- Email: [email protected]
The use of the bivalve Mytilus edulis as a test organism for bioconcentration studies
ECOTOXICOLOGY
AND
ENVIRONMENTAL
SAFETY
9, 17 1- 178 ( 1985)
The Use of the Bivalve Mytilus edulis as a Test Organism for Bioconcentration Studies I. Designing
a Continuous-Flow System and Its Application to Some Organochlorine Compounds
LARS RENBERG,* National
Swedish Laboratory,
MARIA
TARKPEA,~
AND
EVA
LINDI$N~
Environment Protection Board, *Special Analytical Laboratory, Wallenberg S-106 91 Stockholm, and tBrackish Water Toxicology Laboratory, Studsvik, S-61 1 82 Nykiiping, Sweden Received
April
17. I984
Most bioconcentration studies have previously been carried out using fish as a test organism. Equally important is the use of bivalves for this purpose, from both an ecological and an economic point of view. A continuous-tlow system has thus been designed for use also with extremely hydrophobic substances and evaluated using 2,4’,Wichlorobiphenyl. methoxychlor. pentachlorolxnzene, and lindane. The variation of the uptake in the individuals after 3 weeks’ exposure was quite small (relative standard errors varied from 10.1 to 15.3% depending on the test substance), indicating a high degree of reproducibility. The bivalves, however, are known to close their valves under unfavorable conditions, which occasionally may bias the results. To overcome this disadvantage, it is suggested that an internal standard-i.e., a chemically defined compound-be added to the water simultaneously with the test substances. Although there is a principal risk for interactive effects., unexpected variations in the uptake can thus be compensated for by relating the concentration of the test substance to the concentration of the internal standard in the organisms. Comparisons between continuous-flow systems and static systems have also been made. It is concluded that continuous-flow systems are more suitable for studying hydrophobic compounds than static systems. Q 1985 Academx PI=, ITIC.
INTRODUCTION Bioaccumulation of organic substances is a matter of increasing concern. Bioaccumulating substances are able to exert their toxicity during a considerable period of time at increased levels in organisms compared to the surrounding media (direct bioconcentration) or compared to lower trophic levels (indirect bioconcentration or biomagnification). It has been generally accepted to use bioconcentration-a term usually related to studies concerning the aquatic environment-as a measurement of bioaccumulation in general. For quantification of the degree of bioaccumulation, a bioconcentration factor (BCF) is usually used, which is defined as the ratio between the concentration of a specific substance in the organism and the concentration of the substance in the water under steady-state conditions. In order to obtain the steady state, the organisms are usually experimentally exposed in a continuous-flow system, using water with a constant concentration of the substance (e.g., Neely et al., 1974; Branson et al., 1975; OECD, 1981; Veith et al., 1979). Semistatic systems or static systems have also been used, but less frequently. Within the Organization for Economic Cooperation and Development (OECD) 171
0147-6513/85
$3.00
Copyright Q 1985 by Academic Press. Inc. All rights of reproductmn m any form reserved.
172
RENBERG,
TARKPEA,
AND
LINDEN
Chemicals Testing Programme all these types of systems have been adopted (OECD, 198 1). When studying bioconcentration in fish continuous-flow systems have most commonly been used. Mollusks, on the other hand, have been studied mainly in static systems (Ernst, 1977; Hansen et al., 1978) with the exception of the bioconcentration of the organophosphate compound fenitrothion (McLeese et al., 1979). Smaller aquatic organisms (e.g., planktonic organisms) seem to be less used for bioaccumulation tests. As the ratio of the body surface and the body volume increases with decreasing size of the organism, adsorption may-for relatively waterinsoluble compounds-predominate as an “uptake” mechanism compared with transport into the organism. The objective of this investigation was to develop a continuous-flow system in combination with the commonly occurring and important bivalve Myths edulis L. as a test organism for bioconcentration study and to investigate the advantages and disadvantages of such a system using test substances with different water solubilities. The aim was also to include a comparison between a static system and the continuous-flow system using the same test substances. The test system described below has also recently been applied to studies concerning polychlorinated paraffins (L. Renberg, M. Tarkpea, and A. Bergman, in preparation). MATERIALS
AND
METHODS
Test substances. Lindane (y-hexachlorocyclohexane) and pentachlorobenzene were obtained from British Drug House Ltd, pure methoxychlor [ 1, 1, l-trichloro2,2-bis(4-methoxyphenyl)ethane] was obtained after crystallization (ethanol-water) of a technical product, and 2,4’,5trichlorobiphenyl was synthesized as described earlier (Bergman and Wachtmeister, 1977). The purity of the test substances was ascertained by gas chromatography. Test animals. The bivalves were collected at a depth of l-3 m in the Baltic bay Tvaren and were then kept up to 8 months in 60-liter all-glass aquaria with a continuous flow of water (10°C) from the Baltic Sea. Five days a week, the animals were fed with a suspension of the green algae Celenastrum copricornutum. Mussels with an approximate size of 3.5 cm were selected 2-6 days before the start of the experiment; algae, etc., were carefully removed from the shells and the animals were distributed on two glass plates (180 X 330 mm) placed horizontally at the bottom of 60-liter aquaria. After 2 days, the glass plates were raised to a vertical position. At that time almost all organisms were attached to the glass plates. A number of organisms were removed, resulting in about 20 individuals being evenly spread over each glass plate. Within a week, the plates were transferred to the aquarium used for the experiment. Continuous-flow system. The stock solution, consisting of an acetone solution of the test substances, was pumped at a flow rate of 2.9 ml/hr by the aid of a piston pump (Labotron LDP-13, West Germany) to a mixing chamber (id. 72 mm, height 240 mm) where the acetone solution was mixed with brackish water using a magnetic stirrer. The spiked water was then pumped through a polyethylene tubing to a 60-liter silicon-sealed all-glass aquarium with a flow rate of 23.4 -t 0.6 liter/hr. The construction is shown in Fig. 1. Sample preparation. After the exposure, the bivalves were immediately rinsed with brackish water. The two shells were opened with a scalpel and placed onto a
BIOCONCENTRATION
STUDIES IN Mytilus
edulis
173
FIG. 1. Experimental setup for performing the continuous-flow test with Mytilus edulis. (1) Incoming brackish water: (2) contact thermometer; (3) immersion heater; (4) heating system: (5) magnetic stirrer: (6) mixing chamber; (7) piston pump; (8) test solution flask; (9) thermometer; (10) test aquarium.
