Association of CD2AP with dynamic actin on vesicles in podocytes
http://www.100md.com
《美国生理学杂志》
Department of Anatomy and Cell Biology I, University of Heidelberg, and European Molecular Biology Laboratories, Heidelberg, Germany
Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri
ABSTRACT
The docking protein CD2AP (CD2-associated protein) serves a nonredundant function in podocytes as CD2AP knockout mice die of renal failure at the age of 6–7 wk. Furthermore, haploinsufficiency due to mutation of the CD2AP gene is associated with focal segmental glomerulosclerosis in humans. Although CD2AP has been shown to interact with proteins regulating actin polymerization, with proteins of the slit diaphragm, and with the endocytic machinery, its critical function in podocytes remains unclear. In conditionally immortalized mouse podocytes, we demonstrate that CD2AP colocalizes with cortactin and F-actin in spots of 0.5-μm diameter. Confocal time-lapse microscopy in living podocytes expressing GFP-CD2AP or GFP-actin revealed that spots are motile, possess a limited lifetime, and are frequently associated with vesicles. A significant portion of spot-associated vesicles belongs to a later endosomal-sorting compartment, characterized by delayed uptake of fluorescent dextran (10 kDa) and by colocalization with Rab4, but not Rab5 and AP-2. Rapid accumulation of microinjected G-actin in spots and abrogation of spot motility by jasplakinolide demonstrate that spot movements depend on actin polymerization. Furthermore, a high turnover (half-time < 10 s) of CD2AP in spots was demonstrated by FRAP (fluorescence recovery after photobleaching). Our results demonstrate that CD2AP is associated with dynamic actin in a specific late endosomal compartment in podocytes, suggesting that CD2AP might be crucially involved in endosomal sorting and/or trafficking via regulation of actin assembly on vesicles.
glomerular epithelial cells; CMS; green fluorescent protein
CD2AP (CD2-associated protein)/CMS is an 80-kDa adaptor protein containing three SH3 domains in its NH2 terminus, a proline-rich region and a COOH-terminal coiled-coil domain with a putative -thymosin-like actin binding site (7). Because it has been discovered that CD2AP-deficient mice die at the age of 6 to 7 wk because of progressive glomerulosclerosis and massive proteinuria (31), many binding partners and biochemical features of CD2AP have been studied, including molecules involved in actin assembly and signaling pathways (11, 17, 18, 27, 41), proteins of the slit diaphragm complex in podocytes (30, 31), and interactions with the endocytic machinery of the epidermal growth factor (EGF) receptor and ferritin (4, 16, 21). However, the exact role of CD2AP in renal podocytes that prevents lethal glomerular disease still remains unclear.
In our previous work, we showed that CD2AP localizes to podocytes in mouse glomeruli where it is present in foot processes and podocyte cell bodies. Moreover, we used cultured podocytes to discover colocalization of CD2AP with F-actin spots that also stained positive for the actin-nucleating Arp2/3 complex and cortactin (41). Arp2/3 complex, which mediates initiation of new actin filaments and branches on preexisting filaments, is regulated by WASp/Scar proteins and cortactin, which binds and activates Arp2/3 complex (12, 22, 37, 39, 40, 42). Actin filament growth and turnover are essential for cell motility and intracellular motility of bacteria, viruses, and endogenous vesicles (25). In PtK1 cells that overexpress an activated form of the small GTPase ARF6, Schafer et al. (27) observed dynamic actin spots with comet tails, some of which were associated with vesicles. CD2AP localized to the head domain of these actin comet tails.
Recently, CD2AP was found to colocalize with the EGF receptor and cortactin in EGF-induced membrane ruffles in HeLa cells and to form a complex with receptor bound Cbl and endophilin, which is a regulatory component of clathrin-coated vesicles (21). Similarly, recent studies (10, 33) have demonstrated a role for the Cbl interacting protein of 85 kDa (CIN85) in a complex with endophilins and Cbl, in receptor-mediated endocytosis. In this complex, the CIN85 SH3 domains bind to receptor-associated Cbl, whereas a specific proline-rich motif in CIN85 binds endophilin. Because of significant homology in molecular architecture, sequence, and interacting partners (p130Cas, Grb2, Cbl, PI3-kinase), CD2AP and CIN85 represent a common adaptor molecule family (1, 3, 5, 9, 17, 35, 38). Compared with the molecular composition of CD2AP, however, CIN85 lacks an actin-binding domain (20). The regulatory impact of CD2AP is not restricted to the degradative pathway of the EGF receptor. CD2AP in a complex with Cbl was also shown to influence VEGF receptor (Flt-1) internalization (19). Also, haploinsufficiency of CD2AP results in impaired formation of prelysosomal multivesicular bodies (MVBs) in podocytes in response to intravenous ferritin injection compared with wild-type mice (16). Interestingly, CD2AP also binds the small GTPase Rab4, and by interaction with active Rab4 and Cbl controls morphology of EEA1-positve early endosomes. Coexpression of Rab4 and CD2AP in CHO cells resulted in enlarged vesicles that were positive for the late endosome marker Rab7 (4). However, the relationship between CD2AP and the endosomal compartment has not yet been explored in podocytes.
MATERIALS AND METHODS
Podocyte cell culture. Cultivation of conditionally immortalized mouse podocytes was performed as previously reported (29). In brief, podocytes were maintained in RPMI-1640 (Cell Concepts) supplemented with 10% FBS (GIBCO BRL), 100 U/ml penicillin, and 0.1 mg/ml streptomycin (Cell Concepts). To propagate podocytes, cells were first cultivated at 33°C and maintained for at least 1 wk at 38°C to induce differentiation.
Immunofluorescence, confocal laser-scanning microscopy, and microinjection. For immunofluorescence studies, podocytes were cultured on type IV collagen-coated glass coverslips (Biochrom, Berlin, Germany). At room temperature, cells were fixed (2% paraformaldehyde in PBS) for 10 min, permeabilized (0.3% Triton X-100 in PBS) for 8 min, and blocked in blocking solution (2% FBS, 2% BSA, 0.2% gelatine, PBS) for 45 min. Primary antibodies were incubated for 60 min. The following antibodies were used: rabbit anti-CD2AP and anti-Rab4 (Santa Cruz Biotechnology, Heidelberg, Germany), mouse anti-cortactin (4F11, Upstate Biotechnology, Waltham, MA), rabbit anti-AP-2 (kindly provided by Dr. E. Ungewickell, Hannover, Germany). Antigen-antibody complexes were visualized with Cy2- or Cy3-conjugated secondary antibodies (Dianova, Hamburg, Germany). For double-staining of CD2AP and cortactin, both secondary antibodies were from the same host (goat) and preadsorbed to avoid cross reactivity. F-actin was visualized using fluorochrome (Alexa)-conjugated phalloidins (Molecular Probes, Eugene, OR). Coverslips were washed (PBS), rinsed (H2O), and mounted with 15% Mowiol (Calbiochem), 50% glycerol, PBS. Specimens were viewed with a confocal laser-scanning microscope (TCS-SP, Leica Microsystems, Heidelberg, Germany).
Microinjection experiments were done with a Leica IRBE microscope equipped with an Eppendorf InjectMan Ni2 micromanipulator (Eppendorf, Hamburg, Germany). Alexa488-conjugated rabbit G-actin (3.6 μg/μl, Molecular Probes) was injected into cultured podocytes in RPMI-1640 without carbonate using the Eppendorf Femtojet. Injection pressure and injection time were 150 hPa and 0.4 s, respectively. After 1 to 8 min, cells were fixed and prepared for immunohistochemistry as described above.
Plasmids and transfection. Expression vectors coding for the following fluorescent fusion proteins were employed: green fluorescent protein (GFP)--actin (BD Biosciences, Heidelberg, Germany), GFP-CD2AP, CFP-Rab4A, and CFP-Rab5A. Cells were transfected using PerFectin reagent (Peqlab, Erlangen, Germany), according to the manufacturer’s instructions. Transiently transfected cells were taken for experiments 1–3 days after transfection.
Dextran incubation. Cultured podocytes on coverslips were incubated in RPMI-1640 containing 10% FBS and 1 mg/ml Alexa546-conjugated dextran (10 kDa, Molecular Probes) for 30 min, 2 h, or 20 h. Cells were then washed briefly with PBS and mounted in the chamber for live cell microscopy.
