Download - Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

Transcript
Page 1: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

TitleSOCS-1/SSI-1-Deficient NKT Cells Participate inSevere Hepatitis through Dysregulated Cross-TalkInhibition of IFN-γ and IL-4 signaling In Vivo

Author(s) 仲, 哲治

Citation

Issue Date

Text Version ETD

URL https://doi.org/10.18910/43235

DOI 10.18910/43235

rights

Note

Osaka University Knowledge Archive : OUKAOsaka University Knowledge Archive : OUKA

https://ir.library.osaka-u.ac.jp/

Osaka University

Page 2: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

lmmunity, Vol.14,535-545, May,2001,Copyright◎2001 by Cell Press

SOCS-1/SSI-1-Deficient N KT

Severe Hepatitis throughInhibition of IFN一γand IL-4

ellS ar寸ICIpa e 

In

Dysregulated Cross-Talk   Signaling In Vivo

Tetsuji Naka,1,5 Hiroko Tsutsui,3

Minoru Fulimoto,1 Yoshinori Kawazoe,1

Hidetsugu Kohzaki,1 Yoshiaki Morita,¶

Reiko Nakagawa,2 Masashi Narazaki,1

Keishi Adachi,3 Tomohiro Yoshimoto,3

Kenli Nakanishi,3 and Tadamitsu Kishimoto4

1Department of Molecular Medicine

2Department of Microbiology

Osaka University Graduate School of Medicine

3Department of lmmunology and Medical Zoology

Hyogo College of Medicine

40saka University Graduate School

2-2,Yamadaoka, Suita City

Osaka 565-0871Japan

Summary

SupPressor of cytokine signaling-1 {SOCS-1}, also

known as STAT-induced STAT inhibitor-1(SSI-1), is a

negative feedback molec副e for cytokine signaling,

and its in vivo deletion induces fulminant hepatitis.

However, elimination of the STATI or STAT6 gene or

deletion of NKT cells substantially prevented severe

hepatitis in SOCS-1-deficient mice, while administra-

tion of IFN-Y and IL-4 accelerated its development.

SOCS-1 deficiency not only sustained IFN-YllL-4 sig-

naling but also elimmated the cross-inhibito「y action

of IFN-V on IL-4 signaling. TheSe results sug9est that

SOCS-1 deficiency-induced persistent activation of

STATI and STAT6, which would be inhibited by SOCS-1

undemormal cond魔ions, may induce abnormal activa-

tion of NKT cells, thus leading to lethal pathological

changes in SOCS-1-deficient mice.

lntroduction

Most hematopoietic cells are exposed to various cyto-

kines and are systematically regulated by a complicated

network of cytokines. Signaling through Janus kinase

(JAKγSignal transducers and activators of transcription

(STAT)pathway is one of the most important signal path-

ways activated immediately after cytokine stimulation

(for a review, see Darnell et al.,1994;Kishimoto et al.,

1994;lhle,1995). SupPressor of cytokine signaling-1

(SOCS-1), also known as STAT-induced STAT inhibi-

tor-1(SSI-1), is an intracellular protein that inhibits JAK-

mediated cytokine signaling by binding to JAKs(Starr

et al.,1997;Endo et al.,1997;Naka et al.,1997;for a

review, see Naka et al.,1999;Krebs and Hilton,2000;

Chen et al.,2000;Nicola and Greenhalgh,2000;Yasu-

kawa et al.,2000;Naka et al.,2001;Starr,2001).Although

SOCS-1 can be induced by various cytokines, such as

mterferon(IFN)-V, intedeukm(IL)-4, IL-2, and IL-6, the

target molecules of SOCS-1 among members of the JAK

family have not been precisely identified yet(Starr et

5Correspondence:naka@imed3.med.osaka-u.ac.lp

al.,1997;Endo et al.,1997;Naka et al.,1997;Helman

et al.,1998;Narazaki et al.,1998;Yasukawa et al.,1999;

Nicholson et al.,1999;Losman et al.,1999;Pezet et al.,

1999;Morita et al.,2000;Fujimoto et al.,2000;for a

review, see Naka et al.,1999;Krebs and Hilton,2000;

Chen et al.,2000;Nicola and Greenhalgh,2000;Yasu-

kawa et al.,2000).

  SOCS-1-deficient mice(SOCS・1 KO mice)are born

healthy but with growth disclose various kinds of abnor-

malities, including growth retardation, thymic atrophy,

fulminant hepatitis, with serious fatty degeneration and

lung injuries with infiltration of mononuclear cells, and

all die within 3 weeks after birth(Starr et al.,1998;Naka

et al.,1998). T cells from SOCS-1 KO mice show pro-

bnged signaling in response to IL-40r IFN-V, demon-

strathg that SOCS-1 is an inhibitory factor required for

the cessation of IL-40r IFN.V signaling in vivo(Starr et

al.,1998;Naka et al.,1998). Recently, it has been re-

ported that IFN-y-deficient SOCS-1 KO mice are free

from these severe pathological changes(Alexander et

al.,1999;Marine et a1.,1999), indicating that SOCS-1

indeed negatively regulates overshooting of IFN-V sig-

naling. This study provided a new insight into the essen-

tial rde of the IFN-V侶OCS-1 system m the devebpment

and maintenance of the immune system. However, they

do not seem to imply that the lethal changes in SOCS-1

KO mice are solely attributableto oversignaling of lFN-V,

because SOCS-1 also negatively regulates various kinds

of signalings(Naka et al.,1997,1998;Sakamoto et aL,

1998;Song and Shuai,1998;Adams et al.,1998;Hansen

et al.,1999;Losman et al.,1999;Ram and Waxman,

1999;Pezel et al.,1999;Tomic et al.,1999;Morita et

al.,2000;Kawazoe et a1.,2001). l ndeed, a recent study

revealed that IFN-V-induced SOCS-1 plays a critical role

in inhibiting STAT6 activation in vitro(Dickensheets et

al.,1999;Venkataraman, et al.,1999). Thus, without

cross.talk inhib忙ion via SOCS・1,IL-4 and IFN・ッsignal-

ings might be independently, occasionally at the same

time, and persistently transduced to abnormally activate

|ymphocytes, causing injuries of immune organs and

liver. lt is theref◎re▲mportant to d創ermine whether

SOCS-1 deficiency induces pathological changes byabolishing a cross-talk inhibition of IFN-V and IL-4 sig-

nalings or simply by amplifying one signal. lnthis study,

we first tested whether deletion of the STATI gene in-

stead of the IFN-V gene results in similar prevention of

fulminant hepatitis in SOCS-1 KO mice. Second, we

analyzed whether deletion of the STAT6 gene also pre-

vents these pathological changes, as we wished to clar-

ify whether SOCS-1 deficiency facilitates simultaneous

overshooting of IL・4, which, in combination with IFN-V

signaling, induces lethal pathologies. Lastly, we exam-

ined whether SOCS-1 in hepatic NKT cells acts as a

pivotal佗gulatmg factor that limits their hepatocyte-kill-

ing action. This is the first in vivo study to show that

SOCS-1 is a critical molecule for cross-talk inhibition of

lFN-・y and IL-4 signalings, resulting in suppression of

the overactivation of hepatocytotoxic NKT cells.

Page 3: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

lmmunity536

a100

75

05

石≧とコω

25

b

ays)

(O)一工〇一Φ●ロ〉唱Om■

1 3 7 9 12   15   21   28   35 Days

CWT SOCS・1 KO SOCS・1/STAT6 DKO SOCS・1/STATI DKO

Thymus/Body 2Wwei ht ratio

 0.821

SD:0.0213

 0.259

SD:0.0249

 0.502

SD:0.0240

 0.435

SD:0.0251

Liver/Body  2Wwei ht「atio

 3.170

SD:0.1202

 4.431

SD:02340

 3,404

SD:0.1488

 3.640

SD:0.1608

Figure 1. Disruption of STATI or STAT6 Prevents Per}natal Death of SOCS-1 KO Mice

Mice with various gene mutations were kept under SPF conditions. Survival rate(A)and body weight(B}were measured until 90 and 35 days

after birth, respectively, Two weeks after birth, thymus and liver were sampled fπ)m varbus types of mice, and the ratios of the weight of

these organs to body weight were calculated(C). Data in(A)are data of 20 mice in each experimenta|group and are representative of two

independent experiments with similar resuns. Data in(B)and(C)show mean±SD for five mice from each group and are representative of

three independent experiments with simi|ar results.

Results

STATI or STAT6 Deficiency Protects SOCS-1 KO

Mice from Various Kinds of Abnormalities

We generated SOCS-1 KO mice Iacking STAT1(SOCS-

11STATI DKO mice)or SOCS-1 KO mice lacking STAT6

(SOCS-11STAT6 DKO mice)by deletion of the STATI or

STA丁6 gene in SOCS-1 KO mice. All the SOCS-1 KO

mice died within 3 weeks after birth. ln contrast, mor-

tality was significantly reduced in SOCS-1/STATI or

SOCS-11STAT6 DKO mice(Figure I A). As also pre-

viously reported(Starr et al.,1998;Naka et al.,1998),

SOCS-1 KO mice showed a marked growth retardation,

while SOCS-11STATI and SOCS-1/STAT6 DKO micegrew larger than SOCS-1 KO mice but were still smaller

than wild-type(WT)mice(Figure l B}. Moreover, severe

thymic atrophy was partly improved, and hepatomegaly

was almost completely eliminated in SOCS-11STATl

and SOCS-11STAT6 DKO mice(Figure I C).

  Elimination of either STATI or STAT61ed to improve-

ment in pathological changes in various organs of

SOCS-1 KO mice. Histological findings and determina-

tion of the numbers of various cells on day 10after birth

revealed that thymic atrophy had partly improved in

SOCS-1/S丁ATI or SOCS-1/STAT6 DKO mice(Figures

2A and 2B),and that complete histobgical restoration

had been achieved in the livers of both types of DKO

mice(Figure 2C). SOCS-1 KO mice showed fulminant

hepat面s characterized by severe fatty degeneration and

necrosis with massive lymphocyte infiltration(Starr et

al.,1998;Naka et al.,1998;Alexander et al.,1999). ln

contrast, the livers from SOCS-11STATI DKO mice like

those from SOCS’111FN・V DKO mice(Alexander et aL,

1999;Marine et al.,1999)remained almost intact except

for infiltration by a small number of lymphocytes(Figure

2C), thus confirming the pathological role of IFN-V in

this liver injury. To our surprise, SOCS-11STAT6 DKO

mice also had almost intact liver tissue, which suggests

that SOCS-1 might prevent development of fulminant

hepatitis by mhibithg of simultaneous transduction of

both IFN・ッand IL-4 signals.

Relevant Role of SOCS・1-Deficient Lymphocytes

in Severe Hepatitis Observed in SOCS-1 KO Mice

As SOCS-1 is strongly expressed in lymphoid organs,

especially in T cells(Starr et a1.,1997;Naka et al.,1997;

Marine et al.,1999), we compared the proportions of

splenic T cells expressing CD69, an activation marker,

among micewith various genetic mutations. SOCS-1 KO

mice showed an increase inthe population ofactivated T

cells,while SOCS-11STATI orSOCS-11STAT6 DKO mice

Page 4: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

SOCS-1/SS|-1 as a Cross一丁alk lnhibitor537

a b

S

SOCS・1 KO

SO

:・巖怠

.’

