INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS AND MATERIALS RESULTS DISCUSSION REFERENCES

METALLOTHIONEIN (MT) is a low-molecular-weight protein (6,100-7,100 in mammals) with high sulfhydryl content (16). MT shows high affinity for groups IB and IIB metal ions, thereby playing a pivotal role in cellular processes of metal handling and detoxification (15, 29). MT is also involved in different metabolic processes that depend on zinc and copper (6), but a growing body of evidence suggests that the role of MT is not confined to heavy metal homeostasis and exchange.

MT is inducible by heavy metals (15) but also, to various extents, by different chemicals and stressors and, in particular, by oxidants (4). Accordingly, it has been found that the mammal MT gene promoter contains metal response elements (15, 35) and glucocorticoid response elements (17) but also genetic elements responsive to oxidant agents (8, 9). More recent studies have demonstrated that the MT gene promoter of fish shows regulatory sequences that are similar to those of mammals (27, 30).

The evidence that oxidant compounds can promote MT gene activation is in line with the possible involvement of MT in the cellular antioxidant defense system (see reviews in Refs. 18, 31). Such a view is based on different findings: 1) MT shows a remarkable in vitro scavenging activity against different free-radical species (19, 36); 2) the cellular level of MT can be increased by different oxidants (4); 3) cultured cells overexpressing MT can resist oxidative stress (32), whereas cells deficient in MT are more sensitive to oxidants (22); and 4) MT seems able to prevent free-radical injury to biological structures both in vitro (1) and in vivo (26). However, despite a bulk of data arguing for MT-based antioxidant defense, controversy still exists due to contrary observations indicating that the knowledge about the actual antioxidant role of MT is still incomplete (31).

In marine invertebrates, knowledge about MT responses to stress is by far more restricted than in vertebrates. However, bivalve mollusks and mussels in particular are known to accumulate high heavy metal amounts that produce MT induction and eventually heavy metal detoxification (5). Moreover, it has also been shown that strong variations of temperature, oxygen, and salinity can increase the cellular concentration of MT (39). In this study, mussels have been pretreated in vivo with Cd and then, after a period of detoxification, whole animals or isolated cells have been exposed to prooxidant compounds. Previous research has shown that mussels preexposed to Cd for a few days and then detoxified for 1 mo show an increase in MT, which binds most of cytosolic Cd (40). Hence, this experimental system, involving in vivo MT induction followed by oxidative stress, has allowed us to study the possible antioxidant role of MT at the molecular, cellular, and organism levels.

	METHODS AND MATERIALS

TOP ABSTRACT INTRODUCTION METHODS AND MATERIALS RESULTS DISCUSSION REFERENCES

Chemicals. Cytochrome c, EDTA, glutathione reductase, reduced glutathione (GSH), oxidized glutathione (GSSG), HEPES, hypoxanthine, leupeptin, phenylmethylsulfonyl fluoride (PMSF), NADPH, pronase, and xanthine oxidase were from Sigma-Aldrich (Milan, Italy). The oxyradical probe dihydrorhodamine 123 (DHR) and the LIVE/DEAD kit were from Molecular Probes (Eugene, OR). All other reagents were of analytical grade.

Animals and in vivo treatments. Specimens of mussels (Mytilus galloprovincialis, 5- to 6-cm shell length), purchased from a mussel farm (La Spezia, Italy) in March-April, were acclimated for 3 days in aquaria containing recirculating synthetic seawater (1 liter/animal), pH 7.9-8.0, 35 (20) at 15°C. Mussels were exposed for 7 days to Cd (200 µg/l) and then transferred to an unpolluted sea area within cages for a 28-day detoxification period. Subsequently, control and Cd-exposed/detoxified animals were treated with Fe (300-600 µg/l) for 3 days in aquaria. Metal additions and seawater changes were made daily, and Cd was used as CdCl₂ and Fe as FeCl₃. After treatments, digestive glands were rapidly dissected out and immediately used or stored at 80°C until use.

