Báo cáo khoa học: SHP-1 dephosphorylates 3BP2 and potentially downregulates 3BP2-mediated T cell antigen receptor signaling ppt

Because the interactcoimmunoprecipitat-ion of 3BP2 with SHP-1 was first identified from a T cell cDNA library in a modified yeast two-hybrid system in which a Src family kinase c-Src was ex

downregulates 3BP2-mediated T cell antigen receptor

signaling

Zhenbao Yu1, Meryem Maoui1, Zhizhuang J Zhao2, Yang Li1and Shi-Hsiang Shen1,3

1 Health Sector, Biotechnology Research Institute, National Research Council of Canada, Montre´al, Canada

2 Department of Pathology, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA

3 Department of Medicine, McGill University, Montre´al, Canada

Protein tyrosine phosphorylation plays a critical role

in various signal-transduction pathways in T

lympho-cytes [1] For example, ligation of the T cell antigen

receptor (TCR) activates Src family protein tyrosine

kinases (PTKs) such as Lck and Fyn, which in turn

phosphorylate TCR n chain and CD3 e, d and c

sub-units within the immunoreceptor tyrosine-based

activa-tion motif (ITAM), resulting in the recruitment and

activation of the ZAP70 and Syk PTKs [2,3] These

activated PTKs further induce the tyrosine

phosphory-lation of multiple intracellular proteins, including the

adapter proteins LAT [4] and SLP-76 [5]

Phosphoryla-tion of these adapter proteins creates docking sites

for various Src-homology 2 (SH2) domain-containing

proteins such as PLCc, Grb2, Grap, Gads, Nck, Vav,

c-CBL and Tec family tyrosine kinase Itk, leading to

stimulation of downstream signaling pathways, and ultimately to T cell activation [6,7]

Dephosphorylation of these tyrosine-phosphorylated proteins is a necessary counterpart for maintaining a balance between activation and quiescence of TCR signaling [8] SHP-1 is one such enzyme which can counterbalance PTK effects and terminate recep-tor-initiated signaling [9] SHP-1 is expressed primarily

in hematopoietic cells and plays a critical role in the negative regulation of TCR signaling and T cell development Accordingly, thymocytes derived from motheaten (me) mice, which lack the expression of functional SHP-1, hyperproliferate in response to TCR stimulation [10–18] SHP-1 displays its negative func-tion at diverse stages of TCR signaling For instance, SHP-1 constitutively associates with TCR and appears

Keywords

3BP2; protein phosphatases; protein–protein

interaction; SHP-1; T cell-receptor

Correspondence

Zhenbao Yu, Health Sector, Biotechnology

Research Institute, National Research

Council of Canada, Montre´al, Que´bec

H4P 2R2, Canada

Fax: +1 514 496 6319

Tel: +1 514 496 6377

E-mail: zhenbao.yu@nrc.ca

(Received 14 December 2005, revised 28

February 2006, accepted 16 March 2006)

doi:10.1111/j.1742-4658.2006.05233.x

Src homology 2 (SH2) domain-containing protein tyrosine phosphatase-1 (SHP-1) is a critical inhibitory regulator in T cell-receptor (TCR) signaling However, the exact molecular mechanism underlying this is poorly defined, largely because the physiological substrates for SHP-1 in T cells remain elusive In this study, we showed that adaptor protein 3BP2 serves as a binding protein and a physiological substrate of SHP-1 3BP2 is phosphor-ylated on tyrosyl residue 448 in response to TCR activation, and the phos-phorylation is required for T cell signalling, as indicated by transcriptional activation of nuclear factor activated in T cells (NFAT) Concurrently, phosphorylation of Tyr566 at the C-terminus of SHP-1 causes specific recruitment of 3BP2 to the phosphatase through the SH2 domain of the adaptor protein This leads to efficient dephosphorylation of 3BP2 and thereby termination of T cell signaling The study thus defines a novel function of the C-terminal segment of SHP-1 and reveals a new mechanism

by which T cell signaling is regulated

Abbreviations

GST, glutathione S-transferase; IP, immunoprecipitation; ITAM, immunoreceptor tyrosine-based activation motif; PTK, protein tyrosine kinase; PTPase, protein tyrosine phosphatase; pTyr, phosphotyrosine; SH2, src homology 2; SHP, SH2 domain-containing PTPase; TCR,

T cell antigen receptor.

to dephosphorylate the TCR CD3e subunit and more

distal signaling effectors following TCR activation [11]

It has been also reported that SHP-1 is phosphorylated

by activated Src family kinase Lck [19] in T cells and

it, in turn, dephosphorylates and inactivates Lck and

Fyn kinases [12,18,20] SHP-1 is also thought to

regu-late the activity of Syk family kinase ZAP70 [21,22]

Moreover, SHP-1 is known to interact with the adapter

proteins Grb2 and SLP-76, although the physiological

meaning is unclear [11,23,24] In this study, we

identi-fied 3BP2 as a novel SHP-1 substrate and binding

protein

3BP2 was originally described as a PTK c-Abl SH3

domain-binding protein [25] It contains an N-terminal

PH domain, a central proline-rich region that interacts

with c-Abl, and a C-terminal SH2 domain Recently,

the SH2 domain of 3BP2 has been shown to bind to

the PTKs Syk and ZAP70 [26] As a result,

overexpres-sion of 3BP2 in T cells leads to increased nuclear

fac-tor activated T cell (NFAT-) and AP-1-dependent

transcription [26,27] It has also been shown that 3BP2

plays a positive regulatory role in NK-cell-mediated

cytotoxicity [28] and participates in the regulation of

FccR1-mediated degranulation in basophilic cells [29]