10

8

    6           4

(トロx)8§。皇ξ

2

0

C

Figure 2. lmprovement of Tissue Anomalies in SOCS-1 KO Mice by Elimination of STATI or STAT6

Thymus(A and B)and liver(C)were sampled from mice with various genotypes 2 weeks after birth for histological study(A and C)and cell

count(B). Thymocytes were isolated, and the cells were counted(B).(A)Hematoxylin/eosin(HE)staining, original magnification:×25.(C)HE

staining(left four panels}and Sudan川staining(right four panels), original magnification:×80. Data in(A)and(C)are representative of 10

independent experiments with similar results. Data in(B)show mean±SD of five耐ce from each group and are representative of five

independent experiments with simi|ar results.

Page 5: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

lmmunity538

40

30

    20

(ポ)●OO、OOOO

10

0

(ポ}OOO、OOOO

40

30

20

10

0

Figure 3. Disruρtion of STATI or STAT6 Re-

stores Spontaneously Overact}vated T Cells

in SOCS-1 KO Mice

Splenic CD4⊥cells(left panel)or CD8+cells

(right paneりwere isolated from mice wlth var-

ious genotypes by MACS. The proportion of

CD69 expression for each cell group was

measured by FACS. Data are representative

of 10 independent experiments with similar

results.

contained a reduced but still higher proportion of spon-

taneously activated T cells in their spleen when com-

pared with, respectively, the spleens from SOCS-1 KO

and WT mice(Figure 3).

  Next, we investigated thecellularmechanism underly-

ing the pathological changes in SOCS-1 KO mice, with

special attention to severe liver injuries・ To clarify

whetherthese lethal alterations are dueto abnormalities

in lymphocytes, in stromal cells sustaining lymphocyte

maturation, or in parenchymal cells, we transferred he-

matopoietic progenitor cells from the SOCS-1 KO fetal

livers(E16.5days)into RAG2-deficient(RAG2 KO)mice.

All the RAG2 KO mice reconstituted with SOCS-1-defi-

cient lymphocytes died within 7 weeks after transplan-

tation, while RAG2 KO mice reconstituted with WT

lymphocytes survived normally. Furthermore, SOCS-1-

deficient chimeric mice showed various characteristics

similar to those of SOCS-1 KO mice, such as re|ative

weight loss inthe thymus and spleen and relative weight

gain in the liver(Figure 4A). lndeed, histological findings

forvarioustissues from SOCS-1-deficient chimeric mice

obtained 6 weeks after transplantation revealed disap-

pearance of the thymic cortex(data not shown)and

severe liver injuries(Figure 4B). Moreover, the critical

involvement of SOCS-1-deficient lymphocytes in the in-

duction of these pathological changes is well docu-

mented by a recent study in which SOCS-1-deficient

bone marrow ce|ls, when used to reconstitute Jak3-

deficient mice lacking T cells, B cells, and NK cells,

replicated the phenotype seen inSOCS-1-deficient mice

(Marine et al.,1999). lncontrast, RAG2-deficient SOCS-1

KO mice that lacked T cells and B cells but contained

NK cells showed healthy growth(Marine et al.,1999).

Thesefindings incombination with ourresults presented

here strongly indicate that abnormal lymphocytes cause

the multiple pathological changes in SOCS-1 KO mice.

Spontaneously Activated Hepatic NKT Cells Cause

Liver lnluries in SOCS-1 KO Mice

Next, we identified the celltypes and effector molecules

responsible for severe liver injuries in SOCS-1 KO mice・

Since RAG2-deficient SOCS-1 KO mice avoid severe

liver injury(Marine et al.,1999), NK cells seem to be

insignificant as hepatocytotoxic effector cel|s. As the

liver has a unique immune system characterized by the

presence of abundant NKT cells, a potent cytotoxic cell

population(Kawamura et al.,1998), we focused on the

role of hepatic NKT cells in this liver injury. Hepatic

lymphocytes from SOCS-1 KO mice killed hepatocytes

from syngeneic WT mice in a per「orin-dependent but

Fas/Fas ligand(Fas L)-independent manner(Figure 5A).

The hepatocytotoxicity of SOCS-1-deficient hepatic

lymphocytes was reduced after the deletion of NK1.1÷

cells consisting of NK cells and NKT cells(Figure 5B).

NK cells and NKT cells show different susceptibilities

to in vivo treatment with anti-asialo GM1,possibly due

to the difference in their expression of asialo GM1(So-

noda, et al.,1999}. By taking advantage of this differ-

ence, we could selectively deplete NK cells without af-

fecting NKT cells(Figure 5C-2). This treatment did not

eliminate the hepatocytotoxicity of SOCS-1-deficient

hepatic lymphocytes(Figure 5C-1). These results sug-

gest that NKT cells may belong to the effectorcell popu-

lations killing hepatocytes. lnfact, SOCS-1 was induced

in WT NKT cel|s upon stimulation with IFN-V(data not

shown),andthenumberof NKTcells(CD3/NK1.1 doublepositive cells)was markedly elevated in the liver of

SOCS-1 KO mice compared with that of WT mice(Figure

5D). Furthermore, more than 65%of SOCS-1-deficient

NKT cells in the liver spontaneously expressed CD69,

while only less than 20%of WT hepatic NKT cells did

so (Figure 5E). ln contrast, both SOCS-1/STATI and

SOCS-1/STAT6 DKO mice contained comparable num-ber of NKT cells and similar levels of their CD69 expres-

sion in their livers as compared to WT mice(data not

shown).

  To investigate whether SOCS-1-deficient NKT cells

have the potential to cause such severe liver injury in

vivo, we administeredα一galactoceramide(α一GalCer), a

selective activator of NKT cells to pre-onset SOCS-1

KO mice. As shown in Figure 5F, SOCS-1 KO mice but

not WT mice suffered from fulminant hepatitis 12 hr

Page 6: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

SOCS←1/SSI輿1 as a Cross-Talk lnhibitor539

a

SOCS.1+ノ+in RAG2≠ S◎CS.f’in RAG2イ・

Thymus/Body                 6Wweight ratio

  0.227

(+∫-0.0188)

  0.072

(+1・0.0072)

Sple帥1Body   6Wweight ratio

  t427(+1・0.1577)

  0.694

(+ρ0.0724}

Liver IBody

weight ratio    6W  4.460

(+∫・02408)

  6.749

(尋一〇.2646)

b

SOCS-1+1÷in RAG2・ノ・ SOCS-1イ・in RAG2・ρ

Liver(6weeks)

Figure 4. RAG2 KO Mice Reconstituted with SOCS-1-Deficient Lymphocytes Spontaneously Develop Fulminant Hepatitis

(A)RAG2 KO mice were reconstituted with fetal liver stem cellsfrom SOCS-1 KO(SOCS-1+in RAG2-/一)or WT mice(SOCS-1+ノ+in RAG2-/).

Six weeks a行er reconstitution, thymus, spleen, and liver were sampled to measure their weight, and the ratios of the weight of these organs

to body weight were calculated. Data show mean±SD of 10mice in each group and are representative of three independent experiments

with similar results.

(B)Six weeks after reconstitution,[iver was sampled for histological study. HE staining, original magnification:×80. Data are representative

of five independent experiments with similar results.

after theα一GalCer challenge(Figure 6). Thus, SOCS-1-

deficient NKT cells that are highly susceptible to the

stimulation withα一GalCer might have the capacity to

cause severe liver injury upon the apPropriate stimuli,

including cytokines, in vivo, although it is still elusive

thatα一GalCer reactive SOCS-1-deficient T cells solely

mediate the hepatic pathological changes in SOCS-1

KO mice. These results suggest that liver inluries ob-

served in SOCS-1 KO mice might be due to the anoma-

lously activated NKT cells.

Page 7: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

lmmunity540

a b C・1

30.

旦竺」ポ

10

1

 

 

 

 

 

 

 

 

 

    二

 

 

 

   (V.、

  \、礼

  \ ・、

丁「r]..

o 一   三  三5  1;」5  人‘

1

           ユ゜1                                                                                         

         

 1

     

                                                                                             

 1

           、。1

          三 1

_一  ま1

つコ   れ 

::言二: i§§ミヨi§:鶉含L    ,0 11

     

       

      

       

      

       

      

       

      

       

      

       

      

       

      

       

      

       

      

       

      

               1

                                                                                                     1

、品一__ユ   ‘5

   1

    \    +肝     、.    ’≡ll;・1:3-【1-・・一

   丁  \]L_五

               冨∨百

D ・ 一 一「一一一〕 _    50    25   12.5   6    コ

蔓1。、

                     l

                     l

                    

 。↓

㌧1

      \、王

                      、「」1

一ロー一く一【}一一一〔,

   50    25   ’2.~   6

τ灘蕊M1、

舌r

  ’....・董

 、}一つ 一 3   15

c-2

N

PBS寸

口めeN竺’eDo↑

al川・a写ial(,GMI

d‘0、

08

U

Ω

89竺藁「ξ≡5ト菱

o.o.

e1

88

60

  

  

  

  

     コ

      冊

⌒S}」×Z、伊ψOO

ず20

L

T

CI)」

Figure 5. Hepatocytotoxicity of Hepatic NKT Cel|s in SOCS-1 KO Mice

(A)Hepatic|ymphocytes were isolated from WT or SOCS-1 KO mice and were incubated with 100 nM concanamycin A(CMA)or 20μg/ml of

anti-murine Fas L for l hr, and their hepatocytotoxicity was determined by 4 h{5℃d release assay.

(B)NK1.1’cells were removed from hepatic lymphocytes by MACS and their hepatocytotoxicity was determined.

(C-1)Hepatic lymphocytes were treated twice with anti-asiab GMI and complement for depletion of NK cells, and their hepatocytotoxicity

was determined.(C-2)Surface phenotypes of the cells before and after the treatment were determined by FACS.

(D)Hepatic lymphocytes from WT or SOCS-1 KO mice were incubated with biotinylated anti-NK1.1 followed by Cy-streptavidin and FITC-

conjugated anti-CD3, and the NKT cell propoπion(NK1.1↓CD3⊥cell population)was calculated.

(E)Hepatic lymphocytes were incubated with biotinylated anti-NKtl followed by Cy-streptavidin, FITC-conlugated anti-CD3, and PE-conju-

gated anti-CD69. The proportion of CD69 cells gated on NK1.11CD3→cells was calculated. Data show mean±SD of trゆlicate and are

representative of three independent experiments with similar results(A, B, and C-1). Data are representative of five independent experiments

with similar results(C-2, D, and E).

lnvolvement of Simultaneous Signaling of IFN-V/IL-4

in NKT Ce{ls in Severe Liver lnluries

As previous{y reported, treatment of thymocytes from

SOCS-1 KO mice with IL-4 resulted in Iong-term tyrosine

phosphorylation of STAT6, indicating that SOCS-1 can

inhibit STAT6 tyrosine phosphorylation in vivo(Naka et

a「.,1998).This is also the case for STATI signaling. As

shown in Figure 7A, tyrosine phosphorylation of STATl

was abnormally extended in SOCS-1-deficient thymo-

cytes as compared with WT(Figure 7A}, indicating that

disruption ofSOCS-1 sustains activation of both STAT1-

and STAT6-mediated signalings.