Heavy metal determinations. Digestive gland tissues were lyophilized, and digestion was carried out in a microwave oven according to the procedure described by Stripp and Bogen (34). Heavy metal concentrations were determined by atomic absorption spectrophotometry with flame for Fe (Perkin Elmer 3110) and with a graphite furnace for Cd (Perkin Elmer 4110 ZL). The analytical procedure was checked using standard reference material (mussel tissue, BCR 278) provided by the Commission of the European Communities, Community of Bureau Reference. The results are expressed as micrograms per gram dry weight.

Evaluation of oxyradical production in tissue homogenates. Digestive gland tissue was ice-cold homogenized (1:10) in 0.32 M sucrose, 20 mM HEPES (pH 7.4), 1 mM MgCl₂, 0.5 mM PMSF, and 2 µM leupeptine using a Potter-Elvehjem glass/Teflon homogenizer. Homogenates were centrifuged at 20,000 g for 20 min at 4°C. Aliquots of the supernatant (166 µg protein) were incubated with 30 mM HEPES (pH 7.2), 200 mM KCl, 1 mM MgCl₂, and 16 µM DHR in a total volume of 2 ml. DHR is a nonfluorescent derivative that is converted to the fluorescent dye rhodamine 123 on reaction with reactive oxygen species. Probe fluorescence signal was detected by using a Perkin Elmer LS 50B spectrofluorometer [excitation (ex.) = 505 nm, emission (em.) = 534 nm, slit = 2.5 nm].

MT determination. MT was evaluated on digestive gland homogenates according to a previously described spectrophotometric method (41).

Malondialdehyde determination. Digestive gland tissue was ice-cold homogenized (1:4) in 30 mM Tris-HCl (pH 7.4), mixed 1:1 with acetonitril, and centrifuged at 6,000 g for 20 min at 4°C. Malondialdehyde (MDA) concentration in the supernatant was assayed by HPLC as previously described (38).

Antioxidant enzyme assays. Digestive glands were homogenized (1:5) in 0.5 M sucrose and 0.15 M NaCl in 0.02 M Tris-HCl, pH 7.6. The homogenate was centrifuged at 500 g for 15 min at 4°C and the resulting supernatant at 12,000 g for 30 min at 4°C. The 12,000-g pellet (peroxisomal and mitochondrial fraction) was resuspended in a small volume of homogenization buffer. The 12,000-g supernatant (cytosolic fraction) was chromatographed on a Sephadex G-25 column (Pharmacia PD-10); equilibration and elution buffer were the same as the homogenization buffer. Both fractions were used immediately for enzyme assay. Enzyme activities were measured in a Beckman DU 70 spectrophotometer under conditions of saturating substrate concentration at the temperature of 25°C, as described by Lemaire et al. (23). Superoxide dismutase (SOD) and catalase (Cat) were assayed on both the mitochondrial fraction and the cytosolic fraction, glutathione peroxidase (GPX) was assayed in the cytosolic fraction only.

Cat activity was determined by following the decrease in absorbance at 240 nm, due to H₂O₂ consumption (extinction coefficient of 40 M¹ · cm¹). Final reaction mixture (1 ml) was 50 mM H₂O₂ in 50 mM KH₂PO₄/K₂HPO₄, pH 7.0 (13).

SOD activity was evaluated with the xanthine oxidase-cytochrome c method as described by McCord and Fridovich (24). The cytochrome c reduction by superoxide anion radicals generated by the xanthine oxidase-hypoxanthine reaction was monitored at 550 nm. One unit of SOD activity is defined as the amount of sample causing 50% inhibition of cytochrome c reduction under the assay conditions. Final reaction mixture (3 ml) was 43 mM KH₂PO₄/K₂HPO₄, pH 7.6, 0.1 mM EDTA, 0.5 mM hypoxanthine, 28 mU xanthine oxidase, and 30 µM cytochrome c.

