Abstract
ABSTRACT. Megalin is the most abundant endocytic receptor in the proximal tubule epithelium (PTE), where it is concentrated in clathrin-coated pits (CCPs) and vesicles in the brush border region. The heterotrimeric G protein alpha subunit, Gαi3, has also been localized to the brush border region of PTE. By immunofluorescence GIPC and GAIP, components of G protein-mediated signaling pathways, are also concentrated in the brush border region of PTE and are present in megalin-expressing cell lines. By cell fractionation, these signaling molecules cosediment with megalin in brush border and microvillar fractions. GAIP is found by immunoelectron microscopy in CCPs, and GIPC is found in CCPs and apical tubules of endocytic compartments in the renal brush border. In precipitation assays, GST-GIPC specifically binds megalin. The concentration of Gαi3, GIPC, and GAIP with megalin in endocytic compartments of the proximal tubule, where extensive endocytosis occurs, and the interaction between GIPC and the cytoplasmic tail of megalin suggest a model whereby G protein-mediated signaling may regulate megalin’s endocytic function and/or trafficking.
The apical domain of the kidney proximal tubule epithelium (PTE) is specialized for reabsorption of filtered proteins and nutrients and is packed with compartments that are involved in endocytosis, including clathrin-coated pits (CCPs) and clathrin-coated vesicles (CCVs), apical tubules and vesicles, large apical vacuoles, and lysosomes (1,2). Megalin, a member of the LDL receptor (LDLR) family, is highly concentrated in CCPs and CCVs (2–4). Megalin is the most abundant endocytic receptor identified in PTE (4,5), and it is involved in the reabsorption of many components of the glomerular filtrate, including albumin (6), β2- and α1-microglobulin, peptides such as insulin (7) and prolactin (7), and vitamins (8).
Thus there is considerable information on the ligand-binding properties of the ectodomain of megalin, but relatively little is known about the interactions of its cytoplasmic tail or the signaling pathways that regulate its endocytic function and trafficking. Although the ectodomain of megalin shares common structural features with other LDLR family members, the intracellular domain of megalin is unique (3). It contains several putative functional motifs, such as SH3 (Src homology 3), SH2 (Src homology 2), and PDZ-binding motifs, a proline-rich region, and putative phosphorylation sites for casein kinase II and protein kinase C (9). Several cytoplasmic proteins have recently been reported to interact with the cytoplasmic tail (CT) of megalin, including ANKRA (9), which binds to the proline-rich region, Dab1 and Dab2, putative adaptor proteins (10,11), and the PDZ-containing proteins, GIPC, PSD-95, and OMP 25, which presumably interact with the megalin CT through its C-terminal, PDZ-binding motif (12).
In regard to signaling modules, the heterotrimeric G protein alpha subunit, Gαi3, was found to be most highly expressed in the apical region of PTE (S1 segment), concentrated at the base of the brush border (13,14). The function of Gαi3 in this specific location as well as its relationship with megalin are not clear. To gain insight into this question, we investigated the distribution in PTE of two regulatory components of G protein signaling pathways, GAIP and GIPC, which could link Gαi3 to megalin.
GAIP belongs to the regulator of G protein signaling (RGS) protein family; it is a GTPase-activating protein (GAP) for Gαi subunits, that downregulates G protein signaling events (15,16). GAIP is localized on CCPs and CCVs in all tissues and cells investigated, suggesting possible involvement in regulation of endocytosis and vesicular trafficking (17). GIPC is a PDZ domain-containing protein that was found to interact with GAIP (18). On yeast two hybrid, immunoprecipitation, and pull-down assays, GIPC binds to the cytoplasmic domains of several membrane proteins, including megalin (12), the Glut-1 transporter (19), semaphorin-F (20), neuropilin-1 (21), syndecan-4 (22), TrkA (23), and α6 integrin (24). Thus GIPC, like other PDZ domain-containing proteins, may function to cluster signaling molecules and membrane receptors in specific membrane microdomains.
In this study, we investigated the distribution of GAIP and GIPC in the kidney and present findings indicating that GIPC and GAIP are concentrated in the apical region of PTE and in glomerular epithelial cells (GECs) along with megalin and Gαi3. The enrichment of components of G protein-related signaling molecules in this location together with the direct interaction between megalin CT and GIPC suggest that novel G protein-mediated signaling pathways may be involved in the regulation of one or more aspects of megalin’s function in endocytic trafficking.
