Daphnia magna is widespread in many freshwater systems of temperate regions and frequently is used to test metal toxicity. Recently, studies have been performed to determine metal biokinetics and development of tolerance in this important zooplankton species. In the present paper, we review the recent progress in these areas and suggest possible directions for future studies. Substantial differences exist in aqueous uptake, dietary assimilation, and elimination of several metals (Cd, Se, Zn, Ag, Hg, and MeHg) by D. magna. The routes of uptake are metal-specific, with Se and MeHg being accumulated predominantly through diet. All metals except Ag can be biomagnified from algae to D. magna, providing that metal concentrations in algae and algal food density are relatively low. Methylmercury is biomagnified in all situations. As a route for metal elimination in D. magna, maternal transfer is especially important for Se, Zn, and MeHg. On the other hand, the effect of single-generation exposure to metals on D. magna is very different from multigeneration exposure, which often results in a significantly higher metal tolerance. Moreover, D. magna easily loses metal tolerance developed through long-term exposure. Recovery from metal stress can temporarily increase the sensitivity of D. magna to metal toxicity. Finally, metallothionein-like protein is responsible for minimizing metal toxicity in D. magna. The results inferred from these studies can be extrapolated to other aquatic invertebrates as well as to other pollutants in the aquatic environment.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Daphnia magna is a freshwater zooplankton that is widespread throughout many lakes in temperate regions and often is used in aquatic ecotoxicity testing of metals by measuring the concentrations at which adverse effects are observed. Most of these toxicity studies have focused on the aqueous phase of metals, although a limited number have characterized the toxic effects of dietborne metals to D. magna [1,2] or other zooplankton species [3,4]. Quantification of different biokinetic parameters (e.g., aqueous uptake, dietary assimilation, and elimination) and the employment of an appropriate biokinetic model can be very useful in predicting metal accumulation in D. magna as well as in delineating the relative importance of aqueous versus dietary uptake [5–8]. Metals that have been systematically characterized for biokinetics in D. magna include Cd, Se, and Zn [9–11] as well as Ag  and Hg and MeHg [13,14].
Organisms inhabiting metal-contaminated environments may develop elevated tolerance toward metal toxicity . Recent research efforts have focused on characterizing the development of tolerance for metal toxicity by D. magna. In general, the development of metal tolerance is attributed to the induction of metallothionein-like protein (MTLP) [16,17], which may render metals unavailable to cellular receptors. Studying the biokinetics of metals also may help us to understand the underlying mechanisms in metal tolerance. Therefore, it can be informative if the measurement of metallothionein induction is coupled with biokinetic characterization to further understand whether the tolerance change is a result of alteration in biokinetics or other subtle factors.
Daphnia magna has a relatively short life cycle; therefore, it is cost-effective to use D. magna to investigate how multigeneration metal exposures with or without recovery (i.e., no metal added) can differ from single-generation metal exposure. Metals that have been studied for multigeneration exposures include Cu [18–20], Cd [21–23], Zn , Ni [25,26], and Hg [27,28]. Moreover, daphnids are intermediate in the aquatic food chain, located between the primary producer (e.g., algae) and the secondary consumer (e.g., carnivorous fish), and are the key organisms in trophic transfer of metals to freshwater fish. Chen et al.  found that the Zn and Hg concentrations in macrozooplankton (including daphnids) are predictive of those concentrations in fish in different lakes, indicating the trophic transfer of these metals in the systems.
In the present paper, we review recent progress in characterizing the metal biokinetics and tolerance development in D. magna. Results are gathered from previous studies and are synthesized to summarize our current understanding of metal uptake, elimination, and trophic transfer in D. magna. Also, we assess the results of research on the tolerance of D. magna to metals and, most importantly, the significance of multigeneration acclimation and exposure to metals. Finally, we discuss the implications of these studies for the field of metal ecotoxicology and point out some directions for possible future research.