GPX activities were measured by linking the reaction to that of glutathione reductase and following the decrease in NADPH at 340 nm (extinction coefficient of 62,200 M¹ · cm¹). Selenium-dependent activity (Se-GPX) and the sum of Se-GPX and selenium-independent activities (total GPX) were measured by using, respectively, H₂O₂ and cumene hydroperoxide as substrate (21). Final reaction mixture (1 ml) was 65 mM KH₂PO₄/K₂HPO₄, pH 7.5, 2 mM GSH, 1 mM sodium azide (H₂O₂ assay only), 1 U glutathione reductase, 0.12 mM NADPH, and 2 mM H₂O₂ or 8 mM cumene hydroperoxide.

Glutathione assays. For the determination of total glutathione, digestive gland tissue was homogenized (1:5) in cold 1 M perchloric acid containing 2 mM EDTA and centrifuged at 30,000 g for 20 min at 4°C. Supernatants were neutralized with 2 M KOH/0.3 M MOPS and centrifuged at 1,000 g for 10 min at 4°C. The resulting supernatant was used for total GSH content determination through the glutathione reductase enzymatic assay of Akerboom and Sies (2). The reaction mixture contained 1 mM EDTA in 0.1 M KH₂PO₄/K₂HPO₄, pH 7.0, 0.2 mM NADPH, 0.06 mM dithionitrobenzoic acid, and 0.12 U glutathione reductase. Calibration was performed using GSSG instead of sample. The linear increase in absorbance was recorded at 412 nm, temperature was controlled at 25°C.

Determination of GSH was performed according to the method of Asensi et al. (3). Tissue was homogenized (1:10) in 15% trichloroacetic acid containing 1 mM EDTA and centrifuged at 15,000 g for 5 min at 4°C. The final reaction mixture consisted of 25 µl of the acidic supernatant, 10 µl of 1-chloro-2,4-dinitrobenzene (2 mg/ml of ethanol), 10 µl of glutathione S-transferase (500 U/ml), and 825 µl of 0.5 M KH₂PO₄/K₂HPO₄, pH 7.0, containing 1 mM EDTA. Absorbance was recorded at 340 nm until the end point of the reaction.

Digestive gland cell dissociation. Digestive gland was removed from mussels after in vivo treatments (see Animals and in vivo treatments) and rinsed for ~30 s in ice-cold Ca²⁺/Mg²⁺-free saline (CMFS) containing (in mM): 20 HEPES, 500 NaCl, 12.5 KCl, and 5 EDTA. The tissue was then ice-cold minced and stirred into a beaker containing 25 ml of CMFS. Pronase (1 mg/ml) was added to the solution, and the suspension was incubated under stirring for 30 min at 13°C, passed through a 280-µm silk, and then through a 150-µm silk. The obtained cell suspension was centrifuged for 10 min at 200 g, and cells were resuspended in physiological saline containing (in mM): 20 HEPES, 436 NaCl, 53 MgSO₄, 10 KCl, and 10 CaCl₂. This latter step was repeated twice.

Digital imaging of intracellular oxyradical production. Isolated digestive gland cells (see Digestive gland cell dissociation) were settled for 10 min on coverslips previously treated with poly-L-lysine (1:10 in deionized water), loaded with 3 µM DHR for 3 min, and then washed with physiological saline to remove dye excess. Fluorescent images (ex. = 505 nm, em. = 534 nm) were acquired by an Olympus IMT-2 inverted microscope equipped with an IMT2-RFL fluorescence attachment (Olympus Optical, Germany) and with an MTI SIT 68 intensified camera (Oatencourt, England). Images were digitized and analyzed using the CUE2 RMS 4.0 imaging system (Galai Production, Israel).