Live cell microscopy and photobleaching. A custom-built Plexiglas chamber was filled with 400 μl of observation medium (RPMI-1640 w/o phenol red supplemented with 10% FBS). Cell-containing coverslips coated with mouse collagen IV (0.1 mg/ml; BD Biosciences) were attached to the bottom of the chamber. The chamber was sealed and mounted on the stage of an inverted microscope (IRBE, Leica Microsystems). The temperature of the chamber and of the objective was kept at 37°C with the aid of an airstream incubator (ASI 400, Nevtek, Burnsville, VA). Immersion objectives (x63, numerical aperture 1.32) were used and images were obtained by confocal laser-scanning microscopy (TCS-SP, Leica Microsystems). Images (512 x 512 pixel) with a resolution of 12.9 pixel/μm were collected in 5- to 30-s intervals. The 488-nm laser line was used for GFP visualization as well as for bleaching. The bleaching pattern was generated by repetitive (8 times) scanning of a square (100 x 100 pixel). For FRAP analysis, images were collected every 4.8 s after bleaching. Applied laser intensity for bleaching was about 400-fold higher than the intensity used for imaging. Jasplakinolide (Molecular Probes) was injected into the chamber in a small volume of 3 μl to a final concentration of 1 μM. Live cell observations were completed within 1 h after attachment of the coverslip to the chamber.
Transmission electron microscopy. Cells were fixed with 2% paraformaldehyde for 30 min, permeabilized, blocked, and incubated with rabbit anti-CD2AP for 1 h. The speciman was rinsed with washing buffer (PBS containing 0.1% BSA) and exposed to goat anti-rabbit IgG coupled to 10-nm colloidal gold (Sigma) for 1 h at room temperature. Cells were then washed and postfixed with 2% glutaraldehyde and 0.5% tannic acid and counterstained with 2% OsO4 in PBS, treated with 2% uranyl acetate for 2–5 min and finally with 0.003% lead citrate in 2% polyvinyl alcohol (Sigma). The specimens were observed under a Phillips EM 301 electron microscope.
Image processing and data analysis. Image processing for FRAP analysis and video export was done with ImageJ (National Institutes of Health, Bethesda, MD). In FRAP analysis, fluorescence intensities over time were fitted to a monoexponential recovery function using SigmaPlot (SPSS, Chicago, IL). Results are presented as means ± SE if not indicated otherwise.
RESULTS
As we have reported recently (41), CD2AP localizes to characteristic spots of 0.5-μm size that strictly colocalize with F-actin (Fig. 1, a-c). Having shown that both cortactin which is known to be an activator of Arp2/3 complex as well as the Arp2/3 complex itself colocalize with F-actin spots in podocytes (41), we now demonstrate by confocal microscopy that cortactin directly colocalizes with CD2AP in spots. Besides a diffuse perinuclear distribution, CD2AP localized to characteristic cytoplasmic spots. Cortactin resembled the CD2AP spot-like pattern but was also found at cell margins (Fig. 1, d-f). High-power magnification revealed that CD2AP and cortactin exactly colocalized to the same spots (Fig. 1, g-i). Cross reactivity of secondary antibodies with the primary IgGs was unlikely because of the selective staining of cell margins by cortactin, and it was further excluded by control stainings (data not shown).
To evaluate the characteristics of these spots in living podocytes, we transfected cultured podocytes with expression vectors coding for GFP-actin and GFP-CD2AP. We next performed colocalization studies to demonstrate an identical localization of GFP-actin and GFP-CD2AP as the endogenously expressed proteins actin and CD2AP. GFP-actin colocalized with CD2AP and cortactin in spots (Fig. 2, a and b). Moreover, the stainings of endogenous F-actin with fluorochrome-coupled phalloidin and GFP-actin were identical (Fig. 2c). In parallel to those findings, GFP-CD2AP colocalized to cortactin and F-actin spots but not to stress fibers (Fig. 2, e and f) and its staining pattern showed a full overlap with the CD2AP antibody (Fig. 2d). Visualization of GFP-actin and GFP-CD2AP by confocal microscopy in living podocytes revealed that spots are highly motile performing nondirectional movements and that spots possess a limited lifetime that ranges from seconds to minutes (Fig. 3, a-d). Occasionally, dynamic spots formed comet tails (arrow in Fig. 3d).
In addition to spots that were not associated with any obvious intracellular structure, we observed many spots that seemed to be linked to vesicular structures being visible as dark circular areas (DCAs) because of absent cytosolic green fluorescence [Fig. 4; videos 1 and 2 (all supplemental data for this study are available at http://ajprenal.physiology.org/cgi/content/full/00178.2005/DC1)]. To confirm the vesicular nature of the observed DCAs, GFP-CD2AP-transfected podocytes were incubated with Alexa546-conjugated dextran (10 kDa) for different time periods, washed, and chased with confocal time lapse imaging for 10 to 30 min [Fig. 5, a and b; videos 3 and 4 (supplemental data)]. As indicated by z-scans of living cells, dextran-loaded DCAs were located clearly intracellularly with larger dextran-loaded DCAs extending almost through the entire cytoplasm (Fig. 5c). Moreover, the absence of persistent extracellular dextran on z-scan images excludes DCAs from being basal or apical invaginations of the podocyte cell membrane but account for a vesicular nature. The uptake of the endocytic fluid phase marker dextran in DCAs was time dependent. While 29 ± 9% (n = 6 cells) of DCAs were loaded by dextran after a 30-min incubation period, 76 ± 8% of DCAs were loaded with dextran after prolonged exposure for 20 h (n = 6 cells; Fig. 5d). Overall, 32 ± 4% (n = 6 cells) of the total number of CD2AP spots were clearly associated with dextran-loaded DCAs after 2-h incubation. Correlated with the increase in dextran-loaded DCAs, there was an increase in dextran-loaded DCAs associated with CD2AP spots after 2 h compared with incubation for 30 min. After 2-h incubation, 61 ± 9% (n = 6 cells) of dextran-loaded DCAs were associated with CD2AP spots (Fig. 5d). DCAs, however, do not account for all vesicular compartments in GFP-CD2AP expressing podocytes. The maximum of all dextran-filled vesicles that were visible as DCAs was calculated to 26 ± 5% (n = 6 cells) at 20 h. Thus the majority of dextran vesicles could not be associated with DCAs, which indicates that DCAs are a distinct subtype of the vesicular compartments. Conversely, not all CD2AP spots were associated with dextran-labeled vesicles, suggesting association with other (vesicular) structures. Interestingly, the intracellular localization and size of dextran-filled DCAs varied grossly (from 0.3- to 3-μm diameter). Larger DCAs tended to predominate at perinuclear sites while smaller DCAs more often occurred in the cell periphery (Fig. 5, a and b). Of note is that the fluid phase marker dextran is often inhomogenously distributed in DCAs, which proposes an internal membranous suborganization in some DCAs (Fig. 5b).
In conclusion, these findings first strongly support that DCAs are vesicular compartments in the endocytic, degradative pathway at later stages. Second, CD2AP spots are linked to vesicles in the endocytic pathway. This was also analyzed by immunoelectron microscopy of cultured podocytes, confirming the close association of CD2AP with vesicles in nontransfected cells (Fig. 6a, arrow). Furthermore, we noted that association of spots was also detectable by immunofluorescence in nontransfected podocytes (see Fig. 1b, inset). However, DCAs can be detected more easily in GFP-actin- or GFP-CD2AP-transfected cells because of the higher cytosolic background fluorescence. Using CFP-Rab4, which labels sorting endosomes, we found that nearly all Rab4-positive vesicles were decorated with cortactin spots (Fig. 6b). The spatial distribution of Rab4-positive vesicles was enhanced perinuclearly as it was for DCAs. A few cortactin spots were without Rab4-positive vesicle association in the cell periphery (Fig. 6b). Similar results were obtained with an anti-Rab4 antibody (data not shown). Cortactin spots were not colocalized with CFP-Rab5, a marker of early endosomes (Fig. 6c). In addition, GFP-CD2AP spots were not colocalized with AP-2, which labels clathrin-coated vesicles (Fig. 6d).
CD2AP spots often were aligned and moved circumferentially around vesicles [arrows in Fig. 7; videos 1–4 (supplemental data)]. Additionally, CD2AP spots accumulated as tail structures of moving vesicles [arrowheads in Fig. 7; video 1, 3, and 4 (supplemental data)]. This accumulation frequently led the way to enhanced directional movement of vesicles. The velocity of this intracellular "rocketing" motility of vesicles was calculated to 1.6 ± 0.6 μm/min (means ± SD, n = 4 cells). Similar observations were made with GFP-actin (data not shown).
To provide functional evidence that dynamic CD2AP spots are sites of active actin assembly, cultured podocytes were microinjected with Alexa488-coupled G-actin and stained for CD2AP after different time periods. One minute after microinjection, G-actin localized to the cell margin as well as to CD2AP-positve spots in podocytes (arrows in Fig. 8, a-c). Eight minutes after microinjection, the G-actin pattern grossly overlapped with all F-actin structures visualized by phalloidin (data not shown).
