Histochemistry (1992) 98 : 183-197

Histochemistry © Springer-Verlag 1992

Quantitative immunogold localization of Na, K-ATPase along rat nephron T. Takada, A. Yamamoto, K. Omori, and Y. Tashiro Department of Physiology, Kansai Medical University, Moriguchi, Osaka 570, Japan Accepted June 25, 1992 Summary. Ultrastructural localization of Na, K-ATPase e-subunit along rat nephron segments was investigated quantitatively by immunogold electron microscopy on LR-White ultrathin sections using affinity-purified antibody against e-subunit of the enzyme. Ultrathin sections were incubated with the antibody at a saturation level and the number of gold particles bound per gm of the plasma membrane (particle density) of the tubular epithelial ceils from the proximal tubule to the collecting duct was determined. In all the tubular epithelial cells, gold particles were located exclusively on the basolateral surface, and no significant binding of gold particles to the apical surface was observed. Distribution of gold particles on the basolateral membranes was quite heterogeneous; lateral membranes and infolded basal membranes were highly labeled, whereas the basal membranes which are in direct contact with the basal lamina were scarcely labeled. The average particle density on the basal surface was highest in the distal straight tubule cells (11.4 units), very high in the distal convoluted tubule cells (9.8 units), intermediate in the proximal tubule cells (3.3 units), in the connecting tubule cells (4.3 units), and in the principal cells of the collecting duct (5.6-3.8 units), low in the thin limb of Henle's loop (1.0 unit), and at the control level in the intercalated cells in the connecting and collecting duct. The relative number of gold particles/mm nephron segment and the relative number of gold particles in the various nephron segments were calculated using quantitative morphological data. The estimated distribution profile of the former was in good agreement with the Na, K-ATPase activity profile in rat nephron, which was determined biochemically with a microenzymatic method.

Introduction It has been well established that Na, K-ATPase in nephron plays a pivotal role in the active translocation of Correspondence to." Y. Tashiro

Na + and K + across kidney cell membrane as well as in the secondary active transport of a number of other solutes (J6rgensen 1986; Katz 1982). The quantitative localization of Na, K-ATPase along renal tubules is, therefore, very important in order to understand the functions of the various nephron segments. The distribution of Na, K-ATPase along a nephron has been investigated biochemically using tubule microdissection and microanalytical techniques (Doucet 1988; Doucet et al. 1979; Katz 1986, 1988; Katz et al. 1979; Schmidt and Dubach 1969), by cytochemistry (Ernst 1975; Ernst and Schreiber 1981; Kuwahara et al. 1982; Mayahara and Ogawa 1980; Laborde et al. 1990), by radioautographic localization of [3H]ouabain (Shaver and Stirling 1978) and by immunocytochemistry (Baskin and Stahl 1982; Kashgarian etal. 1985; Kyte 1976a, b). In structurally complex organs like kidney, utilization of electron microscopic methods is indispensable to map the distribution of Na, K-ATPase in detail. In fact, immunoelectron microscopic localization ofNa, K-ATPase along nephron segments has been reported using polyclonal (Baskin and Stahl 1982; Kyte 1976a, b) and monoclonal antibodies (Kashgarian et al. 1985). To our knowledge, however, no quantitative immunoelectron microscopic analysis on Na, K-ATPase along nephron segments has been reported. In this report we have determined quantitatively the ultrastructural localization of Na, K-ATPase in all the rat renal epithelial cells by incubating ultrathin sections with affinity-purified antibody against the c~-subunit of rat kidney Na, K-ATPase. The number of gold particles bound per gm (particle density) was determined on the plasma membrane of the various domains of the tubular epithelial cells from the proximal tubules to the collecting duct at a saturation level. A preliminary report of this work was presented at the 42nd Annual Meeting of the Japan Society for Cell Biology (Takada et al. 1989).

184

Materials and methods

Animals Male Sprague-Dawley rats weighing 150-200 g were used and were given a commercial chow and water ad libitum.

Purification of holo Na, K-ATPase and preparation of antibody against the c~-subunit Na, K-ATPase was purified from rat kidney, and rabbit antibody against the holo Na, K-ATPase was prepared as described by Akayama et al. (1986). Monospecific antibody against the c~-subunit of rat kidney Na, K-ATPase was affinity-purified from the antibody by a modification of the method of Smith and Fisher (1984). Briefly, holo Na, K-ATPase was separated into the c~- and/~-subunits by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) according to Laemmli (1970), and transferred to a nitrocellulose sheet (Millipore GV; Nihon Millipore Kogyo Co. Ltd., Yonezawa, Japan) as described by Burnette (1981). The sheet containing the c~-subunit was cut out, then reacted with anti-holo Na, K-ATPase antibody and the specific antibody against the c~subunit was eluted according to McDonough et al. (1982). The immunoglobulin G (IgG) concentration of affinity-purified antie-subunit antibody was determined by enzyme immunoassay (ELISA).

Solubilization of rat kidney proteins and immunoblot analysis Rat kidneys were removed, and all the zones from the cortex to inner medulla were cut out. After washing with phosphate-buffered saline (PBS), they were homogenized in 1% sodium dodecyl sulfate (SDS) with a teflon-glass homogenizer. Insoluble components were removed by centrifugation at 15000 x g for 30 min. The solubilized protein in each zone was quantified according to Lowry et al. (1951), and an aliquot was analyzed by SDS-PAGE (Laemmli 1970). Immunoblot analysis was performed by transferring the electrophoresed protein to a nitrocellulose sheet (Burnette 1981). The sheet was incubated, overnight at 4 ° C with 5% skim milk (Difco, Detroit, Mich., USA) in Tris-HC1buffered saline (pH 7.6), then with either affinity-purified antibody or IgG from preimmune serum (control; 0.5 gg/ml) for i h at room temperature. This was followed by incubation with goat anti-rabbit IgG horseradish peroxidase (HRP) conjugates (6 lag IgG/ml), and was visualized by 3,3'-diaminobenzidine tetrahydrochloride (DAB) color reagent.

