INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

CADMIUM AND CALCIUM are suggested to share a common uptake pathway in the apical membrane of mitochondria-rich cells in fish gills (42, 44, 46). Exposure of freshwater fish to waterborne Cd²⁺ has been well documented to cause both ionic and osmotic imbalance in fish (7, 10, 13, 22, 32, 37). Interference with Ca²⁺ homeostasis resulting in hypocalcemia has been proposed to be the fundamental mechanism of Cd²⁺ toxicity (7, 35).

Verbost et al. (42) reported that Ca²⁺ influx in gills of adult rainbow trout (Salmo gairdneri) was inhibited by 0.1 µM Cd²⁺ after 16 h of exposure (in water containing 0.7 mM Ca²⁺). Our recent study indicated that, within 4 h of exposure, Ca²⁺ uptake in tilapia larvae (Oreochromis mossambicus) was competitively inhibited in water containing 0.18 µM Cd²⁺ and 0.2 mM Ca²⁺ (3). Little is known, however, about the mechanisms of Cd²⁺ inhibition on Ca²⁺ uptake in fish, especially about the sites where interference takes place. Mitochondria-rich cells in the gill epithelium of freshwater fish are believed to be the main site for Ca²⁺ transport (25, 29-31). Verbost et al. (43, 44) suggested that cytosolic Cd²⁺ would inhibit Ca²⁺ uptake by inhibiting the Ca²⁺ pump in the basolateral membrane of Ca²⁺-transporting cells, rather than by interfering with the entrance step at the Ca²⁺ channels in the apical membrane of the gills. They, therefore, hypothesized that this indirect effect of Cd²⁺ would block the Ca²⁺ channels by raising the cytosolic Ca²⁺ level. This effect was observed from the disturbed Ca²⁺ accumulation rate, accompanied by a decreased Cd²⁺ accumulation rate in Salmo gairdneri after a long-term (17 h) exposure to 0.1 µM Cd²⁺, whereas a short-term (4 h) exposure had no effect on either accumulation rate (44). If this hypothesis is applicable to other fish species, we could expect that rates of Ca²⁺ accumulation in tilapia larvae exposed to Cd²⁺ for 24 h or longer would decrease. Contrary to these expectations, we found that Ca²⁺ uptake in larvae was inhibited within 4 h of being transferred to a 0.18 µM Cd²⁺ solution and that Ca²⁺ influxes in larvae recovered compared with controls within 24 h (3). However, we did not investigate variations of Cd²⁺ accumulation rates after preexposure to Cd²⁺ or the impact of exposure to different concentrations of Cd²⁺ on rates of Ca²⁺ accumulation.

Inconsistent sensitivities to Cd²⁺ during development stages have been observed in many species of teleosts (13, 20, 27, 47). Newly hatched tilapia displayed altered sensitivities to Cd²⁺ throughout their larval development. The 96-h half-maximal lethal concentrations (LC₅₀) of 0- and 3-day-old larvae were 1.82 and 0.2 µM, respectively (13). This dramatic change in sensitivities to Cd²⁺ in larval stages is likely due to the different ion uptake efficiency, which, in turn, leads to a different extent of suppressing Ca²⁺ uptake during larval development (3). From our previous study, we know that 3-day-old larvae are the most sensitive, but in 96-h LC₅₀ tests, 0-day-old larvae managed to survive through the 3rd day and survived a Cd²⁺ concentration that would have killed 50% of the larvae at this stage (3 day old) within 4 days. Some acclimation process must have taken place in the larvae at the same time as ion uptake efficiency increased in the course of exposure.

Pretreating individuals with a sublethal dose of certain metals can lead to increased metal tolerance if they are subsequently exposed to a higher metal concentration (2, 17, 23, 34, 48). Therefore, if the acclimation process does exist, then larvae preexposed to a sublethal Cd²⁺ concentration for 3 days from hatching should have a higher tolerance than naive 3-day-old larvae.

