabed (1–2 m depth) with HOBO Pendant^® Temperature/Light data loggers (Onset, Pocasset) and then converted from lux to µmol quanta m⁻² s⁻¹ (Thimijan & Heins, 1983).

Sample preparation

Digestion and analysis

Samples were digested within the test tubes in the presence of 5–10 mL double distilled 9.8 N HNO₃. The closed lid test tubes were placed in acid cleaned glass beakers filled with DDW that were submerged in a water bath and heated to 80–90 °C for 1–7 days. Special care was taken to avoid contact between the outside water and the upper part of the test tube. Once the digested material lost its texture and color and appeared clear, the samples were cooled and supplemented with DDW. Samples with expected high trace element content were also diluted by 5–50 fold by transferring aliquots into new test tubes. Analysis of trace element content (and macro elements) was carried out using an Agilent 7500cx ICP-MS (Agilent Inc.). A set of calibration standards was added to the analysis together with appropriate blanks. An internal standard solution containing Sc, Re and Rh was introduced during the analyses to all standards and samples. The reported values are blank subtracted and averaged over duplicate analysis, with difference in % between duplicate analyses typically smaller than 10%. Results are given in ng trace element per mg dry weight (wt) of anemone.

Statistical analyses

Differences between pedal disc and tentacles in each site, and differences between sites for each body compartment were tested with one-way ANOVA and a multiple comparison test of means (Tukey). Prior to each analysis, all data were checked for normality using the Kolmogorov–Smirnov test and for homogeneity of variances using Cochran’s test. The effect of the site on trace element concentrations in tentacle Symbiodinium and anemone tissue was analyzed using the Kruskal–Wallis test. Post-hoc Mann–Whitney U-tests were run for separation of significant factors. Differences between factors were considered significant for a p value <0.05. Unless otherwise specified, mean values are presented ± SEM. All analyses were performed using SPSS version 20 (SPSS IBM, New York, USA).

Physicochemical properties of the seawater and suspended particles

Next to the vent the pH was significantly lowered and the pCO₂ (calculated) was strongly elevated compared to typical seawater conditions that were recorded at the control site (Fig. 1A). Seawater temperature (18.5–19.5 °C) and light availability at midday (599–642 µmol quanta m⁻² s⁻¹) were similar between sites (1–2 m depth), while TA was elevated at the high pCO₂ site (Fig. 1A; Table S1). The suspended particulate material collected in a net tow (> 100 µm), composed mostly of zooplankton, some of which A. viridis may prey upon, contained more trace elements (in weight per volume) next to the vent than at the control site (Tables 1 and S2). The degree of trace element enrichment in these particles was rather high with some trace elements enriched by 6–17 fold (Fe, Cu, As, Pb, cobalt (Co) and chromium (Cr)), while others by 2–5 fold (Mn, barium (Ba), nickel (Ni), strontium (Sr), Cd and Zn) (Table 1). Macro elements (sodium (Na), calcium (Ca), potassium (K), magnesium (Mg), boron (B) and sulfur (S)) did not show changes between sites, with the exception of phosphorus (P) (3 fold higher next to the vent).

Distribution of trace elements between the anemone pedal disc and tentacles

Our sampling scheme provided us with a paired data set of trace element concentrations (per unit of dry weight) in the pedal disc and the tentacles of each individual anemone. Plotting the concentrations of each of the 12 tested trace elements in the pedal disc vs. that in the tentacles (10 per site), we can visualize differences among sampling sites and body compartments, as well as variability among individual anemones (Figs. 2, 3 and S1). In Fig. 3 we present concentrations of trace elements which showed the most evident changes in bioaccumulation between sites, whereas Fig. S1 shows the remainder 4 trace elements with moderate changes. Averages of these data and statistical analysis of the differences between body compartments and sites are also reported in Table 2.

At the control site, most trace elements were more concentrated in the pedal disc than in the tentacles (p < 0.05; Table 2), with pedal disc to tentacles trace element ratios ranging between 1.6 and 3.6 (p < 0.05; Fig. 2). Few trace elements, including Sr, Mn and Ba were slightly enriched in the tentacles (Table 2), with pedal disc to tentacles trace element ratios of 0.8–0.9 (Fig. 2). At the high pCO₂ site, all trace elements except Sr, were more concentrated in the pedal disc than in the tentacles (Table 2), and the pedal disc to tentacles trace element ratios increased to 3.4–9.8 (Fig. 2). These elevated pedal disc to tentacle ratios reflect favorable accumulation of some trace elements in the pedal disc, while for others it results from depletion of trace elements from the tentacles compared to the control site.

