Copper transport by lobster (Homarus americanus) hepatopancreatic mitochondria

Mitochondria are multifunctional cellular organelles with both energetic and ion-sequestration functions. While considerable information exists about the ion-transporting functions of mammalian mitochondria (Carafoli, 1982; Gunter and Pfeiffer, 1990; Gunter and Gunter, 1994; Gunter et al., 1994), few comparative data are available for these organelles in invertebrate cells. This paucity of mitochondrial ion-transport information from invertebrates is particularly surprising because in some of these animals, such as the crustaceans, considerable ion fluxes occur through epithelial cells during the molt cycle and it is known from structural studies that epithelial mitochondria are a significant site of sequestration of divalent cations during this event.

In crustaceans, the role of mitochondria in Ca²⁺ regulation has been little studied since the early investigations of Chen and Lehninger (1973), Becker et al. (1974) and Chen et al. (1974). These studies investigated the effect of Ca²⁺ on the respiration of mitochondria isolated from crab hepatopancreas and found that the organelles were able to transport and store large amounts of Ca²⁺ and that the movements of the cation across mitochondrial membranes were influenced by inorganic phosphate, ATP, divalent cations and the uncoupling agent p-trifluoromethoxy cyanide phenylhydrazone (FCCP). In addition, ultrastructural studies identified intramitochondrial deposits of calcium phosphate associated with large movements of Ca²⁺ into the organelle.

Recently, the mechanisms of Ca²⁺ transport by lobster hepatopancreatic mitochondria were investigated (Klein and Ahearn, 1999). This study used ⁴⁵Ca²⁺ influx measurements to characterize both the uptake and efflux mechanisms for this divalent cation. The results suggested that, as with mammalian mitochondria (Gunter et al., 1994), the major uptake process was a Ruthenium-Red-sensitive uniporter, while efflux from the organelle was controlled by dual antiport processes: a Na⁺-dependent, diltiazem-sensitive mechanism and a Na⁺-independent, diltiazem-insensitive process. Both influx and efflux processes were inhibited by cytoplasmic zinc.

Crustaceans often ingest considerable quantities of heavy metal ions from their diet and surroundings. At low concentrations, some of these metal ions, such as zinc, iron and copper, may serve important biological roles in enzyme activation or respiratory pigment synthesis, but at higher concentrations they become toxic and may impair nerve and muscle function. It is well known that invertebrates possess at least three epithelial detoxification mechanisms to regulate concentration of heavy metal ions in their bodies: (i) metallothioneins (Engel and Brouwer, 1989; Roesijadi, 1992); (ii) vacuolar concretion metal precipitates (George, 1983a,b; Viarengo et al., 1985, 1987; Vogt and Quinitio, 1994); and (iii) mitochondrial sequestration (Gunter and Pfeiffer, 1990; Gunter and Gunter, 1994; Saris and Niva, 1994; Viarengo, 1994). The specific membrane transport mechanisms responsible for isolating excess metals by way of vacuolar precipitates and mitochondrial sequestration are poorly understood.

In a more recent study, we investigated the carrier-mediated transport processes responsible for transferring copper into isolated lobster hepatopancreatic epithelial cells separated by centrifugal elutriation (Chavez-Crooker et al., 2001). This study did not address the fate of intracellular copper in any hepatopancreatic cell type following passage across the epithelial brush-border membrane. The present investigation extends our recent studies of hepatopancreatic epithelial copper uptake characterization and mitochondrial cation regulation by using the copper-specific fluorescent dye Phen Green to describe the mechanisms by which this heavy metal is accumulated within mitochondria as part of the detoxification processes present in this epithelium.

Male Atlantic lobsters (Homarus americanus H. Milne Edwards, 500–700 g) were purchased from a local market and maintained in a seawater tank at 15°C. All experiments were conducted with intermolt animals. Molt stage was estimated by determining the ratio between the mass of the stomach gastrolith (mg) and the carapace length (cm) (Aiken, 1973).