Light microscopic localization of Na, K-A TPase Light microscopic localization of Na, K-ATPase was performed by two methods: (i) protein A-gold silver enchancement on LRWhite-embedded kidney sections and (ii) avidin-biotin peroxidase complex (ABC) on cryosections. For the former, the post-embedding protein A-gold technique on LR-White sections was used according to Herken et al. (1988) with some modifications. Rats were anesthetized with diethyl ether and perfused from the left ventricle with Hanks' solution for 5 min, followed subsequently with 4% paraformaldehyde containing 0.1% glutaraldehyde in Hanks' solution for 10 min, then washed with PBS containing 50 m M NH,C1 for 5 min. Kidneys were cut and postfixed with 4% paraformaldehyde containing 0.1% glutaraldehyde in Hanks' solution for 20 min, and washed three times for 10 rain with PBS containing 50 m M NH4C1, again postfixed with 4% paraformaldehyde in PBS overnight at 4 ° C, and washed with PBS containing 50 m M NH~C1. Finally the kidneys were dehydrated in a series of graded ethanol at low temperature and embedded in LR-White at --20 ° C. The resin was polymerized by ultraviolet-ray polymerizer TUV-200 (Dosak EM Co., Kyoto, Japan).

Large semi-thin sections containing all renal epithelial cells from cortex to inner medulla were cut with a histodiamond knife (Diatome Ltd., Bienne, Switzerland) from the LR-White-embedded kidney and stained by the protein A-gold silver enhancement procedure, as described by Taatjes et al. (1987). Semi-thin sections were mounted on glass coverslips and subsequently incubated with 0.5% bovine serum albumin (BSA) in PBS for 5 rain, then for 60 rain with anti-Na, K-ATPase c~-subunit antibody (10 gg/ml, 20 lal). The sections were washed six times with 0.5% BSA in PBS, and incubated for 30 min with IgG-gold complex (1 nm diameter, 20 gl, Bio Cell, Cardiff, UK). After washing six times with 0.1 M sodium cacodylate buffer (pH 7.4), the glass slide was incubated for 10 min with 2% glutaraldehyde in cacodylate buffer and washed with distilled water. After silver enhancement, the sections were counterstained with toluidine blue for visualization. As a control, the same concentration of IgG from non-immunized rabbits was used. Since the preservation of the inner medullary segment of rat kidney was quite poor on LR-White-embedded kidney sections, the ABC method on cryosections was used for visualization of Na, K-ATPase in the inner medulla. Rat kidney fixed as described above and washed with PBS containing 50 m M NH4C1 was frozen in liquid nitrogen. Frozen sections were cut with a cryostat and mounted on gelatin-coated glass slides. The sections were preincubated for 30 min with 0.3% H202 in methanol to inactivate endogenous peroxidase activity, washed with PBS, and incubated for 15 min with PBS containing 20% normal goat serum, 0.5% BSA and 0.3% Triton X-100. The sections were then incubated for 30 min with either anti-Na, K-ATPase c~-subunit (1 gg/ml), or IgG from nonimmunized rabbit (1 gg/ml) as control. After incubation, sections were washed three times for 5 min each in PBS containing 0.3% Triton X-100, subsequently reacted for 30 min with biotinylated goat anti-rabbit IgG (5 gg/ml) in PBS containing 0.5% BSA and 0.3% Triton X-100, and washed three times in PBS for 5 min. Bound antibodies were detected with the ABC method (Behringer et al. 1991; Hsu et al. 1981), incubating for 30 min with avidinbiotin peroxidase complex (Vectastain, Vector Lab., Burlingame Calif., USA) in PBS (diluted 1 : 50). After washing three times with PBS for 5 rain, peroxidase activity was visualized by incubating for 2 min with 0.1% DAB in 50 m M Tris-HC1, pH 7.4, supplemented with 0.02% HzO2 and sections were counterstained with toluidine blue.

Immunoelectron microscopic localization of Na, K-ATPase and quantitative analysis Ultra-thin sections were always prepared from LR-White-embedded kidney specimens. Wide rectangular sections, 1.0 x 0.7 mm in area, were prepared from four positions of the kidney specimens : cortex, outer stripe of outer medulla, inner stripe of outer medulla, and outer third of inner medulla. The sections were incubated for 4 h with 0.5% BSA in PBS at room temperature, and overnight ( > ! 5 h) at 4 ° C with anti-Na, K-ATPase e-subunit antibody (10 gg/ml, 40 lal), then subsequently for 30 min with protein A-gold complex (8 nm diameter; optical density at 525 nm=0.08; 100 gl). After washing with sodium cacodylate buffer, pH 7.4, sections were post-fixed with 5% glutaraldehyde, then double stained with uranyl acetate and lead citrate and observed under a Hitachi (Tokyo, Japan) electron microscopy H-7000. As a control, the same concentration of IgG from nonimmunized rabbits was used. Usually photographs were taken at a magnification of x 15 000 and enlarged to a final magnification of x37500. The number of gold particles on the cross-sectional profiles of the plasma membranes or within 13 nm from the center of the membranes was counted, and the length of each membrane was measured on electron microphotographs by a graphic digitizer connected to a personal computer. The labeling density was calculated by subtracting the average number of gold particles per gm of membranes of the control from that of the experimental specimen. Usually 150200 gm of each domain of the plasma membranes of nephron epithelial cells was measured per animal.

185

Determination of the saturation level of antibody for the immunogold labeling assay For quantitative immunogold labeling, immunocytochemical reactions should be performed at a saturation level of antibody (Matsumoto et al. 1987; Matsuura et al. 1979; Tanaka et al. 1986). For this purpose, ultra-thin sections (0.7 x 1.0 mm in area) of cortex were incubated with 40 ILlof increasing concentrations of the antibody solution, after which the labeling density on the basolateral membranes of the distal straight and distal convoluted tubule cells was determined, and plotted against the concentration of antibody (Fig. 3). The saturation level of the antibody thus determined will be effective in the quantitative analyses of any portion of nephron, because the concentration of Na, K-ATPase in the distal straight tubules is highest in all the epithelial cells of nephron (Table 1).

