HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Benmerah, A. Articles by Dautry-Varsat, A. Search for Related Content PubMed PubMed Citation Articles by Benmerah, A. Articles by Dautry-Varsat, A. Social Bookmarking                      What's this?

J Biol Chem, Vol. 275, Issue 5, 3288-3295, February 4, 2000

Mapping of Eps15 Domains Involved in Its Targeting to Clathrin-coated Pits^*

Alexandre Benmerah^§¶, Viviane Poupon, Nadine Cerf-Bensussan, and Alice Dautry-Varsat^§

From  INSERM E9925, Faculté Necker-Enfants Malades, 156 rue de Vaugirard, 75730 Paris, and ^§ Unité de Biologie des Interactions Cellulaires, URA-CNRS 1960, Institut Pasteur, 25 rue de Dr. Roux, 75724 Paris, Cedex 15, France

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Clathrin-coated pit (CCP) formation occurs as a result of the targeting and assembly of cytosolic coat proteins, mainly the plasma membrane clathrin-associated protein complex (AP-2) and clathrin, to the intracellular face of the plasma membrane. In the present study, the mechanisms by which Eps15, an AP-2-binding protein, is targeted to CCPs was analyzed by following the intracellular localization of Eps15 mutants fused to the green fluorescent protein. Our previous results indicated that the N-terminal Eps15 homology (EH) domains are required for CCP targeting. We now show that EH domains are, however, not sufficient for targeting to CCPs. Similarly, neither the central coiled-coil nor the C-terminal AP-2 binding domains were able to address green fluorescent protein to CCPs. Thus, targeting of Eps15 to CCPs likely results from the collaboration between EH domains and another domain of the protein. An Eps15 mutant lacking the coiled-coil domain localized to CCPs showing that Eps15 dimerization is not strictly required. In contrast, Eps15 mutants lacking all AP-2 binding sites showed a dramatic decrease in plasma membrane staining, showing that AP-2 binding sites, together with EH domains, play an important role in targeting Eps15 into CCPs. Finally, the effect of the Eps15 mutants on clathrin-dependent endocytosis was tested by both immunofluorescence and flow cytometry. The results obtained showed that inhibition of transferrin uptake was observed only with mutants able to interfere with CCP assembly.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Clathrin-coated vesicle formation represents the initial step of the major pathway for receptor-mediated endocytosis. The known roles of AP-2,¹ clathrin and dynamin in this process are as follows. AP-2 is believed to drive clathrin assembly at the plasma membrane and binds to tyrosine-based internalization signals, playing a central role in both formation and function of clathrin-coated pits (CCPs); clathrin gives an organizing framework to the pit; and the dynamin GTPase activity is required for membrane fusion events leading to coated vesicle formation (1-5).

The mechanisms by which these soluble proteins are targeted to the plasma membrane to form an organized clathrin coat have been extensively studied. Targeting of AP-2 complexes is mediated by its -adaptin subunit. Studies of chimeras in which equivalent domains of -adaptin from AP-2 and -adaptin from AP-1 were exchanged allowed the identification of a CCP targeting signal within a 200-amino acid region of the N-terminal "head" or "trunk" domain of -adaptin. These studies also established a role for the C-terminal "ear" domain of -adaptin (6, 7), a result confirmed by recent data showing efficient targeting of a GFP-ear construct to CCP (8). Furthermore, the -adaptin N-terminal domain also contains a phosphoinositide binding site (9), which seems necessary for efficient incorporation of AP-2 complexes into CCPs (10). Clathrin assembly at the plasma membrane requires plasma membrane-bound AP-2 complexes (11, 12) and is thought to result from direct AP-2/clathrin interactions. Dynamin directly binds to salt-stripped membranes and to phospholipids containing liposomes (13-15), suggesting that dynamin might directly interact with the plasma membrane in vivo. This hypothesis was confirmed by the fact that dynamin is homogeneously redistributed at the plasma membrane when CCP assembly is inhibited (8). Its localization to CCPs requires its C-terminal proline-rich domain (16), which binds to the SH3 domain of amphiphysin (17). Recently, several new CCP-associated proteins were identified including Eps15 (18), amphiphysin (19, 20), and epsin (21). The mechanisms by which these accessory coat proteins are targeted to CCPs have not yet been investigated.