BIOKINETICS OF METALS
In general, four major processes determine metal concentration in D. magna: Aqueous uptake, dietary uptake, elimination, and growth. Under steady-state conditions (i.e., when the uptake is balanced by the loss), metal concentration (Css) in D. magna can be estimated as follows [5,6]:
where ku is the uptake rate constant (L/g/d), which is a product of dissolved uptake efficiency and filtration rate; Cw is the aqueous metal concentration (μg/L); AE is the dietary assimilation efficiency (%); IR is the weight-specific ingestion rate (g/g/d); Cf is the metal concentration in diet (μg/g); ke is the elimination rate constant (1/d); and g is the growth rate constant (1/d). The term (kuṁCw) represents the aqueous uptake, (AEṁIRṁCf) the dietary uptake, and (ke + g) the elimination. Growth rate (g) may be ignored for adult D. magna, because maximum body size has already been reached . It should be noted that this one-compartment model has been used for both invertebrates and fish [5,30]. For example, concentrations of Cd, Zn, Ag, Co, and Se in field-collected marine copepods were reasonably predicted using experimentally determined biokinetic parameters in the laboratory . Luoma and Rainbow  recently reviewed this model and successfully predicted metal concentrations in diverse aquatic organisms with more than seven orders of magnitude of difference in metal concentrations. In another study, the dynamics of Cd accumulation in D. magna was validated by the biokinetic model even under nonsteady-state conditions . By monitoring the biokinetic parameters from birth to adulthood, the model simulations can accurately predict the experimentally observed temporal Cd accumulation in daphnids. Furthermore, two factors played critical roles in affecting Cd accumulation: Cd partitioning between the aqueous and dietary phases, and growth rate (g). The growth rate constant was comparable to the Cd elimination rate constant (ke) in daphnids and could significantly influence Cd accumulation because of its variation in different developmental stages 
Table Table 1.. Uptake rate constants (ku) of different aqueous metals determined for Daphnia magna during an 8-h uptake period
Concentration during exposure (nM)
Relationship between I (nmol/g/h) and Cw (nM)
aI = influx/uptake rate of metal; Cw = aqueous metal concentration.
The short-term aqueous uptake of Cd, Se, Zn, Ag, Hg, and MeHg by adult daphnids has been characterized [10,12,13]. Also, long-term aqueous metal uptake has been addressed [32–34]. Brief exposure (e.g., 8 h) can minimize the recycling of excreted metals by D. magna. Monson and Brezonik  found that D. magna neonates reached steady-state aqueous MeHg uptake within 20 h. To prevent metal binding to food particles, food was not added to the exposure media during the uptake period . Also, D. magna needs to be gut-evacuated (e.g., 2 h) to minimize fecal materials in the exposure media. The bioconcentration factor (BCF; L/kg) of metals in the daphnids was obtained over time, and the influx/uptake rates of metals (I; nmol/g/h) were then computed for each specific metal concentration. The uptake rate constants (ku; L/g/d) were calculated over a range of experimental metal concentrations. Variations of ku for the studied metals were more than two orders of magnitude, ranging from 0.187 L/g/d for Se to 11.04 L/g/d for MeHg (Table 1). Such variations may be caused by the rate of metal transport across the epithelial membranes in daphnids. It generally is assumed that ku is independent of the ambient metal concentration, although it is possible that aqueous uptake may reach saturation at a very high concentration . In the equation describing the relationship between I and Cw (Table 1), the power coefficients generally are close to unity, with the exception of Se .
The ku of a metal can be influenced by many ambient factors. For example, increasing dissolved organic C concentration and decreasing water hardness were found to elevate the ku of Cd in D. magna . Another example of an increase in the ku of Cd, Se, and Zn in D. magna was caused by an increase in pH and a decrease in Ca concentration in the ambient water . Also, the ku of Ag was reduced if the Na concentration in the ambient water increased, likely caused by the Na ions competing with Ag ions for binding sites on the animals [12,37]. Aqueous metal uptake by D. magna possibly was through direct surface adsorption, as evidenced by determining the total metal content in soft tissue and exoskeleton after the uptake experiment. For Se, Hg, and MeHg, however, tissue was the dominant site for metal uptake [10,13]. For Cd and Zn, approximately 50% of metal absorption was associated with the exoskeleton after short-term aqueous metal uptake .
Assimilation, elimination, and biomagnification
The AEs of Cd, Se, Zn, Cr, Ag, Hg, and MeHg in D. magna were determined using a pulse-chase feeding technique , and these values were significantly dependent on the food C concentrations in the diet [9,11–13]. To accommodate the high variability of metal AEs, Reinfelder et al.  suggested using a range of AEs instead of a single correct AE value. Over a range of food density expressed as C concentration (0.1–10 mg/L), metal AEs were negatively correlated to food density in the ambient water. Linear regression analyses were performed for the seven metals using log-log relationships (Fig. 1). Similarly, C assimilation in zooplankton was negatively correlated to the food density [38,39]. It was hypothesized that higher food density caused changes of filtration activity and reduced gut passage time for both food C and metals and, therefore, that these elements were less efficiently assimilated . By comparing the slope (s) of the linear regressions between AE and food concentration for different elements (Fig. 1), the AE of Ag is most dependent on food density, whereas the AEs of Zn, Hg, and Cr are intermediately dependent on food density. In contrast, the AEs of Cd, Se, MeHg, and C are least dependent on food density. It is interesting to note that the AEs of C do not increase further when the food density drops below 0.1 mg C/L , implying that 0.1 mg C/L is close to the optimal handling capability of food by the digestive systems of D. magna.