In the present study, we first tested the hypothesis proposed by Verbost et al. (44) by comparing the accumulation rates of Ca²⁺ and Cd²⁺ in both Cd²⁺-preexposed tilapia larvae and naive 3-day-old larvae. Second, we examined potential differences in sensitivity to Cd²⁺ of Cd²⁺-preexposed larvae and naive 3-day-old larvae. Ninety-six-hour Cd²⁺ LC₅₀ tests, an index of acclimation to Cd²⁺, were conducted on 3-day-old larvae with or without preexposure to Cd²⁺. Third, we elucidated the underlying physiological basis of the above observation in tilapia larvae acclimated to waterborne Cd²⁺ by examining the effect of Cd²⁺ on Ca²⁺ influx kinetics after preexposing larvae to 0.18 µM Cd²⁺ for 3 days.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Fish. Mature adult tilapia, Oreochromis mossambicus, from the Tainan Branch of the Taiwan Fisheries Research Institute were reared in 182-liter glass aquariums with plastic chips as gravel. Each tank was supplied with dechlorinated, circulated, and aerated local tap water at 27-29°C under a photoperiod of 12-14 h light. Fish were fed with commercial fish food pellets (FwuSow). Fertilized eggs were collected from the mouth of a brooding female 1 day before hatching and incubated in a gently bubbled 1,000-ml container. Larvae were not fed during the experiments.

Cd²⁺ exposure. Completely dried CdCl₂ (Sigma) that had been dissolved in 1 ml concentrated HCl was used to prepare 8.9 µM Cd²⁺ stock solution with double-deionized water. The desired Cd²⁺ concentrations were diluted from the stock solution with local tap water as described by Hwang et al. (13). All containers used in these experiments were cleaned with HNO₃ and thoroughly rinsed with double-deionized water before use. The medium in the test containers was changed daily. Cd²⁺ concentrations and water chemistry in the exposure media were monitored daily. Fluctuations of Cd²⁺ concentration were <5%. Other parameters of the exposure media were (in mM, mean ± SD, n = 26): hardness, 0.28 ± 0.076 calculated as concentration of CaCO₃; dissolved oxygen, 0.23 ± 0.016; Na⁺, 0.24 ± 0.013; K⁺, 0.04 ± 0.002; Ca²⁺, 0.24 ± 0.007; Cl, 0.18 ± 0.005; Mg²⁺, 0.14 ± 0.008; pH 6.9 ± 0.3.

Ca²⁺ influx kinetics. Three tilapia larvae were introduced to each flux chamber after being rinsed in deionized water to remove extra ions from the body surface. For each sampling time, 48 flux chambers were used, representing eight different Ca²⁺ concentrations with three replications in both groups (0 and 0.18 µM Cd²⁺). Flux chambers were 100-ml polyethylene tanks filled with 50-ml tracer media. Artificial Ca²⁺-free solution (in mM: 0.3 NaCl, 0.03 KHCO₃, 0.15 MgSO₄, pH 6.7) was used to prepare two series of media with or without Cd²⁺ (0.18 µM). To these were added CaSO₄ to concentrations of 0.01, 0.02, 0.05, 0.10, 0.50, 1.00, 2.50, and 5.00 mM. Respective amounts of ⁴⁵Ca²⁺-labeled CaCl₂ (Amersham, 616 mCi/mmol) stock solution were added and adjusted to a specific activity of 600 µCi/mmol.

During the 4-h flux experiment, all flux chambers were immersed in a waterbath at the same temperature as the acclimation condition. At the end of the experiment, animals were removed, rinsed in nonradioactive fresh water three times for 1 min each, and killed by an overdose of anesthetic (MS-222, 0.38 mM). Carcasses were digested with 400 µl of liquid tissue solubilizer (Soluene 350, Packard) at 50°C for 3 h. The samples were then mixed with 2.1 ml of scintillation fluor (Hionic-Fluor, Packard) and counted in a scintillation -counter (1211 Rackbeta, LKB). Water samples (400 µl each) were also analyzed with 2.1 ml of scintillation fluor (Hionic-Fluor, Packard) and counted for ⁴⁵Ca²⁺.