filter paper with the tissue against the paper. After 1 min, all wet tissues were removed and transferred to another filter paper and then onto a piece of aluminium foil. After the determination of their weight, the tissues were immediately deepfrozen and kept in a freezer until the chemical analysis was carried out. Water samples (500 ml) were simultaneously withdrawn, frozen, and kept in a freezer until the chemical analysis was carried out. Chemical analysis. The water sample (200 ml) was shaken with hexane, containing aldrin as internal standard (10 ml) in a 250-ml separatory funnel. The extraction was repeated with hexane (10 ml). The combined hexane extracts were gently concentrated down to a final volume of 5 ml using nitrogen and a water bath. The organism sampleswere extracted as previously described (Renberg et al., 1980). Aliquots of the extracts obtained were evaporated into dryness and the fat contents were determined gravimetrically. In order to remove the fat components each hexane extract (2 ml) was gently shaken with sulfuric acid:water (98:18, w/w) (4 ml). After centrifugation, the hexane phase was analyzed on a Varian 3700 gas chromatograph with an electron capture detector. The 170 X 0.2 (i.d.)-cm glass column was packed with a mixture of 8% QFI + 4% SF96 (2:l) on acid-washed silanized Chromosorb W 100/120 and held isothermally at 160°C. The injector and detector temperatures were held at 300°C and carrier gas (N,) flow was 30 ml/min. Recovery experiments for both water and organisms showed recoveries over 85% for all of the test substances.
174
RENBERG,
RESULTS
TARKPEA,
AND
AND
LINDEN
DISCUSSION
Test Organism Bivalves of the genus Mytilus occur in the seas all over the world and are characteristic inhabitants of the tidal zone and shallow waters. The common mussel, M. eddis L., is one of the most abundant representatives of the genus. This species is almost cosmopolite, occurring on hard bottoms along the seashore. The mussels attach themselves by protein (byssus) threads to the substrate. They are nonselective suspension feeders, filtering up to several liters of water per hour. Vahl (1972) demonstrated 80-100% retention by the common mussel of particles in the range 2-8 pm diameter. The common mussel has been successfully used as a monitoring organism for pollution. Studies on pollution levels have included trace metals, petroleum components, and organohalogen compounds (see, e.g., Goldberg et al., 1978, and references cited therein). Most of these studies have been carried out in marine waters, but several investigations have been performed in brackish waters also (Phillips, 1977). The choice of the common mussel as a test organism is due to several factors. It is a common and important organism, from both ecological and economical points of view, and it is easy to collect and to keep as a test animal in the laboratory. The widespread occurrence in both brackish and marine waters makes it possible to compare laboratory experiments with data established in the field. The common mussel can be kept for relatively long periods without any food, which makes it quite suitable for bioaccumulation studies (in our experience up to 4-6 weeks). For longer periods, the mussels are conveniently fed with unicellular green algae. Determination
of Bioconcentration
Factors (BCF)
In order to make a relevant evaluation of the continuous-flow test system suggested in this paper, four organic compounds were chosen, representing a relatively wide range of polarity and water solubility, the most polar being lindane. The use of hydrophobic substances will most probably reveal the difficulties connected with a system for determining BCF values, as problems may arise with the long exposure time necessary to obtain steady-state conditions. It has also been suggested that hydrophobicity is an important criterion for predicting the bioconcentration potential of the common mussel (Geyer et al., 1982). As previously mentioned, it is assumed that the bioconcentration factor is independent of time (steady-state conditions). For clarity, a time-dependent concentration/accumulation factor (AF,) analogous can be defined as C,,,,,/C,,, where C,,;, and C,,,:, are the concentrations of the bioaccumulating substance in the organism and water, respectively, at time t. Two experiments were performed during periods of 21 days. The first was carried out using lindane and 2,4’,5-trichlorobiphenyl as test substances. In the second experiment, methoxychlor and pentachlorobenzene were used. The results are summarized in Tables 1 and 2. In order to visualize whether steady state has been reached or not, log AF, for the test substances is plotted against time in Fig. 2. During the test period steady
