Immunocytochemical Localization of Na +,K + -ATPase in the Calcium-transporting Sternal Epithelium of the Terrestrial Isopod Porcellio scaber L. (Crustacea)

Abstract

Terrestrial isopods store large amounts of calcium carbonate between the epithelium and the old cuticle of the first four anterior sternites before molt. During the formation of these sternal CaCO₃ deposits, large amounts of calcium are transported across the anterior sternal epithelium from the base to the apical side of the integument, and in the reverse direction during resorption of the deposit. A monoclonal antibody against the avian α-subunit of Na⁺,K⁺-ATPase was used to localize Na⁺,K⁺-ATPase in the anterior and the posterior sternal epithelium of Porcellio scaber. Semithin cryosections 0.5 μm thick were used for immunofluorescence microscopy and ultrathin cryosections for immunogold electron microscopy. The Na⁺,K⁺-ATPase was localized in the basolateral plasma membrane of the posterior and anterior sternal epithelium. The apical plasma membrane, including cytoplasmic extensions into the newly secreted cuticle, was virtually devoid of the enzyme. This pattern of immunolocalization was not affected by the direction of transepithelial calcium transport associated with the deposition and resorption phases of the molt cycle.

(J Histochem Cytochem 45:437–446, 1997)

The anterior sternal epithelium (ASE) of the terrestrial isopod Porcellio scaber transports large amounts of calcium and probably also carbonate and protons during the deposition and resorption of sternal deposits. These deposits consist of an amorphous CaCO₃ compound and an organic matrix (Ziegler 1994). They serve as a reservoir for cuticular calcium during the molt cycle of terrestrial isopods (Steel 1993). Isopods molt first the posterior half of the body and, after a short interval of about 1 day, called the intramolt (Price and Holdich 1980), the anterior part of the body (Schöbl 1880). About 2 weeks before molt, the CaCO₃ deposits, which are located in the first four anterior sternites, start to develop, and can be seen as white spots in the intact animal. Messner (1965) and Steel (1993) proposed that during premolt calcium and probably also carbonate ions from the posterior cuticle are first transported across the posterior epithelium into the hemolymph, and then across the ASE into the exuvial space between the epithelium and the old sternal cuticle. During intramolt, these ions are mobilized from the CaCO₃ deposits, transported back into the hemolymph, and used to mineralize the new cuticle. The role of the ASE in the formation and resorption of the CaCO₃ deposits was confirmed by Ziegler (1996), i.e., the ASE, but not the posterior sternal epithelium (PSE), was structurally differentiated for transepithelial ion transport. This morphological differentiation brings about a dramatic increase in the basolateral plasma membrane surface area during late premolt and intramolt, characterized by a system of ramifying invaginations. The latter is designated as an interstitial network (IN). In addition, the apical plasma membranes of the ASE cells are also increased by apical folds during intramolt, and many calcium-containing granules appear within the IN.

It was noted by Cameron (1989) that calcification in crustaceans requires transepithelial transport of Ca²⁺ and HCO^- ₃ from basal to apical poles, and transport of H⁺ in the reverse direction, according to the reaction:

${\text{Ca}}^{\text{2+}}+{\text{HCO}}_3^{-}\rightleftarrows {\text{CaCO}}_3+{\text{H}}^{+}$

This process should be similar in the sternal integument of terrestrial isopods during the formation of the CaCO₃ deposits. Moreover, in the ASE, transport of protons may also be necessary to mobilize the calcium and carbonate ions during resorption of the deposits. Therefore, transport of at least three different kinds of ions across the ASE is necessary during formation and resorption of the sternal CaCO₃ deposits. Nothing is known about the transport mechanisms of the plasma membrane of ASE cells in isopods. However, it is known from many other cell types that, in addition to specific ion pumps, exchange mechanisms play a major role in ion transport (Fröhlich 1989). In the carapace epithelium of Carcinus maenas, an Na⁺,Ca²⁺ exchange mechanism probably contributes to epithelial calcium transport in addition to a Ca-ATPase pump, indicated by ion flux experiments during the mineralization and demineralization of the cuticle (Roer 1980; see also Neufeld and Cameron 1993 for review).

