Chemosphere 144 (2016) 2368–2376
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Comparative studies on the effects of seawater acidification caused by CO2 and HCl enrichment on physiological changes in Mytilus edulis Tianli Sun, Xuexi Tang, Bin Zhou, You Wang∗ Department of Marine Ecology, College of Marine Life Science, Ocean University of China, Qingdao, Shandong Province, China
h i g h l i g h t s • • • • •
We compared HCl and CO2 -induced acidification on physical changes in M. edulis. Both methods had effects on growth, metabolism and carbon sink in M. edulis. CO2 had greater effects on mortality than HCl. CO2 had less effects on metabolism and carbon sink than HCl. IBR is a potential biological monitoring method in evaluating acidification.
a r t i c l e
i n f o
Article history: Received 29 May 2015 Received in revised form 26 October 2015 Accepted 27 October 2015 Available online 22 November 2015 Handling editor: James M. Lazorchak Keywords: Seawater acidification HCl addition CO2 enrichment Physiological changes Mytilus edulis Integrated biomarker response (IBR)
a b s t r a c t The present medium term (21 d) study was performed to evaluate the effects of HCl or CO2 -induced acidified seawater (pH 7.7, 7.1 or 6.5; control: pH 8.1) on the physiological responses of the blue mussel, Mytilus edulis, at different levels of biological organization. The results demonstrate that: (1) either HCl or CO2 enrichment had significant impacts on physiological changes in M. edulis: the mortality increased while condition index (CI) decreased steadily as the pH decreased, those indexes indicate the metabolic activities (e.g. filtering rate, oxygen consumption rate, etc.) underwent similar changes; moreover, the decrease of calcification rate and carbonic anhydrase activity indicate that the carbon sink ability of the mussels was significantly affected. We hypothesize that acidification induced intracellular energy crisis and a decrease in enzyme activities could be a potential explanation for our findings. (2) Comparatively, CO2 enrichment had more severe effects on mortality but caused less stress to the metabolic and carbon sink indexes than HCl adjustment at the same pH level. Apoptosis caused by the ‘intracellular acidification’ in the CO2 group and difference in cytoplasmic Ca2+ concentration between two groups are suggested to be responsible for these results. (3) An integrated biomarker response (IBR) was set up on the basis of the estimated indexes; it was determined that the IBR decreased steadily with the decrease of pH, and a positive relationship was observed between them, inferring that the IBR might be a potential biological monitoring method in evaluating the effects of seawater acidification. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction Ocean acts as a natural ’carbon sink’ and has absorbed up to 50% of the total accumulation of anthropogenic carbon dioxide (CO2 ) that was emitted into the environment (Sabine et al., 2004), which altered the seawater chemistry and led to a series of negative effects (Fabry et al., 2008). One of the most serious manifestations of the changed seawater chemistry is ocean
∗
Corresponding author. 5 Yushan Road, Ocean University of China, China. E-mail addresses: [email protected] (T. Sun), [email protected] (X. Tang), [email protected] (B. Zhou), [email protected] (Y. Wang). http://dx.doi.org/10.1016/j.chemosphere.2015.10.117 0045-6535/© 2015 Elsevier Ltd. All rights reserved.
acidification (OA). Compared with pre-industrial times (pH 8.2), seawater pH has fallen by 0.1 pH units, indicating a 30% increase in the concentration of H+ ions. As a result, it is predicted that the pH of surface seawater would fall by up to 0.4 U before 2100 and a reduction of 0.8 U would occur by 2300 (Caldeira and Wickett, 2003; Orr et al., 2005). Injection of CO2 into the seabed (e.g., depleted gas and oil fields) is an effective way of carbon sequestration at present (Qu et al., 2012), whereas potential CO2 leakage risk is still remained (Hawkins, 2004). Widdicombe and Needham (2007) reported an extreme seawater pH value as low as 5.6 at a natural seep site located at the bay of Naples, Italy. Severe decrease in the seawater pH may re-
T. Sun et al. / Chemosphere 144 (2016) 2368–2376
sult in biological, physiological and evolutionary consequences for all forms of marine biota at many different organizational levels (Bibby et al., 2008). Numerous studies investigating the effects of seawater acidification on marine organisms have been conducted in recent decades, but most of them tended to create acidified seawater by adding mineral acids (such as HCl or H2 SO4 ) to mimic the disposal of CO2 in marine ecosystem (Auerbach et al., 1997; Yamada and Ikeda, 1999). As research continued, it has become evidence that the toxicity of CO2 was largely underestimated (Riebesell et al., 2000; Ishimatsu et al., 2004; Kikkawa et al., 2004; Bibby et al., 2008). In fact, compared with H+ ion, uncharged CO2 molecules are readily permeable to biological membranes (Vandenberg et al., 1994). Therefore, CO2 -induced seawater acidification is considered to have more complex effects on marine organisms than that is induced by HCl or H2 SO4 . Comparisons between different acidified methods have been made in fish and zooplankton (Kikkawa et al., 2004; Zhang et al., 2011), but no attention was paid on bivalves so far. Bivalves have been used as sentinel organisms to monitor the effects of environmental contaminants on aquatic ecosystems. As an important calcifying organism that forms biogenic calcium carbonate deposits through the absorption of Ca2+ and HCO3 − from the seawater, bivalves are particularly vulnerable to seawater acidification (Nagarajan et al., 2006; Bibby et al., 2008). Consequently, we conducted a 21 d-medium term experiment to evaluate the effects of exposure to two kinds of acidified seawater on key aspects of the physiological response of Mytilus edulis. Mussels were exposed to pH levels mimicking near-future ocean acidification (pH 7.7) or CO2 sequestration leak scenarios (pH 7.1 or 6.5), using two acidifying methods of HCl addition and CO2 enrichment. The study aimed at (1) elucidating the effects of acidified seawater on mussels in terms of different biological organizations and (2) comparing the negative outcomes induced by different acidifying methods. A multi-assay approach was used to obtain an overall view of physiological responses in M. edulis under different kinds of acidification stresses. This study for the first time described the effect of seawater acidification on bivalves by systematically comparing CO2 enrichment and HCl adjustment induced seawater acidification, further we provided more evidence on evaluating potential risks of acidification in the coastal ecosystem. 2. Materials and methods
2369
and repeated once again just when the first one was ended. The combination of the repetition was used for the result analysis and statistic evaluation. 2.2. Acidified seawater preparation Two different acidifying methods, CO2 enrichment and HCl adjustment, were applied in the present study. Test seawater was prepared as follows: (1) CO2 group: seawater was bubbled with pure CO2 gas (99.9%); (2) HCl group: seawater was acidified by adding 1 M HCl. Seawater pHNBS was measured and adjusted by pH controllers (pH/ORP-101, HOTEC, Taiwan; pH fluctuations were controlled within 0.08 units), to maintain seawater at one of seven nominal pH levels: 8.1 (ambient seawater pH; pCO2 ≈ 390 ppm), 7.7 (2100, Orr et al., 2005; pCO2 ≈ 1500 ppm), 7.1 (CCS leak, Berge et al., 2006; pCO2 ≈ 5000 ppm), 6.5 (CCS leak; Widdicombe and Needham, 2007; pCO2 ≈ 18,000 ppm). The pHNBS values and salinity were measured daily using a calibrated pH meter (S210k, METTLER TOLEDO, USA) and a handheld salinometer (WY028Y, HUARUI, CHN). Total alkalinity (AT) was measured weekly using an open-cell potentiometric titration technique. All other carbonate system variables were calculated using CO2 SYS (Pierrot et al., 2006). Carbonate chemistry data for these tanks for the experimental exposure are presented in Table 1. 2.3. Experimental design Indexes indicating the physiological changes were from the perspectives of growth, metabolism and carbon sink ability. 2.3.1. From the perspective of growth Individuals suspected to be dead were observed during the experiment. A dissecting needle was used to touch the mantles of the suspected dead mussels, and those that did not respond to the touch were identified as dead. The mortality at the end of the experiment was calculated as:
Motality (% ) = Dead ind/Total ind × 100% The condition index (CI) was measured at the beginning and the end of the experiment. 6 individuals in each tank were randomly picked out, and their soft tissues and shells were sampled. They were dried in a muffle furnace to a constant weight, and the CI was calculated as Walne and Mann (1975):
CI = Dt /Ds × 1000 2.1. M. edulis acclimation and maintenance Blue mussels, M. edulis, were obtained in Laoshan Bay, Qingdao, Shandong Province, China (36°15 N, 120°40 E). The epibionts on the shells were cleaned carefully using metal scrapers and the kept them undisturbed for a 7 day acclimation period prior to the experiment. During the experiment, 30 randomly selected mussels (fitness indexes: 39.82 ± 0.22 mm shell length, 6.05 ± 0.20 g wet weight) were placed in 7 experimental tanks (Vol. = 8 L; 210 mussels in total) which were divided into 3 groups: the HCl adjustment group (3 tanks), the CO2 enrichment group (3 tanks) and the control (1 tank). Natural seawater (pH 8.0 ± 0.1, salinity 31 ± 1.0) was filtered on a 0.45-μm pore size membrane and completely renewed every day. The temperature was maintained at 15 ± 1 °C and illumination was set on a 12 h light- 12 h dark cycle. According to the pre-test results, 200 mg (dry mass·tank−1 day−1 ) food algae, Platymonas helgolandica (Chlorophyta), was diluted in seawater and supplied to the holding tanks by gravity feed (approximate 1 mL min−1 ) and the final density in each tank was 1.5 × 105 cells min−1 . The mortality of M. edulis was measured in similar 7 tanks at the same time. The experiment lasted for 21 d
where ‘Dt ’ and ‘Ds ’ denote the dry weight (g) of the fresh tissues and shells, respectively. 2.3.2. From the perspective of metabolism 3 mussels were randomly taken out from each tank and placed into a 500 mL triangular flask respectively. All flasks were sealed by plastic films to avoid gas exchange. All the experimental flasks were acclimated under the experimental conditions for 20 min before the experiment. 3 replicates were setting up and the experiments were repeated twice. For determining the filtering rate (FR, mL Wwg−1 h−1 ), mussels were fed with the bait microalga of P. helgolandica at a final density of 1.5 × 105 cells mL−1 , and the algal density decreased steadily due to the filtering activity of the mussels. Three water samples (2 ml) from each aquarium were taken out at 30 min after algal feed and the algal density changes were measured using a hemocytometer. The filtration rate was calculated according to Quayle (1948) formula:
FR = V/(w · t ) · loge (c0 /ct )
2370
T. Sun et al. / Chemosphere 144 (2016) 2368–2376
Table 1 Setting seawater pH, measured pH, total alkalinity (TA, μM) and the concentration of HCO3 − , CO3 2− , H2 CO3 , pCO2 (ppm) in seawater in each group both HCl adjustment and CO2 enrichment (Mean ± SD). pH (setting)