BIOCONCENTRATION
STUDIES
TABLE RESULTS FROM THE
Day (1)
1 4 7 4 !I
I.5 1.6
+ 59 = 4) -+ 32 = 3) + 46 = 4) k 21 = 5)
175
edulis
CONTINUOUS
TEST
2.4’,5-Trichlorobiphenyl
Concentration in water (a/k)
-b 235 (n 177 (n 230 (n 204 (n
Myths
I
Lindane Concentration in mussel bcOW
IN
Concentration in mussel (a/W” I50
-h
1.6
I IO
I.7
140
I.9
I IO
41.500 (n 62.300 (?I 125,000 (n I3 1,000 (n
2 = + = + = k =
Average extractable fat content (%j 2 ml SD
Concentration in water (a/kg)
7.100
6.4 1.1
5.400
1.4 + 0.1
4) 3.900 3) 20.000 4) 15.300 5)
5.9
I I,000
0.8 k 0.2
9.0
14.000
1.6 2 0.1
9.9
13.000
I.? + 0.1
states were reached for lindane-for which BCF was found to be 120-and for 2,4’,5-trichloribiphenyl, with a BCF value of 13,500. Apparently, steady state is not completely reached for pentachlorobenzene and methoxychlor and the AF2, values for these were determined to 3900 and 11,500, respectively. In order to investigate the variation of the uptake in the individual organism, the relative standard errors for each single substance were calculated using the values obtained after 3 weeks’ exposure. The relative standard errors were found to vary between 10.1 (lindane) and 15.3% (trichlorobiphenyl), indicating a relatively high degree of reproducibility. The aim of using two organic compounds in the same bioconcentration experiment is based on the possibility of using one defined organic substance as an internal standard in the bioconcentration study. Assuming lindane to be such a substance and that trichlorobiphenyl was the substance to be studied, the relative standard error (based on values obtained after 2 1 days) decreases from 10.1 and 15.3%, respectively, to 4.7%, if the correlation is based on the ratio of the concentrations TABLE RESULTS Methoxychlor
Day (0 0 1 3 4 1 14 21
‘Mean ‘No
Concentration in mussel WW” -b 14,900 37,200 46,900 76,700 96,700 108,000 f 14,000 (n = 5) value samples
+ SE. n = number were taken.
FROM
THE
2
CONTINUOUS-FLOW Pentachlorobenzene
-
Concentration in water b&g) -* 10.6 12.0 7.8 8.5 8.6 9.1
of samples.
AP, (C,.JCw,) -1,400 3,100 6,000 9,000 11,000 12,000
TEST
Concenttation in mussel WW’ -6 1,620 3. I50 2,540 4,550 4,100 5.220 k 560 (n = 5)
Average extractable fat content (%) i rel SD
Concentration in water M’k) -h 1.6 1.7 I.1 1.4 1.3 1.2
1,000 1,900 2,300 3,300 3.200 4,300
1.1 1.7 1.6 1.0 2.4 1.8 2.8 ? 0.6 (n = 5)
RENBERG, TARKPEA,
176
; ,’ /
,I
:
_/ ..
AND LINDl?N
. ........ ._.,,,,,...__.....
_.
Pentachlorobenze,,e
/..~~ ,..’
.:.
*:’
3
.-._._*
.__...
..“.-..--
-.“nacc,imatired
-..’
. ..-
acchnatizsd
/...’
x..
N_ ‘\* ..--. -’ ,,__,,-.
Lindane - --- . ..____ -..-____. 2,
1
I I I 4
I 7
I 14
I 21 days
7-r--
8 days
FIG. 2. Diagram showing the accumulation factors (AF = concn in mussel/concn in water) of investigated compounds as a function of time in (A) flow-through tests and (B) static tests. In the static tests both freshly collected mussels (unacclimatized) and mussels kept for a long time in the laboratory (acclimatized) were tested.
of trichlorobiphenyl and lindane. The conclusion is that the use of an optional substance, which is spiked to the stock solution of the test substance, will reduce errors caused by the variation of the individual test animals. This is particularly useful when using bivalves as test organisms with a capability to avoid the uptake by closing the valves for a shorter or longer period, whenever the conditions are more or less unfavorable for the individual specimen. However, it should be pointed out that the use of two substances in the same experiment implies a risk for interactive effects. Comparison
with a Static System
A static system is, from a technical point of view, a more simple and less expensive system than the continuous-flow system. The water used for the experiment is spiked with the test substance before the exposure to the test animals. No water is exchanged during the experiment. Both the increase of the test substance in the organisms and the decrease in the water are monitored and, when steady-state conditions are reached, the bioconcentration factor is calculated. In order to compare the results obtained in the continuous-flow system described above with results from a static system, mussels were also exposed to a mixture of lindane and methoxychlor during a period of 8 days, mainly according to the experiments described by Ernst (1977). The results of the experiments are summarized in Table 3 and Fig. 2. In the first experiment, the animals were kept in the laboratory during the winter season (8 months) before use; in the second, recently caught organisms were used.