Purification of hepatopancreatic mitochondria followed the methods outlined by Klein and Ahearn (1999), as adopted from Chen and Lehninger (1973) with the addition of a Percoll gradient purification step for the final pellet (Symons and Jonas, 1987). A hepatopancreas was removed from an animal and minced into small pieces with a single-edged razor blade in ice-cold Buffer A (250 mmol l^–1 sucrose plus 1 mmol l^–1 EDTA) followed by homogenization (five strokes at 2800 revs min^–1) with a loose-fitting Teflon pestle in a volume of Buffer A that approximated the mass of the original tissue. The homogenate was diluted 10-fold in Buffer A and then distributed among 4–8 plastic centrifuge tubes followed by centrifugation (two 10-min low-speed spins at 750 g) to remove dense nuclear and particulate material. Fat was removed from the top of the tubes, and the remaining supernatants were poured through two layers of cheesecloth. The filtered supernatants were centrifuged for 15 min at 9000 g to collect membranes and organelles. The resulting pellets were resuspended in 1 ml of ice-cold Buffer B (250 mmol l^–1 sucrose, 25 mmol l^–1 Hepes/Tris, pH 7.5), fully mixed using plastic serum pipettes, diluted to 30 ml in Buffer B, and centrifuged for 15 min at 7000 g. The resulting pellets were resuspended in 1.0 ml of Buffer B using a plastic serum pipette, pooled, and distributed among centrifuge tubes containing 1.5 ml of iso-osmotic Percoll (9 parts Percoll to 1 part Buffer C; 2.5 mol l^–1 sucrose plus 200 mmol l^–1 Hepes, pH 7.5). Buffer B was added to the tubes to bring the final volume to 10 ml, and they were centrifuged for 60 min at 35 000 g. Fractions (1 ml) of the resulting gradient were removed from the top of the tubes with a pipette and collected. Each band was cleaned of Percoll by dilution with ‘inside medium’ (composition varying between experiments) to 10 ml and centrifuged at 18 000 g for 15 min.

The mitochondrial fraction was collected from the Percoll gradient by pipetting off the top 8.5 ml and removing the dark brown band near the bottom of the tubes. The mitochondria were resuspended in appropriate ‘inside medium’ with 20 strokes of a 1 ml tuberculin syringe fitted with a 22-gauge needle and placed on ice until needed after dilution to a working concentration of 20–50 mg protein ml^–1. Protein was determined using the Bio-Rad protein assay with albumin standards in a Beckman DU-640 recording spectrophotometer.

Transport assays were conducted by adding 1 part of mitochondria in ‘inside buffer’ to 8 parts of ‘outside buffer’. Outside buffers were composed of mannitol, sucrose, potassium gluconate, trimethylammonium (TMA⁺) gluconate and Hepes/Tris adjusted to the appropriate pH (Klein and Ahearn, 1999). The copper-specific, fluorescent probe Phen Green and its acetoxymethyl ester (Molecular Probes, Inc.) were used in this study as described previously for whole-cell suspensions of the lobster hepatopancreas (Chavez-Crooker et al., 2001). Phen Green (at 1 μmol l^–1) fluorescence (excitation 490 nm; emission 520 nm) was measured at room temperature (24°C) using hand-stirred samples in a Shimadzu spectrofluorometer (model FR-1501) connected to a DPU-411, Type II, thermal printer. Previous experiments with Phen Green and hepatopancreatic cell suspensions indicated that copper transport could be measured in dye-equilibrated cells following a quantitative decrease in fluorescence (e.g. signal quenching) as external metal ion concentration was increased (Chavez-Crooker et al., 2001). The resulting fluorescence quenching was measured as ΔF mg^–1 protein 120 s^–1 and was quantitatively related to the transport of copper across cell plasma membranes.

In the present study, copper transport experiments were started by equilibrating mitochondria with Phen Green (60 min at 24°C in an orbital shaker at 100 revs min^–1). The preparation was then centrifuged for 10 min (at 9000 g), and the resulting pellet was re-suspended in buffer and maintained on ice for further analysis. During an experiment, a sample of mitochondria was injected into a cuvette, housed in the spectrofluorometer, containing a physiological buffer without copper or another metal ion so that maximal fluorescence would be obtained. Once the dye was stable (at 100 % fluorescence), copper ions were added at different concentrations and the increase in fluorescence quenching (as ΔF mg^–1 protein 120 s^–1) was monitored as an indication of metal transport across mitochondrial membranes.

Curve-fitting and the production of the resulting copper influx kinetic constants were performed using Jandel SigmaPlot 4.01 or 5.0 software. All values reported in this investigation were obtained from triplicate samples, and results are expressed as means ± 1 s.e.m.