The Ca²⁺ influx (J_in; nmol · mg¹ · h¹) was determined from the appearance of ⁴⁵Ca²⁺ radioactivity accumulated during the 4-h period in the whole body of each larva and was calculated according to the following formula where Q_larva is the radioactivity of larva [counts/min (cpm) per individual] at the end of incubation; X_out is the specific activity of the incubation medium (cpm/mmol); t is incubation time (h); and W is average body wet weight (mg) obtained from 10 larvae of the same brood. Preliminary experiments demonstrated that the radioactivity of ⁴⁵Ca²⁺ used in the present study does not have any significant effects on the development of larvae. Differences in the quenching effect between water and tissue were calibrated before calculating the data (14).

Ca²⁺ influxes at different external Ca²⁺ levels were measured as described above. The curves of Ca²⁺ influx in tilapia larvae fit the Michaelis-Menten equation, and values were subjected to the reciprocal Eadie-Hofstee plots to determine Ca²⁺ influx kinetics. The Eadie-Hofstee linear regression generally yielded a higher correlation coefficient (r) than did Lineweaver-Burk plots and therefore was used throughout to calculate the values of the Michaelis constant (K_m) and maximal velocity (V_max). These values were applied to the Michaelis-Menten equation The "actual Ca²⁺ influx", J_in, interpolated from this equation is defined in experiment 3 as the unidirectional Ca²⁺ influx for larvae acclimated at 0.2 mM Ca²⁺.

The values of K_m, V_max, and the actual Ca²⁺ influx with or without 0.18 µM Cd²⁺ were measured in the 4-h experimental period. At the end of this kinetics experiment, each larva represented one datum point. Because for each test there were three larvae in each flux chamber and they were not entirely independent; data from these three larvae were averaged. Therefore, the data per treatment were based on n = 3, instead of n = 9.

Body Ca²⁺ content and Cd²⁺ accumulation. Tilapia larvae were washed in Cd²⁺-free local tap water three times and briefly rinsed in double-deionized water, and water left on the body surface was blotted with filter paper. Each carcass was dried at 37°C overnight and digested in 200 µl HNO₃ (13.1 N) at room temperature (15). The digested solution was diluted with double-deionized water and analyzed by an atomic absorption spectrophotometer (Z-8000, Hitachi, Japan), using air/acetylene flame for calcium analysis and graphite furnace for cadmium analysis. Standard solutions from Merck (Germany) were used for making the standard curve. The standard addition method was used for background correction to eliminate the matrix effect.

Experiment 1: Effects of Cd²⁺ exposure on body calcium content and Cd²⁺ accumulation. Newly hatched tilapia larvae were transferred into 15 1,000-ml polymethylpentene containers and incubated with 0.18, 0.45, or 0.90 µM Cd²⁺ medium with five replications. A control group without Cd²⁺ added in the medium was set up at the same time. Individual body Cd²⁺ accumulation and Ca²⁺ content of five larvae from each container were recorded on the 3rd day of exposure. A total of 25 individuals from each treatment concentration were measured. On the 3rd day after hatching, larvae that had never been exposed to Cd²⁺ from the same brood were then introduced into 0.18, 0.45, and 0.90 µM Cd²⁺ medium with five replications, and body Cd²⁺ accumulation and Ca²⁺ content of larvae were measured in all treatment concentrations as above on day 4. At each sampling time, five larvae from each container were averaged, representing one datum point. Because larvae in each container were not entirely independent, the data per treatment were based on n = 5, rather than n = 25. The amount of Cd²⁺ accumulated in larvae during the last 24 h was obtained by subtracting the average Cd²⁺ content on day 3 from the average Cd²⁺ content on day 4. The same calculation was performed for the analysis of Ca²⁺ content.

Experiment 2: Acute Cd²⁺ challenge. Newly hatched larvae from the same brood were transferred into 1,000-ml polymethylpentene containers and incubated with a 0 (control) or 0.45 µM Cd²⁺ medium, respectively, for 3 days. No mortality was observed during this period of acclimation. On the 3rd day of incubation, the LC₅₀ (96 h) tests of both groups were conducted as described by Hwang et al. (13).

