Clathrin assembly protein, AP180, was originally identified as a brain-specific protein localized to the presynaptic junction. AP180 acts to limit vesicle size and maintain a pool of releasable synaptic vesicles during rapid recycling. In this study, we show that polarized epithelial Madin-Darby canine kidney (MDCK) cells express two AP180-related proteins: the ubiquitously expressed 62-kDa clathrin assembly lymphoid myeloid leukemia (CALM, AP180-2) protein and a novel high-molecular-weight homolog that we have named AP180-3. Sequence analysis of AP180-3 expressed in MDCK cells shows high homology to AP180 from rat brain. AP180-3 contains conserved motifs found in brain-specific AP180, including the epsin NH2-terminal homology (ENTH) domain, the binding site for the α-subunit of AP-2, and DLL repeats. Our studies show that AP180-3 from MDCK cells forms complexes with AP-2 and clathrin and that membrane recruitment of these complexes is modulated by phosphorylation. We demonstrate by immunohistochemistry that AP180-3 is localized to cytoplasmic vesicles in MDCK cells and is also present in tubule epithelial cells from mouse kidney. We observed by immunodetection that a high-molecular-weight AP180-related protein is expressed in numerous cells in addition to MDCK cells.
- clathrin assembly lympoid myeloid leukemia
- kidney epithelial cells
- epsin NH2-terminal homology domain
- DLL repeats
clathrin-mediated endocytosis is essential for diverse cellular functions and is coordinated by numerous proteins acting in a highly complex manner (for review see Refs. 24, 27, 37, 38). Investigations of synaptic vesicle biogenesis at the presynaptic nerve terminal have found that clathrin-mediated endocytosis is the principal mechanism for vesicle recycling (15, 28, 29, 33, 48). Discharge of neurotransmitters into the synaptic cleft results from the fusion of synaptic vesicles with the presynaptic membrane. The vesicular membranes then undergo rapid endocytosis by clathrin-mediated uptake. Once the clathrin-coated vesicles (CCVs) internalize, the clathrin coat is removed, and synaptic vesicles are reloaded with neurotransmitters and delivered to a cytoplasmic pool ready for discharge after a second stimulus (for review see Ref. 41). Hence, clathrin-mediated recycling is required to generate a pool of plasma membrane-derived storage compartments after regulated secretion in neural tissues.
The major structural component of neuronal CCVs is the clathrin triskelion, which consists of three identical clathrin light chains and three heavy chains. Many cytosolic accessory proteins, such as the heterotetrameric adaptor proteins (APs), which sequester membrane cargo into developing clathrin-coated pits, bind to the β-propeller structure of clathrin heavy chain. Another important component of neuronal CCVs, AP180 (previously called NP185, F1-20, AP-3, and pp155) (1, 22) serves to limit synaptic vesicle size during rapid endocytosis (46, 48). AP180 is a 180-kDa glycoprotein that has three main domains: a highly acidic central core, which binds to the heterotetrameric clathrin adaptor AP-2, and two flanking basic domains, which contribute to clathrin binding (30). Repetitive DLL motifs located in the COOH-terminal domain of AP180 promote the assembly of uniform clathrin cages by cross-linking the NH2-terminal domains of three adjacent clathrin triskelions (28, 32). The NH2 terminus of AP180 contains a conserved epsin NH2-terminal homology (ENTH) domain and appears to be required for clathrin-mediated endocytosis, allowing AP180 to bind to the plasma membrane via interaction of the ENTH domain with phosphatidylinositol-4,5-bisphosphate and to recruit clathrin through the COOH-terminal domain (13, 17, 20).
Although originally characterized in mammalian brain, AP180 homologs have been identified in several model organisms, including Like-AP180 (LAP) in Drosophila (48), UNC-11 in Caenorhabditis elegans (32), and Yap180a and Yap180b in yeast (45). LAP is found predominantly in the nervous system, and LAP knockouts exhibit impaired synaptic vesicle endocytosis as a result of deregulation of vesicle size (48). Although UNC-11 is similarly enriched in the nervous system, low-level expression has been detected in other tissues (32). Mammalian AP180 homologs have also been identified in tissues other than brain. In muscle, an AP180 splice variant has been found localized to postsynaptic nuclei near acetylcholine receptor clusters (3). The clathrin assembly lymphoid myeloid leukemia (CALM) protein is a ubiquitously expressed mammalian AP180 homolog that associates with clathrin and participates in clathrin recruitment to developing invaginations in the membrane (11, 42). Hence, although CALM lacks the central acidic region found in AP180, both proteins appear to carry out similar functions. Little is known about the physiological relevance of CALM in the cell compared with the well-established role of AP180 during synaptic vesicle cycling in neurons.