  Next, we investigated whether IFN-v-hduced SOCS-1

is essential for negative regulation of IL-4 Signaling as

well as its own signaling. To test this possibility, we

cultured splenic T cells with IFN-V followed by stim-

ulation with IL-4 and determined the tyrosine phos-

phorylation of STAT6. As shown in Figure 7B, IFN-V-pre-

stimulated SOCS-1-deficient T cells did show tyrosine

phosphorylation of STAT6 in response to IL-4, whereas

the subsequent stimulation did not induce STAT6 tyro-

sine phosph明11ation in WT T cells. An attempt to make

the same comparison by using hepatic NKT cells failed

because of the difficulty in obtaining a sufficient number

of NKT cells from VVT「ivers. A「though we performed the

reverse procedure, we could not obtain the same results

(data not shown). These results suggest that SOCS-1

induced by IL-4 might be insufficient to inhibit STATl

phosphorylation and/or signaling supPressor(s)other

than SOCS-1 might negatively regulate the lL4-induced

inhibition of STATI phosphorylation.

  Finally, we investigated whether SOCS-1-deficient

NKT cells directly cause severe liver injury via dysregu-

lated signalings of IFN-V and IL-4. Pre-onset SOCS-1

KO mice administered with IFN-V and IL-4 resulted in

severe liver injuries similar to those seen in onset SOCS-1

KO mice(Figure 7C}.1n contrast, NKINKT cell-depleted

pre-onset SOCS-1 KO mice were resistant to such treat-

ment(Figure 7C). Pre-onset SOCS-1 KO mice that had

been pretreated with anti-asialo GMI Ab lacked NKce‖s

but maintained NKT cells in theirlivers as in the results

of in vitro treatment shown in Figure 5C(unpublished

data}. They remained sensitive to a challenge with IFN一マ

plus IL-4(data not shown}. Pre-onset SOCS-1 KO mice

did not suffer from severe liver injuries at 24 hr after

challenge with IFN-V or IL-4 alone(data not shown). All

Page 8: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

SOCS-1/SSLI as a Cross-Talk lnhibitor541

Figure 6.α一GalCer Was Administered to Pre-Onset SSI-1 KO Mice or WT Mice

At 12hr, the liver specimens were sampled for histobgical study.(HEstalning, original magnificationl×200)Data are representatlve of five

independent experiments with similar results.

these findings together suggest that SOCS-1 may play

an essential role in regulating cytokine signalings in vivo

and that depbtion of SOCS-1 may facilitate the concur-

rent transmission of multiple cytokine signalings to pro-

duce abnormally activated T cells, including hepatic

NKT cells, thus Ieading to severe liver injuries.

Discussion

lFN-V-deficient SOCS-1 KO mice have been shown to

avoid various abnormalities observed in SOCS-1 KO

mice(Alexander et al.,1999;Marine et al.,1999). This

result suggested the possibility that SOCS-1 might be

aselective inhibitor of IFN-V signaling and that IFN-V is

principally responsib|e for inducing these pathological

changes in vivo. However, previous studies have not

clarified the mechanisms by which SOCS-1 selectively

inhibits IFN-V signaling. Rather, many invivo and invitro

studies have demonstrated that SOCS-1 inhibits a wide

range of signalings of cytokines/hormones, including

lL-4, IL-6, IFN-V, thrombopoietin, leukemia inhibitory

factor, prolactin, insulin, growth hormone, and stem cell

factor(Starr et aL,1997;Naka et al.,1997;Losman et

al.,1999;Adams et al.,1998;Hansen et al.,1999;Ram

and Waxman,1999;Pezet et al.,1999;Tomic et al.,1999;

Sakamoto et aL,1998;Song and Shuai,1998;Morita et

al.,2000;Kawazoe et al.,2001)and that SOCS-1 has

the capacity to bind to a‖JAKs(Starr et al.,1997;Endo

et al.,1997;Naka et aL,1997;Fujimoto et al.,2000;

Losman et al.,1999;Helman et al.,1998;Pezet et al.,

1999;Narazaki et al.,1998;Yasukawa et al.,1999;Mch-

olsonet al.,1999).Alternatively, SOCS-1 shows no spec-

ificity for cytokines or JAKs(for a review, see Naka et

al.,1999;Krebs and Hilton.,2000;Chen et al.,2000;

Nicola and Greenhalgh,2000). ln addition, a previous

study of ours demonstrated that SOCS-1 is capable of

inhibiting activation of STAT6 and the action of lL-4

in vivo(Naka et aL,1998). Moreover, various lesions

observed in SOCS-1 KO mice were eliminated not only

in SOCS-1/STATI DKO but also in SOCS-1/STAT6 DKO

mice(Figures l and 2). On the basis of these findings,

it can be assumed that SOCS-1 is a molecule acting

downstream not only on IFN-y but also on other cyto-

kines, including IL-4 in vivo.

  Our present data, that SOCS-1/STATI DKO avoided

liver injury(Figures l and 2), have further substantiated

the previous conclusion(Alexander et al.,1999;Marine

et al.,1999). To our surprise, however, deletion of the

STAT6 gene similarly preverlted these pathological

changes(Figures l and 2), These results suggest that

persistent activation of both STATI and STAT6, but

not of STATI alone, is responsible for inducing these

pathological changes. IFN-V and IL-4 were found to in-

hibit overactivation of STATI and STAT6 signalings, re-

spectively, via SOCS-1(Figure 7A). Moreover, STAT1-

induced SOCS-1 inhibited STAT6 activation(Figure 7B).

Thus,|FN-V inhibits not only its own signal but also

lL-4 signals via SOCS-1.Deletion of the STATI gene in

SOCS-1 KO mice allows persistent activation of STAT6

but not of STATI and deletion of STAT6 gene vice versa,

Page 9: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

lmmun}ty542

a     SOCS・1+ノ+

.     1.     11「1 

o 1

    SOCS一丁一ノー        Thymus

・:・疹欝欝嚇呼一・Y…AT 1

轟難響口○○喝一ST…3

SOCS-1◇∫◆

6 0

rご診

∴鞭囁

ぶ◇

 膠鷲

 げぺ停沓

 Sp

韓 叉^

戸Pぷぽ

袖 げ@畠☆

1 3 6   St‘mじli;1FN¥

    SOCSdイ・

   轍騰餓←pY今STAT 6

鐸辮糠鋤螺犠嚇矯←・・A・・6  Stim山llし4

C

b1 ; :

ノ浮F㌧恥 ぺい   、

o 1 3

1 : 1 :

    蔭糠ぶ

P _ 霧 再畑妙xw’

6 o

0   2   5

0   0   9

1

2i

3

5  Q  O〔h已wαs⑳imメ【atl。日品hl外Y4  1  4(h>・aftΩrSt≡㎜賦1。nwithlし4

 ・ _㍗拶       ←pY・STAT6

零倭梱皆楊閃陛栢麟番硬暴壕轍轍噸鱈鞘M昧←、,AT6

⑳_ ・ ・一一ぷ  _一一一鰍一r>        熟ぐ× Pr,                       ㈱ ←pY・STAT$

鞭蕪馨繋繍騨一艀●奮雛暫琳謙一・・A・1

一レ<一一→レ        SOCS.1+〆+                              SOCS“]『ρ

Figure 7. lnvolvement of Dysregulated Cross一丁alk lnhibltion of IFN-V and IL-4 Signaling in SOCS-1-Deficient NKT Cells in Severe Hepatitis

(A)Thymocytes were incubated with IFN-V or IL-4, and their STA↑1 and STAT6 phosphorylation was determined by means of immunobiotting

usrng anti-phosphorylated STA丁1 and anti-phosphorylated STAT6, respectively.

(B)Splenic T cellswere incubated with or without IFN-V fo‖owed by additional incubation with IL-4 for the indicated time. The phosphorylated

STATI and STAT6 were determined wlth the same procedure as shown in(A),

(C)SOCS-1 KO mice were administered with anti-NK1.1 mAb or control lgG, and 2 days later were challenged with|FN-V plus IL-4. At 24 hr,

the liver specimens were sampled forhistological study.(HEstaining, original magnification:×200). Data are representative offive independent

experiments with similar results,

resulting inavoiding severe pathological changes. Thus,

activated STATI and STAT6, when collaborated in the

absence of SOCS-1,mediate severe liver injury. Trans-

genic mice overexpressing IFN-V or IL4 intheirlympho-

cytes or hepatocytes have been reported to display

spontaneous thymic atrophy with relatively early death

although they remain entirely free of fulminant hepatitis

(Lewis et al.,1991;Tepper et al.,1990;Erb et al.,1997;

Young et al.,1997;丁oyonagaet al.,1994).These findings

and the results of our study suggest that thymic and

splenicchanges observed inSOCS-1/STATI and SOCS-

1/STAT6 DKO mice could be due to oversignaling of

residual, endogerlous IL-4, and IFN一ツ, respectively.

However, the severe liver injuries seem to require the

participation in simultaneoussignal transduction of mul-

tiple cytokines, at least of both IFN-V and IL-4. Although

lFN一γdeficient SOCS-1 KO mice completely recovered

from the complex disease observed in SOCS-1 KO mice

(Alexander et al.,1999;Marine et al.,1999), the STAT1-

deficient SOCS-1 KO mice in our study did not show

complete recovery in some organs, such as thymus and

spleen.This discrepancy may be dueto the involvement

of another signal pathway besides JAK/STAT signal

transduction in IFN-V signaling.

  Concanavalin A(Con A)-induced hepatitis is a T cell-

mediated fulminant hepatitis model in mice(Tiegs et a【.,

1992;Lohr et al.,1994). Many factors are mvolved in

Con A-induced hepatitis. Since lFN一γdeficient mice are

resistant to Con A-induced hepatitis, and pretreatment

with neutralizing anti-IL-4 protects them against this

form of hepatitis, both IFN-V and IL-4 seem to be in-

volved in this liver injury(Kaneko et al.,2000;Tagawa

et al.,1997;Toyabe et al.,1997;Chen and Paul,1997).

Furthermore, Vα14-deficient mice lacking NKT cells are

resistant to Con A hepatitis(Kaneko et al.,2000), which

suggests an essential role for NKT cells in this disease.

The liver contains a much higher proportion of NKT cells

than do other tissues. Moreover, NKT ce‖s possess re-

ceptors for both IL4 and IFN♂y, so that they can receive

simultaneous signaling from both IFN-V and IL-4. lt is

also known thatthe cytotoxicity of NKT cells isenhanced

by IL-12(Kawamura et al.,1998)and IL-4 via upregula-

tion of their expression of FasL and Granzyme B(Ka-

neko, et al.,2000). As shown in our study, this is also

Page 10: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

SOCS-1/SSI-1 as a Cross-Talk lnhibitor543

the case for fulminant hepatitis in SOCS-1 KO mice.

