Active Osmoregulatory Ion Uptake Across the Pleopods of the Isopod Idotea Baltica (Pallas): Electrophysiological Measurements on Isolated Split Endo- and Exopodites Mounted in a Micro-Ussing Chamber

Most crustaceans live in sea water. In this medium, their haemolymph is nearly iso-osmotic with respect to the environment. Thus, osmoregulation of the body fluid seems to be unnecessary. However, many species migrate to brackish or freshwater habitats. When confronted with these more dilute media, the osmotic concentration of the body fluids is hyperregulated by means of active salt uptake. For decapod crustaceans, it has been shown that this osmoregulatory ion uptake occurs predominantely in the thoracic gills (for reviews, see Mantel and Farmer, 1983; Péqueux et al., 1988; Onken et al., 1994). In isopods, the thoracic gills are extremely reduced, and the branchial functions, including osmoregulatory active ion uptake, are located in the pleopods (Hørlyck, 1973; Bubel and Jones, 1974; Babula, 1977, 1979; Babula and Bielawski, 1981; Kümmel, 1981; Wägele, 1982). These locomotory abdominal extremities consist of thin, oval, dorsoventrally flattened leaf-like exo- and endopodites extending from the protopodite. Measurements of the activity of Na⁺/K⁺-ATPase, the central enzyme in active transepithelial absorption of NaCl in crustacean gills (for a review, see Towle, 1984), in the pleopods of the euryhaline isopods Sphaeroma serratum (Thuet et al., 1969) and Idotea wosnesenskii (Holliday, 1988) indicate that active osmoregulatory ion absorption in these species is located only in the posterior endopodites. The ultrastructure of the endopodites of several osmoregulating brackish-water, freshwater and terrestrial isopods reveals characteristics typical of salt-transporting tissues, as described for the gills of other groups of crustaceans. An epithelial monolayer of ionocytes is covered by a thin cuticle. Within the cells, extensive infoldings of the apical and the interdigitating basolateral plasma membranes are closely associated with numerous elongated mitochondria (Copeland, 1968; Copeland and Fitzjarrell, 1968; Bubel and Jones, 1974; Babula, 1977, 1979; Babula and Bielawski, 1981; Kümmel, 1981; Wägele, 1982; Barra et al., 1983; Péqueux et al., 1987; Compère et al., 1989).

Two methods have been employed successfully in recent years to investigate osmoregulatory gill function in crustaceans. In decapod crustaceans, a preparation of the isolated perfused gill in combination with electrophysiological measurements of transepithelial potential difference (PD_te), tracer flux studies and application of specific ion-transport inhibitors has greatly facilitated the analysis of active ion absorption across the branchial epithelium (Habas and Prosser, 1963; Croghan et al., 1965; Habas Mantel, 1967; King and Schoffeniels, 1969; Péqueux and Gilles, 1978; Siebers et al., 1985). Ten years ago, a preparation consisting of an isolated split-gill half-lamella mounted in a micro-Ussing chamber was developed (Schwarz and Graszynski, 1989) to analyze the mechanisms of active osmoregulatory ion uptake across the gills of the chinese crab Eriocheir sinensis. Since then, several other crab species have been investigated using this method, resulting in a more detailed knowledge of the mechanisms involved in transepithelial ion transport in these species (Schwarz, 1990; Onken and Siebers, 1992; Onken, 1996; Riestenpatt, 1995; Riestenpatt et al., 1996; Postel et al., 1998). By splitting the gill lamella, a single epithelial layer covered by an apical cuticle is obtained, allowing area-related short-circuit currents (I_sc) and conductances (G_te) to be measured. Moreover, advanced electrophysiological techniques such as measurements of membrane potentials using microelectrodes (Onken et al., 1991) or current fluctuation analysis (Zeiske et al., 1992) could be applied. Recently, even measurements of area-specific ion fluxes have been successfully performed in Ussing chamber experiments with crab gill epithelia (Riestenpatt, 1995; Riestenpatt et al., 1996).

