Assembly and targeting of adaptin chimeras in transfected cells

Adaptors are the components of clathrincoated pits and vesicles that attach the clathrin to the membrane. There are two types of adaptors in the cell: one associated with the plasma membrane and one associated with the TGN. Both adaptors are heterotetramers consisting of two adaptins (alpha and beta for the plasma membrane; gamma and beta' for the TGN), plus two smaller proteins. The COOH-terminal domains of the adaptins form appendages that resemble ears, connected by flexible hinges. Unlike the other adaptor components, the COOH termini of the alpha- and gamma-adaptins show no homology with each other, suggesting that they might provide the signal that directs the adaptors to the appropriate membrane. To test this possibility, the COOH-terminal ears were switched between alpha- and gamma-adaptins and were also deleted. All of the constructs contained the bovine gamma-adaptin hinge, enabling them to be detected with a species-specific antibody against this region when transfected into rat fibroblasts. Immunoprecipitation indicated that the engineered adaptins were still fully capable of assembling into adaptor complexes. Immunofluorescence revealed that in spite of their modified ears, the constructs were still able to be recruited onto the appropriate membrane; however, the ear-minus constructs gave increased cytoplasmic staining, and replacing the gamma- adaptin ear with the alpha-adaptin ear caused a small amount of colocalization with endogenous alpha-adaptin in some cells. Thus, the major targeting determinant appears to reside in the adaptor "head," while the ears may stabilize the association of adaptors with the membrane.     

Adaptor self-aggregation, adaptor-receptor recognition and binding of alpha-adaptin subunits to the plasma membrane contribute to recruitment of adaptor (AP2) components of clathrin-coated pits.

Initiation of receptor-mediated endocytosis by nucleation of clathrin-coated pits involves binding of AP2 adaptor molecules to the plasma membrane. This process was reconstituted in vitro, using plasma membrane fragments, prepared by freeze-thaw lysis of cells, and stripped of their endogenous coat proteins, as targets for binding of purified adaptor molecules and their dissociated subunits. The dissociated alpha-adaptin subunit of AP2 bound to plasma membrane fragments, while the dissociated beta-adaptin subunit did not, suggesting that plasma membrane localization of AP2 adaptors is mediated by alpha-adaptin. Membrane binding of intact AP2 adaptor molecules was enhanced by adaptor self-aggregation, which can be modulated by physiological concentrations of inositol phosphates, and may therefore be sensitive to receptor signaling. Adaptor binding was partially inhibited by soluble peptides representing the cytoplasmic domains of the asialoglycoprotein receptor and the polymeric immunoglobulin receptor. These results indicate that direct binding of adaptors to the cytoplasmic domains of receptors contributes to coated pit nucleation but this appears to be a weak interaction, suggesting that an additional recognition signal could be required for high affinity adaptor binding.     

Clathrin interaction and subcellular localization of Ce-DAB-1, an adaptor for protein secretion in Caenorhabditis elegans

Growth factors must be secreted appropriately to co-ordinate cell proliferation, specification and movement during development and to control cell numbers and migrations in adult animals. Previous results showed that the secretion of the Caenorhabditis elegans fibroblast growth factor homologue, EGL-17, from vulval precursor cells in vivo involves the cytoplasmic adaptor protein Ce-DAB-1 and two lipoprotein receptors that bind Ce-DAB-1 and EGL-17. Here, we confirm the Ce-DAB-1 requirement for EGL-17 secretion using mutant animals. In vitro, Ce-DAB-1 binds to clathrin and APT-4, the C. elegans homologue of the alpha-adaptin subunit of adaptor protein 2 (AP2), and weakly to the gamma-appendage domains of APT-1 (AP1gamma-adaptin) and APT-9 (GGA protein). In tissue-culture cells, Ce-DAB-1 localizes to various compartments, including AP2-containing vesicles near the cell surface and perinuclear vesicles that contain AP1. The latter also contain Rab8, but not Rab5 or Rab11, as well as proteins en route from the trans Golgi network (TGN) to the surface. In vivo, EGL-17 secretion was inhibited by depletion of apt-1, apt-9 or ce-rab-8 and partially inhibited by RNAi of ce-rab-5, consistent with an important role for these proteins in the secretion of EGL-17 in vivo. These results suggest that Ce-DAB-1 might co-ordinate the assembly of endocytic or secretory vesicles in vivo and may mediate EGL-17 secretion directly, by recruiting clathrin to lipoprotein receptors at the TGN, or indirectly, by affecting lipoprotein receptor endocytosis and recycling.     