Fig. 1 is a tracing of the time course of change in Phen Green fluorescence that occurred as a result of quantitative quenching of the signal following the addition of three copper concentrations (2, 6 and 12 μmol l^–1) to a spectrofluorometric cuvette containing dye-equilibrated hepatopancreatic mitochondria. The magnitude of quenching was closely correlated with the concentration of copper added to the external buffer, as found previously with isolated hepatopancreatic epithelial cells (Chavez-Crooker et al., 2001), and the rate of initial change in fluorescence over 120 s was used as a quantitative index of transmembrane copper transport. Addition of 1 mmol l^–1 EDTA, a copper chelator, restored the fluorescence signal, suggesting that complexes form between the chelator and copper in solution.

Previous studies with isolated mitochondria from lobster hepatopancreas indicated that Ca²⁺ was transported into these organelles by a combination of a Ruthenium-Red-inhibited uniporter and a diltiazem-sensitive antiporter (Klein and Ahearn, 1999). To determine whether copper was transported by mitochondrial Ca²⁺ carrier proteins, a control experiment was first conducted measuring copper influx kinetics over periods of 120 s at a series of external copper concentrations (0.4, 1.6, 4.08, 8.15 and 12.2 μmol l^–1) (Fig. 2). As indicated by the results in this figure, Cu²⁺ appeared to be translocated to the mitochondrial interior by way of a combination of saturable and non-saturable transport components and the rate of transport followed the equation:

where J_Cu is total copper influx, J_max is apparent maximal carrier-mediated transport, K_Cu is an apparent binding affinity constant, [Cu²⁺] is the external concentration of Cu²⁺ and K_D is the slope of the non-saturable influx component and may represent either diffusion or carrier-mediated transport by a low-affinity process that did not saturate within the range of concentrations selected. A best-fit curve was drawn through the data generated using Jandel SigmaPlot software. The kinetic constants for copper influx by the combination of these processes are as follows: K_Cu=0.17±0.04 μmol l^–1, J_max=129.8±5.9 ΔF mg^–1 protein 120 s^–1 and K_D=4.61±0.62 ΔF mg^–1 protein 120 s^–1 μmol^–1 l Cu (N=3).

Addition of 500 μmol l^–1 calcium chloride to the external medium at each copper concentration resulted in a significant decrease (P<0.01) in the rate of copper influx over the entire copper concentration range (Fig. 2). A best-fit curve is drawn through the experimental data in this figure, and the apparent kinetic constants for copper influx under these conditions are as follows: K_Cu=0.35±0.03 μmol l^–1, J_max=120.0±3.15 ΔF mg^–1 protein 120 s^–1 and K_D= 0.95±0.31 ΔF mg^–1 protein 120 s^–1 μmol^–1 l Cu (N=3). A comparison of the kinetic constants of the control and +500 μmol l^–1 Ca²⁺ conditions suggests that addition of Ca²⁺ doubled (P<0.01) the apparent K_Cu (reduced the apparent copper binding affinity) and significantly (P<0.01) reduced non-saturable copper transport velocity by 80 %, without having a significant (P>0.05) effect on the apparent maximal carrier-mediated copper transport velocity (J_max). An increase in the apparent copper binding affinity as seen in the present set of experiments suggests that Ca²⁺ acted as a competitive inhibitor of carrier-mediated copper influx. The significant reduction of the non-saturable copper transport component in the presence of Ca²⁺ strongly supports the view that this transfer process was carrier-mediated, exhibiting a relatively low apparent copper binding affinity.

In a previous study of Ca²⁺ transport by lobster hepatopancreatic mitochondria, Ruthenium Red was shown strongly to inhibit the influx of this divalent cation through a well-described uniporter carrier mechanism (Klein and Ahearn, 1999). A Dixon plot of the interaction between Ruthenium Red and Ca²⁺ transport indicated that the dye was a strong competitive inhibitor of mitochondrial Ca²⁺ influx as a result of both dye and cation sharing a common binding site on the organellar membranes. Because Fig. 2 suggested that copper and Ca²⁺ may be transported into lobster hepatopancreatic mitochondria by two different carrier processes, it was important to establish whether one or both of these transport mechanisms were also sensitive to Ruthenium Red.

