Physiological/pathophysiological alterations in proximal tubular Pireabsorption are associated with an altered brush-border membrane (BBM) expression of type II Na-Pi cotransporter molecules. Reduction is achieved by an internalization and lysosomal degradation and an increase in Pi reabsorption by new synthesis and BBM insertion of type II Na-Pi cotransporters. In the present study, we investigated by immunohistochemistry and immunogold electron microscopy the routing of internalized rat type II Na-Picotransporters (NaPi-2). In kidney of rats on a chronic low-Pi diet, NaPi-2 is mainly localized in the BBM, in cisterns of the Golgi apparatus and sparsely also in large endocytotic vacuoles and lysosomes. Fifteen minutes after the injection of the 1–34 analog of parathyroid hormone (PTH), the amount of NaPi-2 was decreased in the BBM and increased in endocytotic vesicles. NaPi-2 molecules colocalized with horseradish peroxidase injected prior to the injection of PTH. Vesicles labeled for NaPi-2 were occasionally also labeled for clathrin or the adaptor protein AP2. We conclude that NaPi-2 molecules enter the subapical compartment from where NaPi-2-containing vesicles are segregated off and directed to the lysosomes. A clathrin-mediated pathway may contribute to the PTH-induced internalization of NaPi-2.
- horseradish peroxidase
phosphate reabsorption in the renal proximal tubule is a major mechanism in the maintenance of overall Pihomeostasis; it is a Na+-dependent, secondary active process involving Na-Pi cotransport across the brush-border membrane (BBM) as a rate-limiting step (21, 22). A type II Na-Pi cotransporter seems to determine largely in different species the rate of BBM Pi flux and thus of proximal tubular Pi reabsorption (13, 23). Recent knockout experiments in mice documented that the type II Na-Picotransport-mediated transport contributes up to 70% of BBM Pi flux (1). Alterations in the activity of this cotransporter also account for physiological (e.g., PTH, dietary intake) and pathophysiological (e.g., X-linked hypophosphatemia) alterations in proximal tubular Pi reabsorption. Thereby these alterations are always associated with a different BBM expression of type II Na-Pi cotransporter protein (11, 14, 17, 29).
Parathyroid hormone (PTH) leads to a reduction in the expression of the rat-specific type II Na-Pi cotransporter (NaPi-2) at the BBM, thereby explaining the phosphaturic action of PTH (11, 17). Surprisingly, the decrease in the expression of the NaPi-2 molecules involves also their lysosomal degradation (12). Similar observations were made in studies on OK cells, a cell line with some proximal tubular characteristics (25, 26). The latter study also documented that recovery from PTH-inhibition requires de novo protein synthesis and that a recycling of “retrieved” transporters does not play a major role in the recovery process (26, 18).
Little is known on the mechanisms of initial PTH-dependent internalization of NaPi-2 molecules and on their subsequent intracellular routing to the lysosomes. In the present study, we characterize by immunofluorescence and by immunogold electron microscopy the effects of PTH on the cellular distribution of NaPi-2 molecules in rat kidneys. We infused also horseradish peroxidase (HRP) to label a general endocytotic pathway and performed immunogold labeling against NaPi-2, HRP, clathrin, and the adaptor protein 2 (AP2) to characterize the intracellular routing of NaPi-2.
Experimental animals. The experiments were performed with 12 male Wistar rats (BRL, Basel, Switzerland) of 180–200 g body wt, 3 animals per group. The rats had free access to laboratory chow and tap water. All rats were fed during 5 days a low-Pi diet (0.1% Pi). Rats in group 1 (low Pi) remained untreated and served as controls for groups 2–4. Rats in group 2 were treated with PTH. Rats in group 3were treated with HRP (low Pi + HRP), rats of group 4 were treated with HRP and PTH (HRP + PTH) prior to fixation of the kidneys.