Among different metals, AEs can vary from as much as 98% (MeHg) to as low as 1% (Ag). One explanation for such enormous difference is that metal desorption from food particles in the gut of daphnids can dominantly control the metal AEs. If so, metals that have rapid desorption rates should have relatively high AEs. An alternative explanation is that certain metals can cross the gut lining in greater quantity and at higher rates than others. It is difficult to determine which of the two processes (i.e., desorption from food within the gut and absorption across the gut lining) is the limiting process in controlling metal AEs. Some clues, however, can be found by contrasting the AEs of Hg and MeHg in D. magna . Both forms of Hg were reported to possess similar lipophilicity (i.e., ability to cross the gut lining) when complexed with chloride ions ; therefore, more rapid desorption of MeHg from the food particles may be responsible for its higher AE as compared to that of Hg. Mason et al. , however, demonstrated that the AEs of Hg and MeHg in a marine copepod were significantly correlated with their percentages in the diatom cytoplasm. Similar phenomena were observed for other metals and elements in marine copepods feeding on the diet . To our knowledge, such a cytosolic digestion hypothesis has not been tested for D. magna and freshwater algae. Moreover, the AE of metals in D. magna has not been tested on other food types (e.g., bacteria and ciliates); therefore, further studies are required to quantify these parameters.
Apart from the food density effect, the AEs of metals can be influenced (though to a lesser extent) by metal concentrations in the diet. It has been demonstrated that the AEs of Cd, Se, Ag, Hg, and MeHg significantly decreased with increasing metal concentrations in the diet [11–13]. This relationship was not found for Zn, however, presumably because these animals can regulate Zn uptake. These studies revealed that algal intracellular metal content decreases with increasing metal concentration, which partially explains the decrease of metal AEs with an increase in dietary metal concentration (e.g., Cd) . Other factors, such as nutrients and ambient temperature, did not significantly affect the AEs of metals, including Cd and Zn  and Hg and MeHg . When D. magna fed on P-rich algal cells, however, the AE of Se increased .
The elimination rate constants (ke; 1/d) of Cd, Se, Zn, Cr, Ag, Hg, and MeHg were determined for adult D. magna following either aqueous or dietary metal accumulation. Within each route of accumulation (i.e., aqueous and dietary), the ke values did not differ significantly with food density, ambient temperature, animal age, or metal concentration in the diet [2,9,11–14]. A trade-off, however, exists in the metal elimination in D. magna to maintain a relatively constant ke for specific metals under different conditions. For instance, at a higher food density, the reproduction rate (i.e., more energy reserve) was higher, but the Se content in each offspring decreased. Therefore, the overall Se elimination remained constant for both low- and high-food treatments .
The ke determines the metal loss from the animals after metal is incorporated into the tissues . Several processes contribute to metal elimination in D. magna, including excretion into the aqueous phase, formation of feces, deposition into molt, and transfer to offspring. Planktonic animals, such as cladocerans and copepods, generally have a ke one order of magnitude higher than those of large benthic bivalves  and play important roles in regenerating metals in the surface water.
The trophic transfer factor (TTF) of a metal can be calculated as follows :
A TTF of greater than one indicates that the metal has the potential to be biomagnified in D. magna through diet. We simulate the likely biomagnification potential of different metals over a range of concentrations of C in water and dietary metals. Because the food density influences the AE, but not the ke, only a single ke value was used for modeling. Moreover, factors such as dietary metal concentration (Cd, Se, and Zn ) and ambient temperature (Hg and MeHg ) do not significantly affect the ke of metals in D. magna. At earlier stages, however, D. magna might eliminate metals faster than the adult daphnids (Ag ). In this model, the IR was set at 0.5 g/g/d to indicate possible biomagnification, but the IR may increase with increasing food density in water .