Materials and Methods
Materials
Male Sprague-Dawley rats (175 to 220 g) were purchased from Harlan Sprague Dawley (Indianapolis, IN). Dulbecco’s modified Eagle’s (DME) medium and fetal bovine serum (FBS) were from Life Technologies BRL (Gaithersburg, MD). FITC-conjugated donkey anti-rabbit F(ab′)2 and Texas red-conjugated donkey anti-mouse F(ab′)2 cross-absorbed against human, mouse, rat, chicken, and goat IgG were purchased from Jackson ImmunoResearch Laboratories (West Grove, PA). Goat anti-rabbit and goat anti-mouse IgG conjugates (5 or 10 nm gold) were purchased from Amersham Life Sciences (Arlington Heights, IL). The enhanced chemiluminescence (ECL) detection kit was from Amersham.
Cell Culture
LLC-PK1 cells were maintained in DME high-glucose medium with 10% FBS, 5% CO2/95% air. OK cells were grown in DME-Ham’s F-12 mix (DMEM-F12) containing 10% FBS. Rat L2 cells were grown in DME containing 10% FBS. All media were supplemented with 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM glutamine.
Antibodies
Anti-megalin mAb 20B, which recognizes the ectodomain of megalin (25,26), anti-LBD4 raised against the fourth ligand-binding domain of megalin (27), and anti-megalin (459) raised against an 18-amino acid peptide from the C-terminus of megalin (28), and anti-podocalyxin mAbs 1A and 5A (25) were prepared as described previously. Rabbit antiserum against full-length GIPC (18) and the N-terminus or C-terminus of GAIP (17) were prepared as described. Anti-GIPC IgG was affinity-purified on His-tagged GIPC fusion protein coupled to cyanogen bromide-activated Sepharose 4B (Pharmacia Biotechnology, Inc, Piscataway, NY) following a protocol described previously (29). Anti-FLAG (M2) mAb was purchased from Sigma (St. Louis, MO). Affinity-purified rabbit anti-Gαi3 IgG (EC) was obtained from Dr. Allen Spiegel (National Institutes of Health, Bethesda, MD).
Preparation of Brush Border and Microvillar Fractions
Brush border fractions were prepared according to Thuneberg and Rostgaard (30) with some modifications as described earlier (31). These fractions are enriched in large brush border fragments with intact apical membranes of PTE with microvilli, CCPs, and apical tubules and vesicles.
Microvillar fractions were prepared from kidney cortex as described (32). The resultant fractions are enriched in microvilli and CCPs.
Preparation of Membrane and Cytosolic Fractions from Cultured Cells and Rat Kidney
Cells were harvested in phosphate-buffered saline (PBS) containing protease inhibitors (1 mM phenylmethylsulfonyl fluoride, and 1 μg/ml chymostatin, leupeptin, antipain, and pepstatin-A). A postnuclear supernatant (PNS) was prepared by passing cells through a 30 1/2 G needle (10 times) before centrifugation (1000 × g for 10 min); the PNS was centrifuged at 100,000 × g for 1 h to yield membrane (pellet) and cytosolic (supernatant) fractions.
Rat kidney fractions were prepared as described (33) with some modifications. Briefly, kidneys were homogenized with a motor-driven Teflon pestle fitting tightly into the tube of a Potter-Elvehjem homogenizer for 2 min at 1800 rpm in buffer A (20 mM Tris, pH 8.0, 150 mM NaCl, 1 mM CaCl2) containing protease inhibitors. The homogenate was centrifuged (1000 × g for 10 min) to yield PNS from which membrane (100,000 × g pellet) and cytosolic (100,000 × g supernatant) fractions were prepared. The membrane fraction was resuspended in buffer A in the same volume as the cytosolic fraction. Equal volumes of PNS, membrane, and cytosolic fractions were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Membrane fractions were lysed in buffer B (1% Triton X-100, 50 mM Tris, pH 8.0, 80 mM NaCl, 2 mM CaCl2) containing protease inhibitors before centrifugation at 100,000 × g for 45 min. The protein concentration of the supernatant was determined by BCA Protein Assay (Bio-Rad, Hercules, CA) with bovine serum albumin as a standard.