Experiment 3: Ca²⁺ influx kinetics in Cd²⁺ preexposed larvae. Newly hatched larvae from the same brood were transferred into 1,000-ml polymethylpentene containers and incubated with a 0 (control) or 0.18 µM Cd²⁺ medium, respectively, for 3 days. No mortality was observed during this period of acclimation. Next, Ca²⁺ influx kinetics were measured in both groups. Values of K_m, V_max, and actual Ca²⁺ influx at 0.2 mM Ca²⁺ were determined over a 4-h period with or without 0.18 µM Cd²⁺ added in the tracer medium. The "apparent influx" is defined as the Ca²⁺ influx measured in the 0.18 µM Cd²⁺-preexposed larvae under the interference of the subsequent external Cd²⁺ (11, 12) and the "true influx" as Ca²⁺ influx under a subsequent Cd²⁺-free condition.

Statistical analysis. Data are presented as means ± SD or means ± SE, whichever is appropriate. Results were analyzed by two-way ANOVA and Student's t-test. Multiple-comparison and linear regression analysis followed the method of least-square means. Probit analysis (SAS Institute, Cary, NC) was used for the 96-h LC₅₀ estimation. Statistical significance was accepted for P < 0.05.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Effects of Cd²⁺ exposure on body Ca²⁺ content and Cd²⁺ accumulation. Average body Ca²⁺ content in 3-day-old larvae never exposed to Cd²⁺ was 39.25 nmol per larva and the content increased to 63.5 nmol per larva 24 h later (Fig. 1, sum of the leftmost bars in both panels, 39.25 + 24.5 = 63.5). In those larvae exposed to 0.18 µM Cd²⁺ for both 72 and 96 h from hatching, their average body Ca²⁺ contents were not significantly different from untreated larvae. However, when the exposure concentration increased to 0.45 or 0.90 µM, both 72 and 96 h Ca²⁺ contents were statistically lower than that of the untreated larvae (Fig. 1). Only the last 24-h Ca²⁺ accumulation in larvae exposed to 0.90 µM Cd²⁺ for 96 h from hatching was statistically lower than the control larvae (Fig. 1A). However, when Cd²⁺ exposure started 72 h after hatching, the last 24-h Ca²⁺ accumulation in larvae of all three Cd²⁺-treated groups was always statistically lower than that of the control (Fig. 1A, hatched bars vs. open bar). Moreover, the last 24-h Ca²⁺ accumulation in larvae, no matter when the exposure started, was inversely related to the exposure Cd²⁺ concentration, and the values of Ca²⁺ accumulation within the final 24 h were always higher in larvae exposed to Cd²⁺ from hatching compared with naive larvae exposed from 72 h (Fig. 1A).

View larger version (31K): [in this window] [in a new window]   Fig. 1.   Changes in total body calcium contents in Oreochromis mossambicus larvae exposed to 0, 0.18, 0.45, and 0.90 µM Cd2+ at 0.2 mM Ca2+. A: Ca2+ uptake from 72 to 96 h after hatching. Open bar represents larvae never exposed to Cd2+. Solid bars denote larvae that started their cadmium exposure on day 0 (hatching). Hatched bars represent larvae that did not start exposure to Cd2+ until 72 h after hatching. ** P < 0.01, *** P < 0.001. B: Ca2+ contents at 72 h. Open bar represents larvae never exposed to Cd2+. Crosshatched bars are the larvae exposed to Cd2+ from hatching for 72 h. Bars that do not share letters are significantly different from each other at P < 0.001. Values shown are means ± SE (n = 5).

The increase in Ca²⁺ within the final 24 h in naive 4-day-old larvae (24.3 nmol per larva) was equivalent to 62% (24.3/39.3 nmol per larva) of total body Ca²⁺ content in 72-h-old larvae. Values were 62 to 74% for those larvae treated with 0.18-0.90 µM Cd²⁺ from hatching. However, when Cd²⁺ exposure started on day 3, it decreased from 37% for 0.18 µM Cd²⁺-exposed larvae to 4 and 1% for 0.45 and 0.90 µM Cd²⁺-exposed larvae, respectively.