Results

Immunochemical specificity of the antibody against the c~-subunit of rat kidney Na, K-ATPase The immunological specificity of the affinity-purified antibody against the e-subunit of Na, K-ATPase was tested by immunoblot analysis of the solubilized kidney (Fig. 1). Anti-c~-subunit antibody bound specifically with an approximately 100 kDa component, which corresponds to the ~-subunit of Na, K-ATPase in kidney. According to Sweadner (1979), the e- and c~+-isoforms of Na, K-ATPase were detected in brain by the antibody against the e-subunit of Na, K-ATPase. Although our antibody reacts with both the c~- and e+-isoforms, only the ~-isoform was detected in rat kidney microsomes (Okami et al. 1990). It is noticed in Fig. 1 that the concentration of Na, K-ATPase is low in the cortex and in the inner region of the inner medulla, but high in 2

3i

4I

5i

6i

; i

94--

67 45 m

30-Fig. 1. Characterization of the antibody against the e-subunit of rat kidney Na, K-ATPase by immunoblot analysis. Total rat kidney (lane 1) was solubilized in 1% sodium dodecyl sulfate (SDS), analysed by SDS-polyacrylamide gel electrophoresis (PAGE) in 10% slab gel, and stained with Coomassie brilliant blue. Rat kidney cortex (lane 2), outer stripe (lane 3) and inner stripe (lane 4) of the outer medulla, and outer (lane 5) and inner region (lane 6) of the inner medulla were also solubilized in 1% SDS; aliquots of the specimens containing ~ 25 gg protein each were separated by SDS-PAGE in 10% slab gel, transferred to a nitrocellulose sheet, and immunoblot analysis was performed as described in the Materials and methods. The arrow indicates Na, K-ATPase c~-subunit; the numbers indicate mol.wt, in kDa

the outer and inner stripes of the outer medulla and in the outer region of the inner medulla. This antibody was used in all the following immunocytochemical experiments.

Light microscopic localization of Na, K-ATPase Since large semi-thin sections containing all renal segments from cortex to inner medulla were prepared, it was quite easy to identify the four zones of rat kidney; cortex, outer stripe of outer medulla, inner stripe of outer medulla and inner medulla. Figure 2 A - C shows the distribution of Na, K-ATPase in the cortex, outer stripe and inner stripe of the outer medulla, respectively, as visualized by the protein A-gold silver enhancement procedure. In the cortex (Fig. 2A), the staining reaction was most intense in distal straight and distal convoluted tubules. Proximal tubules, which can be identified by the existence of a brush border, are subdivided into three segments: $1, $2 and $3. In the cortex, $1 and $2 segments predominate. As shown in Fig. 2A, $1 cells were stained weakly. In these cells, silver grains were always observed on the basolateral surface and no significant grains were detected on the apical surface. Connecting tubules and collecting ducts are markedly heterogeneously labeled; especially in the latter, only the basal regions of principal cells were heavily labeled but intercalated cells were not labeled at all. In the outer medulla, thick ascending limbs of the loop of Henle were heavily stained both in the outer (Fig. 2B) and inner stripes (Fig. 2C). However, the $3 segment of the proximal tubules found in the outer stripes (Fig. 2 B) and the thin limbs of the loop of Henle in the inner stripes (Fig. 2 C) were minimally or barely stained. The basal region of the collecting ducts (Figs. 2B and C; stars) are again heterogeneously labeled as in the cortex. The structure of the inner medulla, when embedded in LR-White as described in the Materials and methods, was not well preserved and Na, K-ATPase could not be localized by the protein A-gold silver enhancement procedure. As an alternative approach, therefore, the cryosections were cut and stained using the ABC method. Figure 2D shows the transitional area between outer medulla (upper part) and inner medulla (lower part), indicating that strong immunoreactivity starts at the beginning of the thick ascending limbs, and that weak staining is observed relatively uniformly along the basal surface of collecting duct cells found in the inner medulla. It is noted here that vascular bundles, which are exclusively composed of thin limbs of the loop of Henle and vasa recta, are barely stained. When the semi-thin sections were incubated with IgG of preimmune sera, no positive staining was observed in any part in any cell of the nephron (photographs not shown).

Saturation level of antibody in the immunoelectron microscopic assay Figure 3 shows a binding curve of gold particles to cells of the rat kidney distal straight and convoluted tubules

186

Fig. 2A-D. Light microscopic localization of Na, K-ATPase c~subunit in A rat kidney cortex B the outer stripe and C inner stripe of the outer medulla and D the inner medulla. Semi-thin sections of the kidney embedded in LR-White were incubated with anti-Na, K-ATPase c~-subunit antibody and visualized by the protein A-gold silver enhancement procedure as described in the ' Materials and methods' (A-C). By this procedure localization of Na, K-ATPase e-subunit in A cortex, B outer stripe and C inner stripe

of the outer medulla were visualized. D Inner medulla was visualized by avidin-biotin peroxidase complex (ABC) method of the cryosections. Asterisks, connecting tubule cells; stars, collecting duct cells. Note markedly heterogeneous distribution of silver grains on the basalsurface of the collecting duct cells. G, Glomerulus; S1, S1 segment of proximal tubules; D, distal convoluted tubules; V, vasular bundles. Magnification: A - C x920 (bars 20, gin); D x230 (bar, 40 gin)

187

the inner stripe of outer medulla, which are characterized by a marked reduction in the basal infoldings (Fig. 5A). Numerous gold particles bound to the lateral plasma membranes (Fig. 5A and B). The apical plasma membranes (not shown) and the basal plasma membranes which are in direct contact with the basal lamina (indicated by arrowheads in Fig. 5A) were barely labeled with gold particles.

1412E

\ ~10(3

~ 8-

"S4x-

Z

0 1

2

3

4 5 6 7 8 Concentration of Antibody

9

1'0 ~g/mt

Fig. 3. Saturation of binding of gold particles to the basolateral plasma membranes of the distal straight and convoluted tubules of rat kidney. Relationship between the labeling density of the distal straight (o) and the distal convoluted cells (e) was plotted against antibody concentration

embedded in LR-White. The binding of gold particles to the basolateral membrane was saturated at an antibody concentration of 7.5 gg/ml. The concentration of antibody that was used in the quantitative immunocytochemical experiment (10 gg/ml) was definitely higher than the saturation level. This means that antibody bound to almost all the accessible epitopes of Na, KATPase c~-subunit molecules on the ultra-thin sections. Thus, the labeling densities will be proportional to the number of the c~-subunit molecules in the corresponding domain. Under such conditions, the labeling density of each membrane domain will be proportional to the relative amount of Na, K-ATPase e-subunit. The relationship was also examined between the incubation time and the labeling density. Incubation for 10 h was sufficient for saturation (data not shown). For quantitative analysis, therefore, we usually incubate the kidney specimens overnight ( ~ 15 h).