However, the mechanism by which 3BP2 exerts its

pos-itive effect on downstream signaling molecules remains

elusive

In this study, we demonstrate that the interplay of

3BP2 and SHP-1 has an important role in T cell

signa-ling On the one hand, 3BP2 is phosphorylated on

tyr-osyl residue 448, and the tyrosine phosphorylation is

critical for TCR signaling On the other hand, SHP-1

is phosphorylated on Tyr566 at the C-terminus and

thereby recruits 3BP2 through SH2 domain

inter-action This leads to dephosphorylation of 3BP2 and

termination of T cell signaling This study thus

pro-vides a novel mechanism by which 3BP2 and SHP-1

regulate T cell signaling

Results

3BP2 interacts with SHP-1 in a yeast two-hybrid

screen

To demonstrate the molecular mechanism of

SHP-1-mediated regulation of TCR signaling, we searched

for SHP-1-interacting proteins from human T cells

using a modified yeast two-hybrid screen [30] The

full-length SHP-1 with mutation of Cys455 to Ser

(SHP-1⁄ C455S), which abolishes the protein tyrosine

phosphatase (PTPase) catalytic activity but retains the

binding ability to its substrates, was cloned into

plasmid pBTM-116-Src [31] for two-hybrid screening

Transformation of the plasmid in yeast results in the expression of Lex DNA-binding domain⁄ SHP-1– C455S fusion protein and c-Src kinase Expression of c-Src allows the identification of tyrosine phosphoryla-tion-dependent SHP-1-interacting proteins From 1.1· 107 transformants with a human Jurkat T cell cDNA library, 124 were positive for both HIS3 and LacZ expression Sequence analyses of the 124 positive clones revealed that, among others, 11 independent clones of different lengths represented overlapping cDNAs of the SH3 domain-binding protein 2 (3BP2) 3BP2 was originally characterized as an Abl SH3-interacting protein [25] It is composed of an N-terminal PH domain, a proline-rich region and a C-terminal SH2 domain Interestingly, all of the 3BP2 clones isolated in our two-hybrid screening contained

at least the sequence encoding the entire SH2 domain, suggesting that the SH2 domain of 3BP2 is involved

in mediating the SHP)1 ⁄ 3BP2 interaction Our addi-tional studies demonstrated that the catalytic inactive Cys-to-Ser mutant of SHP-2, an enzyme structurally similar to SHP-1, was incapable of interacting with 3BP2 in the system (data not shown) This indicates a high specificity of the interaction between 3BP2 and SHP-1

3BP2 associates with SHP-1 in 293T cells when coexpressed with Lck

To determine whether 3BP2 associates with SHP-1 in mammalian cells, we carried out a coimmunoprecipitat-ion assay Because the interactcoimmunoprecipitat-ion of 3BP2 with SHP-1 was first identified from a T cell cDNA library in a modified yeast two-hybrid system in which a Src family kinase c-Src was expressed, we cotransfected 293T cells with C-terminal myc-tagged 3BP2 (3BP2–myc), catalyt-ically inactive SHP-1 (SHP-1⁄ C455S), and Src family kinases, Fyn and Lck, which are known to be involved

in TCR signaling 3BP2–myc was immunoprecipitated with an anti-myc IgG As shown in Fig 1A,

SHP-1⁄ C455S was detected by western blot with the anti-(SHP-1) IgG in the anti-myc immunoprecipitant from 293T cells cotransfected with wild-type Lck or catalyti-cally activated Lck (Lck⁄ Y505F) However, under the same conditions, SHP-1⁄ C455S could not be coimmu-noprecipitated with 3BP2–myc from 293T cells cotrans-fected with catalytically activated Fyn (Fyn⁄ Y531F) or the plain control vector (Fig 1A) In reciprocal experi-ments involving immunoprecipitating SHP-1⁄ C445S with anti-(SHP-1) IgG, 3BP2–myc was detected in the immunoprecipitant from cells transfected with cata-lytically activated Lck but not in the immunoprecipi-tant from the cells transfected with wild-type Lck

(Fig 1A, right) although SHP-1 could be

coimmuno-precipitated with 3BP2–myc by anti-myc IgG from the

wild-type Lck-transfected cell This indicates that low

amounts of associated proteins could not be detected

in some coimminoprecipitation experiments

3BP2, SHP-1, and Lck all contain SH2 domains and

potential tyrosine phosphorylation sites that can

medi-ate protein–protein interactions To examine whether

the association of 3BP2 with SHP-1 could be mediated

by Lck, we cotransfected these three proteins into

293T cells and carried out a two-step

immunoprecipi-tation experiment Whole-cell lysates were first

subjec-ted to immunoprecipitation with anti-Lck IgG or

control antibody (mouse IgG or protein

A–Seph-arose 4B beads alone) and the unbound proteins were

then subjected to immunoprecipitation with anti-myc

IgG If the interaction of 3BP2 with SHP-1 is mediated

by Lck, removal of Lck from whole-cell lysates by anti-Lck IgG immunoprecipitation should reduce the amount of SHP-1 coprecipitated with 3BP2–myc However, as shown in Fig 1B, although Lck was essentially depleted from whole-cell lysates by anti-Lck IgG (Fig 1B, upper, lane 4), coimmunoprecipitation

of SHP-1 with 3BP2–myc was not affected (Fig 1B, lower, lanes 8–10) Moreover, neither 3BP2–myc (Fig 1B, middle, lane 7) nor SHP-1 (Fig 1B, lower, lane 7) was detected in the anti-Lck immunoprecipi-tates Parallel reciprocal experiments showed that Lck was not coimmunoprecipitated with 3BP2–myc by anti-myc IgG either (Fig 1B, upper, lanes 8 and 9) Taken together, these results indicate that the associ-ation of 3BP2 with SHP-1 is not mediated by Lck, although its kinase activity is required for the interac-tion in 293 cells