Before determining the effect of Ruthenium Red on copper influx into isolated mitochondria from lobster hepatopancreas, the possible quenching effect of Ruthenium Red on Phen Green fluorescence had to be quantified so that an appropriate concentration of the inhibitor could be determined that had a minimal effect on signal reduction from the fluorescent dye. The data displayed in Fig. 3 indicate that, as Ruthenium Red concentration was increased from 250 to 5000 nmol l^–1 in a spectrofluorometer cuvette containing 1 μmol l^–1 Phen Green in transport buffer, the fluorescent signal was reduced in a linear fashion until, at the highest Ruthenium Red concentration used, approximately 50 % of the fluorescent signal was abolished. At a Ruthenium Red concentration of 500 nmol l^–1, only 5 % of the fluorescent signal was quenched, and this concentration of inhibitor was therefore used for subsequent experiments with copper. This concentration of Ruthenium Red was considered physiologically effective since the K_i for Ruthenium Red inhibition of Ca²⁺ influx by lobster hepatopancreatic mitochondria was found to be 1000 nmol l^–1 (Klein and Ahearn, 1999).

Fig. 4 displays the results of an experiment examining the effects of 500 nmol l^–1 Ruthenium Red on copper influx kinetics into isolated lobster hepatopancreatic mitochondria. As reported above, copper influx was biphasic under control conditions and appeared to be the sum of saturable and non-saturable processes over the range of copper concentrations selected. The kinetic constants for copper influx under control conditions were: K_Cu=0.63± 0.15 μmol l^–1; J_max= 88.17±7.71 ΔF mg^–1 protein 120 s^–1 and K_D=3.05±0.67 ΔF mg^–1 protein 120 s^–1 μmol^–1 l Cu (N=3).

In the presence of 500 nmol l^–1 Ruthenium Red, the control biphasic influx curve was reduced to a single Michaelis–Menten hyperbolic relationship between the variables. A best-fit analysis of copper influx under these conditions resulted in the following kinetic constants: K_Cu=1.05±0.34 μmol l^–1 and J_max=82.94±6 ΔF mg^–1 protein 120 s^–1 (N=3). These results suggest that, in the presence of Ruthenium Red, the low-affinity, non-saturable copper influx component was abolished. In addition, the presence of Ruthenium Red resulted in an approximate doubling of the apparent affinity constant (P<0.01) of the saturable transport component, but had no significant effect (P>0.05) on apparent maximal copper influx by this transporter. These results are compatible with the view that Ruthenium Red acted on both apparent carrier-mediated transport processes responsible for copper uptake into lobster hepatopancreatic mitochondria.

Our previous experiments with hepatopancreatic mitochondria suggested that Ca²⁺ efflux from these organelles might occur through a diltiazem-sensitive, electroneutral 1Ca²⁺/2Na⁺ antiporter (Klein and Ahearn, 1999). To assess whether copper was also transported by this cation exchanger, an experiment was conducted using Phen-Green-equilibrated mitochondria in the presence and absence of 500 μmol l^–1 diltiazem and external concentrations of copper from 0.4 to 12.2 μmol l^–1, as previously employed in this study. As displayed in Fig. 5, the kinetic constants for copper influx under control conditions (lacking diltiazem) were K_Cu=0.16±0.04 μmol l^–1, J_max=126.6±7.07 ΔF mg^–1 protein 120 s^–1 and K_D=5.04±0.75 ΔF mg^–1 protein 120 s^–1 μmol^–1 l Cu (N=3). In the presence of 500 μmol l^–1 diltiazem, the kinetic constants for copper influx were K_Cu=0.23±0.02 μmol l^–1, J_max=108.50±3.08 ΔF mg^–1 protein 120 s^–1 and K_D=5.44±0.31 ΔF mg^–1 protein 120 s^–1 μmol^–1 l Cu (N=3). Addition of diltiazem resulted in a significant (P<0.05) reduction in apparent copper binding affinity and a significant (P<0.05) decrease in apparent maximal transport velocity compared with the control condition, but had no significant (P>0.05) effect on the apparent non-saturable transport process. These results suggest that diltiazem acted as a mixed inhibitor of copper influx into hepatopancreatic mitochondria by the saturable transport system.

The crustacean hepatopancreas is a multifunctional organ responsible for digestion and absorption of dietary nutrients, storage of glycogen and fat, ion transport and osmoregulation, synthesis and release of the oxygen-binding pigment hemocyanin and sequestration and detoxification of heavy metals. Among its many functions, this remarkable organ controls the uptake of divalent cations such as Ca²⁺ and Cu²⁺ from the gastrointestinal tract and stores them within organelles of the hepatopancreatic epithelium, incorporates them into biosynthetic pathways such as that for hemocyanin synthesis or transfers them through the cell layer to the hemolymph, where they may be absorbed by other cell types throughout the organism.