As shown in Figure 2, the variation of AE is large for the seven metals (e.g., 4–67% for Zn). The elimination is the slowest for both Hg and MeHg, followed by Cd, Cr, Se, Zn, and Ag. The high AE and low ke caused MeHg to be definitely biomagnified, a well-known phenomenon in nature [46,47]. The diminution of Hg in Figure 2, however, conforms to the field-collected data reported by Watras et al. . For lakes in Wisconsin (USA), the Hg concentration in the microseston was 170 ng/g, whereas that in the zooplankton was 29 ng/g (i.e., biodiminution) . The MeHg concentrations in the microseston and zooplankton, however, were 34 and 53 ng/g. respectively (i.e., biomagnification) . In natural waters, food density can vary from less than 1 to 30 mg C/L . Therefore, metal AEs can be highly variable depending on the specific food density. From Figure 2, all metals except Ag can be potentially biomagnified from the diet to D. magna if the AEs are at the high end (i.e., low metal and food density conditions). Nonetheless, the diminution in Ag from food to D. magna is mainly caused by its very high ke (i.e., 0.36/d) . It should be emphasized that one of the most important metals, Cu, was not measured for AE in D. magna. This was because of the lack of a suitable radioisotope for such purpose. In one of the limited studies, however, the AE of Cu in a marine copepod (Acartia spp.) ranged from 40 to 50% . Overall, this laboratory-derived information is important in understanding metal movement in the lower trophic level, which currently is much less understood compared to the higher trophic level .
Pathways of uptake and elimination
Apart from quantifying the rates of metal uptake and elimination, it also is important to quantify the relative importance of metal uptake pathways in D. magna. Such information is essential in developing a realistic design for metal toxicity testing, because aqueous and dietary metals can have contrasting toxic effects [1,3,4,50]. The fractions of metal coming from the aqueous phase (fW) and from the dietary phase (fD) can be determined by the following equation :
The bioconcentration factors of metals in phytoplankton are different for different metals because of the metal-specific partition coefficients in phytoplankton in the natural environment. As shown in Figure 3, aqueous metals important for accumulation in D. magna include Hg, Cd, Ag, and Zn (i.e., >50% in most cases), whereas the important dietary metals are Se and MeHg. The major reason for the sole importance of diet for Se accumulation, which is in good agreement with the results of previous studies concerning marine bivalves  and copepods , is its very low uptake rate from the aqueous phase . The fraction of Cd coming from the aqueous phase of metal accumulation in D. magna (60–70%) is similar to the findings of other laboratory experiments that directly determined the relative importance of water and diet to Cd uptake in D. magna  and marine copepods . To our knowledge, no experimental data regarding Cu exist for D. magna, but for this species' counterpart, marine copepods, Chang and Reinfelder  modeled and predicted that more than 75% of Cu comes from dietary sources. In comparison with marine copepods, it was found that aqueous uptake of metal generally is more important for D. magna , possibly because of their large ratio of surface area to volume. Also, adsorption to the exoskeleton is a fast process (i.e., <1 h) and, therefore, contributes significantly to aqueous metal uptake in D. magna [30,56,57].
A recent important finding concerns the relative importance of different routes in eliminating metals from adult D. magna (i.e., excretion, egestion, molting, and reproduction). In general, excretion into water represents a dominant pathway for all metal loss in D. magna; however, the situation is essentially unknown regarding the forms of metals excreted by D. magna. Organic matter complexation may affect the bioavailability of metals and, thus, their recycling in the waters . In general, egestion and molting are only minor routes for metal loss [9,30,56,59]. Metals in the exoskeleton can come from both direct adsorption and metabolic deposition [30,59], but the latter process is much slower. Metals depositing into the exoskeleton often are associated with calcareous compounds .
Most importantly, recent studies show that the offspring can carry a significant portion of maternal metal, of which the amount is metal-specific. Both Se (˜35%) and MeHg (˜30–40%) are maternally transferred to offspring. Other metals, such as Hg (˜10%) and Zn (˜25%), have moderate maternal transfer potential. Only limited maternal transfer for Cd, Ag, and Cr (<10%) occurs in D. magna [9,12,13]. Because Se is an essential metal, maternal transfer of Se can be beneficial to the offspring. In contrast, maternal transfer of MeHg to offspring would translate into multigeneration exposure of such toxic metals in the subsequent generations . Moreover, the amount of metal maternally transferred is directly proportional to the amount of metal that is present in the mother .
Maternal transfer represents an additional route for D. magna to eliminate metals, and to speed up the elimination of certain toxic metals. For example, Tsui and Wang  demonstrate that the comparable ke values of both Hg and MeHg (i.e., 0.052 vs 0.051/d) in D. magna result from the fact that reproduction contributes greatly to the elimination of MeHg. In contrast, previous studies of mussels and fish found that reproduction is not important in eliminating metals within a relatively short experimental duration [61,62]. In experiments with aqueous metal accumulation, aqueous metals can adsorb directly onto the eggs in the brood chamber. Therefore, metals in the offspring do not necessarily come from their mothers . Indeed, it can be difficult to distinguish between metals coming from the mother and metals coming from direct adsorption from the aqueous phase.