The action of brain-specific AP180 during synaptic vesicle endocytosis is tightly regulated by protein phosphorylation (8, 17, 31, 49). Neuronal AP180 can be phosphorylated by protein kinase A (PKA) and casein kinase II, but not by protein kinase C (PKC), and it is phosphorylation by creatine kinase II that results in the loss of AP180/AP-2 binding (8, 17). The changes in AP180 binding partners due to PKA phosphorylation are not known; however, PKA has been shown to regulate synaptic transmission along with PKC (6, 8). PKA and PKC are also implicated in regulation of apical secretion, clathrin-dependent and -independent endocytosis, and transcytosis in Madin-Darby canine kidney (MDCK) cells (5, 7, 12, 16, 19, 26). PKA-stimulated secretion and apically directed transcytosis in MDCK cells appear to be early events compared with PKA-stimulated apical endocytosis (12, 16). Although PKA and PKC have similar roles in endocytosis/exocytosis, the pathways are distinct.
Neurons have an axonal presynaptic domain and a dendritic postsynaptic membrane domain. Similarly, polarized epithelial cells have distinct plasma membrane domains, namely, the apical domain facing a lumen and the basolateral domain mediating interactions between cells and the underlying basement membrane (for review see Ref. 23). Many targeting routes in polarized neurons are recapitulated in MDCK epithelial cells, suggesting that diverse cell types utilize a number of common sorting pathways (23). In this study, we used the MDCK cell model to identify a novel AP180-related protein (AP180-3) that is expressed in polarized epithelial cells along with CALM. We report that AP180-3 forms complexes with AP-2 and clathrin and that membrane recruitment of these complexes is modulated by phosphorylation. Expression of similar AP180-related proteins was detected in most cell cultures and tissues tested, suggesting a new role for this class of accessory molecules in regulation of plasma membrane recycling.
MATERIALS AND METHODS
Cell lines, antibodies, and reagents. MDCK epithelial cells were grown in Eagle's minimum essential medium supplemented with 10% fetal bovine serum, 0.1 mM nonessential amino acids, and 2 mM glutamine at 37°C in a humidified chamber with 5% CO2-95% air. Cells were plated on Falcon tissue culture plastic (Becton Dickinson Labware, Franklin Lakes, NJ) or polycarbonate Transwell permeable filter supports (0.4-μm pore size; Costar, Cambridge, MA). Conditions for establishing electrically resistant cell monolayers on Transwell filters are described elsewhere (18).
The following antibodies were used in the study: mouse monoclonal antibody to the rat AP180 COOH terminus (Transduction Laboratory, Lexington, KY), mouse monoclonal antibody to the α-subunit of AP-2 (Sigma Chemical, St. Louis, MO), goat polyclonal antibody to CALM (Santa Cruz Biotechnology, Santa Cruz, CA), mouse monoclonal antibody to clathrin heavy chain (Transduction Laboratory), and rat monoclonal antibody to ZO-1 (Developmental Studies Hybridoma Bank, Dept. of Biological Sciences, University of Iowa, Iowa City, IA, maintained under National Institutes of Health contract NO1-HD-6-295). Fluorochrome-conjugated secondary reagents were purchased from Jackson ImmunoResearch Laboratories (West Grove, PA) or Molecular Probes (Eugene, OR) and horseradish peroxidase-conjugated secondary reagents from Amersham Pharmacia Biotechnology (Piscataway, NJ).
Forskolin, 3-isobutyl-1-methylxanthine (IBMX), and phorbol 12-myristate 13-acetate (PMA) were purchased from Sigma Chemical and used at 100, 10, and 1 μM, respectively.