Presumably because of the small dimensions of isopod pleopods and particularly the fragility of their rami, physiological experiments using perfusion techniques or split lamellae mounted in an Ussing chamber have not been attempted until now. Our interest focuses on the euryhaline valviferan isopod Idotea baltica (Pallas), which inhabits coastal environments with a wide range of salinities between a minimum of 3–4 ‰ and an upper limit of approximately 85 ‰ (see Sywula, 1964). In the present study, exo- and endopodites of I. baltica were split and mounted in a micro-Ussing chamber, allowing electrophysiological measurements on pleopods from isopod crustaceans for the first time. Furthermore, Na⁺/K⁺-ATPase activity was investigated in the exo- and endopodites of isopods acclimated to different salinities. Histological investigations were included to compare the morphological organization of the podites of I. baltica with the structures present in the gills of crabs.

Isopods, Idotea baltica (Pallas), were caught on the rocky shore of the island of Helgoland in the North Sea. In the laboratory, they were maintained in sea water at a salinity of 30 ‰ and a temperature of 16 °C. Brown algae (Ascophyllum sp.) served as a habitat substratum and source of nutrition. Isopods were acclimated to salinities of 15, 20 and 40 ‰ at least 2 weeks prior to experimentation. Because of their larger body size (approximately 3 cm total length), only males were used in experiments.

The final test volume of 1.125 ml was composed of (in mmol l⁻¹): 100 imidazole/HCl, 150 NaCl, 100 NH₄Cl, 5 MgCl₂, 5 ATP, 2.5 PEP, 0.5 NADH, 20 units ml⁻¹ LDH and 5 units ml⁻¹ PK (pH 7.25). Assays were performed at 25 °C in the absence and in the presence of 5×10⁻³ mol l⁻¹ ouabain, a specific inhibitor of the Na⁺/K⁺-ATPase that blocks its activity completely. The reduction in total ATPase activity resulting from the presence of ouabain represents Na⁺/K⁺-ATPase activity. For calculations of activity, we used the linear changes in absorbance between 5 and 15 min after starting the reaction with 0.05 ml of homogenate. The detection limit of the assays was about 0.05 µmol P_i mg⁻¹ protein h⁻¹. The method was calibrated using the commercial test for inorganic phosphate (Sigma). The protein concentrations of the homogenates were determined using a bicinchoninic acid (BCA) protein assay reagent (Pierce, Rockford, IL, USA). Homogenates were stored at −30 °C for no longer than 1 day prior to analysis of protein content.

After the isopods had been killed by decapitation, the pleopods were excised from the abdomen under a binocular microscope (Olympus, Hamburg, Germany) and the endo- and exopodites were separated from the protopodites. For splitting, isolated podites were held in position using a pair of tweezers while a second pair of tweezers was used to draw the two facing external layers of the podites from each other. In this way, the podites were split into two half-podites, each consisting of an apical and a basolateral epithelial surface. The apical surface of one of the two half-podites is directed towards the sediment, while the apical surface of the other half-podite faces the ventral body wall of the pleon. For clarity, these epithelial layers are termed the sediment-facing and the pleon-facing half-lamella.

One half-lamella was mounted between the two half-chambers of an Ussing-type chamber modified after De Wolf and Van Driessche (1986). The half-chambers were then connected to each other using plastic screws. To minimize edge damage, silicone grease was used to seal the edges of the preparation. Using a peristaltic pump (Ismatec, Zurich, Switzerland), the chamber compartments were perfused continuously with saline (see below) at a rate of 40 ml h⁻¹. In cases where ouabain was used, the drug was added to the basolateral saline and, in one indicated case, to the apical saline. The transepithelial potential differences (PD_te) of the mounted half-podites were measured using Ag/AgCl electrodes (Mettler Toledo, Greifensee, Switzerland) connected to the chamber compartments via agar bridges (3 % agar in 3 mmol l⁻¹ KCl). To short-circuit the PD_te, a second pair of electrodes serving as current electrodes connected the system to an automatic clamping device (VCC 600, Physiologic Instruments, San Diego, CA, USA) through agar bridges. The measured short-circuit current (I′_sc) represents charge movement across both the salines and the preparation. To calculate the area-specific resistance (R′_te) between the tips of the voltage electrodes, small induced voltage pulses (2 mV, ΔPD_te) and the resulting current deflections (ΔI) were used. The relatively low values of R_te require a correction of the R′_te and I′_sc data to obtain values for the preparation alone (R_te, I_sc). To determine the solution resistance (R_s), measurements were carried out without a preparation separating the chamber compartments.

