Abstract: A study was carried out to evaluate the acute and chronic toxicity of lindane and atrazine in Mytilus edulis. The threshold median lethal dose for lindane was approached at long elapsed times (90 days) when the LC50 was 1.7 mg l-1. The LT50 at a concentration of 30 mg l-1 was 8.4 days which appears to be approaching the limiting time. For atrazine the median lethal threshold concentration was not reached even at the lowest dose (1.25 mg l-1) used and after 112 days the LC50 was 1.6 mg l-1. Similarly a limiting time was not reached even at the highest does (30 mg l-1) when the LT50 was 1.7 days. More marginal concentration and time effects (LC90 and LC10; LT90 and LT10) were also estimated for lindane and atrazine using probit analysis. Valve movement response of the mussels during acute and chronic bioassay toxicity of the pesticides was evaluated as biomarker. The data indicated that the rest period increased with increasing concentration and time of exposure. It could be concluded that there is a time-dose-response relationships of lindane and atrazine with mussels (Mytilus edulis).
Introduction
Atrazine is a selective triazine-based herbicide used for control of wideleaf weeds in maize, broomcorn and vine and in apple and pear plants (Neskovic et al., 1993). Similar to other herbicidal triazines, atrazine is a potential contaminant of surface and ground water; it is known to have high mobility though soil, so it may easily find a way to the water. In water, triazines are practically unaffected by microbial or hydrolytic degradation processes (Gamble et al., 1983). Thus data concerning their toxic effects on aquatic organisms are very helpful for the assessment of the dangers these chemicals may pose. Toxicity studies of atrazine on aquatic organisms are acute toxicity on fish, daphnia and carp (Marchini et al., 1988 and Neskovic et al., 1993). There is no study concern on its effect on bivalve. However, lindane (isomer gamma hexschlorocyclohexane) causes serious ecotoxicological problems mainly due to its persistence and high toxicity. Lindane is persisted in soils and may reach the marine environment through erosion processed (Hamza-Chaffia et al., 1993).
Mussels belonging to the genus Mytilus have proved to be model organisms for various physiological, biochemical and genetic investigations (El-Shenawy, 1999 and El-Shenawy et al., 2001a). They are also important economically as food and as biomonitors of coastal water quality (Gosling, 1992; Widdows and Donkin, 1992; Englund and Heino, 1996). Recording the activity of mussels (valves opening and closing) is used for ecotoxicological testing of chemicals under laboratory conditions (Salánki, 1979; Borcherding, 1992; Borcherding and Jantz, 1997; Borcherding and Wolf, 2001). Also, production of byssus thread has been used as an indicator of exposure to pollutants (Roberts, 1975; El-Shenawy et al., 2001b).
Based on the previous studies, the present work was undertaken to determine both median lethal concentration (LC50) at a range of times and median lethal time (LT50) at a range of concentrations for lindane and atrazine using probit analysis. Then, Monitoring and comparison the behaviour -in particular the valve movement- of mussels, M. edulis, exposed to different concentrations of lindane or atrazine for short and long periods of time.
Materials and Methods
Collection and maintenance
Mytilus edulis in the size of 50-60 mm were collected from South
Parade pier on the shore of Southsea, Portsmouth, UK. They were acclimatized
to laboratory conditions for 2 weeks in tanks prior to experimentation. The
tanks were supplied with a continuous flow of sea water from Langstone Harbour.
The temperature varied depending on the time of the year between 8 and 20°C
and salinity was 34‰. A natural photoperiod was maintained. The mussels
were given no food other than that occurring naturally in water surrounding
them. At the start of the experiments, they were transferred to a test cold
room and again were given no additional food.
Test chemicals
Gamma-hexachlorocyclohexane (γ-HCH) purity 99% was obtained from Aldrich,
catalogue no. 23,339-0. Stock solutions were prepared in HPLC grade acetone
(Fisher Laboratory Chemicals, A/0606/17) and stored at -20°C. Stock solutions
of atrazine (purity, 98.8 % from Chem Service) were prepared also by dissolving
in HPLC grade acetone.
