In this study, the amphipod Gammarus aequicauda was evaluated as a test organism for use in sediment toxicity bioassays. Sensitivity to noncontaminant variables, to the reference toxicants and to some contaminated field sediments was analysed. Amphipods were tolerant to various salinity and temperature combinations during a ten-day assay. The organisms tested with different type of diet showed highest survival on the natural diet.
The organism density effect on survival and growth of Gammarus aequicauda during a 28-day assay was examined. The results indicated that the density did not affect survival and production, but the effect of density was significant on the average weight.
No effect on survival was observed on three sediment types during the 10-day exposure. Sensitivity to contaminants was assessed using cadmium chloride and copper chloride as reference toxicants in a 96 h water-only test. Methods were developed for conducting a short-term toxicity test with cadmium chloride-spiked sediment using this species. Water-only testing revealed high sensitivity of amphipods to reference toxicants. Experiments conducted with organisms of four different size classes demonstrated no significant differences in sensitivity.
Sediment contamination in coastal areas is a major environmental issue because of its potential toxic effects on biological resources and often, indirectly, on human health. A large variety of contaminants from industrial, agricultural, urban, and maritime activities are associated with sediment particulates, including bottom sediments. Sediment toxicity bioassays have become more important for evaluating ecological hazards of toxicants in the aquatic environment. They can, for example, be used to determine the sediment toxicity of single chemicals and chemicals mixtures (ASTM, 1996), the potential adverse effects of toxicants on benthic marine organisms, the magnitude and the spatial and temporal distribution of pollution impacts in the field (Long, 2000). For a marine sediment toxicity test a number of properties are required: indigenous and abundant intertidal species with broad salinity tolerance, sensitivity to common sediment contaminants, occupation of microhabitat to ensure a consistent exposure to sediment contaminants, low sensitivity to natural sediment types, such as grain size and organic matter, short life cycles, wide geographic distribution, ease of collection, handling and maintenance in the laboratory, ecological relevance and year-round availability (Chapman, 1988; DeWitt et al., 1989; Smith and Logan, 1993; Ingersoll, 1995).
Amphipods are known to be among the most sensitive species to toxic substances in acute tests and are employed in sediment toxicity tests (Anderson, 1982; Williams et al., 1985; ASTM, 1992; Luoma and Ho, 1993). Several standard methods have been developed for assessing the toxicity of contaminants using amphipods species from Atlantic and Pacific coasts, but few tests have been developed with European species, more specifically in the Mediterranean (Cesar et al., 2002).
In this study Gammarus aequicauda was evaluated as a test organism for use in sediment toxicity bioassay because this species fulfils many of properties listed above. It is a very widespread and frequently recorded species in the Mediterranean and Black Seas (Greze, 1977; Janssen et al., 1979). Although it is considered a marine species, being found most frequently and at higher abundance in sea water, it may penetrate brackish waters. It lives all over the intertidal zone to about 20 m depth, but it is more abundant in shallow waters among macroalgae (Chaetomorpha linum, Ulva ssp.) which provide food and protection from predation for the amphipod. It also feeds upon sediment detritus, pieces of organic material or by scraping the surface of mineral particles (Kevrekidis and Koukouras, 1989). In muddy sediments, G. aequicauda burrows into the sediment surface. Its intimate contact with bottom sediment and overlying waters for extended periods of its life cycle increases the probability for adverse effects occurring in the presence of contaminated sediments. It is important both in the passage of food to other organisms and the natural regeneration of organic matter in underlying sediment (Cottiglia et al., 1983).
The objective of the present research was to determine the sensitivity of G. aequicauda to some non-contaminant variables such as temperature, salinity, grain size, density organism, type of diet as well as sensitivity to a reference toxicant. The data obtained from these tests were integrated in order to define an experimental protocol for acute sediment toxicity test with G. aequicauda.
Materials and methods
Collection of animals
Sediment samples were sieved in the field through 1000-700-500 and 150 μ m screens; the amphipods were grouped in four different size classes (adults, sub-adults, juveniles and newborns) according to mesh size. The amphipods were rinsed into polyethylene buckets containing seawater and immediately transported to the laboratory and placed in aerated glass containers with their native sediment. Table 1 shows the general characteristics of the native sediment. Experimental organisms were acclimated for 3 to 4 d before tests were started. The general design of each test was based largely on the standard guides for conducting acute sediment toxicity tests with marine-estuarine amphipods (ASTM, 1992; SETAC-Europe, 1993).
Sensitivity to non-contaminant variables
The general design of the assays is summarised in Table 2. The parameters tested were: temperature, salinity, organism density, type of diet, particle size.