Overall, the biokinetic model with experimentally determined biokinetic parameters is shown to adequately predict metal concentrations in marine clams (Se ), copepods (Cd, Zn, Ag, Co, and Se ) and mussels (Cd ) collected in the field. Therefore, the biokinetic model can account for most of the important factors governing metal accumulation in aquatic invertebrates in the environment . For D. magna, Guan and Wang  demonstrated that this model accurately predicts Cd accumulation throughout its life span. The next step is to verify if experimentally determined biokinetic parameters can adequately predict metal tissue concentrations in nature.
DEVELOPMENT OF TOLERANCE TO METAL TOXICITY
In an early study, D. magna pre-exposed to Cu at 10 μg/L for 20 h achieved resistance to acute Cu toxicity . Such resistance results from laboratory pre-exposure and can be regarded as acclimation (i.e., physiological changes) instead of adaptation (i.e., genetic changes). Later, it was found that D. magna pre-exposed to Cd and Zn synthesized MTLP, which was accompanied by an elevated tolerance to Cd toxicity . Metal exposure very often causes induction of MTLP in fresh-water invertebrates , including D. magna [66–68], oligochaetes , and clams . Metallothionein-like protein plays important roles in protecting against metal toxicity by sequestering toxic metals (e.g., Cd and Hg) and controlling the availability of essential metals (e.g., Cu and Zn) to other cellular binding ligands. In addition, Xie and Klerks  demonstrated a reduced Cd uptake at 6 mg/L in Cd-tolerant fish compared to control fish. The reduction of uptake by pre-exposed organisms also was observed for D. magna in Hg accumulation, but the uptake reduction only occurred at lethal concentrations. At Hg concentrations below the lethal range, however, control and pre-exposed daphnids had similar Hg uptake rates . One possibility was that the animals developed some defense mechanisms to reduce the Hg uptake when the incoming Hg was lethal to the animals.
Table Table 2.. Parameters and observations that can be determined in single-generation and multigeneration metal exposures using Daphnia magna
a For multigeneration exposure only.
Growth (length, weight)
Acute toxicity (concentration-, time-based)
Short-term aqueous uptake
Assimilation efficiency from food
Elimination rate constant
Metallothionein-like protein levels
Metal body burden
Cellular energy allocation
Effect of recovery from metal stressa
Changes over generationsa
For D. magna, MTLP is the main detoxification pool to sequester toxic metals, such as Hg. Other subcellular pools, including metal-rich granules, which are important for certain mollusks [31,73], may not be important in D. magna. In general, roughly 30 to 70% of accumulated Hg and Cd would be partitioned into the probable MTLP pool in D. magna after reaching a steady-state condition. The remaining metals were distributed in other subcellular fractions, such as heat-sensitive proteins [22,27]. Redistribution of metals is possible, however, during metal uptake and depuration. Indeed, the MTLP pool in D. magna for metal sequestration is far more important than in other aquatic invertebrates, such as marine mussels, in which metals in the MTLP pool normally are less than 20% .
The parameters that can be measured and quantified using D. magna in pre-exposure studies are summarized in Table 2. Because D. magna has a short life span, multigeneration acclimation is feasible in laboratory experiments. Indeed, D. magna offers a fascinating subject for studying the influences of contaminants on multigeneration exposure and toxicity, and we anticipate more studies in this field in the future.
In a Cd pre-exposure experiment, both Cd concentration and pre-exposure duration were found to greatly determine the MTLP induction in D. magna within a single-generation exposure . With a constant duration of pre-exposure (3 d), both the MTLP level and the Cd body burden in D. magna increased significantly with increasing ambient Cd concentration. Above a threshold Cd concentration (5 μg/L), however, MTLP concentration did not increase until after 1 d of exposure, which indicates that Cd accumulation is immediate but also that MTLP synthesis is a response to the metal accumulation . Similarly, Hg pre-exposure resulted in significant MTLP induction in D. magna . Moreover, the capability of MTLP to protect the organism depends not only on the absolute MTLP concentration but also on the distribution of metals in the MTLP pool. A significantly higher fraction of Hg was found in the MTLP pool with a higher Hg pre-exposure in D. magna , suggesting that MTLP becomes more important in detoxification at a higher metal pre-exposure (or in a more contaminated environment). Furthermore, previous studies showed that the ratio between metal and MTLP in D. magna was a better predictor of toxicity than both metal body burden and absolute MTLP concentration , because if more metal than available MTLP binding sites is available, spillover of the metal occurs, which causes cellular toxicity .