Cell viability assay. Digestive gland cells were settled on coverslips, loaded with the LIVE/DEAD viability solution containing ethidium homodimer-1 (4 µM) and calcein AM (2 µM), and incubated for 30 min in a wet chamber at 13°C. In living cells, the membrane-permeant calcein AM is cleaved by esterases to yield cytoplasmic green fluorescence (ex. = 485 nm, em. = 530 nm). In dead cells with damaged membranes, the membrane-impermeant ethidium homodimer-1 labels nucleic acids with red fluorescence (ex. = 485 nm, em. = 590 nm). After staining, cells were washed with physiological saline and incubated for 1 h with different concentrations of H₂O₂. Thereafter, the percentages of red (dead) and green (alive) cells were estimated by sampling at least four coverslip regions.

Anoxic survival assay. After in vivo mussel treatments (see Animals and in vivo treatments), a total of 40 animals for each kind of treatment plus controls were subjected to anoxia by air exposure at 15°C in a humified room. Surviving animals were counted daily by considering open valves or absence of muscular activity as death symptoms until 100% mortality was reached.

Statistics. Data were analyzed by the InStat software package (Graph Pad, San Diego, CA), except survival analysis, which was performed using the Lifetest Procedure, SAS Software (SAS Institute, Cary, NC).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS AND MATERIALS RESULTS DISCUSSION REFERENCES

In the present study, mussel exposure to Cd was used to enhance MT tissue levels, whereas Fe exposure served to induce an oxidative stress (14). Hence, a combination of Cd preexposure/detoxification and subsequent Fe treatment was used to investigate antioxidant properties of MT.

Chemical and biochemical analyses. Mussel exposure to 200 µg/l Cd for 7 days followed by detoxification for 4 wk produced an ~350-fold Cd increase in digestive gland tissue. Such a sharp difference in Cd content between Cd-exposed and nonexposed animals remained essentially unvaried after treatment with 300-600 µg/l Fe (Table 1). Treatments with Fe produced a significant Fe accumulation in digestive gland tissue, both in the presence and in the absence of Cd preexposure (Table 1). MT assays showed a >10-fold increase in the digestive gland of Cd-exposed mussels compared with controls (Fig. 1A). In animals not preexposed to Cd, the treatment with 300 µg/l Fe produced an ~100% MT rise, whereas the treatment with 600 µg/l Fe produced a lower but also significant increase (Fig. 1B). In animals preexposed to Cd, Fe treatments produced slight, nonsignificant decreases in the high MT content induced by Cd (Fig. 1A).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Metallothionein (MT) concentrations in mussel digestive gland tissue. A: mussels exposed to Cd (200 µg/l) for 7 days and then detoxified for 28 days (Cd) show a marked MT rise compared with controls (Contr) (P < 0.01). Mussels exposed to 300-600 µg/l Fe after Cd preexposure and detoxification (Cd/Fe300 and Cd/Fe600) show no significant MT variations compared with Cd. B: mussels exposed to 300-600 µg/l Fe without Cd preexposure (Fe300 and Fe600) show significant MT rises compared with Contr (P < 0.05). Data are means ± SD from 5-9 different measures. Lowercase letters on columns refer to Bonferroni test groups.

Fluorimetric data concerning production rates of reactive oxygen species (ROS) were obtained by incubating a 20,000-g supernatant from digestive gland homogenate with the DHR probe. Bonferroni mean comparisons showed that exposure to 600 µg/l Fe caused a significant rise in ROS production compared with controls (P < 0.05), whereas exposure to Cd did not significantly modify control ROS production rates, and preexposure to Cd abolished the ROS production rise induced by the Fe treatment (Fig. 2A). Parallel data were achieved from an evaluation of the MDA tissue content, obtained through HPLC assay. Also in this case, the Bonferroni analysis showed significant MDA rise in Fe-treated mussels (P < 0.05), whereas differences among controls, Cd-treated, and Cd/Fe-treated animals were not significant (Fig. 2B). In another set of experiments, the sensitivity to oxidative stress in control and Cd-pretreated animals was tested by incubating (in vitro for 60 min) a digestive gland 20,000-g supernatant with 50 µM Fe/100 µM ascorbate, a potent prooxidant mixture (14). Incubation with Fe/ascorbate produced higher ROS production and MDA levels (t-test, P < 0.05 in both cases) in extracts from controls compared with those from Cd-pretreated mussels (Fig. 3).