Temperature and Salinity: this experiment examined effects of four temperatures and five salinity combinations on survival. The temperatures tested were 7, 15, 18, 20°C. Saline solutions were prepared by diluting the water from the reference site (S‰ = 36) with distilled water to obtain 25, 15, 3, 0‰ salinities solutions. No gradual adaptation of the animals to salinity and temperature was performed. Tests were carried out in 1-l glass beakers containing approximately 2 cm of sediment and 800 ml seawater. Twenty amphipods were randomly selected and introduced into each beaker. The exposure time was 10 d with three replicates for each treatment.
Type of diet: the effect of the diet type on survival and growth over a 28 d period was determined. The procedure was based on Costa et al. (1996). The three diet treatments tested were a macroalgae Chaetomorpha linum, Tetra Min® (dried diet) and 1:1 mix of these. Twenty amphipods were placed into each beaker, containing a 2 cm layer of sediment and 800 ml seawater, with three replicates per diet treatment. The water renewal (80% of volume) and food supply ad libitum, took place biweekly.
Organism density: this test is based to verify the effect of organism density on survival and growth of G. aequicauda. The densities tested were 10, 20 and 40 animals per beaker (about 2.5, 5 and 10 individuals dm−2). Five replicates were used for each treatment. The exposures were carried out with a sediment layer about 2 cm and small stones were introduced in each beaker (1 stone per 10 animals). During assays the amphipods were feed with Chaetomorpha linum on ad libitum basis. Amphipods from each treatment were pooled and collectively weighed at the beginning and at the end of the 28 d exposure period. Water replacement and food supply took place at 5 d intervals. At the end of the experiment the sediment was sieved through a 500 μ m sieve and survivors were removed. Production was calculated as the proportional increase in the biomass (Final biomass − Initial biomass/ Initial Biomass). One-way ANOVA analysis was performed with all the data obtained at the end of the experiment.
Sediment grain size and organic matter content: this experiment examined survival of G. aequicauda to different sediment grain size and organic matter content. Sediments were prepared by mixing the required proportion of FF (fine fraction) of clean organically enriched sediment with the native sediment. Sediment samples prior to testing were analysed for grain size and organic matter content. Total organic matter (TOM%) was estimated as percentage of weight lost after ignition of dry sediment at 550°C for 4 h. The natural sediment used in this experiment contained 0.85% TOM and was enriched with the fine fraction of the natural sediment, in different proportions: 0.5, 25, 75 to 100% fine fraction corresponding to 1.91, 2.04, 3.49, and 4.21% of organic matter, respectively. Five replicates for each sediment types including the control sediment were tested. The exposure time was 10 d.
Effects of the sediment geochemical properties on survival were tested by one-way ANOVA.
Sensitivity to contaminants
The general design of the assays is summarised in Table 3. Before sediment testing was initiated, a water-only toxicity test was conducted to determine the sensitivity of G. aequicauda to two reference toxicants: cadmium chloride (CdCl2) and copper chloride (CuCl2) (De Witt et al., 1992; USEPA 1994; ASTM, 1997). These controls, also designated as ‘positive controls,’ consisted in the determination of 96 h LC50 for toxicant in the seawater in the absence of sediment. The water used in the experiments consisted of natural sea water (36‰), filtered through a GFC Whatman® (0.45 μ m) filter. Four different concentrations of 0.2, 0.4, 0.8 and 1.6 mg l−1 were used for CdCl2, five concentrations: 0.4, 0.8, 1.6, 3.2 and 6.4 mg l−1 for CuCl2. One seawater control was used, with four replicates for each treatment. The temperature was maintained at 16 ± 2°C and the amphipods were not provided with additional food during the 96 h exposures.
Differences in the sensitivity to CdCl2 between different size classes (adults, sub-adults, juveniles and newborns) were examined. The cadmium and the copper LC50 values, based on the lethal concentration at which 50% did not survive, were determined by the Spearman–Karber method (Hamilton et al., 1977).
Cadmium in the sediment
The sensitivity of G. aequicauda to Cd in the natural sediment (0.85% TOM) was evaluated by an acute toxicity test (10-day) using three nominal contents: 4.4-8.8-13.6 mg per dry kg sediment.
The spiking technique described in this study is based on the method reported in Swartz et al. (1985) and Costa et al. (1998) with some modifications. The required volume of a stock solution of CdCl2 was directly added to the sediment in order to achieve the three nominal concentrations. The Cd solution was dripped in the sediment, which was then mixed by rolling the beakers at about 25 rpm for 2 h. Then the beakers were removed from the roller machine and stored over night at 4°C, to allow partitioning of the toxicant into the sediment. The control sediment consisted of unspiked sediment.