The Eps15 protein is constitutively and ubiquitously associated with AP-2 (22). Recent data obtained both in vivo and in a perforated cell assay showed that Eps15 is required for the early steps of clathrin-dependent endocytosis (23, 24). The fact that Eps15 is not found in coated vesicles (25) suggested a coated pit-restricted function. Results showing that the -adaptin ear domain, the Eps15 binding site on AP-2 (26, 27), is involved in the targeting of AP-2 to CCPs (6-8) have suggested that Eps15 may be part of the AP-2 docking machinery. This hypothesis is strengthened by the fact that an Eps15 mutant lacking EH domains inhibits CCP assembly (8). Eps15 is a conserved protein (28, 29) organized in three structural domains. The N-terminal domain (DI) contains three repeats of ~70 amino acids homologous to each other and to equivalent domains found in proteins from humans, yeast, and nematodes. These domains were called EH (30), for Eps15 Homology, and defined a novel protein-protein interacting module (31) that recognizes NPF-based motifs (32). The central domain (DII) contains heptad repeats required for coiled-coil structures and is involved in the dimerization of Eps15 (33, 34). The C-terminal domain (DIII) contains the AP-2 binding region (26), which spans ~120 amino acids (24) and contains three independent AP-2 binding sites (27). Therefore, as Eps15 does not contain any known plasma membrane targeting signals, its localization to CCPs is likely to be mediated by protein-protein interactions through its structural domains. The aim of this study was to identify these signals. With this goal in mind, we generated mutant forms of Eps15 and followed their intracellular localization. Their effect on clathrin-dependent endocytosis was further analyzed and compared with their capacity or inability to localize in CCP.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cells and Antibodies-- HeLa cells (ATCC, Manassas, VA) were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum, 2 mM L-glutamine, penicillin, and streptomycin (Life Technologies, Inc.). The mouse monoclonal antibodies AP.6 (anti-AP-2 (35)) and OKT9 (anti-human transferrin receptor (Tf-R)) were obtained from ATCC. Texas Red-conjugated goat anti-mouse immunoglobulins were obtained from Molecular Probes (Eugene, OR), Cy5-conjugated goat anti-mouse immunoglobulins were from Amersham Pharmacia Biotech, and phycoerythrin-conjugated F(ab')2 fragment goat anti-mouse IgG was from Immunotech (Marseille, France).

Generation of the GFP-Eps15 Constructs-- The cDNA of human Eps15 subcloned in pBluescript® II KS (Stratagene, La Jolla, CA) was obtained in the laboratory (22) and used as a template to generate the different cDNA fragments used in this study. The Eps15 constructs encoding each structural domain subcloned in the PGEX5.1 vector (Amersham Pharmacia Biotech) and the EH deleted mutant E95/295 were described previously (8, 24, 26). The deleted mutants were obtained by polymerase chain reactions and introduced within Eps15 subcloned between the BamHI/XhoI sites of the PGEX5.1 vector using appropriate restriction sites. Briefly, deletion of the central coiled-coil domain (ECC) was obtained by deleting a 635-base pair region corresponding to nucleotides 960-1605 of human Eps15, and the mutated DNA fragment was introduced within the Eps15 sequence using the HindIII (825) and ClaI (1937) sites. Mutants of the C-terminal domain were obtained by introducing stop codons in the corresponding lower primers at position 1856 (EA/P/C), 2298 (EP/C), or 2544 (EC). The mutated polymerase chain reaction fragments were then introduced within the Eps15 sequence using the BglII (1337) and XhoI sites. To generate a Eps15 mutant lacking only AP-2 binding sites (EAP-2), a 354-base pair region (1856 to 2298) was deleted as described previously (24). All of the different constructs were excised from the PGEX5.1 vector using the BamHI/XhoI sites, purified on agarose gel, and then subcloned into the BglII/SalI sites of the EGFP-C2 vector (CLONTECH). Restriction enzymes and T4 DNA ligase were from Amersham Pharmacia Biotech. All of the constructs were checked by nucleotide sequencing (Thermosequenase, Amersham Pharmacia Biotech). Sequences of the different primers used to generate the Eps15 mutants are available on request.