Toxicity tests to determine the LC50 for γ-HCH and atrazine
A static system with daily renewal of solutions was used in both acute and
chronic exposures (Ernst 1977). The determination of the LC50s for
lindane and atrazine were carried out many times. A preliminary experiment was
done to determine the LC50 for 24-96 hours exposure. Then the main
experiment was carried out at concentrations lower than those tested at the
preliminary experiment for a period of a month, until 3 months. 10 mussels of
species M. edulis were placed in a glass tank with 2.5 l of sea water
taken directly from Langstone Harbour. Duplicate tanks acted as controls receiving
acetone (the carrier solvent) and others contained different concentrations
of toxicant. A wide range of concentrations was used in order to determine the
tolerance distribution in both short and long term experiments.
Measurement of valve movement
Measurement of this behavioral activity response of mussels was following
the method of Salánki (1979) modified by Kramer et al. (1989)
and Salánki et al. (1991).
Byssal production
Measurement of byssus formation has been used as indicator of toxic effects
on the activity of mussels. In this study, the method of Martin et al.
(1975) was used to measure the byssal thread growth.
Statistical analysis
The data were analyzed using probit analysis to determine LC50,
LT50 and the concentration necessary for a 50% inhibition of valve
movement (ED50) of both pesticides. One way analysis of variance
(ANOVA) and Duncan’s multiple range were used to test significance of differences
between the effects of individual concentrations (Saunders and Tarp, 1990).
Results
The determination of LC50 of γ-HCH was repeated on a number of time scales (acute, sub-chronic and chronic bioassays). In acute bioassay, mussels were treated with the following concentrations of γ-HCH; 0.1, 1 and 10 mg l-1 for 96 h. However, in a sub-chronic toxicity test, the experiment was repeated using different concentrations (30, 15, 7.5 and 3.75 mg l-1) of γ-HCH for 24 days. A range of lower doses (1.8, 0.94 and 0.47 mg l-1) was added to extend the exposure to 3 months. The LC50 at 7 days was 61.6 mg l-1; at 15 days, 4.8 mg l-1; at 30 days, 2.3 mg l-1; at 60 days, 1.8 mg l-1 and at the end of the experiment (75 and 90 days), 1.7 mg l-1. Limits in time and concentration were approached (Table 1 and 2). At the 4th day in the acute toxicity test the mussels were not affected by any of the concentrations (0.1, 1 and 10 mg l-1) of γ-HCH. With the higher range of concentrations (30, 15, 7.5 and 3.75 mg l-1) of γ-HCH applied for 24 days (Fig. 1). The LT50 was estimated as 8.4 days at 30 mg l-1; 8.8 days at 15 mg l-1; 11.4 days at 7.5 mg l-1; 19.4 days at 3.75 mg l-1; 66.4 days 1.87 mg l-1, 1.87; and 143 days at 0.9 mg l-1 (Table 2). No mortality was observed in the mussels exposed to a concentration of 0.45 mg l-1 γ-HCH even after 3 months exposure. More marginal effects of concentration and time (LC90 and LC10; LT 90 and LT10) were also estimated for lindane using probit analysis (Table 3 and 4).
| Table 1: | Concentration-response analysis of lindane. Median Lethal Concentrations (LC50) (mg l-1) and slope function (± SE) at a range of exposure time (days) of mussels (5-6 cm) |
| Table 2: | Time-response analysis of lindane treated mussels. Median period of survival (L50-days) and slope function (± SE) at a range of concentrations |
| Table 3: | Concentration-response analysis of lindane. Ten and ninety percent lethal concentrations (LC10 and LC90) (mg l-1) at a range of exposure times (days) of mussels (5-6 cm) |
| Table 4: | Time-response analysis of lindane treated mussels. Ten and ninety percent period of survival (LT10 and LT90-days) (days) at a range of concentrations |
| Fig. 1: | Concentration-response analysis of lindane in the marine mussel, M. edulis |
It is clear that the medium lethal threshold concentration is approached at 91-101 days, but atrazine was still killing mussels at a concentration of 1.25 mg l-1 at the end of the chronic experiment (4 months). However, no mortality was reported for control population during the course of the experiment. The LC50 for atrazine decreased with increasing time of exposure (Table 5 and Fig. 2).