3BP2-myc

SHP-1

+ + + + +

kinase

-F 5 Y

-n k

F 5 Y -k

IP: anti-myc IP: anti-SHP-1

Western blot:

anti-SHP-1

82

62

47

82

62

47

3BP2-myc

SHP-1

+ + + + +

-F 5 Y

-n k

F 5 Y -k

Western blot:

anti-myc

A

IP: anti-myc IP: anti-SHP-1 Whole cell lysates

3BP2-myc

SHP-1

+

kinase

-F 5 Y

-n k

F 5 Y -k

+

-F 5 Y

-n k

F 5 Y -k

+

-F 5 Y

-n k

F 5 Y -k

Western blot:

anti-pTyr

3BP2-myc

SHP-1 82

62

47

160

W s

first IP

Lck

unbound precipitant second IP: anti-myc

Western blot: anti-SHP-1 3BP2

SHP-1

82

62

47

82

62

47

82

62

47

Western blot: anti-myc

Western blot: anti-Lck

B

Fig 1 3BP2 associates with 1 when coexpressed with Lck in 293T cells (A) 293T cells were transfected with 3BP2–myc,

SHP-1 ⁄ C445S and Src family of kinases Fyn or Lck as indicated The cells were grown in Dulbecco’s modified Eagle’s medium with 10% fetal bovine IgG for 48 h after transfection and then lysed without any treatment Whole-cell lysates were subjected to immunoprecipitation and western blot analysis with anti-myc IgG, anti-SHP-1 IgG or anti-phosphotyrosine IgG Molecular mass (kDa) is indicated to the left of the gel (B) 293T cells were transfected with 3BP2–myc, SHP-1 ⁄ C445S and autoactivated Lck (Lck ⁄ Y505F) Forty-eight hours after transfection the cells were lysed and the whole-cell lysates were subjected to immunoprecipitation with anti-Lck IgG, IgG or without antibody as control The unbound proteins after the first immunoprecipitation were subjected to immunoprecipitation with anti-myc IgG The proteins collected in each step were analyzed by western blot as indicated.

3BP2 interacts with SHP-1 through the SH2

domain of 3BP2

Because both 3BP2 and SHP-1 contain SH2 domains,

the interaction of 3BP2 with SHP-1 might be through

either binding of the SH2 domains of SHP-1 to

tyro-sine-phosphorylated 3BP2 or that of the SH2 domain

of 3BP2 to phosphorylated SHP-1 We thus

construc-ted a glutathione S-transferase (GST) fusion protein of

the SHP-1 SH2 domains (GST–SHP-1–2SH2) and also

of the 3BP2 SH2 domain (GST)3BP2–SH2), and

carried out GST pull-down experiments to determine

which of these two possibilities accounts for the

observed association As shown in Fig 2A,

SHP-1⁄ C455S was precipitated by GST)3BP2–SH2 In the

same condition, SHP-2 could not be pulled down by

GST)3BP2–SH2 Western blot with

anti-phophotyro-sine IgG indicates that both SHP-1 and SHP-2 were

phosphorylated These results suggest that 3BP2

specif-ically interacts with SHP-1 but not SHP-2 In contrast,

3BP2–myc could not be pulled down by GST–SHP-1–

2SH2 (Fig 2B) Note that, under the same conditions,

GST–SHP-1–2SH2 was able to pull down S2V, a siglec

family receptor previously identified as an

SHP-1-bind-ing protein [32], suggestSHP-1-bind-ing that the GST–SHP-1–2SH2

fusion protein was properly folded These results

sug-gest that 3BP2 interacts with SHP-1 through the SH2

domain of 3BP2 and, presumably,

tyrosine-phosphor-ylated SHP)1

Interaction of the SH2 domain of 3BP2 with SHP-1 is mediated by phosphorylation

of SHP-1 at Tyr566

We next determined which tyrosine residue(s) of

SHP-1 is (are) involved in the interaction using tyrosine-to-phenylalanine mutants SHP-1 contains a C-terminal noncatalytic tail that bears three potential phosphoryl-ated tyrosine residues (Tyr538, Tyr543 and Tyr566)

We mutated each of them and cotransfected the result-ing mutants with 3BP2–myc and Lck into 293T cells

As shown in Fig 3B, although both SHP-1⁄ Y538F and SHP-1⁄ Y543F were detected in the anti-myc immunoprecipitants, SHP-1⁄ Y566 was not detectable, suggesting that 3BP2 binds to the phosphorylated Tyr566 of SHP-1 Interestingly, we found that both wild-type and catalytically inactive SHP-1

(SHP-1⁄ C455S) could be coprecipitated by 3BP2–myc with anti-myc IgG (Fig 2B and data not shown), indicating that Tyr566, the 3BP2-binding site of SHP-1 was not dephosphorylated by SHP-1 in this condition How-ever, western blot analyses of the anti-SHP-1 immunoprecipitants with anti-phosphotyrosine IgG showed that the tyrosine phosphorylation level of SHP-1⁄ Y566F mutant is much lower than that of wild-type SHP-1, SHP-1⁄ Y538F, SHP-1 ⁄ Y543F and SHP-1⁄ C455S ⁄ Y566F (Fig 3F), suggesting that Tyr566 is the major phosphorylation site of SHP-1 in this condition The anti-phosphotyrosine western blot

Western blot:

anti-SHP-1

Western blot:

anti-SHP-2

L

C

W

T

-2 P 3-T S

SHP-1

SHP-2

IP :anti-SHP-2

Western blot:

anti-pTyr

Western blot:

IgG SHP-1

SHP-2

pervanadate - +

Western blot:

anti-myc

c m -2 P 3

c m -V 2 S

c m -V 2 S + y m -2 P 3

c m -2 P 3

c m -V 2 S

c m -V 2 S + y m -2 P 3

3BP2-myc S2V-myc

WCL GST-SHP-1-2SH2

A

B

IP :anti-SHP-1

Fig 2 SHP-1 associates with 3BP2 through the SH2 domain of 3BP2 293T cells were transfected with SHP-1 ⁄ C455S,

SHP-2 ⁄ C459S, 3BP2–myc and 3BP2–myc plus S2V-myc, respectively Forty-eight hours after transfection, the cells were treated with 0.5 m M pervanadate for 30 min The whole-cell lysates were incubated with GST, GST )3BP2–SH2 or GST–SHP-1–2SH2 bound on glutathione Sepharose and subjec-ted to immunoprecipitation as indicasubjec-ted The proteins precipitated were analyzed by western blot with anti-SHP-1, anti-SHP-2, anti-phosphotyrosine or anti-myc IgG.