Immunoblotting and immunoprecipitation. Immunoblotting was carried out using commercially available total protein samples prepared from rat tissue or cell lines (Transduction Laboratory) or cytosolic and membrane cell fractions prepared as follows. Cells were washed with PBS supplemented with 2 mM EDTA and 5 mM EGTA (PBS-CH) and then scraped in ice-cold homogenization buffer (HB) consisting of 10 mM HEPES, pH 7.4, 0.25 M sucrose, 1 mM EDTA, and protease inhibitors (0.2 mM phenylmethylsulfonyl fluoride and 1 μM leupeptin). Cells were collected by centrifugation, resuspended in HB, and homogenized with 22 strokes of a Dounce homogenizer. Postnuclear supernatants were separated into cytosolic and membrane fractions by high-speed centrifugation (14,000 g). Membrane pellets were solubilized with 1% (wt/vol) Triton X-100 in HB and centrifuged to remove all insoluble material. Samples were resolved by SDS-PAGE and transferred to nitrocellulose with the use of standard techniques (25, 43). Nitrocellulose blots that had been blocked with 5% dry milk dissolved in Tris-buffered saline + Tween 20 (TBS-T; 10 mM Tris, pH 7.4, 150 mM NaCl, and 1% Tween 20) were incubated overnight at 4°C with specific primary antibodies diluted in blocking solution. After the blots were washed extensively with Tris-buffered saline + Tween 20, they were incubated with the appropriate horseradish peroxidase-conjugated secondary antibody diluted in blocking solution for 1 h at room temperature and subjected to detection using enhanced chemiluminescence (Amersham Life Sciences).
For immunoprecipitations, samples were incubated with primary antibodies overnight at 4°C and then for 1 h with protein A-Sepharose CL-4B beads (Sigma Chemical) at room temperature. Where appropriate, protein A beads were pre-incubated with species-specific rabbit IgG to facilitate binding of the primary antibody. After they were washed thoroughly with STN buffer (10 mM Tris, pH 7.4, 150 mM NaCl, and 0.25% NP-40), immunoprecipitates were solubilized with Laemmli buffer, subjected to SDS-PAGE, and transferred to nitrocellulose for detection by enhanced chemiluminescence, exactly as described above.
Data analysis. Densitometric analysis was carried out using a UMAX Astra 2100U scanner and Scion Image 1.62C software and NIH Image software. Further data processing was performed with Excel (Microsoft). All quantitative data are expressed as means ± SD. Comparison between two groups was analyzed using Student's two-tailed paired t-test.
Reverse transcription-polymerase chain reaction. AP180-related cDNA sequences corresponding to residues 1-2341 were obtained by reverse transcription (RT) of RNA from MDCK cells or rat cortex made using TRIzol reagent (GIBCO-BRL Life Technologies, Gaithersburg, MD) followed by polymerase chain reaction (PCR). RT was performed with SuperScript RNase H- reverse transcriptase (GIBCO-BRL) and oligo(dT)16 (Perkin Elmer, Boston, MA) for 1 h at 37°C using 1 μg of total RNA. After an additional 5-min incubation at 100°C, samples were diluted to 1:20 with distilled H2O and amplified by PCR using Taq polymerase (Hoffmann-La-Roche, Basel, Switzerland) or Expand high-fidelity PCR system (Roche Molecular Systems, Mannheim, Germany). The four forward and reverse primer sets, based on sequences from rat neuronal AP180 (accession no. X68877), were as follows: 1) 5′-ATGTCGGGCCAAACGCTCACG-3′ and 5′-GTCGGCCCCTTTCTTCACTCTGG-3′, respectively, designed to anneal to sequences beginning at the 5′-start codon (in italics) and ending at nucleotide 515; 2)5′-GAAAGGGGCCGACGGTGTAATGAG-3′ and 5′-GTTGTGGCTGGAGAAGACT-3′, respectively, designed to anneal to sequences beginning at nucleotide 503 and ending at nucleotide 967; 3) 5′-GAAAGGGGCCGACGGTGTAATGAG-3′ and 5′-CTACCTTCAGATGGGGCAAA-3′, respectively, designed to anneal to sequences beginning at nucleotide 503 and ending at nucleotide 1482; and 4) 5′-GAAAGGGGCCGACGGTGTAATGAG-3′ and 5′-CCAAGATTGCCTACTAAGCTGGC-3′, respectively, designed to anneal to sequences beginning at nucleotide 503 and ending at 2341. PCR products were resolved on a 1% agarose gel for size determination and ligated to a pT-Adv PCR cloning vector using AdvantTAge PCR cloning kit (Clontech, Palo Alto, CA). Competent TOP10F′ Escherichia coli was transformed with ligation products, and cloned inserts were sequenced using M13, SP6, and T7 primers plus internal primers when necessary (Cleveland Genomics, Cleveland, OH). DNA and amino acid homology searches were carried out using the BLAST program from the National Institutes of Health.