As was expected, tilapia larvae exposed to 0.45 µM Cd²⁺ accumulated more Cd²⁺ than those exposed to 0.18 µM Cd²⁺ for 72 h (Fig. 2A) or 96 h [Fig. 2, sum of the values of crosshatched (Fig. 2B) and solid (Fig. 2A) bars at each Cd²⁺ concentration] from hatching (Fig. 2). However, the accumulation of Cd²⁺ in larvae exposed to 0.90 µM Cd²⁺ for 72 h was not significantly higher than that in 0.45 µM Cd²⁺-exposed larvae at 72 h of exposure (Fig. 2A), and on 96 h of exposure the total Cd²⁺ accumulation in 0.90 µM Cd²⁺-exposed larvae [Fig. 2, sum of the values of crosshatched (Fig. 2B) and solid (Fig. 2A) bars at each Cd²⁺ concentration] was even statistically lower than that in larvae exposed to 0.45 µM Cd²⁺ (P = 0.008). Naive fish exposed to 0.18-0.90 µM Cd²⁺ at 72 h after hatching accumulated about the same amount of Cd²⁺, 15.1-17.8 pmol per larva (Fig. 2A, crosshatched bars). The amounts of Cd²⁺ accumulation within 72-96 h in larvae exposed to 0.45 and 0.90 µM Cd²⁺ from hatching (Fig. 2A) were statistically higher than naive larvae exposed to Cd²⁺ from 72 h (Fig 2A, hatched bars), possibly because of change in Ca²⁺ influx kinetics (see below).

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Changes in Cd2+ accumulation in Oreochromis mossambicus larvae exposed to 0, 0.18, 0.45 and 0.90 µM Cd2+ at 0.2 mM Ca2+. A: Cd2+ uptake from 72 to 96 h after hatching. Bar on left represents the larvae never exposed to Cd2+. Solid bars denote larvae that started their cadmium exposure on day 0. Hatched bars represent the larvae that did not begin their exposures to Cd2+ until 72 h after hatching. *** P < 0.001. B: Cd2+ contents at 72 h. Bar on left represents the larvae never exposed to Cd2+. Crosshatched bars are the larvae exposed to Cd2+ from hatching for 72 h. Bars that do not share letters are significantly different from each other at P < 0.001. Values shown are means ± SE (n = 5).

Acute Cd²⁺ challenge. Tilapia larvae preexposed to Cd²⁺ developed higher tolerance to potentially lethal concentrations of Cd²⁺. The 96-h LC₅₀ of Cd²⁺ was estimated to be 0.19 µM in naive 3-day-old larvae. A 12-fold increase in 96-h LC₅₀ (to 2.28 mM) was found in tilapia larvae exposed to a sublethal Cd²⁺ concentration of 0.45 µM for 3 days immediately after hatching.

Ca²⁺ influx kinetics in Cd²⁺ preexposed larvae. Cadmium preexposure concentration, external Cd²⁺ concentration, and the interaction of these two factors had significant effects on both K_m and V_max of Ca²⁺ uptake and actual Ca²⁺ influx (Table 1). Without the presence of external Cd²⁺ in the tracer medium, the true K_m values in the Cd²⁺-preexposed larvae were determined by examining the effect of internally accumulated Cd²⁺ on the kinetics of Ca²⁺ influx. They were not statistically higher than those in naive 3-day-old larvae. The presence of external Cd²⁺ in the tracer medium resulted in a 17.4-fold increase of K_m in naive 3-day-old larvae (comparison 1 in Fig. 3A). Nevertheless, under the interference of external Cd²⁺ in the subsequent tracer medium, the apparent K_m values measured in Cd²⁺-preexposed larvae were not statistically higher than the true K_m values (comparison 2 in Fig. 3A).

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Comparison of Michaelis constant (Km) values (A), maximal velocity (Vmax) values (B), and actual Ca2+ influxes (Jin; C) at a 0.2 mM external Ca2+ level in Ca2+ uptake between naive and 0.18 µM Cd2+ preexposed 3-day-old Oreochromis mossambicus larvae under the interference of 0.18 µM Cd2+. Values shown are means ± SD (n = 3). ** P < 0.01, *** P < 0.001. Nos. above bars are different comparisons.