Recently, we investigated the mechanisms of uptake and efflux of Ca²⁺ by lobster hepatopancreatic epithelium using purified brush-border and basolateral membrane vesicles (Ahearn and Zhuang, 1996; Zhuang and Ahearn, 1996, 1998). These studies described three brush-border membrane proteins involved in Ca²⁺ uptake from dietary elements: (i) an amiloride-sensitive, electrogenic 1Ca²⁺/1H⁺ antiporter that was shared with Na⁺; (ii) an amiloride-insensitive, electroneutral 1Ca²⁺/2Na⁺ antiporter; and (iii) a verapamil-inhibited Ca²⁺ channel. On the basolateral membrane of this epithelium, three additional Ca²⁺ transport proteins were also described: (i) a high-affinity, ATP-dependent Ca²⁺ pump; (ii) a low-affinity, electrogenic 1Ca²⁺/3Na⁺ antiporter; and (iii) a verapamil-sensitive Ca²⁺ channel. Other authors have also reported the presence and characteristics of these Ca²⁺ transport systems in epithelia from freshwater crayfish (Wheatly, 1996, 1999; Wheatly et al., 1998, 1999), marine crabs (Flik et al., 1994) and the European lobster Homarus gammarus (Flik and Haond, 2000). It has been proposed that these six membrane proteins are responsible for most, if not all, Ca²⁺ movements into and through crustacean epithelia during feeding, excretion and ion regulation under conditions that the animals may encounter during their daily activities and during the different molt stages that they undergo periodically.

Heavy metals, such as zinc and cadmium, are known to share one or more of these Ca²⁺ transport proteins in epithelia from crustaceans and in related transporters in echinoderm epithelia (Ahearn et al., 1994; Zhuang et al., 1995). In addition, a recent study, using the copper-sensitive dye Phen Green and centrifugal elutriation to separate individual hepatopancreatic cell types cleanly into purified suspensions, showed that copper uptake into each of the four hepatopancreatic epithelial cell types occurred via a brush-border cation antiport process that was shared by Zn²⁺ and exchanged with intracellular Ca²⁺ or other cations (Chavez-Crooker et al., 2001). In this latter study, a model was suggested for the transapical uptake of heavy metals from dietary elements using one or both of the previously described Ca²⁺ antiporters shown to occur in this membrane. In a more recent study using Phen Green to investigate copper influx into individual lobster hepatopancreatic epithelial cells, amiloride was employed as a potential inhibitor of copper uptake by the amiloride-sensitive electrogenic 1Ca²⁺/1H⁺ antiporter (P. Chavez-Crooker, N. Garrido and G. A. Ahearn, unpublished observations). No effect of this inhibitor was observed during copper influx into each of the four hepatopancreatic epithelial types, suggesting that the entry mechanism for this metal was probably via the brush-border electroneutral, amiloride-insensitive 1Ca²⁺/2Na⁺ antiporter. None of these studies provided evidence for the intracellular fate of metals or Ca²⁺ after passage across the respective epithelial brush-border membranes.

One fate of Ca²⁺ and metal ions transported by hepatopancreatic epithelial cells might be the sequestration of these divalent cations in organelles such as the mitochondria, lysosomes or endoplasmic reticulum. In vertebrates, Ca²⁺ and a variety of other cations are known to be transported across mitochondrial membranes by several different proteins. The most thoroughly investigated process is the electrogenic, Ruthenium-Red-inhibited uniporter that transfers Ca²⁺ from the cytosol to the mitochondrial interior (Carafoli, 1982; Bronner, 1992; Gunter et al., 1994). Several studies have indicated that this uptake process also supports the transport of Zn²⁺ (Saris and Niva, 1994), Mn²⁺ (Pushkin et al., 1976; Baker and Schlaepfer, 1978), Sr²⁺ (Carafoli, 1965), Ba²⁺ (Akerman et al., 1977), Fe²⁺ (Romslo and Flatmark, 1973) and La³⁺ (Reed and Bygrave, 1974). Ruthenium Red is a non-competitive inhibitor of the vertebrate Ca²⁺ uniporter protein and, while reducing the rate of cation transfer across the membrane, does not interfere with the Ca²⁺ binding process (Reed and Bygrave, 1974). Several hypotheses have been proposed for the presence of a Ca²⁺ uniporter uptake process that transports divalent cations from the cytoplasm to sequestration sites within the organelle (Gunter et al., 1994). The suggested functions of this carrier include (i) regulation of cytosolic cation concentrations, (ii) the generation of a cation sink in times of excess cytoplasmic concentrations of substances that might otherwise be toxic, (iii) service as a reusable source of activator Ca²⁺ or essential metals needed for enzymatic function and (iv) regulation of mitochondrial free Ca²⁺ concentration to control the activation of Ca²⁺-dependent dehydrogenases.