Clathrin adaptor analysis. Cellular subfractionation for clathrin adaptor localization was carried out as described previously (36) with modifications. Briefly, cells were washed and scraped using PBS-CH supplemented with protease inhibitors. Cells were collected by centrifugation and resuspended in lysis buffer (0.1 M MES, pH 6.5, 20 mM EGTA, 0.5 mM MgCl2, 0.02% NaN3, and 10 mg/ml BSA) supplemented with 1% (wt/vol) NP-40 and protease inhibitors. Postnuclear supernatants were centrifuged at 60,000 g for 30 min at 4°C in a fixed-angle rotor (model TL 100.3, Beckman Instruments, Palo Alto, CA). Supernatants containing cytosolic proteins were saved, and membrane pellets were resuspended in lysis buffer supplemented with 1% (wt/vol) NP-40. Peripheral membrane proteins were released from cell membranes by the addition of 50 mM Na2CO3. Samples were centrifuged for 30 min at 50,000 rpm at 4°C, and supernatants containing alkaline-sensitive peripheral membrane proteins were saved. Alkaline-stripped membrane pellets were resuspended in lysis buffer supplemented with RIPA detergent [1% (wt/vol) NP-40, 0.5% (wt/vol) deoxycholate, and 0.1% (wt/vol) SDS] to solubilize alkaline-resistant membrane proteins.
Confocal laser scanning microscopy. Kidney and brain from C57BL/6J mice were placed in embedding solution and quick-frozen in liquid nitrogen. Cryostat sections (10-μm) were placed on poly-l-lysine-coated slides. Tissue sections and filter-grown MDCK cells were fixed with a solution of freshly prepared 3% paraformaldehyde (Electron Microscopy Sciences, Fort Washington, PA) for 10 min at room temperature, washed with PBS, and then incubated with 50 mM NH4Cl for 10 min to quench residual paraformaldehyde. Samples were permeabilized with 0.2% (wt/vol) Triton X-100 for 10 min and then incubated with 3% BSA to block nonspecific binding. Samples were incubated with primary antibody overnight at 4°C, washed with PBS containing 5% horse serum, and incubated with 5% normal serum from the host animal used to raise the secondary antibody for 15 min. Samples were incubated with appropriate secondary antibodies conjugated to a fluorescent marker for 1 h at 37°C and then rinsed extensively with PBS-5% horse serum. For dual staining, samples were incubated in the second primary antibody for 1 h at 37°C before addition of an appropriate secondary antibody conjugated to a separate fluorescent marker. Polycarbonate membranes were excised from plastic inserts and mounted cell side up on a glass slide. Tissue sections and filters were covered with a drop of SlowFade mounting reagent (Molecular Probes) and then a coverslip. Cells were viewed using a scanning laser confocal microscope (model LSM 410, Zeiss, Gottingen, Germany) using the 488/568-nm wavelength lines of an argon-krypton laser and a ×100 Plan-Neofluor oil objective. The cell monolayer was optically sectioned every 1 μm at a resolution of 512 × 512 pixels, and z sections were digitally compiled using Zeiss LSM software.
Expression of AP180-related proteins in MDCK cells. MDCK cells derived from canine kidney distal tubules are widely used as a model for the study of epithelial cell polarity. We initially sought to determine the expression of 62-kDa CALM protein in MDCK using an antibody that recognizes a shared amino-terminal epitope found in CALM and brain-specific AP180 protein (Fig. 1A). When used for immunoblotting, this antibody unexpectedly detected a second protein in addition to CALM, with an approximate molecular weight of 160 kDa (data not shown). Therefore, to confirm that CALM and a second AP180-related protein are expressed in MDCK cells, immunoblots with crude fractions of soluble and membrane proteins from Dounce-homogenized cells were probed with antibodies specific for the COOH terminus of each protein (Fig. 1A). We found that MDCK cells do express a high-molecular-weight protein specifically recognized by a monoclonal antibody to the COOH terminus of AP180 (Fig. 1B). The high-molecular-weight AP180-related protein was detected predominantly in the crude soluble fraction, whether cells were grown on Transwell filter inserts or on tissue culture plastic (Fig. 1C). This is similar to the finding that rat brain AP180 is enriched in the cytosol relative to CCVs under basal conditions (44). An antibody against the COOH terminus of CALM was specific for a 62-kDa protein (Fig. 1D). In contrast to high-molecular-weight AP180-related protein, however, a significant amount of CALM was also present in the crude membrane fraction under basal conditions. Thus, although a high-molecular-weight AP180-related protein and CALM are expressed in MDCK cells, they are found in different fractions of the cell under basal conditions.