Application of HRP and PTH and fixation. Rats were anesthetized with thiopental (Pentothal, 100 mg/kg body wt) injected intraperitoneally, their abdominal cavity was opened, and the aorta and vena cava were exposed. Animals of group 2 received 100 μg PTH (dissolved in 250 μl of 0.9% saline; Bachem, Bubendorf, Switzerland) as a single bolus injected into the vena cava. In animals of group 3, 2 ml of HRP-PBS solution (25 mg HRP, Sigma type II, dissolved in 2 ml PBS; Sigma, Buchs, Switzerland) was injected within 2 min into the vena cava. In rats of group 4, PTH (as in group 2) was injected immediately following the HRP injection. Fifteen minutes after injection, all animals were fixed by vascular perfusion via the abdominal aorta, at a pressure of 1.38 hectopascals, as described previously (5). The fixative consisted of 3% paraformaldehyde and 0.05% picric acid in 0.6 M cacodylate buffer (pH 7.4; containing 3 mM MgCl2 and adjusted to 300 mosmol/kgH2O with sucrose) and 4% hydroxyethyl starch in saline (HAES steril; Fresenius, Stans, Switzerland). After 5 min, the fixative in the right kidney, destined for immunohistochemistry, was washed out by perfusion for 5 min with cacodylate buffer. The left kidney, designated for immunogold electron microscopy, after perfusion fixation was postfixed for 2 h in the same fixative as described above, added with 0.1% glutaraldehyde.
Immunohistochemistry. Slices of fixed kidneys were frozen in liquid propane, cooled with liquid nitrogen, and mounted onto thin cork slices. Sections, 3 μm thick, were cut at −22°C in the cryomicrotome, mounted on chromalum/gelatine-coated glass slides, thawed, and stored in cold PBS until use. For immunofluorescence staining, sections were pretreated with 3% milk powder in PBS for 10 min. They were incubated overnight at 4°C with a rabbit anti-rat polyclonal antiserum against the NaPi-2 protein (4) diluted 1:500 in 3% milk powder in PBS containing 0.3% Triton X-100. Mouse monoclonal antibodies against HRP (Sigma) were diluted 1:250; monoclonal antibodies against the α-subunit of the adaptor protein 2 (AP2) (Sigma) and against clathrin (Progen, Bad Zurzach, Switzerland) were diluted 1:50. For double staining, the monoclonal antibodies were diluted in PBS/milk powder containing the NaPi-2 antiserum. Sections were then rinsed three times with PBS and covered for 45 min at 4°C with the secondary antibodies. Swine anti-rabbit IgG conjugated to fluorescein isothiocyanate (FITC; Dakopatts, Glostrup, Denmark) was diluted 1:50, goat anti-mouse IgG conjugated to Cy3 (Jackson ImmunoResearch Laboratories, West Grove, PA) was diluted 1:500 in PBS/milk powder, and these were used as secondary antibodies. Finally, the sections were rinsed three times with distilled water, coverslipped using DAKO-Glycergel (Dakopatts) containing 2.5% 1,4-diazabicyclo[2.2.2]octane (DABCO; Sigma, St. Louis, MO) as a fading retardant and studied by an epifluorescence microscope (Polyvar, Reichert-Jung, Austria).
Unspecific binding of the secondary antibodies to the tissue was tested by omitting the primary antibody. All control incubations were clearly negative.
Labeling for immunoelectron microscopy. For NaPi-2 labeling, we have affinity purified the polyclonal rabbit antiserum (4) (dilution 1:150) and have used the same mouse monoclonal antibodies as for immunohistochemistry. Dilution of monoclonal antibodies against HRP was 1:300. The monoclonal antibodies against clathrin and AP2 were diluted 1:100. Primary and secondary antibodies were diluted in PBS containing 3% bovine serum albumin, 0.02% Triton X-100, and 0.01% Tween-20. The goat anti-rabbit antiserum was coupled to 12-nm gold particles, and the goat anti-mouse serum was coupled to 6-nm gold particles. Gold-coupled secondary antibodies were diluted to an OD525 of 0.1–0.2 (12-nm gold particles) and of 0.06 (6-nm gold particles). Conjugation of the goat serum to gold particles was performed according to Slot and Geuze (28). Unspecific binding of the secondary antibodies to the tissue was tested by replacement of the primary antibody by either preimmune serum or PBS.