Transfections, Immunofluorescence, and Endocytosis-- Subconfluent HeLa cells were used for transient expression of the different constructs. Transfections were performed using the CalPhos Maximizer transfection kit from CLONTECH.

For immunofluorescence studies, transfected HeLa cells were grown on coverslips and used the day after transfection. The cells were washed in PBS and fixed in 3.7% paraformaldehyde and 0.03 M sucrose for 30 min at 4 °C. The cells were then washed once in PBS and, after quenching for 10 min in 50 mM NH₄Cl in PBS, washed again in PBS supplemented with 1 mg/ml bovine serum albumin (BSA). The cells were then incubated with the AP.6 antibody in permeabilization buffer (PBS with 1 mg/ml BSA and 0.05% saponin) for 45 min at room temperature. After two washes in the permeabilization buffer, the presence of antibodies was revealed by incubating the cells for 45 min at room temperature in permeabilization buffer containing labeled secondary antibody. After two washes in permeabilization buffer and one in PBS, the cells were mounted on microscope slides in 100 mg/ml Mowiol (Calbiochem), 25% glycerol (v/v), and 100 mM Tris-HCl, pH 8.5. To wash out cytosolic GFP constructs, the cells were briefly permeabilized before fixation with a 2-5-min incubation with 0.03% saponin in cytosolic buffer (100 mM potassium acetate, 1 mM MgCl₂, 20 mM Hepes) at 4 °C. The cells were then washed twice in cytosolic buffer and processed for immunofluorescence as described above. For surface staining, transfected cells were incubated with antibodies in PBS, 1 mg/ml BSA at 4 °C.

Human transferrin (Tf) was conjugated to Cy3 dye using the CyDye FluoroLink reactive dye kit from Amersham Pharmacia Biotech following the manufacturer's instructions. Endocytosis of Cy3-conjugated Tf was performed on subconfluent HeLa cells grown on coverslips 1 day after transfection. The cells were first incubated for 20 min at 37 °C in Dulbecco's modified Eagle's medium and 20 mM Hepes, pH 7.2, to eliminate receptor-bound Tf and then incubated in Dulbecco's modified Eagle's medium, 20 mM Hepes, pH 7.2, and 1 mg/ml BSA containing 100 nM Cy3-conjugated Tf. After incubation at 37 °C for 15 min, the cells were rapidly cooled to 4 °C, washed twice in cold PBS, and then fixed as described above. The samples were examined under an epifluorescence microscope (Zeiss, Oberkochen, Germany) attached to a cooled CCD camera (Photometrics, Tucson, AZ).

Quantification of Cell Surface Transferrin Receptors-- Cell surface Tf-Rs were quantified by flow cytometry. HeLa cells were detached from culture plates 1 day after transfection using a Costar cell lifter. Cells were washed twice in PBS, 1 mg/ml BSA and then incubated at 4 °C for 45 min with anti-Tf-R antibody OKT9 at saturating concentrations in PBS 1 mg/ml BSA. The cells were then washed twice in PBS 1 mg/ml BSA and incubated at 4 °C for 45 min in PBS 1 mg/ml BSA with a phycoerythrin-conjugated goat anti-mouse IgG. The cells were then washed twice in PBS 1 mg/ml BSA, and the levels of expression of both GFP and Tf-R on the cell surface were assessed using a FACSCAN (Becton Dickinson, San Jose, CA). Data were analyzed using the CELLQuest program (Becton Dickinson).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Endogenous Eps15 is constitutively found in CCPs (18, 36). Addition of the GFP to its N terminus does not modify its intracellular localization (8, 24), with the resulting GFP-Eps15 construct showing a punctate staining at the plasma membrane that colocalizes with AP-2 (Fig. 2, a, b, and insets). In this study, the GFP fusion system was used to map the domains of Eps15 involved in its targeting to CCPs. The different mutants derived from Eps15 (see Fig. 6) were fused to GFP, the resulting constructs were transiently transfected into HeLa cells, and their intracellular distribution was compared with that of AP-2.