In Vitro Binding Assays
GST-GIPC (18) or GST-RAP (34) were prepared as described. Ten micrograms of the respective GST fusion protein or GST were incubated with 25 μl glutathione-sepharose beads (Boehringer Mannheim, Summerville, NJ) in buffer B (final volume, 500 μl) at 4°C for 2 h. Beads were washed with buffer B (3×) and with buffer C (0.1% Triton X-100, 10 mM Tris, pH 7.5, 150 mM NaCl, 1 mM MgCl2, 1 mM CaCl2, 1 mM MnCl2) before incubation with 250 μg of rat kidney membrane lysate in buffer C (final volume, 500 μl) at 4°C overnight as described (12). Beads were washed with buffer C (3× over 15 min) and suspended in 20 μl of 2 × Laemmli sample buffer.
Immunoblotting
Cell and rat kidney membrane lysates, subcellular fractions, or bead-bound protein complexes were boiled in Laemmli sample buffer, separated on 6% or 10% SDS-polyacrylamide gels and transferred to PVDF membranes (Millipore, Bedford, MA). After blocking with PBS containing 5% calf serum and 0.1% Tween-20, PVDF membranes were incubated with primary antibodies for 1 h at room temperature before incubation for 1 h with horseradish peroxidase-conjugated goat anti-rabbit or goat anti-mouse secondary antibodies (Jackson ImmunoResearch Laboratories) and detection by ECL.
Immunofluorescence
Rat kidneys were perfusion-fixed in 8% paraformaldehyde (PFA), 100 mM phosphate buffer, pH 7.4, for 15 min, followed by 4% PFA in phosphate buffer for 1 h, after which samples were cryoprotected and frozen in liquid nitrogen as described (35). Semithin cryosections (0.5 to 1.0 μm) were cut on an Ultracut UCT microtome (Leica, Wetzlar, Germany) equipped with an FCS cryoattachment at −100°C, placed on gelatin-coated microscope slides and incubated with primary antibodies for 3 h at 4°C before incubation in cross-absorbed FITC-conjugated donkey anti-rabbit or Texas red-conjugated donkey anti-mouse F(ab′)2 for 1 h at 4°C. Samples were observed with a Zeiss Axiophot microscope (Carl Zeiss Inc., Thornwood, NY) equipped for epifluorescence. Images were captured by a Cohu camera and processed as tiff files using SCION Image software (Scion Corp., Frederick, MA). Images were colorized and superimposed using Adobe Photoshop 5.0 (Adobe Systems, Mountain View, CA) to show overlap in the fluorescence patterns.
Alternatively, samples were observed by deconvolution microscopy with the DeltaVision imaging system (Applied Precision, Issaquah, WA) coupled to a Zeiss S100 fluorescence microscope. For cross-sectional images of kidney semithin sections, stacks were obtained with a 150-nm step width to optimize reconstruction of the center plane image. Deconvolution was done on an SGI workstation (Mountain View, CA) using Delta Vision reconstruction software, and images were processed as tiff files using Adobe Photoshop 5.0.
Immunogold Labeling of Brush Border Fractions
For incubation in suspension, brush border fractions were fixed with 4% PFA in 100 mM phosphate buffer, pH 7.4, for 45 min, blocked with PBS containing 10% fetal calf serum and 0.07% glycine, and incubated with primary antibodies in suspension in blocking solution overnight at 4°C. They were then extensively washed overnight with PBS containing 5% fetal calf serum and 0.07% glycine, incubated with goat anti-rabbit (5 nm) and/or anti-mouse (10 nm) IgG-gold conjugates at 4°C, and processed for routine electron microscopy as described previously (31).
For ultrathin cryosectioning, brush border fractions were fixed in 4% PFA in 100 mM phosphate buffer, pH 7.4, for 45 min as above and then in 8% PFA in phosphate buffer for 15 min. Samples were then cryoprotected and frozen as described (35). Ultrathin (80-nm) cryosections were cut at −110°C and transferred to nickel grids coated with formvar and carbon. After quenching free aldehyde groups with PBS containing 0.01 M glycine, sections were incubated with primary antibody overnight at 4°C followed by 5 or 10 nm gold-conjugated goat anti-rabbit or mouse IgG for 1 h. Grids were stained for 20 min in 2% (vol/vol) neutral uranyl acetate, absorption-stained, and embedded in 0.1% uranyl acetate/0.2% methyl-cellulose/3.2% (vol/vol) polyvinyl alcohol before examination in the electron microscope.