The true V_max and apparent V_max values measured in Cd²⁺-preexposed larvae were also not statistically different (comparison 2 in Fig. 3A), but both were significantly higher than V_max measured in Cd²⁺-free medium of naive 3-day-old larvae (comparisons 3 and 4 in Fig. 3B). However, the presence of external Cd²⁺ in the tracer medium caused a significant increase in the maximal Ca²⁺ transport rate in naive 3-day-old larvae (comparison 1 in Fig. 3B).

The presence of external Cd²⁺ in the tracer medium significantly decreased J_in in naive 3-day-old larvae (comparison 1 in Fig. 3C), but the true and apparent actual Ca²⁺ influxes calculated in the Cd²⁺-preexposed larvae were not statistically different (comparison 2 in Fig. 3C). Both these values were significantly higher than the actual Ca²⁺ influx measured in Cd²⁺-free medium of naive 3-day-old larvae (comparisons 3 and 4 in Fig. 3C).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Four major conclusions can be drawn from the present study. First, the inhibitory effect of Cd²⁺ on Ca²⁺ uptake is concentration dependent. Second, the concentration of Cd²⁺ that suppresses Ca²⁺ uptake does not inhibit Cd²⁺ accumulation. Third, Ca²⁺ influx in Cd²⁺-preexposed larvae is not inhibited by either external or internal Cd²⁺. Fourth, Cd²⁺-preexposed larvae have a higher tolerance to Cd²⁺ and higher Ca²⁺ uptake efficiency than naive 3-day-old larvae.

Effects of Cd²⁺ exposure on body Ca²⁺ content and Cd²⁺ accumulation. There was no significant change of total body Ca²⁺ content in larvae exposed to 0.18 µM Cd²⁺ from hatching. This can be explained from our previous finding that Ca²⁺ influxes in larvae may be restored to levels of their respective controls within 24 h of being transferred to 0.18 µM Cd²⁺ (3). Although we do not know exactly how long it would take for larvae exposed to 0.45 or 0.90 µM Cd²⁺ to restore their Ca²⁺ influxes, our data indicate that it is likely to be >24 h. However, it is apparent that the inhibitory effect of Cd²⁺ on Ca²⁺ uptake results in a concentration-dependent decrease of body Ca²⁺ content. Similar findings have been reported by Rombough and Garside (37) and Hwang and Yang (16).

The net calcium increase in the last 24 h of larvae exposed to Cd²⁺ on day 3 was significantly lower than that of larvae starting the exposure right after hatching. In other words, larvae pretreated with Cd²⁺ from hatching are acclimated to Cd²⁺ and the inhibition of high concentrations of Cd²⁺ on Ca²⁺ uptake is less than that on the larvae starting the exposure 3 days after hatching. The ratio of the amount of calcium accumulated within day 4 to the total body calcium content of larvae on day 3 is equivalent to the ionic uptake efficiency during the last 24 h of experiment. When exposure started on day 3, the uptake efficiency decreased significantly. This resulted in a significantly decreased total body Ca²⁺ content in larvae that started the treatment on day 3, whereas when the exposure started right after hatching, it did not change significantly (Fig. 1).