| Table 5: | Concentration-response analysis. Median lethal concentrations (LC50) mg l-1 of atrazine at a range of exposure times (days) of mussels (5-6 cm) |
| Fig. 2: | Concentration-response analysis of atrazine in the marine mussel, M. edulis |
In the acute toxicity test, the LT50 was estimated as 1.7 days at 30 mg l-1. In the chronic tests (exposure for up to 4 months) the LT50 was estimated as 9.6 days at 20 mg l-1; and 26.3 days at 10 mg l-1. However, at the three lowest concentrations (5, 2.5 and 1.25 mg l-1) of atrazine the LT50 estimates were very similar, all being in the range 90 to 100 days and there was no trend with increasing concentration (Table 6). Ten and ninety percent lethal concentrations (LC10 and LC90 mg l-1) of atrazine at a range of exposure times (days) of mussels (5-6 cm) were presented in Table 7. Also, ten and ninety percent period of survival (LT10 and LT90) (days) of the marine mussel exposed to a range of concentration of atrazine were presented in Table 8.
| Table 6: | Time-response analysis. Median period of survival (LT50) (days) and slope function (± SE) of the mussels exposed to a range of concentrations of atrazine |
| Table 7: | Concentration-response analysis. Ten and ninety percent lethal concentrations (LC10 and LC90 (mg l-1) of atrazine at a range of exposure times (days) of mussels (5-6 cm) |
| Table 8: | Time-response analysis. Ten and ninety percent period of survival (LT10 and LT90) (days) of the marine mussel exposed to a range of concentration of atrazine |
A threshold median lethal concentration for atrazine appeared to be approached at 14 days, but then the LC50 started to fall with time and endpoint had not been reached by 112 days (Fig. 2) when the LC50 was 1.6 mg l-1. Similarly a limiting time had not been approached at the highest concentration (30 mg l-1) by the end of this study.
Evaluation of the valve movement response of the mussels during short-term (4 days) response to lindane using 1, 5 and 10 mg l-1 γ-HCH and a slow speed recorder. Additional measurements of control activities were made during the course of the whole experimental period. The normal valve movement behaviour of M. edulis was shown in Fig. 3. It was characterized by long periods of open state, with short intermittent periods closed state. The valves re-open within a few minutes of any closure. As time elapsed following the addition of toxicant a gradual increase in the time spent with the valve closed was observed. At longer exposure time, they responded by a total closure of the shell valves. Mean duration of activity, expressed as a percentage of time spent in the open state, at a range of periods of exposure to different concentrations of γ-HCH 1, 5 and 10 mg l-1 were presented in Fig. 3. The mussel activity at zero time was 97.2%. After 12 h the duration of the activity was 95.8, 85.2, 66.3 and 45.8% for control, 1, 5 and 10 mg l-1 lindane, respectively. Changes in valve movement patterns were seen in all three concentrations of lindane during continuous recording for 4 days. By the end of the acute toxicity study valve open time in animals in a concentration of 5 mg l-1 lindane, was decreased to 29.7 ± 2.8 % of control levels. At the highest lindane concentration (10 mg l-1) activity was markedly decreased and closure period increased until the animals became moribund and were active for only 10 minutes in every 24 h (Fig. 3). The data indicate that the rest period increased with increasing concentration of lindane and time of exposure (Table 9).