Results
Megalin, GAIP, and GIPC Colocalize in the Apical Region of Proximal Tubule Epithelial Cells by Immunofluorescence
Megalin is highly enriched in CCPs located at the base of the microvilli in PTE (36,37). To investigate where in the kidney GAIP and GIPC are localized, we performed immunofluorescence on semithin cryosections of rat kidney. We found that both GAIP (Figure 1A) and GIPC (Figure 1D) are also concentrated in the apical region of PTE, where their staining partially overlaps (Figure 1, C and F) with that of megalin (Figure 1, B and E). GAIP also showed some diffuse cytoplasmic staining in PTE. GAIP (Figure 1, G and I) and GIPC (Figure 1, J and L) are also expressed in glomerular epithelial cells (GEC), where megalin is also expressed (32,37). To achieve better resolution of structures in the brush border region at the light microscope level, we carried out deconvolution microscopy of immunofluorescence specimens. Both GAIP (Figure 2A) and GIPC (Figure 2D) showed a punctate staining pattern in PTE corresponding to their vesicular distribution (17,18), and staining was most concentrated in the brush border region beneath the microvilli, where it partially overlapped with megalin (Figure 2, C and F). These results indicate that GIPC and GAIP are concentrated in the brush border region of PTEs and in GECs, where their distribution overlaps with that of megalin.
Figure 1. Distribution of GIPC, GAIP, and megalin in proximal tubule epithelium (PTE) and glomerular epithelial cells (GEC) of rat kidney. In PTE, staining for GAIP (A) and megalin (B) is concentrated at the base of the brush border. Overlap in the staining for GAIP and megalin at the base of the brush border is indicated by the yellow signals (arrows) in the merged image (C). GIPC (D) is also concentrated at the base of the brush border, where it partially overlaps with megalin (E), as indicated (arrows) by the yellow signals in the merged image (F). GAIP (G) and GIPC (J) are distributed throughout the cytoplasm of GCE. GEC are indicated by staining for podocalyxin (H and K). (I and L) Enlargement of GEC showing punctate staining (asterisks) for GAIP (I) and diffuse staining (asterisks) for GIPC (L). Rat kidneys were fixed and prepared for semithin cryosectioning, and sections were double incubated with anti-megalin mAb 20B, and either affinity-purified anti-GIPC or anti-GAIP (C-terminus) IgG followed by FITC-conjugated goat anti-rabbit or Texas red conjugated goat anti-rabbit F(ab′)2 as described in Materials and Methods. Scale bar: 5 μm in I and L; 10 μm in A through H, J, and K.
Figure 2. Codistribution of GIPC and GAIP with megalin in PTE observed by deconvolution microscopy. Punctate-staining for GAIP (A) and GIPC (D) is found in the cytoplasm of PTEs but is most concentrated at the base of the brush border, where there is significant overlap with megalin (B and E). Overlap is indicated by the yellow signals in the merged images (C and F). Enlargement of the brush border is shown on the right of each panel. Semithin cryosections were labeled as for Figure 1, and samples were analyzed by deconvolution microscopy as described in Materials and Methods. Scale bar, 4 μm.
GAIP, GIPC, and Gαi3 are Present in Cell Lines Expressing Endogenous Megalin
Several stable cell lines that express endogenous megalin have been used for studying ligand binding, endocytosis, and trafficking of megalin, including L2 rat yolk sac carcinoma cells (26,35), OK cells derived from opossum PTE (38), and LLC-PK1 cells from porcine PTE (39). The expression levels of megalin as determined by immunoblotting of cell lysates varies widely in these cell lines (Figure 3A). L2 cells express the highest levels, whereas OK and LLC-PK1 cells express lower but detectable levels of endogenous megalin. We found that Gαi3, GAIP, and GIPC are expressed in all three cell lines. By immunoblotting, the levels of expression of GIPC and Gαi3 are comparable (Figure 3A). GAIP is expressed at comparable levels in L2 and LLC-PK1 cells, but its detection in OK cells required prolonged exposure by ECL. GAIP is seen as two bands, 28- and 25-kD, in these as well as in other cell lines (40). We have previously shown that the 28-kD band represents the phosphorylated form of GAIP and the 25-kD band the nonphosphorylated form (40). We conclude that Gαi3, GAIP, and GIPC are co-expressed with megalin in the L2, OK, and LLC-PK1 cell lines.