The Na⁺,Ca²⁺ exchanger, like the Na⁺,H⁺ exchanger, utilizes the high Na⁺ gradient across the plasma membrane of the cell to transport ions against an electrochemical gradient. The Na⁺ gradient is established and maintained by the Na⁺,K⁺ pump, which is an integral membrane protein that transports Na⁺ out via exchange for K⁺ into the cell, using metabolic energy derived from ATP hydrolysis (for review see Horisberger et al. 1991). An Na⁺,K⁺-ATPase has been found in a variety of epithelial cells. However, it has never been demonstrated in the integument of terrestrial isopods.

Because of its potential role in Ca²⁺ and H⁺ transport during formation and resorption of CaCO₃ deposits, identification of the cellular localization and distribution of Na⁺,K⁺-ATPase was undertaken. The enzyme was localized in the ASE during late premolt and intramolt, and in the PSE, where the rate of calcium transport is rather low. The results show that Na⁺,K⁺-ATPase is present in the basolateral side of the ASE and PSE during both formation and resorption of CaCO₃ deposits.

Materials and Methods

Animals

Porcellio scaber with a body length of 9–12 mm, maintained as described earlier (Ziegler 1996), was used for investigation. Specimens with visible sternal CaCO₃ deposits were placed individually in small plastic containers and used after the desired late premolt or intramolt stage was entered.

Antibody, Gel Electrophoresis, and Immunoblotting

Mouse monoclonal antibody IgG₅ raised against the α-sub-unit of the avian sodium pump (Takeyasu et al. 1988) was kindly provided by D. M. Fambrough (Baltimore, MD).

The sternal integument and the central neural tissue of the pereions 2–8 of animals with well-developed sternal CaCO₃ deposits were quickly removed and homogenized in ice-cold 2 X sample buffer (1 X = 62.5 mM Tris-HCl, pH 6.8, 10% glycerol, 2% SDS, 650 mM β-mercaptoethanol, 0.025% bromophenol blue), boiled for 2 min, sonicated by two short ultrasonic bursts, and centrifuged at 16,000 X g for 5 min. The supernatant and molecular weight markers (Sigma; St Louis, MO) were diluted with 1 X sample buffer and used for SDS-PAGE. Proteins were electrotransferred onto nitrocellulose sheets.

The blots were preincubated overnight at 4C in Tris-buffered saline (TBST: 10 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.1% Tween 20, 0.005% thimerosal) containing 5% skim milk to block nonspecific binding sites. The blots were rinsed in TBST, incubated overnight at room temperature in primary antibody diluted at 5 μg/ml in TBST containing 1% skim milk, washed three times for 10 min in TBST, and reacted for 1 hr with the secondary antibody (HRP-coupled goat anti-mouse IgG_H&L; HyClone, Logan, UT) diluted 1:10,000 in TBST. Bound antibodies were visualized by incubating the blots in a chemiluminescent detection system (ECL Western blotting detection reagents; Amersham, Little Chalfont, UK) for 1 min, followed by exposure to autoradiographic film (Hyperfilm-ECL; Amersham) for 5–30 sec.

Immunofluorescence Microscopy on Semithin Cryosections

Porcellio scaber was injected with a small amount (10–15 μl) of either 1.25%, 3.125%, or 12.5% glutaraldehyde (GA). These small volumes of GA are quickly diluted by the vigorous circulating hemolymph, by an estimated factor of 4-5. About 1 min after injection the anterior and posterior sternal plates were removed and fixed in a mixture of either 0.25% or 0.5% GA, 2.0% paraformaldehyde (PFA), and 2.25% or 2.0% sucrose in 0.1 M Na-cacodylate buffer, pH 7.5, for 1-2 hr. Cryofixation was performed by a method similar to that described by Tokuyasu (1980). Specimens were immersed in 0.76 M sucrose in 0.1 M PBS for 15 min and in 2.3 M sucrose in PBS for at least another 90 min, dipped and agitated in a droplet of tissue glue (Tissue-Tec; Miles, Elkhart, IN) to improve cutting properties, clamped individually into incisions of prepared pieces of cork, and frozen in liquid propane.