BIOCONCENTRATION
STUDIES IN TABLE
Mytilus
177
edulis
3
RESULTS FROM THE STATIC TESTS Lindane
Methoxychlor Average
Day (0
Concentration in mussel WW”
Concentration in water Wk)
Concentration in muse1 WW”
&L
Musselskept 8 ma#nthsin 0 I 2 4 8
-b 210 280 312 + (n = 252 * (n =
23 4) 22 5)
1.8 -b 1.6 1.3
1.1
180 240 230
the laboratory 3500 4000
5010 f 530
Concentration in water (/.&kg)
Kz”,:t,
extractable fat content (%I) k rel SD
before use 4.2 -b 1.4 1.3
2900 3900
-6 1.1 1.3 1.6 k 0.3
0.71
8000
1.4 f 0.3
3.9 -b 0.71 0.62
6100 5800
-b 2.5 2.2
1.5 + 0.3
0.51
8400
1.3 + 0.2
(n = 4) 5700 k 400 (n = 5)
Musselsrecentlycaught before use 0 1 2 4 8
-b 650 430 355 + (n = 372 k (n =
1.7 -b 1.1 74 4) 52 5)
‘Mean value+ SE.n = number
1.0
391 355
0.9
413
3500 4300 35x0 + 550 (n = 4) 4300 + 490 (n = 5)
of samples.
’ NO samples were taken.
The results indicate that steady states were reached in respect to lindane with BCF values calculated to 240 and 4 10 for “old” and “new” mussels, respectively. The corresponding accumulation factors for methoxychlor after 8 days were 8020 and 8400, which cannot be regarded as significantly different. However, steady state was not obtained for methoxychlor as a result of the short experimental period. The results indicate that, although a static system may be well suited for relatively watersoluble compounds (e.g., lindane), unpolar compounds will demand longer uptake periods than is possible with a static system with its restricted amount of oxygen available for the test animals. In the experiments described here, 8 days was estimated to be the maximum period of time without oxygen deficiency occurring, considering the amount of water and number of organisms in the system. It is also obvious that different results are obtained for lindane when using “old” and “new” mussels, respectively, while the results for methoxychlor are more constant. The reason for this phenomenon is not fully understood. ACKNOWLEDGMENTS The authors are indebted to Ulla Tjarnlund and Margareta Hansson for skillful technical assistance and to Goran Sundstriim for valuable criticism. The gift of 2,4’,5-trichlorobiphenyl from Ake Bergman is gratefully acknowledged.
REFERENCES BERGMAN, A., AND WACHTMEISTER, C. A. (1977). Synthesis of [‘%]biphenyl and of some chlorinated [%]biphenyls containing 4&loro-, 2,4dichloro- and 2,3,6-trichlorophenyl nuclei from the corresponding labelled anilines. Chemosphere 6, 759-76’7. ERNST, W. (1977). Determination of the bioconcentration potential of marine organisms-A steady state approach. I. Bioconcentration data for seven chlorinated pesticides in mussel (Mytilus edulis) and their relation to solubility data. Chemosphere II, 731-740.
178
RENBERG,
TARKPEA,
AND
LlNDI?N
H., SCHEEHAN, P., KOTZIAS, D., FREITAG, D., AND KORTE, F. (I 982). Prediction of ecotoxicological behaviour of chemicals. Relationship between physico-chemical properties and bioaccumulation of organic chemicals in the mussel Myfilus edulis. Chemosphere 11, 112 1- 1134. GOLDBERG, E. D., BOWEN, V. T., FARRINGTON, J. W., HARVEY, G., MARTIN, J. H.. PARKER, P. L., RISEBROUGH, R. W., ROBERTSON, W., SCHNEIDER, E., AND GAMBLE, E. (1978). The musselwatch. Environ. Conserv. 5, 10 1- 125. HANSEN, N., JENSEN, V. B., APPELQUIST, H., AND MARCH, E. (1978). The uptake and release of petroleum hydrocarbons by the marine mussel Mytilus edulis. Prog. Water Technol. 10, 35 1-359. MCLEESE, D. W., ZITKO, V., AND SERGEANT, D. B. (1979). Uptake and excretion of fenitrothion by clams and mussels. Bull. Environ. Contam. Toxicol. 22, 800-806. OECD Chemicals Testing Programme. Degradation/Accumulation Group (1979). Inter Comparison Laboratory Work on Bioaccumulation Tests. PHILLIPS, D. J. H. (1977). The common mussel Mytilus edulis as an indicator of trace metals in Scandinavian Waters. I. Zinc and cadmium. Mar. Biol. 43, 283-291. RENBERG, L., SVANBERG, O., BENGTSSON,B-E., AND SUNDSTR~M, G. (1980). Chlorinated guaiacols and catechols. Bioaccumulation potential in Bleaks (Alburnus alburnus, Pisces) and reproductive and toxic effects on the harpacticoid Nifocra spinipes (Crustacea). Chemosphere 9, 143-150. VAHL, 0. (1972). Efficiency of particle retention in Myrilus edulis L. Ophelia 10, 17-25. VEITH, G. D., DEFOE, D. L., AND BERGSTEDT,B. V. (1979). Measuring and estimating the bioconcentration factor of chemicals in fish. J. Fish. Res. Board Canad. 36, 1040-1048. GEYER,
AND
ENVIRONMENTAL
SAFETY
9, 17 1- 178 ( 1985)
The Use of the Bivalve Mytilus edulis as a Test Organism for Bioconcentration Studies I. Designing
a Continuous-Flow System and Its Application to Some Organochlorine Compounds
LARS RENBERG,* National
Swedish Laboratory,
MARIA
TARKPEA,~
AND
EVA
LINDI$N~
Environment Protection Board, *Special Analytical Laboratory, Wallenberg S-106 91 Stockholm, and tBrackish Water Toxicology Laboratory, Studsvik, S-61 1 82 Nykiiping, Sweden Received
April
17. I984
Most bioconcentration studies have previously been carried out using fish as a test organism. Equally important is the use of bivalves for this purpose, from both an ecological and an economic point of view. A continuous-tlow system has thus been designed for use also with extremely hydrophobic substances and evaluated using 2,4’,Wichlorobiphenyl. methoxychlor. pentachlorolxnzene, and lindane. The variation of the uptake in the individuals after 3 weeks’ exposure was quite small (relative standard errors varied from 10.1 to 15.3% depending on the test substance), indicating a high degree of reproducibility. The bivalves, however, are known to close their valves under unfavorable conditions, which occasionally may bias the results. To overcome this disadvantage, it is suggested that an internal standard-i.e., a chemically defined compound-be added to the water simultaneously with the test substances. Although there is a principal risk for interactive effects., unexpected variations in the uptake can thus be compensated for by relating the concentration of the test substance to the concentration of the internal standard in the organisms. Comparisons between continuous-flow systems and static systems have also been made. It is concluded that continuous-flow systems are more suitable for studying hydrophobic compounds than static systems. Q 1985 Academx PI=, ITIC.