SOCS・1-deficient hepatic lymphocytes were intrinsi-

cally cytotoxiC against syngeneic hepatocytes(Figure

5)and lost theirhepatocytotoxicity after deletion of NKT

cells(Figure 5). Furthermore, in vivo deletion of NKT

cells rescued pre-onset SOCS-1 KO mice from IFN-y

plus IL-4-induced liver injuries(Figure 7C), while control

lgG-administered pre-onset SOCS-1 KO mice suffered

from liver inluries similar to those observed in onset

SOCS-1 KO mice after administration of lFN一ッplus IL-4

(Figure 7C).

  We observed spontaneous development of fulminant

hepatitis in SOCS-1 KO mice. At present, we do not

know whether this change is induced intrinsically or

extrinsically, but we succeeded in shortening the period

required for development of the lethal pathological

changes by administration of anti-CD3(data not shown)

orα一GalCer(Figure 6). Thus, we can assume that

SOCS・1 KO mice may be hypersensitive to potentially

contaminating pathogens even under SPF cond爾ons.

  The study presented here demonstrated that IFN-V-

induced SOCS-1 is essential for the inhibition of IL-4

signaling in vivo and that the disturbance of this cross-

talk inhibition abnormally activates NKT cells, particu-

larly their hepatocytotoxic activity against se廿一driven

ce川s, consequently leading to fulminant hepatitis. How-

ever, these results cannot exclude the possibility that

other cells besides NKT cells are involved. SOCS-11

CDI double KO mice will be expected to provide direct

evidence of the essential role of NKT cells in liver inluries

of SOCS-1 KO mice. Our present study allows us to

investigate the possible mvolvement of SOCS-1 dys-

function in autoimmunity and infection-hduced mtracta-

ble organ failures. Further analysis of SOCS-1 might

lead to new therapies even for cancer by focusing on

strengthening the cytotoxic action of NKT cells. We are

now clarifying how SOCS-1 takes part in cross-talk inhi-

bition for other cytokines and what roles dysregulated

SOCS-1 plays m other tissue damage.

Experimenta甲rocedures

MiceSTAT1(Meraz et al.,1996), ST酊6πakeda et al.,1996}, and SOCS-1

KO mice(Naka et al.,1998)were established as previously de-

scribed. All SOCS-1 KO mice were on the C57BL/6 background

(backcrossed more than seven generations).

Cell Preparation and Cell Counts

Single-cell suspensions were obtained from thymi and spleens after

the cel|s had been passed through mesh filters. Spleen ce|ls were

also treated with Ack buffer(0.15MNH4α,1mM KHCO3, and O.1

mM Na2EDTA)to lyse RBC.The total number of cells was determined

by microscopic observation of廿ypan blue-stained cells and using

hemocytometers

Flow Cytometric Analysis

Cells were stained with the following monoclonal Abs(mAbs):FITC-,

PE-, or APC-conlugated anti-CD3∈, a耐i-CD4, anti-CD8, anti-CD69,

and anti-B220(PharMingen, San Diego CA). Stained cells were ana-

lyzed on a FACSCalibur(Becton Dickinson, San Jose, CA)using

Ce‖Quest softwa佗{Becton Dickinson). Live Iymphocytes we陀

gated according to their forward and side scatter profiles.

Westem Blot AnalysisSplenic T cells, which were sorted by using anti-B220, anti-Gr1,and

anti-Macl Ab mAbs(Miltenyi Biotec, Bergisch Gladbach, Germany},

were treated with 10μglml of IFN-Y,10μglml of IL-4,0r 10μglml

of lFN-7 followed by 10帖glml of lL-4 for the times indicated. The

cells were lysed with an NP-401ysis buffer(20 mM Tris-HCI【pH 7.5],

150mM NaCl,1%NP-40,10mM Na2VO4,0.5 mM dithiothreitol,1

μglml PepstatinA,1μg/ml Leupeptin, and l mM PMSF)and clarified

by centrifugation. The supernatants collected were sublected to an

8%SDS-PAGE. The proteins were transferred onto a nitrocellulose

filter(Schleier&Schuell, Dassel, Gemany}. After blocking with 5%

skim milk, the filterwas incubated with anti-phospho-specific STATl

or phospho-specific STAT6 Abs(New England Biolab, MA, USA},

and then with the horseradish peroxidase-conjugated anti-rabbit

lgG Ab. The signals were detected with an ECL system(Amersham,

Bukinghamshire, England). A肚er the Abs had been stripped, the

filter was reprobed with antトSTAT1(Transduction Labolatry, KY,

USA}or anti-STAT6 Abs{Santa Cruz, CA, USA).

Assay for Hepatocytotoxicity

Hepatocytotoxicity was determined by the method described else-

where with some modificationπsutsui et al.,1992). ln brief, liver

parenchymal cells(1×1051ml)prepared from WT C57BL/6 mice

were incubated in a 96-well-collagen type l-coated plate overnight,

washed twice with the culture medium kept at 37°C, and|abeled

with【51Cr1-sodium chromate for l hr. Hepatic lymphocytes from

mice with various mutants were hcubated with 20μglml of anti・

murine Fas L or 100 nM concanamycin A(Wako, Osaka, Japan)for

lhrand incubated with the labeled liver parenchymal cells at various

effector-target cell ratios for4hr. In some experime耐s, we depleted

NK1.1+cells from hepatic lymphocytes by MACS after incubation

with biotinylated anti-NK1.1{PharMingen}followed by additional

incubation with streptavidin・microbeads(Miltenyi Biotec), We also

depleted the NK cell population by mcubating the cells twice with

100脾g/ml of anti-asialo GM1(Wako}plus complement for 30 min,

fo|lowed by determination of their hepatocytotoxicity. Hepatocyto-

toxicity was calculated as previously reported(Tsutsui et al.,1992).

Spontaneousrelease of hepatocytes was less than 8%of theirmaxi-

mum時lease.

Depletion of NKT Cells and ln Vivo Stimulation

of IFN・ツPlus IL-4

Mice at 3 days after birth were inlected intraperitoneally with O.1 ml

PBS containing 2.5 mg anti-NK1.1 mAb(PK136}. Two days after

treatme耐with NK1.1 mAb, these mice were injected intraperitone-

ally with O.1 ml PBS containing both 3μg IFN-V and 1口IL-4.

Administration ofα一GalCer

Mice(3 days after birth)were injected intraperitoneally with O.1 ml

PBS containing either 1μgα一GalCer or control vehicle, which were

provided by KIRIN(Tokyo, Japan}.

Adoptive TransferRecipient RAG2 KO mice, kind|y provided by Dr. ltoh at CIEA(Kana・

gawa, Japan}, were irradiated with 700 rad. Donor cells from embry-

onic day 16.5(E16.5)feta川iver of SOCS-1 KO or WT mice were

suspended in DMEM supplemented with 2%fetal calf serum(2×1071ml}. Recipient animalswere inlected i.v. with 2×106 donor cells.

Acknowbdgments

We gratefuny acknowledge the provision of STATI KO mice by Dr.

R.D. Schreiber at Washington University, STAT6 KO mice by Dr. S.

Akira at Osaka University(Osaka, Japan), RAG2 KO mice by Dr. M.

ltoh(CIEA, Kanagawa, Japan), and anti-murine Fas L mAb by Dr.

N.Kayagaki(Juntendo Unive信ity, Tokyo, Japan). We thank Dr. S.

Nagata(Osaka University)for helpful discussions, Dr. T. Hoshino

(Kurume University, Kurume, Japan)for the kind gift of PK136, Ms.

KSato(KAC Corp, Shiga, Japan}for expert management of the

mice, and Ms. A. Saito and Ms. R. Harada fortheirsecretarial assis-

tance. This work was suppor寸ed by a Grant-in-Aid from the Ministry

of Education, Science, and Culture of Japan.

Received November 7,2000;revised March 29,2001.

Page 11: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

lmmunity544

References

Adams, T.E, Hansen, J.A., Starr, R., Nicola, N.A., Hilton, D.J., and

Billestrup, N.(1998). Growth hormone preferentially induces the

rapid, transient expression of SOCS-3, a novel inhibitor of cytokine

receptor signaling. J. BioL Chem.273,1285-1291.

Alexander, W.A., Starr, R., Fenner, J且, Scott, C.L, Handman, E.,

Springg, N.S., Corbin, J.E., Cornish, A.L., Darwiche, R., Owczarek,

C.M.,et al.(1999).SOCSI iscritical inhibitor of interferon V signaling

and prevents the potential|y fatal neonatal actions of this cytokine.

Cell 98,597-608.

Chen, H., and Paul, W.E.(1997}. Cultured NK1.1+CD4+Tcells pro-

duce|arge amounts of IL-4 and IFN-V upon activation by anti-CD3

0r CD1.J. lmmunol.759,2240-2249.

Chen, X.P., Losman, J.A., and Rothman, P.(2000}. SOCS proteins,

regulatαs of intracellωar signaling. Immunity 73,287-290,

Darnell, J.E., Jr., Kerr,1.M., and Stark, G.R.(1994). Jak-STAT path-

ways and transcriptional activation in response to IFNs and other

extracellular signaling protein. Science 264,1415-1421.

Dickensheets, H⊥., Venkataraman, C., Shindler, U., and Donnelly,

R.P. n 999). lnter「erons inhibit activation of STAT6 by interleukin 4

in human monocytes by inducing SOCS-1 gene expression. Proc.

Natl. Acad. Sci. USA 96,10800-10805.

Endo, T.A., Masuhara, M., Yokouchi, M., Suzuki, R., Sakamoto, H.,

Mitsui, K., Matsumoto, A., Tanimura, S., Ohtsubo, M., Misawa, H.,

et al.(1997). A new protein containing an SH2 domain that inhibits

JAK kinases. Nature 387,921-924.

Erb, KJ., Ruger, B., Von Brevern, M., Ryffel, B., Schimpl, A., and

Rive廿, K.(1997). Constitutive expression of interleukin(IL)-4 in vivo

causes autoimmune-type disorders in mice. J.Exp.Med 785,

329-339.

Fujimoto, M., Naka, T., Nakagawa, R., Kawazoe, Y., Morita, Y., Ta-

teishi,A.,Okumura, K.,Narazaki, M.,and Kishimoto,T.(2000). Defec-

tive thymocyte devebpment and perturbed homeostasis of T cells

in SOCS-1/SOCS-1 transgenic mice, J」mmunoL 765,1799-1806.

Hansen, J.A., Lindberg, K., Hilton, D.J., Nielsen, J.H., and Billestrup,

N.(1999).Mechanism of inhibition of growth h◎rmone receptor sig-

naling by suppressor of cytokine signaling proteins. Mol. Endocrinol.

73,1832-1838.

Helman,D.,Sandowski, Y.,Cohen, YっMatsumoto,AりYoshimura, A.,

Merchav, S., and Gertler, A.(1998}. Cytokine-inducibb SH2 protein

(αS3)and binding protein(JAB}abolish prolactin receptor-medト

ated STAT5 signaling. FEBS 447,287-291.

lhle,」.N.(1995). Cytokine receptor signaling. Nature 377,591-594.

Kaneko, K., Harada, M., Kawano, T., Yamashita, M門Shibata, Y.,

Gelyo, F., Nakayama, T., and Taniguchi, M.(2000). Augmentation of

Vα14NKT cell-mediated cytotoxicity by interleukin 4 in an autocrine

mechanism resulting in the development of concanavalin A・induced

hepatitis. J. Exp. Med 797,105-114.