The rapid recruitment of G-actin to spots suggested that spots are sites of high actin turnover leading to the question whether actin assembly is responsible for the motility of CD2AP spots. Therefore, CD2AP spots were tracked by confocal time-lapse imaging before and after treatment with jasplakinolide (1 μM). Jasplakinolide is a membrane-permeable cyclic fungal peptide that stabilizes F-actin and lowers free actin monomer concentration (2). Dynamics of CD2AP spots were immediately abrogated after exposure to jasplakinolide [n = 3 experiments; Fig. 8, d-f; video 5 (supplemental data)], which strongly advocates that intracellular motility of CD2AP spots employs actin polymerization.
We finally investigated whether CD2AP itself features a high turnover and whether it is a stable or a dynamic component of spots by analyzing the fluorescence recovery after photobleaching (FRAP) of GFP-CD2AP (Fig. 9). Intensities were calculated for a 80 x 80-pixel square, averaging spots and diffuse cytoplasmic CD2AP fluorescence (method A), as well as for single spots (method B). FRAP followed monoexponential functions. Curve-fitting of FRAP revealed 81 ± 4% (A) and 81 ± 8% (B) fluorescence recovery (mobile fraction) with a half-time of 9 ± 2 s (n = 6 experiments, A) and 7 ± 3 s (n = 4 experiments, B), respectively. Thus CD2AP is rapidly turned over in spots and elsewhere.
DISCUSSION
In the present study, we demonstrate that CD2AP spots in podocytes are distinct structures that are composed of cortactin and F-actin. In our previous work, we showed that F-actin spots colocalized with either CD2AP, cortactin or Arp2/3 complex in podocytes (41). Although it is known that cortactin colocalizes to, binds, and activates Arp2/3 complex with an NH2-terminal acidic domain (37), according to our earlier data we could not rule out that CD2AP and cortactin localize to different F-actin spots. By double staining of cortactin and CD2AP, we now demonstrate that both proteins colocalize to identical spots. As we did not observe spots that selectively stained for either CD2AP or cortactin, both molecules appear to be essential components of spots and can be used as marker proteins. Colocalization of CD2AP and cortactin in spots may be the result of a direct interaction, as a binding site for cortactin has been recently mapped to the second proline-rich region of CD2AP (21).
Using GFP fusion proteins for actin and CD2AP, which labeled the same spots as the endogenous proteins, we observed that 50% of spots were associated with vesicles visible as dark circular areas (DCAs). Association of spots with vesicles was also confirmed in nontransfected podocytes by immunofluorescence and immunoelectron microscopy. Whether the other half of CD2AP spots is associated with vesicles that are too small to be visible as DCAs or whether some CD2AP spots are not linked to vesicles at all remains an open question. Using GFP-Arp3 and GFP-capping protein, Schafer and co-workers (28) observed similar spots containing F-actin and CD2AP that mainly formed finlike projections on the cell surface in the absence of membrane vesicles in the vicinity. We characterized DCAs as vesicles of a late endosomal compartment, because the fluid phase marker dextran was trapped in less than one-third of DCAs after 30 min. Dextran-loaded vesicles that were associated with CD2AP spots frequently exhibited an inhomogeneous luminal distribution of dextran, being consistent with a membranous suborganization in the vesicle interior. Budding and internalized membrane is the hallmark of multivesicular bodies (MVBs), which are a distinct late endosomal compartment (26). The fact that CD2AP/cortactin spots did not colocalize with AP-2 and Rab5 in podocytes provides further evidence that CD2AP spots do not associate with early endosomes. Moreover, colocalization of cortactin spots in podocytes with Rab4-positive vesicles places these vesicles at a later stage in the endosomal pathway. Colocalization of spots with Rab4 is consistent with the work of Cormont et al. (4), who discovered CD2AP to be a binding partner of the GTP-bound form of Rab4. Rab4 has been implicated in endosomal membrane sorting into the recycling and degradative pathway (23, 32). Based on several findings, Cormont et al. (4) speculated that Rab4, besides other functions, controls sorting from early to late endosomes. First, morphologically changed endosomes after Rab4 and CD2AP coexpression included the late endosome marker Rab7. Second, horseradish peroxidase uptake was altered by CD2AP overexpression only beyond 15 min, and, third, CD2AP was not involved in transferrin receptor recycling which is an early recycling pathway.
It is noteworthy that a large fraction of CD2AP spots is associated with vesicles in podocytes. In contrast, spots were rarely associated with vesicles in PtK1 fibroblasts (28). Furthermore, overexpression of a constitutive active form of the small GTPase ARF6 in PtK1 fibroblasts and CHO cells was needed to induce vesicle-associated actin spots and comet tails, the head of which stained for CD2AP (27). In addition, association of CD2AP with vesicular structures was only detectable in COS-7 cells after transfection with CD2AP (17) and in CHO cells after cotransfection with CD2AP and Cbl or Rab4 (4). In contrast, we detected association of CD2AP with vesicles in nontransfected podocytes. These findings may indicate that podocytes possess specifically enlarged endosomal compartment(s).
Our findings are consistent with the concept that CD2AP in spots around vesicles could be required for formation of MVBs and for protein sorting into the degradative pathway. According to this concept, Kim et al. (16) observed a twofold increase in MVBs in podocytes of wild-type mice after intravenous injection of ferritin, which failed to stimulate MVB formation in CD2AP+/– mice. Simultaneous overexpression of CD2AP with Cbl or Rab4 induced enlargement of EEA1-positive early endosomes in CHO cells (4), suggesting that CD2AP is involved in a sorting or transport step distal to early endosomes. MVB formation is required for downregulation of activated signaling receptors (15). Accordingly, CD2AP has been shown to be involved in EGF receptor trafficking (21), in PDGF receptor (4), and in VEGF receptor (Flt-1) degradation (19).
Our confocal time-lapse microscopy of GFP-actin and GFP-CD2AP in podocytes revealed the dynamic nature of spots in podocytes and a rather short but variable lifetime. Spot motility is nondirectional and does not seem to be linked to intracellular filaments. Spots are characterized by rapid accumulation of microinjected G-actin, demonstrating a high rate of actin polymerization in actin spots. Actin polymerization drives spot motility, as demonstrated by inhibition of motility with jasplakinolide. The high turnover rate of CD2AP in spots would be consistent with an involvement of CD2AP in regulating actin polymerization. In fact, it has been shown that CD2AP interacts with F-actin (20) and with several proteins known to regulate actin filament assembly: cortactin (21), capping protein (13), and the p85 regulatory subunit of PI3-kinase (11).
The actin cytoskeleton is importantly involved in various pathways of vesicular trafficking (34). Recently, it has become evident that actin polymerization participates in internalizing clathrin-coated pits (24). However, CD2AP spots were not colocalized with AP-2, an adaptor protein in clathrin-dependent endocytosis, in our experiments. On the other hand, the role of the actin cytoskeleton in later endosomal compartments, to which we localized CD2AP spots in podocytes, is not well defined. GFP-CD2AP as well as GFP-actin spots on vesicles exhibited either a circumferential movement of one or multiple spots around the vesicle or were concentrated at one point on the circumference generating a propulsive force for vesicular movement. The latter kind of movement has been reconstituted with vesicles of HeLa cells in Xenopus laevis egg extract (36). Interestingly, 80% of vesicles with actin tails were MVBs. Propulsive movement of vesicles has also been demonstrated for cortactin spots by Kaksonen et al. (14). Propulsive movements may enhance the chance of endosomes to reach targets within shorter times. On the other hand, we can only speculate about the function of the circumferential movement of vesicle-associated spots. By targeting cofilin to late endosomes in Dictyostelium, Drengk et al. (6) showed that loss of the late endosomal actin coat leads to aggregation of late endosomal vacuoles. Thus F-actin could function as a docking barrier. For docked vesicles, in contrast, actin polymerization has been suggested to forcibly promote fusion (8). Finally, actin could be involved in the inward budding of endosomal membranes and in pinching off, processes that are poorly understood (26).
GRANTS
This study was supported by grants of the German Research Foundation (DFG, En 329/7–3).
ACKNOWLEDGMENTS
We thank C. Kocksch for skilled technical assistance, H. Hosser for electron microscopy, T. Berger for continuous technical support regarding in vivo microscopy set-up, and R. Nonnenmacher for excellent graphical work.
FOOTNOTES
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.
REFERENCES
Borinstein SC, Hyatt MA, Sykes VW, Straub RE, Lipkowitz S, Boulter J, and Bogler O. SETA is a multifunctional adapter protein with three SH3 domains that binds Grb2, Cbl, and the novel SB1 proteins. Cell Signal 12: 769–779, 2000.
Bubb MR, Spector I, Beyer BB, and Fosen KM. Effects of jasplakinolide on the kinetics of actin polymerization. An explanation for certain in vivo observations. J Biol Chem 275: 5163–5170, 2000.
Chen B, Borinstein SC, Gillis J, Sykes VW, and Bogler O. The glioma-associated protein SETA interacts with AIP1/Alix and ALG-2 and modulates apoptosis in astrocytes. J Biol Chem 275: 19275–19281, 2000.