| Fig. 3: | The duration of the active period (% of total observation
time) of Mytilus edulis over 4 days at a range of concentrations
of lindane. Each point represents the mean±SE of the duration of
activity (%) (n=17-20). Activity periods for all treatment levels at all
exposure times were significantly different from the respective control
value (ANOVA with Dunnett’s test) p<0.001 |
Probit analysis has been used, to estimate ET50s, the times necessary to inhibit 50% of the mussels activity, at each of the various concentrations used in this study (Table 10) and EC50s, the concentrations of lindane necessary for 50% inhibition at each of a range of exposure times (Table 11). A 50% reduction in activity in the present study was reached after only 13.6 h in mussels exposed to a concentration of 10 mg l-1 γ-HCH (Table 10). EC50s of γ-HCH was 10.4 mg l-1 at 12 h and 4.8 mg l-1 at 24 h. After 2 days this had fallen to 8.6 mg l-1 (Table 11).
| Table 9: | The Effect of lindane on the active and rest periods of the mussel, Mytilus edulis |
| A: represents the mean ± SE of the change of duration
of the active periods expressed as percent of control at the same time (n
= 17-20) R: represents the mean ± SE of the change of duration of the rest periods expressed as percent of control at the same time (n = 17-20) |
|
| Table 10: | Time-response analysis. Median effective period (LT50) (days) and slope functions for inhibition of the valve opening activity of mussels exposed to a range of concentrations of γ-HCH |
| Table 11: | Concentration- response analysis. Median effective concentrations (EC50) (mg l-1) of γ-HCH for the inhibition of valve opening activity in M. edulis at a range of exposure times |
In this chronic study the length of time spent in an active state was monitored in animals exposed to either lindane or atrazine over a long period (up to 3 months). As in the earlier study duration of activity was expressed as the percentage of time spent in an open state. During this 3 months study of chronic toxicity, LC50 was estimated at a range of elapsed times and activity was measured after 7, 14, 28, 56 and 84 days of exposure. Changes in behaviour were measured at a range of concentrations of lindane (1.8, 0.9 and 0.45 mg l-1) (Table 12) and atrazine (30, 20, 10, 5, 2.5, 1.25 and 0.625 mg l-1). Additional control measurements were made during the whole time course. Valve movements were recorded continuously for 24 h at each of the time points listed above.
| Table 12: | Effect of exposure to sublethal concentrations of lindane on the valve movement activity of the marine mussel, M. edulis, over 3 months |
| a The mean±SE of the duration of activity
(%) of 7 mussels bSignificantly different from the respective control (ANOVA with Dunnett’s test p≤ 0.001 |
|
At intermediate concentration (0.9 mg l-1) of γ-HCH the periodic activity of the mussels was markedly changed compared with control (Plate 1a). There was an obvious change in the relative length of active and rest periods; the duration of the activity shortened and rest periods occurred more frequently after 7 days of exposure (Plate 1b). The proportion of the activity at this intermediate concentration had decreased to less than 50 % of the control by 84 days (Plate 1c-1f). At the highest concentration (1.8 mg l-1) γ-HCH a closure response occurred within 3-4 h of the start of the experiment and the amplitude of valve opening was reduced. These responses became more marked at the experiment prolonged, by 3 months the animals were moribund and zero activity was recorded (plate 2a-2f). The effect of a range of concentrations of γ-HCH on the mean duration of active periods, with exposure time is presented in Table 12. The effects produced by γ-HCH on the duration of the mussel activity are time and dose dependent.