Figure 3. Megalin, GIPC, GAIP, and Gαi3 are expressed in rat kidney and in L2, LLC-PK1, and OK cells. (A) Megalin (top panel) is expressed at highest levels in L2 cells, at lower levels in OK cells, and has the lowest level of expression in LLC-PK1 cells. GIPC and Gαi3 are expressed in comparable amounts in all three cell lines. GAIP is expressed at comparable levels in L2 and LLC-PK1 cells, but OK cells have a lower level of expression (lane 2). After a prolonged exposure of the same blot (5 min), GAIP is also detected (lane 4) in OK cells. Fifty micrograms of total cell lysate was immunoblotted with anti-megalin (459), anti-GIPC, anti-GAIP (N-terminus), or anti-Gαi3 (EC) before incubation with HRP-conjugated goat anti-rabbit IgG and detection by enhanced chemiluminescence. (B) GIPC and GAIP are present in both cytosolic (S) and membrane fractions (P) from rat kidney and L2, LLC-PK1, and OK cells. In rat kidney, approximately 50% of the GIPC and 70% of the GAIP are associated with membranes. The affinity-purified GAIP antibody recognizes two bands, 28-and 25-kD. The 28-kD form is seen primarily in the membrane fraction and represents phosphorylated GAIP, whereas the 25-kD form is in cytosolic fractions and represents nonphosphorylated GAIP (40). Two bands, 40- and 38-kD, are detected with affinity-purified GIPC IgG. The lower band seen only in kidney fractions may represent the GIPC homolog SEMCAP-2 (20). PNS prepared from L2, LLC-PK1, or OK cells or rat kidney was centrifuged at 100,000 × g to yield a crude membrane fraction (P) and a soluble fraction (S). These fractions were normalized to volume, immunoblotted with either anti-GIPC or anti-GAIP (N-terminus) followed by detection by enhanced chemiluminescence.
GIPC and GAIP are Present in Both Cytosolic and Membrane Fractions
Two pools—cytosolic and membrane-associated—of GAIP and GIPC have been found in all cultured cell lines examined to date (18,41). The association of GAIP with membranes is presumably via protein-protein interactions as well as lipid anchoring (palmitoylation) (41), whereas only protein-protein interaction is involved in the case of GIPC (18). To determine the distribution of GAIP and GIPC between membrane and cytosolic pools, we carried out immunoblotting on membrane (100,000 × g pellet) versus cytosolic (100,000 × g supernatant) fractions from the rat kidney and megalin-expressing cell lines.
GIPC and GAIP were present in both membrane and cytosolic fractions generated from rat kidney (Figire 3B, lanes 1 to 3). GIPC was equally distributed between the two fractions, whereas GAIP was mainly (approximately 70%) membrane-bound. The GIPC antibody recognized two bands at 40- and 38-kD in kidney membranes but not in other membranes. The lower band may represent the GIPC homologue SEMCAP-2 (20).
Similarly, GAIP and GIPC were also detected in both membrane and cytosolic fractions prepared from L2, OK, and LLC-PK1 cells (Figure 3B, lanes 4 to 12). GIPC was primarily distributed in cytosolic fractions, whereas more GAIP was present in membrane fractions: 40% (LLC-PK1 cells) to 90% (L2 and OK cells) of the GAIP and only 10% (OK and LLC-PK1cells) to 20% (L2 cells) of the GIPC was associated with membrane fractions. Interestingly, the 28-kD phosphorylated form predominated in the membrane fraction, whereas the 25-kD nonphosphorylated form was found mainly in the cytosolic fraction. This is in keeping with previous findings that phosphorylated GAIP is exclusively membrane-associated (40).
These results indicate that (1) there are cytosolic and membrane-associated pools of both GIPC and GAIP in rat kidney and in cell lines expressing megalin; (2) the amount of GIPC or GAIP associated with membranes varies from one cell type to another; and 3) in the rat kidney’s considerable GIPC (50%) and GAIP (70%) are associated with membranes.