A novel adaptor-related protein complex

Coat proteins are required for the budding of the transport vesicles that mediate membrane traffic pathways, but for many pathways such proteins pathways, but for many pathways such proteins have not yet been identified. We have raised antibodies against p47, a homologue of the medium chains of the adaptor complexes of clathrin-coated vesicles (Pevsner, J., W. Volknandt, B.R. Wong, and R.H. Scheller. 1994. Gene (Amst.). 146:279-283), to determine whether this protein might be a component of a new type of coat. p47 coimmunoprecipitates with three other proteins: two unknown proteins of 160 and 25 kD, and beta-NAP, a homologue of the beta/beta'-adaptins, indicating that it is a subunit of an adaptor-like heterotetrameric complex. However, p47 is not enriched in preparations of clathrin-coated vesicles. Recruitment of the p47-containing complex onto cell membranes is stimulated by GTP gamma S and blocked by brefeldin A, indicating that, like other coat proteins, its membrane association is regulated by an ARF. The newly recruited complex is localized to non-clathrin-coated buds and vesicles associated with the TGN. Endogenous complex in primary cultures of neuronal cells is also localized to the TGN, and in addition, some complex is associated with the plasma membrane. These results indicate that the complex is a component of a novel type of coat that facilitates the budding of vesicles from the TGN, possibly for transporting newly synthesized proteins to the plasma membrane.     

Structural basis for the accessory protein recruitment by the gamma-adaptin ear domain

The adaptor proteins AP-1 and GGA regulate membrane traffic between the trans-Golgi network (TGN) and endosomes/lysosomes through ARF-regulated membrane association, recognition of sorting signals, and recruitment of clathrin and accessory proteins. The gamma 1-adaptin subunits of AP-1 and GGA possess homologous ear domains involved in the recruitment of accessory proteins, gamma-synergin and Rabaptin-5. The crystal structure of the human gamma 1-adaptin ear domain consists solely of an immunoglobulin-like fold, unlike the alpha-adaptin ear domain. Structure-based mutational analyses reveal a binding site for the accessory proteins that is composed of conserved basic residues, indicating that the recruitment mechanism in gamma 1-adaptin and GGA is distinct from that in alpha-adaptin.     

100-kD proteins of Golgi- and trans-Golgi network-associated coated vesicles have related but distinct membrane binding properties

The 100-110-kD proteins (alpha-, beta-, beta'-, and gamma-adaptins) of clathrin-coated vesicles and the 110-kD protein (beta-COP) of the nonclathrin-coated vesicles that mediate constitutive transport through the Golgi have homologous protein sequences. To determine whether homologous processes are involved in assembly of the two types of coated vesicles, the membrane binding properties of their coat proteins were compared. After treatment of MDBK cells with the fungal metabolite Brefeldin A (BFA), beta-COP was redistributed to the cytoplasm within 15 s, gamma-adaptin and clathrin in the trans-Golgi network (TGN) dispersed within 30 s, but the alpha-adaptin and clathrin present on coated pits and vesicles derived from the plasma membrane remained membrane associated even after a 15-min exposure to BFA. In PtK1 cells and MDCK cells, BFA did not affect beta-COP binding or Golgi morphology but still induced redistribution of gamma-adaptin and clathrin from TGN membranes to the cytoplasm. Thus BFA affects the binding of coat proteins to membranes in the Golgi region (Golgi apparatus and TGN) but not plasma membranes. However, the Golgi binding interactions of beta-COP and gamma-adaptin are distinct and differentially sensitive to BFA. BFA treatment did not release gamma-adaptin or clathrin from purified clathrin-coated vesicles, suggesting that their distribution to the cytoplasm after BFA treatment of cells was due to interference with their rebinding to TGN membranes after a normal cycle of disassembly. This was confirmed using an in vitro assay in which gamma-adaptin binding to TGN membranes was blocked by BFA and enhanced by GTP gamma S, similar to the binding of beta-COP to Golgi membranes. These results suggest the involvement of GTP-dependent proteins in the association of the 100-kD coat proteins with membranes in the Golgi region of the cell.     