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Free oxygen radical production and malondialdehyde (MDA) content in mussel digestive gland tissue. A: oxyradical production has been evaluated fluorimetrically on 20,000-g supernatant from tissue homogenates (total protein 166 µg, see METHODS AND MATERIALS). Data are expressed as increase in fluorescence intensity recorded after 1,000 s of sample incubation with dihydrorhodamine (DHR) probe. B: MDA concentrations have been evaluated by HPLC on tissue homogenates (see METHODS AND MATERIALS). Mussels exposed to Cd and then detoxified (Cd) show no variation in oxyradical production and MDA levels compared with Contr. Mussels exposed to 600 µg/l Fe without Cd preexposure (Fe) show a significant rise in both oxyradical production and MDA levels, whereas mussels exposed to Fe after Cd preexposure (Cd/Fe) show no significant variations compared with both controls and Cd-exposed mussels. Data are means ± SD from 9-13 (A) and 4-6 (B) different measures. Lowercase letters on columns refer to Bonferroni test groups (P < 0.05). ww, Wet wt.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Susceptibility to oxidative stress tested through in vitro incubation of digestive gland extracts with 50 µM Fe/100 µM ascorbate followed by fluorimetric evaluation of reactive oxygen species (ROS) or HPLC-based MDA assay (see METHODS AND MATERIALS). Mussels preexposed in vivo to Cd and then treated in vitro with Fe/ascorbate (Cd/Fe-asc) show significantly lower ROS production and MDA levels than mussels treated only in vitro with Fe/ascorbate (Fe-asc). Fluorimetric data are expressed as in Fig. 2. Data are means ± SD from 5 different measures; * P < 0.05. Int, intensity.

To exclude that the resistance to oxidative stress acquired by Cd-treated mussels could have derived from an induction of antioxidant systems other than MT, we assayed main antioxidant enzyme activities, viz. Cat, SOD, GPX, and total glutathione levels in digestive gland extracts. No significant differences between control and Cd-exposed animals were found for each of these components of the antioxidant defense system, with the exception of a minor component of Cat measured in the peroxisome fraction, which increased after Cd exposure (Table 2). Moreover, we also checked the glutathione redox status by evaluating reduced glutathione. The percentage present in the reduced form remained at 97.6% in controls and 96.7% in Cd-exposed mussels. Hence, these data globally indicate no significant induction of main antioxidant enzymes and compound after Cd exposure.

Analyses on isolated cells. The results of the above biochemical analyses seem to indicate that MT plays a role as antioxidant. However, a confirmation of these data was sought in experiments on isolated digestive gland cells obtained from mussels exposed to Cd and/or to Fe. In a first experiment concerning the sensitivity to oxidative stress, cells were loaded with DHR and analyzed by digital imaging to point out intracellular ROS production rates. In cells from Cd-exposed animals, treatment with 0.5 mM H₂O₂ elicited a lower ROS production rise than observed in controls, as shown by t-test comparison (P < 0.01, Fig. 4). In addition, basal ROS production rates prior to H₂O₂ addition were similar in both samples (Fig. 4), thus confirming the data about ROS production in extracts from digestive gland homogenates (see Fig. 2). A second experiment, based on cell loading with the LIVE/DEAD viability solution followed by incubations with different H₂O₂ concentrations, showed a significant drop at 1 mM H₂O₂ in the viability of cells deriving from animals treated with Fe (Bonferroni test, P < 0.01, Fig. 5), whereas cells from mussels preexposed to Cd and then treated with Fe showed a viability similar to controls (Fig. 5).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Digital imaging fluorescence microscopy data concerning digestive gland cells loaded with DHR probe. Oxyradical production rates are initially similar in cells from control and Cd-exposed/detoxified mussels, whereas, after exposure to 0.5 mM H2O2, there is a significantly higher oxyradical rise in control cells. Data are means ± SD from 4 different cells; * P < 0.01 in Cd vs. control comparison.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Cell viability assay on digestive gland cells loaded with LIVE/DEAD viability solution and incubated for 1 h with increasing H2O2 concentrations (see METHODS AND MATERIALS). At H2O2 concentrations higher than 0.5 mM, cell viability shows a marked decrease in cells from mussels treated with Fe (Fe) compared with cells from control mussels and from mussels treated with Fe after Cd preexposure/detoxification (Cd/Fe). Data are means ± SD of percentage of living cells found on 4 different slide areas. * P < 0.01.