pH (measured)
HCl group 8.1 7.7 7.1 6.5 CO2 group 8.1 7.7 7.1 6.5
HCO3 −
TA
0.07 0.08 0.02 0.08
1910 1686 1292 570
± ± ± ±
48 82 89 54
1625 1569 1231 561
± ± ± ±
79 118 121 64
109 41 10 2
± ± ± ±
52 13 2 1
16 35 103 156
± ± ± ±
1 5 39 35
439 1266 3186 4737
± ± ± ±
40 88 256 384
8.03 7.64 7.11 6.55
± ± ± ±
0.07 0.07 0.01 0.05
1910 1918 1910 1918
± ± ± ±
48 61 64 53
1625 1779 1835 1907
± ± ± ±
79 89 74 16
109 54 16 4
± ± ± ±
52 22 4 2
16 43 152 565
± ± ± ±
1 9 31 42
439 1352 4718 17,539
± ± ± ±
40 109 322 782
where ‘D0 ’ and ‘Dt ’ denote the dissolved oxygen (ppm) in each flask at the beginning and at time ‘t’ after exposure, respectively. The ammonia-N excretion rate (RN , μmol Wwg−1 h−1 ) was determined by the changes between the initial and the final concentrations of NH4 + -N in mussel individual at 2 h after exposure by using the sodium bromate oxidation method, and the ammonia-N excretion rate was calculated according to the following formula:
RN = {(Nt − N0 ) × V × 1000}/(w × t × 14 ) where ‘N0 ’ and ‘Nt ’ are the concentrations of NH4 + -N (ppm) in each flask at the beginning and at time ‘t’ after exposure, respectively. The O:N ration was calculated as Widdows (1978):
O : N = RO /RN The respiration rate (RC , μmol Wwg−1 h−1 ) was obtained according to the changes in the dissolved inorganic carbon (DIC) between the beginning and at 2 h after exposure. DIC was estimated using a TOC meter (TOC-vcpn, SHIMADZU, JPN) and Rc was calculated as Zhang et al. (2011):
Rc = {(DIC + DICG ) × V }/(w × t ) where ‘DIC’ is the DIC variation (μM); ‘DICG ’ could be calculated through following formula:
DICG = TAi − TA f /2 where ‘TAi ’ and ‘TAf ’ are the total alkalinity (μM) at the beginning and at time ‘t’ after exposure. 2.3.3. From the perspective of carbon sink ability Calcification was estimated by using an ‘alkalinity anomaly technique’ (Rodoni, 1993). The calcification rate (G, μmol Wwg−1 h−1 ) was calculated according to the following equation:
pCO2
± ± ± ±
Ro = {(D0 − Dt ) × V × 1000}/(w × t × 16 )
G=
H2 CO3
8.03 7.61 7.11 6.58
where ‘V’ is the total experimental volume (mL); ‘w’ is the wet weight (g) per mussel; ‘t’ is the duration (h) of the experiment; ‘c0 ’ and ‘ct ’ denote the cell densities in each flask at the beginning and at time ‘t’ after exposure, respectively. The determination of oxygen consumption rate (Ro, μmol Wwg−1 h−1 ) was performed in the same experimental system only except that the experiment lasted for 2 h, and the dissolved oxygen (DO) was measured by using a dissolved oxygen meter (DO-200, YSI, USA). Ro was calculated as follows:
CO3 2−
TAi − TA f × V /(2 × w × t )
The CA activity was measured according to the improved method of pH (Henry, 1991). The gills of the mussels in the
treated groups were sampled and CA activity units (U) was calculated according to Henry and Watts (2001):
U = t0 /tenz − 1 where ‘t0 ’ and ‘tenz ’ are response time (s) of the blanks and each pH level. CA activity was expressed as ‘μmol CO2 ·mg protein−1 ·min−1 ’ according to Wheatly and Henry (1987). The soluble protein was analyzed using the Coomassie brilliant blue G-250 method using bovine serum albumin (BSA) as a standard. 2.4. Statistics The mean values and standard deviations were calculated from the different replicates (n = 6), and figures were generated using Sigmaplot 12.5. Differences between the treated groups and the control were analyzed by a paired t-test in SPSS 22.0, with the significant difference was set at p < 0.05. General linear model (GLM) univariate ANOVA was used to investigate the effects of pH and time on each parameter. Where GLM analysis identified a significant effect of Time but not pH, the data from the different pH treatments were pooled before ANOVA analysis. Where the GLM analysis identified a significant effect of Time, the effect of pH at each sampling interval was examined separately. Hence the comparison between HCl and CO2 could be classified into a concentration-dependent manner or a time-dependent manner. 3. Results 3.1. Effects of acidified seawater on growth of M. edulis The control mussels remained in a good condition and responded to stimuli during the exposure period, and their shell length and wet weight were increased to 40.16 mm ± 0.37 mm and 6.12 ± 0.49 g respectively at the end of the experiment. Dead individuals were observed in both treated groups during the experiment, and the mortality increased steadily with the decline of pH. The mortalities were markedly higher in the CO2 group than the HCl group at the same levels of pH vales (Table 2). The gaps between two groups became wider from 6.7% to 16.6% with the pH decrease. The results suggested that CO2 enrichment might be more toxic than HCl adjustment for the mortality at the same pH level. The decrease of CI values suggested growth inhibitions when mussels exposed to acidifying seawater. However, the inhibition varied with the acidified methods. The CI values decreased significantly in the HCl group and were maintained at low levels. The largest drop (−22.2%) was observed at pH 7.7 (p < 0.05). In the CO2 group, a significant decrease (−44.4%) was also observed at pH 7.7 (p < 0.01), however, little changes of CI values at pH 7.1
T. Sun et al. / Chemosphere 144 (2016) 2368–2376 Table 2 The mortality (%) and condition indexes of M. edulis after 21 d exposure in HCl adjustment groups and CO2 enrichment groups (mean ± SD). Mortality (%)
8.1 7.7 7.1 6.5
CI
HCl group
CO2 group
HCl group
0 3.3 3.3 6.7
0 10 13.3 23.3
36.1 27.0 28.1 28.2
± ± ± ±
CO2 group
6.6 3.6∗ 12.7∗ 6.8∗
36.1 19.3 35.9 31.0
± ± ± ±
6.6 1.5∗ ∗ 1.9 9.8
Note: ‘∗ ’ means significant difference (p < 0.05); ‘∗ ∗ ’ means extremely significant difference (p < 0.01).