The majority of studies have focused on the changes in resistance to metal toxicity in D. magna after pre-exposure, but a limited number have quantified the alteration of metal biokinetics. After pre-exposure to Cd at 20 μg/L for 3 d, the aqueous uptake rates (I) of Cd increased by 50%, whereas the AE from ingested algae decreased by 55% . The elevated uptake was mainly caused by the induced MTLP, which may provide additional ligands for binding Cd in tissue. The metals taken up, however, may not be available to other cellular ligands, because they are sequestered by the MTLP. In contrast to Cd pre-exposure, Hg pre-exposure only slightly reduced the aqueous uptake (by 28%) and the AEs (by 12%), whereas the MTLP level and the acute tolerance to Hg toxicity increased significantly . Therefore, the effects of pre-exposure on the metal biokinetics in D. magna are metal-, concentration-, and duration-specific.
It is interesting to note that the aqueous uptake rate (I) of metals increases after metal pre-exposure (e.g., Cd) while, at the same time, tolerance to metal toxicity increases. At a first glance, this seems to contradict the expectation that a higher tolerance should accompany a decrease, not an increase, in I. More incoming metals in the pre-exposed organisms, however, can be partitioned to the MTLP pool in D. magna. Therefore, even though more metals are taken up, the pre-exposed organisms suffer less from metal toxicity than unexposed organisms do. In contrast to I and AE, the ke of both Cd and Zn were not significantly affected by Cd pre-exposure in D. magna . The uptake and elimination processes are uncoupled, because pre-exposure only affects the former process in D. magna. Therefore, it can be inferred that the changes in certain biokinetic parameters, such as I, AE, and IR (Eqn. 1), can result in changes in the relative importance of uptake pathways (i.e., aqueous vs dietary) (Eqn. 3) and also the trophic transfer factor (Eqn. 2) of metals.
Regarding the tolerance development of metal toxicity, a number of studies found that after short-term exposure to sublethal metal concentrations, D. magna developed higher resistance to acute toxicity for some metals, including Cu , Cd [67,68], and Hg [27,28]. These studies, however, employed metals at concentrations in the pre-exposure media well above environmentally realistic levels. Bossuyt and Janssen  demonstrated that D. magna gained Cu tolerance in the laboratory during acclimation to environmentally realistic Cu concentrations (i.e., 0–100 μg/L). Therefore, it is important in future studies to determine threshold metal concentrations as well as the minimum duration for the development of tolerance to metal toxicity.
Most of the pre-exposure studies with D. magna employed waterborne exposure with clean algae provided as food. The metals spiked in the media would partition to the algal surface and partly contribute to dietary metal uptake [1,35]. Indeed, there can be a major difference in tolerance development because of the pre-exposure of metals through aqueous and dietary phases. Canli  showed that D. magna neonates acclimating to aqueous Zn (1 μM) and algae-loaded with Zn (1 μM in algal media) for 4 d achieved a higher 48-h median lethal concentration (LC50) for Zn (71 μM) compared with those acclimated to higher aqueous Zn (50 μM) or to dietary Zn only (60 μM). The 48-h LC50 of Zn was the lowest for the control animals (48 μM). By comparing the aqueous Zn–only and dietary Zn–only treatments, it can be concluded that the dietary Zn promotes higher Zn tolerance. Therefore, pre-exposure studies with different metal uptake routes can possibly yield different results and have important implications for future studies.
During multigeneration exposure of metals, D. magna are exposed to sublethal metal concentrations for several successive generations. Recovery from metal stress can be tested to understand the persistence of metal toxicity tolerance and altered metal biokinetics simply by transferring newly reproduced individuals from the metal-spiked solution to the clean water. Overall, multigeneration exposure is used to mimic long-term contamination in the environment . The majority of multigeneration exposure studies have focused on the toxic effect and the development of tolerance by the animals to the metal toxicity. Other studies have determined changes in metal uptake and biokinetics.