The results obtained in our previous study showed that EH domains are required for CCP targeting of Ep15. Indeed, an Eps15 construct lacking the second and third EH domains showed a diffuse cytosolic staining (8). The putative role of EH domains as a sufficient coated pit targeting signal was first tested. As shown in Fig. 1a, a GFP construct encoding EH domains (DI) presented a diffuse intracellular staining, suggesting that the DI construct was not targeted to CCP. This was confirmed by its lack of colocalization with AP-2 at the plasma membrane (Fig. 1, a, b, and insets), showing that EH domains were unable to specifically target GFP to CCPs. From these and previous results (8) it can be concluded that EH domains are required but not sufficient for CCP targeting.

View larger version (95K): [in this window] [in a new window]   Fig. 1.   The structural domains of Eps15 are not sufficient coated pit targeting signals. HeLa cells transiently transfected with DI (a and b), DII (c and d), or DIII (e and f) GFP constructs were fixed, permeabilized, processed for fluorescence microscopy using the AP.6 antibody directed against AP-2, and revealed by a Texas Red-labeled secondary antibody as described under "Experimental Procedures." The cells were then observed under an epifluorescence microscope attached to a cooled CCD camera. The focus was on AP-2 dots on the planar plasma membrane adherent to the coverslip. a, c, and e, green fluorescence emitted by GFP. b, d, and f, red fluorescence emitted by Texas Red corresponding to AP-2 complexes. Insets show higher magnifications of representative areas of the plasma membrane; in c and e, a region under the nucleus was selected because of its lack of cytosolic GFP staining (see also Fig. 2, a and g and Fig. 3a).

The central coiled-coil and C-terminal AP-2 binding domains, DII and DIII constructs, respectively, were then tested for their capacity to bring GFP to CCPs. The DII construct was diffusely distributed in cytosol and did not colocalize with AP-2 at the plasma membrane (Fig. 1, c, d, and insets). A similar cytosolic staining was observed with the DIII construct (Fig. 1e). However, expression of the DIII construct, previously shown to inhibit clathrin-dependent endocytosis (24), induced a mislocalization of AP-2 complexes. Cells expressing DIII showed less AP-2 dots at the plasma membrane and an increase in cytosolic staining compared with neighboring untransfected cells (Fig. 1f, indicated by arrows). These results are reminiscent of those obtained with the E95/295 mutant (8) and indicated that expression of the DIII construct also inhibits clathrin-coated pit assembly. Nevertheless, the DIII construct did not colocalize with the AP-2 dots still present at the plasma membrane (Fig. 1, e, f, and insets), showing that the AP-2 binding domain is not a sufficient coated pit targeting signal. The lack of plasma membrane punctate staining observed for each structural domain was not due to an excess of cytosolic GFP, because brief permeabilization of the transfected cells before fixation effectively washed out cytosolic GFP constructs but did not reveal any specific plasma membrane-associated staining (data not shown). Altogether, these results show that none of the three structural domains of Eps15 is sufficient for CCP targeting.

The fact that EH domains are necessary but not sufficient for coated pit targeting suggests the involvement of another domain(s) in this process. We first investigated the role of the C-terminal domain. Indeed, the C-terminal domain of Eps15 contains both AP-2 binding sites and a proline-rich region. The proline-rich region is found between amino acids 768 and 849 and contains a PALPPK (768/774) sequence that binds to SH3 domain-containing proteins (37, 38). Interestingly, a proline-rich domain is responsible for the targeting of dynamin to CCPs (16) through interaction with the SH3 domain of amphiphysin (39). Such a mechanism could therefore also be involved in Eps15 targeting. Eps15 constructs lacking either the 849-896 (EC; data not shown) or the 763-896 region (EP/C; Fig. 2c) showed a bright plasma membrane punctate staining. Furthermore, the plasma membrane punctate staining completely colocalized with AP-2 (Fig. 2, c, d, and insets), showing that EP/C is targeted to CCPs. Thus, the proline-rich region of Eps15, responsible for binding to SH3 domain containing proteins, is not necessary for its targeting to CCPs.