analysis of the anti-myc immunoprecipitants showed

that 3BP2 was phosphorylated and dephosphorylated

by wild-type SHP-1, and the SHP-1⁄ Y538F and

SHP-1⁄ Y543F mutants (Fig 3E), but not by the

catalyti-cally inactivated mutants (C455S and C455S⁄ Y566F)

3BP2 was also partially dephosphorylated by

SHP-1⁄ Y566F mutant (Fig 3E,D, lane 7) although it did

not associate with this mutant (Fig 3B, lane 7),

sug-gesting that SHP-1 may be also able to directly

de-phospharylate 3BP2 without association of the two

proteins through the SH2 domain–phosphotyrosine

interaction in the condition with the overexpression of

the two proteins

To exclude the possibility that the major

tyrosine-phosphorylated protein in the anti-myc precipitants is

not 3BP2–myc but another protein of similar

mole-cular mass that might be comimmunoprecipitated with

3BP2–myc, we treated the whole-cell lysates by adding

SDS to 1% and heating the samples at 100
C for

10 min before immunoprecipitation This should dis-rupt protein–protein interactions Treated samples were then diluted 10 times with lysis buffer and subjected to immunoprecipitation Such treatment is expected to eliminate the coimmunoprecipitation of any 3BP2-binding proteins from 3BP2 with anti-myc IgG As shown in Fig 3G, the tyrosine-phosphoryl-ated protein with the same molecular mass as 3BP2– myc was detected in the anti-myc precipitants from the SDS-treated samples as well as in those from non-treated samples This result further confirms that 3BP2 was tyrosine phosphorylated More significantly, the phosphorylated 3BP2 was nearly completely de-phosphorylated by wild-type SHP-1, but not by its catalytically inactive mutant SHP-1⁄ C455S (Fig 3E), suggesting that 3BP2 is a potential substrate for SHP-1

A

whole cell lysates IP: anti-myc IP: anti-SHP-1

Western blot: anti-pTyr

3BP2-myc + + + + + + +

Lck +

+

+

-SHP-1 W T

S

5

C

F 5 Y F 5 Y F 5 Y

F 5 Y / S 5 C T

W

-+ + + + + + +

+ + +

-T W S 5 C F 5 Y F 5 Y F 5 Y

F 5 Y / S 5 C T

W

-+ + + + + + +

+ + +

-T W S 5 C F 5 Y F 5 Y F 5 Y

F 5 Y / S 5 C T

W

-120

60

40

SHP-1

3BP2-myc + + + + + + +

Lck +

+

+

-SHP-1 W T

S 5 C

F 5 Y F 5 Y F 5 Y

F 5 Y / S 5 C T

W

-+ + + + + + +

+ + +

-T W S 5 C F 5 Y F 5 Y F 5 Y

F 5 Y / S 5 C T

W

-120

60

40

SHP-1

IP: anti-myc IP: anti-myc

WB: anti-myc WB: anti-SHP-1

+ + + + + + +

+ + +

-T W S 5 C F 5 Y F 5 Y F 5 Y

F 5 Y / S 5 C T

W

-IP: anti-SHP-1 WB: anti-SHP-1

E

3 4 5 6 7 8

2

1 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8

3 4 5 6 7 8 2

1

3 4 5 6 7 8 2

1

3 4 5 6 7 8

2

1

3BP2-myc + + +

Lck + + + +

-SHP-1

S 5 C T W

+ + +

-S 5 C T W

-+ -+ + + + + +

-S 5 C T W

-SDS treatment - +

WCL IP: anti-myc

WB: anti-pTyr

3BP2-myc + + +

Lck + + + +

-SHP-1

S 5 C T W

+ + +

-S 5 C T W

-WCL WCL WB: anti-Lck WB: anti-SHP-1

82 62 47

82 62 47

WB: anti-myc

82 62 47

G

Fig 3 SHP-1 associates with 3BP2 through the phosphorylated tyrosine residue 566 of SHP-1 (A–F) 293T cells were cotransfected with 3BP2–myc, Lck ⁄ Y505F and SHP-1 or its mutants Forty-eight hours after transfection, the cells were lysed and whole-cell lysates were sub-jected to immunoprecipitation and western blot with the indicated antibodies (G) 293T cells were cotransfected with 3BP2 myc, Lck⁄ Y505F and SHP-1 or its catalytically inactive mutant (SHP-1 ⁄ C455S) Forty-eight hours after transfection, the cells were lysed The whole-cell lysates were equally divided into two portions One portion of the lysates was treated with 1% SDS at 100
C for 10 min The other portion was left untreated The lysates were then diluted 10 times with lysis buffer and subjected to immunoprecipitation and western blot as des-cribed.

Tyr448 is the major phosphorylated residue of

3BP2 in response to TCR engagement and is

critical for 3BP2 function in TCR signaling

3BP2 is a positive regulator of TCR signaling and is

phosphorylated on tyrosine residues in response to

TCR engagement [26] To further study the function

of 3BP2 phosphorylation in the regulation of TCR

signaling, we determined which tyrosine residue(s) of

3BP2 can be phosphorylated Four potential

phos-phorylation sites, namely Tyr174, Tyr183, Tyr448

and Tyr485, were predicted (http://kinasephos.mbc

nctu.edu.tw) We mutated these tyrosine residues to

phenylalanine and transfected these mutants into

Jur-kat cells The tyrosine-phosphorylation status of these

mutants was examined following stimulation with

anti-CD3 IgG OKT3 As shown in Fig 4, although none

of the mutations Y174F, Y183F and Y485F exerted

any evident effect on the tyrosine phosphorylation of

3BP2, mutation Y448F almost completely abolished

tyrosine phosphorylation, suggesting that Tyr448 is the

major tyrosine-phosphorylated residue in response to

TCR activation

We next determined the effects of these 3BP2

mutants on NFAT activation To do so, we

cotransfected 3BP2 or its mutants with

NFAT-lucif-erase (firefly) reporter into Jurkat cells and stimulated

the cells with anti-CD3 IgG, and PMA plus

ionomy-cin, respectively To determine whether the expression

of 3BP2 and its mutants affects stimulation of the

expression of NFAT-driven luciferase by PMA⁄

iono-mycin, we transfected the Jurkat cells with

NFAT-luciferase vector, pRL-TK vector which expresses TK

promoter-driven Renilla luciferase and 3BP2 or its mutants, and then stimulated the cells with PMA plus ionomycin Firefly luciferase activity was normalized