Based on the findings above, we compare between sites in a compartment-specific manner. Such comparisons across all tested trace elements yielded three separate behaviors (Figs. 3A–3H, S1; Table 2): (1) Group I: Trace elements that are favorably accumulated in both body compartments at the high pCO₂ site (Fe, Pb, Cu, Co and Sr), (2) Group II: Trace elements that are accumulated in the pedal disc but are slightly depleted in the tentacles at the high pCO₂ site (Mn, Ba, Ni and Cr), and (3) Group III: Trace elements that are depleted in both body compartments at the high pCO₂ site (Zn, As and Cd). Statistical differences between sites are found for most trace elements mentioned above (p < 0.05), except Co and Group II elements in the tentacles, and Cd and As in the pedal disc (Table 2). The most significant change between sites was observed for Fe, which concentrations increased from 67 ± 7 to 1,021 ± 159 ng/mg dry wt in the pedal disc, and from 38 ± 2 to 195 ± 24 ng/mg dry wt in the tentacles (Table 2). Other metals in Group I, which included Pb, Cu and Co also showed substantial (3–9 fold) accumulation in both body compartments. In Group II, Mn and Ba accumulated 3–5 fold more in A. viridis pedal disc compared to control animals, while Ni and Cr showed very little change. However, in the tentacles, the concentrations of these trace elements were lower next to the vent than at the control site (Fig. S1; Table 2). This pattern of trace element depletion next to the vent was even more pronounced in Group III, comprising Zn, As and Cd, with lower concentrations in both body compartments at the high pCO₂ site than at the control (Figs. 3F–3H; Table 2).

Macro element concentrations (Na, Ca, K, Mg, B and S) did not change in both body compartments between sites. P, however, significantly increased in the pedal disc at the high pCO₂ site from 7,673 ± 179 to 8,279 ± 224 ng/mg dry wt (p < 0.05), whereas there was a significant decrease from 3,162 ± 168 to 2,201 ± 197 ng/mg dry wt in the tentacles (p < 0.05).

Distribution of trace elements among the algae and animal tissue of the tentacles

Data analysis

Having analyzed the algae and host fractions as well as the whole tentacle, we can probe for contamination or trace element loss in the separation procedure, by comparing the sum of the separated fractions to the whole tentacle (once weight normalized). Six of the trace elements (Ni, Cr, Co, Mn, Ba and Sr) were screened out due to contamination or detection problems. In three trace elements (Fe, Pb and Cu), that were mostly associated with the algae, the sum of the separated fractions balanced out adequately with the total. In the other trace elements (Cd, Zn and As), which were mostly associated with the host, the total greatly exceeded the sum of the separated fractions, indicative of significant metal loss. Hence, we chose not to include the measured host fraction. Instead, it was calculated from the difference between the total and algal fraction (Figs. 4A–4D; Table 3). Since this is a calculated measure, we did not perform statistical analysis on these results. In addition to Table 3, we present data for several trace elements in Fig. 4 demonstrating different partitioning between algae and animal tissue.

Table 3:

Trace element partitioning between algae and animal tissue in the tentacles of A. viridis collected from sites 1 (control) and 2 (high pCO2).

For each site we analyzed the trace element concentrations in total tentacle, algae and animal host tissue. Mean concentrations (ng/mg dry wt), standard error (between brackets) and number of replicates (in italics) are presented for each element. Host tissue values are calculated as the difference between total and algal trace element concentrations. The partitioning of trace elements between the algae (Alg) and the animal (Ani) fractions are presented as percentage of the tentacles’ total trace element concentration. Letters indicate significant differences in mean trace element concentrations (Mann–Whitney, p < 0.05).