and 6.5 were found comparing with the control group (p > 0.05) (Table 2). 3.2. Effects of acidified seawater on the metabolic activities of M. edulis There were significant effects of both pH and Time on of M. edulis significantly in either HCl or CO2 groups, (Table 3, Fig. 1(a)). Filtering rates at pH 7.1 and 6.5 in both treatments decreased with decreasing pH since 7 d. The lowest levels were 39.6% of the control in the HCl group on 7 d and 18.4% in the CO2 group on 21 d at pH 7.1 (p < 0.01), 11.6% and 11.3% on 7 d at pH 6.5 (p < 0.01). Filtering rates at pH 7.7 in the CO2 group declined on 14 d and 21 d, but no significant effect in the HCl group was observed at pH 7.7. Compared to the HCl group, mussels in the CO2 group showed higher filtering rates on 7 d but lower rates on 14 d and 21 d. Consequently, the inhibiting ability of CO2 enrichment was stronger than HCl adjustment as the increase of exposure time. Obvious pH-dependent manner and time-dependent manner were found in both acidified groups on the oxygen consumption
Table 3 General linear model (GLM) univariate ANOVA for CI, filtering rates, oxygen consumption rates, ammonia-N excretion rates, O:N ration, carbonic anhydrase activity, calcification rates and respiratory rates, against pH and time. Indexes
HCl group F
CI Time 2.487 pH 6.997 Filtering rates Time 7.608 pH 70.962 Oxygen consumption rates Time 48.406 pH 72.232 Ammonia-N excretion rates Time 25.070 pH 30.883 O:N ration Time 24.388 pH 4.827 Respiratory rates Time 5.656 pH 891.769 Calcification rates Time 23.477 pH 54.236 Carbonic anhydrase activity Time 4.096 pH 9.208 Note: ‘ns’ means not significant.
2371
rate of M. edulis (Table 3). Oxygen consumption rate levels declined with decreasing pH since 7 d and the minimum value of in each pH level was found on 21 d (Fig. 1(b)). Although mussels in reduced pH treatments in the CO2 group increased their oxygen consumption rate on 14 d, this recovery did not continue to 21 d. Similar results were indicated by the filtering rates, i.e. oxygen consumption rates of M. edulis were affected by seawater acidification, and the inhibiting ability of CO2 enhanced with the increase of exposure time, compared to HCl addition. Both acidified methods on pH and time were confirmed to significantly influence the ammonia-N excretion rate of M. edulis (Table 3), in which mussels in the CO2 group showed a stronger inhibition than in the HCl group (Fig. 1(c)). The highest inhibitions in the CO2 group were appeared on 21 d, with the ammonia-N excretion rate at pH 7.7, 7.1, 6.5 showed a decrease of 73.4%, 65.9% and 50.0% comparing with the control (p < 0.01), respectively. The highest inhibitions in the HCl group appeared on 14 d, with the ammonia-N excretion rate at pH 7.1 and 6.5 displayed a decrease of 75.2% and 45.5% comparing with the control (p < 0.01), respectively. No significant effect of pH 7.7 in the HCl group was detected on the ammonia-N excretion rate. Time, but not pH, affected the O:N ratios in both methods of acidification (Table 3). Overall, there was an decrease in the O:N ratio with time from 30.5 to 22.6 in the HCl group, but there existed a ‘low-promoting and high-repressing’ manner in the CO2 group, in which the O:N ratios increased from 30.5 to 34.3 on 7 d and 31.9 on 14 d, respectively, then decreased to 23.7 on 21 d (Table 4). An O:N ratio in approximately 20 indicates that amino acid catabolism dominates the M. edulis aerobic metabolism. Pure protein metabolism could be expected at a ratio of 3–16, and balanced lipid and protein degradation occurs at O:N = 50–60 (Mayzaud and Conover, 1988). These results showed that acidification might lead to an increase in protein consumption and eventually changes in the metabolic type of M. edulis. Visible enhancement in the respiratory rate were found between the HCl and CO2 groups at the same pH level and time (Fig. 1(d)). In the HCl group, the respiratory rate showed a peak value at pH 7.1 on 21 d (380.2% of the control), while the peak value moved to a lower pH value (pH 6.5 on 14 d, 283.4% of the control) in the CO2 group. Therefore, the respiratory rate of M. edulis was more sensitive to the stimulation of HCl than CO2 .