by Renilla luciferase activity As the results show that 3BP2 and its mutants did not affect the T cell response

to the PMA⁄ ionomycin stimulation (data not shown),

we normalized the transfection efficiencies determined

by the stimulation with PMA plus ionomycin As shown in Fig 5, although mutation on other tyrosine residues of 3BP2 did not exert appreciable effects on 3BP2-mediated NFAT activation, the Y448F mutation reduced the effect of 3BP2 on NFAT activation These results suggest that phosphorylation of 3BP2 on Tyr448 plays an important role for its function in TCR signaling

SHP-1 dephosphorylates 3BP2 in TCR signaling and negatively regulates 3BP2-induced NFAT activation

Because tyrosine phosphorylations of SHP-1 at the C-terminal residues can activate its phosphatase activity [33] and 3BP2 interacts with phosphorylated SHP-1, it is expected that the tyrosine-phosphorylated 3BP2 is a potential substrate for activated SHP-1 during their interaction The substrate characteristic

of 3BP2 for SHP-1 was primarily demonstrated in 293T cells where the tyrosine-phosphorylated 3BP2 was largely dephosphorylated by wild-type SHP-1, but not by catalytically inactive mutant SHP-1⁄ C455S (Fig 3E) To further investigate the substrate nature

of 3BP2 for SHP-1 during T cell signaling, we cotrans-fected 3BP2–myc with wild-type SHP-1, its

catalyti-3BP2-myc

3BP2-myc

0 2' 10' 0 2' 10' 0 2' 10' 0 2 10' 0 2' 10'

3BP2 WT 3BP2/Y174F 3BP2/Y183F 3BP2/Y448F 3BP2/Y485F

62 82

62 82

IP: anti-myc

WB: anti-pTyr

IP: anti-myc

WB: anti-myc

OKT3

Fig 4 Identification of Tyr448 as the major phosphorylated residue of 3BP2 in response to TCR engagement Jurkat T cells were

transfect-ed with 3BP2 or its mutants as indicattransfect-ed Forty-eight hours after transfection, the cells (5 · 10 7 ) were stimulated with anti-CD3 IgG (OKT3)

at 37
C for 0, 2 or 10 min as described in Experimental procedures Whole-cell lysates were immunoprecipitated with anti-myc IgG and the precipitated proteins were subjected to western blot analysis with anti-phosphotyrosine (pTyr) and anti-Myc IgG, respectively.

cally inactive mutant SHP-1⁄ C455S, and mutant

SHP-1⁄ Y566F, which is not capable of associating with

3BP2, into human Jurkat T cells, respectively The

tyrosine-phosphorylation status of 3BP2 was examined

following anti-CD3 IgG stimulation in the transfected

cells As shown in Fig 6, the tyrosine-phosphorylation

level of 3BP2 was dramatically reduced when

cotrans-fected with wild-type SHP-1 In contrast, neither the

catalytically inactive SHP-1 (SHP-1⁄ C455S) nor

mutant SHP-1⁄ Y566F exerted any detectable effect on

the tyrosine phosphorylation of the cotransfected 3BP2

in Jurkat cells Because Tyr448 is the major

phosphor-ylated site of 3BP2, SHP-1-mediated

dephosphoryla-tion of 3BP2 is expected to take place mainly on this

tyrosine residue These results suggest that SHP-1 via

Tyr566 recruits 3BP2 as its potential substrate for

dep-hosphorylation during TCR signaling

To further determine if the SHP-1-mediated

de-phosphorylation of 3BP2 affects its function in TCR,

we cotransfected Jurkat cells with NFAT-luciferase

reporter, 3BP2 and SHP-1 or SHP-1 mutants As

shown in Fig 7, expression of 3BP2 resulted in both

constitutive and anti-CD3 IgG-induced NFAT

activa-tion Expression of SHP-1, however, inhibited

anti-CD3-induced NFAT activation SHP-1 also nearly

completely inhibited 3BP2-mediated NFAT activation

in response to anti-CD3 stimulation in

3BP2-transfect-ed cells Furthermore, SHP-1 also inhibit3BP2-transfect-ed the

consti-tutive NFAT activation in 3BP2-transfected cells but

not the basal NFAT activity in the cells without 3BP2

transfection In contrast, the catalytically inactive mutant SHP-1⁄ C455S or mutant SHP-1 ⁄ Y566F, which abolished its interaction with 3BP2, was incapable of suppressing the 3BP2-mediated NFAT activation in 3BP2 transfected cells Taken together, these results suggest that SHP-1 negatively regulates the function of 3BP2 in TCR signaling through dephosphorylation of 3BP2 on its Tyr448 residue

Discussion

It has been reported that SHP-1 plays a negative role

in TCR signaling However, the precise mechanism by which SHP-1 regulates TCR signaling is largely unknown In this study, we reported the identification

of a novel SHP-1-interacting adapter protein 3BP2 3BP2 is composed of an N-terminal PH domain, an SH3-binding proline-rich region, and a C-terminal SH2 domain In addition to SHP-1 reported here, the SH2 domain of 3BP2 has been shown to bind to sev-eral phosphorylated proteins including ZAP70, PLCc, LAT, Grb2 and Cbl26 3BP2 was initially identified as

an Abl SH3 domain-binding protein of unknown func-tion [25] Recently, 3BP2 has been shown to interact with the Syk and ZAP70 proteins of the Syk family of tyrosine kinases In addition, 3BP2 plays a positive adapter function on basal and TCR-mediated NFAT and AP-1 transcriptional activation in human Jurkat

Fig 5 Effect of the mutation of tyrosine residues on 3BP2-induced

NFAT activation Jurkat T cells were cotransfected with

NFAT-lucif-erase reporter and 3BP2–myc or its mutants as indicated Twenty

hours after transfection, the cells were incubated with either no

addition, anti-CD3 IgG (OKT3) or PMA plus ionomycin at 37
C for

6 h as described in Experimental procedures Luciferase activity in

cell extracts was assayed and the data were normalized by the

maximal response obtained in the presence of PMA plus

ionomy-cin The results shown are means ± SE from three independent

assays performed in two separate experiments.