AP180 sequence comparisons. On the basis of results in Fig. 1, we reasoned that MDCK cells should express an AP180-related mRNA species with high homology to published sequences for the brain-specific protein AP180. We therefore carried out RT-PCR amplifications using total RNA extracted from MDCK cells and primer sets chosen to obtain PCR fragments corresponding to nucleotides 1-2341, of a total of 3682 nucleotides, in the rat brain AP180 sequence (Fig. 2A). This strategy was devised to obtain a sequence corresponding to the entire NH2-terminal region of the MDCK protein, where the sequences of other AP180 homologs are most conserved, as well as sequences leading into the acidic central domain (Fig. 2D). As a control, experiments were also carried out using total RNA extracted from rat brain. Amplification of the RNA directly by PCR without RT yielded no detectable product on ethidium bromide-stained agarose gels, confirming that the RNA was not contaminated with cDNA or genomic DNA (data not shown).
A total of six PCR products obtained from MDCK-derived RNA were sequenced and compared with the reported sequence for rat AP180 (accession no. NM031728) and rat CALM (accession no. AF041374; Fig. 2B). The nucleotide sequence for MDCK-1 is from PCR products obtained with primer sets 1, 2, and 3, and that for MDCK-2 is from PCR products obtained with primer sets 1, 2, and 4 (Fig. 2B). Overall analysis of the nucleotide sequences for MDCK-1 and MDCK-2 revealed 89% homology with rat AP180 mRNA. The most obvious differences between the MDCK sequences and the rat brain AP180 were internal deletions. The nucleotide sequence of MDCK-1 lacked a span of 180 nucleotides corresponding to rat brain AP180 nucleotides 1140-1320, and the nucleotide sequence of MDCK-2 showed a deletion of 558 nucleotides corresponding to rat brain AP180 nucleotides 1140-1698. These data suggest that MDCK cells express at least two mRNAs that are splice variants of the rat brain AP180 message. MDCK-1 and MDCK-2 have a common 5′-nucleotide splice site that was obtained in four independent clones. However, the sequences have variable 3′-splice sites. We predict that, under basal conditions, MDCK-2 is the major transcript expressed because of lack of clones obtained for the MDCK-1 sequence with the use of primer set 4. The nucleotide sequence of MDCK-1 was detected using a 3′ primer corresponding to sequences that are absent from MDCK-2. MDCK-1 and MDCK-2 showed a 75% identity to rat CALM in a high-homology region (nucleotides 1-848), whereas the same region of high homology in rat AP180 demonstrated a 72% identity with rat CALM.
Except for a predicted internal deletion of 186 residues starting at amino acid 379, the deduced amino acid sequence of the AP180-related protein sequence corresponding to the MDCK-2 nucleotide sequence was 96% homologous to rat brain AP180. Known attributes of rat brain AP180 are highly conserved in the MDCK protein. These domains include the NH2-terminal ENTH domain (21) and the binding site for the α-subunit of AP-2 (2, 5, 16, 19, 26) (Fig. 2C). The COOH-terminal sequences obtained for the MDCK protein also contain several DLL repeats found in the rat brain AP180 (Fig. 2C), although one of the DLL repeats found in AP180 has been lost in the MDCK protein because of the internal deletion (28, 32). Immunodetection of the COOH terminus of the MDCK protein with AP180-specific antibody (Fig. 1B) suggests a strong homology in that region (amino acids 706-896). Hence, the MDCK protein appears to be a novel species that is more highly related to rat brain AP180 than other homologs that have been reported in the literature (Fig. 2D). The major difference between the two proteins is an internal deletion of 186 amino acids in the MDCK protein involving the central region of the brain-specific protein. We have named the MDCK protein AP180-3 on the basis of its homology to rat brain AP180 and CALM (also known as AP180-2).