Depending on the diameter of the area between the two half-chambers used (0.7, 1.1 and 1.6 mm), R_s varied between 8.1±0.5 and 14.5±0.3 Ω cm² (means ± S.E.M., N=10). The corrected R_te was obtained by subtracting R_s from R′_te, while the correction of I′_sc followed Ohm’s law. The transepithelial conductance, G_te, was calculated as G_te=1/R_te. In the Results section, only the corrected values are given.

After separation from the thorax, the pleon was immediately immersed in 6 % glutaraldehyde in 0.1 mol l⁻¹ phosphate buffer, pH 7 at 4 °C. While the pleon was still immersed in this fixative, the endopodites of the fifth pleopods were excised from the protopodite and kept in the fixation solution for 24 h. Fixed endopodites were washed with phosphate buffer (0.1 mol l⁻¹, pH 7) and then post-fixed in buffered 1 % OsO₄ for 1 h at 4 °C. Endopodites were then washed three times for 10 min with phosphate buffer followed by dehydration in an ethanol series increasing from 35 to 100 %. Samples were then treated twice for 15 min in propylene oxide before being embedded in Epon. Semi-thin (1 µm) and ultra-thin (70 nm) sections were cut with an ultramicrotome (Ultracut, Reichert-Jung, Vienna, Austria) using a glass knife and a diamond knife, respectively. For light microscopic (Olympus BH-2, Hamburg, Germany) investigation, the semi-thin sections were stained with Toluidine Blue (0.8 % Toluidine Blue, 0.8 % borax, 0.2 % pyronin) for 1 min at 70 °C, followed by two washes with distilled water for 1 min at 70 °C. Ultra-thin sections were contrasted with uranyl acetate and lead citrate and viewed in a transmission electron microscope [EM 902, Zeiss(Leo), Oberkochen, Germany].

To acclimate the isopods to various salinities, sea water from the North Sea was diluted with distilled water or concentrated with sea salt (Instant Ocean). The haemolymph-like saline employed in the electrophysiological measurements was composed of (in mmol l⁻¹): 248 NaCl, 5 KCl, 5 CaCl₂, 4 MgCl₂, 2 NaHCO₃, 5 Hepes and 2 glucose. The pH was adjusted to 7.75 with Tris before use.

All values are presented as means ± S.E.M. Differences between groups were tested using a Student’s t-test, assuming statistical significance when P ⩽ 0.05.

Isopods were maintained in sea water of 30 ‰ salinity, and three groups were acclimated for at least 2 weeks to salinities of 15, 20 and 40 ‰. We did not succeed in acclimating specimens of I. baltica to salinities below 15 ‰. To measure Na⁺/K⁺-ATPase activity in the endo- and exopodites of the pleopods, the right and left endo- and exopodite, respectively, of each pair of pleopods were pooled, homogenized and analyzed for enzyme activity. In all exopodites and in the two anterior endopodites (numbers 1 and 2) of seawater-adapted (30 ‰ salinity) specimens, Na⁺/K⁺-ATPase activity was less than 1 µmol P_i mg⁻¹ protein h⁻¹ (Fig. 1C). The posterior endopodites (numbers 3–5) exhibited significantly higher activities of between 2 and 4 µmol P_i mg⁻¹ protein h⁻¹. Following acclimation of isopods to brackish water of 20 ‰ salinity, the Na⁺/K⁺-ATPase activity in posterior endopodites 3–5 was greatly elevated, on average to 8.4±1.2 µmol P_i mg⁻¹ protein h⁻¹ (N=16) (Fig. 1B). Further significant increases to 18.2±1.4 µmol P_i mg⁻¹ protein h⁻¹ (N=18) were observed in posterior endopodites 3–5 of individuals acclimated to 15 ‰ salinity (Fig. 1A). Remarkably, an increase in the acclimation salinity to 40 ‰ also resulted in elevated enzyme activities in the posterior endopodites (Fig. 1D). In contrast to the high enzyme activities of posterior endopodites 3–5, which were shown to be greatly dependent on the acclimation salinity, Na⁺/K⁺-ATPase activities in all other podites were much lower and were independent of ambient salinity levels.