Cloning and expression of gamma-adaptin, a component of clathrin-coated vesicles associated with the Golgi apparatus

Adaptins are the major components of adaptors, the protein complexes that link clathrin to transmembrane proteins (e.g., receptors) in coated pits and vesicles. The plasma membrane adaptor contains an alpha- adaptin subunit and a beta-adaptin subunit, while the Golgi adaptor contains a gamma-adaptin subunit and a beta'-adaptin subunit. A partial cDNA clone encoding gamma-adaptin was isolated from a bovine brain expression library by screening with antibodies, and was used to obtain a cDNA clone from a mouse brain library containing the full coding sequence. The identity of the clones was confirmed by protein sequencing. The deduced amino acid sequence of gamma-adaptin was found to be homologous to that of alpha-adaptin, with several stretches of identical amino acids or conservative substitutions in the first approximately 70 kD, and 25% identity overall. Weaker homology was seen between gamma- and beta-adaptins. Like both alpha- and beta-adaptins, gamma-adaptin has a proline and glycine-rich hinge region, dividing it into NH2- and COOH-terminal domains. A chimeric gamma-adaptin was constructed from the mouse and bovine cDNAs and transfected into Rat 1 fibroblasts. Immunofluorescence microscopy was carried out using an mAb which recognizes an epitope present on the chimera but not found on the rodent protein. The construct was found to have a distribution typical of endogenous gamma-adaptin. Using this transfection system, it should now be possible to exchange domains between alpha- and gamma-adaptins, to try to find out how adaptors are targeted to the appropriate membrane compartment of the cell, and how they recruit the appropriate receptors into the coated vesicle.     

Similar subunit interactions contribute to assembly of clathrin adaptor complexes and COPI complex: analysis using yeast three-hybrid system.

Clathrin adaptor protein (AP) complexes are heterotetramers composed of two large, one medium, and one small subunits. By exploiting the yeast three-hybrid system, we have found that an interaction between the two large subunits of the AP-1 complex, gamma-adaptin and beta1-adaptin, is markedly enhanced in the presence of the small subunit, sigma1. Similarly, two large subunits of the AP-4 complex, epsilon-adaptin and beta4-adaptin, are found to interact with each other only in the presence of the small subunit, sigma4. Furthermore, we have found that an interaction between two large subunits of the COPI F subcomplex, gamma-COP and beta-COP, is detectable only in the presence of zeta-COP. Because these COPI subunits have common ancestral origins to the corresponding AP subunits, these three-hybrid data, taken together with the previous two-hybrid data, suggest that the AP complexes and the COPI F subcomplex assemble by virtue of similar subunit interactions.     

Subunit interaction and function of clathrin-coated vesicle adaptors from the Golgi and the plasma membrane

Clathrin in coated vesicles is linked to transmembrane receptors by adaptor protein complexes. The Golgi-associated adaptor complex HA1 is a tetramer, made up of beta', gamma, 47-kDa, and 20-kDa subunits, whereas the tetrameric plasma membrane adaptor, HA2, contains alpha, beta, 50-kDa, and 16-kDa subunits (Ahle, S., Mann, A., Eichelsbacher, U., and Ungewickell, E. (1988) EMBO J. 7, 919-929). Here we report on the structural organization of adaptor subunits as revealed by proteolytic dissection. We show that the beta' and gamma subunits of HA1 are cleaved into 60-67-kDa "trunk" and 32-44-kDa "head" fragments. Interactions between adaptor subunits involve the trunk domains only. In overall organization of their domains, the Golgi and plasma membrane adaptors are very similar. The similarity encompasses also the location of phosphorylated serine residues in the alpha a, beta, beta', and gamma subunits, which are found in the head domains in all cases. In the alpha a and beta subunits they probably occur in the proline- and glycine-rich hinge region, which connects the head to the trunk. Identical adaptor fragments were obtained by controlled digestion of clathrin-coated vesicles. Under conditions that did not affect the integrity of the clathrin heavy chain, the adaptor head fragments were always quantitatively released from coated vesicles. The release of the bulk of the adaptors occurred concomitantly with the cleavage of their beta-type subunits (beta and beta') and under buffer conditions that prevent aggregation of adaptors. These observations taken together with the results of reconstitution experiments confirm and extend previous data (Ahle, S., and Ungewickell, E. (1989) J. Biol. Chem. 264, 20089-20093) which suggested that adaptors attach to clathrin through their beta-type (beta and beta') subunits. Moreover, high affinity interaction between adaptors and clathrin requires the participation of regions from both the head and trunk domains of the beta-type subunits.     

Direct interaction of the trans-Golgi network membrane protein, TGN38, with the F-actin binding protein, neurabin.