AP180-3 expressed in MDCK cells associates with clathrin and AP-2. On the basis of the high degree of sequence homology to AP180 from rat brain, we hypothesized that AP180-3 might have a function similar to that of the neuronal form of the protein. The role of AP180 in neuronal cells is to regulate CCV assembly through association with clathrin (28, 30, 47) and AP-2 (17). To ascertain whether AP180-3 interacts with the same molecules in MDCK cells, cell fractions were incubated with monoclonal antibodies directed against clathrin heavy chain or the α-subunit of AP-2, and immunoprecipitates were analyzed by immunoblotting with the AP180-specific antibody. MDCK cells were lysed using relatively mild conditions to better preserve low-affinity membrane-protein interactions, and high-speed membrane pellets were then extracted with NaCO3 to release alkaline-sensitive peripheral membrane proteins (36). Cell fractions containing soluble, peripheral alkaline-sensitive membrane, or alkaline-insensitive membrane proteins were immunoprecipitated with monoclonal antibodies to clathrin heavy chain (Fig. 3A) or the α-subunit of AP-2 (Fig. 3B). Recovered proteins were resolved by SDS-PAGE and transferred to nitrocellulose filters for immunoblot analysis. AP180-3 was coimmunoprecipitated by antibodies to clathrin (Fig. 3A) or AP-2 (Fig. 3B), but not with an isoform-matched monoclonal antibody to cathepsin D (Fig. 3C). On the basis of the amount of AP180-3 that was coimmunoprecipitated, we conclude that AP180-3-clathrin complexes are stable in the soluble cell fraction, in contrast to AP180-3-AP-2 complexes, which are stable in peripheral membrane as well as soluble protein fractions. Interestingly, MDCK cells appear to express two isoforms of the α-subunit of AP-2: a ubiquitous C isoform (αC) and an A isoform (αA) abundantly expressed in certain tissues including brain (35).
Analysis of AP180-3 localization in forskolin- and PMA-treated cells. Because of the known effect of phosphorylation of neuronal AP180, we examined the effect of PKA and PKC on AP180-3 localization in MDCK cells using confocal laser scanning microscopy. Filter-grown cells were fixed, permeabilized, and costained with an AP180-specific antibody and an antibody to the tight junction protein ZO-1 to distinguish apical and basolateral membrane domains. MDCK cells under basal conditions showed primarily cytosolic AP180 localization with little vesicular-like staining (Fig. 4A). On the basis of the PKA treatment of MDCK cells described by Hansen and Casanova (16), forskolin was added for 15 min to promote maximal exocytotic events. MDCK cells showed a general increase in staining of AP180-3 and, more specifically, to areas beneath the apical plasma membrane (Fig. 4B). The increase in AP180-3 staining suggests that AP180-3 has undergone a conformational change that allows for increased immunoreactivity due to epitope exposure. PKC stimulation was performed by the addition of 1 μM PMA for 15 min for optimal apical endocytosis as stated by Holm et al. (19). PKC showed no change in the immunostaining pattern of AP180-3 in MDCK cells (Fig. 4C).
To determine whether AP180-3 protein is being relocated in the cell as a result of PKA stimulation, a time course of an IBMX-forskolin treatment was performed on MDCK cells. Cells were fractionated and analyzed for AP180-3 enrichment in the cytosolic fractions (Fig. 5). We observed a transient increase of 17% at 10 min after stimulation in the amount of soluble cytosolic AP180-3 compared with the control value at time 0, reflecting the release of AP180-3 into the cytosol due to activated PKA. The amount of soluble AP180-3 then declines after the peak at 10 min, suggesting that AP180-3 becomes recruited back to membrane fractions (Fig. 5, A and D). To test whether this effect was specific for AP180-3, immunoblot analysis of soluble cytosolic proteins was also performed with an antibody to the COOH terminus of CALM. In contrast to AP180-3, the abundance of CALM in the soluble fraction did not change in response to forskolin stimulation (Fig. 5, B and D). The overall effect of forskolin was not due to an increase in protein expression, because the total amount of AP180-3 present in the cell was unchanged by the activation of PKA (Fig. 5C). Taken together, these data suggest that forskolin-activated PKA results in a release of AP180-3 from membranes to the cytosol.