Semithin (0.5-μm) cryosections were used for immunocytochemistry (Takizawa and Robinson 1994) to locate the antigen with a high spatial resolution. Sagittal sections were cut with a Reichert Ultracut S microtome equipped with a FCS cryochamber (Wien, Austria), using glass knives and an antistatic device (Diatome; Biel, Switzerland) at temperatures of −50C for the specimen holder, knife, and cryochamber. Good sections were teased away from the knife edge with an eyelash and transferred to lysine-covered glass slides (Polyprep; Sigma) with a droplet of 2.3 M sucrose in PBS, using a wire loop with a diameter of about 1 mm. Care was taken not to touch the knife with the sucrose solution to avoid freezing of the droplet onto the surface of the knife.

The sucrose was washed off with PBS and the sections were treated successively with (a) 0.01% Tween 20, 150 mM NaCl in 10 mM phosphate buffer, pH 7.3, once for 10 min; (b) 50 mM NH₄Cl in PBS, once for 5 min, (c) blocking solution (BS) containing 1% BSA, and 0.1% gelatin in PBS for 10 min. Small droplets (10 μl) of primary antibody diluted in BS at 20–40 μg/ml were placed on the sections and incubated for 2 hr at RT in a wet chamber. Control sections were incubated in BS without primary antibody. The sections were washed six times for 5 min in BS, incubated for 1 hr in small droplets of secondary antibody (Cy3-conjugated donkey anti-mouse IgG (H & L) (Jackson Immunoresearch; West Baltimore MD) diluted 1:100 or 1:200 in BSA-PBS, washed in BS six times for 5 min, and mounted in 80% glycine, 20% PBS plus 2% N-propyl-gallate (to retard fading) and examined with a Zeiss Axiophot (Jena, Germany). Micrographs were taken on Agfapan APX 25 film.

Figure 1.

Ultrastructure of the anterior sternal epithelium during formation (a) and resorption (b) of the CaCO₃ deposits. bl, basal lamina; ce, cellular extensions; cu, cuticle; f, folds; IN, interstitial network; m, mitochondria; n, nucleus.

Figure 2.

Western blot analysis of homogenized sternal tissue of Porcellio scaber. The monoclonal antibody IgG₅ raised against the α-sub-unit of the avian Na⁺,K⁺-ATPase specifically labels one band of about 110 kD.

Immunogold Electron Microscopy

Ultrathin cryosections were cut with a diamond knife (Diatome) from the same specimens as for immunofluorescence microscopy, at a temperature of −120C. Immunolabeling was performed similar to the method described by Tokuyasu (1980). Good (∼80-nm-thick) sections were transferred as described for semithin sections (see above) onto nickel grids covered with a carbon-coated formvar film. Grids were collected and placed on 2% gelatin containing 50 mM glycine on ice. The grids were then placed on droplets of 50mM glycine in PBS once for 5 min and 1% BSA-PBS three times for 5 min, transferred to small (10-μl) droplets of primary antibody (anti-Na,K-ATPase diluted to 25 μg/ml in BSA/PBS), and incubated for 2 hr at RT in a wet chamber. Control sections were incubated in BSA-PBS without primary antibody. The sections were washed six times for 5 min in BSA-PBS, and reacted with 10-nm gold-conjugated goat anti-mouse IgG for 1 hr at RT in a wet chamber. Samples were washed once for 5 min in BSA-PBS and three times for 5 minutes in PBS, postfixed in 2.5% GA in 0.1 M phosphate buffer pH 7.3, for 10 min, washed once for 5 min in PBS, and three times for 5 min in H₂O. Sections were stained on a droplet of 1.8% polyvinyl alcohol (PVA) and 0.2% uranyl acetate (UA) for 10 min. The grids were removed from the droplets using a wire loop slightly larger than the diameter of the grid. Excess solution was absorbed with filter paper, and the grids were dried in the loop and examined in a Philips 400 or Zeiss CEM 902 electron microscope. The number of colloidal gold particles on membrane segments 1 μm long were determined on micrographs of the sternal epithelia. One-way ANOVA was used to analyze the statistical significance of differences among various membranes.