TGN38 is a type I integral membrane protein that constitutively cycles between the trans-Golgi network (TGN) and plasma membrane. The cytosolic domain of TGN38 interacts with AP2 clathrin adaptor complexes via the tyrosine-containing motif (-SDYQRL-) to direct internalization from the plasma membrane. This motif has previously been shown to direct both internalization and subsequent TGN targeting of TGN38. We have used the cytosolic domain of TGN38 in a two-hybrid screen, and we have identified the brain-specific F-actin binding protein neurabin-I as a TGN38-binding protein. We demonstrate a direct interaction between TGN38 and the ubiquitous homologue of neurabin-I, neurabin-II (also called spinophilin). We have used a combination of yeast two-hybrid and in vitro protein interaction assays to show that this interaction is dependent on the serine (but not tyrosine) residue of the known TGN38 trafficking motif. We show that TGN38 interacts with the coiled coil region of neurabin in vitro and binds preferentially with the dimeric form of neurabin. TGN38 and neurabin also interact in vivo as demonstrated by coimmunoprecipitation from stably transfected PC12 cells. These data suggest that neurabin provides a direct physical link between TGN38-containing membranes and the actin cytoskeleton.     

The AP-1A and AP-1B clathrin adaptor complexes define biochemically and functionally distinct membrane domains

Most epithelial cells contain two AP-1 clathrin adaptor complexes. AP-1A is ubiquitously expressed and involved in transport between the TGN and endosomes. AP-1B is expressed only in epithelia and mediates the polarized targeting of membrane proteins to the basolateral surface. Both AP-1 complexes are heterotetramers and differ only in their 50-kD mu1A or mu1B subunits. Here, we show that AP-1A and AP-1B, together with their respective cargoes, define physically and functionally distinct membrane domains in the perinuclear region. Expression of AP-1B (but not AP-1A) enhanced the recruitment of at least two subunits of the exocyst complex (Sec8 and Exo70) required for basolateral transport. By immunofluorescence and cell fractionation, the exocyst subunits were found to selectively associate with AP-1B-containing membranes that were both distinct from AP-1A-positive TGN elements and more closely apposed to transferrin receptor-positive recycling endosomes. Thus, despite the similarity of the two AP-1 complexes, AP-1A and AP-1B exhibit great specificity for endosomal transport versus cell polarity.     

Functional and physical interactions of the adaptor protein complex AP-4 with ADP-ribosylation factors (ARFs)

AP-4 is a member of the family of heterotetrameric adaptor protein (AP) complexes that mediate the sorting of integral membrane proteins in post-Golgi compartments. This complex consists of four subunits (epsilon, beta4, mu4 and sigma4) and localizes to the cytoplasmic face of the trans-Golgi network (TGN). Here, we show that the recruitment of endogenous AP-4 to the TGN in vivo is regulated by the small GTP-binding protein ARF1. In addition, we demonstrate a direct interaction of the epsilon and mu4 subunits of AP-4 with ARF1. epsilon binds only to ARF1-GTP and requires residues in the switch I and switch II regions of ARF1. In contrast, mu4 binds equally well to the GTP- and GDP-bound forms of ARF1 and is less dependent on switch I and switch II residues. These observations establish AP-4 as an ARF1 effector and suggest a novel mode of interaction between ARF1 and an AP complex involving both constitutive and regulated interactions.     

A novel clathrin homolog that co-distributes with cytoskeletal components functions in the trans-Golgi network

A clathrin homolog encoded on human chromosome 22 (CHC22) displays distinct biochemistry, distribution and function compared with conventional clathrin heavy chain (CHC17), encoded on chromosome 17. CHC22 protein is upregulated during myoblast differentiation into myotubes and is expressed at high levels in muscle and at low levels in non-muscle cells, relative to CHC17. The trimeric CHC22 protein does not interact with clathrin heavy chain subunits nor bind significantly to clathrin light chains. CHC22 associates with the AP1 and AP3 adaptor complexes but not with AP2. In non-muscle cells, CHC22 localizes to perinuclear vesicular structures, the majority of which are not clathrin coated. Treatments that disrupt the actin-myosin cytoskeleton or affect sorting in the trans-Golgi network (TGN) cause CHC22 redistribution. Overexpression of a subdomain of CHC22 induces altered distribution of TGN markers. Together these results implicate CHC22 in TGN membrane traffic involving the cytoskeleton.     