Twenty amphipods were added in random to the beakers, about 24 h after the Cd solution was initially added to the sediment. All treatments had four replicates for concentration. At the end of the experimental period, the sediment was sieved through a 500 μ m sieve and survivors were counted. Missing organisms as well as those that did not move even after gentle stimulation were as dead.
An acute 10-d sediment toxicity test was performed to evaluate the sensitivity of G. aequicauda to field sediments. Four sampling sites were selected, two of them (stations 1 and 2) were located in the First Inlet of Mar Piccolo in vicinity of pulp mill industries, these sediments are known to be contaminated with a mixture of toxicants, one in the Second Inlet of Mar Piccolo (station 3) and one in the Gulf of Taranto (station 4) chosen as a reference station since this area is far from anthropogenic source of contaminants. Four replicates were prepared per treatment, twenty young adult amphipods were randomly selected and introduced into the test chambers. At end of exposure time the survivors were counted and missing organisms were considered dead.
The bioassay was considered valid if the mortality in the control sediment was < 20%, while the sediment was considered toxic if mortality significantly differs from mortality of the control (t-test; p < 0.05) (Ennas et al., 2002).
Table 4 shows the mean amphipod survival for combinations of temperature and salinity on 10-day assay. The tolerance of G. aequicauda to various levels of salinity and temperature revealed that survival was high (98%) at the salinities of 15–25–36‰ at 18°C. However, for the salinities of 0 and 3 ‰, survival decreased to 32.5% and 40% at 15°C respectively, but was still high at 20°C (80% and 90%). Therefore this species may be considered as euryhaline and such as other species of gammaridae showed different tolerance limits (Neuparth et al., 2002; Costa et al., 1998). Such species preferred a range of 18 to 20°C and 15 to 36%, depending on according to the water temperature and salinity range of the sampling sites.
There was a difference in the effect of diets on the survival of G. aequicauda at 28 d exposures. Survival was high (90%) for all treatments with the macroalgae Chaetomorpha linum (fresh diet) and low (60%) for dried diet (Tetra Min®) and 1:1 mixture of the two diets.
In the density experiment, survival was generally high (80–90%), however the highest values were found at 5 and 10 individuals dm−2. One-way ANOVA indicated that the density did not affect survival (df = 11, p > 0.05), although a significant effect was observed on average weight increase (df = 11, p < 0.05). For production, the effects of density were not significant (df = 5, p > 0.05) (Figure 1).
The one-way ANOVA showed no significant differences in survivorship among three sediments, ranging in texture from sand to clay, at the end of 10 days assay (df = 5, p > 0.05) (Table 5). The percent of survival in these tests was high, ≥85%.
The 96 h nominal LC50 of Cd and Cu in seawater were 0.71 mg l−1 (95% confidence intervals: 0.44–1.14 mg l−1) and 0.82 mg l−1 (95% confidence intervals: 0.53–1.28), respectively. Mean percentage of survival in the controls of all experiments (n = 8) was high (97.1%; SD = ± 0.81)
In addition, experiments were carried out using organisms of four different classes (150–500–750 and 1000 μ m) to investigate body-size related differences in sensitivity for Cd. The LC50 values obtained showed that new borns and juveniles were more sensitive than adults and sub-adults (Figure 2).
The 10 d LC50 value of the Cd-spiked sediment was 5.62 mg per dry kg sediment (10.96–2.88 mg per dry kg sediment) for the sediment with 0.85% TOM.
The results of experiments conducted for evaluating the sensitivity of G. aequicauda to field sediments showed that it did not tolerate sediments from contaminated sites near the pulpmill industries in which mortality was very high (51.6 ± 1.52 to 68.3 ± 1.15%) compared with mortality in the control sediment (Table 6). The tested samples were significantly toxic (p < 0.05) compared to the control mortality in the Gulf of Taranto sediment that was 6.6 ± 0.58%.
Contaminant intolerant species include most amphipod crustaceans which are highly sensitive to toxic chemicals (DeWitt et al., 1988; Swartz et al., 1994; Costa et al., 1996, 1998; Peterson et al., 1996). Since amphipods are dominant members of all major marine and freshwater assemblage systems, they are very good candidates for ambient reference indicators. Standardized sediment toxicity tests have been developed for burrowing amphipods, including Rhepoxynius abronius (Swartz et al., 1985), Leptocheirus plumulosus (Schlekat et al., 1992), Monoporeia affinis and Pontoporeia femorata (Eriksson-Wiklund, 2002) and tube-dwelling species, including Ampelisca abdida and Grandidiella japonica (Kohn et al., 1994), and Corophium volutator (Ciarelli, 1994; Ciarelli et al., 1997; Conradi and Depledge, 1998). Corophium orientale is one of the few Mediterranean amphipods which have been examined for sediment toxicity testing (Onorati et al., 1999).