Conventional Electron Microscopy

The animals were prefixed by injecting 10–15 μl of 12.5% GA in 0.1 M cacodylate buffer. The sternites were removed and fixed in 2.5% GA and 2% PFA in 0.1 M cacodylate buffer (Mead et al. 1976) for 1 hr. Specimens were postfixed in 2% OsO₄ + 0.8% K₄Fe(CN)₆ in 0.1 M cacodylate buffer, dehydrated in a series of propanol concentrations and embedded in Epon. Ultrathin sections were cut on a Reichert Ultracut microtome and stained with 2% uranyl acetate in H₂O and 0.3% lead citrate. Specimens were viewed in a Philips 400T electron microscope at 80 kV.

Results

Morphology of the Integument

The ultrastructural organization of the anterior sternal integument during premolt and intramolt is shown in Figure 1. The integument consists of a thin (≤1-μm) cuticular layer, the epithelial cells, and a basal lamina. The electron micrographs show the enormous increase of the basolateral plasma membrane surface by the interstitial network (IN), which is especially prominent in the basal half of the anterior sternal epithelium (ASE). During intramolt, many folds considerably increase the apical surface of the ASE. No IN and no apical folds occur in the posterior sternal epithelium (PSE) (not shown), and the cuticular layer of the PSE is much thicker than in the ASE (see Figures 3a and 3g). For a more detailed description see Ziegler (1996a).

Immunolocalization of the Na⁺,K⁺-ATPase

A monoclonal antibody, IgG₅, was used to identify the α-subunit of the Na⁺,K⁺-ATPase in Porcellio scaber. In Western blots of solubilized sternal tissue, the antibody binds specifically to a single band (Figure 2). The apparent molecular weight of this protein was 110 kD.

All combinations of the various injection and fixation procedures tested yielded good structural preservation and excellent antigenity. Immunofluorescence microscopy shows that the localization of the Na⁺,K⁺-ATPase in regions of the plasma membrane of the epithelial cells is distinct and is well-resolved in semithin cryosections. Strong immunoreactivity of the antibody was observed along the entire basolateral sides of the ASE and PSE cells but was absent from the apical plasma membrane below the cuticle (Figure 3). In the basolateral region of the ASE, anti Na⁺,K⁺-ATPase binds to a larger area than in the PSE (Figures 3b, 3d, and 3f). This corresponds to the larger surface area of the IN formed by the basolateral plasma membrane of the ASE. The immunostaining pattern in the ASE and the PSE was maintained during premolt and intramolt (Figures 3b and 3d; not shown for the PSE). Immunolabeling for Na⁺,K⁺-ATPase was also present in the plasma membranes of the neural tissue of the sternal ganglia (Figure 4). Only very weak immunofluorescence was seen in the plasma membrane of blood cells (Figures 3a and 3b). All controls were devoid of any specific binding (e.g., Figures 3g and 3h).

At the ultrastructural level, the distribution of the Na⁺,K⁺-ATPase corresponded well with that observed by immunofluorescence microscopy. Gold particles were well-confined to the basolateral plasma membrane of the ASE and the PSE (Figures 5 and 7). In areas at which the plasma membrane was cut perpendicular to the surface, gold particles were present exclusively at the cytoplasmic side of the plasma membrane (Figures 5c-5f).

Figure 3.

Localization of the Na⁺,K⁺-ATPase. (a,b) Anterior sternal epithelium during late premolt. (c,d) Anterior sternal epithelium during intramolt. (e,f) Posterior sternal epithelium during intramolt. (g,h) Control. (b,d,f,h) Fluorescence micrographs; (a,c,e,g) corresponding phase-contrast pictures. Arrows, apical side of the plasma membrane. bc, blood cells; cu, cuticle; n, nucleus.

Figure 4.

Localization of the Na⁺,K⁺-ATPase in the sternal ganglion (G) and anterior sternal epithelium (E).