The finding that endopodites 3–5 show enhanced Na⁺/K⁺-ATPase activity that is markedly dependent on ambient salinity suggests that these podites are the main sites of osmoregulatory ion absorption. To investigate further, we applied an independent methodological approach used previously by other authors on crab gills (Schwarz and Graszynski, 1989). Isolated endo- and exopodites 1–5 were split and mounted in an Ussing-type chamber. Electrophysiological measurements were carried out while superfusing the preparation symmetrically with 248 mmol l⁻¹ NaCl standard saline. To investigate the osmoregulatory abilities of the pleopods of I. baltica, transepithelial potential differences (PD_te), short-circuit currents (I_sc) and transepithelial conductances (G_te) were measured in the endo- and exopodites of specimens acclimated to brackish water of 20 ‰ salinity. While endopodites 3–5 exhibited an inside-negative PD_te of −6.7±0.6 mV (N=15) (Fig. 2), almost no PD_te was generated by preparations of exopodites and endopodites 1 and 2. Among endopodites 3–5, the potential differences measured did not differ significantly. When the PD_te was clamped to zero, a mean I_sc of −445±34 µA cm⁻² was observed in endopodites 3–5 (N=15) (Fig. 2). In contrast, mean I_sc values close to zero were measured in all exopodites and in endopodites 1 and 2. A mean transepithelial conductance (G_te) of 84.1±4.7 mS cm⁻² was measured in the three posterior endopodites. Exopodites 2–5 and endopodites 2 showed comparable values for G_te (94.4±8.7 mS cm⁻², N=23), while the first exo- and endopodites exhibited a significantly lower conductance of 13.9±4.1 mS cm⁻² (N=8). Differences in PD_te, I_sc and G_te between the sediment-facing and the pleon-facing half-lamellae of endopodites 3–5 were not detectable (Fig. 3).

To determine the interrelationship between the pronounced Na⁺/K⁺-ATPase activities and the short-circuit currents in posterior endopodites, the effects of ouabain on I_sc and G_te of endopodites 5 of specimens acclimated to 20 ‰ salinity were examined. Application of ouabain (5 mmol l⁻¹) in the basolateral superfusate resulted in an almost total loss of I_sc from −531±28 to −47±5 µA cm⁻² (N=10) (Fig. 4). Following the removal of ouabain, I_sc recovered to −226±20 µA cm⁻². The transepithelial conductance (G_te), was only slightly sensitive to ouabain, being significantly reduced from 85.4±3.9 to 75.5±2.5 mS cm⁻² (N=10) (Fig. 4). Following removal of ouabain, G_te did not recover but decreased further during the following 45 min to 69.7±2.8 mS cm⁻². For a more detailed analysis of the blocking effect of ouabain, the inhibitor concentration was elevated in a stepwise fashion. In response, I_sc decreased in a concentration-dependent manner. Expressing the I_sc data as a Hanes–Woolf plot, saturation kinetics for ouabain were obtained (Fig. 5). The linear correlation (r=0.999) of the data indicates simple Michaelis–Menten kinetics with a half-maximum inhibitor concentration K_oua of 1.3×10⁻⁴±0.14×10⁻⁴ mol l⁻¹. The maximum current decrease ΔI_max is −545±82 µA cm⁻². Apical application of 10 mmol l⁻¹ ouabain had no effect on I_sc and G_te.

What does the procedure of splitting the podites into two halves, to be mounted in an Ussing chamber, mean in terms of the structural organization of the gills? Histological studies were performed on pleopods using both microscopic and ultramicroscopic techniques. To facilitate handling and to allow comparisons with electrophysiological results, the largest endopodites (number 5) were selected. As shown in Figs 6 and 7, the flattened endopodites of I. baltica consist of two half-lamellae of epithelial monolayers of ionocytes located opposite each other and covered by cuticle. A fenestrated lamellar septum of connective tissue was present between the sediment-facing and the pleon-facing half-lamellae. At intervals, the sediment-facing and the pleon-facing sides of the podites are connected by bundles of opposed pillar cells forming column-like structures.