Analyses on whole organisms. Finally, an experiment based on mussel anoxic survival was aimed at verifying, also at the organism level, the protective effects of preexposure to Cd against oxidative stress. After in vivo treatments with Cd and/or Fe, animals were left in air under controlled conditions of temperature and humidity. As previously demonstrated, survival in air is lower in stressed mussels (37). In our experiment, survival curves allowed us to point out that animals preexposed to Cd and then treated with Fe were more resistant to air exposure than animals treated with Fe without Cd preexposure (Fig. 6). The LT₅₀ (50% lethal time) of Fe-treated mussels was 6.65 days, whereas the LT₅₀ of Cd-preexposed/Fe-treated mussels was 8.8 days, very close to the LT₅₀ of 9 days for controls. Survival analysis showed significant differences among curves (Fig. 6).

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Survival curves obtained from mussels exposed to anoxic conditions under air. Control animals showed highest resistance to air exposure, whereas mussels preexposed to Cd and then treated with Fe (Cd/Fe) were more resistant than animals treated with Fe without Cd preexposure (Fe). According to survival analysis, differences among curves were highly significant (P < 0.01) or significant (Control vs. Cd/Fe; P < 0.05).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS AND MATERIALS RESULTS DISCUSSION REFERENCES

A main question arising from experimental data consists of the extent to which they are representative of real phenomena. In previous research, MT antioxidant properties were indeed explored at different levels, from in vitro chemical reactions to whole organisms. Hence, multilevel information is available, but it must be reconstituted from a variety of data obtained in different experiments. In addition, many studies actually indicated an antioxidant role of MT, but some did not; thus contributing to build up a complex and debatable picture about the MT role in cellular antioxidant defense (31).

In this study, we attempted to achieve an integrated picture of MT antioxidant effects, spanning from the molecular to the organism level. We, therefore, chose the mussel as an experimental model organism as this bivalve mollusc, besides showing fair MT inducibility, is particularly resistant to manipulations. The mussel seemed, therefore, particularly suitable for an experimental design involving a sequence of in vivo treatments. Moreover, the developed resistance of mussels to air exposure allowed us to perform sensitive anoxic survival tests for an assessment of the physiological status of whole organisms at the end of heavy metal incubations.

The first step of this research consisted of the checkup of heavy metal tissue contents on in vivo mussel exposures to metals. Data showed that Cd and Fe incubations caused an increase of either metal in digestive gland tissue and, moreover, that these metals did not largely interfere with the tissue accumulation of each other. More in detail, Fe treatments produced significant Fe increases compared with controls in both Cd-preexposed and nonpreexposed mussels; and in Cd-preexposed mussels, treatment with Fe produced no variation of the enhanced Cd content (see Table 1). The high Cd content found in mussels after the detoxification period and even after subsequent Fe treatments is a confirmation of the particularly long half-life of this heavy metal in mussel tissue (40).

The assay of MT after Cd exposure and detoxification confirmed that Cd produced a strong and persistent MT increase in mussel digestive gland tissue. Moreover, such an increase remained nearly unaltered after treatment with Fe, showing that mussels preexposed to Cd were endowed with high MT tissue levels while experiencing Fe accumulation. On the other hand, Fe treatments made on mussels that had not been preexposed to Cd also produced a lower but significant MT increase. This is to some extent in line with previous data on the chick, in which Fe parenteral injections led to marked MT rise (10, 25). It has also been shown that Fe can bind in vitro to MT, but the physiological meaning of this interaction has not been established yet (12).