Cormont M, Meton I, Mari M, Monzo P, Keslair F, Gaskin C, McGraw TE, and Le Marchand-Brustel Y. CD2AP/CMS regulates endosome morphology and traffic to the degradative pathway through its interaction with Rab4 and c-Cbl. Traffic 4: 97–112, 2003.
Dikic I. CIN85/CMS family of adaptor molecules. FEBS Lett 529: 110–115, 2002.
Drengk A, Fritsch J, Schmauch C, Ruhling H, and Maniak M. A coat of filamentous actin prevents clustering of late-endosomal vacuoles in vivo. Curr Biol 13: 1814–1819, 2003.
Dustin ML, Olszowy MW, Holdorf AD, Li J, Bromley S, Desai N, Widder P, Rosenberger F, van der Merwe PA, Allen PM, and Shaw AS. A novel adaptor protein orchestrates receptor patterning and cytoskeletal polarity in T-cell contacts. Cell 94: 667–677, 1998.
Eitzen G. Actin remodeling to facilitate membrane fusion. Biochim Biophys Acta 1641: 175–181, 2003.
Gout I, Middleton G, Adu J, Ninkina NN, Drobot LB, Filonenko V, Matsuka G, Davies AM, Waterfield M, and Buchman VL. Negative regulation of PI 3-kinase by Ruk, a novel adaptor protein. EMBO J 19: 4015–4025, 2000.
Haglund K, Shimokawa N, Szymkiewicz I, and Dikic I. Cbl-directed monoubiquitination of CIN85 is involved in regulation of ligand-induced degradation of EGF receptors. Proc Natl Acad Sci USA 99: 12191–12196, 2002.
Huber TB, Hartleben B, Kim J, Schmidts M, Schermer B, Keil A, Egger L, Lecha RL, Borner C, Pavenstadt H, Shaw AS, Walz G, and Benzing T. Nephrin and CD2AP associate with phosphoinositide 3-OH kinase and stimulate AKT-dependent signaling. Mol Cell Biol 23: 4917–4928, 2003.
Hufner K, Higgs HN, Pollard TD, Jacobi C, Aepfelbacher M, and Linder S. The verprolin-like central (vc) region of Wiskott-Aldrich syndrome protein induces Arp2/3 complex-dependent actin nucleation. J Biol Chem 276: 35761–35767, 2001.
Hutchings NJ, Clarkson N, Chalkley R, Barclay AN, and Brown MH. Linking the T cell surface protein CD2 to the actin-capping protein CAPZ via CMS and CIN85. J Biol Chem 278: 22396–22403, 2003.
Kaksonen M, Peng HB, and Rauvala H. Association of cortactin with dynamic actin in lamellipodia and on endosomal vesicles. J Cell Sci 113: 4421–4426, 2000.
Katzmann DJ, Odorizzi G, and Emr SD. Receptor downregulation and multivesicular-body sorting. Nat Rev Mol Cell Biol 3: 893–905, 2002.
Kim JM, Wu H, Green G, Winkler CA, Kopp JB, Miner JH, Unanue ER, and Shaw AS. CD2-associated protein haploinsufficiency is linked to glomerular disease susceptibility. Science 300: 1298–1300, 2003.
Kirsch KH, Georgescu MM, Ishimaru S, and Hanafusa H. CMS: an adapter molecule involved in cytoskeletal rearrangements. Proc Natl Acad Sci USA 96: 6211–6216, 1999.
Kirsch KH, Georgescu MM, Shishido T, Langdon WY, Birge RB, and Hanafusa H. The adapter type protein CMS/CD2AP binds to the proto-oncogenic protein c-Cbl through a tyrosine phosphorylation-regulated Src homology 3 domain interaction. J Biol Chem 276: 4957–4963, 2001.
Kobayashi S, Sawano A, Nojima Y, Shibuya M, and Maru Y. The c-Cbl/CD2AP complex regulates VEGF-induced endocytosis and degradation of Flt-1 (VEGFR-1). FASEB J 18: 929–931, 2004.
Lehtonen S, Zhao F, and Lehtonen E. CD2-associated protein directly interacts with the actin cytoskeleton. Am J Physiol Renal Physiol 283: F734–F743, 2002.
Lynch DK, Winata SC, Lyons RJ, Hughes WE, Lehrbach GM, Wasinger V, Corthals G, Cordwell S, and Daly RJ. A Cortactin-CD2-associated protein (CD2AP) complex provides a novel link between epidermal growth factor receptor endocytosis and the actin cytoskeleton. J Biol Chem 278: 21805–21813, 2003.
Machesky LM and Insall RH. Scar1 and the related Wiskott-Aldrich syndrome protein, WASP, regulate the actin cytoskeleton through the Arp2/3 complex. Curr Biol 8: 1347–1356, 1998.
McCaffrey MW, Bielli A, Cantalupo G, Mora S, Roberti V, Santillo M, Drummond F, and Bucci C. Rab4 affects both recycling and degradative endosomal trafficking. FEBS Lett 495: 21–30, 2001.
Merrifield CJ. Seeing is believing: imaging actin dynamics at single sites of endocytosis. Trends Cell Biol 14: 352–358, 2004.
Pollard TD and Borisy GG. Cellular motility driven by assembly and disassembly of actin filaments. Cell 112: 453–465, 2003.
Raiborg C, Rusten TE, and Stenmark H. Protein sorting into multivesicular endosomes. Curr Opin Cell Biol 15: 446–455, 2003.
Schafer DA, D’Souza-Schorey C, and Cooper JA. Actin assembly at membranes controlled by ARF6. Traffic 1: 892–903, 2000.
Schafer DA, Welch MD, Machesky LM, Bridgman PC, Meyer SM, and Cooper JA. Visualization and molecular analysis of actin assembly in living cells. J Cell Biol 143: 1919–1930, 1998.
Schiwek D, Endlich N, Holzman L, Holthofer H, Kriz W, and Endlich K. Stable expression of nephrin and localization to cell-cell contacts in novel murine podocyte cell lines. Kidney Int 66: 91–101, 2004.
Schwarz K, Simons M, Reiser J, Saleem MA, Faul C, Kriz W, Shaw AS, Holzman LB, and Mundel P. Podocin, a raft-associated component of the glomerular slit diaphragm, interacts with CD2AP and nephrin. J Clin Invest 108: 1621–1629, 2001.
Shih NY, Li J, Karpitskii V, Nguyen A, Dustin ML, Kanagawa O, Miner JH, and Shaw AS. Congenital nephrotic syndrome in mice lacking CD2-associated protein. Science 286: 312–315, 1999.
Somsel Rodman J and Wandinger-Ness A. Rab GTPases coordinate endocytosis. J Cell Sci 113: 183–192, 2000.
Soubeyran P, Kowanetz K, Szymkiewicz I, Langdon WY, and Dikic I. Cbl-CIN85-endophilin complex mediates ligand-induced downregulation of EGF receptors. Nature 416: 183–187, 2002.
Stamnes M. Regulating the actin cytoskeleton during vesicular transport. Curr Opin Cell Biol 14: 428–433, 2002.
Take H, Watanabe S, Takeda K, Yu ZX, Iwata N, and Kajigaya S. Cloning and characterization of a novel adaptor protein, CIN85, that interacts with c-Cbl. Biochem Biophys Res Commun 268: 321–328, 2000.
Taunton J, Rowning BA, Coughlin ML, Wu M, Moon RT, Mitchison TJ, and Larabell CA. Actin-dependent propulsion of endosomes and lysosomes by recruitment of N-WASP. J Cell Biol 148: 519–530, 2000.
Uruno T, Liu J, Zhang P, Fan Y, Egile C, Li R, Mueller SC, and Zhan X. Activation of Arp2/3 complex-mediated actin polymerization by cortactin. Nat Cell Biol 3: 259–266, 2001.
Verdier F, Valovka T, Zhyvoloup A, Drobot LB, Buchman V, Waterfield M, and Gout I. Ruk is ubiquitinated but not degraded by the proteasome. Eur J Biochem 269: 3402–3408, 2002.
Weaver AM, Karginov AV, Kinley AW, Weed SA, Li Y, Parsons JT, and Cooper JA. Cortactin promotes and stabilizes Arp2/3-induced actin filament network formation. Curr Biol 11: 370–374, 2001.
Weed SA, Karginov AV, Schafer DA, Weaver AM, Kinley AW, Cooper JA, and Parsons JT. Cortactin localization to sites of actin assembly in lamellipodia requires interactions with F-actin and the Arp2/3 complex. J Cell Biol 151: 29–40, 2000.
Welsch T, Endlich N, Kriz W, and Endlich K. CD2AP and p130Cas localize to different F-actin structures in podocytes. Am J Physiol Renal Physiol 281: F769–F777, 2001.