The ionocytes of the epithelial layer exhibit three structurally different zones. An apical region located directly under the cuticle is characterized by numerous infoldings of the apical plasma membrane (zone 1) (Fig. 8). On the cytosolic sides of the folds, the distances between the plasma membranes often increase to form bleb-like endings which enlarge the extended subcuticular space between the folds. Numerous mitochondria are concentrated in the middle zone (zone 2) consisting of a relatively homogeneous cytoplasm without infoldings. The mitochondria are often elongated, and most of them are closely associated with the cytoplasmic endings of the apical folds. A few of them are found in close contact with basolateral invaginations. In the areas located close to the apical and basolateral cell surfaces, the folds are almost free of mitochondria. Occasional vesicle-like structures are located in the cytoplasmic region. The basolateral region (zone 3) shows many interdigitations of the plasma membranes of adjacent cells. This area is separated from the haemolymph space by a basal lamina. The infoldings of the apical and basolateral plasma membranes are located at right angles to the side of the cell (Fig. 8). The comparatively large cells of the lamellar septum in which, in contrast to the ionocytes, the apical and the basolateral sides cannot be distinguished contain numerous glycogen granules (not illustrated). This tissue separates the extracellular space into an upper and a lower sinus containing the circulating haemolymph.

The pillar cells (Figs 6, 7, 9) traverse the whole epithelium. Their apical sides adjoin the cuticle and are structured in the same way as the ionocytes, with numerous elongated mitochondria located within the perpendicular infoldings of the strongly invaginated plasma membranes. In the region of their basal sides, the two facing groups of pillar cells penetrate the lamellar septum and contact in the middle of the connective tissue in a zone characterized by dense bundles of microtubules forming a zig-zag junctional complex.

The highest Na⁺/K⁺-ATPase specific activities detected in the posterior endopodites of isopods acclimated to 15 ‰ salinity were roughly 20 µmol P_i mg⁻¹ protein h⁻¹. Comparable specific activities have been found in the posterior gills of shore crabs Carcinus maenas acclimated to salinities of 10 and 20 ‰ (Siebers et al., 1982). The central role of the enzyme as the driving force in the active osmoregulatory net ion uptake of crabs has been demonstrated by tracer flux experiments (Lucu and Siebers, 1987; Riestenpatt et al., 1996). The increased activities of the Na⁺/K⁺-ATPase measured in the three posterior endopodites of I. baltica compared with activities in all the exopodites and the anterior endopodites, in combination with the large salinity-dependent alterations in activity observed only in the posterior endopodites (Fig. 1), indicate that these podites are the main sites of active osmoregulatory NaCl absorption.

This mode of distribution of Na⁺/K⁺-ATPase activity and its salinity-dependence has also been described in the pleopods of other isopods, e.g. Sphaeroma serratus (Thuet et al., 1969; Philippot et al., 1972) and Idotea wosnesenskii (Holliday, 1988). Our deduction of the important role of the posterior endopodites of I. baltica in active ion absorption corresponds with the results of silver-staining experiments performed by Hørlyck (1973), who found that, in this species, only these three podites are stained by silver and therefore considered to be permeable to chloride. As in isopods, the salinity-dependent Na⁺/K⁺-ATPase in various euryhaline decapods is also located in posterior branchial appendages (Mantel and Olson, 1976; Péqueux and Gilles, 1978; Spencer et al., 1979; Neufeld et al., 1980; Siebers et al., 1982).