INTRODUCTION Bioaccumulation of organic substances is a matter of increasing concern. Bioaccumulating substances are able to exert their toxicity during a considerable period of time at increased levels in organisms compared to the surrounding media (direct bioconcentration) or compared to lower trophic levels (indirect bioconcentration or biomagnification). It has been generally accepted to use bioconcentration-a term usually related to studies concerning the aquatic environment-as a measurement of bioaccumulation in general. For quantification of the degree of bioaccumulation, a bioconcentration factor (BCF) is usually used, which is defined as the ratio between the concentration of a specific substance in the organism and the concentration of the substance in the water under steady-state conditions. In order to obtain the steady state, the organisms are usually experimentally exposed in a continuous-flow system, using water with a constant concentration of the substance (e.g., Neely et al., 1974; Branson et al., 1975; OECD, 1981; Veith et al., 1979). Semistatic systems or static systems have also been used, but less frequently. Within the Organization for Economic Cooperation and Development (OECD) 171
0147-6513/85
$3.00
Copyright Q 1985 by Academic Press. Inc. All rights of reproductmn m any form reserved.
172
RENBERG,
TARKPEA,
AND
LINDEN
Chemicals Testing Programme all these types of systems have been adopted (OECD, 198 1). When studying bioconcentration in fish continuous-flow systems have most commonly been used. Mollusks, on the other hand, have been studied mainly in static systems (Ernst, 1977; Hansen et al., 1978) with the exception of the bioconcentration of the organophosphate compound fenitrothion (McLeese et al., 1979). Smaller aquatic organisms (e.g., planktonic organisms) seem to be less used for bioaccumulation tests. As the ratio of the body surface and the body volume increases with decreasing size of the organism, adsorption may-for relatively waterinsoluble compounds-predominate as an “uptake” mechanism compared with transport into the organism. The objective of this investigation was to develop a continuous-flow system in combination with the commonly occurring and important bivalve Myths edulis L. as a test organism for bioconcentration study and to investigate the advantages and disadvantages of such a system using test substances with different water solubilities. The aim was also to include a comparison between a static system and the continuous-flow system using the same test substances. The test system described below has also recently been applied to studies concerning polychlorinated paraffins (L. Renberg, M. Tarkpea, and A. Bergman, in preparation). MATERIALS
AND
METHODS
Test substances. Lindane (y-hexachlorocyclohexane) and pentachlorobenzene were obtained from British Drug House Ltd, pure methoxychlor [ 1, 1, l-trichloro2,2-bis(4-methoxyphenyl)ethane] was obtained after crystallization (ethanol-water) of a technical product, and 2,4’,5trichlorobiphenyl was synthesized as described earlier (Bergman and Wachtmeister, 1977). The purity of the test substances was ascertained by gas chromatography. Test animals. The bivalves were collected at a depth of l-3 m in the Baltic bay Tvaren and were then kept up to 8 months in 60-liter all-glass aquaria with a continuous flow of water (10°C) from the Baltic Sea. Five days a week, the animals were fed with a suspension of the green algae Celenastrum copricornutum. Mussels with an approximate size of 3.5 cm were selected 2-6 days before the start of the experiment; algae, etc., were carefully removed from the shells and the animals were distributed on two glass plates (180 X 330 mm) placed horizontally at the bottom of 60-liter aquaria. After 2 days, the glass plates were raised to a vertical position. At that time almost all organisms were attached to the glass plates. A number of organisms were removed, resulting in about 20 individuals being evenly spread over each glass plate. Within a week, the plates were transferred to the aquarium used for the experiment. Continuous-flow system. The stock solution, consisting of an acetone solution of the test substances, was pumped at a flow rate of 2.9 ml/hr by the aid of a piston pump (Labotron LDP-13, West Germany) to a mixing chamber (id. 72 mm, height 240 mm) where the acetone solution was mixed with brackish water using a magnetic stirrer. The spiked water was then pumped through a polyethylene tubing to a 60-liter silicon-sealed all-glass aquarium with a flow rate of 23.4 -t 0.6 liter/hr. The construction is shown in Fig. 1. Sample preparation. After the exposure, the bivalves were immediately rinsed with brackish water. The two shells were opened with a scalpel and placed onto a
BIOCONCENTRATION
STUDIES IN Mytilus
edulis
173
FIG. 1. Experimental setup for performing the continuous-flow test with Mytilus edulis. (1) Incoming brackish water: (2) contact thermometer; (3) immersion heater; (4) heating system: (5) magnetic stirrer: (6) mixing chamber; (7) piston pump; (8) test solution flask; (9) thermometer; (10) test aquarium.