							Fraction from total
	Control ng/mg dry wt			High pCO2 (ng/mg dry wt)			Control (%)		High pCO2 (%)
Element	Total	Algal	Animal	Total	Algal	Animal	Alg	Ani	Alg	Ani
Fe	39a	33a	5.7	184b	118b	66	85	15	64	36
	(5.1)	(6.3)	(10)	(34)	(27)	(46)
	4	4	4	5	5	5
Pb	0.0082a	0.005a	0.0031	0.022b	0.0068a	0.015	63	37	33	67
	(0.001)	(0.0022)	(0.0021)	(0.013)	(0.001)	(0.013)
	4	4	4	4	4	4
Cu	0.97a	0.45b	0.52	1.19a	0.7ab	0.48	46	54	60	40
	(0.17)	(0.1)	(0.16)	(0.22)	(0.1)	(0.2)
	5	5	5	5	5	5
Zn	19a	0.086b	19	13c	0.075b	13	0	100	1	99
	(2)	(0.015)	(2)	(2.6)	(0.018)	(2.6)
	5	5	5	5	5	5
As	19a	0.62b	18	11c	0.41d	10	3	97	4	96
	(2.3)	(0.052)	(2.2)	(2.4)	(0.051)	(2.4)
	5	5	5	5	5	5
Cd	0.013a	0.0012b	0.012	0.0037c	0.0016b	0.002	9	91	45	55
	(0.001)	(0.00037)	(0.0012)	(0.0011)	(0.0004)	(0.0014)
	5	5	5	5	5	5

DOI: 10.7717/peerj.538/table-3

The “trace element environment” nearby the vent

In order to analyze the trace element profiles in A. viridis growing at normal and high pCO₂ sites, it is first necessary to compare their “trace element environments”. Anemones can accumulate their trace elements from seawater, sediments, and suspended particulate material (including prey). The trace element accumulation pathways consist of: absorbance across the body surface, ingestion of particles, and/or active predation (Harland & Nganro, 1990; Harland, Bryan & Brown, 1990).

Previous reports have indicated that the seawater close to the CO₂ vent was enriched in Fe, Cu, Zn, Cr, and Mn (Table 1). Most of these metals were present as organic and/or inorganic complexes in the nano-particles size range (<100 nm fraction), while Fe was mostly associated with larger (>100 nm) colloids or particles (Kadar et al., 2012; Boatta et al., 2013). Fe, Mn and Cr show the most pronounced increase at the high pCO₂ site, while Sr, Ba and Zn remain unchanged (Table 1). Seawater concentrations of other trace elements included in our study, namely Cd and As, were not analyzed. Sediments collected from the high pCO₂ site were enriched in trace elements compared to the control site (Vizzini et al., 2013). Elements showing the highest enrichments were Ba, Ni, Zn and As (Table 1). Next to the vent, the low pH increases the solubility of some metals in the sediments, resulting in increased bioavailability and/or toxicity (Roberts et al., 2013). Similarly, the low pH and other elements released by the vents are likely to alter the speciation of trace elements in the seawater and subsequently their availability to the anemones (Byrne, Kump & Cantrell, 1988; Millero et al., 2009). While the study by Kadar et al. (2012) provided some information on the speciation of trace elements next to the vent, we currently cannot link the speciation of trace elements to their availability to anemones, which bioaccumulate trace elements through multiple pathways.

Here we added the analysis of trace elements in particles larger than 100 µm, which presumably represent zooplankton and other particles that the anemone can ingest or prey upon. All tested trace elements were substantially enriched in the particles collected next to the CO₂ vent compared to the control site (by a factor of 2–18; Table 1). All in all, the above mentioned data, both from the literature and this study, indicates that the “trace element environment” next the CO₂ vent, that is seawater, sediments and suspended particulate material, is different to the control site. Subsequently, regardless of the trace element uptake pathways in the anemones, their trace element content likely differs between sites.

Trace element accumulation patterns at the high pCO₂ site (highlighting Fe, Cu and Pb)