Megalin, Gαi3, GAIP, and GIPC are Enriched in Renal Brush Border and Microvillar Fractions.
We next determined if GAIP and GIPC cosediment in brush border (30) and microvillar (32) fractions prepared from rat kidney. Both fractions contain CCPs as well as other components of the brush border. By immunoblotting (50 μg of protein), megalin, GAIP, GIPC, and Gαi3 were found to be enriched over cortical homogenate in both microvillar (Figure 4A, lane 2) and brush border (Figure 4B, lane 4) fractions. These results indicate that GAIP, GIPC, and Gαi3 cosediment with megalin in kidney fractions, which correlates with their colocalization by immunofluorescence.
Figure 4. Distribution of megalin, GIPC, GAIP, and Gαi3 in rat kidney fractions. (A) GIPC, GAIP, and Gαi3 are enriched in microvillar fractions (MV) compared with total kidney cortical homogenate (H). Megalin is only slightly enriched in microvillar fractions. (B) Megalin, GIPC, GAIP, and Gαi3 are enriched in brush border fractions (BB) compared with total kidney cortical homogenate (CH). Rat kidney microvillar or brush border fractions were prepared as described in Materials and Methods, and 50 μg of protein from each fraction was separated by SDS-PAGE before immunoblotting with anti-megalin (LBD4), anti-GIPC, anti-GAIP (N-terminus), or anti-Gαi3 (EC) and detection by autoradiography. (C) Megalin binds to GST-GIPC (lane 2) and GST-RAP (lane 3) but not to GST alone (lane 1). GST-GIPC, GST-RAP, or GST (10 μg) were prebound to glutathione-Sepharose beads and incubated with 250 μg rat kidney membrane lysate. Proteins bound to the beads were separated by SDS-PAGE before immunoblotting with anti-megalin (LBD4).
Interaction between GIPC and Megalin
It was previously shown that the CT of megalin binds GIPC and two other PDZ proteins, PSD -95 and OMP 25 (12). To determine if GIPC interacts with endogenous megalin, we performed in vitro binding assays in which we incubated rat kidney membrane lysates with GST-GIPC bound to glutathione-sepharose beads. We found that megalin binds to GST-GIPC (Figure 4C, lane 2).
Localization of Megalin, GAIP, and GIPC by Immunoelectron Microscopy
The brush border region contains a variety of compartments involved in endocytosis including CCVs, apical tubules, endosomes, and lysosomes (1). Megalin is most concentrated in CCVs, but it is also present at lower concentrations in apical tubules and endosomes. To find out with which organelles GAIP or GIPC are associated in the renal brush border, we performed immunogold labeling for GAIP and GIPC on ultrathin cryosections of brush border fractions (Figure 5) or those fixed and incubated in suspension with antibodies before embedding (31). We found that GAIP was largely localized to CCPs and CCVs (Figure 5, A and C), corresponding to its localization in other cell types. GIPC was localized mainly to apical tubules and vesicles of endocytic compartments (Figure 5E) and was occasionally seen on CCPs (Figure 5F). To find out if GAIP or GIPC and megalin are localized in the same structures, we carried out double immunogold labeling on brush border fractions. GAIP could be colocalized with megalin in CCPs and in some of the vesicles and tubules in the brush border region (Figure 5D). GIPC was predominantly localized in small tubules and vesicles where relatively little megalin labeling was detected under our conditions6 (Figure 5B). These results reveal that GAIP and GIPC are associated with endocytic compartments in very close proximity to megalin in the brush border region, where extensive and rapid endocytosis and recycling occur. Some GAIP and megalin are also found on the same CCPs.