Electron microscopic localization of Na, K-ATPase by immunog old procedure Proximal tubule cells. The proximal tubule cells in $1 and $2 segments are characterized by a well-developed brush border (densely packed microvilli) on their apical surface and well-developed basal infoldings, usually sandwiching elongated mitochondria (Kriz and Kaissling 1985). Gold particles were localized in considerable numbers along the lateral plasma membranes and infolded basal plasma membranes (Fig. 4A), but barely detectable on the basal plasma membranes which are in direct contact with the basal lamina. No significant binding of gold particles was detectable along the apical plasma membranes including microvilli, whereas the lateral plasma membranes were heavily labeled (Fig. 4B). When incubated with control IgG, gold particle were barely detectable on the basal infoldings (Fig. 4 C). Figure 5A and B show $3 proximal tubule ceils in

Thin descending limb of loop of Henle. In the thin descending limb of the inner stripe of the outer medulla, gold particles were predominantly bound to the lateral surfaces, and barely detectable on the apical surface of the tubule cells as shown in Fig. 6. The basal surface of the tubule cells, which are in direct contact with the basal lamina, was only slightly labeled. The binding of gold particles to the tubule cells in the other regions of the loop of Henle was similar (electron micrographs not shown). Distal straight tubules. The distal straight tubule cells are characterized by very well-developed basal infoldings usually sandwiching elongated mitochondria (Kriz and Kaissling 1985). These infolded plasma membranes were very densely labeled as shown in Fig. 7A; in marked contrast, the apical plasma membranes were scarcely labeled (Fig. 7B). The apical cytoplasma of the distal straight tubule cells is characterized by a number of small vesicles, which are heavily loaded with TammHorsfall glycoprotein (Bachmann et al. 1985). Distal convoluted tubule and connecting tubule. In the segment of the distal convoluted tubule cells, the degree of basal infolding and the association of mitochondria to the infolded plasma membranes are most prominent of all nephron segments (Kriz and Kaissling 1985). As shown in Fig. 8, numerous gold particles bound to these basolateral plasma membranes. It is noted that gold particles scarcely bound to the basal plasma membranes which are in association with the basal lamina. In marked contrast to the dense labeling of the basolateral membranes, few gold particles bound to the apical plasma membranes (not shown). Connecting tubules are composed of connecting tubule cells and intercalated ceils (Kriz and Kaissling 1985). Gold particles scarcely bound to the latter cells. Figure 9 shows the basal infoldings of a connecting tubule cell. The majority of the infolded basal membranes are not associated with mitochondria but are markedly studded with numerous gold particles. Collecting duct. Collecting ducts in the cortex and outer medulla are composed of collecting duct cells (principal cells) and intercalated cells as shown in Fig. 10A (Kriz and Kaissling 1985); gold particles scarcely bound to the intercalated cells, compared to cells in the connecting tubule. The cellular organization of the collecting duct cells is similar to that of the connecting tubule cells but with fewer basal infoldings. Figure 10 B and C show basal infoldings of collecting duct cells in the cortical collecting duct. Gold particles numerously bound to the

188

Fig. 4A-C. Immunogold labeling of the $1 segment of the proximal tubules of rat kidney. A Basolateral plasma membranes with marked basal infoldings are labeled with numerous gold particles. Note paucity of gold particles on the basal membranes which are in direct contact with the basal lamina (arrowheads). B Apical plas-

ma membranes are barely labeled. In marked contrast, the lateral plasma membrane is heavily labeled (arrowheads). C Basolateral plasma membranes incubated with IgG from non-immunized rabbit serum (control). Magnification, x 37500; bars, I gm

189

Fig. 5A, B. Immunogold labeling of the $3 segment of the basal and lateral plasma membranes of the proximal tubules ($3). A The basal plasma membrane, which is directly opposed to the basal lamina is barely labeled (arrowheads), while the lateral plasma membrane indicated by small arrowsis labeled with gold particles. B The lateral plasma membrane labeled with gold particles. Magnification, A x 37500; B x 75000; bars, 0.5 gm

Fig. 6. Immunogold labeling of the thin descending limb of Henle's loop in the inner segment of the outer medulla. Gold particles are numerous on the lateral plasma membrane. The apical plasma membranes are barely labeled. The basal plasma membranes, which are in direct contact with the basal lamina (arrowheads),are only slightly labeled. Magnification, x 42 500; bar, 0.5 lam

infolded basal plasma m e m b r a n e s (basal labyrinth). In marked contrast, few gold particles bound to the apical membranes of the collecting duct cells (Fig. 10D). The collecting duct cells in the collecting duct undergo gradual changes f r o m the cortex to the tip of the papilla. The basal infoldings decrease gradually and in the inner medulla, only remnants of the basal labyrinth are found (Kriz and Kaissling 1985). Gold particles bound to the infolded basolateral plasma membranes as shown in Fig. 10E.

Quantitative analysis of the labeling of Na, K-A TPase In order to analyse quantitatively the distribution of the enzyme, the average n u m b e r of gold particles b o u n d per gm of each domain of the plasma m e m b r a n e (labeling density) of the tubular epithelial cells along rat nephron was estimated (Table 1). The data are shown schematically in Fig. 11. Specific binding (Table 1, last column) was calculated by subtracting the labeling density of gold particles of the experimental specimens f r o m that

190

N Fig. 7A, B. Immunogold labeling of the distal straight tubules. A Many gold particles bound to the basal infoldings which are sandwiching elongated mitochondria. B In marked contrast, few gold particles bound to the apical plasma membranes. Note vesicu-

lar structures in the apical cytoplasma, which is characteristic of the distal straight tubule cells. Magnification; A, B x 42500; bars, 0.5 gm

of the control. Statistical analysis using Student's paired t-test indicated that the labeling densities of all the basolateral surfaces of the epithelia are statistically significant but that those of all the apical surfaces are not. Labeling density was highest on the basolateral surface of the distal straight tubule, very high on that o f the distal convoluted tubule, and intermediate on that of the proximal tubules. The labeling density on the basolateral membranes of the collecting duct cells was as high as that in the proximal tubule cells in both the cortex and

medulla. The labeling density of the intercalated cell membranes was very low and was not measured because they were not clearly visible in the LR-White sections. As described above, labeling of the basal plasma membranes was quite heterogeneous; few gold particles were bound in the region of the basal membranes that are in direct contact with basal lamina both in the proximal tubule cells (Figs. 4 A and 5A) and in the distal tubule cells (Fig. 8). We estimated the labeling density of the contact region of the basal membranes in the