Yarar D, To W, Abo A, and Welch MD. The Wiskott-Aldrich syndrome protein directs actin-based motility by stimulating actin nucleation with the Arp2/3 complex. Curr Biol 9: 555–558, 1999.(Thilo Welsch, Nicole Endl)
Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri
ABSTRACT
The docking protein CD2AP (CD2-associated protein) serves a nonredundant function in podocytes as CD2AP knockout mice die of renal failure at the age of 6–7 wk. Furthermore, haploinsufficiency due to mutation of the CD2AP gene is associated with focal segmental glomerulosclerosis in humans. Although CD2AP has been shown to interact with proteins regulating actin polymerization, with proteins of the slit diaphragm, and with the endocytic machinery, its critical function in podocytes remains unclear. In conditionally immortalized mouse podocytes, we demonstrate that CD2AP colocalizes with cortactin and F-actin in spots of 0.5-μm diameter. Confocal time-lapse microscopy in living podocytes expressing GFP-CD2AP or GFP-actin revealed that spots are motile, possess a limited lifetime, and are frequently associated with vesicles. A significant portion of spot-associated vesicles belongs to a later endosomal-sorting compartment, characterized by delayed uptake of fluorescent dextran (10 kDa) and by colocalization with Rab4, but not Rab5 and AP-2. Rapid accumulation of microinjected G-actin in spots and abrogation of spot motility by jasplakinolide demonstrate that spot movements depend on actin polymerization. Furthermore, a high turnover (half-time < 10 s) of CD2AP in spots was demonstrated by FRAP (fluorescence recovery after photobleaching). Our results demonstrate that CD2AP is associated with dynamic actin in a specific late endosomal compartment in podocytes, suggesting that CD2AP might be crucially involved in endosomal sorting and/or trafficking via regulation of actin assembly on vesicles.
glomerular epithelial cells; CMS; green fluorescent protein
CD2AP (CD2-associated protein)/CMS is an 80-kDa adaptor protein containing three SH3 domains in its NH2 terminus, a proline-rich region and a COOH-terminal coiled-coil domain with a putative -thymosin-like actin binding site (7). Because it has been discovered that CD2AP-deficient mice die at the age of 6 to 7 wk because of progressive glomerulosclerosis and massive proteinuria (31), many binding partners and biochemical features of CD2AP have been studied, including molecules involved in actin assembly and signaling pathways (11, 17, 18, 27, 41), proteins of the slit diaphragm complex in podocytes (30, 31), and interactions with the endocytic machinery of the epidermal growth factor (EGF) receptor and ferritin (4, 16, 21). However, the exact role of CD2AP in renal podocytes that prevents lethal glomerular disease still remains unclear.
In our previous work, we showed that CD2AP localizes to podocytes in mouse glomeruli where it is present in foot processes and podocyte cell bodies. Moreover, we used cultured podocytes to discover colocalization of CD2AP with F-actin spots that also stained positive for the actin-nucleating Arp2/3 complex and cortactin (41). Arp2/3 complex, which mediates initiation of new actin filaments and branches on preexisting filaments, is regulated by WASp/Scar proteins and cortactin, which binds and activates Arp2/3 complex (12, 22, 37, 39, 40, 42). Actin filament growth and turnover are essential for cell motility and intracellular motility of bacteria, viruses, and endogenous vesicles (25). In PtK1 cells that overexpress an activated form of the small GTPase ARF6, Schafer et al. (27) observed dynamic actin spots with comet tails, some of which were associated with vesicles. CD2AP localized to the head domain of these actin comet tails.
Recently, CD2AP was found to colocalize with the EGF receptor and cortactin in EGF-induced membrane ruffles in HeLa cells and to form a complex with receptor bound Cbl and endophilin, which is a regulatory component of clathrin-coated vesicles (21). Similarly, recent studies (10, 33) have demonstrated a role for the Cbl interacting protein of 85 kDa (CIN85) in a complex with endophilins and Cbl, in receptor-mediated endocytosis. In this complex, the CIN85 SH3 domains bind to receptor-associated Cbl, whereas a specific proline-rich motif in CIN85 binds endophilin. Because of significant homology in molecular architecture, sequence, and interacting partners (p130Cas, Grb2, Cbl, PI3-kinase), CD2AP and CIN85 represent a common adaptor molecule family (1, 3, 5, 9, 17, 35, 38). Compared with the molecular composition of CD2AP, however, CIN85 lacks an actin-binding domain (20). The regulatory impact of CD2AP is not restricted to the degradative pathway of the EGF receptor. CD2AP in a complex with Cbl was also shown to influence VEGF receptor (Flt-1) internalization (19). Also, haploinsufficiency of CD2AP results in impaired formation of prelysosomal multivesicular bodies (MVBs) in podocytes in response to intravenous ferritin injection compared with wild-type mice (16). Interestingly, CD2AP also binds the small GTPase Rab4, and by interaction with active Rab4 and Cbl controls morphology of EEA1-positve early endosomes. Coexpression of Rab4 and CD2AP in CHO cells resulted in enlarged vesicles that were positive for the late endosome marker Rab7 (4). However, the relationship between CD2AP and the endosomal compartment has not yet been explored in podocytes.
MATERIALS AND METHODS
Podocyte cell culture. Cultivation of conditionally immortalized mouse podocytes was performed as previously reported (29). In brief, podocytes were maintained in RPMI-1640 (Cell Concepts) supplemented with 10% FBS (GIBCO BRL), 100 U/ml penicillin, and 0.1 mg/ml streptomycin (Cell Concepts). To propagate podocytes, cells were first cultivated at 33°C and maintained for at least 1 wk at 38°C to induce differentiation.
Immunofluorescence, confocal laser-scanning microscopy, and microinjection. For immunofluorescence studies, podocytes were cultured on type IV collagen-coated glass coverslips (Biochrom, Berlin, Germany). At room temperature, cells were fixed (2% paraformaldehyde in PBS) for 10 min, permeabilized (0.3% Triton X-100 in PBS) for 8 min, and blocked in blocking solution (2% FBS, 2% BSA, 0.2% gelatine, PBS) for 45 min. Primary antibodies were incubated for 60 min. The following antibodies were used: rabbit anti-CD2AP and anti-Rab4 (Santa Cruz Biotechnology, Heidelberg, Germany), mouse anti-cortactin (4F11, Upstate Biotechnology, Waltham, MA), rabbit anti-AP-2 (kindly provided by Dr. E. Ungewickell, Hannover, Germany). Antigen-antibody complexes were visualized with Cy2- or Cy3-conjugated secondary antibodies (Dianova, Hamburg, Germany). For double-staining of CD2AP and cortactin, both secondary antibodies were from the same host (goat) and preadsorbed to avoid cross reactivity. F-actin was visualized using fluorochrome (Alexa)-conjugated phalloidins (Molecular Probes, Eugene, OR). Coverslips were washed (PBS), rinsed (H2O), and mounted with 15% Mowiol (Calbiochem), 50% glycerol, PBS. Specimens were viewed with a confocal laser-scanning microscope (TCS-SP, Leica Microsystems, Heidelberg, Germany).
Microinjection experiments were done with a Leica IRBE microscope equipped with an Eppendorf InjectMan Ni2 micromanipulator (Eppendorf, Hamburg, Germany). Alexa488-conjugated rabbit G-actin (3.6 μg/μl, Molecular Probes) was injected into cultured podocytes in RPMI-1640 without carbonate using the Eppendorf Femtojet. Injection pressure and injection time were 150 hPa and 0.4 s, respectively. After 1 to 8 min, cells were fixed and prepared for immunohistochemistry as described above.
Plasmids and transfection. Expression vectors coding for the following fluorescent fusion proteins were employed: green fluorescent protein (GFP)--actin (BD Biosciences, Heidelberg, Germany), GFP-CD2AP, CFP-Rab4A, and CFP-Rab5A. Cells were transfected using PerFectin reagent (Peqlab, Erlangen, Germany), according to the manufacturer’s instructions. Transiently transfected cells were taken for experiments 1–3 days after transfection.
Dextran incubation. Cultured podocytes on coverslips were incubated in RPMI-1640 containing 10% FBS and 1 mg/ml Alexa546-conjugated dextran (10 kDa, Molecular Probes) for 30 min, 2 h, or 20 h. Cells were then washed briefly with PBS and mounted in the chamber for live cell microscopy.