The prooxidant effects of transition metals such as Cu and Fe on marine organisms have been assessed, showing that these metals can induce oxyradical production leading to lipid peroxidation (33, 38). This was confirmed by our experiments, demonstrating increases of both oxyradical production and MDA in the tissue of mussels exposed to Fe. By contrast, no evidence for an alteration of the redox status was found in Cd treated/detoxified mussels. Moreover, in the present study, it was also found that Cd preexposure was able to prevent prooxidant processes due to subsequent Fe treatment or to lessen the susceptibility to oxidative stress of tissue extracts, suggesting an antioxidant effect of MT.

Yet, these data do not exclude that cellular antioxidant systems other than MT could be involved in the prevention of Fe-induced oxidative stress. For this reason, we also tested the effect of Cd exposure on main antioxidant enzymes and glutathione levels. Glutathione is considered a main hydrophilic oxyradical scavenger able to protect cells from oxydative damage (7). The concentration and redox status of this low-molecular- weight thiol, which is effective against oxidative damage also in the mussel (28), showed no significant variation in Cd-exposed mussels compared with controls. Main enzymes involved in detoxification from reactive oxygen species, i.e., SOD for superoxide anion radical, Cat for H₂O₂, and GPX for hydroperoxides, have been shown to contribute to antioxidant defense also in the mussel (11). Our data showed negligible variations for these enzymes, with the only exception of some Cat increase that cannot account for the sharp resistance to prooxidant processes developed by Cd-preexposed mussels. Hence, these assays globally indicated that main mussel antioxidant systems remained unaltered after Cd treatment.

Biochemical assays involve the use of tissue homogenates, in which the cellular distribution of MT could be altered and oxyradical production rates modified, thus possibly providing biased information about the actual antioxidant potentials of MT at the cellular level. However, data concerning isolated digestive gland cells have provided a confirmation to biochemical assays. Firstly, exposure to H₂O₂ strongly enhanced oxyradical production, as detected by DHR, in control cells but not in cells from Cd-exposed mussels. Secondly, the viability on H₂O₂ incubations observed in cells from mussels preexposed to Cd clearly indicated that the reduction in intracellular oxyradical production due to Cd preexposure actually protected cells from oxidative damage. Hence, data collected on whole cells combined with evidence from biochemical assays provide a clear indication that MT has a physiological role in the protection of cells against the toxic effects of free oxygen radicals.

Yet, data obtained on cells in primary culture, although extremely indicative, cannot provide exhaustive information about a possible repercussion at the organism level of MT antioxidant properties. However, in our experiments, such a gap has been filled by anoxic survival tests, which allowed us to ascertain a protective effect of Cd preexposure against noxious effects produced by Fe treatments at the organism level. This clearly indicates that the resistance to oxidative stress conferred by MT rise, as evaluated at the subcellular and cellular levels, reflects the occurrence of a physiological process, which is effective also at the organism level.

Taking into account the very high in vitro reactivity of MT with hydroxyl radical · OH (36), it has been speculated that the antioxidant role of MT could mainly consist in a scavenging activity of this extremely reactive and dangerous oxyradical species, a view also reinforced by the fact that no specific enzyme for hydroxyl radical inactivation is known (31). Our data provide support to such a view, as Fe is known to act as a catalyst for the Fenton reaction, thereby facilitating the production of hydroxyl radicals (14). In our experiments, MT tissue rise corresponded to increased resistance against Fe and H₂O₂, and in both cases, a reduction in oxyradical production rates was also found.

In conclusion, this study has provided evidence arguing for an antioxidant role of MT in the mussel. The significant contribution to antioxidant defense provided by MT seems to derive from oxyradical scavenging activity and is effective in protecting cells and the entire organism from oxidative stress.

Perspectives