View larger version (76K): [in this window] [in a new window]   Fig. 2.   AP-2 binding sites are required for coated pit targeting of Eps15. HeLa cells transiently transfected with Eps15 (a and b), EP/C (c and d), EA/P/C (e and f), or EAP-2 (g and h) GFP constructs were processed for fluorescence microscopy as for Fig. 1. The same fields are shown in a and b, c and d, e and f, and g and h. a, c, e, and g, green fluorescence emitted by GFP. b, d, f, and h, red fluorescence emitted by Texas Red corresponding to AP-2 complexes. Insets show higher magnifications of representative areas.

The role of AP-2 binding sites was investigated using two different constructs; EA/P/C lacking the entire C-terminal region after amino acid 620 and EAP-2 lacking only the AP-2 binding sites (621-738 region). As shown in Fig. 2, these two constructs showed a very faint plasma membrane staining (Fig. 2, e and g). This very faint staining still presented some colocalization with AP-2 (Fig. 2, e-h and insets), showing that the constructs could be found in CCPs. However, the plasma membrane staining was highly reduced compared with that observed for both wild type Eps15 (Fig. 2a) and for all of the constructs containing both EH domains and AP-2 binding sites (Fig. 2c and Fig. 3, a and c), showing that AP-2 binding sites play an important role in CCP targeting of Eps15. Noticeably, in contrast to endogenous Eps15² and transfected GFP-Eps15 (Fig. 2a), some constructs (including EP/C (Fig. 2c) and EA/P/C (Fig. 2e)) were found in the nucleus. Interestingly, the Eps15-related protein, Eps15r, is found both in plasma membrane CCPs and in the nucleus (40). Thus, Eps15 mutants may reveal a possible nucleocytoplasmic traffic for Eps15.³

Finally, the role of the central dimerization domain in coated pit localization was also tested. The ECC construct lacking the coiled-coil domain colocalized with AP-2 at the plasma membrane (Fig. 3, a, b, and insets), showing that the coiled-coil domain is not required for CCP targeting of Eps15. However, we repeatedly noticed that the CCP staining observed for ECC (Fig. 3a) was fainter than for the full-length Eps15 (Fig. 2a). Compared with EP/C (Fig. 2c), which only lacks the C-terminal and proline-rich regions, weaker CCP staining was also observed for the ECC+P/C (Fig. 3c) lacking the coiled-coil domain, the C-terminal, and the proline-rich regions. Altogether, these results suggest that the coiled-coil domain does not provide specific targeting information but rather helps to enhance the number of Eps15 molecules present in a given CCP.

View larger version (136K): [in this window] [in a new window]   Fig. 3.   The coiled-coil domain is not required for coated pit targeting of Eps15. HeLa cells transiently transfected with ECC (a and b) or ECC+P/C (c and d) constructs were fixed, permeabilized, and processed for fluorescence microscopy as described for Figs. 1 and 2. The same fields are shown in a and b and c and d. a and c, green fluorescence emitted by GFP. b and d, red fluorescence emitted by Texas Red corresponding to AP-2 complexes. Insets show higher magnifications of representative areas.

We next checked the effect of all of the Eps15 mutants (described above) on clathrin-dependent endocytosis. These mutants can be classified in two groups. Group I includes DIII and E95/295 that affect CCP assembly, and group II includes all of the other mutants that do not affect CCP assembly (Fig. 6). Indeed, in the latter group, clathrin-dependent endocytosis could be affected at a later step, i.e. CCP invagination and/or coated vesicle formation. The effect of these mutants on Tf uptake was therefore tested. Expression of group II mutants did not inhibit internalization of Cy3-conjugated Tf (Fig. 4B and data not shown) whereas a strong inhibition was observed with group I mutants (Fig. 4A and data not shown).