S 5 4 C / 1-P S

F 6 Y / 1-P S

T W / 1-P S

OKT3 - + - + - + - +

r o t c V

82

62

82

62

3BP2-myc 3BP2-myc

WB:

anti-myc

WB:

anti-pTyr

Fig 6 3BP2 is dephosphorylated by SHP-1 in activated Jurkat

T cells Jurkat T cells were transfected with 3BP2 and SHP-1 or its mutants as indicated Forty-eight hours after transfection, the cells (5 · 10 7 cell equivalents) were stimulated with anti-CD3 antibody (OKT3) at 37
C for 2 min as described in Experimental procedures Whole-cell lysates were immunoprecipitated with anti-myc IgG and the precipitated proteins were subjected to western blot analysis with anti-phosphotyrosine (pTyr) and anti-myc IgG.

T cells [26] However, the molecular mechanism by

which 3BP2 regulates TCR signal transduction remains

unclear We found that 3BP2 is a potential substrate

of SHP-1 and SHP-1 is likely to negatively regulate

3BP2-mediated NFAT activation in TCR signaling In

addition, we identified the major tyrosine

phosphoryla-tion site of 3BP2, Tyr448 Mutaphosphoryla-tion of this tyrosine

residue reduced 3BP2-mediated NFAT activation

Thus, tyrosine phosphorylation is crucial for 3BP2

function in TCR signaling and dephosphorylation of

the phosphorylated 3BP2 by SHP-1 negatively

regu-lates 3BP2 activity Tyrosine phosphorylation of 3BP2

has been also demonstrated in mast cells in response

to aggregation of high affinity IgE receptor [29,34], in

NK cells upon stimulation with anti-FcR IgG [28] and

recently in T cells upon TCR activation [35] In

RBL-2H3 mast cells, phosphorylation of Tyr448 of 3BP2

creates a binding site for the SH2 domain of Lyn, a

Src family protein tyrosine kinase, and interaction of

Lyn with 3BP2 positively regulates the kinase activity

of Lyn [34] In NK cells, Tyr183 of 3BP2 is

phosphor-ylated and binds Vav and PLCc during activation of

NK cells through natural cytotoxicity receptors and

this phosphorylation is necessary for the enhancement

of natural cytotoxicity by 3BP2 [28] Qu et al [35]

recently found that both Tyr183 and Tyr448 could be

phosphorylated in response to TCR activation by

anti-CD3 IgG together with PMA However, in our study, mutation of Tyr183 to phenylalanine did not have obvious effect on 3BP2 phosphorylation in response to anti-CD3 IgG-induced TCR activation in the absence

of PMA This suggests that both cross-linking of TCR and direct activation of protein kinase C are required for the phosphorylation of Tyr183 of 3BP2 Thus, it is likely that 3BP2 is selectively activated in response to various upstream signalings

Usually, the SH2 domain of SHP-1 associates with tyrosine-phosphorylated proteins during its interac-tions with other signal molecules Interestingly, the association of 3BP2 with SHP-1 is through the SH2 domain of 3BP2 and the tyrosine-phosphorylated phosphatase SHP-1 contains three tyrosine residues (Tyr538, Tyr543 and Tyr566) in its C-terminal tail It has been reported that at least two of these tyrosine residues could be phosphorylated in response to the stimulation of T cell-receptor [19], CSF receptor and c-Kit [36] However, the biochemical consequence and physiological significance of tyrosine-phosphorylation

on SHP-1 remain elusive It has been suspected that tyrosine phosphorylation of SHP-1 may regulate its phosphatase activity as observed in other phosphatases [33] In this study, however, we found that phosphory-lation of SHP-1 at tyrosine residues on its C-terminal tail confers to the phosphatase an ability to recruit adapter protein 3BP2 and thereby affects signaling Site-directed mutation experiments further revealed that 3BP2 interacts with SHP-1 through its phosphor-ylated Tyr566 residue The sequence surrounding Tyr566 (Tyr566⁄ Glu567 ⁄ Asn568) is strikingly similar

to the optimal 3BP2 SH2 domain-binding motif (Tyr⁄ Glu⁄ Asn) [37]

In this study, we demonstrated that SHP-1 interacts with 3BP2 through the tyrosine-phosphorylated C-ter-minal segment of the former and the SH2 domain of the latter This interaction allows 3BP2 to be dephosphoryl-ated more efficiently by the catalytic domain of SHP-1

We thus defined a novel function for the C-terminal seg-ment of SHP-1 It has been known that tyrosine phos-phorylation of SHP-1 at its C-terminal segment also initiates interaction with adapter protein Grb2 and mSOS [23] However, this does not seem involve the cat-alytic activity of the enzyme and thus the physiological meaning remains unclear Furthermore, like SHP-1, SHP-2 is also known to be phosphorylated at its C-ter-minal segment However, because these two enzymes share minimum sequence identity at their C-termini, in contrast to high sequence homologies in their SH2 and catalytic domains, we believe this may allow the enzymes interact with distinct proteins This may explain the often-opposite functions of the two enzymes

r

o

t

3

1-P S

+ P 3

1-P S

+ P 3

S / C 1-P S

+ P 3

F 6 Y 1-P S

S / C 1-P S

F 6 Y 1-P S

Fig 7 SHP-1 negatively regulates 3BP2-induced NFAT activation.