191

Fig. 8. Dense immunogold labeling of the basal infoldings in the distal convoluted tubule cells. Arrowheads indicate that the basal plasma membranes directly opposing the basal lamina are barely labeled. Magnification, x 37500; bar, 0.5 gm

Fig. 9. Dense immunogold labeling of the basal infoldings in the connecting tubule cells. Magnification, x 42500; bar, 0.5 gm

192

Fig. 10A-D. Collecting duct cells (principal cells) and intercalated cells of the cortical collecting duct. A Conventional electron micrograph of collecting duct cells (C) and intercalated cells (/) in the cortical collecting duct. B-D Immunogold labeling of the basolateral plasma membranes (B, C) and the apical plasma membrane

(D) of the collecting duct cells in the cortical collecting duct. E Immunogold labeling of the infolded basolateral plasma membranes (labyrinth) of the collecting duct cells in the inner medullary collecting duct. Magnification: A x 10000; bar 1 gm; B x 37500; bar 0.5 gm; C, E x 75000; bars 0.5 gm; D x 42500, bar 0.5 gm

193 Table 1. Labeling density of gold particles on various cell surfaces of rat kidney epithelial cells (number of gold particles/gm)a Cells and cell surfaces

Proximal tubule:

Antibody to Na,K-ATPase e-subunit

Nonimmunized IgG

Specific binding

$1, $2 segment (cortex) Basolateral * 3.51 ± 0.68 Apical 0.10 -+0.04

0.21 ± 0.18 0.05_+0.03

3.30 ± 0.67 c 0.04_+0.03 b

$3 segment (outer medulla outer stripe) Lateral 4.38 -+0.48 Apical 0.33 -+0.15

0.29 _+0.06 0.08_+0.03

4.09 _+0.50 c

0.27 _+0.06 0.19 _+0.11 0.07_+ 0.05

1.00 -+0.26 c 2.41 ± 0.49 ° 0.15_+ 0.07b

Distal straight tubule (outer medulla inner stripe) BasolateraI* 11.67-+ 1.75 Apical 0.21 -+0.03

0.32_+ 0.09 0.11 ± 0.04

11.35 ± 1.74 ~ 0.10 + 0.06 b

Distal convoluted tubule (cortex) Basolateral * 10.02 _+1.09 Apical 0.25_+0.08

0.24 _+0.06 0.11 -t-0.05

9.78 __1.12 ° 0.14_+0.11 b

Connecting tubule (cortex) Basolateral* 4.52_+ 0.40 Apical 0.09 _+0.02

0.23_+ 0.05 0.07 _+0.02

4.29_+ 0.43 c 0.02_+ 0.03 b

Collecting duct (cortex) principal cell Basal 6.01 _+0.40 Apical 0.20 -+0.04

0.40_+ 0.02 0.09 _+0.04

5.62-+ 0.39 c 0.10 _+0.06 b

Collecting duct (outer third of inner medulla) Basal 4.03 _+0.64 Lateral 4.92 _+0.72 Apical 0.08 _+0.03

0.19 -+0.11 0.20 _+0.05 0.04_+ 0.01

3.84 ± 0.54 c 4.72 ± 0.69 c 0.03 _+0.03 b

Descending thin limb of Henle's loop (outer medulla inner stripe) : Basal Lateral Apical Distal tubule:

Collecting system:

Mean -+ standard deviation (n = 3 rats) b Statistically not significant, p>0.02

Table 2. Labeling density of gold particles on the contact domain of the basal plasma membranes of rat kidney epithelial cells (number of gold particles/gm) a

1.27 _+0.27 2.59 ± 0.46 0.22_+ 0.08

0.25 +0.18 b

c Statistically significant, p < 0.02 * Basal plus lateral plasma membranes

Antibody to Na,K-ATPase e-subunit

Nonimmunized IgG

Specific binding

Proximal tubule S1,$2 segment (cortex)

0.06 ±0.01

0.06 _+0.02

0.00 + 0.00 b

Distal tubule convoluted tubule (cortex)

0.06_+ 0.02

0.05 _+0.01

0.00 -}-0,00 b

Cells and cell surfaces

" Mean _+ standard deviation (n = 3 rats) b Statistically not significant, p > 0.02

$1 a n d $2 segments of the p r o x i m a l t u b u l e cells a n d in the distal c o n v o l u t e d cells. N o significant b i n d i n g o f gold particles was detected (Table 2).

Discussion

Because o f the f u n c t i o n a l a n d cellular heterogeneity o f the n e p h r o n segment, q u a n t i t a t i v e localization o f Na, K - A T P a s e in discrete t u b u l e segments o f renal t u b u l e s

should help define the role of this e n z y m e in t u b u l a r c a t i o n t r a n s p o r t . F o r this p u r p o s e we utilized a q u a n t i t a tive i m m u n o g o l d electron microscopic analysis, which c o m b i n e s high specificity a n d q u a n t i t a t i o n o f the imm u n o c h e m i c a l r e a c t i o n with the high r e s o l u t i o n of elect r o n microscopy. Firstly we have c o n f i r m e d the mono-specificity of the a n t i - N a , K - A T P a s e ~ - s u b u n i t a n t i b o d y by i m m u n o b l o t analysis (Fig. 1). By i m m u n o g o l d light m i c r o s c o p y a n d electron microscopy, gold particles were s h o w n to b i n d

194

T

Distal Con Proximal

Cortex

t outer stripe

+ Outer Medulla I inner stripe

Thin Limb Henle's L(

Inner Medulla

~)

t0 gold particles ,/l~m

Fig. 11. Schematic representation of the labeling density of Na, K-ATPase along rat nephron. Thickness of the tubular wall in the various renal segments is proportional to the labeling density of the basal or basolateral plasma membrane of the epithelial cells at each segment