Effect of forskolin and PMA treatment on complex formation. To ascertain the effect of PKA and PKC on the localization of AP180-3 in subfractionations of MDCK cells and the ability of AP180-3 in each fraction to form a complex with clathrin, coimmunoprecipitations were performed on untreated cells, cells treated with IBMX and then with forskolin, and cells treated with PMA. Cell fractions containing soluble proteins, alkaline-sensitive peripheral membrane proteins, or detergent-solubilized membrane protein fractions were immunoprecipitated with antibody to AP180. Under basal conditions, AP180-3 is present in all three fractions, with smaller portions contained in the alkaline-sensitive peripheral membrane fraction and detergent-solubilized membrane fraction (Fig. 6A). Forskolin stimulation increases the pool of cytosolic AP180-3 and reduces the alkaline-sensitive peripheral membrane fraction and the detergent-solubilized membrane fraction (Fig. 6, B and G). The addition of PMA caused a relocation of AP180-3 from the cytosolic fraction to the membrane fractions (Fig. 6, C and G). The ability of AP180-3 to associate with clathrin was then determined by immunoblot analysis using a clathrin heavy chain antibody (Fig. 6, D-F). Similar to untreated cells, clathrin was coimmunoprecipitated by the AP180-specific antibody from the soluble fraction in forskolin-treated cells (Fig. 6, D, E, and H), suggesting that the ability of the proteins to form a complex was unaffected by activation of PKA. In contrast to PKA, the amount of clathrin coimmunoprecipitated with AP180-3 in cells stimulated with PMA decreased in the soluble fraction, whereas the peripheral and membrane fractions showed no change (Fig. 6, F and H). Clathrin decrease from the soluble fraction is not carried over to the other fractions, suggesting that the AP180-3-clathrin complex is compromised.
Localization of AP180-3 in the mouse kidney. To establish whether a similar AP180-related protein is present in native kidney epithelial tissue, mouse kidney and brain were excised and sectioned for immunohistochemistry using an AP180-specific antibody. Kidney sections were also stained with a tight junction-specific antibody to identify epithelial cells. Figure 7A shows the localization of a cross-reactive protein to the AP180 antibody to vesicles in the kidney epithelial cells. As a control, a serial section of the kidney was incubated with secondary antibody alone along with the tight junction marker, and vesicular staining was not detected (Fig. 7B). Similarly, brain sections, stained as a control, showed extensive AP180 reactivity but no staining with secondary antibody alone (Fig. 7, C and D). The ability to detect an AP180 cross-reactive protein in epithelial cells from intact kidney cells strengthens the conclusion that AP180-related protein is expressed in cell types other than neuronal tissue, specifically the epithelial cells of the kidney. Our study of MDCK cells would suggest that this AP180-related protein is AP180-3.
High-molecular-weight AP180-related protein expression is ubiquitous. The presence of AP180-3 in kidney epithelial cells suggests that functions associated with the brain-specific protein are not restricted to synaptic vesicle recycling. Because diverse cell types are known to have common sorting pathways, we sought to determine the extent of AP180-related protein expression in other cell culture systems and organ tissue. Figure 8A shows the expression of a high-molecular-weight AP180-related protein in different tissue culture cells. All cells tested were positive for AP180-related protein expression by immunoblot analysis with specific antibody to AP180, except for mouse macrophages. All organ tissue sources tested also showed AP180 reactivity, except for liver (Fig. 8B). The size variation among tissue samples may reflect splice variants or different extents of AP180 phosphorylation (31). The expression of AP180-related protein in cell culture and organ tissue remained relatively low compared with the expression in neuronal tissue, inasmuch as the concentration of rat cerebrum was 1/10th lower in the immunoblot analysis to allow for development without interference to other lanes. Cell culture systems, which represent a homogeneous population of cell types, maintained AP180-related protein expression. Although it can be argued that AP180 cross-reactivity in tissues is due to contaminating neuroendocrine cells, the ability to detect high cross-reactive species in homogeneous tissue cultured cells suggests that the expression of high-molecular-weight AP180-related protein is relatively widespread in mammalian cells.