Jurkat T cells were cotransfected with the NFAT-luciferase reporter

gene and empty vector, 3BP2–myc and SHP-1 or its mutants as

indicated Twenty hours after transfection, the cells were incubated

with either no addition, anti-CD3 IgG (OKT3) or PMA plus

iono-mycin at 37
C for 6 h as described in Experimental procedures.

Luciferase activity in cell extracts was assayed and the data were

normalized by the maximal response obtained in the presence of

PMA plus ionomycin The results shown are means ± SE from

three independent assays performed in two separate experiments.

Experimental procedures

Reagents and antibodies

Rabbit anti-SHP-1 polyclonal IgG was generated as

des-cribed previously [38] Mouse anti-SHP-1 and anti-SHP-2

monoclonal IgG were obtained from Transduction

Labor-atories (Lexington, KY) An anti-(human CD3-a) (OKT3)

monoclonal IgG was purified from the culture medium

of OKT3 hybridomas by protein A–Sepharose affinity

chromatography Rabbit anti-(mouse IgG) was obtained

from BD Biosciences Pharmingen (San Diego, CA)

Anti-phosphotyrosine (4G10) and anti-myc (9E10) monoclonal

IgG were purchased from Santa Cruz Biotechnology (Santa

Cruz, CA) Anti-hemagglutinin (anti-HA) monoclonal IgG

(clone 12CA5) was prepared from the culture medium of

hybridomas (ATCC, Manassas, VA) Anti-(mouse

IgG-horseradish peroxidase) and anti-(rabbit IgG-IgG-horseradish

peroxidase) were from Bio-Rad Laboratories (Hercules,

CA) Nitrocellulose membrane Hybond-ECL was from

Amersham Pharmacia Biotech (Little Chalfont, UK)

West-ern Lightning Chemiluminescence Reagent kit was

pur-chased from Perkin–Elmer Life Sciences Inc (Boston, MA)

Protease inhibitor cocktail tablets were from Roche

Diag-nostics (Mannheim, Germany)

Plasmids

Plasmids expressing SHP-1 and its mutants were

construc-ted as described previously [31,38] Myc-tagged 3BP2

plas-mid and HA-tagged 3BP2 plasplas-mid were constructed by

amplifying the full-length 3BP2 encoding region using total

RNA from Jurkat cells and inserting the amplified PCR

product into the HindIII site of pcDNA3.1⁄ myc-His (–) C

vector (Invitrogen, Carlsbad, CA) and pACTAG-2 vector

(kindly provided by M Tremblay, McGill University)

3BP2 mutants were generated by PCR-based mutagenesis

Fyn kinase in pRK5 vector was a kind gift from S Stamm

(Max Planck Institute of Biochemistry, Germany) Lck and

its activated mutant (Lck⁄ Y505F) constructs were kindly

provided by B Sefton and G Chiang (The Salk Institute for

Biological Studies) NFAT-luciferase reporter was kindly

provided by G Crabtree (Stanford University School of

Medicine)

Cell culture and transfection

293T cells and Jurkat T cells were maintained as described

previously [31,39] 293T cells were transfected with different

sets of plasmid DNAs using standard calcium phosphate

pre-cipitation methods In some experiments, the transfected cells

were treated with 0.5 mm sodium pervanadate in regular

medium for 30 min Sodium pervanadate was prepared by

mixing 100 mm sodium orthovanadate (Sigma, St Louis,

MO) and 50 mm H2O2(Sigma) and incubating the mixture

at room temperature for at least 30 min Jurkat T cells (107

in 400 lL of medium) were transfected with 20–25 lg DNA

by electroporation using a gene pulser (BTX Corp., San Diego, CA) at 260 V for 50 ms Empty vector was added to some samples to make an equal amount of DNA in each transfection Forty-eight hours after transfection, Jurkat

T cells were washed and suspended in NaCl⁄ Pi For stimula-tion, cells were incubated with 2 lgÆmL)1of OKT3 on ice for

5 min and then with 10 lgÆmL)1of rabbit anti-(mouse IgG) for an additional 5 min The samples were then incubated at

37
C for the indicated times

Yeast two-hybrid screen The cDNA encoding the full-length of SHP-1 with Cys455

to Ser mutation (SHP-1-C455S) was PCR-amplified from the corresponding plasmid [40] and cloned in-frame downstream

of the DNA binding domain of Lex A in pBTM-116-src vec-tor [30] to form the bait construct (Lex A–SHP-1–C455S) [31] The human Jurkat cDNA library expressed as fusion proteins with the activation domain of GAL4 in the pACT2 vector was obtained from Clontech Laboratories (Palo Alto, CA) The bait DNA and library DNA were sequentially transformed into yeast strain L40a and 1.1· 107

primary transformants were screened for growth on medium lacking leucine, tryptophan and histidine The positive colonies were further screened for the expression of b-galactosidase The plasmid DNA was recovered from His+⁄ LacZ+

colonies and identified by DNA sequencing

Immunoprecipitation and immunoblot analysis Immunoprecipitation and western blot experiments were carried out as described previously [31] Briefly, cells were washed with cold NaCl⁄ Pionce and lysed in a lysis buffer containing 50 mm Hepes (pH 7.4), 150 mm NaCl, 1% Tri-ton X-100, 5 mm b-mercaptoethanol, 0.5 mm vanadate and

an EDTA-free mixture of protease inhibitors The samples

were centrifuged at 20 000 g for 10 min at 4
C An aliquot

of this whole-cell lysate was removed and the remaining lysate was subjected to immunoprecipitation For immuno-precipitations, cell lysates were incubated with optimal con-centrations of antibodies for 2 h at 4
C, followed by incubation with 50 lL of 50% suspension of Protein A– Sepharose CL-4B beads for 1 h The Sepharose CL-4B beads were washed at 4
C with lysis buffer four times The proteins were resolved on a SDS⁄ PAGE gel and transferred

to nitrocellulose membranes (Hybond-ECL) The mem-branes were blocked with 5% milk in Tris-buffered saline (TBS) (pH 7.6) overnight and then incubated with the first antibodies for 2 h After washing four times with TBS con-taining 0.05% Tween-20 (TBS-T), the membranes were incubated with the second antibody conjugated to horserad-ish peroxidase for 1 h and then washed four times with TBS-T The blots were developed using the western