Chronic toxicity was monitored over a 4 month exposure using concentrations of 1.25, 2.5, 5, 10, 20 and 30 mg l-1 of atrazine. The results of chronic exposure on valve movements were presented in plates 3-4. After 60 days of atrazine exposure, the behaviour patterns in animals of the three lowest concentrations (5, 2.5 and 1.25 mg l-1) were all different from the control, but no differences could be distinguished between the patterns observed in the three individual concentrations (Plate 3c, 3d, 3f, respectively). The reaction of the mussels to a 24 h exposure to 5 mg l-1 was a high frequency opening and closing (Plate 3e). After 15 days treatment with 10 mg l-1 atrazin, the number of valve openings had fallen to 3-4 h-1 (Plate 4a) and decreased more by increasing the time of exposure (Plate 4b). Mussels in a concentration of 20 mg l-1 atrazine tended to exhibit prolonged resting periods (Plate 4c). It was clear that the closure periods increased with increasing length of exposure. After 14 days the percentage survival of mussels decreased to 30% (Plate 4d). The activity period of the mussels was only one hour in every 24 h and the opening time was characterised by low amplitude gape. In the mussels treated with 30 mg l-1 atrazine a reduction in the amplitude of the valve opening compared to the controls was observed after only 24 h exposure (plate 4e). After this 24 h exposure shell opening decreased markedly and mussel remained gaping and 60% of the mussels died.
In a preliminary study of the toxicity of atrazine in M. edulis, it was observed that the average number of byssal thread produced by each of a total of 30 mussels in a range of concentrations of atrazine (20, 10, 5, 2.5, 1.5 and 0.0 mg l-1) after a fixed exposure time of 24 h at 15°C were 3.3, 3.5, 47, 9.8, 9.9 and 14 respectively.
| Plate 1: | a; normal valve movement behaviour of M. edulis. Downward
deflections of the pen indicate valve closure, upward deflections indicate
valve opening. The major division on the time axis of the chart represents
one hour. b, c, d, e and f; response of M. edulis exposed to 0.9
mg l-1 lindane for 7, 14, 28, 56 and 84 days, respectively. |
| Plate 2: | Recording (24 h at each time point) of valve movement response
of M. edulis exposed to 1.8 mg l-1 lindane; a (after 14
days), b and c (after 28 days), d (after 56 days) and e-f (84-90 days). |
| Plate 3: | Real-time valve movement patterns of M. edulis exposed
to 1.25, 2.5 and 5 mg l-1 atrazine. a and b; after 14 and 28
days exposure to 1.25 mg l-1 atrazine, c and d after 60 days
exposure to 1.25 and 2.5 mg l-1 respectively. E and f after 24
h and 60 daye exposure to 5 mg l-1. |
| Plate 4: | Typical recording of valve movement patterns of M. edulis
exposed to 10, 20 and 30 mg l-1 atrazine. a and b represent 10
mg l-1 for 15 and 60 days, respectively. c and d represent 20
mg l-1 exposure for 24 h and 14 days, respectively. e; mussels
exposed to 30 mg l-1 for 24 h |
Also, the reduction in the number of the animals attached was observed with increasing lindane concentration despite being no obvious decline in the number of byssal threads formed. In lindane (0.1 mg l-1), all the animals attached to each other in a similar way to the control individuals. After 48 h exposure to 1 and 10 mg l-1 lindane, the percentage of experimental animals attached were 100 and 80% respectively and these had decreased to 50 and 0% respectively by 96 h. All individuals lost the ability to attach to each other after 48-h exposure to the highest concentration (15 mg l-1) of lindane.
Discussion
It is the first time to detect the chronic lethal concentration of lindan and atrazine for mussels. The threshold median lethal dose for lindane was approached at long elapsed times (90 days) when the LC50 was 1.7 mg l-1. The LT50 at a concentration of 30 mg l-1 was 8.4 days which appears to be approaching the limiting time. For atrazine the median lethal threshold concentration was not reached even at the lowest dose (1.25 mg l-1) used and after 112 days the LC50 was 1.6 mg l-1. Similarly a limiting time was not reached even at the highest does (30 mg l-1) when the LT50 was 1.7 days.
The concentration and time required to produce an adverse effect in the mussel vary from lindane to atrazine. For lindane there is a concentration or threshold level below which there was No Observed Effect (NOEL). But for atrazine there was a response at all concentrations with no threshold. One possible explanation of these observations is that more than one mechanism of toxicity may be involved, one short term and one long term.