In the absence of any external electrophysiological driving force, PD_te measured in the posterior three half-endopodites superfused on both sides (symmetrically) with the same NaCl saline (Fig. 2) must be considered to be an active transport potential. The transepithelial short-circuit currents (I_sc) measured under the same conditions (Fig. 2) also provides evidence for active electrogenic net ion movement (Riestenpatt et al., 1996). The polarity of the PD_te (inside-negative) and the negative, inwardly directed I_sc correspond with the findings of other authors using crab gills in Ussing chamber experiments (Eriocheir sinensis, Schwarz, 1990; Onken et al., 1991; Zeiske et al., 1992; Riestenpatt, 1995; Uca tangeri, Schwarz, 1990; Carcinus maenas, Onken and Siebers, 1992; Riestenpatt, 1995; Riestenpatt et al., 1996). The comparatively consistent absence of measurable PD_te and I_sc in the other podites suggests that these play little part in active osmoregulatory ion uptake. This result correlates with the distribution of Na⁺/K⁺-ATPase activity (Fig. 1). The dominant role of the posterior endopodites in active ion uptake is also obvious from ultrastructural investigations in the isopods Asellus aquaticus (Babula, 1979) and Mesidotea entomon (Babula and Bielawski, 1981). The strict functional differentiation within the two rami of the pleopods suggests that, besides the known morphological segmentation in arthropods, a physiological tagmatization exists in sequential organs such as pleopods. This is not immediately obvious but, as our physiological experiments suggest, is present and fully operative. The functional heterogeneity in the pleopods of isopods is remarkably similar to that described in the gills of crabs. The complete elimination of the I_sc in the posterior endopodites of I. baltica by basolateral application of ouabain (Fig. 4) corresponds with the results on posterior gills of C. maenas (Riestenpatt, 1995; Riestenpatt et al., 1996). In isolated posterior gills of C. maenas, the complete abolition of PD_te by ouabain coincides with the abolition of net influx of Na⁺ and Cl⁻ after treatment with this drug (Lucu and Siebers, 1987). Furthermore, the transbranchial I_sc measured in isolated gill half-lamellae of C. maenas was correlated with Cl⁻ influxes (Riestenpatt, 1995; Riestenpatt et al., 1996). From these and other results, it was proposed that the PD_te and I_sc measured in the gills of C. maenas are based on active electrogenic uptake of Na⁺ and Cl⁻ across the cells of the gill epithelium and that the underlying transport mechanism is similar to that described for the thick ascending limb of Henle’s loop of the mammalian kidney. Uptake of NaCl across the apical membrane proceeds via apical Na⁺/K⁺/2Cl⁻ symport. This cotransporter is secondarily energized by the activity of the Na⁺/K⁺-ATPase, which generates a Na⁺ gradient directed into the cell. Branchial Na⁺/K⁺-ATPase in C. maenas was shown to be the only driving force energizing active, electrogenic ion uptake across this epithelium (for details, see Onken et al., 1994; Riestenpatt et al., 1996). Taken together, the pattern of Na⁺/K⁺-ATPase activity (Fig. 1) and the results of the Ussing chamber experiments (Figs 2, 4) suggest that the active ion uptake mechanism located in the three posterior endopodites of I. baltica closely resembles the coupled active electrogenic absorption of NaCl in the posterior gills of C. maenas.

The half-maximal inhibition of I_sc by basolateral ouabain (K_oua=1.3×10⁻⁴±0.14×10⁻⁴ mol l⁻¹) (Fig. 5) measured in the endopodites of I. baltica agrees with ouabain inhibitor constants obtained for gill tissues of other crustaceans such as C. maenas (5.6×10⁻⁵ mol l⁻¹) (Riestenpatt et al., 1996), E. sinensis (5×10⁻⁵ to 6×10⁻⁵ mol l⁻¹) (Schwarz and Graszynski, 1990), several species of fiddler crabs (7×10⁻⁵ to 8×10⁻⁵ mol l⁻¹) (Graszynski and Bigalke, 1986), Callinectes sapidus (1.5×10⁻⁴ mol l⁻¹) (Neufeld et al., 1980) and isopods, e.g. S. serratum (10⁻⁵ mol l⁻¹) (Philippot et al., 1972) and I. wosnesenskii (6.3×10⁻⁵ mol l⁻¹) (Holliday, 1987). In general, Na⁺/K⁺-ATPases in crustaceans are much less ouabain-sensitive than this enzyme from mammalian tissues, which have inhibitor constants in the submicromolar range.

A comparison of electrophysiological data from the branchial epithelia of osmoregulating crustaceans (Table 1) shows that the highest transepithelial conductivity (G_te) of the posterior endopodites is that of I. baltica. The findings indicate that the epithelium of the posterior endopodites of I. baltica is very leaky and allows comparatively large diffusive transepithelial losses of ions into the dilute ambient medium. These losses can only be compensated for by large ion uptake rates, as is evident from the comparatively high transepithelial short-circuit current (I_sc). I. baltica can therefore be considered to be a comparatively moderate osmoregulator, more comparable to C. maenas, which cannot tolerate salinities below approximately 9 ‰, than to the strongly osmoregulating crab E. sinensis acclimated to fresh water. This deduction agrees with observations of the moderate hyperregulatory ability of I. baltica by Hørlyck (1973). In contrast to I. baltica, E. sinensis exhibits a much higher PD_te and much lower G_te and I_sc (Table 1). The ultrastructural results (see below) also demonstrate the leakiness of the epithelium of the posterior endopodites of I. baltica when considering the paracellular pathways in terms of intercellular junctions, which are largely lacking in septate desmosomes (Fig. 8A,B).