View larger version (47K): [in this window] [in a new window]   Fig. 4.   Clathrin-dependent endocytosis of transferrin is inhibited by expression of group I, but not of group II, Eps15 mutants. HeLa cells transiently transfected with E95/295 (A) or DI (B) constructs were incubated in the presence of Cy3-conjugated Tf at 37 °C for 15 min and then washed and fixed. To detect surface Tf-R, cells were incubated first with anti-transferrin receptor antibody followed by a Cy5-conjugated anti-mouse immunoglobulin. a, green fluorescence emitted by GFP. b, red fluorescence emitted by Cy3. c, far red fluorescence emitted by Cy5. The same field is shown in a, b, and c. d, black and white images were colored and superimposed using Adobe Photoshop software; green is for GFP, red is for Tf, and blue is for cell surface Tf-R.

We next verified that the lack of inhibition of Tf uptake by group II mutants was not due to lower expression levels of group II mutants compared with group I mutants. We took advantage of the fact that inhibition of clathrin-dependent endocytosis induces an increase in the number of Tf-R at the cell surface (41-44). Indeed, as shown in Fig. 4A, cells expressing the E95/295 construct (a) presented both an inhibition of internalization of Cy3-Tf (b) and an increase of cell surface Tf-R (c) compared with neighboring untransfected cells. In contrast, no increase of cell surface Tf-R was observed in cells expressing group II mutants that did not inhibit Cy3-Tf uptake (Fig. 4B and data not shown). An increase of cell surface Tf-R was then used to follow the inhibition of clathrin-dependent endocytosis. HeLa cells transiently transfected with group I, group II, and control constructs were analyzed by flow cytometry. Cells expressing identical levels of GFP constructs were selected (Fig. 5a, region 2) and analyzed for cell surface Tf-R expression. As shown in Fig. 5b, cells expressing the DII construct presented similar levels of cell surface Tf-R compared with cells expressing the control construct D32, which do not inhibit clathrin-dependent endocytosis (24). Similar results were also found for cells expressing the ECC construct (data not shown). Cells expressing either DI or EA/P/C constructs did not show increased cell surface Tf-R (Fig. 5b). Rather they expressed a lower level of cell surface Tf-R compared with that found for both DII and D32 constructs. Only cells expressing either DIII (Fig. 5b) or E95/295 (data not shown) showed a clear increase of cell surface Tf-R compared with cells expressing the control D32 construct. Therefore, the quantitative results obtained by flow cytometry confirm those obtained by Tf uptake experiments and show that, at identical expression levels, group I but not group II mutants inhibit CCP functions.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   The number of cell surface transferrin receptors is increased in cells expressing group I, but not by group II, Eps15 mutants. HeLa cells were transiently transfected with DI, DII, DIII, D32, and EA/P/C GFP constructs and stained for cell surface Tf-R using the OKT9 antibody revealed by a phycoerythrin-conjugated anti-mouse IgG. a, cells expressing identical high levels of GFP constructs were selected (region 2). They were then analyzed for phycoerythrin fluorescence corresponding to cell surface Tf-R expression (b). The mean fluorescence intensity for phycoerythrin is indicated for each mutant. Results shown are representative of four experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The present results extend our previous study and indicate that EH domain-binding protein(s), together with the AP-2 complex, play an important role in coated pit targeting of Eps15 (summarized in Fig. 6). Furthermore, they show that the inhibitory effect of Eps15 mutants correlates with inhibition of CCP assembly.

View larger version (35K): [in this window] [in a new window]   Fig. 6.   Structural organization of Eps15 mutants and summary of the results. A, P, and C stand for AP-2 binding, proline-rich, and C-terminal regions, respectively.

Results obtained by several groups have shown that Eps15 EH domains bind to numerous proteins through NPF-based motifs (21, 32). Some of the identified proteins including RAB/RABr are not found in CCPs nor involved in clathrin-dependent endocytosis and therefore can be excluded as physiological partners for Eps15 targeting. The most interesting candidate is the recently identified EH domain-binding protein, epsin. Like Eps15, epsin is a constitutive component of CCPs that is associated with AP-2 and required for clathrin-dependent endocytosis. Interestingly, overexpressed wild type epsin is localized to the cytosolic face of the plasma membrane in regions devoid of clathrin coat (21). This is in agreement with a role of epsin as a plasma membrane-associated docking protein, which could therefore link Eps15 to the plasma membrane. Furthermore, Eps15 recruitment into CCPs also requires binding to AP-2. The fact that AP-2/Eps15 interaction sites (26) are involved in their respective targeting to CCPs (6-8) suggests that Eps15 is targeted to CCPs complexed to AP-2.