Live cell microscopy and photobleaching. A custom-built Plexiglas chamber was filled with 400 μl of observation medium (RPMI-1640 w/o phenol red supplemented with 10% FBS). Cell-containing coverslips coated with mouse collagen IV (0.1 mg/ml; BD Biosciences) were attached to the bottom of the chamber. The chamber was sealed and mounted on the stage of an inverted microscope (IRBE, Leica Microsystems). The temperature of the chamber and of the objective was kept at 37°C with the aid of an airstream incubator (ASI 400, Nevtek, Burnsville, VA). Immersion objectives (x63, numerical aperture 1.32) were used and images were obtained by confocal laser-scanning microscopy (TCS-SP, Leica Microsystems). Images (512 x 512 pixel) with a resolution of 12.9 pixel/μm were collected in 5- to 30-s intervals. The 488-nm laser line was used for GFP visualization as well as for bleaching. The bleaching pattern was generated by repetitive (8 times) scanning of a square (100 x 100 pixel). For FRAP analysis, images were collected every 4.8 s after bleaching. Applied laser intensity for bleaching was about 400-fold higher than the intensity used for imaging. Jasplakinolide (Molecular Probes) was injected into the chamber in a small volume of 3 μl to a final concentration of 1 μM. Live cell observations were completed within 1 h after attachment of the coverslip to the chamber.
Transmission electron microscopy. Cells were fixed with 2% paraformaldehyde for 30 min, permeabilized, blocked, and incubated with rabbit anti-CD2AP for 1 h. The speciman was rinsed with washing buffer (PBS containing 0.1% BSA) and exposed to goat anti-rabbit IgG coupled to 10-nm colloidal gold (Sigma) for 1 h at room temperature. Cells were then washed and postfixed with 2% glutaraldehyde and 0.5% tannic acid and counterstained with 2% OsO4 in PBS, treated with 2% uranyl acetate for 2–5 min and finally with 0.003% lead citrate in 2% polyvinyl alcohol (Sigma). The specimens were observed under a Phillips EM 301 electron microscope.
Image processing and data analysis. Image processing for FRAP analysis and video export was done with ImageJ (National Institutes of Health, Bethesda, MD). In FRAP analysis, fluorescence intensities over time were fitted to a monoexponential recovery function using SigmaPlot (SPSS, Chicago, IL). Results are presented as means ± SE if not indicated otherwise.
RESULTS
As we have reported recently (41), CD2AP localizes to characteristic spots of 0.5-μm size that strictly colocalize with F-actin (Fig. 1, a-c). Having shown that both cortactin which is known to be an activator of Arp2/3 complex as well as the Arp2/3 complex itself colocalize with F-actin spots in podocytes (41), we now demonstrate by confocal microscopy that cortactin directly colocalizes with CD2AP in spots. Besides a diffuse perinuclear distribution, CD2AP localized to characteristic cytoplasmic spots. Cortactin resembled the CD2AP spot-like pattern but was also found at cell margins (Fig. 1, d-f). High-power magnification revealed that CD2AP and cortactin exactly colocalized to the same spots (Fig. 1, g-i). Cross reactivity of secondary antibodies with the primary IgGs was unlikely because of the selective staining of cell margins by cortactin, and it was further excluded by control stainings (data not shown).
To evaluate the characteristics of these spots in living podocytes, we transfected cultured podocytes with expression vectors coding for GFP-actin and GFP-CD2AP. We next performed colocalization studies to demonstrate an identical localization of GFP-actin and GFP-CD2AP as the endogenously expressed proteins actin and CD2AP. GFP-actin colocalized with CD2AP and cortactin in spots (Fig. 2, a and b). Moreover, the stainings of endogenous F-actin with fluorochrome-coupled phalloidin and GFP-actin were identical (Fig. 2c). In parallel to those findings, GFP-CD2AP colocalized to cortactin and F-actin spots but not to stress fibers (Fig. 2, e and f) and its staining pattern showed a full overlap with the CD2AP antibody (Fig. 2d). Visualization of GFP-actin and GFP-CD2AP by confocal microscopy in living podocytes revealed that spots are highly motile performing nondirectional movements and that spots possess a limited lifetime that ranges from seconds to minutes (Fig. 3, a-d). Occasionally, dynamic spots formed comet tails (arrow in Fig. 3d).
In addition to spots that were not associated with any obvious intracellular structure, we observed many spots that seemed to be linked to vesicular structures being visible as dark circular areas (DCAs) because of absent cytosolic green fluorescence [Fig. 4; videos 1 and 2 (all supplemental data for this study are available at http://ajprenal.physiology.org/cgi/content/full/00178.2005/DC1)]. To confirm the vesicular nature of the observed DCAs, GFP-CD2AP-transfected podocytes were incubated with Alexa546-conjugated dextran (10 kDa) for different time periods, washed, and chased with confocal time lapse imaging for 10 to 30 min [Fig. 5, a and b; videos 3 and 4 (supplemental data)]. As indicated by z-scans of living cells, dextran-loaded DCAs were located clearly intracellularly with larger dextran-loaded DCAs extending almost through the entire cytoplasm (Fig. 5c). Moreover, the absence of persistent extracellular dextran on z-scan images excludes DCAs from being basal or apical invaginations of the podocyte cell membrane but account for a vesicular nature. The uptake of the endocytic fluid phase marker dextran in DCAs was time dependent. While 29 ± 9% (n = 6 cells) of DCAs were loaded by dextran after a 30-min incubation period, 76 ± 8% of DCAs were loaded with dextran after prolonged exposure for 20 h (n = 6 cells; Fig. 5d). Overall, 32 ± 4% (n = 6 cells) of the total number of CD2AP spots were clearly associated with dextran-loaded DCAs after 2-h incubation. Correlated with the increase in dextran-loaded DCAs, there was an increase in dextran-loaded DCAs associated with CD2AP spots after 2 h compared with incubation for 30 min. After 2-h incubation, 61 ± 9% (n = 6 cells) of dextran-loaded DCAs were associated with CD2AP spots (Fig. 5d). DCAs, however, do not account for all vesicular compartments in GFP-CD2AP expressing podocytes. The maximum of all dextran-filled vesicles that were visible as DCAs was calculated to 26 ± 5% (n = 6 cells) at 20 h. Thus the majority of dextran vesicles could not be associated with DCAs, which indicates that DCAs are a distinct subtype of the vesicular compartments. Conversely, not all CD2AP spots were associated with dextran-labeled vesicles, suggesting association with other (vesicular) structures. Interestingly, the intracellular localization and size of dextran-filled DCAs varied grossly (from 0.3- to 3-μm diameter). Larger DCAs tended to predominate at perinuclear sites while smaller DCAs more often occurred in the cell periphery (Fig. 5, a and b). Of note is that the fluid phase marker dextran is often inhomogenously distributed in DCAs, which proposes an internal membranous suborganization in some DCAs (Fig. 5b).
In conclusion, these findings first strongly support that DCAs are vesicular compartments in the endocytic, degradative pathway at later stages. Second, CD2AP spots are linked to vesicles in the endocytic pathway. This was also analyzed by immunoelectron microscopy of cultured podocytes, confirming the close association of CD2AP with vesicles in nontransfected cells (Fig. 6a, arrow). Furthermore, we noted that association of spots was also detectable by immunofluorescence in nontransfected podocytes (see Fig. 1b, inset). However, DCAs can be detected more easily in GFP-actin- or GFP-CD2AP-transfected cells because of the higher cytosolic background fluorescence. Using CFP-Rab4, which labels sorting endosomes, we found that nearly all Rab4-positive vesicles were decorated with cortactin spots (Fig. 6b). The spatial distribution of Rab4-positive vesicles was enhanced perinuclearly as it was for DCAs. A few cortactin spots were without Rab4-positive vesicle association in the cell periphery (Fig. 6b). Similar results were obtained with an anti-Rab4 antibody (data not shown). Cortactin spots were not colocalized with CFP-Rab5, a marker of early endosomes (Fig. 6c). In addition, GFP-CD2AP spots were not colocalized with AP-2, which labels clathrin-coated vesicles (Fig. 6d).
CD2AP spots often were aligned and moved circumferentially around vesicles [arrows in Fig. 7; videos 1–4 (supplemental data)]. Additionally, CD2AP spots accumulated as tail structures of moving vesicles [arrowheads in Fig. 7; video 1, 3, and 4 (supplemental data)]. This accumulation frequently led the way to enhanced directional movement of vesicles. The velocity of this intracellular "rocketing" motility of vesicles was calculated to 1.6 ± 0.6 μm/min (means ± SD, n = 4 cells). Similar observations were made with GFP-actin (data not shown).
To provide functional evidence that dynamic CD2AP spots are sites of active actin assembly, cultured podocytes were microinjected with Alexa488-coupled G-actin and stained for CD2AP after different time periods. One minute after microinjection, G-actin localized to the cell margin as well as to CD2AP-positve spots in podocytes (arrows in Fig. 8, a-c). Eight minutes after microinjection, the G-actin pattern grossly overlapped with all F-actin structures visualized by phalloidin (data not shown).
The rapid recruitment of G-actin to spots suggested that spots are sites of high actin turnover leading to the question whether actin assembly is responsible for the motility of CD2AP spots. Therefore, CD2AP spots were tracked by confocal time-lapse imaging before and after treatment with jasplakinolide (1 μM). Jasplakinolide is a membrane-permeable cyclic fungal peptide that stabilizes F-actin and lowers free actin monomer concentration (2). Dynamics of CD2AP spots were immediately abrogated after exposure to jasplakinolide [n = 3 experiments; Fig. 8, d-f; video 5 (supplemental data)], which strongly advocates that intracellular motility of CD2AP spots employs actin polymerization.