At least two characterized antiporter proteins occur on vertebrate mitochondrial membranes that control the rate of Ca²⁺ efflux from the organelle (and potentially that of other sequestered cations as well). One antiporter is Na⁺-dependent and diltiazem-inhibited (Vaghy et al., 1982), while the other is Na⁺-independent and not affected by diltiazem (Gunter and Pfeiffer, 1990). The diltiazem-sensitive exchanger displays a stoichiometry of 1Ca²⁺/2Na⁺ and is therefore electroneutral. The Na⁺-independent Ca²⁺ efflux antiporter also is electroneutral and exhibits a stoichiometry of 1Ca²⁺/2H⁺. The driving force for these two electroneutral exchangers appears to be the transmembrane Ca²⁺ activity difference between the mitochondrial matrix and the cytoplasm, which is believed to be a factor of approximately 2 (Gunter and Pfeiffer, 1990; Gunter et al., 1994). Both efflux mechanisms display some specificity for other divalent cations. The 1Ca²⁺/2Na⁺ antiporters can substitute Sr²⁺, but not Mn²⁺, for Ca²⁺ (Crompton et al., 1976; Gavin et al., 1990), while the 1Ca²⁺/2H⁺ exchangers are able to transport Sr²⁺ (Gunter et al., 1988), Ba²⁺ (Lukacs and Fonyo, 1985) and Mn²⁺ (Gavin et al., 1990). While both antiporters exist in mitochondria from a wide variety of vertebrate tissues, the Na⁺-dependent exchanger predominates in heart, skeletal muscle, brain and other tissues, while the Na⁺-independent transporter predominates in liver and kidney (Wingrove and Gunter, 1986a,b).

Recently, the Ca²⁺ transport mechanisms of lobster hepatopancreatic mitochondria have been characterized (Klein and Ahearn, 1999). This study showed that, as with vertebrate mitochondria, the crustacean organelles displayed a combination of a Ruthenium-Red-sensitive uniporter for Ca²⁺ uptake from the cytoplasm and two Ca²⁺ efflux mechanisms: a diltiazem-inhibited, electroneutral 1Ca²⁺/2Na⁺ antiporter, and a diltiazem-insensitive, electroneutral 1Ca²⁺/2H⁺ exchanger. Zinc appeared to interact with both uptake and efflux pathways.

In the present investigation, the mechanisms of copper uptake by lobster hepatopancreatic mitochondria were studied in an effort to determine whether any of the previously described Ca²⁺ transport systems of this organelle would accommodate the heavy metal. Fig. 6 is a working model of the combined results obtained in the present study, in our recent characterization of mitochondrial Ca²⁺ transport (Klein and Ahearn, 1999) and in our previous study examining the nature of copper influx into hepatopancreatic epithelial cells across brush-border membranes (Chavez-Crooker et al., 2001). As shown in this figure, heavy metals, such as copper, may enter hepatopancreatic epithelial cells by an antiporter carrier system in exchange for intracellular Na⁺ or other cations such as Ca²⁺ (Chavez-Crooker et al., 2001). This transport system may be activated by Ca²⁺ derived from the exterior through verapamil-sensitive channels or Ca²⁺ may serve as an antiport substrate in exchange with the external cation. After copper, or another transportable heavy metal ion, has entered the cytoplasm, it may be translocated into mitochondria either through the Ruthenium-Red-sensitive uniporter or by way of one or both Ca²⁺ antiporters, displacing the normal substrates (e.g. 2Na⁺ or 2H⁺) for these exchangers.