In a recent study , D. magna were exposed to Cd at 3 μg/L for six successive generations. Both Cd biokinetics and physiological performance (e.g., MTLP, IR, growth, and reproduction) were measured for the groups recovering from Cd stress at the third and fifth generations . As shown in Figure 4a, the MTLP concentration in the exposed group (in which neonates and developing embryos were continuously exposed to Cd) was almost twofold that of the unexposed group for the second generation. It increased to fourfold in the unexposed group from the fourth to sixth generations. Presumably, this surge of MTLP may result from both long-term Cd acclimation and handling stress during the experiments (e.g., renewal, food addition, and animal transfer). In Figure 4b, the AEs of Cd were relatively constant during the first four generations but then decreased significantly concurrent with a decrease of IR (i.e., feeding depression) (Fig. 4c). Moreover, when D. magna recovered from Cd stress in the fifth generation, IR increased significantly, but MTLP decreased .
In a study of fish, it was demonstrated that the mRNA of MTLP in a Cd-tolerant fish can be maternally transferred to the next generation . Maternal transfer of MTLP itself, however, was not observed in a previous study with D. magna continuously acclimating to Hg . Whether transfer of MTLP mRNA occurs in D. magna is unknown. Although it was shown for D. magna that metal tolerance for Hg  and Cd  can be maternally transferred, this is not the case for all metals. For example, after chronic Ni (which is not an inducer of MTLP) contamination, the offspring from exposed D. magna did not show significantly higher Ni tolerance when compared to the offspring from the unexposed group .
In previous studies with D. magna, it has often been observed that individuals recovering from metal stress lost metal tolerance (e.g., Cd and Hg), showed reduced MTLP levels, and had poorer performance (i.e., growth and reproduction) than their continuously acclimated counterparts [23,28]. When D. magna that had been acclimated to Hg were transferred into clean water, they lost their Hg tolerance gradually over life stages, as quantified by time-to-death assays . Therefore, tolerance development through acclimation in D. magna is only a physiological acclimation, and removal of the metal stress soon results in the loss of such tolerance. Indeed, individuals recovering from metal stress had MTLP levels lower than those of individuals with continuous metal acclimation and, thus, became more sensitive to acute metal toxicity . These results suggest that cleanup of a metal-contaminated site may temporarily reduce the metal tolerance. Moreover, organisms in a system with fluctuating metal exposures (i.e., episodic exposure) may be more vulnerable than those in a system with constant metal exposure. This is certainly an interesting topic requiring further study.
The effects of multigeneration exposure with essential and nonessential metals can be very different on D. magna. The former is required for normal metabolism; therefore, deficiency of these metals (e.g., Cu, Zn, and Se) can have a great impact . With five generations of exposure to Zn at different ambient concentrations (3–800 μg/L), daphnids grew and reproduced significantly better than the unexposed counterparts when acclimated at 300 to 450 μg/L of Zn . Such response was observed only after the first generation of Zn acclimation. Similarly, for daphnids acclimated to Cu at 0.5 to 100 μg/L over three successive generations, Cu tolerance increased only after the second and third generations . This also occurred for nonessential metals; for example, acclimation of Cd for eight generations significantly increased the 48-h LC50 for daphnids from 61 to 180 μg/L, a finding that may be more related to genetic adaptation . These results suggest that multigeneration exposure can help us to understand the long-term acclimation to metals. In contrast to what has been observed with essential metals, hormesis (i.e., stimulated animal growth or other physiological performance) was observed in multigeneration exposure studies with Cd  and Hg . Similarly, the response for these nonessential metals was observed only after at least one generation of metal acclimation.
In addition to MTLP induction, energy reserves in acclimated daphnids can be another important factor for the enhancement of metal tolerance. Muyssen and Janssen  found that zooplankton acclimated at optimal Zn concentrations for seven generations had significantly higher amounts of cellular energy allocation, which is a summation of carbohydrate, protein, and lipid contents. Similarly, an increase in cellular energy allocation was observed in daphnids acclimated to optimal Cu concentrations after 14 generations  but also after only three generations of exposure in another, separate study . Thus, the response was greatly dependent on the generation of the animals. Indeed, Canli  demonstrated a direct relationship between energy reserves and the tolerance of D. magna to acute metal toxicity (e.g., Zn). Therefore, cellular energy allocation is an important basis for tolerance development in D. magna. In addition, previous studies suggested that other end points, such as whole-body adenosine triphosphate and glycogen, were affected in D. magna exposed to sublethal concentrations of metals .