Valve movement and byssal formation has been used for monitoring the effect of pollutants (Kramer and Botterweg, 1991; Rajagopal et al., 1997; El-Shenawy el al., 2001b). Most of the concentrations used in the present study of γ-HCH (1, 5, 10 mg l-1) had a significant influence on the periodic activity of the mussels after 12 h exposure. Considerable shortening of the active periods (85, 66 and 45%) was observed (Fig. 3) and this resulted in a decrease in the amount of water filtered by the animal and an associated depression of oxygen and food uptake. In turn this would lead to a fall in the rate of retardation of growth (El-Shenawy et al., 1999). Although atrazine and lindane may be expected to have different sites and modes of action, the secondary lesions resulting from the respective primary lesion have similar manifestation at the whole organism level. The whole animal symptoms measured in this work may act as general indicators of poisoning.
The results of the present study of the effect of lindane on the activity of the mussel were in agreement with those obtained from Salánki and Varanka (1978) in an investigation of Anodonta cygnea L (n = 9-10) for a week. In their study a commercial formulation, Hungaria L-7, of lindane was used. A significant shortening (by 50%) of the active periods was observed at a concentration of 6 mg l-1 Hungaria L-7 equivalent to 0.4 mg l-1 lindane on the survival and pumping behaviour of freshwater mussel Anodonta cygnea L. The shorting of active periods was used as indicator of the sub-lethal effects. Similar sensitivities were observed in the present study where valve movements of M. edulis were reduced to 55% of control levels after 96 h exposure to a concentration of 1 mg l-1 lindane, similar reductions in activity to 50.8% were observed following an 84 day exposure to 0.45 mg l-1 lindane.
The present study revealed that the importance of measuring both the concentration and time of exposure to the toxicants when assessing toxicity. The patterns of symptoms of poisoning with lindane and atrazine change with time and concentration and probably reflect the accumulation of toxicant and hence exposure of internal tissues, including any sites of action. In M. edulis, the inhibitory effects are manifested first of all in the shortening of active periods and characteristic changes in the frequency of rhythmic pumping activity and these observations has been previously reported by Jenner et al. (1992). Valve movement reaction to toxic substances can be distinguished as: (1) fast closure of the valves for minutes or hours, alternated with quick opening for water quality detection, after which, depending on type and concentration of toxic component, the valves may be closed again. (2) fast movement of the valves (not closing) for extra pumping capacity in an attempt to refresh the mantle cavity which is followed by a closing reaction when pumping fails to refresh.
The concentrations of lindane and atrazine necessary for 50% inhibition were considerably higher than the concentrations normally reached in highly polluted natural waters. However, these values should be used because of the limited time scale of laboratory based studies, but mimic the accumulation over longer times in mussels where exposure can be prolonged because of the slow decomposition rate of lindane. The presence of this compound in sediment and lower organisms from where mussels can assimilate it with the food enhances the uptake still further (Salánki and Varanka, 1978). Byssogenesis in M. edulis was decreased significantly by exposure to lindane or atrazine and by increasing the time of exposure due to decrease the pedal activity (El-Shenawy et al., 2001b).
In conclusion, it was clear from this study that M. edulis was more robust and easier to place in the environment for monitoring water quality. Moreover, atrazine appears to have two types of toxicity, one short-term operating over a few weeks and one long-term operating over months of exposure. The full time-dose response study undertaken here has defined toxicological parameters such as the median lethal threshold concentration and the slope of the dose response curve which are necessary to identify sub-lethal concentrations that can be used in studies of chronic toxicity. In short term experiments mussels proved to be insensitive to the low concentrations of the pesticides used but over longer exposure when accumulation of the toxicant mussels occurred the animals showed symptoms of poisoning. Also, Valve movement and byssal formation were early warning indicator for pollution.