filter paper with the tissue against the paper. After 1 min, all wet tissues were removed and transferred to another filter paper and then onto a piece of aluminium foil. After the determination of their weight, the tissues were immediately deepfrozen and kept in a freezer until the chemical analysis was carried out. Water samples (500 ml) were simultaneously withdrawn, frozen, and kept in a freezer until the chemical analysis was carried out. Chemical analysis. The water sample (200 ml) was shaken with hexane, containing aldrin as internal standard (10 ml) in a 250-ml separatory funnel. The extraction was repeated with hexane (10 ml). The combined hexane extracts were gently concentrated down to a final volume of 5 ml using nitrogen and a water bath. The organism sampleswere extracted as previously described (Renberg et al., 1980). Aliquots of the extracts obtained were evaporated into dryness and the fat contents were determined gravimetrically. In order to remove the fat components each hexane extract (2 ml) was gently shaken with sulfuric acid:water (98:18, w/w) (4 ml). After centrifugation, the hexane phase was analyzed on a Varian 3700 gas chromatograph with an electron capture detector. The 170 X 0.2 (i.d.)-cm glass column was packed with a mixture of 8% QFI + 4% SF96 (2:l) on acid-washed silanized Chromosorb W 100/120 and held isothermally at 160°C. The injector and detector temperatures were held at 300°C and carrier gas (N,) flow was 30 ml/min. Recovery experiments for both water and organisms showed recoveries over 85% for all of the test substances.
174
RENBERG,
RESULTS
TARKPEA,
AND
AND
LINDEN
DISCUSSION
Test Organism Bivalves of the genus Mytilus occur in the seas all over the world and are characteristic inhabitants of the tidal zone and shallow waters. The common mussel, M. eddis L., is one of the most abundant representatives of the genus. This species is almost cosmopolite, occurring on hard bottoms along the seashore. The mussels attach themselves by protein (byssus) threads to the substrate. They are nonselective suspension feeders, filtering up to several liters of water per hour. Vahl (1972) demonstrated 80-100% retention by the common mussel of particles in the range 2-8 pm diameter. The common mussel has been successfully used as a monitoring organism for pollution. Studies on pollution levels have included trace metals, petroleum components, and organohalogen compounds (see, e.g., Goldberg et al., 1978, and references cited therein). Most of these studies have been carried out in marine waters, but several investigations have been performed in brackish waters also (Phillips, 1977). The choice of the common mussel as a test organism is due to several factors. It is a common and important organism, from both ecological and economical points of view, and it is easy to collect and to keep as a test animal in the laboratory. The widespread occurrence in both brackish and marine waters makes it possible to compare laboratory experiments with data established in the field. The common mussel can be kept for relatively long periods without any food, which makes it quite suitable for bioaccumulation studies (in our experience up to 4-6 weeks). For longer periods, the mussels are conveniently fed with unicellular green algae. Determination
of Bioconcentration
Factors (BCF)
In order to make a relevant evaluation of the continuous-flow test system suggested in this paper, four organic compounds were chosen, representing a relatively wide range of polarity and water solubility, the most polar being lindane. The use of hydrophobic substances will most probably reveal the difficulties connected with a system for determining BCF values, as problems may arise with the long exposure time necessary to obtain steady-state conditions. It has also been suggested that hydrophobicity is an important criterion for predicting the bioconcentration potential of the common mussel (Geyer et al., 1982). As previously mentioned, it is assumed that the bioconcentration factor is independent of time (steady-state conditions). For clarity, a time-dependent concentration/accumulation factor (AF,) analogous can be defined as C,,,,,/C,,, where C,,;, and C,,,:, are the concentrations of the bioaccumulating substance in the organism and water, respectively, at time t. Two experiments were performed during periods of 21 days. The first was carried out using lindane and 2,4’,5-trichlorobiphenyl as test substances. In the second experiment, methoxychlor and pentachlorobenzene were used. The results are summarized in Tables 1 and 2. In order to visualize whether steady state has been reached or not, log AF, for the test substances is plotted against time in Fig. 2. During the test period steady