Results from the kinetic analyses of copper influx into hepatopancreatic mitochondria (Figs 2, 4, 5) generally agree with previous studies characterizing the mechanisms of Ca²⁺ uptake by these organelles (Klein an Ahearn, 1999) and support the view that Ca²⁺ and copper share the same uptake processes. Klein and Ahearn (1999) described an apparent low-affinity, non-saturable, Ruthenium-Red-inhibited Ca²⁺ uptake process in hepatopancreatic mitochondria that did not saturate over a Ca²⁺ activity range of 14.06–450 μmol l^–1. Similarly, in the present investigation, mitochondrial copper influx kinetics suggested the presence of saturable and non-saturable transport components (Figs 2, 4, 5). Ca²⁺ and Ruthenium Red essentially abolished the non-saturable uptake process (Figs 2, 4), while diltiazem was ineffective as an inhibitor of this process (Fig. 5). It is likely, therefore, that this non-saturable uptake mechanism was the mitochondrial uniporter transport system and appeared to be capable of transporting Ca²⁺, Cu²⁺ and Zn²⁺ (Klein and Ahearn, 1999).

Diltiazem produced a small, but significant, inhibition of copper influx through a mixed-type effect on the saturable influx process (Fig. 5). In addition, both Ca²⁺ and Ruthenium Red acted as competitive inhibitors of copper influx by the saturable transport mechanism (Figs 2, 4). It can, therefore, be postulated that this saturable uptake process is probably one or both of the previously characterized Ca²⁺ antiporters, can transport both Ca²⁺ and Cu²⁺ and is inhibited by both diltiazem and Ruthenium Red.

Ca²⁺ and Cu²⁺ uptake by hepatopancreatic mitochondria represent one possible means by which these cells are able to control the concentrations of these divalent cations in the cytoplasm and in the blood and, thereby, detoxify their potential debilitating effects on nerves and muscles. It is well known that Ca²⁺, Mn²⁺ and Sr²⁺ may be stored at high concentrations in these organelles in combination with phosphate (Chen et al., 1974). This study and that of Klein and Ahearn (1999) suggest that both zinc and copper may also be accumulated within these organelles, thereby limiting their cytoplasmic and hemolymph concentrations.

Other means may be employed by hepatopancreatic cells to limit intracellular concentrations of metals such as copper. Metallothioneins are low-molecular-mass proteins possessing high-affinity cation-binding sites that are synthesized in response to increased cellular concentrations of metals. Cytoplasmic metals may bind to these proteins during periods of excess dietary metal and contribute to their detoxification (Engel and Brouwer, 1989; Roesijadi, 1992; Viarengo et al., 1985, 1987). Electron micrographs of numerous epithelial cells of the gastrointestinal tract and kidneys of a variety of invertebrates show the presence of membrane-bound vacuoles within the cytoplasm that, upon examination by electron microprobe analysis, are found to contain concretions of calcium and several heavy metals including zinc and copper (Al-Mohanna and Nott, 1985, 1987; George, 1983a,b; George et al., 1982; Loret and Devos, 1992). In some instances, these metal-containing vacuolar concretions are stored in the respective epithelial cells and later released by exocytosis to the gastrointestinal lumen as the cell ages; they may eventually be found in the feces (Vogt and Quinitio, 1994). It is likely that, through the combined contribution of heavy metal uptake and sequestration by the mitochondria, cytoplasmic metallothioneins and intracellular metal-containing concretions, cellular regulation of these potentially toxic elements can be managed.

An alternative fate of transported Cu²⁺ in hepatopancreatic epithelial cells, at low concentrations of the metal, is the inclusion of these ions in biosynthetic pathways for a variety of essential enzymes and the oxygen-binding pigment hemocyanin. In the last 10 years, a variety of elegant molecular biological methods (Spindler et al., 1992; Rainer and Brouwer, 1993; Durstewitz and Terwilliger, 1997) have shown the site of hemocyanin synthesis in crustaceans to be the hepatopancreatic epithelium. The biosynthetic characteristics of this important respiratory pigment, and inclusion of copper within its molecular make-up, is related both to the developmental stage of the organism, when animals undergo the transition from larval to adult blood types (Durstewitz and Terwilliger, 1997), and to the molting cycle, when large fluctuations in protein synthesis take place (Spindler et al., 1992). What is still unclear about the inclusion of copper within the biochemistry of hemocyanin synthesis is which type of hepatopancreatic epithelial cell is responsible for this activity. It is hoped that the cell purification methods described in our recent paper (Chavez-Crooker et al., 2001) may help define the locus of this biological process.

This study was supported by National Science Foundation grant number IBN99-74569.