exclusively to the basolateral plasma membrane of all the nephron tubule cells except for intercalated cells. Also, the distribution of gold particles on the basal membranes was quite heterogeneous; the basal membranes that are in direct contact with the basal lamina were barely labeled (Table 2). This observation was first described by Kashgarian et al. (1985), who used monoclonal antibody for immunoperoxidase localization of Na, K-ATPase along nephron segments. Koob e t al. (1987) have shown co-localization of ankyrin and Na, KATPase in kidney epithelial cells and suggested that linkage of Na, K-ATPase via ankyrin to the actin-spectrin lattice is responsible for such a heterogeneous distribution of Na, K-ATPase. The present immunocytochemical reactions were always carried out at a saturation level of antibody (Fig. 3). Under such a condition, the number of gold particles bound to a specific domain of plasma membranes should be proportional to the number of Na, K-ATPase molecules at that domain. The labeling densities of various plasma membrane domains of rat kidney epithelial cells determined thus are shown in Table 1 and represented schematically in Fig. 11. Several interesting suggestions are deduced from the data of Table 1. In the proximal tubule, it has been shown by biochemical analyses of isolated tubule segments that the activity of Na, K-ATPase decreases along the length of the proximal tubule (Doucet 1988). The labeling density in the lateral domain in the $3 cells was similar to that in the basolateral domain in the $1 and $2 cells. This result suggests that the decrease in the Na, K-ATPase activity along the proximal tubules is not due to the

decrease in the labeling density on the basolateral domain of the membranes; rather it is due to the decrease in the free surface areas of the basal plasma membranes that are not in direct contact with the basal lamina and are heavily loaded with Na, K-ATPase. The labeling density of the descending thin limb of Henle's loop was very low, and gold particles were mainly localized on the lateral surface. Such distribution of Na, K-ATPase may be effective for reabsorption of water through the paracellular pathway by standing-gradient osmotic flow (Diamond and Bossert 1967). In the distal straight and distal convoluted tubules, the labeling density of the basolateral plasma membranes was very high: approximately three times higher han that in the proximal tubules. Na, K-ATPase in the microsomal membranes, presumably derived from distal renal tubules, shows crystalline arrays as reported by Deguchi et al. (1977), Haase and Koepsell (1979), and Zampighi et al. (1984). Such a large amount of Na, K-ATPase may be necessary to reabsorb large amounts of Na ÷ against a concentration gradient. In the proximal tubule cells, large amounts of the other transport proteins such as glucose and amino acid transporters also exist on the basolateral domains. It is suggested, therefore, that the labeling density of Na, K-ATPase in the proximal tubules cannot increase beyond this level because of possible steric hindrance in this domain. In the distal tubules, however, relative paucity in the other transport proteins may make it possible to increase the concentration of Na, K-ATPase maximally to the level of ~ 12 gold particle/gm. We can estimate the labeling efficiency of Na, KATPase in the distal membranes. According to Deguchi et al. (1977) and Haase and Koepsell (1979), the frequency of particles of negatively stained Na, K-ATPase membrane from kidney outer renal medulla, was 12 50019000/txm 2. According to Kellenberget et al. (1987), the surface of thin sections of aldehyde-fixed biological materials shows a specimen-related relief of ~ 4 nm with Lowicryl. We assume here that the same value is applicable for LR-White. Only the epitopes on the surface of the ultra-thin section, which were supposed to be laid open, can react with the corresponding antibodies and subsequently with protein A-gold particles. So the labeling density represents the number of gold particles bound to a small rectangle, 0.004 gm x 1.0 gm in area. The number of Na, K-ATPase molecules in this rectangle is calculated to be 50 76. Since the labeling density in the distal straight tubule cells was 11.4, the labeling efficiency is calculated to be ~24-16%. A similar value (~20%) has been obtained for the labeling efficiency of microsomal cytochrome P-450IIB by the quantitative immunogold method (Fukui et al. 1992). Taking into consideration the possibilities that the epitopes of Na, K-ATPase are inactivated during fixation, embedding, and the other procedures for specimen preparation, this labling efficiency of 20% appears to be reasonable. The connecting tubule and collecting duct epithelia are composed of principal and intercalated cells (Kriz and Kaissling 1985; Holth6fer et al. 1987). The principal cells showed labeling of the basal infoldings, whereas

195

A

y,

0 S1,S2

S3

DST(OMIS)DST(OMOS)

DCT

CCD

IMCD

OMCD

IMCD

B

,x3

0 S1,S2

$3

TLH

DST

DCT

CCD

Fig. 12A, B. Approximate estimation of A relative number of gold particles/mm nephron segment and B relative number of gold particles in the various nephron segments. Ratios of the various segments given in B are relative to those of Sl and $2 segments. The values in cortical, and outer medullary and inner medullary collecting duct cells were calibrated by the percentages of Na, KATPase immunoreactive cells (principal cells), which were given by Holth6fer et al. (1987) as ~ 70%, ~ 50%, ~ 80%, respectively. A In order to calculate the relative number of gold particles/ram nephron segment, the labeling densities on the basolateral membrane surface (or lateral surface) of the various nephron cells were multiplied with the corresponding surface areas, as given in Table 22 of Pfaller's report on rat kidney cells (Pfaller 1982). Labeling density on the outer stripe of outer medullary distal straight tubule cells [DST(OMOS)] was assumed to be similar to that on the inner stripe of outer medullary distal straight tubule cells [DST(OMIS)]. B In order to calculate the relative numbers of gold particles in the various nephron segments, the labeling densities on the basolateral (or lateral) membrane surface of the various nephron cells were multiplied by the absolute surface area of basolateral membranes per standard sized kidney of 1320 gl, as given in Table 30 of Pfaller's report (Pfaller 1982). Labeling density of DST(OMOS) and of OMCD were assumed to be similar to that of DST(OMIS) and CCD, respectively. $1, $2, $1 plus $2 segments of proximal tubule; $3, $3 segment of proximal tubule; TLH, thin limb of Henle's loop ; DST(OMIS), distal straight tubule in outer medullar inner stripe; DST(OMOS), distal straight tubule in outer medulla outer stripe; DCT, distal convoluted tubule; CCD, cortical collecting duct; OMCD, outer medullary collecting duct; IMCO, inner medullary collecting duct