Figure 5. Immunogold labeling of megalin, GAIP, and GIPC in rat kidney brush border. GAIP (A and C) is localized on CCPs (arrowheads). GIPC (E and F) is localized to apical tubules (arrowheads) and CCPs (arrow). (B) After double labeling, megalin (10 nm gold) and GIPC (5 nm gold) are seen on a clathrin-coated pit (CP), and GIPC is associated primarily with apical tubules (AT) (arrowhead). (D) Megalin (10 nm gold, arrow) and GAIP (5 nm gold, arrowhead) colocalize in the same CCP. (A through D) Renal brush border fractions were fixed as described in Materials and Methods and incubated with primary antibodies in suspension overnight at 4°C before overnight incubation with goat anti-rabbit (5 nm) and/or anti-mouse (10 nm) IgG-gold conjugates at 4°C and processed for routine electron microscopy. (E and F) Ultrathin cryosections of brush border fractions were prepared and incubated with affinity-purified rabbit anti-GIPC or anti-GAIP IgG and mouse mAb anti-megalin (20B) before incubation with 5 or 10 nm of gold-conjugated, goat anti-rabbit or anti-mouse IgG, respectively. Scale bar, 0.1 μm.
Discussion
Proximal tubules are the major site for the reabsorption of filtered proteins via receptor-mediated endocytosis. The brush border region of PTE is particularly rich in CCPs, small vesicles, endosomes, and apical tubules, which are components of the endocytic apparatus (1,2). The multiligand receptor, megalin, is the most abundant receptor in the brush border of the PTE and is the major receptor for uptake of filtered nutrients (42). It binds a variety of proteins such as albumin (6), insulin, prolactin (7), and vitamin-binding proteins (8,43,44). Megalin’s importance for the formation and function of the endocytic apparatus is indicated by the findings that PTEs of megalin knockout mice have impaired endocytosis and reduced numbers of coated pits, endosomes, and apical tubules in the brush border region (45). The role of megalin in endocytosis is now well established; however, the mechanism for regulation of megalin’s endocytic activities has remained unknown.
In this article, we present evidence suggesting that megalin’s function may be regulated in part by a G protein-mediated signaling pathway that includes Gαi3, GAIP, and GIPC. By immunofluorescence, we found both GIPC and GAIP in the coated pit region at the base of the brush border of PTE in the same location as megalin. We also found that GIPC binds to the cytoplasmic tail of megalin, confirming previous work by Gottherdt et al. (12). Both GAIP and GIPC are associated with components of the endocytic pathway with GAIP located predominantly on CCPs and GIPC on apical tubules. GAIP is a GAP for Gαi proteins, and GIPC is a scaffold protein that binds to the C-terminus of GAIP. We have recently shown that after stimulus of the δ-opioid receptor, a G protein coupled receptor, GAIP remains associated with clathrin-coated regions at the plasma membrane, whereas the receptor is internalized into endosomes (Elenko E, unpublished data). Moreover, when endocytosis was blocked by expression of a dominant interfering dynamin mutant (K44A) (46), GAIP and the receptor accumulated on the same CCPs after ligand stimulation. In addition, we previously showed that GIPC and GAIP can form a complex with the TrkA growth factor receptor, and GIPC but not GAIP colocalizes with retrogradely transported TrkA (23).
These findings together with our current findings in kidney suggest the following model (Figure 6). During internalization of megalin, GIPC is recruited to CCPs through an unknown mechanism that may be regulated by G proteins, where it binds GAIP and megalin. CCVs are rapidly internalized, GAIP remains on CCPs, GIPC dissociates from GAIP, and megalin and GIPC traffic together to endosomes. The fact that we seldom find GIPC on CCPs is likely due to its transient presence there. The concentration of a set of G protein-related signaling molecules in endocytic compartments in the brush border region together with the evidence that GIPC can interact with the cytoplasmic tail of megalin suggest that novel G-protein signaling pathways mediated via an unknown seven-transmembrane domain receptors may be involved in the modulation of one or more aspects of megalin’s function such as endocytosis and/or intracellular trafficking and delivery of ligands for degradation.
Figure 6. Model for the sequential association of GAIP and GIPC with megalin during receptor mediated endocytosis. Megalin and GAIP are associated with CCPs. Upon binding of ligand to megalin, GIPC is recruited to CCP, where it binds to the cytoplasmic tail of megalin. Upon internalization, GIPC traffics with megalin to recycling endosomes, but GAIP remains on CCPs. After ligand dissociation, megalin recycles back to the plasma membrane and GIPC returns to cytosol.