Recruitment of coat proteins onto Golgi membranes in intact and permeabilized cells: effects of brefeldin A and G protein activators.

Brefeldin A (BFA) causes a rapid redistribution of coat proteins (e.g., gamma-adaptin) associated with the clathrin-coated vesicles that bud from the trans-Golgi network (TGN), while the clathrin-coated vesicles that bud from the plasma membrane are unaffected. gamma-Adaptin redistributes with the same kinetics as beta-COP, a coat protein associated with the non-clathrin-coated vesicles that bud from the Golgi complex. Upon removal of BFA, however, gamma-adaptin recovers its perinuclear distribution more rapidly. Redistribution of both proteins can be prevented by pretreating cells with AlF4-. Recruitment of adaptors from the cytosol onto the TGN membrane has been reconstituted in a permeabilized cell system and is increased by addition of GTP gamma S and blocked by addition of BFA. These results suggest a role for G proteins in the control of the clathrin-coated vesicle cycle at the TGN and further extend the similarities between clathrin-coated vesicles and non-clathrin-coated vesicles.     

Trafficking of the Menkes copper transporter ATP7A is regulated by clathrin-, AP-2–, AP-1–, and Rab22-dependent steps

The transporter ATP7A mediates systemic copper absorption and provides cuproenzymes in the trans-Golgi network (TGN) with copper. To regulate metal homeostasis, ATP7A constitutively cycles between the TGN and plasma membrane (PM). ATP7A trafficking to the PM is elevated in response to increased copper load and is reversed when copper concentrations are lowered. Molecular mechanisms underlying this trafficking are poorly understood. We assess the role of clathrin, adaptor complexes, lipid rafts, and Rab22a in an attempt to decipher the regulatory proteins involved in ATP7A cycling. While RNA interference (RNAi)–mediated depletion of caveolin 1/2 or flotillin had no effect on ATP7A localization, clathrin heavy chain depletion or expression of AP180 dominant-negative mutant not only disrupted clathrin-regulated pathways, but also blocked PM-to-TGN internalization of ATP7A. Depletion of the μ subunits of either adaptor protein-2 (AP-2) or AP-1 using RNAi further provides evidence that both clathrin adaptors are important for trafficking of ATP7A from the PM to the TGN. Expression of the GTP-locked Rab22a_Q64L mutant caused fragmentation of TGN membrane domains enriched for ATP7A. These appear to be a subdomain of the mammalian TGN, showing only partial overlap with the TGN marker golgin-97. Of importance, ATP7A remained in the Rab22a_Q64L-generated structures after copper treatment and washout, suggesting that forward trafficking out of this compartment was blocked. This study provides evidence that multiple membrane-associated factors, including clathrin, AP-2, AP-1, and Rab22, are regulators of ATP7A trafficking.     

Stabilization of clathrin coats by the core of the clathrin-associated protein complex AP-2

AP-2 is the class of clathrin-associated protein complex found in coated vesicles derived from the plasma membrane of eukaryotic cells. We demonstrate here, using a chemical method, that an AP-2 complex is an asymmetric structure consisting of one large alpha chain, one large beta chain, one medium AP50 chain, and one small AP17 chain. The complex has been shown to contain a core and two appendages. The AP core includes the small AP17 and the medium AP50 chains together with the amino-terminal domains of the large alpha and beta chains. One appendage corresponds to the carboxy-terminal domain of the beta chain. We find that as in the case of the beta chains, the carboxy-terminal portion of the alpha chains is an independently folded domain corresponding to the second appendage. We use limited tryptic proteolysis of clathrin/AP-2 coats to show the release of the appendages from the interior of the coats and the retention of the AP core by the remaining clathrin lattice. In addition, we find that the AP core stabilizes the coat and prevents its depolymerization. These results are consistent with the proposal that the AP core contains the binding site(s) for clathrin, while the alpha- and beta-chain appendages interact with membrane components of coated pits and coated vesicles.     

The Salmonella transmembrane effector SteD hijacks AP1-mediated vesicular trafficking for delivery to antigen-loading MHCII compartments

SteD is a transmembrane effector of the Salmonella SPI-2 type III secretion system that inhibits T cell activation by reducing the amounts of at least three proteins –major histocompatibility complex II (MHCII), CD86 and CD97 –from the surface of antigen-presenting cells. SteD specifically localises at the trans-Golgi network (TGN) and MHCII compartments; however, the targeting, membrane integration and trafficking of SteD are not understood. Using systematic mutagenesis, we identify distinct regions of SteD that are required for these processes. We show that SteD integrates into membranes of the ER/Golgi through a two-step mechanism of membrane recruitment from the cytoplasm followed by integration. SteD then migrates to and accumulates within the TGN. From here it hijacks the host adaptor protein (AP)1-mediated trafficking pathway from the TGN to MHCII compartments. AP1 binding and post-TGN trafficking require a short sequence in the N-terminal cytoplasmic tail of SteD that resembles the AP1-interacting dileucine sorting signal, but in inverted orientation, suggesting convergent evolution.     

Cross-talk between clathrin-dependent post-Golgi trafficking and clathrin-mediated endocytosis in Arabidopsis root cells

Coupling of post-Golgi and endocytic membrane transport ensures that the flow of materials to/from the plasma membrane (PM) is properly balanced. The mechanisms underlying the coordinated trafficking of PM proteins in plants, however, are not well understood. In plant cells, clathrin and its adaptor protein complexes, AP-2 and the TPLATE complex (TPC) at the PM, and AP-1 at the trans-Golgi network/early endosome (TGN/EE), function in clathrin-mediated endocytosis (CME) and post-Golgi trafficking. Here, we utilized mutants with defects in clathrin-dependent post-Golgi trafficking and CME, in combination with other cytological and pharmacological approaches, to further investigate the machinery behind the coordination of protein delivery and recycling to/from the TGN/EE and PM in Arabidopsis (Arabidopsis thaliana) root cells. In mutants with defective AP-2-/TPC-dependent CME, we determined that clathrin and AP-1 recruitment to the TGN/EE as well as exocytosis are significantly impaired. Likewise, defects in AP-1-dependent post-Golgi trafficking and pharmacological inhibition of exocytosis resulted in the reduced association of clathrin and AP-2/TPC subunits with the PM and a reduction in the internalization of cargoes via CME. Together, these results suggest that post-Golgi trafficking and CME are coupled via modulation of clathrin and adaptor protein complex recruitment to the TGN/EE and PM.     

The beta2-adaptin clathrin adaptor interacts with the mitotic checkpoint kinase BubR1

The adaptor AP2 is a heterotetrameric complex that associates with clathrin and regulatory proteins to mediate rapid endocytosis from the plasma membrane. Here, we report the identification of the mitotic checkpoint kinase BubR1 as a novel binding partner of beta2-adaptin, one of the AP2 large subunits. Using two-hybrid experiments and in vitro binding assays, we show that beta2-adaptin binds to BubR1 through its amino-terminal beta2-'trunk' domain, while the beta2-binding region of BubR1 maps to the carboxy-terminal kinase domain. Subcellular immunolocalization studies suggest that the interaction between BubR1 and beta2-adaptin could take place in the cytosol at any time during the cell cycle. In addition, we found that BubR1 and the BubR1-related kinase, Bub1, also bind to beta-adaptins of other AP complexes. Together, these results support a model in which the mitotic checkpoint kinases BubR1 and BuB1, by binding to beta-adaptins, may play novel roles in the regulation of vesicular intracellular traffic.     

A novel class of clathrin-coated vesicles budding from endosomes

Clathrin-coated vesicles transport selective integral membrane proteins from the plasma membrane to endosomes and from the TGN to endosomes. Recycling of proteins from endosomes to the plasma membrane occurs via unidentified vesicles. To study this pathway, we used a novel technique that allows for the immunoelectron microscopic examination of transferrin receptor-containing endosomes in nonsectioned cells. Endosomes were identified as separate discontinuous tubular-vesicular entities. Each endosome was decorated, mainly on the tubules, with many clathrin-coated buds. Endosome-associated clathrin-coated buds were discerned from plasma membrane-derived clathrin-coated vesicles by three criteria: size (60 nm and 100 nm, respectively), continuity with endosomes, and the lack of labeling for alpha-adaptin. They were also distinguished from TGN-derived clathrin-coated vesicles by their location at the periphery of the cell, size, and the lack of labeling for gamma-adaptin. In the presence of brefeldin A, a large continuous endosomal network was formed. Transferrin receptor recycling as well as the formation of clathrin-coated pits at endosomes was inhibited in the presence of brefeldin A. Together with the localization of transferrin receptors at endosome-associated buds, this indicates that a novel class of clathrin-coated vesicles serves an exit pathway from endosomes. The target organelles for endosome-derived clathrin-coated vesicles remain, however, to be identified.