Clathrin- and AP-2–binding sites in HIP1 uncover a general

assembly role for endocytic accessory proteins.

Sanjay K. Mishra*, Nicole R. Agostinelli*, Tom J. Brett¶, Ikuko Mizukami†,

Theodora S. Ross† and Linton M. Traub*§.

*Department of Cell Biology and Physiology, University of Pittsburgh School of

Medicine, Pittsburgh, PA 15261

¶Department of Pathology, Washington University School of Medicine, St. Louis

MO 63110

†Department of Internal Medicine, University of Michigan Comprehensive

Cancer Center, Ann Arbor, MI 48109

§ To whom correspondence should be addressed at:

Department of Cell Biology and Physiology

University of Pittsburgh School of Medicine

3500 Terrace Street, S325BST

Pittsburgh, PA 15261

Tel: (412) 648-9711

Fax: (412) 648-9095

e-mail: traub+@pitt.edu

Copyright 2001 by The American Society for Biochemistry and Molecular Biology, Inc.

JBC Papers in Press. Published on September 27, 2001 as Manuscript M108177200 by guest on A

pril 13, 2019http://w

ww

.jbc.org/D

ownloaded from

2

Running title: HIP1 binds core endocytic components

Abbreviations

BSA, bovine serum albumin

ENTH, epsin N-terminal homology

GSH, glutathione

GST, glutathione S-transferase

HC, heavy chain

HIP1, huntingtin-interacting protein 1

HIP1R, huntingtin-interacting protein 1 related protein

LC, light chain

PtdIns(4,5)P2, phosphatidylinositol(4,5) bisphosphate

SDS-PAGE, SDS-polyacrylamide gel electrophoresis

WT, wild type

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

3

Clathrin-mediated endocytosis is a major pathway for the internalization of

macromolecules into the cytoplasm of eukaryotic cells. The principle coat

components, clathrin and the AP-2 adaptor complex, assemble a polyhedral

lattice at plasma membrane bud sites with the aid of several endocytic accessory

proteins. Here, we show that huntingtin-interacting protein 1 (HIP1), a binding

partner of huntingtin, co-purifies with brain clathrin-coated vesicles and

associates directly with both AP-2 and clathrin. The discrete interaction

sequences within HIP1 that facilitate binding are analogous to motifs present in

other accessory proteins, including AP180, amphiphysin and epsin. Bound to a

phosphoinositide-containing membrane surface via an ENTH domain, HIP1

associates with AP-2 to provide coincident clathrin-binding sites that together

efficiently recruit clathrin to the bilayer. Our data implicate HIP1 in endocytosis

and the similar modular architecture and function of HIP1, epsin and AP180

suggest a common role in lipid-regulated clathrin lattice biogenesis.

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

4

Endocytosis entails the preferential recruitment of select molecules into a patch

of plasma membrane that will bud into the cytoplasm. At the bud site, an

assembling clathrin coat links the mechanical process of invagination to cargo

selection. The core endocytic components, clathrin trimers and the AP-2

heterotetramer, as well as lipids (1) and multiple protein cofactors, collectively

termed endocytic accessory proteins (2), participate in bud nucleation, lattice

assembly and invagination, and in the final scission event (1-3). The precise role

of many of the accessory proteins remains poorly understood however. One

common feature of several of the accessory proteins is the capacity to bind to

both AP-2 and clathrin (4-13). Associations with the AP-2 adaptor complex

generally involve the independently folded appendage domains of the large α

(αA or αC isoform) and β2 subunits, each separated from the heterotetrameric

adaptor core by a flexible hinge. Despite only ~10% sequence identity, the fold of

the β2 appendage (14) is structurally analogous to that of the α appendage

(15,16), and, indeed, the αC and β2 appendages interact with an overlapping

group of partner proteins (14).

The privileged association of accessory factors with the core endocytic machinery

is due to discrete interaction sequences located within each accessory protein. In

epsin and eps15, the AP-2 αC-binding sequence is Asp-Pro-Trp (DPW) and Asp-

Pro-Phe (DPF) respectively, each protein bearing multiple triplets arrayed in a

tandem fashion. The minimal region of amphiphysin I (12) or AP180 (8) required

to bind the αC appendage does not contain a DPF/W sequence however. Instead,

in these proteins, and the long-splice isoform of phosphoinositide

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

5

polyphosphatase synaptojanin, SJ170, an alternate binding motif is based upon

the di-aromatic consensus sequence FXDXF (where X is any amino acid)1.

Several of the designated AP-2–binding accessory proteins also interact directly

with clathrin. The most prevalent binding motif that facilitates clathrin

association is based upon the consensus L[L,I][D,E,N][L,F][D,E] (6,17,18), the so-

called clathrin box (19). Variations on this type I consensus are found adjacent to

the AP-2 binding sequences in amphiphysin (6,12), epsin (7,9,10), AP180 (13,20)

and β-arrestin (21). The extended type I clathrin-binding sequence interacts with

an elongated shallow cleft in the globular amino-terminal domain of the clathrin

heavy chain (19). Here, we identify an additional protein, huntingtin-interacting

protein 1 (HIP1), which displays canonical AP-2- and clathrin-binding motifs as

well as overall domain organization and functional properties akin to epsin,

AP180 and amphiphysin. Our data strongly implicate HIP1 in clathrin-mediated

endocytosis.

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

6

EXPERIMENTAL PROCEDURES

Construct preparation—Glutathione S-transferase (GST)-HIP1M1 (residues

311–394) and M2 (residues 311–369) were prepared using appropriate PCR

primers and human EST xd79f07.x1 (Research Genetics) as the template. The

inserts were ligated, after digestion, into EcoRI/XhoI-cleaved pGEX-4T-1. A

similar PCR-based strategy was used to insert the human HIP1M1 segment into

pcDNA3.1 with an amino-terminal myc-epitope tag. The GST-mHIP1 (1-533)

fusion construct was prepared to contain in-frame insertion of nucleotide 1–1599

of mouse HIP1. GST-SJ170C2 (residues 1454–1530) and GST-epsin 1 (1–407) were

prepared similarly using human EST nf51b08.s1 (Incyte Genomics) and rat epsin

1 cDNA, kindly provided by Pietro De Camilli, respectively. Mutations were

generated using appropriate mutagenic primers with the QuikChange kit

(Stratagene) as described elsewhere (10,16). All constructs and mutations were

confirmed by automated dideoxynucleotide sequencing.

Cytosol and Protein purification—Cytosol was prepared from frozen rat brains

exactly as described previously (22). The rat brain detergent extract was prepared

as described elsewhere (12) and filtered into assay buffer (25 mM Hepes-KOH,

pH 7.2, 125 mM potassium acetate, 5 mM magnesium acetate, 2 mM EDTA, 2

mM EGTA, 1 mM DTT) containing 0.5% Triton X-100 before use. GST-fusion

proteins were produced in E. coli using a standard IPTG induction protocol

(10,16). After lysis in B-PER (Pierce), soluble protein was purified on glutathione

(GSH) Sepharose, eluted with 25 mM Tris-HCl, pH 8.0, 10 mM GSH, 5 mM DTT

and then dialyzed into PBS, 1 mM DTT. Human HIP1M1 (residues 311-369),

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

7

HIP1M2 (residues 311–369) and epsin DPW (residues 249-407) were cleaved

from GST with thrombin, followed by addition of the irreversible thrombin

inhibitor PPACK (Calbiochem) to a final concentration of 25 µM. For the GST-

mHIP1 (1-533), the purified recombinant protein was first dialyzed to remove

GSH and then treated with thrombin to cleave off the HIP1 segment. Liberated

GST was removed by adding a second aliquot of GSH-Sepharose and collecting

the unbound fraction as mHIP1 (1-533). Clathrin and AP-2 were purified from rat

brain clathrin-coated vesicles by Tris extraction followed by sequential

chromatography over Superose 6 and hydroxylapatite (23). Pooled clathrin- or

AP-2-containing fractions were gel-filtered into assay buffer and centrifuged at

134,000 Xgmax before use in binding assays.

Binding assays—For pull-down-type assays, GST or GST- fusion proteins were

first immobilized on GSH Sepharose, washed in assay buffer and then mixed

with clarified rat brain cytosol/detergent extract to give a final concentration of

~7.5 mg/ml in 300 µl total volume (10,24). After incubation at 4°C for 60 min, the

beads were separated by centrifugation and aliquots corresponding to 1/80 of

each supernatant and 1/8 of each washed pellet were analyzed by SDS-PAGE

and immunoblotting exactly as described (10,16,24).

The two-stage liposome binding assays used synthetic liposomes composed of

10% (w/w) cholesterol, 30% (w/w) phosphatidylethanolamine, 30% (w/w)

phosphatidylcholine and 30% (w/w) phosphoinositides (Sigma) prepared as

described elsewhere (25). Aliquots of 10 µg of mHIP1 (1-533), GST-epsin (1-407),

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

8

AP-2, or combinations thereof, were first preincubated with 0.5 mg/ml

liposomes for 60 min at 4°C. After centrifugation at 20,000 Xgmax for 15 min, the

lipid pellets were resuspended in assay buffer, then 5 µg clathrin added and the

volume adjusted to 200 µl. Both stages contained 0.1 mg/ml BSA as a protein

carrier. After a further 60 min incubation at 4°C, the liposomes were again

centrifuged and the pellets resuspended in SDS-sample buffer. Aliquots

corresponding to 1/25 of each supernatant and 1/4 of each pellet were analyzed

by SDS-PAGE.

Transient transfections—COS-7 cells were grown at 37°C in DMEM supplemented

with 10% fetal calf serum and 2 mM L-glutamine. Cells plated on poly-L-lysine-

coated round glass coverslips were transfected with DEAE dextran and, after 48

hours, incubated in serum-free DMEM for 60 min. Biotinylated human

transferrin (25 µg/ml) was then added and incubation continued for 15 min at

37°C. Cells were fixed with 3.7% formaldehyde and processed for

immunofluorescence microscopy as described (10).

Antibodies—Polyclonal antibodies directed against HIP1 were generated by

immunizing rabbits with the GST-HIP1M1 fusion protein. Anti-HIP1 antibodies

were affinity purified on thrombin-cleaved HIP1M1 coupled to CNBr-activated

Sepharose 4B using standard procedures. The sources of the antibodies against

clathrin, the AP-2 subunits, epsin, amphiphysin, AP180 and eps15 have been

described (10,16,24). Anti-myc mAb 9E10 was from BAbCo and the anti-

synaptotagmin I mAb was purchased from Transduction Laboratories.

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

9

RESULTS

Searching the databases for additional proteins that contain the AP-2 binding

FXDXF motif we identified HIP1. This protein was initially discovered in screens

for binding partners of huntingtin (26,27), the product of the gene that undergoes

pathogenic CAG-codon expansion in Huntington’s disease. HIP1 displays a

modular domain architecture: an epsin N-terminal homology (ENTH) domain

precedes a central region predicted to have a high propensity for coiled-coil

formation, followed by a C-terminal talin-homology I/LWEQ domain (Fig. 1A).

Within the central region of the protein, before the coiled-coil segment, a stretch

of three interwoven FX[N/D/S]X[F/L] motifs proceeds a single DPF triplet (Fig.

1B). Significantly, this region of HIP1 is quite divergent from HIP1R (Fig. 1B), a

related protein roughly 50% identical to HIP1 that localizes to clathrin-coated

pits (28). The presence of these motifs within HIP1 is intriguing as the

intracellular staining pattern of HIP1 is highly reminiscent of AP-2 and clathrin

(29), and the Saccharomyces cerevisiae HIP1/HIP1R orthologue, Sla2p/End4p, is

implicated in endocytic control (30,31). In fact, we find that in rat brain extracts,

HIP1 copurifies with clathrin-coated vesicles (Fig. 2, lane d). Affinity purified

anti-HIP1 antibodies detect a major ~120-kDa polypeptide with a distribution

that, on subcellular fractionation, clearly parallels that of both clathrin and the

AP-2 adaptor complex (Fig. 2). Like HIP1, a major fraction of AP180 is also

recovered in the clathrin-coated vesicle fraction also containing the recycling

synaptic vesicle transmembrane protein synaptotagmin I (right panel, lane d)

(5,23). By contrast, although present, other accessory proteins, including epsin 1

and amphiphysin are not correspondingly enriched within the purified coated

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

10

vesicle fraction (lane c compared to lane d) and substantial cytosolic pools of

these proteins exist (lane c) (5). These results show that HIP1 is significantly

enriched in clathrin-coated endocytic vesicles found in brain.

HIP1 associates directly with the AP-2 adaptor—To assess whether the region

harboring the FXDXF/DPF-type sequences does actually facilitate an interaction

with AP-2, brain cytosol was incubated with GST-fusion proteins containing this

region of human HIP1. The HIP1M1- and M2-fusion proteins (see Fig. 1A) both

pull down AP-2 near quantitatively (Fig. 3A, lane f and h), whereas GST alone

fails to interact with the adaptor complex (lane b), which remains in the

supernatant (lane a). The presence of the µ2 adaptor subunit verifies that

heterotetrameric AP-2 adaptor complexes bind and, in these assays, the extent of

AP-2 association with the HIP1 fusions is similar to that seen with the GST-

SJ170C2 fusion (lane d), which contains residues 1454-1530 of human SJ170

harboring the sequence 1462GFKDSF.

Importantly, the soluble, native HIP1 protein also binds to the isolated αC-

appendage domain of AP-2. When incubated with brain cytosol, GST-αC binds to

AP180, epsin, (Fig. 3B, lane f), eps15, and amphiphysin I and II (see Fig. 3C)

(15,16) but, as there is very little HIP1 in cytosolic extracts (Fig. 2), only low level

HIP1 association is observed (Fig. 3B, right panel, lane f). HIP1 is more abundant

in a rat brain detergent extract (12) (Fig. 3B, lane c) and, after incubation with this

extract, a prominent additional ~120-kDa band is recovered together with the

common binding partners on the GST-αC beads (Fig. 2B, left panel, lane h

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

11

compared to lane f). This major polypeptide is confirmed to be HIP1 by

immunoblotting (right panel, lane h). The substantially enhanced recovery of

native HIP1 on the GST-αC beads without a concomitant increase in the other

appendage partners again argues that the HIP1–AP-2 interaction is direct.

To confirm that HIP1 engages the same binding surface on the αC appendage

utilized by other endocytic accessory proteins, we tested the capacity of the AP-

2-binding region of HIP1 to inhibit the association of other cytosolic accessory

proteins with the immobilized GST-αC appendage. As a control, supplementing

brain cytosol with 20 µM epsin 1 DPW domain (which contains eight tandemly

arrayed DPW triplets) completely abrogates amphiphysin and AP180 binding

and only low level epsin and eps15 association remains (Fig. 3C, lane f) (15).

Since the added DPW domain abolishes associations when bound

substoichiometrically to the αC appendage (left panel, lane f), a single DPW

domain is capable of engaging multiple appendages simultaneously. By contrast,

addition of 20 µM HIP1M2 affects neither epsin nor eps15 binding (lane h).

Instead, this concentration of the HIP1 fragment totally inhibits AP180 binding

and partially interferes with the amphiphysin interaction (lane h). Thus, the HIP1

fragment inhibits the αC partner associations in the reverse order of their

apparent affinity for GST-αC (epsin ≈ eps15 > amphiphysin > AP180)(16); the

cooperative effect of the multiple DPF/W triplets in eps15 and epsin resists

efficient HIP1 competition under these conditions2.

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

12

DPF and FXDXF sequences both bind directly to the αC appendage1. To gain

some insight into the relative contribution of these two sequences within HIP1 to

AP-2 binding, the DPF triplet was altered to KPS using site-directed

mutagenesis. This substitution severely compromises the ability of the GST-

HIP1M2 fusion to associate with soluble AP-2 (Fig. 3A, lane i and j) but clearly

does not abolish adaptor binding totally. We conclude from these experiments

that HIP1 binds to AP-2 directly by engaging the common binding surface on the

platform subdomain of the αC appendage and that the distal DPF sequence

appears to be the dominant ligand. We attribute the remaining adaptor binding

observed with the DPF mutant to the proximal FXDXF motif. Significantly, all of

the known accessory proteins containing the FXDXF motif (amphiphysin, AP180,

SJ170, HIP1) have one or more DXF/W triplets located adjacent to this sequence.

HIP1 also binds to clathrin trimers directly— In addition to binding to AP-2

adaptors, the HIP1–GST-fusion proteins also affinity purify a prominent ~180-

kDa polypeptide (Fig. 3A, left panel, lane f and h). Antibodies identify this

protein as the clathrin heavy chain and the cognate light chains, although not

visible on the stained gel, are also detected on the blots3. It is well established

that the hinge and appendage regions of the AP-2 β2-subunit interact with

clathrin directly (14,32), yet the observed association of GST-HIP1 with clathrin is

not simply due to the presence of bound AP-2. No clathrin associates with the

GST-SJ170C2 fusion protein, despite it binding similar amounts of AP-2 (lane d).

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

13

Closer inspection of the HIP1 sequence reveals a putative clathrin box proximal

to the AP-2-binding elements (Fig. 1B). The sequence, LMDMD, is an unusual

variation on the type I consensus L[L,I][D,E,N][L,F][D,E] (32) which binds to a

shallow groove located between blades 1 and 2 of the clathrin heavy chain β-

propeller (19). The role of this region in clathrin binding was probed by altering

the sequence to AAAMD (LMD→AAA; Fig. 4). Compared to the native sequence

(lane d), this mutation abolishes clathrin binding completely but leaves the AP-2

interaction intact (lane f). This result implicates the LMDMD in binding clathrin

directly but, interestingly, in the GST-HIP1M2 (DPF→KPS) mutant, which still

contains an intact clathrin box, clathrin binding is strongly diminished (Fig. 3A,

lane j). These results are identical to the behavior of the clathrin-binding

sequence 257LMDLADV located within the DPW domain of rat epsin 1, proximal

to the tandemly arrayed eight DPW repeats (10). There, again, robust clathrin

binding is dependent upon AP-2 recruitment, providing coincident binding

motifs for the terminal domain of the clathrin heavy chain. Transient transfection

of a myc-tagged HIP1M1 segment (residues 311-394) into COS-7 cells, like the

AP-2/clathrin-binding DPW domain of epsin 1 (5), inhibits endocytic transferrin

uptake (Fig. 5), although complete inhibition is only seen in less than a third of

all transfected cells.

HIP1 as an endocytic accessory protein—HIP1 also resembles endocytic accessory

proteins in other respects (Fig. 1A). Like amphiphysin and epsin, there is a

unique functional module at the carboxyl terminus; in HIP1 the talin-like

I/LWEQ domain. And like epsin and AP180, sequence analysis algorithms

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

14

predict HIP1 to have ENTH domain. Importantly, the critical Lys and His

residues in AP180 required to electrostatically coordinate the negatively charged

phosphatidylinositol(4,5)bisphosphate (PtdIns(4,5)P2) headgroup (13) are

conserved in both HIP1 and HIP1R (Fig. 6A) and indeed murine HIP1 (1-533)

binds to polyphosphoinositides (see below).

To test the hypothesis that one general function of accessory proteins with this

type of domain architecture is to cooperate with AP-2 to drive clathrin lattice

assembly, we assayed clathrin recruitment onto synthetic liposomes containing

PtdIns(4,5)P2. The ENTH domain allows the amino-terminal segments of mouse

HIP1 (residues 1-533) (Fig. 6B, lane d) and epsin 1 (33) (GST-epsin 1 (1-407), lane

j) each to bind directly to phosphoinositide-containing liposomes. AP-2 (lane f)

also associates with the liposomes directly (13), due to a PtdIns(4,5)P2-binding

determinant found at the amino terminus of the α subunit (34). Attached to the

liposome surface, each of these proteins is able to interact with and recruit

soluble clathrin (lane d, f and j), while no clathrin sediments with the liposomes

in their absence (lane b). Mixing either the HIP1 or epsin protein together with

AP-2 increases the recovery of the adaptor complex with the liposome (lane h

and l), validating the interaction surfaces mapped using the affinity interaction

assays with GST (Fig. 3). Importantly, soluble clathrin recruitment onto the

liposome surface is markedly more efficient in the presence of both AP-2 and an

AP-2– and clathrin-binding accessory protein (HIP1, lane h and epsin 1, lane l

compared to lanes d, f and j). These results resemble very closely data obtained

recently using AP180 and AP-2 to assemble clathrin lattices upon synthetic lipid

membranes (13). Thus, several membrane-bound accessory proteins can similarly

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

15

cooperate with AP-2 to effect efficient clathrin assembly. The close similarity in

the behavior of HIP1 and epsin/AP180 in these assays clearly includes HIP1

within the family of endocytic accessory proteins.

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

16

DISCUSSION

Huntington disease is a neurodegenerative disorder principally affecting striatial

neurons, yet the mutated gene product, huntingtin, is not a brain-restricted

protein. Both mRNA and protein is widely expressed in mammalian tissues

(35,36). Broad tissue distribution is true also of both HIP1 and HIP1R (28,29),

hinting at a more general function for these proteins. The intracellular

distribution of huntingtin overlaps partly with clathrin (37,38) and a direct

association between huntingtin and the αC subunit of AP-2 has been reported

(39). HIP1R colocalizes more precisely with clathrin and with AP-2 in several cell

types (28) and the link between the HIP protein family and endocytosis is

validated by Sla2p, the S. cerevisiae orthologue of HIP1/HIP1R. Sla2p regulates

endocytosis and stabilizes actin organization (30,31,40). Like HIP1, HIP1R is also

enriched in brain clathrin-coated vesicles (28) but the region of HIP1 we identify

here to interact directly with the core endocytic machinery is notably absent from

HIP1R (Fig. 1A). The conserved sequence LFDQTF might facilitate binding to the

α appendage, but neither an adjacent LMDMD nor a DPF is present in HIP1R.

The sequence 348LIEIS in HIP1R does resemble the β-arrestin 2 type I clathrin-box

sequence 374LIEFE and could possibly confer on HIP1R the capacity to bind to the

clathrin heavy chain β-propeller. Alternatively, as the central coiled-coil domain

found in both proteins likely mediates dimerization (31,40) and, as there is some

evidence for an interaction between HIP1 and HIP1R (29), heterodimers of HIP1

and HIP1R, like amphiphysin I and II, may exist.

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

17

The precise role of the AP-2 and clathrin binding sequences imbedded within

many endocytic accessory proteins is a matter of debate (2,32). Instead of simply

targeting accessory proteins to an existing clathrin bud site, our findings and

others (13) indicate rather that when linked to a lipid-binding module, tandemly

arrayed AP-2- and clathrin-binding sequences could play a more critical role in

clathrin-coat biogenesis. Because accessory proteins like AP180, epsin and HIP1

bind to clathrin directly in the absence of AP-2, these proteins could initiate

lattice assembly in a phosphoinositide-dependent fashion. S. cerevisiae epsin

orthologues Ent1p and Ent2p both have essential ENTH domains and clathrin-

binding motifs (41), so the coat promoting properties of the accessory proteins

could explain the surprising presence of clathrin-coated vesicles and the viability

of yeast strains engineered to lack all functional adaptor complexes and AP180

(42,43). Clustered recruitment of AP-2 together with or augmented by AP180,

epsin and/or HIP1 will provide a network of clathrin-binding sequences for local

amplification of clathrin lattice assembly. By linking physiological clathrin

recruitment and assembly to phospholipid metabolism, these accessory proteins

allow for precise control of the intracellular location of bud assembly and also

link the process to multiple regulatory inputs. Strikingly, while amphiphysin

also displays centrally located AP-2 and clathrin binding sequences (6,12,24)

analogous to epsin, AP180 and HIP1, an ENTH domain is not present in this

protein. Nevertheless, the N-terminal segment (the BAR domain; Bin1,

amphiphysin, Rvs161/167) of amphiphysin I displays inherent phospholipid

binding properties (44).

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

18

The diversity of lipid-dependent coat-assembly regulators might possibly

facilitate the assembly of functionally different clathrin coats at the cell surface

(45) or impart discrete functions to a single assembling lattice. The carboxyl-

terminal I/LWEQ domain of HIP1 is 70% identical to HIP1R and 34% identical to

Sla2p, proteins known to bind to actin filaments (28,46), so it seems probable that

HIP1 also associates directly with assembled actin. In yeast, a temperature

sensitive mutant allele of Sla2p (end4-1) was identified in a screen for endocytic

mutants (30) but the I/LWEQ domain does not appear necessary to sustain

endocytosis (31). This is in line with our observation that the coat assembly

properties of mammalian HIP1 reside within the first 533 amino acids.

Nonetheless, positioned within the assembling lattice, the I/LWEQ domain in

HIP1 could link the clathrin-bud site to the adjacent actin cytoskeleton. Given the

enrichment of HIP1 in isolated clathrin coated vesicles, this interaction could

potentially explain the dramatic increase in the radial mobility of GFP-clathrin-

marked bud sites at the plasma membrane after depolymerization of the actin

cytoskeleton with jasplakinolide (47). The ability of HIP1 to bind to

phosphoinositides and actin simultaneously could also be linked to the

observation that in neurons derived from synaptojanin 1 knock-out animals,

clathrin-coated vesicles accumulate at the nerve terminal enmeshed within an

actin-like matrix (48). Irrespective, the modular arrangement of proteins like

AP180, epsin, amphiphysin and HIP1 appears to dictate related function in

clathrin-coat assembly.

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

19

Acknowledgements

We are grateful to Stuart Kornfeld and Gerry Apodaca for constructive

comments on our manuscript and to Ora Weisz for the COS-7 cells and for help

with the transfection procedure. We also thank Juan Bonifacino, Pietro De

Camilli, Reinhard Jahn and Ernst Ungewickell for providing important reagents.

This work was supported in part by NIH grants RO1 DK53249 (L.M.T.), KO8

CA76025-01 (T.S.R.) and RO1 CA82363-01A1 (T.S.R.). T.S.R. is currently

supported by the Cancer Research Fund of the Damon Runyon-Walter Winchell

foundation Award, DRS-22.

Footnotes

1. T.J. Brett, L.M. Traub and D.H. Fremont, manuscript submitted for publication.

2. If the density of the GST-αC appendage immobilized on GSH Sepharose is

reduced 5–10 fold, increasing the relative spacing of the individual appendages,

then the FXDXF-bearing SJ170C2 fragment is able to inhibit soluble eps15 and

epsin binding, T.J. Brett, D.H Fremont and L.M. Traub, unpublished

observations. This supports the notion that the cooperativity of the tandemly

arrayed DPF/W triplets in eps15 and epsin prevents efficient competition by the

HIP1M1 segment.

3. . While cytosolic clathrin binds avidly to the HIP1 GST fusions, we note that

HIP1M1 reproducibly interacts more efficiently with soluble trimers than the

HIP1M2 segment.

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

20

REFERENCES

1. Cremona, O., and De Camilli, P. (2001) J. Cell Sci. 114, 1041-1052.

2. Slepnev, V. I., and De Camilli, P. (2000) Nat. Rev. Neurosci. 1, 161-172

3. Marsh, M., and McMahon, H. T. (1999) Science 285, 215-220

4. McMahon, H. T., Wigge, P., and Smith, C. (1997) FEBS Lett. 413, 319-322

5. Chen, H., Fre, S., Slepnev, V. I., Capua, M. R., Takei, K., Butler, M. H., Di

Fiore, P. P., and De Camilli, P. (1998) Nature 394, 793-797

6. Ramjaun, A. R., and McPherson, P. S. (1998) J. Neurochem. 70, 2369-2376

7. Yamabhai, M., Hoffman, N. G., Hardison, N. L., McPherson, P. S.,

Castagnoli, L., Cesareni, G., and Kay, B. K. (1998) J. Biol. Chem. 273, 31401-

31407

8. Hao, W., Luo, Z., Zheng, L., Prasad, K., and Lafer, E. M. (1999) J. Biol.

Chem. 274, 22785-22794

9. Rosenthal, J. A., Chen, H., Slepnev, V. I., Pellegrini, L., Salcini, A. E., Di

Fiore, P. P., and De Camilli, P. (1999) J. Biol. Chem. 274, 33959-33965

10. Drake, M. T., Downs, M. A., and Traub, L. M. (2000) J. Biol. Chem. 275,

6479-6489

11. Haffner, C., Di Paolo, G., Rosenthal, J. A., and De Camilli, P. (2000) Curr.

Biol. 10, 471-474

12. Slepnev, V. I., Ochoa, G. C., Butler, M. H., and De Camilli, P. (2000) J. Biol.

Chem. 275, 17583-17589

13. Ford, M. G., Pearse, B. M., Higgins, M. K., Vallis, Y., Owen, D. J., Gibson,

A., Hopkins, C. R., Evans, P. R., and McMahon, H. T. (2001) Science 291,

1051-1055.

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

21

14. Owen, D. J., Vallis, Y., Pearse, B. M., McMahon, H. T., and Evans, P. R.

(2000) EMBO J. 19, 4216-4227

15. Owen, D. J., Vallis, Y., Noble, M. E., Hunter, J. B., Dafforn, T. R., Evans, P.

R., and McMahon, H. T. (1999) Cell 97, 805-815

16. Traub, L. M., Downs, M. A., Westrich, J. L., and Fremont, D. H. (1999)

Proc. Natl. Acad. Sci. USA 96, 8907-8912

17. Goodman, O. B., Jr., Krupnick, J. G., Gurevich, V. V., Benovic, J. L., and

Keen, J. H. (1997) J. Biol. Chem. 272, 15017-15022

18. Dell'Angelica, E. C., Klumperman, J., Stoorvogel, W., and Bonifacino, J. S.

(1998) Science 280, 431-434

19. ter Haar, E., Harrison, S. C., and Kirchhausen, T. (2000) Proc. Natl. Acad.

Sci. U S A 97, 1096-1100

20. Morgan, J. R., Prasad, K., Hao, W., Augustine, G. J., and Lafer, E. M. (2000)

J. Neurosci. 20, 8667-8676

21. Krupnick, J. G., Goodman, O. B., Jr., Keen, J. H., and Benovic, J. L. (1997) J.

Biol. Chem. 272, 15011-15016

22. Traub, L. M., Bannykh, S. I., Rodel, J. E., Aridor, M., Balch, W. E., and

Kornfeld, S. (1996) J. Cell Biol. 135, 1801-1804

23. Ahle, S., and Ungewickell, E. (1986) EMBO J. 5, 3143-3149

24. Drake, M. T., and Traub, L. M. (2001) J. Biol. Chem. 276, 28700-28709

25. Zhu, Y., Drake, M. T., and Kornfeld, S. (1999) Proc. Natl. Acad. Sci. USA 96,

5013-5018

26. Kalchman, M. A., Koide, H. B., McCutcheon, K., Graham, R. K., Nichol, K.,

Nishiyama, K., Kazemi-Esfarjani, P., Lynn, F. C., Wellington, C., Metzler,

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

22

M., Goldberg, Y. P., Kanazawa, I., Gietz, R. D., and Hayden, M. R. (1997)

Nat. Genet. 16, 44-53

27. Wanker, E. E., Rovira, C., Scherzinger, E., Hasenbank, R., Walter, S., Tait,

D., Colicelli, J., and Lehrach, H. (1997) Hum. Mol. Genet. 6, 487-495

28. Engqvist-Goldstein, A. E., Kessels, M. M., Chopra, V. S., Hayden, M. R.,

and Drubin, D. G. (1999) J. Cell Biol. 147, 1503-1518

29. Chopra, V. S., Metzler, M., Rasper, D. M., Engqvist-Goldstein, A. E.,

Singaraja, R., Gan, L., Fichter, K. M., McCutcheon, K., Drubin, D.,

Nicholson, D. W., and Hayden, M. R. (2000) Mamm. Genome 11, 1006-1015

30. Raths, S., Rohrer, J., Crausaz, F., and Riezman, H. (1993) J. Cell Biol. 120,

55-65

31. Wesp, A., Hicke, L., Palecek, J., Lombardi, R., Aust, T., Munn, A. L., and

Riezman, H. (1997) Mol. Biol. Cell. 8, 2291-2306

32. Dell'Angelica, E. C. (2001) Trends Cell. Biol. 11, 315-318

33. Itoh, T., Koshiba, S., Kigawa, T., Kikuchi, A., Yokoyama, S., and

Takenawa, T. (2001) Science 291, 1047-1051

34. Gaidarov, I., Chen, Q., Falck, J. R., Reddy, K. K., and Keen, J. H. (1996) J.

Biol. Chem. 271, 20922-20929

35. Sharp, A. H., Loev, S. J., Schilling, G., Li, S. H., Li, X. J., Bao, J., Wagster, M.

V., Kotzuk, J. A., Steiner, J. P., Lo, A., and et al. (1995) Neuron 14, 1065-1074

36. Trottier, Y., Devys, D., Imbert, G., Saudou, F., An, I., Lutz, Y., Weber, C.,

Agid, Y., Hirsch, E. C., and Mandel, J. L. (1995) Nat. Genet. 10, 104-110

37. Velier, J., Kim, M., Schwarz, C., Kim, T. W., Sapp, E., Chase, K., Aronin,

N., and DiFiglia, M. (1998) Exp. Neurol. 152, 34-40

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

23

38. Kim, M., Velier, J., Chase, K., Laforet, G., Kalchman, M. A., Hayden, M. R.,

Won, L., Heller, A., Aronin, N., and Difiglia, M. (1999) Neuroscience 89,

1159-1167

39. Faber, P. W., Barnes, G. T., Srinidhi, J., Chen, J., Gusella, J. F., and

MacDonald, M. E. (1998) Hum. Mol. Genet. 7, 1463-1474

40. Yang, S., Cope, M. J., and Drubin, D. G. (1999) Mol. Biol. Cell 10, 2265-2283

41. Wendland, B., Steece, K. E., and Emr, S. D. (1999) EMBO J. 18, 4383-4393

42. Huang, K. M., D'Hondt, K., Riezman, H., and Lemmon, S. K. (1999) EMBO

J. 18, 3897-3908

43. Yeung, B. G., Phan, H. L., and Payne, G. S. (1999) Mol. Biol. Cell. 10, 3643-

3659

44. Takei, K., Slepnev, V. I., Haucke, V., and De Camilli, P. (1999) Nat. Cell

Biol. 1, 33-39

45. Di Fiore, P. P., and De Camilli, P. (2001) Cell 106, 1-4

46. McCann, R. O., and Craig, S. W. (1999) Biochem. Biophys. Res. Commun. 266,

135-140.

47. Gaidarov, I., Santini, F., Warren, R. A., and Keen, J. H. (1999) Nat. Cell Biol.

1, 1-7

48. Cremona, O., Di Paolo, G., Wenk, M. R., Luthi, A., Kim, W. T., Takei, K.,

Daniell, L., Nemoto, Y., Shears, S. B., Flavell, R. A., McCormick, D. A., and

De Camilli, P. (1999) Cell 99, 179-188

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

24

Figure legends

Fig. 1. HIP1 structure and sequence alignment.

A. Domain organization of human (Hs) HIP1. The location of the FXDXF (red),

DPF (cyan) and clathrin-binding (green) sequences is indicated with vertical bars

while the regions corresponding to the GST-HIP1-fusion proteins used in this

study are shown below.

B. Local sequence alignment of human and mouse (Mm) HIP1 and HIP1R

sequences. Identical residues are colored pink and conservative

substitutions yellow. The locations of the FXDXF (red), DPF (cyan) and LMDMD

(green) motifs in HIP1 are indicated above.

Fig. 2. HIP1 is a component of clathrin-coated vesicles.

Aliquots of 20 µg each of rat brain homogenate, crude microsomes, cytosol and

purified clathrin-coated vesicles were fractionated by SDS-PAGE and either

stained with Coomassie blue (left panel) or transferred to nitrocellulose (right

panels) and probed with antibodies directed against the clathrin light chains

(LC), the µ2 subunit of the AP-2 complex, HIP1, AP180, epsin 1, amphiphysin, or

synaptotagmin I. The position of the molecular mass standards (in kDa) and the

~180-kDa clathrin heavy chain (HC) (left panel) are indicated on the left and

right respectively. Only the relevant portion of each blot (right panels) is shown.

Proteolysis of the clathrin light chains in the coated vesicle preparation suggests

the lower ~80-kDa band detected by the anti-HIP1 antibodies is likely a

degradation product.

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

25

Fig. 3. HIP1 interacts with the core endocytic machinery.

A. Immobilized GST (lane a and b) or GST-SJ170C2 (lane c and d), GST-HIP1M1

(lane e and f), GST-HIP1M2 (lane g and h) or GST-HIP1M2 (DPF→KPS) (lane i

and j) was incubated with rat brain cytosol at 4°C for 60 min. After

centrifugation, aliquots of 1/80 of each supernatant (S) and 1/8 of each washed

pellet (P) were resolved by SDS-PAGE and either stained with Coomassie blue

(left panel) or immunoblotted (right panels) with antibodies directed against the

AP-2 α- or µ2 subunit or the clathrin heavy (HC) or light chain (LC).

B. Immobilized GST (lane a-d) or GST-αC appendage (lane e-h) were incubated

with either rat brain cytosol (lane a, b, e and f) or a rat brain detergent extract

(lane c, d, g and h). After centrifugation, aliquots of 1/80 of each supernatant (S)

and 1/8 of each washed pellet (P) were resolved by SDS-PAGE and either

stained with Coomassie blue (left panel) or immunoblotted (right panels) with

antibodies directed against HIP1, AP180 or epsin 1.

C. Immobilized GST (lane a and b) or GST-αC appendage (lane c-h) were

incubated with rat brain cytosol alone (lane a-d) or supplemented with 20 µM

epsin 1 DPW domain (lane e and f) or 20 µM HIP1M2 fragment (lane g and h).

After centrifugation, aliquots of 1/80 of each supernatant (S) and 1/8 of each

washed pellet (P) were resolved by SDS-PAGE and either stained with

Coomassie blue (left panel) or immunoblotted (right panels) with antibodies

directed against epsin 1, eps15, amphiphysin or AP180. The immunoreactive

signal below the epsin band after addition of the DPW domain (right panel, lane

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

26

f) reflects strong reactivity of the antibody with this portion of epsin, to which it

was raised.

Fig. 4. A functional clathrin box within HIP1.

A. Immobilized GST (lane a and b) or GST-HIP1M1 (lane c and d) or GST-

HIP1M1 (LMD→AAA) (lane e and f) was incubated with rat brain cytosol. After

centrifugation, aliquots of 1/80 of each supernatant (S) and 1/8 of each washed

pellet (P) were resolved by SDS-PAGE and either stained with Coomassie blue

(left panel) or immunoblotted (right panels) with antibodies directed against the

AP-2 β-or µ2-subunit or the clathrin heavy (HC) or light chain (LC). Note that

using brain cytosol, the bound AP-2 complexes contain both the faster migrating

β2 subunit as well as lower amounts of the AP-1 β1 subunit (right panel, lane d

and f). The promiscuity of β subunit incorporation into AP-1 and AP-2 has been

described before (22).

Fig. 5. Effect of the overexpressed HIP1M1 segment on endocytosis.

COS-7 cells transiently expressing myc-tagged HIP1M1 for 48 hours were serum

starved for 60 min and then incubated with biotinylated transferrin for 15 min at

37°C prior to fixation. Endocytosed transferrin was visualized with streptavidin-

Alexa 594 (panel A, red) and the transfected cells identified with anti-myc mAb

9E10 and then an anti-mouse Alexa 488 conjugate (panel B, green). The merged

(panel C) image showing both endocytic inhibited (small arrows, no

accumulation of transferrin within perinuclear recycling endosomes) and

uninhibited (large arrowhead) cells is also shown.

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

27

Fig. 6. Phosphoinositide-dependent clathrin-coat assembly.

A. Sequence alignment of the amino-terminal segments of rat (Rn) AP180 and

human HIP1 and HIP1R. Amino acid conservation is colored as in Fig. 1B. The

location of the first two α helices of the AP180 ENTH domain (13) is shown

above while the basic side chains required to coordinate PtdIns(4,5)P2 are

indicated below with vertical arrows.

B. Phosphoinositide-containing liposomes were first preincubated with HIP1

(1-533) (lanes c,d and g,h), AP-2 (lanes e-h and k,l) and GST-epsin (1-407) (lanes i-

l) at 4°C for 60 min as indicated. After recovery by centrifugation, each liposome

pellet was resuspended and then incubated at 4°C for 60 min with purified

clathrin trimers. After centrifugation, aliquots of 1/25 of each supernatant (S)

and 1/4 of each pellet (P) were resolved by SDS-PAGE and stained with

Coomassie blue.

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

HIP1(1-533)HIP1M1HIP1M2

ENTH coiled coil I/LWEQ

Hs HIP1

1 1034

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

Hs HIP1

305

ISPVVVIPAEA--------------SSPDSEPVLEKDDLMDMDASQQN

Mm HIP1

308

ISPVVVIPAEV--------------SSPDSEPVLEKDDLMDMDASQQT

Hs HIP1R

299

IKPVVVIPEEAPEDEEPENLIEISTGPPAGEPVVVAD-----------

Mm HIP1R

299

IKPVVVIPEEAPEEEEPENLIEISSAPPAGEPVVVAD-----------

Hs HIP1

339

LFDNKFDDIFGSSFSSDPFNFNSQNGVNKDEKDHLIERLYRE

Mm HIP1

342

LFDNKFDDVFGSSLSSDPFNFNNQNGVNKDEKDHLIERLYRE

Hs HIP1R

336

LFDQTF---------------GPPNGSVKDDRDLQIESLKRE

Mm HIP1R

336

LFDQTF---------------GPPNGSVKDDRDLQIENLKRE

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

AB

C

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

α1

α2

Rn AP180

17

VTGSAVSKTVCKATTHEVMGPKKKHLDYLIQATNE

Hs HIP1

32

SFERTQTVSINKAINTQEVAVKEKHARTCILGTHH

Hs HIP1R

26

QFDKTQAISISKAINTQEAPVKEKHARRIILGTHH

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from

and Linton M. TraubSanjay K. Mishra, Nicole R. Agostinelli, Tom J. Brett, Ikuko Mizukami, Theodora S. Ross

endocytic accessory proteinsClathrin- and AP-2-binding sites in HIP1 uncover a general assembly role for

published online September 27, 2001J. Biol. Chem. 

  10.1074/jbc.M108177200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

by guest on April 13, 2019

http://ww

w.jbc.org/

Dow

nloaded from