Kawamura, T.,Takeda, K.,Mendiratta, S.K., Kawamura, H.,van Kaer,

L.,Yagita, H., Abo, T., and Okumura, K.(1998}. Critical role of NK1+

Tcells in lし12-induced immune responses in vivo. J. lmmunol.’60,

16-19.

Kawazoe, Y., Naka, T., Fulimoto, M., Kohzaki, H., Morita, Y,, Na陥一

zaki, M., Okumura, K., Saitoh, H., Nakagawa, R., Uchiyama, Y., et

aL(2001). Signal transducer and activator of transcription(STAT)-

induced STAT inhibitor 1(SSI-1 ysuppressor of cytokine signaling 1

(SOCS1)inhib吐s insulin signal transduction pathway through modu-

|ating insulin receptor substrate 1(IRS-1}phosphorylatbn. J. Exp.

Med.793,263-269.

Kishimoto, T., Taga, T., and Akira, S.(1994). Cytokine signa|trans-

duction. Cell 76,253-262.

Krebs, D.L, and Hilton, D.J.(2000). SOCS:physiological suppres-

sors of cytokine signaling. J. Cell Sci.773,2813-2819.

Lewis, D.B.,Yu, C.C., Forbush, K.A.,Carpenter, J.,Sato, T.A., Gross・

man, A., Liggitt, D.H., and Perlmutter, R.M.(1991). lnterleukin 4

expressed in situ selectively alters thymocyte development. J. Exp.

Med.773,89-100.

Lohr, H.F., Schlaak, J.F., Gerken, G., Fleischer, B., Dienes, H.P., and

Meyer zum BOschemfelde, K.H.(1994}. Phenotypical analysis and

cytokine release of∬ver infiltratmg and peripheral blood T lympho-

cytes from patient with chronic hepatitis of different etiology. Liver

74,161-167.

Losman, J.A., Chen, X.P., Hilton, D.J., and Rothman, P.(1999}.

SOCS-1 is a potent inhibitor of lL-4 signal transduction. J. lmmunol.

762,3770べ3774.

Marine, J.C., MacKay, C., Wang, D., Topham, D.J., Parganas, E.,

Nak司ima, H., Pendeville, H., Yasukawa, H., Sasaki, A., Yoshimura,

A.,and lhb, J.N.σ999). SOCSI deficiency causes a lymphocytes-

dependent perinatal lethality. Cell 98,609-616.

Meraz, M.A., White, J.M., Sheehan, K.C., Bach, E.A., Rodig, S.J.,

Dighe, A.S., Kaplan, D.H., Riley, J.K., Greenlung, A.C., Canpbell, D.,

et al.(1996). Targeted disruption of the STATI gene in mice reveals

unexpected physiologic specificity in the JAK-STAT signaling path-

way. Cell 84,431-442,

Morita, Y., Naka, T., Kawazoe, Y., Fujimoto, M., Narazaki, M., Naka-

gawa, R., Fukuyama, H.,Nagata, S.,and Kishimoto,T.(2000).Signals

transducers and activators of transcription(STAT)-induced STAT

inhibitor-1(SSI-1 ysuppressor of cytokine signaling-1(SOCS-1)sup-

presses tumor necrosis factor a-induced ce冊death m fibroblasts.

Proc. Natl, Acad. Sci. USA.97,5405-5410.

Naka, T., Narazaki, M., Hirata, M., Matsumoto, T., Minamoto, S.,

Aono, A., Nishimoto, N., Kεりita, T., Yoshizaki, K., et al.(1997). Struc・

ture and function of a new STAT-induced STAT inhibitor. Nature

387,924-929.

Naka, T., Matsumoto, T., Narazaki, M., Fulimoto, M., Morita, Y.,

Ohsawa, Y., Saito, H., Nagasawa, T., Uchiyama, Y., and Kishimoto,

T.(1998).Accelerated apoptosis of lymphocytes by augmented in-

duction of Bax in SSI(STAT-induced STAT inhibitor-1)deficient

mice. Proc. Natl, Acad. Sci. USA.95,15577-15582.

Naka, T., Narazaki, M., and Kishimoto, T. Negative feedback regula-

tions of cytokine signals.(2001). Cytokine inhibitors (IRBM),

241-260.

Naka, T,, Fujimoto, M., 

and Kishimot◎, T.{1999}. Negative regulation

of cytokine signaling:STAT-induced STAT inhibitor. TiBS 24,

394-398.

Narazaki, M., Fulimoto, M., Matsumoto, T., Morita, Y., Saito, H.,

Kalita, T., Yoshizaki, K., Naka, T., and Kishimoto, T.(1998}. Three

distinct domains of SOCS-1/SOCS-1/JAB protein are required for

its suppression of interleukin 6 signaling. Proc. Natl. Acad. Sci. USA

95,13130-13134.

Nicholson,S.E., Willson, T.A., Farley, A.,Starr, R., Zhang, J.G., Baca,

M.,Alexander, W.S., Metcalf, D.,Hilton, D,J.,and Nicola, N.A.(1999).

Mutational analyses of the SOCS proteins suggest a dual domain

requirement but distinct mechanisms for inhibition of LIF and IL-6

signal transduction. EMBO J.78,375-385.

Nicob, N.A., and Greenhalgh, C.J.(2000). The suppressors of cyto-

kine signaling(SOCS)proteins:important feedback inhibitor of cyto-

kine action. Exp. Hematol.28,1105-1112.

Pezet, A., Favre, H., Kelly, P.A., and Edery, M.(1999}. lnhibition

and restoration of prolactin signal transduction by supPressors of

cytokine signaling. J. Biol. Chem.274,24497-24502.

Ram, P.A., and Waxman, D.J.(1999}. SOCS/αS prot引n inhibition

of growth hormone-stimulated STAT5 signaling by multiple mecha-

nism. J. Biol. Chem.274,35553-35561.

Sakamoto, H., Yasukawa, H., Masuhara, M., Tanimura, S., Sasaki,

A.,Yuge, K., Ohtsubo, M., Ohtsuka, A., Fujita, T., Ohta, T., et al.

(1998).Alanus kinase inhibitor, JAB, is an interferon-gamma-induc-

ible gene and c◎nfers resistance to interferons. Blood 92,1668-

1674.

Song, M.M.,and Shuai, K.(1998).The suppressorof cytokine signal-

ing(SOCS)1and SOCS 3 but not SOCS 2 proteins hhibit interferon-

mediated antiviral and antiproliferative activities. J. Biol. Chem.273,

35056-35062.

Sonoda, K.H., Exley, M., Snapper, S., Balk, S.P., and Streilein, J.S.

(1999}.CD1-reactive natural killer T cells are required for develop-

ment of systemic tolerance through an immune-privileged site. J.

Exp. Med、790,1215-1225.

Page 12: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

SOCS-11SSI-1 as a Cross・Talk lnhibitor545

Starr, R., Willson, T.A., Viney, E.M., Murray, L.J.L, Rayner, J.R.,

Jenkinsβ.J., Gonda, T.J.,Alexander, W.S.,Metcalf, D.,Nicola, N.A.,

et al.(1997}. A family of cytokine-inducible inhibitors of signaling.

Nature 387,917-921.

Starr, R.,Metcalf, D., Elefanty, A.G., B呼sha, M.,Willson,T.A,, Nicola,

NA, Hilton, DJ., and Abxander, W.S.(1998). Liver degeneration

and lymphoid deficiencies in mice lacking suppressor of cytokine

signaling・1.Proc. Natl. Acad. Sci. USA.95,14395-14399.

Starr, R.(2001). lnhibition oftheJanus kinase-signal transducers and

activators of transcription(JAK/STAT)signaling pathway. Cytokine

inhibitor(IRBM),261-284.

Tagawa, Y., Sekikawa, K., and lwakura, Y.(1997}. Suppressor of

concanavalin A-induced hepatitis in IFN-V’ノ’mice, but not in TNF一

α“みmice. J. Immunol.759,1418-1428.

Takeda, K., Tanaka, T., Shi, W., Matsumoto, M., Minami, M., Kashi-

wamura, S., Nakanishi, K., Yoshida, K., Kishim◎to,丁., and Akira, S.

n996). Essential role of STAT6 in lL-4 signaling. Nature 380,627-630.

Tepper, R.1., Levinson, D.A., Stanger, B.Z., Campos-Torres, J., Ab-

bas, A.K., and Leder, D.(1990). IL-4 induces allergic-like inflamma-

tory disease and alters T cell development in transgenic mice. Cell

62,457-467.

Tiegs, G., Hentschel, J., and Wendel, A.(1992). A T cell-dependent

experimental liver inlury inmice inducible by concanavalin A. J.Clin.

hvest.90,196-201.

Tomic, S., Chughtai, N., and Ali, S.(1999). SOCS-1,-2,-3:selective

targets and functions downstream of the prolactin receptor. Mol.

Cell. Endocrinol.758,45←52.

Toyabe, S., Seki, S., liai, T., Takeda, K., Shirai, K., Watanabe, H.,

Hiraide, H., Uchiyama, M., and Abo, T.(1997). Requirement of IL-4

and liver NK1+Tcells for concanavalin A-induced hepatic injury in

mice. J. lmmunol.759,1537-1542.

Toyonaga, T., Hin◎, O., Sugai, S., Wakasugi, S., Abe, K., Shichiri,

M.,and Yamamura, S.(1994). Chronic active hepatitis in transgenic

mice expressing interferon-Y in the liver. Proc. Natl. Acad. USA 97,

614-618.

Tsutsui, H., Mizoguchi, Y., and Morisawa, S.(1992). lmportance of

direct hepatocytolysis by liver macrophages in experimental fulmi-

nant hepatitis. Hepato・Gastroenterol、39,553-558、

Venkataraman, C., Leung, S., Salvekar, A., Mano, H., and Schindler,

U.(1999).Repression of IL-4-induced gene expression by IFN-V

requires Statl activation. J. lmmunol.762,4053-4061.

Yasukawa, H., Misawa, H”Sakamoto, H. Masuhara., Sasaki, A.,

Wakioka, T., Ohtsuka, S., lmaizumi, T., Matsuda, T., lhle, J.N., and

Yoshimura, A.(1999). The JAK-binding protein JAB inhibits Janus

tyrosine kinase activity through binding in the activation loop. EMBO

J.78,1309-1320.

Yasukawa, H.,Sasaki, A.,and Yoshimura, A.(2000). Negative regula-

tion of cytokine signaling pathways. Annu. Rev. lmmunol.78,

143-164.

Young, H.A., Klinman, D.M., Reynold, D.A., Gロegorzewski, K.J., Nii,

A.,Robin, J.M.,John, W-P. Kenny, O.JJ.,and Komsclies, K.L(1997}.

Bone marrow and thymus expression of interferon-V results in se-

vere B-cel川ngeage reduction,T-cell Iineage alterations, and hema-

topoietic progenitor deficiencies. Blood 89,583-595.