CO2 group p
F
p
ns
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Comparative studies on the effects of seawater acidification caused by CO2 and HCl enrichment on physiological changes in Mytilus edulis Tianli Sun, Xuexi Tang, Bin Zhou, You Wang∗ Department of Marine Ecology, College of Marine Life Science, Ocean University of China, Qingdao, Shandong Province, China
h i g h l i g h t s • • • • •
We compared HCl and CO2 -induced acidification on physical changes in M. edulis. Both methods had effects on growth, metabolism and carbon sink in M. edulis. CO2 had greater effects on mortality than HCl. CO2 had less effects on metabolism and carbon sink than HCl. IBR is a potential biological monitoring method in evaluating acidification.
a r t i c l e
i n f o
Article history: Received 29 May 2015 Received in revised form 26 October 2015 Accepted 27 October 2015 Available online 22 November 2015 Handling editor: James M. Lazorchak Keywords: Seawater acidification HCl addition CO2 enrichment Physiological changes Mytilus edulis Integrated biomarker response (IBR)
a b s t r a c t The present medium term (21 d) study was performed to evaluate the effects of HCl or CO2 -induced acidified seawater (pH 7.7, 7.1 or 6.5; control: pH 8.1) on the physiological responses of the blue mussel, Mytilus edulis, at different levels of biological organization. The results demonstrate that: (1) either HCl or CO2 enrichment had significant impacts on physiological changes in M. edulis: the mortality increased while condition index (CI) decreased steadily as the pH decreased, those indexes indicate the metabolic activities (e.g. filtering rate, oxygen consumption rate, etc.) underwent similar changes; moreover, the decrease of calcification rate and carbonic anhydrase activity indicate that the carbon sink ability of the mussels was significantly affected. We hypothesize that acidification induced intracellular energy crisis and a decrease in enzyme activities could be a potential explanation for our findings. (2) Comparatively, CO2 enrichment had more severe effects on mortality but caused less stress to the metabolic and carbon sink indexes than HCl adjustment at the same pH level. Apoptosis caused by the ‘intracellular acidification’ in the CO2 group and difference in cytoplasmic Ca2+ concentration between two groups are suggested to be responsible for these results. (3) An integrated biomarker response (IBR) was set up on the basis of the estimated indexes; it was determined that the IBR decreased steadily with the decrease of pH, and a positive relationship was observed between them, inferring that the IBR might be a potential biological monitoring method in evaluating the effects of seawater acidification. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction Ocean acts as a natural ’carbon sink’ and has absorbed up to 50% of the total accumulation of anthropogenic carbon dioxide (CO2 ) that was emitted into the environment (Sabine et al., 2004), which altered the seawater chemistry and led to a series of negative effects (Fabry et al., 2008). One of the most serious manifestations of the changed seawater chemistry is ocean
∗
Corresponding author. 5 Yushan Road, Ocean University of China, China. E-mail addresses: [email protected] (T. Sun), [email protected] (X. Tang), [email protected] (B. Zhou), [email protected] (Y. Wang). http://dx.doi.org/10.1016/j.chemosphere.2015.10.117 0045-6535/© 2015 Elsevier Ltd. All rights reserved.
acidification (OA). Compared with pre-industrial times (pH 8.2), seawater pH has fallen by 0.1 pH units, indicating a 30% increase in the concentration of H+ ions. As a result, it is predicted that the pH of surface seawater would fall by up to 0.4 U before 2100 and a reduction of 0.8 U would occur by 2300 (Caldeira and Wickett, 2003; Orr et al., 2005). Injection of CO2 into the seabed (e.g., depleted gas and oil fields) is an effective way of carbon sequestration at present (Qu et al., 2012), whereas potential CO2 leakage risk is still remained (Hawkins, 2004). Widdicombe and Needham (2007) reported an extreme seawater pH value as low as 5.6 at a natural seep site located at the bay of Naples, Italy. Severe decrease in the seawater pH may re-
T. Sun et al. / Chemosphere 144 (2016) 2368–2376
sult in biological, physiological and evolutionary consequences for all forms of marine biota at many different organizational levels (Bibby et al., 2008). Numerous studies investigating the effects of seawater acidification on marine organisms have been conducted in recent decades, but most of them tended to create acidified seawater by adding mineral acids (such as HCl or H2 SO4 ) to mimic the disposal of CO2 in marine ecosystem (Auerbach et al., 1997; Yamada and Ikeda, 1999). As research continued, it has become evidence that the toxicity of CO2 was largely underestimated (Riebesell et al., 2000; Ishimatsu et al., 2004; Kikkawa et al., 2004; Bibby et al., 2008). In fact, compared with H+ ion, uncharged CO2 molecules are readily permeable to biological membranes (Vandenberg et al., 1994). Therefore, CO2 -induced seawater acidification is considered to have more complex effects on marine organisms than that is induced by HCl or H2 SO4 . Comparisons between different acidified methods have been made in fish and zooplankton (Kikkawa et al., 2004; Zhang et al., 2011), but no attention was paid on bivalves so far. Bivalves have been used as sentinel organisms to monitor the effects of environmental contaminants on aquatic ecosystems. As an important calcifying organism that forms biogenic calcium carbonate deposits through the absorption of Ca2+ and HCO3 − from the seawater, bivalves are particularly vulnerable to seawater acidification (Nagarajan et al., 2006; Bibby et al., 2008). Consequently, we conducted a 21 d-medium term experiment to evaluate the effects of exposure to two kinds of acidified seawater on key aspects of the physiological response of Mytilus edulis. Mussels were exposed to pH levels mimicking near-future ocean acidification (pH 7.7) or CO2 sequestration leak scenarios (pH 7.1 or 6.5), using two acidifying methods of HCl addition and CO2 enrichment. The study aimed at (1) elucidating the effects of acidified seawater on mussels in terms of different biological organizations and (2) comparing the negative outcomes induced by different acidifying methods. A multi-assay approach was used to obtain an overall view of physiological responses in M. edulis under different kinds of acidification stresses. This study for the first time described the effect of seawater acidification on bivalves by systematically comparing CO2 enrichment and HCl adjustment induced seawater acidification, further we provided more evidence on evaluating potential risks of acidification in the coastal ecosystem. 2. Materials and methods
2369
and repeated once again just when the first one was ended. The combination of the repetition was used for the result analysis and statistic evaluation. 2.2. Acidified seawater preparation Two different acidifying methods, CO2 enrichment and HCl adjustment, were applied in the present study. Test seawater was prepared as follows: (1) CO2 group: seawater was bubbled with pure CO2 gas (99.9%); (2) HCl group: seawater was acidified by adding 1 M HCl. Seawater pHNBS was measured and adjusted by pH controllers (pH/ORP-101, HOTEC, Taiwan; pH fluctuations were controlled within 0.08 units), to maintain seawater at one of seven nominal pH levels: 8.1 (ambient seawater pH; pCO2 ≈ 390 ppm), 7.7 (2100, Orr et al., 2005; pCO2 ≈ 1500 ppm), 7.1 (CCS leak, Berge et al., 2006; pCO2 ≈ 5000 ppm), 6.5 (CCS leak; Widdicombe and Needham, 2007; pCO2 ≈ 18,000 ppm). The pHNBS values and salinity were measured daily using a calibrated pH meter (S210k, METTLER TOLEDO, USA) and a handheld salinometer (WY028Y, HUARUI, CHN). Total alkalinity (AT) was measured weekly using an open-cell potentiometric titration technique. All other carbonate system variables were calculated using CO2 SYS (Pierrot et al., 2006). Carbonate chemistry data for these tanks for the experimental exposure are presented in Table 1. 2.3. Experimental design Indexes indicating the physiological changes were from the perspectives of growth, metabolism and carbon sink ability. 2.3.1. From the perspective of growth Individuals suspected to be dead were observed during the experiment. A dissecting needle was used to touch the mantles of the suspected dead mussels, and those that did not respond to the touch were identified as dead. The mortality at the end of the experiment was calculated as:
Motality (% ) = Dead ind/Total ind × 100% The condition index (CI) was measured at the beginning and the end of the experiment. 6 individuals in each tank were randomly picked out, and their soft tissues and shells were sampled. They were dried in a muffle furnace to a constant weight, and the CI was calculated as Walne and Mann (1975):
CI = Dt /Ds × 1000 2.1. M. edulis acclimation and maintenance Blue mussels, M. edulis, were obtained in Laoshan Bay, Qingdao, Shandong Province, China (36°15 N, 120°40 E). The epibionts on the shells were cleaned carefully using metal scrapers and the kept them undisturbed for a 7 day acclimation period prior to the experiment. During the experiment, 30 randomly selected mussels (fitness indexes: 39.82 ± 0.22 mm shell length, 6.05 ± 0.20 g wet weight) were placed in 7 experimental tanks (Vol. = 8 L; 210 mussels in total) which were divided into 3 groups: the HCl adjustment group (3 tanks), the CO2 enrichment group (3 tanks) and the control (1 tank). Natural seawater (pH 8.0 ± 0.1, salinity 31 ± 1.0) was filtered on a 0.45-μm pore size membrane and completely renewed every day. The temperature was maintained at 15 ± 1 °C and illumination was set on a 12 h light- 12 h dark cycle. According to the pre-test results, 200 mg (dry mass·tank−1 day−1 ) food algae, Platymonas helgolandica (Chlorophyta), was diluted in seawater and supplied to the holding tanks by gravity feed (approximate 1 mL min−1 ) and the final density in each tank was 1.5 × 105 cells min−1 . The mortality of M. edulis was measured in similar 7 tanks at the same time. The experiment lasted for 21 d
where ‘Dt ’ and ‘Ds ’ denote the dry weight (g) of the fresh tissues and shells, respectively. 2.3.2. From the perspective of metabolism 3 mussels were randomly taken out from each tank and placed into a 500 mL triangular flask respectively. All flasks were sealed by plastic films to avoid gas exchange. All the experimental flasks were acclimated under the experimental conditions for 20 min before the experiment. 3 replicates were setting up and the experiments were repeated twice. For determining the filtering rate (FR, mL Wwg−1 h−1 ), mussels were fed with the bait microalga of P. helgolandica at a final density of 1.5 × 105 cells mL−1 , and the algal density decreased steadily due to the filtering activity of the mussels. Three water samples (2 ml) from each aquarium were taken out at 30 min after algal feed and the algal density changes were measured using a hemocytometer. The filtration rate was calculated according to Quayle (1948) formula:
FR = V/(w · t ) · loge (c0 /ct )
2370
T. Sun et al. / Chemosphere 144 (2016) 2368–2376
Table 1 Setting seawater pH, measured pH, total alkalinity (TA, μM) and the concentration of HCO3 − , CO3 2− , H2 CO3 , pCO2 (ppm) in seawater in each group both HCl adjustment and CO2 enrichment (Mean ± SD). pH (setting)