Essentially no gold labeling was observed in the apical plasma membrane (Figure 7), either below the cuticle (Figures 5a and b) or in the cytoplasmic extensions that protruded into the cuticle (Figure 6 a). However, in two cases only, a dense structure in the center of the cytoplasmic extensions was intensely labeled (Figure 6 c). This is noteworthy because structures of similar size were labeled within the inner epicuticle (Figure 6 b), where cellular extensions are absent.

Discussion

Methodological Considerations

The use of semithin sections in immunofluorescence microscopy yielded high spatial resolution, which allowed visualization of the plasma membranes of the structurally complex interstitial network (IN) of the anterior sternal epithelium (ASE) (Figures 3b and 3d). The reduced signal intensity in semithin sections (0.5 μm) compared to thick sections was partly compensated for by the use of cyanine fluochrome-conjugated secondary antibody (Cy3). The latter has about twice the intensity of tetramethyl rhodamine (TRITC) conjugates.

Specifity of the Immunolabeling

The Na⁺,K⁺-ATPase is an integral transmembrane protein that maintains the Na⁺ and K⁺ gradients in almost all animal cells. It occurs in high concentrations in excitable and epithelial transport tissues. The protein consists of two subunits, a catalytic α-subunit of about 110 kD, which lies predominantly at the cytoplasmic side, and a lighter, predominantly extracellular glycoprotein β-subunit (for reviews see Fambrough et al. 1994 and Horisberger et al. 1991). The α-subunit is very well conserved throughout evolution. In Drosophila, 80% of the amino acid residues are similar to α-subunit sequences reported for vertebrates (Lebovitz et al. 1989), and the α-subunit of the crustacean Artemia sanfranciscana has about 69–72% homology (Baxter-Lowe et al. 1989).

The present study used an antibody that had been raised against the cytoplasmic domain of the α-subunit of chicken Na⁺,K⁺-ATPase (Takeyasu et al. 1989). This antibody binds specifically to all three known α-subunits (α₁–α₃). Crossreactivity with the α-subunit of epithelia in insects includes Drosophila melanogaster (Lebovitz et al. 1989), Apis mellifera (Baumann and Takeyasu 1993), Calliphora erythrocephala (Baumann et al. 1994), Periplaneta americana (Just and Walz 1994), and vertebrates: Oncorhynchus mykiss (Witters et al. 1996).

In this study the antibody specifically recognized a single band with an apparent molecular weight of about 110 kD. This is within the range reported for α-subunits from vertebrate tissues. Furthermore, the antibody recognized an epitope at the cytoplasmic face of the plasma membrane. These properties strongly indicate that the antibody binding polypeptide corresponds to the Na⁺,K⁺-ATPase α-subunit.

Localization of the Na⁺,K⁺-ATPase

The restricted distribution of Na⁺,K⁺-ATPase to the basolateral side of the epithelium correlates well with immunocytochemical and histochemical localization in non-excitable epithelial tissues of a variety of animals (insects: Just and Walz 1994; Zimmermann 1992; crustacea: Sun et al. 1991; Warburg and Rosenberg 1989; teleosts: Witters et al. 1996; and mammals: Nakazawa et al. 1995; Iwano et al. 1990; Siegel et al. 1984). Asymmetric distribution of Na⁺,K⁺-ATPase to the basolateral membrane of epithelial cells allows the vectorial transepithelial transport of ions and water across cell layers (Horisberger et al. 1991).

Figure 5.

Immunogold localization of the Na⁺,K⁺-ATPase on ultrathin cryosections. (a,b) Two examples of the apical region of the anterior sternal epithelium during intramolt. (c-e) Basolateral plasma membrane of the anterior sternal epithelium during late premolt (c) and intramolt (d,e) Arrowhead (e) points to a tangentially cut portion of the plasma membrane. (f) Basolateral plasma membrane of the posterior sternal epithelium during intramolt. CP, cytoplasm; cu, cuticle; IN, intercellular network; m, mitochondria; og, osmiophilic granules; za, zonula adherens.

Figure 6.

Immunogold localization in ultrathin cryosections of the freshly secreted anterior sternal cuticle (cu). (a) Tangentially cut section; Bar = 0.5 μm. (b) Sagittal section of the epi-and exocuticle. (c) Oblique section through the apical region of the sternal epithelium. ec, epithelial cell; ep, epicuticle; ex, exocuticle; large arrows, cytoplasmic extensions of the epithelial cells into the cuticle; small arrows, membrane plaques.