Lightning Chemiluminescence Reagent kit (Roche)

accord-ing to the manufacturer’s instruction

Expression, purification of GST fusion proteins,

and GST pull-down

For the construction of a plasmid expressing GST)3BP2–

SH2 domain fusion protein, the cDNA fragment encoding

amino acid residues 452–561 of 3BP2 was amplified by

PCR and inserted into pGEX-5X1 vector (Amersham

Phar-macia Biotech) Construction of GST–SHP-1–2SH2 has

been described previously [41] Fusion proteins were

expressed in Escherichia coli strain DH5a by induction with

25 lm isopropyl-d-thiogalactopyranoside at 25
C for 16 h

and purified as described previously [42] For binding

assays, glutathione–Sepharose beads with 1 lg of bound

GST or GST fusion protein were incubated at 4
C for 2 h

with 1 mL of cell lysates The beads were washed four

times with the lysis buffer and the bound proteins were

analyzed by SDS⁄ PAGE and western blot

NFAT reporter assay

Jurkat T cells (2· 107

) were transiently transfected with

5 lg of pNFAT-luciferase and 20 lg of indicated plasmids

by electroporation Twenty hours after transfection, cells

were aliquoted into a 12-well plate in 1 mL of culture

med-ium and triplicate samples were either left unstimulated,

stimulated with OKT3 (2 lgÆmL)1) or with PMA

(50 ngÆmL)1) plus 1 lm ionomycin for 6 h Cells were then

harvested and washed with 1 mL of NaCl⁄ Pi Harvested

cells were lysed and assayed for luciferase activity as

previ-ously described [40] Luciferase activity was determined in

triplicate for each experimental condition and normalized

by the transfection efficiencies determined by the maximum

stimulation with PMA plus ionomycin

Acknowledgements

This study was supported in part by the National

Sci-ence and Engineering Research Council of Canada

Grant 0GP0183691 We thank Dr J.A Cooper for

kindly providing the pBTM-116-src vector, Dr S

Stamm for Fyn kinase vector, Dr B.M Sefton and Dr

G.G Chiang for Lck constructs, Dr G.R Crabtree for

NFAT-luciferase reporter and Dr M Tremblay for

pACTAG-2 vector

References

1 Hermiston ML, Xu Z, Majeti R & Weiss A (2002)

Reci-procal regulation of lymphocyte activation by tyrosine

kinases and phosphatases J Clin Invest 109, 9–14

2 Zamoyska R, Basson A, Filby A, Legname G, Lovatt

M & Seddon B (2003) The influence of the Src-family kinases, Lck and Fyn, on T cell differentiation, survival and activation Immunol Rev 191, 107–118

3 Chu DH, Morita CT & Weiss A (1998) The Syk family

of protein tyrosine kinases in T-cell activation and development Immunol Rev 165, 167–180

4 Zhang W, Sloan-Lancaster J, Kitchen J, Trible RP & Samelson LE (1998) LAT: the ZAP-70 tyrosine kinase substrate that links T cell receptor to cellular activation Cell 92, 83–92

5 Jackman JK, Motto DG, Sun Q, Tanemoto M, Turck

CW, Peltz GA, Koretzky GA & Findell PR (1995) Molecular cloning of SLP-76, a 76-kDa tyrosine phos-phoprotein associated with Grb2 in T cells J Biol Chem

270, 7029–7032

6 Samelson LE (2002) Signal transduction mediated by the T cell antigen receptor: the role of adapter proteins Annu Rev Immunol 20, 371–394

7 Koretzky GA (2003) T cell activation I: proximal events Immunol Rev 191, 5–6

8 Mustelin T, Rahmouni S, Bottini N & Alonso A (2003) Role of protein tyrosine phosphatases in T cell activa-tion Immunol Rev 191, 139–147

9 Zhang J, Somani AK & Siminovitch KA (2000) Roles of the SHP-1 tyrosine phosphatase in the negative regula-tion of cell signalling Semin Immunol 12, 361–378

10 Shultz LD, Schweitzer PA, Rajan TV, Yi T, Ihle JN, Matthews RJ, Thomas ML & Beier DR (1993) Muta-tions at the murine motheaten locus are within the hematopoietic cell protein-tyrosine phosphatase (Hcph) gene Cell 73, 1445–1454

11 Pani G, Fischer KD, Mlinaric-Rascan I & Siminovitch

KA (1996) Signaling capacity of the T cell antigen receptor is negatively regulated by the PTP1C tyrosine phosphatase J Exp Med 184, 839–852

12 Lorenz U, Ravichandran KS, Burakoff SJ & Neel BG (1996) Lack of SHPTP1 results in Src-family kinase hyperactivation and thymocyte hyperresponsiveness Proc Natl Acad Sci USA 93, 9624–9629

13 Zhang J, Somani A, Yuen D, Yang Y, Love P & Simi-novitch K (1999) Involvement of the SHP-1 tyrosine phosphatase in regulation of T cell selection J Immunol

163, 3012–3021

14 Plas DR, Williams CB, Kersh GJ, White LS, White JM, Paust S, Ulyanova T, Allen PM & Thomas ML (1999) Cutting edge: the tyrosine phosphatase SHP-1 regulates thymocyte positive selection J Immunol 162, 5680–5680

15 Johnson KG, LeRoy FG, Borysiewicz LK & Matthews

RJ (1999) TCR signaling thresholds regulating T cell development and activation are dependent upon SHP-1

J Immunol 162, 3802–3813

16 Carter JD, Neel BG & Lorenz U (1999) The tyrosine phosphatase SHP-1 influences thymocyte selection by