In addition to their well-known functions in regulating seven-transmembrane receptors signaling, G proteins have also been implicated in control of membrane trafficking (47–49). For example, Gα subunits have been implicated in formation of endocytic vesicles, fusion of incoming vesicles with endosomes (50), transcytosis (51), and fusion of secretion granules with the plasma membrane (52). Moreover, βγ subunits of G-proteins can inhibit dynamin GTPase (53), which regulates budding of CCVs from plasma membrane (54). Sequestration of βγ subunits by overexpression of Gαi1 resulted in decreased rates of receptor-mediated endocytosis of both LDL and transferrin receptors (53). Particularly relevant to our study, Gαi3 has been shown to regulate receptor-mediated endocytosis of albumin in OK cells (38). Megalin is the major receptor for albumin uptake in PTE (2,6); it is therefore reasonable to assume that Gαi3 might be involved in regulating megalin’s endocytic function.
It has been shown that different Gα subunits have distinctive localizations in the kidney (13,14). Gαs was found to be concentrated in glomeruli and in the basolateral membrane of cells of the thick ascending limb as well as PTE and principal and intercalated cells of the collecting duct, Gαi1 to the apical pole of the cells of the thick ascending limb and the papillary epithelium, and Gαi2 to principal cells of the collecting duct. Gβ was localized to the brush border of PTE and in glomeruli. Of particular interest for this work is the finding that Gαi3 is concentrated in intracellular vesicles in the apical region of PTE (13,14), where megalin is also located. It would be of great interest to carry out double labeling for Gαi3 and megalin, GAIP, or GIPC; however, Gαi3 antibodies suitable for immunoelectron microscopy are not currently available.
GIPC is ubiquitously expressed and has been reported to interact with the cytoplasmic domain of a number of membrane receptors or transporters other than megalin (12,19–21,23). The interaction between GIPC and its membrane partners could be determined by their tissue distribution and compartmentalization within cells. We found GIPC to be abundant in brush border and microvillar fractions and to be expressed in megalin-expressing cell lines. In general, the cytoplasmic domain of membrane receptors regulates the receptor’s endocytic trafficking or signal transduction cascades downstream of the receptor. GIPC interacts with the cytoplasmic domain of receptors from several different gene families with diverse functions, indicating that GIPC might be involved in rather general cellular functions, such as endocytic trafficking of receptors or crosstalk between different signaling networks (23).
Both megalin and GAIP contain PDZ-interacting motifs and presumably interact with the PDZ domain of GIPC. GIPC has been shown to be able to interact with itself likely through its N-terminal region (19). Therefore, it is possible for GIPC to form multimers and to cluster molecules such as megalin and GAIP.
A connection between G protein signaling and megalin was also suggested by previous studies showing that a pool of NHE3 in the brush border is associated with megalin (55) and the activity of NHE3 is regulated at least in part by cAMP-dependent protein kinase A (56,57). Therefore, it is logical to hypothesize that G proteins might indirectly regulate megalin’s function by regulating the level of cAMP and the activity of protein kinase A and NHE3.
It is notable that G protein signaling has been linked to several receptors other than GPCRs through unknown mechanisms. For example, PDZ-RGS3 mediates Ephrin-B reverse signaling and inhibits G protein-coupled chemoattraction (23). GIPC and GAIP form a complex with the nerve growth factor receptor TrkA, suggesting a link between G protein and growth factor receptor signaling (23). It has been established that some ligands of GPCRs such as lysophosphatidic acid can transactivate receptor tyrosine kinases such as the EGF receptor (23). Moreover, G protein signaling has been suggested to regulate the LDLR-related protein (LRP), also a member of the LDLR family (58,59). Our findings that several components of G protein signaling pathways are concentrated together with megalin suggest a possible linkage between G protein signaling and megalin’s functions or trafficking in kidney.
Acknowledgments
This work was supported by an American Heart Association California Affiliate Predoctoral Fellowship to Xiaojoing Lou and National Institutes of Health Research Grant DK17724 to Marilyn Gist Farquhar. Xiaojoing Lou is a graduate student in the Biomedical Sciences Graduate Program at the University of California San Diego. We thank Dr. Scott Emr, University of California San Diego, for use of his deconvolution microscope.
Footnotes
-
Dr. Orlando’s present affiliation: Department of Biochemistry and Molecular Biology, University of New Mexico Health Sciences Center, Albuquerque, New Mexico
- © 2002 American Society of Nephrology