Instead of applying ¹⁰⁹Cd²⁺ for measuring the Cd²⁺ influx (44), the methodology used in the present study is based on the variation of Cd²⁺ content. The advantage is that we examined not only the average Cd²⁺ influx but the body Cd²⁺ accumulation at the same time. On day 3, the body Cd²⁺ accumulation in Cd²⁺-preexposed larvae was certainly higher than in naive larvae. According to the hypothesis proposed by Verbost et al. (44), naive larvae should have higher Ca²⁺ and Cd²⁺ accumulation rates than larvae preexposed to Cd²⁺, because the Ca²⁺ channels may be inhibited indirectly by the elevated cytosolic Cd²⁺ in the latter. However, our data indicate that the Cd²⁺ accumulation rate of Cd²⁺ preexposed larvae in the following 24 h was not lower than that of larvae that started the exposure on day 3, and the Cd²⁺ accumulation rate was not dependent on the Cd²⁺ concentrations exposed. That is, the uptake sites were likely saturated at the Cd²⁺ concentration higher than 0.18 µM. These results suggest that the exposure of Cd²⁺ that decreases the uptake of Ca²⁺ does not suppress the accumulation rate of Cd²⁺ at the same time. However, Verbost et al. (44) found that there was a 75% decrease of Cd²⁺ influx in rainbow trout (Salmo gairdneri) exposed to 0.1 µM Cd²⁺ for 17 h; with 1 µM Cd²⁺ in the water, an 80% inhibition occurred within 2 h of exposure. There may be two explanations for this contradiction. First, it took tilapia larvae a much shorter period of time than rainbow trout to acclimate to waterborne Cd²⁺. Second, obtaining sufficient Ca²⁺ is critically necessary for larvae, and simultaneous Cd²⁺ uptake can be described as a situation of compromising. The compromise is that fish need Ca²⁺ and cannot distinguish between Ca²⁺ and Cd²⁺. Fish preexposed to Cd²⁺ do better in maintaining Ca²⁺ homeostasis, presumably by altering Ca²⁺ uptake kinetics. By increasing Ca²⁺ uptake, fish preexposed to Cd²⁺ also accumulated more Cd²⁺, but, even so, they survive better (greater LC₅₀).

Increased tolerance to Cd²⁺. In the present study, increasing Cd²⁺ tolerance was observed in Cd²⁺-preexposed larvae. The K_m value of Ca²⁺ influx in Cd²⁺-preexposed larvae was not affected by waterborne Cd²⁺. In addition, their true K_m value, rather than the apparent K_m value [the value measured in the presence of Cd²⁺ (11, 12)], did not differ significantly from that of naive larvae. This suggests that the Ca²⁺ uptake in Cd²⁺-preexposed larvae is not suppressed by internally accumulated Cd²⁺. On the contrary, the actual Ca²⁺ influx at 0.2 mM Ca²⁺ in Cd²⁺-preexposed larvae was higher than that in naive 3-day-old larvae. The facilitated uptake for Ca²⁺ did not change even in the presence of external Cd²⁺. Pretreating individuals with a sublethal dose of metals usually leads to an increased metal tolerance if they are subsequently exposed to a higher metal concentration (2, 17, 23, 34, 48). Such acclimation phenomenon was also observed in the present study.

Three metal acclimation mechanisms have been proposed by McDonald and Wood (26). The first step is to reduce metal uptake by increasing mucus secretion or decreasing metal binding properties. The second step is to induce the production of a heavy metal binding protein such as metallothionein (MT). The third step is associated with decreased metal sensitivity by compensatory synthesis of ion-transporting ATPases. There is evidence that Cd²⁺-exposed rainbow trout had reduced uptake of Cd²⁺ by gill (44). Similar findings in Zn²⁺-acclimated juvenile rainbow trout were described by Hogstrand et al. (11, 12). One-half the control Zn²⁺ influx was measured in Zn²⁺-exposed fish that had been exposed to 2.3 µM Zn²⁺ for 15, 29, and 56 days. Contrary to these results, we found no reduction in the Cd²⁺ accumulation rate in tilapia larvae. Those larvae that had been exposed to Cd²⁺ for 3 days, did not accumulate less Cd²⁺ within the following 24 h, and in fact they accumulated more Cd²⁺ than did naive fish exposed to Cd²⁺. Verbost et al. (44) and Hogstrand et al. (11, 12) also demonstrated that the metal in the water inhibited the influxes of both the metal itself and Ca²⁺. In addition, Hogstrand and colleagues (11, 12) indicated that the influence of external Zn²⁺ on the actual Ca²⁺ influx was negligible, because the elevated K_m values (from 0.038 to 0.23 mM) they obtained from their studies remained much below the water Ca²⁺ concentration (1 mM). However, the influx of Zn²⁺ was reduced because of the decreased binding affinity. In other words, the reduction in binding affinity inhibits only the uptake of Zn²⁺ itself, and the Ca²⁺ transport rate could be maximized even under the interference of Zn²⁺. However, this was not the case in the present study. First, the elevated K_m values of 0.29 mM in 3-day-old larvae exposed to 0.18 µM Cd²⁺ were higher than the water Ca²⁺ concentration (0.2 mM). This reduce in affinity (high K_m) will lead to a decrease in calcium uptake. Second, the apparent K_m, V_max, and actual Ca²⁺ influx in Cd²⁺-preexposed larvae were not statistically different from their true ones. Both the true V_max value and true actual Ca²⁺ influx were significantly higher than their respective values of naive 3-day-old larvae, and these represented promoted Ca²⁺ uptake. This inconsistency between the observations of Hogstrand et al. (11, 12) or Verbost et al. (44) and our data is possibly due to the age of the fish, the sampling time of the experiments, or variations among metals and species. Furthermore, we have found that Ca²⁺ influxes in 0- and 3-day-old larvae could quickly be restored to their respective control levels within 24 h after transfer to 0.18 µM Cd²⁺ (3). This in not the case in Reid and McDonald's study (33) in which they examined the inhibitory effect of a 12-h Cd²⁺ exposure on the Ca²⁺ uptake of rainbow trout (Salmo gairdneri) and found that the inhibition could not be eliminated within another 12 h under a Cd²⁺-free condition. To meet the needs of sufficient Ca²⁺ uptake, which is critically important for a normal larval growth, this observation implies the superior modulation ability of tilapia larvae in response to the interfered calcium homeostasis.