The results of the present paper have provided information about a free-living epibenthic amphipod, G. aequicauda, as a potential test species in the sediment toxicity assessment. Gammarus aequicauda was chosen for many reasons. It is readily available and easy to handle. Also, it shows a high tolerance to different salinities, temperatures, sediment textures yet is sensitivity to the reference toxicants. Mean survival of G. aequicauda was 99.4% at 18°C in salinities of 15–25–36‰ after 10 d exposure. According to Grosse et al. (1986) the adult estuarine gammarids are fairly tolerant to a wide salinity range.
Grain size is considered a ‘confounding factor’ because it can be stressful to amphipods interfering with their ability to burrow and remain in contact with the sediment and cause additional mortality. In this study mean survival of G. aequicauda for all sediment treatments was high (91%) after 10 d. Nevertheless, further reference controls are necessary when the tested sediments are outside of the range used in this study.
Sensitivity to the fine fraction (clay content) was reported for R. abronius, a phoxocephalid amphipod used in marine sediment toxicity. Its mortality was attributed to the adhesion of very fine particles to the gills or to the feeding appendages (DeWitt et al., 1988). Schlekat et al. (1992) reported a mean survival of 97% for L. plumulosus after 10 d in sediment ranging from sand to clay. Survival of E. estuarius was 92.4% in sediments with 80% silt-clay content (DeWitt et al., 1989).
The appropriate organism density is one of the factors to consider when developing toxicity tests (Buikema and Benfield, 1979; DeWitt et al., 1992). It is necessary in order to establish the species optimum condition and good quality of assurance for the statistical analysis of the results. The test results can be compared with G. locusta (Costa et al., 1996) assuming the density of 7.5 ind.dm−2 as optimum.
Effects of different diets on the survival of this species are important in toxicity assays, the major mortality of G. aequicauda cultured with Tetra Min suggest that it does not like a dried diet, but this species can be conveniently cultured with a natural diet. In Mar Piccolo it is always associated with macrophytes, and in particular with drifting mats of the green alga Chaetomorpha linum, these formations seem to be advantageous for the amphipod, because the algae protect it against predation and it can find plenty of food there (E. Prato, pers. observ.).
The sensitivity of G. aequicauda to aqueous Cd is within the range of values reported for other amphipod species proposed for estuarine or marine sediment toxicity tests. According to the LC50 (0.71 mg Cd l−1 and 0.82 mg l−1), it appears to be a highly sensitive species, since data on amphipod exposure to Cd in water indicate a range of LC50 values from 0.19 to 11.4 mg l−1 (DeWitt et al., 1992; Schlekat et al., 1992). Little information was found about the toxicity of Cu in seawater, Costa et al. (1998) reported for G. locusta a 96-h LC50 of 0.3 mg l−1.
Sensitivities also differ within the same organisms depending on life-stage. Smaller G. aequicauda were more sensitive to Cd. Robinson et al. (1988) reported differences in sensitivity (to CdCl2) in relation to various life-stages for R. abronius, and found that the early life stage was more sensitive than the adults. On the other hand, Ciarelli et al. (1997) reported no significant differences in sensitivity between Corophiumvolutator of different size classes.
In this study, the results of Cd-spiked sediment (0.85% TOM) showed that G. aequicauda responded in a dose-dependent manner in acute exposures. No background information was found of published data concerning G. aequicauda sensitivity to Cd.
Previous benthic community studies, in Mar Piccolo estuary, revealed a significant decrease in number and abundance of species in the proximity of pulp mill industry (Prato et al., 1999, 2000). Results obtained on the field sediments from these sites confirmed that this species may be suitable for use in sediment bioassays in order to assess the quality of sediments. Survival was very low in sediments from areas of urban growth and developed industrial complexes that have relatively high toxic chemical content (E. Prato, unpubl. data).
In conclusion, this species fulfills most of the criteria required for suitable sediment toxicity tests. It may be an alternative or complementary test species for ecotoxicological studies in Mediterranean ecosystems. However additional experiments should be carried to examine the species sensitivity to natural factors called ‘false positives’ or ‘confounding factors’ such as high ammonia or sulphide, that cause mortality and can lead to at inaccurate conclusions during assays.