Of the 12 trace elements tested, the concentrations of five elements (Fe, Cu, Pb, Co and Sr) increased in both body compartments at the high pCO₂ site compared to the control site (Group I; Table 2), while four other trace elements (Mn, Ni, Ba and Cr) were elevated only in the pedal disc at the high pCO₂ site (Group II; Table 2). An increase in trace element concentrations of anemones (whole body) exposed to elevated loads of dissolved trace elements in the water was frequently observed in laboratory studies (Harland & Nganro, 1990; Harland, Bryan & Brown, 1990; Mitchelmore et al., 2003; Mitchelmore, Ringwood & Weis, 2003; Main, Ross & Bielmyer, 2010; Brock & Bielmyer, 2013). Additionally, differences in the overall metabolic activity of the anemones under elevated pCO₂ (Towanda & Thuesen, 2012; Suggett et al., 2012) may contribute to increased trace element bioaccumulation. This may be due to the higher rates of respiration in the more metabolically active anemone (Suggett et al., 2012), which directly affects feeding rate and thus the uptake of waterborne trace elements (Zamer, 1986). It is commonly accepted that the excess trace elements are kept in a “chemically safe” form through binding by proteins or other organic molecules and/or formation of metal granules, and in some cases are stored in a specialized organ (i.e., hepatopancreas or digestive gland in crustaceans, Depledge & Rainbow, 1990; Rainbow, 1990; Rainbow, 1993). Our measured trace element profiles in different body compartments and sites may serve to illustrate where trace elements are stored. Below we focus on three trace elements—Fe, Pb, and Cu, which were found in high concentrations in suspended particulate material, as well as seawater and sediments nearby the vent (Tables 1 and S2).

Trace element depletion patterns at the high pCO₂ site (highlighting Zn, Cd and As)

In contrast to the trace element enrichment in the suspended particulate material and sediments next to the vent (Table 1), the concentrations of some of the trace elements in the anemones were lower at the high pCO₂ site than in those of the control site. Three trace elements - Zn, Cd and As, showed the most pronounced depletion in both the tentacles and the pedal disc (Group III; Table 2). Four other trace elements - Mn, Ni, Ba and Cr, were depleted to a lesser extent and in the tentacles only (Group II; Table 2). Intrigued by this unexpected depletion, we tested how the trace elements in the tentacles partition between the algae and the host tissue. Since anemones were shown to expel their algae when subjected to trace element stress (Harland & Nganro, 1990; Mitchelmore et al., 2003), trace elements that occur predominately in the algae may be removed via this mechanism. The trace element depletion in the anemones may reflect their efficient elimination system, as was previously observed in laboratory studies during the recovery stages of anemones from high trace element loads (Mitchelmore et al., 2003; Brock & Bielmyer, 2013). It is also possible that changes in ion speciation due to the low pH diminish the bioavailability of these trace elements (Millero et al., 2009). Below, we explore the profiles of Group III elements among body compartments and sites and briefly speculate on their modes of regulation.

Environmental implications

The strong gradients in trace element concentrations under OA-like conditions in the relatively small spatial scale of Levante Bay enabled us to detect distinct patterns of trace element accumulation and depletion in A. viridis from contrasting environments. These, in turn, may help predict the ability of anemones to adapt to such conditions, as coastal marine invertebrates will have to cope with OA together with other anthropogenic stressors such as chronic environmental pollution in the future (Lewis, Clemow & Holt, 2013). It was not within the scope of our study to discern whether trace element uptake is biologically-driven or based on physicochemical changes of the metals and metalloids in the water column. Our results, however, demonstrate the potential resilience of A. viridis to acidification combined with trace element pollution, and moreover, they seem to thrive in this environment (Suggett et al., 2012). Despite potentially harmful levels of trace elements within the region of our study site (260 m from the vent), we found no apparent toxicity to the anemones although there was clear bioaccumulation. The conditions of high pCO₂/low pH and elevated trace elements may pose some metabolic costs to the anemones. Animals subjected to high pCO₂/low pH are required to compensate for acid–base imbalance in intra- and extracellular spaces (Fabry et al., 2008). Anemones subjected to high trace element fluxes are forced to invest in metal regulation and compartmentalization (Harland & Nganro, 1990; Mitchelmore et al., 2003; Brock & Bielmyer, 2013). On the other hand, high pCO₂ conditions are metabolically favorable for anemones due to increased photosynthesis (Towanda & Thuesen, 2012). Although we are unable to determine these metabolic costs, we assume gain outweighs costs in this environment based on their prevalence and large dimensions (Suggett et al., 2012). We think these are valuable considerations that merit further study for predicting anemones’ performance under increasing trace element loads and low pH. To our knowledge, this is the first study performed in situ in an effort to better understand and describe the potential impact of trace element pollutants under OA conditions on an ecologically important coastal invertebrate. This study is unique in terms of long-term chronic exposure to multiple trace elements and high pCO₂/low pH conditions, and as such bears significant implications to current predictions of OA impacts on coastal marine communities.