The Eps15 mutants can be classified as group I mutants, which inhibit both CCP assembly and clathrin-dependent endocytosis and group II mutants, which do not (Fig. 6). The group I mutants contain the AP-2 binding sites. However, the presence of AP-2 binding sites is not sufficient for these inhibitory effects because the ECC construct that contains both EH domains and the AP-2 binding domain did not inhibit coated pit assembly nor clathrin-dependent endocytosis. The ECC construct was still targeted to CCP (Fig. 3) whereas DIII and E95/295 were not (see Fig. 1 and Ref. 8). Therefore, the inhibitory effects of Eps15 mutants on clathrin-dependent endocytosis requires both the presence of AP-2 binding sites and a lack of CCP localization.

Comparable domain requirements were also reported for dynamin. Dynamin mutants lacking the pleckstrin homology domain were recently shown to inhibit clathrin-dependent endocytosis of Tf (45-47). As previously shown for GTPase-deficient mutants (48), the C-terminal amphiphysin binding region is required for the inhibitory effect of the mutants on clathrin-dependent endocytosis (47). Furthermore, as observed for Eps15 with EH domains, the autonomously expressed dynamin pleckstrin homology domain does not inhibit clathrin-dependent endocytosis (49, 50).

Recently, the ear domain of -adaptin, the Eps15 binding site on AP-2, was shown to bind to many other proteins including epsin, amphiphysin, auscilin, and AP-180 (51). However, auxillin and AP-180 are only expressed in neuronal cells (52, 53); Eps15, epsin, and amphiphysin seemed to be the main -ear domain partners in peripheral cells. Interestingly, all of these different proteins bear DPF repeats first found in the AP-2 binding domain of Eps15 (26) and were shown to share with Eps15 the same -ear binding site (51). We could not formally exclude the possibility that the effects that we observed by expression of AP-2 binding sites containing Eps15 mutants on CCPs were because of a general inaccessibility of the -ear domain. Nevertheless, the increasing number of -ear binding proteins suggests a very important role of this AP-2 domain in clathrin coat function, an hypothesis confirmed by the fact that expression of -ear inhibits clathrin-dependent internalization of Tf. What could be these functions? The -ear domain could bring accessory proteins to the pit as it is thought for dynamin via amphiphysin (39). Another possibility, and both are not exclusive, is that -ear plays an active role in the recruitment of AP-2 complexes onto the plasma membrane. This hypothesis is sustained by the fact that the ear domains of -, - and -adaptins are the most divergent parts of assembly protein complexes (54-57) and could therefore discriminate between distinct docking machinery on their target membranes. The fact that an -ear domain GFP fusion protein is efficiently targeted to CCP (8) shows that this domain effectively acts as a sufficient coated pit targeting signal. In this model, epsin and Eps15 seemed to be interesting potential candidates as constituents of the plasma membrane-associated AP-2 docking machinery.

	ACKNOWLEDGEMENTS

We thank V. Collin and V. Mallarde for helpful technical assistance and D. Ojcius for careful reading of the manuscript.

	FOOTNOTES

^* This work was supported by grants from the Association pour la Recherche contre le Cancer (ARC), from the Fondation Princesse Grace de Monaco, and from Human Frontier Science Program (grant number RG404/96).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.

^¶ To whom correspondence should be addressed: INSERM E9925, Faculté Necker-Enfants Malades, 156 rue de Vaugirard, 75730 Paris, Cedex 15, France. Tel.: 33-1-40-61-56-38. Fax: 33-1-40-61-56-38. E-mail: benmerah@necker.fr.

Supported by Ligue Nationale contre le Cancer.

² A. Benmerah, unpublished observation.