Page 13: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

nal weeklyjournal ofscience

Structure and function of a new

STAT-induced STAlT inhibitor

   Tets可i Naka*1,Masashi Narazaki*1,Moritoshi Hirata*1,

Tbmoshige Matsumoto*1, SeUiro Minamoto*1,AtsufUmi Aono*1,

  Norihiro Nishimoto*2, Tadahiro K勾ita*3, Tbtsuya Taga*4,

Kazuyuki Ybshizaki*2, Shizuo Akira*5&Tadamitsu Kishimoto*1

       *10saka University Medical School Department of Medicine III.

             2-2,Yamada-Oka, Suita, Osaka 565, Japan

   *2Medical Science I, School of Health and Sports Sciences, Osaka University.

             2-1,Yamada-Oka, Suita, Osaka 565, Japan

    *3Research and Development Center, International Reagents Corporation,

          1-2,1-Chome, Murotani, Nishiku, Kobe 651-22, Japan

*4Department of Molecular Cell Biology, Medical Research Institute, Tokyo Medical and

    Dental University.2-3-10, Kanda-Surugadai, Chiyoda-ku, Tokyo lO1,Japan

        *5Department of Biochemistry, Hyogo College of Medicine,

            1-1,Mukogawa, Nishinomiya, Hyogo 663, Japan

Reprinted from Nature, VbL 387, No.6636, pp.924-929,26 June 1997

◎Macmillan Magazines Ltd.,1997

Page 14: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

OIBt:ructure and function of a new

SIA■■induced S’『賀「inMbitor

Tbt8uj川aka㍉Ma8ashl Narazakl㌔Morltosh川irata㌔1bmoshigo Matsumoto㍉Soljiro Mlnamoto㍉Atsu佃ml Aono㍉Norlhlro Nlshlmoto†, Tadahlro Kajlta‡,1●tsuya T㌔ga$, Kazuyuki VbshlzaM†, Shl茗uo Aklrall

&1恒dam脆8u Kishlmotぴ

・OsαkαUniv¢rsltγMεdlCdl School D¢pαr畑e川(lfMεdicifle HI.2-2,

YOmΩd@≡Okα’S“itα,05αkα565,∫αpΩn

†Medlc侃I Scl¢”cε1, School O∫Heolth飢d Sports Sc{¢nces, Depロr加e耐q∫Osロkロ

Univer5三ウ.2-1, 

Ya,ηodα一〇kロ, Sμ{tロ, Os“k4565,∫σpαn

‡Resεακhαnd Dεvelopment Ce川εち1寵¢mαtiomi R孤gents Co,’porαtlon,

1-2,1-Cho1れe, Murotαn輌, Nis}1輌kt↓, Kobe 651-22,∫αPαn

§Depぴητ¢’舷(ヅMρ1ε‘L‘1αr Cεll Biolo筋Medim!R¢鰯rch Ins蹴姥,

哀}kyo M¢dicdα”d Dεnml U”iver5{リノ,2-3-10, Kαnd4-Surugαdαi, Ch輌γod6-kt’,

τbkyo 101,∫αpan

ll D¢ραr伽θ解(ザB輌oζ加”ε輌5rrγHγogo Collε9εqfMε4ic加θ,1-1, Mし‘た09ロ願,

Nis}1inoη1iγ偽Hγogo 663,∫αPαn

unknown genes, two of which contained an SH2 domain. Onewas later ide煎i6ed as CIS16 and the oth亨r was a new gene(named

SSI-1). SSLlcDNA has a single open reading frame that encodes a

212-amino-acid polypeptide and contains an SH2 domain in the

middle of its sequence(at codons 79-167). No other consensus

motifS like SH3 domains were fbund in SSH(Fig. la). The SH2

domain ofSSI-1was homologous to that ofCIS 36%but showed no

signi6cant homology with those of Stat30r Stat6 except at the

phosphotyrosine recognition site4’5・18・19(Fig.1b).

  SSI-1 expression was examined in various murine tissues. SSI-

lmRNA was ubiquitously expressed, with expression being strong

in lung, spleen and testis, and weak in all other tissues(Fig.2a). We

examined SSI-l induction in several factor-dependent cell lines

(Fig.2b). Hybridoma MH60 and myeloid leukaemia M l cells both

expressed SSI-l mRNA, with expression peaking 60-120 min a丘er

treatment with IL6 plus soluble IL-6 receptor(slL-6R). IL-4-

dependent CT4S cells20 and G-CSF-dependent NFS60 cells

expressed SSI-l mRNA in response to IL-4 and G-CSE respectivelγ

These results show that SSI-1is induced not only by IL-6 and G-CSE

both ofwhich activate Stat3(re烏4,21),but also by IL-4, a cytokine

The signaning pathwaγthat comprise⑨IAK kinases and STATproteins(fbr signal transdu㏄r and activator of transαiption)is

important f6r relaying signals fbm various cytokines outside the

ceU to the㎞sideト3. The fヒedback mechanism responsible fbr

switch輌ng oκthe(7tokine signal has not been elucidated. We now

report the cloning and characterization of an inhibitor of STAT

activat輌on which we name SSI-1(fbr STAT-induced STATinhibitor-1). We fbund that SSI・1 messenger RNA was induced

by the cγtokines intedeukins 4 and 6 (IL-4, IL-6), leukaemia-

inhibitory factor(UF), and granulo《Wte colony-stimulating食ctor

(G-CSF). Stimulation by IL-60τLIF ofmurine myeloid leukaemia

cells(MI cells)induced SSI-l mRNA expression wllich was

blocked by transfごction of a dominant-negative mutant of Stat3,

ind忙at輌ng that the SSI-1gene is a target ofStat3(爬fS 4-7). Forced

overexpre⑨sion of SSI-l complementary DNA inter伝ed with肌一

6-and LIF-mediated apoptosis and maαophage di飽rentiation of

Ml cells, as well as IL-6 induced tyrosine-phosphorylation of a

receptor凪ycoprotehl component, gp l 30, and ofStat3. When SSI-

1is overexpressed i皿COS7 ceUs, it can associate with the kinases

Iak2 and Tyk2. These血n《1ings indicate that SSI-1is responsible fbr

negative一食edback regulation of the IAK-STAT pathway induced     クby《力o㎞e stimulation.

  Most cytokine receptors consist of two or three polypeptide

chains, namely a ligand-specific receptor chain and a signal trans-

ducer that is commonly used by various cytokines8-ll.The nature of

this receptor system explains the fUnctional redundancy of cyto-

kines. Homo-or heterodimerization of receptor components with

ligands or ligand-receptor complexes stimulates a unique cytokine

signalling cascade, the IAK-STAT pathwayl-3」2, which causes the

activation of target genes in the cell nucleus;little is㎞own about

the direct targets ofthe STAT family↓One feature ofcytokines is the

transient expression of their activityl which suggests that negative-

feedback regulation operates in cytokine-signal transduction:pos-

sible candidates fbr exerting this control are the SH2-domain-

containing phosphotyrosine phosphatase SHP-1, which associates

with the tyrosine-phosphorylated IL-3 receptorβ一chain and with

the erythropoietin receptor(EPO-R)13-15, and a cγtokine-inducible

SH2-containing protein(CIS)which binds to the Stat5-binding

sites of EPO-R(refS 16,17, and A. Ybshimuraε口1., personal

communication). We have now cloned a cDNA encoding an SH2-

domain-containing protein that is inducible by Stat3 and inhibits

Stat3 fUnction.

  τbclone other members of the STAT family, we prepared a

monoclonal antibody against a sequence motif(GTFLLRFS)fbund in the SH2 domain of Stat3(refS 4,5). Using this antibody}

we screened a murine thymus cDNA library and isolated 20

a99CCCCしCgagヒaggaヒ99しagCaCgCaaCCag9仁gqCaσCCgaCaaヒgCgaヒCしCCC             M  V  A  R  N  Q  V  A  A  D  N  A  工  S  P

eggeagCagagCCCCgaCg9(:qg仁CagagCCCしeCヒCgヒC⊂ヒCgヒCしヒCgしCCヒCgCCag

 A A E P R R R S E P S S S S S S S S P A

CggCCCCCgヒgCgしCCCCgqCCCヒ9己CCg9999ヒ⊂CCagCCCCagCCCCし99CgaCaCヒe

 APVRPRPCPGVPAPAPGDTHaCヒヒCCgCaCeしヒCCgCしCCCaCヒCCgaししaCCggCgCaしCaCgCggaC⊂agCgCgC仁CC  F  R  T  F  R  S  H  S  D  Y  R  R  I  T  R  T  S  A  L  L

ヒ99aegCCヒgCggCヒヒCヒaヒヒ9999aCe⊂CヒgageqしgCaCg9⊆rgCgCaCgageggCヒgC D A C G P Y W G P L S V H G A H B R L R

9A

Cヒ

しF

Cヒ

ヒF

C9しC

CaaN

99CR

aa

CQ

CgCR

し9aS

CagD

CgCR

9ヒ

9V

9ヒ

ヒL

Cヒ

しF

CCaT

C99G

9し

qV

eeeP

ga9菖

CCqA

ヒ9

CgCヒCagCgヒgaagaし99eヒしCg99CCCCaCgagCaヒeCgCgヒgCaeヒヒCCaggCCggeC L S V K H A S G P τ S 工 R V H F Ω ム G R

gCヒヒCCaCしし99aCggCaaCCg⊂gaga⊂CヒしCgaCしqCCヒししヒCgagCヒOCヒ99agCaCヒ

 FHLDGNRETFDCLP菖LLEHYaCgヒ99CggCgCCgCgCCgCaヒ9ヒヒqqqggCCeCgCヒgCgCCagCgCCgCg七gCggCCgC V A A P R R M L O A P L R Q R R V R P L

ヒgCagOagCヒ9ヒ9しCgCCagCqCaヒCgヒ99eCgeCgヒgqqヒCgCgagaaCCし99CgCgCa Q E L C R Q R I V A A V G R E N L A R 工

しCCCしCヒヒaaCCeg9しaCしCCgしgaCヒaCCヒgagヒヒCCヒヒCCCCヒヒCCagaヒCしgaCCgg

 PLNPVLRDYしSSF2FQ工★CしgCCgCヒ9ヒqeCgCagCaしヒaagし09999CgCCヒヒaヒヒaヒししCヒしaヒしaヒヒaaヒヒaヒヒa

ヒヒaしヒししヒcし⊆rgaaccacqヒgggagGecしc⊂⊂cgecヒqggしcggagggagヒggヒヒgヒgga

999しgagaヒgCCヒCCCaCヒヒCヒ99CしgqagaCCこCaヒCCCaC⊂ヒCヒCag999ヒgqg99し9

CヒCCCCヒCCしgqヒgCヒCeCし⊂CO99ヒCCCCCCヒ99ヒヒ9ヒageagCヒヒ9し9ヒCヒ999⊆rCea

99aCCヒgaaヒヒeCaCヒeCヒaeCヒCヒCCaヒ9ヒヒヒaCaヒaヒヒ⊂CCagヒaヒCヒヒヒgCaCaaaC

eag口99ヒCg999ag99ヒCヒCヒ99CヒヒCaしtヒヒtCヒgCヒ9ヒgCagaaヒatCeヒaヒしヒしaヒa

ヒヒしヒヒaCagCCagししヒag9しaaヒaaaCヒヒヒaヒヒaこqaaagヒヒヒヒヒヒヒヒヒaaaagaaaCaa

acaaagaヒヒ

C:S

SSI-1STAT3STAT6

CIS

SS1-1STAT3

ST品16

 WYwGS ITASEARQHLQKHPEG雪■LvRDSTHPSYLFTLSVKTTRGPTNvRIEYADSSPRLGFYWGPLSvHGJ、HERLRAEPvGTFLVRDSRQRNCFFALSVKMASGPTS工RVHFQAGRPHLG   G  S      L   P GTPLLRFS

   G  S       L   P G宝■LLRFS                 S

OSNCLSRPRILAPPDvvSLVQHYvASCAADTRSDSPDDGN-一一一⇔RE-TPDCLFELLEHYVAAPRRHLGAPL                  A

D               L

Figure l a, Nucleotide and deduced amino-acid sequences of SSI-1 cDNA. The

asterisk indicatesthe stopcodon.The ami∩o-acid sequence ofthe SH2domainis

shown in bold. b, Sequence alignment of the SH2 domain of SSI-1 with those of

CIS, Stat3 and Stat6, Bold letters show amino・acid residues that are identical ln

SSI-1, CIS, Stat3 and Stat6.