When a parental generation of daphnids ingested MeHgladen algae to a lethal level, a high mortality rate was observed among the parents, and their offspring performed poorly even without further exposure to MeHg, which implies a transfer of metal toxicity through generations . Similarly, Pane et al.  revealed that daphnids exposed to Ni for two successive generations reproduced offspring with lower tolerance to Ni toxicity than the unexposed individuals. The number of offspring produced by the exposed daphnids was similar to that produced by the control daphnids, which indicates a reduced fitness of the offspring of the Ni-acclimated mothers. Such a phenomenon also was demonstrated in studies with D. magna exposed to a nonmetallic insecticide (diazinon) . The nutritional status of parents is greatly affected during exposure to a toxicant at sublethal levels; therefore, their offspring may be smaller in size or of poorer quality in terms of macromolecular stoichiometry compared with normal daphnids . This rarely is considered in reproduction tests with D. magna .
Different clonal strains of D. magna can have different tolerances to metal toxicity, but the underlying mechanism is not well understood. One study revealed that variation in the MTLP level was a cause for the difference in Cd tolerances between two clones of D. magna . Through analyzing amplified fragment-length polymorphisms of Cd-acclimated daphnids over eight successive generations, Ward and Robinson  demonstrated a significant decrease in genetic diversity compared to unexposed daphnids.
In contrast to the studies performed in the laboratory, Lopes et al.  showed that daphnids (Daphnia longispina) from metal-contaminated waters were significantly more tolerant (both at lethal and sublethal levels) to a mixture of toxic metals than the daphnids collected from reference sites, regardless of whether or not these animals were acclimated in the laboratory before the tests. This strongly suggests that animals metal-adapted for many generations would not easily lose their tolerances compared to the metal-acclimated individuals. More studies are needed to examine the differences between zooplankton collected from contaminated and clean environments.
In recent years, an intensive research effort has been made to develop a biotic ligand model of various metals (e.g., Cu, Zn, and Ag) for fish and Daphnia sp. . The major pathway for metal uptake is water; therefore, a need exists for toxicity testing with water-only chemicals. In realistic situations, metals are partitioned into a particulate phase, such as living phytoplankton cells, and then taken up by feeding zooplankton as part of their diet. Therefore, a model that incorporates diet-borne metal toxicity along with aqueous metal toxicity, such as a biotic ligand model, can be one step forward for environmental science. The bioenergetic-based kinetic model for predicting metal accumulation in D. magna has its own merits and, moreover, the potential to be modified for predicting metal toxicity for this highly sensitive zooplankton species .
For pre-exposure studies, it would be interesting to determine whether MTLP induced by one metal can infer protective effects against exposure to another metal. Currently, this is poorly understood for D. magna. LeBlanc  showed that D. magna pre-exposed to Cu did not develop tolerance to the acute Zn and Pb toxicity. Another study demonstrated that Cd-acclimated daphnids were more tolerant of Pb toxicity but more sensitive to phenol toxicity . A study of marine mussels showed that pre-exposure to one metal can affect the aqueous uptake rate for another metal, but such influence was metal-specific . In addition, the interrelationship is complicated. Daphnia magna can be a suitable candidate for addressing these research questions and can be employed in a factorial design to test the intermetal influences. It also will be very informative if specific metallothionein in D. magna can be isolated and identified to determine the specificity of metallothionein in protecting the organisms against metal toxicity. Future studies of the biokinetics and toxicity using D. magna should focus on areas such as mixtures of similar and dissimilar chemicals (see, e.g., ). Mixtures of metals can yield toxicity results significantly different from those predicted by the additive model .
It is necessary to study the biokinetics and multigeneration exposure of metals in freshwater zooplankton other than D. magna, including other species of Daphnia. A recent toxicity study  found that D. magna did not perform similarly to other cladoceran species, in which case D. magna was the most tolerant species to both Cd and Zn toxicity. Other subjects for this particular study included Daphnia pulex, Daphnia ambigua, and Ceriodaphnia dubia . It is presumed that D. magna has a unique survival strategy, possibly based on metal biokinetics. Therefore, research should be expanded to examine other zooplankton species in terms of biokinetics and tolerance development to better predict metal concentration and development of tolerance in the field. Whether D. magna is representative of other zooplankton commonly found in nature is a subject of future research. Finally, the results from D. magna studies with metals can be extrapolated to other aquatic invertebrates as well as other classes of pollutants in the aquatic ecosystems. We conclude that studies regarding the metal ecotoxicology of D. magna continue to be exciting and offer many opportunities for both fundamental and applied research.
Our research described in this review was supported by several Competitive Earmarked Research Grants from the Hong Kong Research Grants Council (HKUST6097/02M, HKUST6405/05M, and HKUST6420/06M) to W.-X. Wang.