Figure 7.

The number of colloidal gold particles per μm membrane (means ± SEM). (A) Basolateral plasma membrane of the ASE during intramolt. (B) Basolateral plasma membrane of the ASE during late premolt. (C) Basolateral plasma membrane of the PSE during intramolt. (D) apical plasma membrane of the ASE during intramolt. (E) Membrane of the nuclear envelope of the ASE during intramolt. (F) Outer membrane of the mitochondria of the ASE during intramolt. The numbers of membrane segments analyzed are given in brackets: ns; not significant.

In the ASE and the posterior sternal epithelium (PSE) of Porcellio scaber, the basolateral localization is maintained during premolt and intramolt. This indicates that the direction of transepithelial transport of calcium, and probably carbonate as well as protons, which changes direction during the transition from CaCO₃ deposit formation to resorption, is independent of the location of the Na⁺,K⁺-ATPase. However, a change in the location of the Na⁺,K⁺-ATPase is not expected. Mechanisms such as the Na,Ca exchanger or the Na,H exchanger require the Na,K-ATPase in an indirect way by utilizing the Na gradient maintained by the pump.

The quantitative differences between the gold labeling density in the ASE and the PSE (Figure 7) must be considered with some degree of caution. The complex morphology of the ASE cell surface may lead to an overestimation of the number of colloidal gold particles compared to that of the PSE. However, it is clear from the areas at which the ASE cells were cut perpendicular to the surface of the plasma membrane (Figures 5c-5f), and from the immunofluorescent labeling (Figure 3), that the density of the Na⁺,K⁺-ATPase within the plasma membrane of the ASE is at least not less than that in the PSE. This leads to the conclusion that the total Na⁺,K⁺- ATPase is much greater in the ASE than in the PSE owing to the enormously increased basolateral plasma membrane of the ASE during pre-and intramolt. Because the basolateral membrane is not increased during early premolt stages (Ziegler 1997), an increased net synthesis of Na⁺,K⁺-ATPase molecules must accompany the development of the IN during later premolt stages.

The function of an increased amount of Na⁺,K⁺-ATPase may be to compensate for the presumed increase in Na⁺ influx and K⁺ efflux that would result from an increased basolateral plasma membrane surface area. In addition, the Na⁺,K⁺-ATPase may function to balance an increased Na⁺ load, possibly caused by an increased Na⁺-dependent ion exchange during formation and resorption of CaCO₃ deposits. A role for the Na⁺,K⁺-ATPase in transepithelial ion transport in the ASE is further supported by an intense immunofluorescent labeling comparable to that in isopod neural tissue (Figure 4). The strong labeling in the PSE suggests that, during molt, the Na/K-ATPase may also play a role in the mineralization and demineralization of the cuticle.

The densely labeled structures within the epicuticle may represent nonspecific binding of the antibody to a cuticular protein that is not solubilized by the sample buffer. Alternatively, the densely labeled structures in the inner epicuticle and in the core of cytoplasmic extensions into the cuticle suggest that α-subunits may be deposited into the inner epicuticle. The inner epicuticle of crustaceans contains a wide variety of proteins (Skinner et al. 1992). However, little is known about the tissue specifity of cuticular proteins. It is unlikely that the α-subunit of the Na⁺,K⁺-ATPase within the epicuticle is functionally involved in transepithelial ion transport. Therefore, localization of α-subunit epitopes in the inner epicuticle may indicate that functionally irrelevant proteins or protein fragments add to the proteinous composition of the epicuticle. It is possible that the epicuticle serves as a site for discarding proteins that are destined for degradation in the course of protein turnover.

In conclusion, these results suggest that an Na,K-ATPase localized at the basolateral side of the ASE maintains an Na gradient that contributes to the transepithelial transport of at least one of the ions involved in formation and deposition of CaCO₃ deposits. The present study provides a firm basis for further investigations on vectorial Na-dependent ion transport across crustacean calcium transporting epithelia.