Page 15: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

that activates Stat6(re£19). Consistent with these丘ndings, the

promoter region of the SSI-lgene was fbund to contain Stat3 and

Stat6 binding sequences4」9・2z(data not shown).

  We next examined SSI-1 induction in M l cells trans飴cted with

wild-type Stat3(Ml/Stat3)and a Stat3 mutant(M1/Y705F)(Fig.

2c). Stat3(Y705F)is a dominant-negative fbrm of Stat3, in which a

tyrosine at residue 705 that is phosphorylated by a IAK kinase has

been substituted by phenylalanine6・7. In this experiment, transfec.

tants containing only the neomycin-resistance gene(M1/Neo)were

used as controls. SSI-1 mRNA was more strongly induced by IL-6

plus slL6R or LIF in Ml/Stat3 cells than in M1/Neo control cells,

but was not induced in M l1Y705F cells. These results indicate that

the SSI-l gene is one ofthe target genes of Stat3 and is induced by

theハAK-STAT pathwa}㌧

  Tb test the ef琵ct of SSI-10n the gp 130-mediated signalllng

pathwa拓we established M l transfとctants that constitutlvely express

wild-type SSI-1(Ml/SSI-1)or a mutant SSI-1(Ml!TR)that is

truncated at the C-terminal regionwhich includes the SH2 domain.

As shown in Fig.3a,b, M 1/Neo alld M 1/TR underwentgrowth arrest

and cell death aftertreatment with IL-6 plus slL-6Ror LIEas did the

parenta1 Mlcells(Ml).The dead cells showed飴atures ofapoptosis

such as chromatin condensation and apoptotic bodies(Fig.3c).In

contrast, growth ofM1/SSI-lcells did not arrest 6)llowing stimula-

tion(Fig.3a-c).Expression ofthe receptor R)r the immunoglobulin

恒gment Fc7(FcVR)oll days O,1,2and 3 was examined with aow

cytometry in M 1/Neo, Ml/SSI-1and M l/TR cells treated with lL6

plus slL-6R. Despite a partial inhibition of Stat3 phosphorylatiol1

(Fig.4a), SSI-1 almost completely blocked IL-6-induced FcVR

expression, which is a direct target gene of Stat3 in M lcells(Fig.

3d)6. These results suggest that SSI-1 inhibits the Stat3-mediated

b

黍●吉

.  一ぐ=.

a

・ 崎暴書ssエー・

垣輪■培■●蜘一ac七i・H B S I、u L Sm K Te

τ奪.剰肇苓蕃

酷●●●鼻壺ξ

一亨冨幕SS・一・        (L3kb)

    ■一●拳O●13-actin

誓看時.●●特幽悟一m・・

1

sセim.:   IL-6

ce11:   MH60

C

工L-6

M1

ce1⊥: M1/STAT3

   冨噺寧書

1L“・6 5tin.言 0 30 60 120

           弱

XL-4CT4S

》41/Y705F

懸:弓

●●●

3060120

G-CSFNFS60

M1/Neo→.

箏6SSエ.1

●●●1う一actin

30 60 120 m工n

●- SSΣ_1

国■屠駒鯵LIF Sヒim.: 0  30 60 120   30 60 120  30 60 120

β一actin

min

signalling Pathway

  Stimulation of cytokines of the IL-6 family induces tyrosine-

phosphorylation of the signal-transducing receptor componentgpl30 to activate Stat3 by∫AK kinases23-25. Tb determine how SSI-

lexerts its inhibitory effヒct on the gp130-mediated signalling

pathway, we measured the tyrosine-phosphorylation of gp130 and

Stat3 in M l cells a丘er stimulation with lL-6 plus slL-6R As shown

in Fig.4a, tyrosine-phosphorylation ofthese molecules was much

reduced in Ml/SSI-l cells compared wlth that in control M l/Neo

and M l!TR cells. Next we tested whether SSI-1 interacts directly

with IAK kinases. The cDNAs encoding SSI-land Iak20rTyk2 were

co-transfected into COSフcells;as shown in Fig.4b, SSI-1 co-

precipitated with Iak20r Tyk2. These results indicate that the

activity of IAK kinases is suppressed by SSI-1.]R)see whether the

inhibitory effect of SSI-1 is speci6c to∫AK kinases, we immuno-

blotted M 1/Neo, M 1/SSLland M l!TR cell lysates befbre alld a6er

IL-6 plus sIL6R stimulation with antじphosphotyrosine. In the case

ofMl/SSH,the intellsity ofthe bands corresponding to gp130 and

Stat3 was signi6cantly reduced compared with the same bands加m

control M 1/Neo and M l/TR cells. The intensity ofother bands was

comparable among M l/Neo, M 1!SSI-1 and M 1/TR ce田ysates. In

addition, there was no difference in the phosphorylation of Flt-3

and the insulin-receptor一βamong these cells a丘er stimulation with

the correspollding ligands. These results indicate that SSI-1

speci6cally inhibits the IAK-STAT signalling pathway(Fig.4c).

  Cytokine binding to the receptor induces homo-or hetero-

dimerization of the receptor components and activates IAKkinases that are constitutively associated with the intracellular

domains ofthe receptor components;once activated, these kinases

tyrosine-phosphorylate receptor components. STArs then associate

with phosphotyrosine residues on the receptor components thr皿gh

their SH2 domains and in turn are tyrosine-phosphorylated by∫AK

kinasesl-3・18・25. Here we have cloned the cDNA encoding SSI-1, an

SH2-domain-containing protein that is inducible by Stat3 and

inhibits Stat3. It has been reported that CIS participates in negative一

丘edback regulation of theハAK-STAT signalling pathwayl6・17. CIS

inhibits Stat5 fUnctionby directly associating with the Stat5-binding

region of the EPO receptor, but the mechanism of SSI-mediated

inhibition of the IAK-STAT pathway is quite different. SSI-1

reduces tyrosine-phosphorylation of Stat3 as well as of gp l 30;as

tyrosine-phosphorylation of gP 130 is a prerequisite五)r its associa-

tion with SSLl at the Stat3-binding region, SSI-l is unlikely to

compete with Stat3 fbr binding to gp130. SSI-lmaγblock an earlier

step in the IAK-STAT signalling pathway than CIS, probably by

inhibiting IAK kinases. We have shown that SSI-l associated with

Iak2 and Tソk2, and this molecule has recently been cloned as a IAK-

binding protein in a yeast two-hybrid system and also shqwn to bind

Iakl and Iak3(A. Ybshimuraε口1., personal communication).

Another point is that SSI-1 is not induced by all STATs, ruling out

arole as a general inhibitor of the IAK-STAT pathway

  We have shown that SSI-1is induced in response to cytokines that

act through gp 130, and to IL-4. In our experiments, NFS60 and TF-1

cells did not express SSI-1 in response to IL-30r EPO stimulation,

respectively(data not shown). The physiological signi丘cance ofthe

cγtokine-speci丘c induction ofSSI-1 needs to be fUrther investigated・

Nb’θ藺∂θ4加ρro(≠Elsewhere in this issue,∫AK inhibitors similar

to SSI-1 are reported under the names SOCS26 and∫AB2フ.   口

1 Figure 2 a, Expression of SS|-1 mRNA in murine tissues、 H, heart;B, brainl S,

spleen;Lu, lung;L、 liver;Sm, skeletal muscle;K, kidney;Tさ, testis. b,1∩duction of

SSI-1 mRNA in factor-depe∩dent cells, MH60 and Mτcells were stimulated with

lし6plus slL-6R. CT4S and NFS60 ce‖s were stimulated with IL・4 and G-CSF

respectively. cイηγc mRNA is shown as a contro|for stimulation, c, lnduction of

SSI-1mRNA泊Mlcells transfected with the indicated Stat3 constructs. M1/Stat3:

Mlcells with wild-type Stat3. M1!Y705F:Mlce【ls with a dominant-negative form

ofStat3. M1/Neo:Mlce|lswith a neomycinイesistance gene alone. lna-c,β一Actm

mRNA is i∩cluded as a loadi∩g contro1.

Methods

Preparation of the monoclonal antibody. A hybridoma clone producing the

monoclonal antibody(mAb)FL-238, which is reactive against the synthetic

oligopeptide(Fmo⊂)GTFLLRFS(this sequence is highly conserved in the STAT

白mily), was established by血sing spleen cells fξom BALB/c mice immunized

with the synthetic oligopeptide TKPPGTFLLRFSESSKEG(amino acids 600 to

617in the SH2 domain ofStat3)coupled to keyhole limpet haemocyanin with

mouse myeloma ceUs P3-X63-Ag8-653. The mAb was puri丘ed by protein A

affinity chromatography from the ascitic nuid of BALB/c mice.

Page 16: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

Isolation of SSI’1 cDNA. Using the monoclonal antibody against the

GTFLLRFS moti£SSI-l cDNA was isolated fVom a murine thymus cDNA

library(lambda ZAR Stratagene)by using a PicoBlue Immunoscreening kit.

Cell culture. Myeloid leukaem捻Ml cells were cultured in Eagle’s minimal

essential medium supplemented with double亡he normal concentrations of

amino acids and vitamins and 10%(vol/vol)fetal calfserum(FCS).The IL-6-

dependent myeloma MH60 cells were maintained in RPMU640 medium

supplemented with IL-6(5ng ml-1)and 10%FCS, IL-2/IL-4-dependent

CT4S cellsユo were cultured in RPMI I640 medium supplemented with IL-4

(10Umrl)and 10%FCS. IL-3-dependent myeloid NFS60 cells were main-

tained in RPMI 1640 medium supplemented with 10%FCS and IO%condi-

tioned medium加m the WEHI-3B cell line as a source of IL-3, We had

previously constructed dominant-negative fbrms of Stat3 and established MI

ce肛ines that constitし1Uvelγexpress them6. Tねns飴ctant M l cells were cultured

with 250μg m「l Genetidn(Gibco).