The results reported in this study were obtained from specimens of I. baltica caught in the North Sea, which are very seldom exposed to osmotic stress. In the laboratory, these individuals could not be acclimated to salinities below 15 ‰ for periods longer than approximately 1 month. However, I. baltica occupies brackish-water habitats in the Baltic Sea with salinities down to 4 ‰ (Sywula, 1964). This author mentioned local populations with minimum salinity requirements of 3.06–15 ‰. Hørlyck (1973) reported difficulty in keeping specimens of this species collected in the Baltic Sea north of Helsingør in salinities below 10 ‰. A range of osmoregulatory abilities have also been reported for populations of other crustacean species, e.g. for the isopods Gnorimosphaeroma oregonensis (Riegel, 1959) and Mesidotea (Saduria) entomon (Lockwood and Croghan, 1957; Croghan and Lockwood, 1968) and for the green crab C. maenas (Theede, 1969). We want to emphasize that, in the present study, specimens of I. baltica belonging to North Sea populations were investigated because Baltic specimens, which would probably have exhibited higher levels of ion uptake, were considerably smaller, making the preparation, if feasible at all, for Ussing chamber experiments much more difficult.

The idea of splitting the rami of the pleopods of I. baltica and using this preparation for electrophysiological experiments in an Ussing-type chamber originated from the structural similarity between the pleopod rami of several isopod species (Babula, 1977, 1979; Babula and Bielawski, 1981; Bubel and Jones, 1974; Kümmel, 1981; Wägele, 1982) and the gill lamellae of crabs such as C. maenas (Goodman and Cavey, 1990) and E. sinensis (Barra et al., 1983), in which ion uptake has already been examined successfully using this type of experimental approach (Schwarz and Graszynski, 1989; Onken et al., 1991; Onken and Siebers, 1992; Riestenpatt, 1995; Onken, 1996; Riestenpatt et al., 1996; Postel et al., 1998). The endopodite of the fifth pleopod of I. baltica exhibited a microscopic structure (Fig. 8) remarkably similar to that described for the lamellae of phyllobranchiate gills of brachyurans (Drach, 1930; Copeland, 1968; Copeland and Fitzjarell, 1968; Barra et al., 1983; Cioffi, 1984; Compère et al., 1989; Goodman and Cavey, 1990; Taylor and Taylor, 1992; Lawson et al., 1994; Luquet et al., 1997).

The posterior pleopods of I. baltica consisted of two facing monolayers of ionocytes, each covered on the apical side by cuticle. Bundles of pillar cells located at regular intervals between the layers of ionocytes conceivably maintain the structural stability and the uniform width of the haemocoel. The ionocyte layers are separated by a fenestrated lamellar septum of connective tissue. We assume that splitting the podite resulted in the disruption of the lamellar septum and the pillar cells approximately in the area where they traverse it. According to Bubel and Jones (1974), the pleon-facing (‘dorsal’) side of the pleopods of the isopod Jaera nordmanni is folded to enclose a series of haemolymph spaces that are absent on the sediment-facing (‘ventral’) side. In I. baltica, however, we found neither structural nor electrophysiological differences between the two epithelial sides. The absence of physiological differences (Fig. 3) between the sediment-facing and the pleon-facing half-lamellae indicates the functional homogeneity of these sides and suggests that the lamellar septum of connective tissue that remains part of the preparation during the process of splitting a podite does not contribute major polarity properties. Furthermore, we suggest that, during the preparation, the pleopod was disrupted in the region of the lamellar septum and pillar cells without major effects on the physiological functions of the epithelial monolayer. The similarity of the structural elements between crab gills and the pleopods of isopods suggests that a direct comparison of the electrophysiological data obtained from measurements on gill and pleopod half-lamellae mounted in an Ussing chamber is permissible.

This article is based on a doctoral study performed by Ute Postel at the Faculty of Biology, University of Hamburg.