We finally investigated whether CD2AP itself features a high turnover and whether it is a stable or a dynamic component of spots by analyzing the fluorescence recovery after photobleaching (FRAP) of GFP-CD2AP (Fig. 9). Intensities were calculated for a 80 x 80-pixel square, averaging spots and diffuse cytoplasmic CD2AP fluorescence (method A), as well as for single spots (method B). FRAP followed monoexponential functions. Curve-fitting of FRAP revealed 81 ± 4% (A) and 81 ± 8% (B) fluorescence recovery (mobile fraction) with a half-time of 9 ± 2 s (n = 6 experiments, A) and 7 ± 3 s (n = 4 experiments, B), respectively. Thus CD2AP is rapidly turned over in spots and elsewhere.
DISCUSSION
In the present study, we demonstrate that CD2AP spots in podocytes are distinct structures that are composed of cortactin and F-actin. In our previous work, we showed that F-actin spots colocalized with either CD2AP, cortactin or Arp2/3 complex in podocytes (41). Although it is known that cortactin colocalizes to, binds, and activates Arp2/3 complex with an NH2-terminal acidic domain (37), according to our earlier data we could not rule out that CD2AP and cortactin localize to different F-actin spots. By double staining of cortactin and CD2AP, we now demonstrate that both proteins colocalize to identical spots. As we did not observe spots that selectively stained for either CD2AP or cortactin, both molecules appear to be essential components of spots and can be used as marker proteins. Colocalization of CD2AP and cortactin in spots may be the result of a direct interaction, as a binding site for cortactin has been recently mapped to the second proline-rich region of CD2AP (21).
Using GFP fusion proteins for actin and CD2AP, which labeled the same spots as the endogenous proteins, we observed that 50% of spots were associated with vesicles visible as dark circular areas (DCAs). Association of spots with vesicles was also confirmed in nontransfected podocytes by immunofluorescence and immunoelectron microscopy. Whether the other half of CD2AP spots is associated with vesicles that are too small to be visible as DCAs or whether some CD2AP spots are not linked to vesicles at all remains an open question. Using GFP-Arp3 and GFP-capping protein, Schafer and co-workers (28) observed similar spots containing F-actin and CD2AP that mainly formed finlike projections on the cell surface in the absence of membrane vesicles in the vicinity. We characterized DCAs as vesicles of a late endosomal compartment, because the fluid phase marker dextran was trapped in less than one-third of DCAs after 30 min. Dextran-loaded vesicles that were associated with CD2AP spots frequently exhibited an inhomogeneous luminal distribution of dextran, being consistent with a membranous suborganization in the vesicle interior. Budding and internalized membrane is the hallmark of multivesicular bodies (MVBs), which are a distinct late endosomal compartment (26). The fact that CD2AP/cortactin spots did not colocalize with AP-2 and Rab5 in podocytes provides further evidence that CD2AP spots do not associate with early endosomes. Moreover, colocalization of cortactin spots in podocytes with Rab4-positive vesicles places these vesicles at a later stage in the endosomal pathway. Colocalization of spots with Rab4 is consistent with the work of Cormont et al. (4), who discovered CD2AP to be a binding partner of the GTP-bound form of Rab4. Rab4 has been implicated in endosomal membrane sorting into the recycling and degradative pathway (23, 32). Based on several findings, Cormont et al. (4) speculated that Rab4, besides other functions, controls sorting from early to late endosomes. First, morphologically changed endosomes after Rab4 and CD2AP coexpression included the late endosome marker Rab7. Second, horseradish peroxidase uptake was altered by CD2AP overexpression only beyond 15 min, and, third, CD2AP was not involved in transferrin receptor recycling which is an early recycling pathway.
It is noteworthy that a large fraction of CD2AP spots is associated with vesicles in podocytes. In contrast, spots were rarely associated with vesicles in PtK1 fibroblasts (28). Furthermore, overexpression of a constitutive active form of the small GTPase ARF6 in PtK1 fibroblasts and CHO cells was needed to induce vesicle-associated actin spots and comet tails, the head of which stained for CD2AP (27). In addition, association of CD2AP with vesicular structures was only detectable in COS-7 cells after transfection with CD2AP (17) and in CHO cells after cotransfection with CD2AP and Cbl or Rab4 (4). In contrast, we detected association of CD2AP with vesicles in nontransfected podocytes. These findings may indicate that podocytes possess specifically enlarged endosomal compartment(s).
Our findings are consistent with the concept that CD2AP in spots around vesicles could be required for formation of MVBs and for protein sorting into the degradative pathway. According to this concept, Kim et al. (16) observed a twofold increase in MVBs in podocytes of wild-type mice after intravenous injection of ferritin, which failed to stimulate MVB formation in CD2AP+/– mice. Simultaneous overexpression of CD2AP with Cbl or Rab4 induced enlargement of EEA1-positive early endosomes in CHO cells (4), suggesting that CD2AP is involved in a sorting or transport step distal to early endosomes. MVB formation is required for downregulation of activated signaling receptors (15). Accordingly, CD2AP has been shown to be involved in EGF receptor trafficking (21), in PDGF receptor (4), and in VEGF receptor (Flt-1) degradation (19).
Our confocal time-lapse microscopy of GFP-actin and GFP-CD2AP in podocytes revealed the dynamic nature of spots in podocytes and a rather short but variable lifetime. Spot motility is nondirectional and does not seem to be linked to intracellular filaments. Spots are characterized by rapid accumulation of microinjected G-actin, demonstrating a high rate of actin polymerization in actin spots. Actin polymerization drives spot motility, as demonstrated by inhibition of motility with jasplakinolide. The high turnover rate of CD2AP in spots would be consistent with an involvement of CD2AP in regulating actin polymerization. In fact, it has been shown that CD2AP interacts with F-actin (20) and with several proteins known to regulate actin filament assembly: cortactin (21), capping protein (13), and the p85 regulatory subunit of PI3-kinase (11).
The actin cytoskeleton is importantly involved in various pathways of vesicular trafficking (34). Recently, it has become evident that actin polymerization participates in internalizing clathrin-coated pits (24). However, CD2AP spots were not colocalized with AP-2, an adaptor protein in clathrin-dependent endocytosis, in our experiments. On the other hand, the role of the actin cytoskeleton in later endosomal compartments, to which we localized CD2AP spots in podocytes, is not well defined. GFP-CD2AP as well as GFP-actin spots on vesicles exhibited either a circumferential movement of one or multiple spots around the vesicle or were concentrated at one point on the circumference generating a propulsive force for vesicular movement. The latter kind of movement has been reconstituted with vesicles of HeLa cells in Xenopus laevis egg extract (36). Interestingly, 80% of vesicles with actin tails were MVBs. Propulsive movement of vesicles has also been demonstrated for cortactin spots by Kaksonen et al. (14). Propulsive movements may enhance the chance of endosomes to reach targets within shorter times. On the other hand, we can only speculate about the function of the circumferential movement of vesicle-associated spots. By targeting cofilin to late endosomes in Dictyostelium, Drengk et al. (6) showed that loss of the late endosomal actin coat leads to aggregation of late endosomal vacuoles. Thus F-actin could function as a docking barrier. For docked vesicles, in contrast, actin polymerization has been suggested to forcibly promote fusion (8). Finally, actin could be involved in the inward budding of endosomal membranes and in pinching off, processes that are poorly understood (26).
GRANTS
This study was supported by grants of the German Research Foundation (DFG, En 329/7–3).
ACKNOWLEDGMENTS
We thank C. Kocksch for skilled technical assistance, H. Hosser for electron microscopy, T. Berger for continuous technical support regarding in vivo microscopy set-up, and R. Nonnenmacher for excellent graphical work.
FOOTNOTES
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.
REFERENCES
Borinstein SC, Hyatt MA, Sykes VW, Straub RE, Lipkowitz S, Boulter J, and Bogler O. SETA is a multifunctional adapter protein with three SH3 domains that binds Grb2, Cbl, and the novel SB1 proteins. Cell Signal 12: 769–779, 2000.
Bubb MR, Spector I, Beyer BB, and Fosen KM. Effects of jasplakinolide on the kinetics of actin polymerization. An explanation for certain in vivo observations. J Biol Chem 275: 5163–5170, 2000.
Chen B, Borinstein SC, Gillis J, Sykes VW, and Bogler O. The glioma-associated protein SETA interacts with AIP1/Alix and ALG-2 and modulates apoptosis in astrocytes. J Biol Chem 275: 19275–19281, 2000.
Cormont M, Meton I, Mari M, Monzo P, Keslair F, Gaskin C, McGraw TE, and Le Marchand-Brustel Y. CD2AP/CMS regulates endosome morphology and traffic to the degradative pathway through its interaction with Rab4 and c-Cbl. Traffic 4: 97–112, 2003.