BIOCONCENTRATION
STUDIES
TABLE RESULTS FROM THE
Day (1)
1 4 7 4 !I
I.5 1.6
+ 59 = 4) -+ 32 = 3) + 46 = 4) k 21 = 5)
175
edulis
CONTINUOUS
TEST
2.4’,5-Trichlorobiphenyl
Concentration in water (a/k)
-b 235 (n 177 (n 230 (n 204 (n
Myths
I
Lindane Concentration in mussel bcOW
IN
Concentration in mussel (a/W” I50
-h
1.6
I IO
I.7
140
I.9
I IO
41.500 (n 62.300 (?I 125,000 (n I3 1,000 (n
2 = + = + = k =
Average extractable fat content (%j 2 ml SD
Concentration in water (a/kg)
7.100
6.4 1.1
5.400
1.4 + 0.1
4) 3.900 3) 20.000 4) 15.300 5)
5.9
I I,000
0.8 k 0.2
9.0
14.000
1.6 2 0.1
9.9
13.000
I.? + 0.1
states were reached for lindane-for which BCF was found to be 120-and for 2,4’,5-trichloribiphenyl, with a BCF value of 13,500. Apparently, steady state is not completely reached for pentachlorobenzene and methoxychlor and the AF2, values for these were determined to 3900 and 11,500, respectively. In order to investigate the variation of the uptake in the individual organism, the relative standard errors for each single substance were calculated using the values obtained after 3 weeks’ exposure. The relative standard errors were found to vary between 10.1 (lindane) and 15.3% (trichlorobiphenyl), indicating a relatively high degree of reproducibility. The aim of using two organic compounds in the same bioconcentration experiment is based on the possibility of using one defined organic substance as an internal standard in the bioconcentration study. Assuming lindane to be such a substance and that trichlorobiphenyl was the substance to be studied, the relative standard error (based on values obtained after 2 1 days) decreases from 10.1 and 15.3%, respectively, to 4.7%, if the correlation is based on the ratio of the concentrations TABLE RESULTS Methoxychlor
Day (0 0 1 3 4 1 14 21
‘Mean ‘No
Concentration in mussel WW” -b 14,900 37,200 46,900 76,700 96,700 108,000 f 14,000 (n = 5) value samples
+ SE. n = number were taken.
FROM
THE
2
CONTINUOUS-FLOW Pentachlorobenzene
-
Concentration in water b&g) -* 10.6 12.0 7.8 8.5 8.6 9.1
of samples.
AP, (C,.JCw,) -1,400 3,100 6,000 9,000 11,000 12,000
TEST
Concenttation in mussel WW’ -6 1,620 3. I50 2,540 4,550 4,100 5.220 k 560 (n = 5)
Average extractable fat content (%) i rel SD
Concentration in water M’k) -h 1.6 1.7 I.1 1.4 1.3 1.2
1,000 1,900 2,300 3,300 3.200 4,300
1.1 1.7 1.6 1.0 2.4 1.8 2.8 ? 0.6 (n = 5)
RENBERG, TARKPEA,
176
; ,’ /
,I
:
_/ ..
AND LINDl?N
. ........ ._.,,,,,...__.....
_.
Pentachlorobenze,,e
/..~~ ,..’
.:.
*:’
3
.-._._*
.__...
..“.-..--
-.“nacc,imatired
-..’
. ..-
acchnatizsd
/...’
x..
N_ ‘\* ..--. -’ ,,__,,-.
Lindane - --- . ..____ -..-____. 2,
1
I I I 4
I 7
I 14
I 21 days
7-r--
8 days
FIG. 2. Diagram showing the accumulation factors (AF = concn in mussel/concn in water) of investigated compounds as a function of time in (A) flow-through tests and (B) static tests. In the static tests both freshly collected mussels (unacclimatized) and mussels kept for a long time in the laboratory (acclimatized) were tested.
of trichlorobiphenyl and lindane. The conclusion is that the use of an optional substance, which is spiked to the stock solution of the test substance, will reduce errors caused by the variation of the individual test animals. This is particularly useful when using bivalves as test organisms with a capability to avoid the uptake by closing the valves for a shorter or longer period, whenever the conditions are more or less unfavorable for the individual specimen. However, it should be pointed out that the use of two substances in the same experiment implies a risk for interactive effects. Comparison
with a Static System
A static system is, from a technical point of view, a more simple and less expensive system than the continuous-flow system. The water used for the experiment is spiked with the test substance before the exposure to the test animals. No water is exchanged during the experiment. Both the increase of the test substance in the organisms and the decrease in the water are monitored and, when steady-state conditions are reached, the bioconcentration factor is calculated. In order to compare the results obtained in the continuous-flow system described above with results from a static system, mussels were also exposed to a mixture of lindane and methoxychlor during a period of 8 days, mainly according to the experiments described by Ernst (1977). The results of the experiments are summarized in Table 3 and Fig. 2. In the first experiment, the animals were kept in the laboratory during the winter season (8 months) before use; in the second, recently caught organisms were used.