Ultrathin cryosections were prepared based on the methods of Griffiths et al. (6). Briefly, after tissue fixation as described above, small pieces of rat kidney were infiltrated with 2 M sucrose containing 10% (wt/vol) polyvinylpyrrolidone. The samples were rapidly frozen in liquid nitrogen; ultrathin sections of 80 nm were cut at −120°C using a Leica Ultracut UCT. The sections were transferred to polyvinyl chloride/polyvinyl acetate (0.3% in chloroform)-covered nickel grids, and stored on 1% gelatin at 4°C overnight. The nickel grids with the ultrathin cryosections were first incubated for 5 min in 50 mM NH4Cl to quench free aldehyde groups, then they were incubated for 10 min in PBS, containing 3% bovine serum albumin, 0.02% Triton X-100, and 0.01% Tween-20. Thereafter, the grids were transferred for 1 h at 37°C onto droplets of the primary antibodies, washed in PBS, and incubated for 1 h at room temperature on droplets with the respective secondary antibodies. Double labeling was performed by mixing both primary antibodies and both secondary antibodies (goat anti-rabbit 12-nm gold and goat anti-mouse 6-nm gold). Following incubations with secondary antibodies, the grids were rinsed with PBS and postfixed for 10 min in 1% glutaraldehyde in PBS. After postfixation, the grids were rinsed first with PBS, second with cacodylate buffer, and third with distilled water. The sections were contrasted in 2% (wt/vol) methylcellulose containing 0.2% uranyl acetate for 10 min and examined on a Philips model CM 100 electron microscope.
All experiments were made with rats, fed for 5 days a diet with low phosphate (Pi) content (0.1%). Previous studies have shown that this condition (similar to parathyroidectomy; 17) leads to marked upregulation of NaPi-2 in the BBM (14). Therefore the high NaPi-2 abundance in the BBM of rats adapted to low Pi intake is appropriate as a starting point for studying pathways involved in downregulation of NaPi-2 abundance in BBM, acutely induced by PTH injection. The results of the different treatments were homogeneous in each group. The micrographs in Figs. 1-4 are representative for the respective groups.
Distribution pattern of NaPi-2 in proximal tubules. In rats fed for 5 days a low-Pi diet (group 1), labeling for NaPi-2 is strong in the BBM. Very weak immunofluorescence is seen in the cytoplasm, immediately below the brush border. The perinuclear region is distinctly labeled (Fig.1 A). The NaPi-2 distribution observed in ultrathin cryostat sections extends above data. Gold particles, indicating sites of NaPi-2, are abundant on the brush-border microvilli (Fig. 1 C), some are seen on the membranes, invaginated into the apical cytoplasm, and in large vesicles in the subapical cytoplasm. Cisterns of the Golgi apparatus are distinctly labeled (Fig. 1 E), and occasionally gold particles are found in lysosomes (Fig. 1 F).
Animals of group 2 were injected intravenously with PTH. This treatment is known to downregulate NaPi-2 abundance in the BBM (11,17). Fifteen minutes after PTH injection, the brush border reveals a weaker immunofluorescence for NaPi-2 (Fig. 1 B) than in rats ofgroup 1 (Fig. 1 A). Yet, in contrast to the control group, a broad immunofluorescent rim is apparent in the cytoplasm below the brush border (Fig. 1 B). In ultrathin sections, the lower abundance of NaPi-2 in the brush border and the higher abundance in the subapical compartment (Fig. 1 D) is also apparent. The subapical region comprises the so-called vacuolar (endocytotic) apparatus, which consists of different-sized endocytotic vesicles and “dense” apical tubules (19). After PTH treatment, the dense apical tubules seem to be expanded. The gold particles are accumulated over the enlarged tubular profiles (Fig. 1 D). These tubulovesicular profiles of the vacuolar apparatus are the structural correlate for the subapical fluorescent rim in the proximal tubules of the PTH-treated rats (Fig. 1 B).
Colocalization of NaPi-2, AP2, and clathrin in PTH-treated rats. To elucidate whether proteins, which play an essential role in receptor-mediated endocytosis, are involved in NaPi-2 downregulation, we studied the distribution of clathrin and the adaptor protein AP2, together with NaPi-2. The heterotetrameric AP2 complex interacts with clathrin and the cytoplasmic domain of receptor proteins (15).