Dikic I. CIN85/CMS family of adaptor molecules. FEBS Lett 529: 110–115, 2002.
Drengk A, Fritsch J, Schmauch C, Ruhling H, and Maniak M. A coat of filamentous actin prevents clustering of late-endosomal vacuoles in vivo. Curr Biol 13: 1814–1819, 2003.
Dustin ML, Olszowy MW, Holdorf AD, Li J, Bromley S, Desai N, Widder P, Rosenberger F, van der Merwe PA, Allen PM, and Shaw AS. A novel adaptor protein orchestrates receptor patterning and cytoskeletal polarity in T-cell contacts. Cell 94: 667–677, 1998.
Eitzen G. Actin remodeling to facilitate membrane fusion. Biochim Biophys Acta 1641: 175–181, 2003.
Gout I, Middleton G, Adu J, Ninkina NN, Drobot LB, Filonenko V, Matsuka G, Davies AM, Waterfield M, and Buchman VL. Negative regulation of PI 3-kinase by Ruk, a novel adaptor protein. EMBO J 19: 4015–4025, 2000.
Haglund K, Shimokawa N, Szymkiewicz I, and Dikic I. Cbl-directed monoubiquitination of CIN85 is involved in regulation of ligand-induced degradation of EGF receptors. Proc Natl Acad Sci USA 99: 12191–12196, 2002.
Huber TB, Hartleben B, Kim J, Schmidts M, Schermer B, Keil A, Egger L, Lecha RL, Borner C, Pavenstadt H, Shaw AS, Walz G, and Benzing T. Nephrin and CD2AP associate with phosphoinositide 3-OH kinase and stimulate AKT-dependent signaling. Mol Cell Biol 23: 4917–4928, 2003.
Hufner K, Higgs HN, Pollard TD, Jacobi C, Aepfelbacher M, and Linder S. The verprolin-like central (vc) region of Wiskott-Aldrich syndrome protein induces Arp2/3 complex-dependent actin nucleation. J Biol Chem 276: 35761–35767, 2001.
Hutchings NJ, Clarkson N, Chalkley R, Barclay AN, and Brown MH. Linking the T cell surface protein CD2 to the actin-capping protein CAPZ via CMS and CIN85. J Biol Chem 278: 22396–22403, 2003.
Kaksonen M, Peng HB, and Rauvala H. Association of cortactin with dynamic actin in lamellipodia and on endosomal vesicles. J Cell Sci 113: 4421–4426, 2000.
Katzmann DJ, Odorizzi G, and Emr SD. Receptor downregulation and multivesicular-body sorting. Nat Rev Mol Cell Biol 3: 893–905, 2002.
Kim JM, Wu H, Green G, Winkler CA, Kopp JB, Miner JH, Unanue ER, and Shaw AS. CD2-associated protein haploinsufficiency is linked to glomerular disease susceptibility. Science 300: 1298–1300, 2003.
Kirsch KH, Georgescu MM, Ishimaru S, and Hanafusa H. CMS: an adapter molecule involved in cytoskeletal rearrangements. Proc Natl Acad Sci USA 96: 6211–6216, 1999.
Kirsch KH, Georgescu MM, Shishido T, Langdon WY, Birge RB, and Hanafusa H. The adapter type protein CMS/CD2AP binds to the proto-oncogenic protein c-Cbl through a tyrosine phosphorylation-regulated Src homology 3 domain interaction. J Biol Chem 276: 4957–4963, 2001.
Kobayashi S, Sawano A, Nojima Y, Shibuya M, and Maru Y. The c-Cbl/CD2AP complex regulates VEGF-induced endocytosis and degradation of Flt-1 (VEGFR-1). FASEB J 18: 929–931, 2004.
Lehtonen S, Zhao F, and Lehtonen E. CD2-associated protein directly interacts with the actin cytoskeleton. Am J Physiol Renal Physiol 283: F734–F743, 2002.
Lynch DK, Winata SC, Lyons RJ, Hughes WE, Lehrbach GM, Wasinger V, Corthals G, Cordwell S, and Daly RJ. A Cortactin-CD2-associated protein (CD2AP) complex provides a novel link between epidermal growth factor receptor endocytosis and the actin cytoskeleton. J Biol Chem 278: 21805–21813, 2003.
Machesky LM and Insall RH. Scar1 and the related Wiskott-Aldrich syndrome protein, WASP, regulate the actin cytoskeleton through the Arp2/3 complex. Curr Biol 8: 1347–1356, 1998.
McCaffrey MW, Bielli A, Cantalupo G, Mora S, Roberti V, Santillo M, Drummond F, and Bucci C. Rab4 affects both recycling and degradative endosomal trafficking. FEBS Lett 495: 21–30, 2001.
Merrifield CJ. Seeing is believing: imaging actin dynamics at single sites of endocytosis. Trends Cell Biol 14: 352–358, 2004.
Pollard TD and Borisy GG. Cellular motility driven by assembly and disassembly of actin filaments. Cell 112: 453–465, 2003.
Raiborg C, Rusten TE, and Stenmark H. Protein sorting into multivesicular endosomes. Curr Opin Cell Biol 15: 446–455, 2003.
Schafer DA, D’Souza-Schorey C, and Cooper JA. Actin assembly at membranes controlled by ARF6. Traffic 1: 892–903, 2000.
Schafer DA, Welch MD, Machesky LM, Bridgman PC, Meyer SM, and Cooper JA. Visualization and molecular analysis of actin assembly in living cells. J Cell Biol 143: 1919–1930, 1998.
Schiwek D, Endlich N, Holzman L, Holthofer H, Kriz W, and Endlich K. Stable expression of nephrin and localization to cell-cell contacts in novel murine podocyte cell lines. Kidney Int 66: 91–101, 2004.
Schwarz K, Simons M, Reiser J, Saleem MA, Faul C, Kriz W, Shaw AS, Holzman LB, and Mundel P. Podocin, a raft-associated component of the glomerular slit diaphragm, interacts with CD2AP and nephrin. J Clin Invest 108: 1621–1629, 2001.
Shih NY, Li J, Karpitskii V, Nguyen A, Dustin ML, Kanagawa O, Miner JH, and Shaw AS. Congenital nephrotic syndrome in mice lacking CD2-associated protein. Science 286: 312–315, 1999.
Somsel Rodman J and Wandinger-Ness A. Rab GTPases coordinate endocytosis. J Cell Sci 113: 183–192, 2000.
Soubeyran P, Kowanetz K, Szymkiewicz I, Langdon WY, and Dikic I. Cbl-CIN85-endophilin complex mediates ligand-induced downregulation of EGF receptors. Nature 416: 183–187, 2002.
Stamnes M. Regulating the actin cytoskeleton during vesicular transport. Curr Opin Cell Biol 14: 428–433, 2002.
Take H, Watanabe S, Takeda K, Yu ZX, Iwata N, and Kajigaya S. Cloning and characterization of a novel adaptor protein, CIN85, that interacts with c-Cbl. Biochem Biophys Res Commun 268: 321–328, 2000.
Taunton J, Rowning BA, Coughlin ML, Wu M, Moon RT, Mitchison TJ, and Larabell CA. Actin-dependent propulsion of endosomes and lysosomes by recruitment of N-WASP. J Cell Biol 148: 519–530, 2000.
Uruno T, Liu J, Zhang P, Fan Y, Egile C, Li R, Mueller SC, and Zhan X. Activation of Arp2/3 complex-mediated actin polymerization by cortactin. Nat Cell Biol 3: 259–266, 2001.
Verdier F, Valovka T, Zhyvoloup A, Drobot LB, Buchman V, Waterfield M, and Gout I. Ruk is ubiquitinated but not degraded by the proteasome. Eur J Biochem 269: 3402–3408, 2002.
Weaver AM, Karginov AV, Kinley AW, Weed SA, Li Y, Parsons JT, and Cooper JA. Cortactin promotes and stabilizes Arp2/3-induced actin filament network formation. Curr Biol 11: 370–374, 2001.
Weed SA, Karginov AV, Schafer DA, Weaver AM, Kinley AW, Cooper JA, and Parsons JT. Cortactin localization to sites of actin assembly in lamellipodia requires interactions with F-actin and the Arp2/3 complex. J Cell Biol 151: 29–40, 2000.
Welsch T, Endlich N, Kriz W, and Endlich K. CD2AP and p130Cas localize to different F-actin structures in podocytes. Am J Physiol Renal Physiol 281: F769–F777, 2001.
Yarar D, To W, Abo A, and Welch MD. The Wiskott-Aldrich syndrome protein directs actin-based motility by stimulating actin nucleation with the Arp2/3 complex. Curr Biol 9: 555–558, 1999.(Thilo Welsch, Nicole Endl)