BIOCONCENTRATION
STUDIES IN TABLE
Mytilus
177
edulis
3
RESULTS FROM THE STATIC TESTS Lindane
Methoxychlor Average
Day (0
Concentration in mussel WW”
Concentration in water Wk)
Concentration in muse1 WW”
&L
Musselskept 8 ma#nthsin 0 I 2 4 8
-b 210 280 312 + (n = 252 * (n =
23 4) 22 5)
1.8 -b 1.6 1.3
1.1
180 240 230
the laboratory 3500 4000
5010 f 530
Concentration in water (/.&kg)
Kz”,:t,
extractable fat content (%I) k rel SD
before use 4.2 -b 1.4 1.3
2900 3900
-6 1.1 1.3 1.6 k 0.3
0.71
8000
1.4 f 0.3
3.9 -b 0.71 0.62
6100 5800
-b 2.5 2.2
1.5 + 0.3
0.51
8400
1.3 + 0.2
(n = 4) 5700 k 400 (n = 5)
Musselsrecentlycaught before use 0 1 2 4 8
-b 650 430 355 + (n = 372 k (n =
1.7 -b 1.1 74 4) 52 5)
‘Mean value+ SE.n = number
1.0
391 355
0.9
413
3500 4300 35x0 + 550 (n = 4) 4300 + 490 (n = 5)
of samples.
’ NO samples were taken.
The results indicate that steady states were reached in respect to lindane with BCF values calculated to 240 and 4 10 for “old” and “new” mussels, respectively. The corresponding accumulation factors for methoxychlor after 8 days were 8020 and 8400, which cannot be regarded as significantly different. However, steady state was not obtained for methoxychlor as a result of the short experimental period. The results indicate that, although a static system may be well suited for relatively watersoluble compounds (e.g., lindane), unpolar compounds will demand longer uptake periods than is possible with a static system with its restricted amount of oxygen available for the test animals. In the experiments described here, 8 days was estimated to be the maximum period of time without oxygen deficiency occurring, considering the amount of water and number of organisms in the system. It is also obvious that different results are obtained for lindane when using “old” and “new” mussels, respectively, while the results for methoxychlor are more constant. The reason for this phenomenon is not fully understood. ACKNOWLEDGMENTS The authors are indebted to Ulla Tjarnlund and Margareta Hansson for skillful technical assistance and to Goran Sundstriim for valuable criticism. The gift of 2,4’,5-trichlorobiphenyl from Ake Bergman is gratefully acknowledged.
REFERENCES BERGMAN, A., AND WACHTMEISTER, C. A. (1977). Synthesis of [‘%]biphenyl and of some chlorinated [%]biphenyls containing 4&loro-, 2,4dichloro- and 2,3,6-trichlorophenyl nuclei from the corresponding labelled anilines. Chemosphere 6, 759-76’7. ERNST, W. (1977). Determination of the bioconcentration potential of marine organisms-A steady state approach. I. Bioconcentration data for seven chlorinated pesticides in mussel (Mytilus edulis) and their relation to solubility data. Chemosphere II, 731-740.
178
RENBERG,
TARKPEA,
AND
LlNDI?N
H., SCHEEHAN, P., KOTZIAS, D., FREITAG, D., AND KORTE, F. (I 982). Prediction of ecotoxicological behaviour of chemicals. Relationship between physico-chemical properties and bioaccumulation of organic chemicals in the mussel Myfilus edulis. Chemosphere 11, 112 1- 1134. GOLDBERG, E. D., BOWEN, V. T., FARRINGTON, J. W., HARVEY, G., MARTIN, J. H.. PARKER, P. L., RISEBROUGH, R. W., ROBERTSON, W., SCHNEIDER, E., AND GAMBLE, E. (1978). The musselwatch. Environ. Conserv. 5, 10 1- 125. HANSEN, N., JENSEN, V. B., APPELQUIST, H., AND MARCH, E. (1978). The uptake and release of petroleum hydrocarbons by the marine mussel Mytilus edulis. Prog. Water Technol. 10, 35 1-359. MCLEESE, D. W., ZITKO, V., AND SERGEANT, D. B. (1979). Uptake and excretion of fenitrothion by clams and mussels. Bull. Environ. Contam. Toxicol. 22, 800-806. OECD Chemicals Testing Programme. Degradation/Accumulation Group (1979). Inter Comparison Laboratory Work on Bioaccumulation Tests. PHILLIPS, D. J. H. (1977). The common mussel Mytilus edulis as an indicator of trace metals in Scandinavian Waters. I. Zinc and cadmium. Mar. Biol. 43, 283-291. RENBERG, L., SVANBERG, O., BENGTSSON,B-E., AND SUNDSTR~M, G. (1980). Chlorinated guaiacols and catechols. Bioaccumulation potential in Bleaks (Alburnus alburnus, Pisces) and reproductive and toxic effects on the harpacticoid Nifocra spinipes (Crustacea). Chemosphere 9, 143-150. VAHL, 0. (1972). Efficiency of particle retention in Myrilus edulis L. Ophelia 10, 17-25. VEITH, G. D., DEFOE, D. L., AND BERGSTEDT,B. V. (1979). Measuring and estimating the bioconcentration factor of chemicals in fish. J. Fish. Res. Board Canad. 36, 1040-1048. GEYER,