In proximal tubules of PTH-treated rats, the immunofluorescent distribution of AP2 is approximately congruent with the extension of the subapical rim (Fig. 2 A; compare with Fig. 1 B). Immunofluorescence for clathrin is restricted to a narrow rim at the top of the AP2-positive region (Fig.2 D). The brush border is negative for both proteins. Immunogold double labeling for NaPi-2 and AP2 reveals occasional colocalization of them at the base of the invaginations of the microvillous membrane (Fig. 2 B). Both proteins, AP2 and NaPi-2, are detectable in endocytotic vesicles (Fig. 2 C). Occasional colocalization of NaPi-2 and clathrin is seen exclusively in the membrane invaginations at the base of the brush border (Fig. 2 E).
Double labeling for HRP and NaPi-2. The distribution pattern of NaPi-2 in the proximal tubule after HRP injection is in all respects identical to that in group 1 (Fig.3 A; compare with Fig. 1 A). Fifteen minutes after HRP injection, a weak HRP immunofluorescence is apparent in the subapical cytoplasm below the unstained brush border. HRP seems to be accumulated in large vesicles immediately below this rim and in a few basally located vesicles (Fig. 3 B). The immunofluorescence pattern of NaPi-2 after injection of HRP followed by a PTH injection is identical to that after PTH alone (Fig. 3 C; compare with Fig. 1 B). In proximal tubules of PTH-treated kidneys, HRP is accumulated in rather coarse granules in the subapical cytoplasm, corresponding to the region of the enlarged structures of the vacuolar apparatus (Fig. 3 D; compare with Fig. 1,B and D). A few large fluorescent granules are seen below this rim and occasionally also in basal cytoplasm.
Double labeling of NaPi-2 and HRP at the electron microscopic level confirms that HRP and NaPi-2 occur in the same structures: in the membrane invaginations at the base of the brush-border microvilli (Fig.4 A), in profiles of small vesicles and tubules of the vacuolar apparatus (Fig. 4 B), in large endocytotic vacuoles (Fig. 4 C), and also in the lysosomes (Fig.4 D).
In the present study on rats, we trace by immunofluorescence and immunoelectron microscopy the pathway and fate of PTH-induced internalized NaPi-2 (rat type II Na-Pi cotransporter) in proximal tubular cells. By colabeling of NaPi-2 and of proteins, known to be involved in endocytosis, and by labeling the fluid phase endocytosis pathway by HRP injection, we demonstrate that the intracellular routing of NaPi-2 follows the pathway of internalized HRP. This pathway involves small endocytotic vesicles, which fuse with the cisterns and tubes (“early endosomes”) of the vacuolar apparatus, situated in the subapical compartment. From there both HRP and NaPi-2 are targeted via large endocytotic vacuoles (“late endosomes”) into lysosomes. These observations are in agreement with previous data from our laboratory which have clearly revealed that internalized NaPi-2 enters the degradative pathway (12). As an alternative, PTH-induced downregulation and subapical accumulation of NaPi-2 transporters could be explained by a reduced rate of insertion of newly and constantly synthesized NaPi-2 proteins into the BBM. Such a mechanism, however, would require a PTH-induced degradation of NaPi-2 cotransporters at the brush border site. Evidence for such a mechanism has not been obtained to date. The involvement of the subapical vacuolar apparatus in the pathway of an internalized membrane protein, not belonging to the class of receptors or receptor-bound proteins, had been suggested by prior data of our laboratory (16) and is substantiated by the present study. The vacuolar apparatus consists of a network with a complex three-dimensional structure, composed of so-called dense apical tubules, with their ends occasionally adorned with clathrin-coated buds. The dense apical tubules have a few connections to the apical membrane and are sometimes fused with larger endocytotic vacuoles (7). Clathrin-coated vesicles derived from the Golgi apparatus may fuse with this system and may also pinch off to fuse with the apical membrane. On the other side, clathrin-coated vesicles, derived from the apical membrane carrying cargo of endocytosed proteins fuse as well with this compartment (20, 24). Thus this compartment may be seen as a “meeting point” of protein trafficking to and from the apical membrane. Hence, the coincidence of HRP and internalized NaPi-2 in the same compartment is not surprising. A striking feature, however, is the transient marked enlargement of this compartment during NaPi-2 internalization. Fifteen minutes after injection of PTH, to initiate internalization of NaPi-2, the dense apical tubules are enormously enlarged and the vacuolar apparatus forms a large, even in light microscopic preparations, well-recognizable rim in the apical cytoplasm of proximal tubules. Studies from our laboratory (17a) have demonstrated that 1 h after PTH treatment the structure of this region has returned to its normal aspect. Cui et al. (3) recently documented similar structural changes in proximal tubules of rats after injection of the cytological stain light green, often used for staining of tubules in vivo. This stain is rapidly taken up into the cells. Cui et al. (3) interpreted their findings as a toxic effect of the light green. In the case of PTH-induced NaPi-2 downregulation, we assume that the rapid and transient enlargement of this compartment is rather an equivalent for rapid and massive membrane traffic than of toxicity. In rats adapted to a low phosphate intake and subsequently fed with high-phosphate diet, the enlargement of this compartment, filled with NaPi-2 transporters, is also observed at early time periods but not after prolonged feeding of high-Pidiets (27).