Northern hybridization. Cells(ex⊂eptMlcells)were factor-depleted負)r 4 h in

RPMI medium containing l%BSA, and then all cells were stimulated with

cγtokines at the fbllowing concentrations brvarious periods:IL-2,10ng m「㌧

IL-3,5ngml-1;IL-4,10Umrl;IL-6,正00ngm「1;slL-6R,50ngm「1;G-CSF,

20ng m「[;LIE 1、000 U m「1. Cγtoplasmic RNA was extracted using工so-Gen

(N口ponGene).]R)tal RNA(5μ.g rnl-1)was eleαrophoresed onagarose gels and

transferred to a nylon membrane(Hybond N+, Amersham). The membrane

was hybridized with radioIabelled cDNA probes.

Flow cytometry analysis.丁士ansfヒctants(1×10s)were cultured with IL-6

(300ng m「【)plus slL-6R(500 ng ml-1)fbr 3 days. After collection, cells were

incubatedwith lμ.g mouse lgG2a(CapeD fbr 20 min on ice. A負er washing with

PBS, cells were incubated with fluroescein-isothiocyanate-conjugated anti-

mouse IgG antibody(Zymed)fbr 20 min on ice. Cells were analysed in a

fluorescence-activated ceU sorter(Becton Dickinson FACS).

Plasmid construction and DNA transfection. Constructs were clolled into

pEF-BOS, a mammalian expression vector. Brieny, SSI-lcDNA was digested

with the restriction enzymes X加I and Pγ1‘H and then inserted into the X6‘11 site

of the pEF-BOS vector (pEF-BOS/SSI-1 (SSI-1)). For the constru⊂tion oF

pEF-BOS/SSI-1(TR),360 bp of a B55HII-digested f白gment was deleted.

cDNAs encoding lak2 and Tyk2 were inserted into the X加l site ofthe pEF-BOS

vector.

  MI ceUs were transfεcted by electroporation with the SSI-1(SSI-1/TR)

expressbnvector andpSV2 Neo at a 20:lratio, and neomycin-resistant dolles

a100

_  80

言言 609

示0  40

b

20

-.

lL-6 Stimulatlon

       0

       0

       

C

       10

E.●.。)⊂oる」098εΦ⊂三ε〉£プムエ“・)

0

d

D1 D2

lL-6 Stimulation

D3

100

BO

60

LIF Stimulation

  

20

10.000

o

Dl D2

LIF Stimulation

D3

Ml∫Nco Day O MUNeo Day l M1/N。o Day 2

、.‘ 1.

Ml/Neo Day 3

「lnl≧‘!{  ち  、   \^

C

M1/D3●

M1/SS卜1 D3

M1/NeoD3

・OM1πRD3

」3己」O己一一㊤OO>口目一〇

Ml/SSI-l Day O Ml!SSじI Day l MuSSLIDay 2 Ml1SS【-I Day 3

MlπRDayO  MlπRDay lA

Log fluorescence intensity 一FCYR

’’’’”榊  oon【r《ハ[

Figure 31L-6-or LIF-induced growth arrest and apoptosis of M l cb∩es. M1,

parenta「M]cellsl M 1/Neol Mlcel[s carryi∩g a∩eomycin-r’esistance ge∩e;M1/

SSI-1, MIcells with wild-type SSI-1;M1/TR, MI cells carrying muta∩t SSト1. a,

Parental a∩d tra∩sfected Mlce「lswere seeded with IL-6(300∩g ml-1)plus s[L-6R

(500∩gm「1)or LIF(11000Um「1)at day O, and巾e viabil[ty of the ce日s was

determlned on days(D)1,2and 3. b,[3H]thymidine incOrporation was measured

on day l a丘er stimulation, Each column represents the mea∩ of three

experiments.c, May-Grunwald-Giemsa stalning of parental and transfected Ml

cells on day 3. d, IL-6-induced expressiOn of the Fcvイeceptor, as determined by

¶ow cytometry.

Page 17: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

1

a

      r-■■■1 r-一一1r■■■■■■11L-6     十  _  十  _  十

b

鞠1.蕎 難

麟撲瀧薩.邊嚢

培嬉鱒壕oo

lcDNA3

軍但卿1鱈肇

 IP:α一gp130

Blot:α一pT>r

Blot:α一gp130

IP:α一Stat3

Blot:α一pTyr

Blot:α一Stat3

IP:α一Jak2

        、、    Blot:α一SSI-1

縫.轍捲,←SSI.1                                          

ll嚢1覆

一 一

Lysate

Blot:α一Jak2

Blot:α一SSI-1

Figure 4 a, IL・6 induced tyroshe-phosphorylatio∩ol gp130 and Stat3 in Ml

transfectants. Cells were either unstimulated orstimulated with IL-6 plus slL-6R.

1mmunOprecipitated (IP)gp130 and Stat3 were immunobloπed with anti-

phosphotyrosine antibody(α一pTyrl. Blots wel’e reprobed with the respective

a∩tibodies. b, Association of SSI-1 with」ak2 and Tyk2. COS7 cells were

transfected with the indicated combinations of plasmids encodlng SSI-1,」ak2

and 丁yk2. lmmunoprecipitates with anti-」ak2 0r antiTyk2 antibody were

immunoblotted with anti-SSI-1 antibody. The expression of these proteins was

examined by immunoblotti∩g the cel川ysates. c, The effect of SSI-10∩IL-6, F|t-3

1iga∩d and insuli∩-induced tyrosine-phosphorylat}on, Le硅:whole-cell iysates,

before and after lL-6 stimulatlon,were immunobloπed with anti-phosphotyrosine

a∩tibody. Right;cells were either unsWmulated or stimulated with Flt・311ga∩d or

i∩sulln for 5 min. lmmunoprecipitated Flt3 and i∩su}inイeceptor一βwere immuno-

bbπed with anti-ph6sphotyrosi∩e antibody and reprobed with the respective

a∩tibodies.

竈裏壁響一’.    IP:α一Tyk2

               Blot:α一SSI_1

-⊆濠≒警藁団簾ぐ一SSI-1

に二藷薗覇■i

軸 齢

1・ysate

Blot:α一Tyk2

Blot:α一SSI_1

C ざぎξlL-6’:一で「=でごで

”r(K)

  220一

97.4一

66一

46一

30一

21.5一

14.5一

” :

聾二:毒懇難

Whole-cell

  lysates

Blot:α一pTyr

Flt-3

hσand o

蕊喜ξr■■■■1「一■■■1「一■■■1

 _  十  _  十  _  十

   葛 覇 “

眉■曽■■■

     ぷ㌔e“、ぷ

[・・山n「ニーで「=で:一で

    ぎ罐鷲.簸鵜        ・ 1 :-  5 ← ・・署   1

    智●憎憎楢増

 IP:α一Flt-3

Bbt:α一pTyr

Blo【:α一Flt-3

 IP:α一【nsulin Rβ

Blot二α一pTyr

Blot:α一lnsulin Rβ

were selected in growth medium containing GeneUcin(Gibco)at 750時m「1.

Western bbt analysis. Cells were treated with IL-6(300 ng m「1)plus slL6R

(500ng ml-1), Flt-3∬gand(300 ng m「1;Genzyme)or msulin(1μM;Becton

Di⊂kinson)氏)r 5 min or were untreated, and were solubilized with lysis buf丘r

(05%Nonidet P-40,10mM Tris, pH 7.4,150 mM NaCl,1mM EDTA,1mM

NazVO4)containing protease inhibitors. Immunoprecipitates obtained with

anti-gp130(Upstate Biotechnology),anti-Stat3(re£4),anti-Flt-3(Santa Cruz

Biotechnologγ)or anti-insulin-receptor一β antibody (Santa Cruz Biotech-

nology)were resolved by SDS-PAGE ullder reducirlg conditions and trans一

丘rred to nitrocellulose. Filters were probed with anti-phosphotyrosine

monoclonal antibody(4GlO;Upstate Biotechnology)and reprobed with

anti-9P130, antCStat3, anti-Flt-30ranti-insulin-receptor一βantibody, respec卜

ively, afterstripping the firstblots. Blots were visualized with the ECL detection

system(Amersham).COS7 cells(1×106)were transfごcted with the designated

plasmids encoding SSI-1(2μg)Jak2〔20 Ug), or Tyk2〔20μg)by using the

DEAE-dextran method. A丘er 48 h, cells were solubilized ill lysis buffとr and the

正ysates used fbr western b[ot ana[ysis.

Received 26 Mar⊂h;accePted 70vlay 199フ.

1. Darllell,1. E. Ir, Kピrr、 L M,&Stark. G. R. Iak-STAT pathways and transcriptiαul activatio|l m

  responseωIFNs洲d other ext口⊂el|uhr signa|ing Proteins・5dcノκど264,1415-1Ω1U994L

2. Ih|∈」. N, Cytokine receptor sig1)alling. Nσ’tげ’377,591-594(1995).

3, Kishimoto, T.,Akirユほ,,Namzaki, M.8d、ga、 T. Interleukil卜6 family ofcytokines and gpl30. Bb〆

Page 18: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

   86,1243-1254(1995).

4.Akira, S.’rα’. Mol㏄ular clon拍g of APR耳anovel IFN-stimula【ed gene畠ctor 3 p9トrelat¢d

   tmnscription倫ctor mvolved in the gpl30-mediated signaling pathway.α〃77,63-71(1994).

5.Zhong, Z., WeぴZ.&Darnell, L E. lr Stat3:aS脚「f這milγmember activa匡ed by tyrosine

   phosphorγlation in response 【o epide’rnal growth 倫ctor and in【erleukin-6.5‘∫e’1cε 264,95-98

   “994).

6,Minami, M.’M’. Sm3 a‘【ivation is a critical step in gp130-mediated terminal dif勧entiation and

  growth arre5t ofamyeloid cell lme. Prθ‘.ハ倫r’Aα34.5‘輌. L6A 93,3963-3966(1996).

7.Nakalima, K.どfαt Acmtra〔role fb【Sta6 iロ[L-6-induced re8ulation ofgrowth and dif丘ren蛭aUon in

  Mlleukemia cells.εMβ0/15,3651-3658(1996).

8.1、ga, T&Kishimoto, T Cytokine receptors and signa[transduction.昂SεB J 6,3387-3396(1992).

9.[p,N,Y¢rα’. CNTF and LIFact on neuronakells via shared 5igna∬ng pathways that involve the IL’6

  signal【ransducin8 re㏄ptor componentgpl30.(乏”69,1121-1132(1992).

10.Kondo, M.εは}. PuncUonal participation of【he IL-2 receptor Y chain in Iレ7 receptor comp[exe5.

   S‘i’η‘ε262,1453-1454(1994),

lLSakamaki, K., Miya戸ma.1., Kitamura, T&Miyalima, A. Critical cytoplasmic domaisn orthe common

   βsubunit ofthe human GM・CSE IL・3and 1L-5 receptors lbr growth si6nahransduction and tyrosine

   phosphorylation. Eハイβ0∫.11,3541-3549(1992).

12.Heldin, C. H, Dimerization ofcell surf己ce receptors in signalぬnsduction.α’∫80,213-223(1995}.

13.Adachi. Mε’α’. Mammalian SH2-containing protein tγrosine phosphatase5.α〃85,15(1996).

14.Yi, T. Mui, A. L.-R, KrystaしG.&lhle,1. N. