1

EPIYA MOTIF IS A MEMBRANE TARGETING SIGNAL OF

HELICOBACTER PYLORI VIRULENCE FACTOR CagA IN MAMMALIAN

CELLSHideaki Higashi‡, Kazuyuki Yokoyama‡, Yumiko Fujii‡, Shumei Ren‡, Hitomi Yuasa‡, Iraj

Saadat‡, Naoko Murata-Kamiya‡, Takeshi Azuma§ and, Masanori Hatakeyama‡

From the ‡Division of Molecular Oncology, Institute for Genetic Medicine, Hokkaido University,

Sapporo 060-0815, Japan, and the §International Center for Medical Research and Treatment,

Kobe University Graduate School of Medicine, Kobe 650-0017, Japan

Running title: EPIYA motif, a membrane targeting signal of H. pylori CagA

Address correspondence to: Masanori Hatakeyama, Division of Molecular Oncology, Institute for Genetic

Medicine, Hokkaido University, Kita-15, Nishi-7, Kita-ku, Sapporo 060-0815, Japan, TEL/FAX. 81-11-

706-7544; e-mail: mhata@igm.hokudai.ac.jp

Helicobacter pylori (H. pylori) contributes to

the development of peptic ulcers and atrophic

gastritis. Furthermore, H. pylori strains

carrying the cagA gene are more virulent than

cagA-negative strains and are associated with

the development of gastric adenocarcinoma.

The cagA gene product, CagA, is translocated

into gastric epithelial cells and localizes to the

inner surface of the plasma membrane, where it

undergoes tyrosine phosphorylation at the Glu-

Pro-Ile-Tyr-Ala (EPIYA) motif. Tyrosine-

phosphorylated CagA specifically binds to and

activates SHP-2 tyrosine phosphatase at the

membrane, thereby inducing an elongated cell

shape termed the hummingbird phenotype.

Accordingly, membrane tethering of CagA is an

essential prerequisite for the pathogenic activity

of CagA. We show here that membrane

association of CagA requires the EPIYA-

containing region but is independent of EPIYA

tyrosine phosphorylation. We further show that

specific deletion of the EPIYA motif abolishes

the ability of CagA to associate with the

membrane. Conversely, reintroduction of an

EPIYA sequence into a CagA mutant that lacks

the EPIYA-containing region restores

membrane association of CagA. Thus, the

presence of a single EPIYA motif is necessary

for the membrane localization of CagA. Our

results indicate that the EPIYA motif has a dual

function in membrane association and tyrosine

phosphorylation, both of which are critically

involved in the activity of CagA to deregulate

intracellular signaling, and suggest that the

EPIYA motif is a crucial therapeutic target of

cagA-positive H. pylori infection.

Helicobacter pylori (H. pylori) is a causative

agent of gastroduodenal diseases such as atrophic

gastritis and peptic ulcers. Furthermore, chronic

infection with H. pylori in the stomach is an

important risk factor for the development of

gastric cancer and the World Health Organization

International Agency for Research on Cancer

(WHO/IARC) classified H. pylori as a definite

carcinogen in 1994 (1-6). The clinical outcome of

H. pylori infection is dependent on both host and

bacterial factors. H. pylori strains carrying the cag

gene, which localizes at one end of the cag

pathogenicity island (cagPAI), a horizontally

acquired 40 kilo base-pair DNA segment, are

associated with increased levels of inflammation

and severe atrophic gastritis (7, 8). As a

consequence, cagA-positive H. pylori strains are

more virulent than cagA-negative strains, and

infection with the cagA-positive strain greatly

increases the risk of developing gastric carcinoma

(9-13). However, the molecular mechanisms that

underlie the development of mucosal lesions

caused by cagA-positive H. pylori infection

remain largely unknown.

The CagA protein is delivered into gastric

epithelial cells by the type IV secretion system,

which is encoded by genes present in cagPAI (14-

18). Translocated CagA localizes to the inner

surface of the plasma membrane and undergoes

tyrosine phosphorylation by Src family kinases

(SFKs) (18-20). Although SFKs are cytoplasmic

JBC Papers in Press. Published on April 13, 2005 as Manuscript M503583200

Copyright 2005 by The American Society for Biochemistry and Molecular Biology, Inc.

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

2

tyrosine kinases, they are N-terminally

myristoylated and this lipid modification

facilitates the attachment of SFKs to the

membrane. Accordingly, membrane localization of

CagA may be required for CagA phosphorylation

or at least promote the process by SFKs. Tyrosine

phosphorylation sites of CagA are characterized by

a Glu-Pro-Ile-Tyr-Ala (EPIYA) motif present in its

C-terminal variable region (19-22). Tyrosine-

phosphorylated CagA specifically binds to Src

homology 2 (SH2)-containing tyrosine

phosphatase (SHP-2) or C-terminal Src kinase

(Csk) and stimulates their enzymatic activity (22-

24). Since both SHP-2 and Csk are involved in

intracellular signaling, disturbance of their

functions by CagA has been thought to play a role

in the development of gastric disorders that are

associated with cagA-positive H. pylori infection.

In vitro infection of gastric epithelial cells with

cagA-positive H. pylori strains, but not cagA-

negative strains, induces a unique morphological

change that is characterized by spreading and

elongation of cell shape (25). The cell

morphological change, termed “the hummingbird

phenotype”, is associated with cell scattering and

increased cell motility (21, 26). Induction of the

hummingbird phenotype is mediated by SHP-2,

which is deregulated through CagA-SHP-2

interaction (26). A constitutively active form of

SHP-2 can also induce the hummingbird

phenotype, but the morphological change requires

membrane association of SHP-2 (22). This

observation indicates that the plasma membrane

recruitment is essential for SHP-2 function and

suggests that a primary role of CagA in host cell is

to translocate cellular proteins such as SHP-2 from

the cytoplasm to the membrane and activate them

at the membrane. In this sense, CagA functionally

mimics mammalian scaffolding adaptor proteins,

such as IRS and Gab/Dos family members. Also,

CagA has been reported to co-localize with ZO-1

and JAM, components of the junctional complex,

and destroy tight junctions in polarized MDCK

epithelial cells in a manner independent of

tyrosine phosphorylation (27). These findings

indicate that CagA induces junctional dysfunction,

which alters epithelial cell permeability and

thereby promotes mucosal inflammation.

The above studies indicate that membrane

association of CagA is important for the

pathogenic action of CagA in gastric epithelial

cells. In this work, we investigated the CagA

structure that is involved in membrane localization

of CagA using a series of CagA mutants. We

found that the EPIYA motif, which is known to be

the site of CagA tyrosine phosphorylation, is also

responsible for the association of CagA with the

membrane. We discuss the bifunctional role of the

EPIYA motif in membrane localization and

tyrosine phosphorylation in the context of CagA

pathogenicity.

EXPERIMENTAL PROCEDURES

Antibodies—Anti-HA rabbit polyclonal

antibody (Y-11) and anti-Myc mouse monoclonal

antibody (9E10) were purchased from Santa Cruz

Biotechnology. Anti-phosphotyrosine mouse

monoclonal antibody (4G10) was purchased from

Upstate Biotechnology. Alexa Fluor 546-

conjugated anti-rabbit and anti-mouse antibodies

were purchased from Invitrogen.

Expression Vectors—Mammalian expression

vectors for HA-tagged wild-type CagA derived

from H. pylori NCTC11637 strain (WT CagA) and

its mutants, phosphorylation-resistant (PR) mutant

CagA and CagA N mutant, have been described

previously (26, 28). CagA N C and CagA

N ABCCC mutants were generated from CagA

N by deletion of amino-acid residues 1087-1247

and 869-1086, respectively. CagA N, CagA

ABCCC, CagA BCCC, CagA AB, CagA

ABC, CagA ABCC-s, CagA ABCCC-s and

CagA CCC mutants were generated by deleting

amino acid sequences 613-1247, 869-1086, 901-

1086, 869-940, 869-974, 869-1008, 869-1042 and

941-1086 of WT CagA, respectively. CagA

ABCC or CagA ACCC mutant lacks 2 regions

of WT CagA, which correspond to amino acid

sequences 869-1008 and 1043-1086 ( ABCC) or

869-900 and 941-1086 ( ACCC). CagA

BCCC[ EPIYA], CagA ACCC[ EPIYA] and

CagA ABCC[ EPIYA] were generated by

deleting five amino-acids composing EPIYA motif

from CagA BCCC, CagA ACCC and CagA

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

3

ABCC, respectively. CagA ABCCC+EPIYA

and CagA ABCCC+3EPIYA were prepared from

WT CagA by substituting amino-acid residues

869-1086 with one and three copies of the EPIYA

sequence, respectively. The DNA fragments

encoding WT CagA and its mutants were cloned

into a pSP65SR mammalian expression vector.

Constitutively active mutants of SHP-2 with

(Myr-SHP-2 SH2-Myc) or without (SHP-2 SH2-

Myc) myristoylation signal sequence from v-Src

have been described previously (22). EPIYA-SHP-

2 SH2-Myc was constructed by replacing the v-

Src-derived myristoylation signal sequence with

the Glu-Pro-Ile-Tyr-Ala (EPIYA) sequence. The

DNA fragment encoding the mutant was cloned

into a pSP65SR vector.

Cell staining—Human gastric epithelial AGS

cells were cultured in RPMI1640 medium

supplemented with 10% FCS. For transient

transfection, 160 ng of plasmids was transfected

into AGS cells (4 x 104

cells /0.8 cm2) by using 0.4

μl of Lipofectamine 2000 reagent (Invitrogen)

according to the manufacturer's protocol. The

transfected AGS cells were fixed with 3%

paraformaldehyde at 17 h post-transfection. Cells

were then treated with anti-HA (Y-11) or anti-Myc

(9E10) antibody. Primary antibody was localized

by Alexa Fluor 546-conjugated anti-rabbit or anti-

mouse antibody (Invitrogen). Images were

acquired using a confocal microscope system

(Fluoview, Olympus).

Induction of Cell morphological change and

Western blotting—Eight μg of plasmid was

transfected into AGS cells (1.2 x 105

cells / 35-mm

dish) by using 5 μl of Lipofectamine 2000 reagent.

Morphology of the AGS cells was examined at 17

h after transfection. After examination of cell

morphology, the cells were harvested, and then

lysed in lysis buffer as described previously (29).

Total cell lysates were subjected to SDS-8%

polyacrylamide gel electrophoresis (PAGE).

Proteins transferred to poly (vinylidene difluoride)

membrane filter (Millipore) were soaked in

solutions of appropriate antibody and then

visualized using Western blot chemiluminescence

reagent (PerkinElmer Life Sciences).

RESULTS

Requirement of the EPIYA-containing region for

membrane localization of CagA—To determine

the CagA region that is responsible for membrane

localization, we made a series of CagA mutants

from H. pylori NCTC11637 CagA. NCTC11637

CagA (Wild-type CagA) possesses five EPIYA

motifs in the C-terminal EPIYA-containing region

as schematically depicted in Fig. 1A. We first

generated CagA mutants that lack the N-terminal

region (CagA N), C-terminal region (CagA N) or

EPIYA-containing region (CagA ABCCC) by

internal deletion. From the CagA N mutant, we

also made additional mutant derivatives, CagA

N C and CagA N ABCCC, which lack the C-

terminus region and EPIYA-containing region,

respectively (Fig. 1A). These CagA mutants were

all C-terminal hemagglutinin (HA) epitope-tagged

and were transiently expressed in AGS human

gastric epithelial cells. Expression of these CagA

mutants was confirmed by immunoblotting with

an anti-HA antibody (Fig. 1B). Expectedly, CagA

mutants possessing the EPIYA-containing region

were efficiently phosphorylated, while those

lacking the EPIYA-repeat region were not (Fig.

1B). Wild-type CagA induces the hummingbird

phenotype in AGS cells (22). Accordingly, we

examined the ability of these CagA mutants to

induce the hummingbird phenotype. The CagA

N and CagA N C mutants were capable of

inducing elongation of cells, which was

indistinguishable from the elongation of cells

induced by wild-type CagA (Fig. 1C, left panel).

On the other hand, CagA mutants lacking the

EPIYA-containing region failed to induce

morphological changes in cells as was shown with

phosphorylation-resistant CagA (PR CagA), in

which all of the tyrosine residues in the five

EPIYA motifs were replaced by phenylalanine

residues (Fig. 1C, right panel). These results

confirm our previous results showing that CagA

induces the hummingbird phenotype in a tyrosine

phosphorylation-dependent manner (22).

We next investigated sub-cellular localization of

each CagA mutant in AGS cells by

immunostaining. We found that wild-type CagA as

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

4

well as the CagA N and CagA N C mutants,

both of which retained the ability to induce the

hummingbird phenotype, were localized to the

plasma membrane. In contrast, CagA N ABCCC,

CagA N and CagA ABCCC, all of which failed

to induce the hummingbird phenotype, were

primarily detected in the cytoplasm (Fig. 1C, right

panel). Notably, PR CagA, which does not induce

the hummingbird phenotype, still retained the

ability to localize to the membrane. From these

observations, we concluded that stable association

of CagA with the membrane requires the EPIYA-

repeat region but is independent of tyrosine

phosphorylation at the EPIYA sites.

Determination of EPIYA segments that

correspond to the genetic units of homologous

recombination—NCTC11637 wild-type CagA

contains five EPIYA motifs, which are designated

EPIYA-A, EPIYA-B and 3 x EPIYA-C sites based

on the sequence surrounding each of the EPIYA

motifs (Fig. 2A) (29). Accordingly, we subdivided

the EPIYA-containing region of CagA into the

EPIYA-A segment (residues 869-900), the EPIYA-

B segment (residues 901-940), and 3 copies of the

EPIYA-C segment, a repeatable 34-amino-acid

sequence that contains the EPIYA-C site (residues

941-974, residues 975-1008 and residues 1009-

1042) (Fig. 2B). The border of each fragment was

determined by comparing sequences of the

EPIYA-containing regions, which were made by

extensive genetic recombination among various

CagA isolates (29, 30).

Delineation of EPIYA sites involved in CagA-

membrane interaction—To elucidate which

EPIYA site is involved in membrane localization

of CagA, we generated a series of CagA mutants

that lack the EPIYA segments in various

combinations (Fig. 3A). We have already shown

that EPIYA-C is the major site of CagA tyrosine

phosphorylation in AGS cells (29). Consistently,

the levels of tyrosine phosphorylation among the

CagA mutants were proportional to the number of

EPIYA-C (Fig. 3B). As expected, CagA mutants

such as CCC, ACCC and BCCC were hardly

tyrosine-phosphorylated in AGS cells because

they do not possess the EPIYA-C site. We next

examined the ability of each EPIYA mutant to

interact with the membrane by immunostaining

(Fig. 3C). The results indicate that CagA mutants

possessing at least a single EPIYA fragment are

capable of membrane localization. However, there

was no difference among EPIYA-A, EPIYA-B

and EPIYA-C segments in their ability to confer

membrane localization upon CagA (Fig. 3C).

Hence, EPIYA-A, EPIYA-B and EPIYA-C

segments have comparable activities to confer

membrane association of CagA. We also

compared membrane association among CagA

mutants possessing different numbers of EPIYA-C

fragments (CagA AB, CagA ABC and CagA

ABCC) (Fig. 3A). Again, we did not find any

quantitative correlation between membrane

association of CagA and the number of EPIYA

sites. From these observations, we concluded that

the presence of a single EPIYA fragment is

necessary for membrane association of CagA.

EPIYA motif confers membrane localization on

CagA—To further elucidate the structural basis for

the membrane association of CagA, we compared

amino-acid sequences among EPIYA-A, EPIYA-

B and EPIYA-C segments and concluded that the

EPIYA motif is the only conserved sequence

among the EPIYA segments. Accordingly, we

suspected that the five-amino-acid sequence is

responsible for membrane association of CagA.

CagA BCCC, CagA ACCC and CagA ABCC

mutants possess a single EPIYA fragment and

localize to the membrane. To address the above

possibility, we specifically deleted the EPIYA

sequence from these CagA mutants and examined

their membrane localization in AGS cells (Fig. 4A

and B). As shown in Fig. 4C, deletion of the

EPIYA motif abolished membrane interaction of

these CagA mutants. The results indicate that the

EPIYA sequence itself is required for membrane

association of CagA. To further substantiate the

conclusion, we made CagA ABCCC+EPIYA and

CagA ABCCC+3EPIYA mutants, in which the

entire EPIYA-repeat region (amino acids 869-

1086) was replaced with a single copy and three

copies of the EPIYA sequence, respectively, and

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

5

expressed the mutants in AGS cells (Fig. 5A and

B). In contrast to CagA ABCCC, which

exclusively localized in the cytoplasm, both of the

CagA mutants exhibited membrane localization

(Fig. 5C). Again, in this case, the CagA mutants

with a single and triple EPIYA sequence exhibited

comparable levels of membrane localization,

indicating that the presence of a single EPIYA

motif is necessary for membrane association of

CagA. We concluded from these observations that

the EPIYA motif is a membrane localization

signal of H. pylori CagA in gastric epithelial cells.

Finally, we investigated whether the EPIYA

motif can specify the membrane localization of an

unrelated cytoplasmic protein. To do so, we

introduced the EPIYA sequence into a constitutive

active mutant of SHP-2, SHP-2 SH2-Myc, which

lacks the N-terminal SH2 domains and therefore

localizes exclusively in the cytoplasm and nucleus

(Fig. 6A). As has been previously reported,

addition of the myristoylation signal to the N-

terminal region of the SHP-2 mutant directed

membrane localization (22). On the other hand,

addition of the EPIYA motif to the N-terminal

region of the SHP-2 mutant did not specify

membrane association (Fig. 6B). The result

indicates that the EPIYA motif per se is not

sufficient to direct membrane localization of

unrelated proteins.

DISCUSSION

Upon translocation into gastric epithelial cells,

CagA localizes to the plasma membrane, where it

undergoes tyrosine phosphorylation at EPIYA

sites so as to generate docking sites for cellular

proteins such as SHP-2 and Csk (31). Both SHP-2

and Csk are cytoplasmic proteins and they are

activated upon membrane recruitment.

Accordingly, membrane localization is an essential

prerequisite for CagA to deregulate intracellular

signaling. CagA also associates with tight junction

proteins, such as ZO-1 and JAM, and disturbs

junctional functions (27). Again, in this case,

membrane localization is crucial for CagA to

interact with junctional complexes.

In this work, we investigated the CagA region

that is responsible for membrane association and

found that the EPIYA motif is a membrane

targeting-sequence of CagA. We confirmed the

essential role of EPIYA in membrane association

both by deletion and insertion of the EPIYA

sequence in CagA. However, we also note that the

EPIYA motif is insufficient to direct membrane

localization of unrelated proteins. The result

indicates that another CagA structure is required in

addition to the EPIYA motif to make a stable

association of CagA with the membrane. The

EPIYA motif has already been shown to be the

site of CagA tyrosine phosphorylation by Src

family kinases (19, 20). However, the results of

the present study rule out the possibility of a role

of EPIYA phosphorylation in CagA-membrane

interaction because substitution of the tyrosine

residue with the non-phosphorylatable

phenylalanine residue did not affect the ability of

CagA to associate with the membrane.

Given that virtually all of the CagA isolates

possess multiple EPIYA sites, a dual role of the

EPIYA motif in membrane integration and

tyrosine phosphorylation is intriguing. Since

presence of a single EPIYA motif has been shown

to be sufficient for the membrane association of

CagA, one of the multiple EPIYA sites may be

utilized for membrane tethering of CagA, while

others are employed for signal generation though

interacting with cellular proteins such as SHP-2.

The EPIYA sites of CagA are sub-classified into

EPIYA-A, -B, -C and -D based on the sequences

surrounding individual EPIYA motifs. A prevalent

form of CagA species isolated in Europe, North

America, Africa and Australia is the “A-B-C” type

CagA (Western CagA), whereas that in East Asia

is the “A-B-D” type CagA (East Asian CagA)

based on the composition of EPIYA sites (30, 32,

33). We previously reported that EPIYA-C and

EPIYA-D sites are major sites of tyrosine

phosphorylation in Western and East Asian CagA

species, respectively, and that SHP-2 specifically

interacts with the tyrosine-phosphorylated EPIYA-

C or EPIYA-D site (29). However, it should be

also noted that EPIYA-A and EPIYA-B sites are

conserved virtually in all CagA isolates, whereas

there are some CagA species that contain neither

an EPIYA-C site nor EPIYA-D site. The

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

6

observations indicate that EPIYA-A and EPIYA-B

sites are also important in the function of CagA,

and the present work suggests that EPIYA-A or

EPIYA-B site is primarily responsible for

membrane association of CagA.

The present work raises the possibility that the

EPIYA motif binds a cellular protein that tethers

CagA to the membrane in a manner independent of

EPIYA tyrosine phosphorylation. Notably, ZO-1, a

component of the tight junction, has been reported

to associate with CagA in polarized MDCK

epithelial cells independent of CagA

phosphorylation (27). Accordingly, a junctional

component protein such as ZO-1 may be a

potential candidate that tethers CagA to the

membrane, although we could not demonstrate a

direct complex formation between CagA and ZO-1

in cells2. In addition to this basal CagA activity, a

fraction of CagA species may have developed

additional pathological activity that disturbs

intracellular signaling in a tyrosine

phosphorylation-dependent manner through

acquisition of an EPIYA-C or EPIYA-D site. It has

also been reported that CagA is enriched in a lipid

raft through an immuno-receptor tyrosine-based

activation motif (ITAM)-like sequence present in

the EPIYA-containing region (34). Since

translocation of CagA to a lipid raft was shown to

be dependent on tyrosine phosphorylation of CagA,

the ITAM-like sequence is not likely to be

involved in the CagA-membrane interaction. Also,

c-Met HGF receptor has been reported to associate

with CagA in cells (35). The interaction was

independent of CagA tyrosine phosphorylation,

although c-Met phosphorylation, which is caused

by HGF treatment or H. pylori infection, was

required. However, we detected neither tyrosine

phosphorylation of c-Met nor physical interaction

between CagA and c-Met in cagA-transfected

AGS cells, arguing against the involvement of c-

Met in membrane association of CagA3. There

remains the possibility that membrane association

of CagA involves direct interaction of the EPIYA

motif with lipid components of the membrane,

although we do not have any evidence that

supports this idea.

Further elucidation of molecular mechanisms

involved in membrane association of CagA will

not only improve our understanding of cagA-

positive H. pylori-induced gastroduodenal

disorders but also facilitate identification of

potential therapeutic targets. Specific inhibition of

EPIYA-mediated membrane association of CagA

should prevent development of gastric

pathological changes caused by cagA-positive H.

pylori infection, which eventually lead to gastric

adenocarcinoma.

REFERENCES

1. Parsonnet, J., Friedman, G. D., Orentreich, N., and Vogelman, H. (1997) Gut 40, 297-301

2. Huang, J. Q., Sridhar, S., Chen, Y., and Hunt, R. H. (1998) Gastroenterology 114, 1169-1179

3. Watanabe, T., Tada, M., Nagai, H., Sasaki, S., and Nakao, M. (1998) Gastroenterology 115, 642-648

4. Honda, S., Fujioka, T., Tokieda, M., Satoh, R., Nishizono, A. & Nasu, M. (1998) Cancer Res. 58,

4255-4259

5. Danesh, J. (1999) Aliment. Pharmacol. Ther. 13, 851-856

6. Uemura, N., Okamoto, S., Yamamoto, S., Matsumura, N., Yamaguchi, S., Yamakido, M., Taniyama,

K., Sasaki, N., and Schlemper, R. J. (2001) N. Engl. J. Med. 345, 784-789

7. Censini, S., Lange, C., Xiang, Z., Crabtree, J. E., Ghiara, P., Borodovsky, M., Rappuoli, R., and

Covacci, A. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 14648-14653

8. Akopyants, N. S., Clifton, S. W., Kersulyte, D., Crabtree, J. E., Youree, B. E., Reece, C. A., Bukanov,

N. O., Drazek, E. S., Roe, B. A., and Berg, D. E. (1998) Mol. Microbiol. 28, 37-53

9. Covacci, A., Censini, S., Bugnoli, M., Petracca, R., Burroni, D., Macchia, G., Massone, A., Papini, E.,

Xiang, Z., Figura, N., and Rappuoli, R. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 5791-5795

10. Blaser, M. J., Perez-Perez, G. I., Kleanthous, H., Cover, T. L., Peek, R. M., Chyou, P. H.,

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

7

Stemmermann, G. N., and Nomura, A. (1995) Cancer Res. 55, 2111–2115

11. Kuipers, E. J., Perez-Perez, G. I., Meuwissen, S. G., and Blaser, M. J. (1995) J. Natl. Cancer Inst. 87,

1777-1780

12. Peek, R. M., Jr, Miller, G. G., Tham, K. T., Perez-Perez, G. I., Zhao, X., Atherton, J. C., and Blaser, M.

J. (1995) Lab. Invest. 73, 760-770

13. Rugge, M., Busatto, G., Cassaro, M., Shiao, Y. H., Russo, V., Leandro, G., Avellini, C., Fabiano, A.,

Sidoni, A., and Covacci, A. (1999) Cancer 85, 2506-2511

14. Covacci, A., Telford, J. L., DelGiudice, G., Parsonnet, J., and Rappuoli, R. (1999) Science 284, 1328-

1333

15. Asahi, M., Azuma, T., Ito, S., Ito, Y., Suto, H., Nagai, Y., Tsubokawa, M., Tohyama, Y., Maeda, S.,

Omata, M., Suzuki, T., and Sasakawa, C. (2000) J. Exp. Med. 191, 593-602

16. Backert, S., Ziska, E., Brinkmann, V., Zimny-Arndt, U., Fauconnier, A., Jungblut, P. R., Naumann, M.,

and Meyer, T. F. (2000) Cell. Microbiol. 2, 155-164

17. Odenbreit, S., Puls, J., Sedlmaier, B., Gerland, E., Fischer, W., and Haas, R. (2000) Science 287,

1497-1500

18. Stein, M., Rappuoli, R., and Covacci, A. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 1263-1268

19. Selbach, M., Moese, S., Hauck, C. R., Meyer, T. F., and Backert, S. (2002) J. Biol. Chem. 277, 6775-

6778

20. Stein, M., Bagnoli, F., Halenbeck, R., Rappuoli, R., Fantl, W. J., and Covacci, A. (2002) Mol.

Microbiol. 43, 971-980

21. Backert, S., Moese, S., Selbach, M., Brinkmann, V., and Meyer, T. F. (2001) Mol. Microbiol. 42, 631-

644

22. Higashi, H., Tsutsumi, R., Muto, S., Sugiyama, T., Azuma, T., Asaka, M., and Hatakeyama, M. (2002)

Science 295, 683-686

23. Tsutsumi, R., Higashi, H., Higuchi, M., Okada, M., and Hatakeyama, M. (2003) J. Biol. Chem. 278,

3664-3670

24. Yamazaki, S., Yamakawa, A., Ito, Y., Ohtani, M., Higashi, H., Hatakeyama, M., and Azuma, T. (2003)

J. Infect. Dis. 187, 334-337

25. Segal, E. D., Cha, J., Lo, J., Falkow, S., and Tompkins, L. S. (1999) Proc. Natl. Acad. Sci. U. S. A. 96,

14559-14564

26. Higashi, H., Nakaya, A., Tsutsumi, R., Yokoyama, K., Fujii, Y., Ishikawa, S., Higuchi, M., Takahashi,

A., Kurashima, Y., Teishikata, Y., Tanaka, S., Azuma, T., and Hatakeyama, M. (2004) J. Biol. Chem.

279, 17205-17216

27. Amieva, M. R., Vogelmann, R., Covacci, A., Tompkins, L. S., Nelson, W. J., and Falkow, S. (2003)

Science 300, 1430-1434

28. Umehara, S., Higashi, H., Ohnishi, N., Asaka, M., and Hatakeyama, M. (2003) Oncogene 22, 8337-

8342

29. Higashi, H., Tsutsumi, R., Fujita, A., Yamazaki, S., Asaka, M., Azuma, T., and Hatakeyama, M.

(2002) Proc. Natl. Acad. Sci. U. S. A. 99, 14428-14433

30. Azuma, T., Yamazaki, S., Yamakawa, A., Ohtani, M., Muramatsu, A., Suto, H., Ito, Y., Dojo, M.,

Yamazaki, Y., Kuriyama, M., Keida, Y., Higashi, H., and Hatakeyama, M. (2004) J. Infect. Dis. 189,

820-827

31. Hatakeyama, M. (2004) Nat. Rev. Cancer 4, 688-694

32. Yamaoka, Y., Osato, M. S., Sepulveda, A. R., Gutierrez, O., Figura, N., Kim, J. G., Kodama, T.,

Kashima, K., and Graham, D. Y. (2000) Epidemiol. Infect. 124, 91-96

33. Azuma, T. (2004) J. Gastroenterol. 39, 97-103

34. Asahi, M., Tanaka, Y., Izumi, T., Ito, Y., Naiki, H., Kersulyte, D., Tsujikawa, K., Saito, M., Sada, K.,

Yanagi, S., Fujikawa, A., Noda, M., and Itokawa, Y. (2003) Helicobacter 8, 1-14

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

8

35. Churin, Y., Al-Ghoul, L., Kepp, O., Meyer, T. F., Birchmeier, W., and Naumann, M. (2003) J. Cell

Biol. 161, 249-255

FOOTNOTES

* This work was supported by grants-in-aid for science research and JSPS Fellows from the Ministry of

Education, Science, Sports, and Culture of Japan, and by a grant of the Virtual Research Institute of Aging

of Nippon Boehringer Ingelheim, and by a grant from the Sagawa foundation for promotion of cancer

research.1The abbreviation used are: H. pylori, Helicobacter pylori; cagA, cytotoxin-associated gene A; cag PAI,

cytotoxin associated gene pathogenicity island; SHP-2, SH2 domain-containing protein-tyrosinephosphatase-2; Csk, C-terminal Src kinase; EPIYA, glutamic acid-proline-isoleucine-tyrosine-alanine;WT, wild-type; PR, phosphorylation-resistant; HA, hemagglutinin.

2S. Ren, and M. Hatakeyama, unpublished observation.

3H. Higashi, Y. Fujii, and M. Hatakeyama, unpublished observation.

FIGURE LEGENDS

FIG. 1. Subcellular localization of CagA mutants in AGS cells. A, A schematic view of the wild-

type CagA and its mutants. Information of each mutant is described in the EXPERIMENTAL

PROCEDURES. All proteins were C-terminal hemagglutinin (HA) epitope-tagged. Numbers indicate

amino-acid positions in the wild-type CagA (WT CagA). Black boxes indicate locations of the EPIYA

sequence in CagA. The open box indicates the EPIYA-containing region that spans amino-acid residues

869-1086. B, AGS cells were transfected with the indicated CagA expression vector or control empty

vector. The cells were harvested and lysed at 17 h post-transfection. Total cell lysates were

immunoblotted (IB) with anti-HA or anti-phosphotyrosine (anti-pY) antibody. Mr, relative molecular mass.

C, AGS cells transfected with the indicated expression vector were fixed and treated with anti-HA

antibody (Y-11). Primary antibody was then localized by Alexa Fluor 546-conjugated anti-rabbit antibody.

Stained cells were then examined under a confocal microscope system. Bars indicate 100 μm (black) and

25 μm (white).

FIG. 2. Subdivision of the EPIYA-containing region of CagA. A, Structural comparison of the

EPIYA-containing region among CagA species from different H. pylori Western strains (Upper parts) and

the schematic view of the EPIYA-containing region of NCTC11637 CagA (lower). In NCTC11637 CagA,

the EPIYA-containing region consists of three distinct segments (EPIYA-A, -B and -C segments), each of

which includes one EPIYA motif (EPIYA-A, -B or –C site). The border of each segment was determined

by comparing sequences of the EPIYA-containing regions among various CagA isolates (Upper parts). B,

Amino acid sequence for each EPIYA segment.

FIG. 3. The role of EPIYA segments in membrane localization of CagA. A, A schematic view of

wild-type CagA (WT) and its mutants, which lack EPIYA-A, -B, and/or -C segments in various

combinations. Information of each mutant is described in the EXPERIMENTAL PROCEDURES. All

CagA construct were HA-tagged at C-terminal. Black boxes indicate locations of the EPIYA motif in

CagA. B, AGS cells were transfected with WT CagA, each of the EPIYA-segment mutants shown in (A),

or a control vector. At 17 h after transfection, the cells were harvested and lysed. The cell lysates were

immunoblotted (IB) with an antibody to HA or phosphotyrosine (anti-pY). Mr, relative molecular mass. C,

AGS cells transfected with the indicated expression vector were stained with an anti-HA antibody and

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

9

were examined under a confocal microscope system. Bars indicate 25 μm.

FIG. 4. A single EPIYA motif is necessary and sufficient for membrane localization of CagA. A, a

schematic representation of CagA mutants, which contain a single EPIYA segment with or without the

EPIYA sequence. All CagA constructs were HA-tagged at C-terminal. Black boxes indicate locations of

the EPIYA motif in CagA. B, AGS cells were transfected with wild-type CagA (WT), each of the EPIYA-

segment mutants shown in (A), or a control vector. Cell lysates were immunoblotted (IB) with an antibody

to HA or phosphotyrosine (anti-pY). C, AGS cells transfected with the indicated expression vector. At 17

h after transfection, cells were stained with an anti-HA antibody and were examined under a confocal

microscope system. Bars indicate 25 μm. Mr, relative molecular mass.

FIG. 5. Introduction of the EPIYA motif restores membrane localization of a CagA mutant that

lacks the EPIYA-containing region. A, a schematic view of CagA ABCCC, CagA ABCCC+EPIYA and

CagA ABCCC+3EPIYA. ABCCC+EPIYA and ABCCC+3EPIYA mutants were made by replacing

the entire EPIYA-containing region (amino-acid residues 869-1086) with a single or triple-repeats of the

EPIYA sequence, respectively. All CagA constructs were HA-tagged at C-terminal. Black boxes indicate

locations of the EPIYA motif in CagA. B, AGS cells were transfected with the indicated construct and the

cell lysates were immunoblotted (IB) with antibody to HA or phosphotyrosine (anti-pY). C, AGS cells

transfected with the indicated expression vector were stained with an anti-HA antibody and were

examined under a confocal microscope system. Bars indicate 25 μm.

FIG. 6. EPIYA motif alone is insufficient to specify the membrane localization of an unrelated

cytoplasmic protein. A, a schematic view of wild-type SHP-2 and its mutants. B, AGS cells were

transfected with constitutively active SHP-2 (SHP-2 SH2-Myc), constitutively active SHP-2 with the

myristoylation signal from v-Src (Myr-SHP-2 SH2-Myc), or constitutively active SHP-2 with the EPIYA

motif (EPIYA-SHP-2 SH2-Myc). At 17 h after transfection, cells were fixed and treated with anti-Myc

antibody (9E10). Primary antibody was localized by Alexa Fluor 546-conjugated anti-mouse antibody.

Stained cells were then examined under a confocal microscope system. Bars indicate 25 μm.

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Contro

lΔN ΔN

ΔC

N WT

CagA-HA

ΔNΔA

BCCC

Contro

lΔN ΔN

ΔC

N WT

CagA-HA

ΔNΔA

BCCC

IB: anti-HA IB: anti-pY

HACagA ΔN12471 57 587

HANCTC11637

CagA (Wid-type)

12471EPIYA-containing

regionCOOHNH2

HACagA ΔNΔC10861 57 587

HACagA ΔNΔABCCC12471 57 587 868 1087

A

HACagA N6121

HACagA ΔABCCC

12471 868 1087

Figure 1 (Higashi et al. )

165(kD

115

84

61

55

B

: EPIYA motif

IB: anti-HA IB: anti-pYW

T

ΔABCCC

Contro

lW

T

ΔABCCC

Contro

l

175(kDa)

115

Mr

Mr

CagA N

CagAΔNΔC

CagAΔNΔABCCC

CagA ΔN

WT CagA

PR CagA

Control

CagAΔABCCC

C

10

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

EPIYA-Csegment

EPIYA-Csegment

NCTC11637 CagA EPIYA-containig region

EPIYA-Asite EPIYA motifEPIYA-B

siteEPIYA-C

site

EPIYA-Asegment

EPIYA-Bsegment

EPIYA-Csegment

ANCTC11637

(ABCCC)

26695(ABC)

F79(ABCC)

CCUG17874(AB)

ATCC49503(AA’C)

A’

A B C C C

Figure2 (Higashi et al.)

B

B siteTGQVASPEEPIYAQVAKKVNAKIDRLNQAASGLGGVGQAG

EPIYA-B segment

C siteFPLKRHDKVDDLSKVGRSVSPEPIYATIDDLGGP

EPIYA-C segment

KKELNEKFKNFNNNNNNGLENEPIYAKVNKKKA site

EPIYA-A segment

11

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

165(kDa)

115

165(kDa)

115

Contro

lW

T

ΔABCC

IB: anti-pY

IB: anti-HA

CagA-HA

Contro

lW

TΔA

BCCC

CagA-HA

ΔABCCC-s

ΔABCC-s

ΔABC

ΔAB

ΔCCC

ΔACCC

ΔBCCC

175(kDa)

115

IB: anti-pY

175(kDa)

115

IB: anti-HA

B

Mr

Mr

HA12471

B

1087868901 940

HA12471

A B

1087940

HAWT12471

COOHNH2A

A CC CB

HA12471 1087900

A

HA12471

CC C

941868

HA12471

C C

975868

HA12471

C

1009868

HA12471

C

108786810421009

ΔCCC

ΔBCCC

ΔACCC

ΔAB

ΔABC

ΔABCC-s

ΔABCC

CagA

HA12471 1043868

ΔABCCC-s

: EPIYA motif

Figure 3 (Higashi et al.)

ΔABCCC-s

ΔCCC

ΔBCCC

ΔACCC

ΔAB

ΔABC

ΔABCC-s

ΔABCC

EPIYA motifs: 3

EPIYA motifs: 2

EPIYA motif: 1 EPIYA motif: 0C

Mr

Mr

EPIYA-containingregion

12

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

HA12471ΔACCC

[ΔEPIYA]

EPIYA-containingregion COOHNH2

A

HA12471

HA12471

C

108786810421009

ΔABCC

12471 1087868HAΔABCC

[ΔEPIYA]

ΔBCCC[ΔEPIYA]

CagA

HA12471 1087900

AΔBCCC

HA12471

B

1087868901 940

ΔACCC

Figure 4 (Higashi et al.)

ΔACCC [ΔEPIYA]

ΔBCCC [ΔEPIYA]

ΔABCC [ΔEPIYA]

C

B

Contro

l

ΔBCCC

[ΔEPIY

A]

WT

ΔACCC

[ΔEPIY

A]

Contro

lW

T

165(kDa)

115

165(kDa)

115

IB: anti-pY

IB: anti-HA

ΔABCC

[ΔEPIY

A]

: EPIYA motif

Mr

Mr

13

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

IB: anti-pY

IB: anti-HA

CagA-HAΔA

BCCC+EPIY

A

ΔABCCC+3

EPIYA

Control

EPIYA-containingregion COOHNH2

A CagA

HA12471 1087868

ΔABCCC

HA12471 1087868ΔABCCC

+EPIYA

HA12471 1087868ΔABCCC

+3EPIYA

B

ΔABCCC

ΔABCCC+EPIYA

ΔABCCC+3EPIYA

C

Figure 5 (Higashi et al.)

: EPIYA motif

14

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

SHP-2ΔSH2-Myc Myr-SHP-2ΔSH2-Myc EPIYA-SHP-2ΔSH2-Myc

Figure 6 (Higashi et al.)

5941

COOHNH2

Myc1 32 192 593

SHP-2ΔSH2-Myc

wild-type SHP-2

Myc1 192 593

Myr-SHP-2ΔSH2-Myc

: myristoylation signal sequence

Myc1

EPIYA-SHP-2ΔSH2-Myc

: EPIYA motif

A

B

SH2 SH2

192 593

phosphatasedomain

phosphatasedomain

phosphatasedomain

phosphatasedomain

15

by guest on April 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Saadat, Naoko Murata-Kamiya, Takeshi Azuma and Masanori HatakeyamaHideaki Higashi, Kazuyuki Yokoyama, Yumiko Fujii, Shumei Ren, Hitomi Yuasa, Iraj

mammalian cellsEpiya motif is a membrane targeting signal of Helicobacter pylori CagA in

published online April 13, 2005J. Biol. Chem. 

  10.1074/jbc.M503583200Access 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 16, 2018

http://ww

w.jbc.org/

Dow

nloaded from