the intercalated cells were barely labeled. This marked difference in the distribution of Na, K - A T P a s e in the principal and intercalated cells m a y be due to the functional difference of the two cells as pointed out by Kashgarian et al. (1985). Recently biochemical and immunocytochemical evidence has been accumulating that intercalated cells contain H +, K + - A T P a s e (Cheval et al. 1991 ; Palanelles et al. 1991 ; Wingo et al. 1990), and suggests that they are probably not involved in the transepithelial m o v e m e n t of N a +. In contrast, the collecting duct cell (principal) cells in the cortical, inner and outer medullary collecting ducts showed marked labeling of the basolateral infoldings. This result is in good agreement with the immunoperoxidase labeling experiment by Kashgarian et al. (1985). The principal cells respond to potassium and sodium adaptation and mineralo-corticoids with an increase in the basolateral m e m b r a n e surface area and an increase in Na, K-ATPase (Kaissling et al. 1985; Kaissling and Stanton 1988; Rastegar et al. 1980; Stanton et al. 1981, 1985a, b; Zalups et al. 1985). Taken together, it is suggested that Na, K - A T P a s e in the infolded plasma membranes in the medullary collecting duct cells plays an important role in regulating urinary sodium and potassium secretion and in the maintenance of N a gradient which is necessary for counter current mechanism. In order to compare the immunoelectron microscopic data shown in Table 1 with the biochemical data on the distribution of Na, K - A T P a s e along the nephron (Katz et al. 1979; Katz 1986), the labeling density of a domain should be multiplied by the total surface area of that domain. Fortunately extensive quantitative morphological data on rat kidney have been published by Pfaller (1982) and it is possible to calculate the relative number of gold particles/ram of nephron and the relative number of gold particles in the various nephron segments as shown in Fig. 12A and B, respectively. Figure 12A is in good correlation with the Na, K - A T P a s e activity profile in rat reported by Katz et al. (1979) and the correlation coefficient of the two profiles was calculated to be 0.95. Figure 12B indicates that the total a m o u n t of Na, K-ATPase in the proximal tubules and distal tubules in rat kidney is approximately 1 : 1.6, indicating the relative contribution of Na, K - A T P a s e in the functions of the nephron.

Acknowledgements. One of the authors (T. Takada) thanks Dr. Takaya Tanaka, Emergency Care Unit of Kansai Medical University, for his encouragement. This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture of Japan and by the Science Promotion Fund from Japan Private School Promotion Foundation. We thank Ms. K. Miki for assistance with the manuscript.

References Akayama M, Nakada H, Omori K, Masaki R, Taketani S, Tashiro Y (1986) (The (Na +, K+)ATPase of rat kidney: purification, biosynthesis, and processing. Cell Struct Funct 11:259 271 Bachmann S, Koeppen-Hagemann I, Kriz W (1985) Ultrastructural localization of Tamm-Horsfall glycoprotein (THP) in rat kidney

196 as revealed by protein A-gold immunocytochemistry. Histochemistry 83:531 538 Baskin DG, Stahl WL (1982) Immunocytochemical localization of Na 4, K ÷-ATPase in the rat kidney. Histochemistry 73:535548 Behringer DM, Meyer K-H, Veh RW (1991) Antibodies against neuroactive amino acids and neuropeptides. II. Simultaneous immunoenzymatic double staining with labeled primary antibodies of the same species and a combination of the ABC method and the hapten-anti-hapten bridge (HAB) technique. J Histochem Cytochem 39:761-770 Burnette WN (1981) "Western blotting": electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal Biochem 112:195-203 Cheval L, Barlet-Bas C, Khadouri C, Feraille E, Marsy S, Doucet A (1991) K +-ATPase-mediated Rb + transport in rat collecting tubule: modulation during K ÷ deprivation. Am J Physiol 260 : F800-F805 Deguchi N, Jorgensen PL, Maunsback AB (1977) Ultrastructure of the sodium pump. Comparison of thin sectioning, negative staining, and freeze-fracture of purified, membrane-bound (Na +, K ÷)ATPase. J Cell Biol 75:619-634 Diamond JM, Bossert WH (1967) Standing-gradient osmotic flow: A mechanism for coupling of water and solute transport in epithelia. J Gen Physiol 50:2061-2083 Doucet A (1988) Function and control of Na-K-ATPase in single nephron segments of the mammalian kidney. Kidney Int 34:749 760 Doucet A, Katz AI, Morel F (1979) Determination of Na, KATPase activity in single segments of the mammalian nephron. Am J Physiol 237:F105-F113 Ernst SA (1975) Transport ATPase cytochemistry: ultrastructural localization of potassium-dependent and potassium-independent phosphatase activites in rat kidney cortex. J Ceil Biol 66 : 586-608 Ernst SA, Schreiber JH (1981) Ultrastructurai localization of Na 4, K+-ATPase in rat and rabbit kidney medulla. J Cell Biol 91 : 803-813 Fnkui Y, Yamamoto A, Masaki R, Miyauchi K, Tashiro Y (1992) Quantitative immunocytochemical analysis of the induction of cytochrome P450IIB in rat hepatocytes. J Histochem Cytochem 40 : 73-82 Haase W, Koepsell H (1979) Substructure of membrane-bound Na+-K+-ATPase protein. Pfliigers Arch 381:127-135 Herken R, Fussek M, Thies M (1988) Light and electron microscopical post-embedding lectin histochemistry for WGA-binding sites in the renal cortex of the mouse embedded in polyhydroxy aromatic resins LR-white and LR-gold. Histochemistry 89 : 277-282 Holth6fer H, Schulte BA, Pasternack Gary, Siegel GJ, Spicer SS (1987) Three distinct cell populations in rat kidney collecting duct. Am J Physiol 253 : C323-C328 H su SM, Raine L, Fanger H (1981) Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem 29 : 577-580 J6rgensen PL (1986) Structure, function and regulation of Na, KATPase in the kidney. Kidney Int 29:10-20 Kaissling B, Stanton BA (1988) Adaptation of distal tubule and collecting duct to increased sodium delivery. I. Ultrastructure. Am J Physiol 253 :F1256~1268 Kaissling B, Bachmann S, Kriz W (1985) Structural adaptation of the distal convoluted tubule to prolonged furosemide treatment. Am J Physiol 248 :F374-F381 Kashgarian M, Biemesderfer D, Caplan M, Forbush B (1985) Monoclonal antibody to Na, K-ATPase: immunocytochemical localization along nephron segments. Kidney Int 28:899-913