pH (measured)
HCl group 8.1 7.7 7.1 6.5 CO2 group 8.1 7.7 7.1 6.5
HCO3 −
TA
0.07 0.08 0.02 0.08
1910 1686 1292 570
± ± ± ±
48 82 89 54
1625 1569 1231 561
± ± ± ±
79 118 121 64
109 41 10 2
± ± ± ±
52 13 2 1
16 35 103 156
± ± ± ±
1 5 39 35
439 1266 3186 4737
± ± ± ±
40 88 256 384
8.03 7.64 7.11 6.55
± ± ± ±
0.07 0.07 0.01 0.05
1910 1918 1910 1918
± ± ± ±
48 61 64 53
1625 1779 1835 1907
± ± ± ±
79 89 74 16
109 54 16 4
± ± ± ±
52 22 4 2
16 43 152 565
± ± ± ±
1 9 31 42
439 1352 4718 17,539
± ± ± ±
40 109 322 782
where ‘D0 ’ and ‘Dt ’ denote the dissolved oxygen (ppm) in each flask at the beginning and at time ‘t’ after exposure, respectively. The ammonia-N excretion rate (RN , μmol Wwg−1 h−1 ) was determined by the changes between the initial and the final concentrations of NH4 + -N in mussel individual at 2 h after exposure by using the sodium bromate oxidation method, and the ammonia-N excretion rate was calculated according to the following formula:
RN = {(Nt − N0 ) × V × 1000}/(w × t × 14 ) where ‘N0 ’ and ‘Nt ’ are the concentrations of NH4 + -N (ppm) in each flask at the beginning and at time ‘t’ after exposure, respectively. The O:N ration was calculated as Widdows (1978):
O : N = RO /RN The respiration rate (RC , μmol Wwg−1 h−1 ) was obtained according to the changes in the dissolved inorganic carbon (DIC) between the beginning and at 2 h after exposure. DIC was estimated using a TOC meter (TOC-vcpn, SHIMADZU, JPN) and Rc was calculated as Zhang et al. (2011):
Rc = {(DIC + DICG ) × V }/(w × t ) where ‘DIC’ is the DIC variation (μM); ‘DICG ’ could be calculated through following formula:
DICG = TAi − TA f /2 where ‘TAi ’ and ‘TAf ’ are the total alkalinity (μM) at the beginning and at time ‘t’ after exposure. 2.3.3. From the perspective of carbon sink ability Calcification was estimated by using an ‘alkalinity anomaly technique’ (Rodoni, 1993). The calcification rate (G, μmol Wwg−1 h−1 ) was calculated according to the following equation:
pCO2
± ± ± ±
Ro = {(D0 − Dt ) × V × 1000}/(w × t × 16 )
G=
H2 CO3
8.03 7.61 7.11 6.58
where ‘V’ is the total experimental volume (mL); ‘w’ is the wet weight (g) per mussel; ‘t’ is the duration (h) of the experiment; ‘c0 ’ and ‘ct ’ denote the cell densities in each flask at the beginning and at time ‘t’ after exposure, respectively. The determination of oxygen consumption rate (Ro, μmol Wwg−1 h−1 ) was performed in the same experimental system only except that the experiment lasted for 2 h, and the dissolved oxygen (DO) was measured by using a dissolved oxygen meter (DO-200, YSI, USA). Ro was calculated as follows:
CO3 2−
TAi − TA f × V /(2 × w × t )
The CA activity was measured according to the improved method of pH (Henry, 1991). The gills of the mussels in the
treated groups were sampled and CA activity units (U) was calculated according to Henry and Watts (2001):
U = t0 /tenz − 1 where ‘t0 ’ and ‘tenz ’ are response time (s) of the blanks and each pH level. CA activity was expressed as ‘μmol CO2 ·mg protein−1 ·min−1 ’ according to Wheatly and Henry (1987). The soluble protein was analyzed using the Coomassie brilliant blue G-250 method using bovine serum albumin (BSA) as a standard. 2.4. Statistics The mean values and standard deviations were calculated from the different replicates (n = 6), and figures were generated using Sigmaplot 12.5. Differences between the treated groups and the control were analyzed by a paired t-test in SPSS 22.0, with the significant difference was set at p < 0.05. General linear model (GLM) univariate ANOVA was used to investigate the effects of pH and time on each parameter. Where GLM analysis identified a significant effect of Time but not pH, the data from the different pH treatments were pooled before ANOVA analysis. Where the GLM analysis identified a significant effect of Time, the effect of pH at each sampling interval was examined separately. Hence the comparison between HCl and CO2 could be classified into a concentration-dependent manner or a time-dependent manner. 3. Results 3.1. Effects of acidified seawater on growth of M. edulis The control mussels remained in a good condition and responded to stimuli during the exposure period, and their shell length and wet weight were increased to 40.16 mm ± 0.37 mm and 6.12 ± 0.49 g respectively at the end of the experiment. Dead individuals were observed in both treated groups during the experiment, and the mortality increased steadily with the decline of pH. The mortalities were markedly higher in the CO2 group than the HCl group at the same levels of pH vales (Table 2). The gaps between two groups became wider from 6.7% to 16.6% with the pH decrease. The results suggested that CO2 enrichment might be more toxic than HCl adjustment for the mortality at the same pH level. The decrease of CI values suggested growth inhibitions when mussels exposed to acidifying seawater. However, the inhibition varied with the acidified methods. The CI values decreased significantly in the HCl group and were maintained at low levels. The largest drop (−22.2%) was observed at pH 7.7 (p < 0.05). In the CO2 group, a significant decrease (−44.4%) was also observed at pH 7.7 (p < 0.01), however, little changes of CI values at pH 7.1
T. Sun et al. / Chemosphere 144 (2016) 2368–2376 Table 2 The mortality (%) and condition indexes of M. edulis after 21 d exposure in HCl adjustment groups and CO2 enrichment groups (mean ± SD). Mortality (%)
8.1 7.7 7.1 6.5
CI
HCl group
CO2 group
HCl group
0 3.3 3.3 6.7
0 10 13.3 23.3
36.1 27.0 28.1 28.2
± ± ± ±
CO2 group
6.6 3.6∗ 12.7∗ 6.8∗
36.1 19.3 35.9 31.0
± ± ± ±
6.6 1.5∗ ∗ 1.9 9.8
Note: ‘∗ ’ means significant difference (p < 0.05); ‘∗ ∗ ’ means extremely significant difference (p < 0.01).