As was suggested above, the most important criterion for determining Cd²⁺ toxicity (i.e., the survival of larvae) is the degree of interference with Ca²⁺ homeostasis in tilapia larvae rather than the absolute amount of Cd²⁺ accumulated (Figs. 1, 2). A similar suggestion was proposed by us earlier (13). Some intracellular response resistant to the toxic effect was suggested to serve as protection during heavy metal exposure (1, 8, 9). One such protective response has been proposed to be the production of certain proteins. MT, one of the most important metal-binding proteins, can sequester most of the intracellular free toxic metal ions by preventing their binding with functional proteins (18, 19, 28, 36). Induction of MT synthesis by pretreating fish with a low dose of metals, which increases the tolerance of subsequent metal exposure, has been well demonstrated (2, 17, 23). The expression of MT mRNA was concentration dependent in 0-day-old tilapia larvae after transfer to Cd²⁺ for 48 h (Hwang, unpublished data). Other protection mechanisms such as cytochrome P-450 and glutathione, involved in detoxification have been reported (40, 41). Moreover, some endocrine response correlated with the increased prolactin cell activity and the increased plasma cortisol concentration involved in the increased-tolerance tilapia has been reported (5, 6).

Almost all previous studies indicated that heavy metals, including Cd²⁺, exerted either a small or no inhibitory effect on V_max values (11, 12, 38, 39, 42, 43, 45). However, that is not the case in the present study. It is surprising to find the elevated V_max value in the presence of waterborne Cd²⁺. An interpretation could be that the larvae are somehow able to detect the decrease in Ca²⁺ uptake. Consequently, the larvae may increase in the number and/or activity of Ca²⁺ transporters to restore the normal Ca²⁺-transporting efficiency of the organism. An accelerated development of mitochondria-rich cells has been found in fish gills to compensate for the loss in ion uptake capacity caused by toxic Cd²⁺ (21, 22). Compensation for Ca²⁺ deficiency by increasing the binding affinity (lower K_m) and maximal transport rate (higher V_max) of Ca²⁺ was observed in tilapia larvae acclimated to the very low external Ca²⁺ medium within 1 wk (15). On the contrary, it took more than 10 wk for adult fish to reestablish a positive Ca²⁺ balance after transfer to a low-Ca²⁺ environment (4). The superiority of tilapia larvae in the modulation of Ca²⁺ uptake also seems true when the ion imbalance is caused after Cd²⁺ exposure. Compared with the previous studies on adult fish, tilapia larvae show a greater capability in modulating the physiological disturbance on Ca²⁺ influx induced by environmental interference such as Cd²⁺ and low Ca²⁺.