Clathrin-coated invaginations are found at the base of the microvilli. The occasionally observed colocalizations of NaPi-2 and the adaptor protein AP2 and clathrin suggest that PTH-induced internalization of NaPi-2 may involve clathrin-mediated endocytosis. The involvement of a clathrin-mediated internalization is supported by previous studies on OK opossum kidney cells showing that a reduction of the endocytosis rate by high medium osmolarity reduced the PTH effect on Na-Pi cotransport (10). Hypertonicity is known to inhibit clathrin-coated endocytosis (8). Furthermore, in most recent experiments on OK cells, we could document that OK cell-specific type II Na-Pi cotransporter (NaPi-4) labeled at the apical surface with biotin is internalized after PTH treatment (Jankowski, unpublished observations). PTH-induced internalization of type II Na-Pi cotransporters via a non-clathrin-mediated endocytotic pathway cannot be ruled out by our presented data on rat proximal tubules or on OK cells. Clathrin-mediated endocytosis is involved in regulating the membrane expression of other transport systems, e.g., the vasopressin-dependent water channel aquaporin-2 (AQP2) in collecting duct principal cells and LLC-PK1 cells (9, 2).
In the present study, we have used the fluid-phase marker HRP to label the endocytotic pathway and to analyze whether NaPi-2 follows the same route. HRP is filtrated in the glomerulus and immediately internalized in the proximal tubule via fluid-phase endocytosis at the base of the microvilli, subsequently found in subapical tubular structures and later on in the lysosomes (30). After 15 min, HRP is visible in the endocytotic apparatus, including coated pits, apical tubules, and apical vacuoles, whereas HRP is sparsely detectable in the lysosomes. The downregulation of NaPi-2 in the BBM and its increase in the subapical compartment coincides with an accumulation of HRP in the same subapical zone after PTH injection. This points to a cointernalization of NaPi-2 and HRP to the same cellular compartments.
In summary, we have shown in our study that PTH leads to an internalization of NaPi-2 from the BBM via the endocytotic invaginations between the base of the microvilli (Fig.5). Furthermore, after internalization, NaPi-2 and HRP have the same intracellular fate. They are directed via small endocytotic vesicles to the subapical tubulovesicular compartment, where they accumulate. This is associated with an expansion of the respective subapical structures. After passing through this compartment, both proteins are carried by large vacuoles, possibly derived from the subapical compartment, to the lysosomes for degradation. The precise mechanisms involved in moving NaPi-2 molecules from the top of a microvillus down to the clefts and the mechanisms involved in the PTH-induced retrieval of NaPi-2 from the BBM have yet to be determined.
We thank Heinz Sonderegger and Christian Gasser for professional assistance in preparing Figs. 1-5 for this report and Dr. Bruno Guhl for helpful assistance in performing the electron microscopy.
The study was supported by Swiss National Science Foundation Grant 31-47742-96 (to B. Kaissling) and Grant 31-46523-95 (to H. Murer).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
- Copyright © 2000 the American Physiological Society