and 6.5 were found comparing with the control group (p > 0.05) (Table 2). 3.2. Effects of acidified seawater on the metabolic activities of M. edulis There were significant effects of both pH and Time on of M. edulis significantly in either HCl or CO2 groups, (Table 3, Fig. 1(a)). Filtering rates at pH 7.1 and 6.5 in both treatments decreased with decreasing pH since 7 d. The lowest levels were 39.6% of the control in the HCl group on 7 d and 18.4% in the CO2 group on 21 d at pH 7.1 (p < 0.01), 11.6% and 11.3% on 7 d at pH 6.5 (p < 0.01). Filtering rates at pH 7.7 in the CO2 group declined on 14 d and 21 d, but no significant effect in the HCl group was observed at pH 7.7. Compared to the HCl group, mussels in the CO2 group showed higher filtering rates on 7 d but lower rates on 14 d and 21 d. Consequently, the inhibiting ability of CO2 enrichment was stronger than HCl adjustment as the increase of exposure time. Obvious pH-dependent manner and time-dependent manner were found in both acidified groups on the oxygen consumption
Table 3 General linear model (GLM) univariate ANOVA for CI, filtering rates, oxygen consumption rates, ammonia-N excretion rates, O:N ration, carbonic anhydrase activity, calcification rates and respiratory rates, against pH and time. Indexes
HCl group F
CI Time 2.487 pH 6.997 Filtering rates Time 7.608 pH 70.962 Oxygen consumption rates Time 48.406 pH 72.232 Ammonia-N excretion rates Time 25.070 pH 30.883 O:N ration Time 24.388 pH 4.827 Respiratory rates Time 5.656 pH 891.769 Calcification rates Time 23.477 pH 54.236 Carbonic anhydrase activity Time 4.096 pH 9.208 Note: ‘ns’ means not significant.
2371
rate of M. edulis (Table 3). Oxygen consumption rate levels declined with decreasing pH since 7 d and the minimum value of in each pH level was found on 21 d (Fig. 1(b)). Although mussels in reduced pH treatments in the CO2 group increased their oxygen consumption rate on 14 d, this recovery did not continue to 21 d. Similar results were indicated by the filtering rates, i.e. oxygen consumption rates of M. edulis were affected by seawater acidification, and the inhibiting ability of CO2 enhanced with the increase of exposure time, compared to HCl addition. Both acidified methods on pH and time were confirmed to significantly influence the ammonia-N excretion rate of M. edulis (Table 3), in which mussels in the CO2 group showed a stronger inhibition than in the HCl group (Fig. 1(c)). The highest inhibitions in the CO2 group were appeared on 21 d, with the ammonia-N excretion rate at pH 7.7, 7.1, 6.5 showed a decrease of 73.4%, 65.9% and 50.0% comparing with the control (p < 0.01), respectively. The highest inhibitions in the HCl group appeared on 14 d, with the ammonia-N excretion rate at pH 7.1 and 6.5 displayed a decrease of 75.2% and 45.5% comparing with the control (p < 0.01), respectively. No significant effect of pH 7.7 in the HCl group was detected on the ammonia-N excretion rate. Time, but not pH, affected the O:N ratios in both methods of acidification (Table 3). Overall, there was an decrease in the O:N ratio with time from 30.5 to 22.6 in the HCl group, but there existed a ‘low-promoting and high-repressing’ manner in the CO2 group, in which the O:N ratios increased from 30.5 to 34.3 on 7 d and 31.9 on 14 d, respectively, then decreased to 23.7 on 21 d (Table 4). An O:N ratio in approximately 20 indicates that amino acid catabolism dominates the M. edulis aerobic metabolism. Pure protein metabolism could be expected at a ratio of 3–16, and balanced lipid and protein degradation occurs at O:N = 50–60 (Mayzaud and Conover, 1988). These results showed that acidification might lead to an increase in protein consumption and eventually changes in the metabolic type of M. edulis. Visible enhancement in the respiratory rate were found between the HCl and CO2 groups at the same pH level and time (Fig. 1(d)). In the HCl group, the respiratory rate showed a peak value at pH 7.1 on 21 d (380.2% of the control), while the peak value moved to a lower pH value (pH 6.5 on 14 d, 283.4% of the control) in the CO2 group. Therefore, the respiratory rate of M. edulis was more sensitive to the stimulation of HCl than CO2 .
CO2 group p
F
p
ns