This study provides evidence that polarized kidney epithelial cells express a novel AP180-related protein with extensive homology to brain-specific AP180 that we have called AP180-3. The proteins share many common features, including conserved ENTH domain, AP-2 binding domain, and DLL repetitive elements. The most striking difference between the major form of the AP180-3 protein found in MDCK cells and brain-specific AP180 is the internal deletion of 184 amino acids in the AP180-3 protein involving the central core region of the brain-specific protein. The known AP180 phosphorylation sites for casein kinase II map to this region (17), suggesting that the two proteins may be regulated in different manners in the two cell types. The proposed function of AP180 in neurons is to promote the assembly of uniform clathrin cages during rapid retrieval of synaptic vesicles by clathrin-mediated endocytosis (47, 48). AP180-related proteins may fulfill an analogous role in nonneuronal cells, under circumstances when rapid uptake of plasma membrane is necessary to maintain steady state (5). Our data suggest that MDCK-expressed AP180-3 is localized predominantly to the cytosol in association with clathrin. PKA stimulation, which regulates secretion primarily by increasing the rate of exocytosis (5), leads to an increase of AP180-3 in the cytosol from the membrane fraction. This increase is followed by a decline in levels, which may represent recruitment back to the membrane, consistent with a role for AP180-3 during rapid recycling after regulated secretion.
A summary of results obtained in MDCK cells also draws on results from studies carried out on neuronal AP180. Secretory vesicles release their contents from the apical cell surface after fusion with the plasma membrane by a mechanism that is stimulated by PKA. The vesicle membrane is then retrieved by clathrin-mediated endocytosis. Because AP180-3-clathrin and AP180-3-AP-2 complexes are found in the cytosol, the signal for CCV assembly may recruit each of the protein complexes to the membrane, creating a high-density pool of clathrin and adaptor proteins available to be sequestered into the developing clathrin-coated pit. It is highly likely that AP180-3 and AP-2 act coordinately to bind clathrin, because neuronal AP180 and AP-2 complexes assemble clathrin more efficiently than either protein alone (17). AP180-3 and/or AP180-3-AP-2 complexes link clathrin to the plasma membrane by attaching to phosphatidylinositol-4,5-bisphosphate by the ENTH domain of AP180-3, during which time AP180-3 could regulate uniform vesicle size (13, 14, 20). CCVs then detach from the plasma membrane, and endocytosed vesicles undergo uncoating, releasing clathrin and associated adaptor proteins into a cytosolic pool, where they can be reused during vesicle cycling after a second stimulus. It is also possible that cytosolic AP180-3 serves to block inappropriate clathrin lattice formation by competing for the clathrin-binding site on the β-subunit of AP-2, because neuronal AP180 has been shown to bind to the α- and β-subunits (17). Our study of coimmunoprecipitations shows that AP180-3-clathrin complexes in the cytosol are reduced by PKC (Fig. 6F), consistent with the finding that PKC stimulates endocytosis in MDCK cells (19). Hence, although PKA and PKC are implicated in the apical-regulated endocytosis/exocytosis in MDCK cells, our data indicate that these kinases act on AP180-3 by two separate mechanisms.
MDCK cells also express CALM, a ubiquitously expressed homolog of AP180. CALM participates in clathrin-mediated endocytosis by binding directly to clathrin and, similar to AP180, contains an ENTH domain that aides in binding to phosphatidylinositols. We found that CALM is more tightly associated with the cell membrane than AP180-3 and that its membrane association is not regulated by PKA. This suggests that CALM may be involved in constitutive trafficking events, whereas AP180-3 participates in rapid recycling required to generate a pool of plasma membrane endosome-derived storage compartments after regulated secretion. Hence, although the function of the two proteins is thought to be similar, differences in their subcellular distribution may imply distinct roles during clathrin-mediated trafficking.
An increased number of molecules involved in membrane trafficking once believed to be restricted to certain cell types are now found in several cell types. For example, synapsin I, first identified as a component of the synaptic vesicle trafficking machinery in neurons, has recently been found in epithelial cells (4). The failure to identify AP180 homologs in tissues other than brain in the original studies that defined AP180 as a neuronal-specific protein may be due to the available antibody reagents that have been raised against rat brain cortex (39, 40). Because many AP180 homologs are splice variants of the brain protein, they may not be recognized by these antibodies. Also, the expression level of AP180-3 in MDCK cells is considerably lower than the level of AP180-3 expressed in brain and, therefore, required a greater sensitivity for detection. Although the functional significance of AP180 homolog expression in diverse cell types remains unknown, it will be interesting to decipher specialized trafficking steps regulated by this growing family of endocytosis accessory molecules.
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants P50-DK-54178 and P50-DK-57306 (to C. Carlin). L. Kusner was supported by National Heart, Lung, and Blood Institute Training Grant T32-HL-07653.
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. Section 1734 solely to indicate this fact.
- Copyright © 2003 the American Physiological Society