Katz AI (1982) Renal Na-K-ATPase: its role in tubular sodium and potassium transport. Am J Physiol 242:F207-F219 Katz AI (1986) Distribution and function of classes of ATPase along the nephron. Kidney Int 29:21-31 Katz AI (1988) Overview: Role of Na-K-ATPase in kidney. Prog Clin Biophys Res 268A:20~232 Katz AI, Doucet A, Morel R (1979) Na-K-ATPase activity along the rabbit, rat, and mouse nephron. Am J Physiol 237: Fl14FI20 Kellenberger E, Durrenberger M, Villiger W, Carlemalm E, Wurtz M (1987) The efficiency of immunolabel on Lowicryl sections compared to theoretical predictions. J Histochem Cytochem 35 : 959-969 Koob R, Zimmermann M, Schoner W, Drenckhahn D (1987) Colocalization and coprecipitation of ankyrin and Na +, K +ATPase in kidney epithelial cells. Eur J Celi Biol 45 : 230-237 Kriz W, Kaissling B (1985) Structural organization of the mammalian kidney. In: Seldin DW, Giebish G (eds) The kidney: physiology and pathology. Raven Press, New York, pp 265-306 Kuwahara T, Mayahara H, Kawai C, Ogawa K (1982) Histochemical study on the ouabain-sensitive potassium-dependent P-nitrophenyl phosphatase activity along the nephron in several mammals. Acta Histochem Cytochem 15:717-733 Kyte J (1976a) Immunoferritin determination of the distribution of (Na +, K +)ATPase over the plasma membranes of renal convoluted tubules. I. Distal segment. J Cell Biol 68 : 287-303 Kyte J (1976b) Immunoferritin determination of the distribution of (Na +, K +)ATPase over the plasma membranes of renal convoluted tubules. II. Proximal segment. J Celi Biol 68:304-318 Laborde K, Bussieres L, Smet De, Dechaux M, Sachs C (1990) Quantification of renal Na-K-ATPase activity by image analysing system. Cytometry 11 : 859-868 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 : 680-685 Lowry OH, Rosebrough H J, Farr AL, Randall RJ (1951) Protein measurement with the folin phenol reagent. J Biol Chem 193:265 275 Matsumoto S, Tanaka K, Yamamoto A, Nakada H, Uchida M, Tashiro Y (1987) Immunoelectron microscopic localization of dopamine /~-hydroxylase and chromogranin A in adrenomedullary chromaffin cells. Cell Struct Funct 12:483-496 Matsuura S, Fujii-Kuriyama Y, Tashiro Y (1979) Quantitative immunoelectron microscopic analyses of the distribution of cytochrome P-450 molecules on rat liver microsomes. J Cell Sci 36:413-435 Mayahara H, Ogawa K (1980) Ultracytochemical localization of ouabain sensitive, potassium-dependent p-nitrophenyl-phosphatase activity in the rat kidney. Acta Histochem Cytochem 13:90-102 McDonough AA, Hiatt A, Edelman IS (1982) Characteristics of antibodies to guinea pig (Na +, K+)-adenosine triphosphatase and their use in cell-free synthesis studies. J Membr Biol 69 : 1322 Okami T, Yamamoto A, Omori K, Takada T, Uyama M, Tashiro Y (1990) Immunocytochemical localization of Na +, K +ATPase in rat retinal pigment epithelial cells. J Histochem Cytochem 38 : 1267-1275 Palanelles G, Anagnostopoulos T, Cheval L, Doucet A (1991) Biochemical and functional characterization of H +-K +-ATPase in distal amphibian nephron. Am J Physiol 260:F806-F812 Pfaller W (1982) Structure function correlation on rat kidney. Adv Anat Embryol Ceil Biol 70:1 106 Rastegar AB, Biemesderfer D, Kashgarian M, Hayslett JP (1980) Changes in membrane surfaces of collecting duct cells in potassium adaptation. Kidney Int 18:293 301 Schmidt U, Dubach UC (1969) Activity of (Na +, K+)-stimulated adenosine-triphosphatase in rat nephron. Pflugers Arch 306:219 226 Shaver JLF, Stifling C (1978) Ouabain binding to renal tubules of the rabbit. J Cell BioI 76:278-292

197 Smith DE, Fisher PA (1984) Identification, developmental regulation, and response to heat shock of two antigenically related forms of a major nuclear envelope protein in Drosophila embryos : application of an improved method for affinity purification of antibodies using polypeptides immobilized on nitrocellulose blots. J Cell Biol 99:20 28 Stanton BA, Biemesderfer B, Wade JB, Giebisch G (1981) Structural and functional study of the rat distal nephron: effects of potassium adaptation and depletion. Kidney Int 19:3648 Stanton B, Giebisch G, Klein-Robbenhaar G, Wade J, DeFronzo RA (1985 a) Effects of adrenalectomy and chronic adrenal corticosteroid replacement on potassium transport in rat kidney. J Clin Invest 75:1317-1326 Stanton B, Janzen A, Klein-Robbenhaar G, DeFronzo R, Giebisch G, Wade J (1985b) Ultrastructure of rat initial collecting tubule. J Clin Invest 75:1327-1334 Sweadner KJ (1979) Two molecular forms of (Na +, K +)-stimulated ATPase in brain. Separation and difference in affinity for strophanthidin. J Biol Chem 254:6060-6067

Note added in proof. In the calculation of the relative number of gold particles/mm nephron segment of DCT shown in Fig. 12A, the correct value of 4.52 was used as the basolateral membrane surface per unit length of nephron (mm 2/ram) (W. Pfaller, personal communication), and not 5.66, which is given in Table 22 (distal tubule in cortex 1) by W. Pfaller (Adv Anat Embryo1 Cell Biol 70:1 106 (1982)).

Taatjes DJ, Schaub U, Roth J (1987) Light microscopical detection of antigens and lectin binding sites with gold labeled reagents on semi-thin Lowicryl K4M sections : usefulness of the photochemical silver reaction for signal amplification. Histochem J 19:235 245 Takada T, Yamamoto A, Omori K, Tashiro Y (1989) Quantitative analysis of Na, K-ATPase along rat nephron by immunoelectron microscopy. Cell Struct Funct 14:921 Tanaka K, Omori K, Tashiro Y (1986) Quantitative immunoferritin localization of leucine aminopeptidase on canine hepatocyte cell surface. J Histochem Cytochem 34 : 775 784 Wingo CS, Madsen KM, Smolka A, Tisher CC (1990) H-KATPase immunoreactivity in cortical and outer medullary collecting duct. Kidney Int 38:985-990 Zalups RK, Stanson BA, Wade JB, Giebisch G (1985) Structural adaptation in initial collecting tubule following reduction in renal mass. Kidney Int 27 : 636 642 Zampighi G, Kyte J, Freytag W (1984) Structural organization of (Na +, K+)ATPase in purified membranes. J Cell Biol 98:1851 1864