³ V. Poupon, N. Cerf-Bensussan, A. Dautry-Varsat, and A. Benmerah, manuscript in preparation.

	ABBREVIATIONS

The abbreviations used are: AP-2, plasma membrane clathrin-associated protein complex; AP-1, Golgi clathrin-associated protein complex; CCP(s), clathrin-coated pit; EH, Eps15 homology; GFP, green fluorescent protein; Tf, transferrin; Tf-R(s), transferrin receptor; PBS, phosphate-buffered saline; BSA, bovine serum albumin.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				Y. Pu and X. Zhang Mouse Hepatitis Virus Type 2 Enters Cells through a Clathrin-Mediated Endocytic Pathway Independent of Eps15 J. Virol., August 15, 2008; 82(16): 8112 - 8123. [Abstract] [Full Text] [PDF]

				S. Chi, H. Cao, J. Chen, and M. A. McNiven Eps15 Mediates Vesicle Trafficking from the trans-Golgi Network via an Interaction with the Clathrin Adaptor AP-1 Mol. Biol. Cell, August 1, 2008; 19(8): 3564 - 3575. [Abstract] [Full Text] [PDF]

				L. Maldonado-Baez, M. R. Dores, E. M. Perkins, T. G. Drivas, L. Hicke, and B. Wendland Interaction between Epsin/Yap180 Adaptors and the Scaffolds Ede1/Pan1 Is Required for Endocytosis Mol. Biol. Cell, July 1, 2008; 19(7): 2936 - 2948. [Abstract] [Full Text] [PDF]

				A. Uezu, A. Horiuchi, K. Kanda, N. Kikuchi, K. Umeda, K. Tsujita, S. Suetsugu, N. Araki, H. Yamamoto, T. Takenawa, et al. SGIP1{alpha} Is an Endocytic Protein That Directly Interacts with Phospholipids and Eps15 J. Biol. Chem., September 7, 2007; 282(36): 26481 - 26489. [Abstract] [Full Text] [PDF]

				A. A. Kolokoltsov, D. Deniger, E. H. Fleming, N. J. Roberts Jr., J. M. Karpilow, and R. A. Davey Small Interfering RNA Profiling Reveals Key Role of Clathrin-Mediated Endocytosis and Early Endosome Formation for Infection by Respiratory Syncytial Virus J. Virol., July 15, 2007; 81(14): 7786 - 7800. [Abstract] [Full Text] [PDF]

				A. G. Ramsay, M. D. Keppler, M. Jazayeri, G. J. Thomas, M. Parsons, S. Violette, P. Weinreb, I. R. Hart, and J. F. Marshall HS1-Associated Protein X-1 Regulates Carcinoma Cell Migration and Invasion via Clathrin-Mediated Endocytosis of Integrin {alpha}v{beta}6 Cancer Res., June 1, 2007; 67(11): 5275 - 5284. [Abstract] [Full Text] [PDF]

				T. H. T. Tran, Q. Zeng, and W. Hong VAMP4 cycles from the cell surface to the trans-Golgi network via sorting and recycling endosomes J. Cell Sci., March 15, 2007; 120(6): 1028 - 1041. [Abstract] [Full Text] [PDF]

				J. Z. Rappoport, S. Kemal, A. Benmerah, and S. M. Simon Dynamics of clathrin and adaptor proteins during endocytosis Am J Physiol Cell Physiol, November 1, 2006; 291(5): C1072 - C1081. [Abstract] [Full Text] [PDF]

				M. A. McNiven and H. M. Thompson Vesicle formation at the plasma membrane and trans-Golgi network: the same but different. Science, September 15, 2006; 313(5793): 1591 - 1594. [Abstract] [Full Text] [PDF]

				C. Leterrier, J. Laine, M. Darmon, H. Boudin, J. Rossier, and Z. Lenkei Constitutive activation drives compartment-selective endocytosis and axonal targeting of type 1 cannabinoid receptors. J. Neurosci., March 22, 2006; 26(12): 3141 - 3153. [Abstract] [Full Text] [PDF]

M. P. Smith, V. J. Ayad, S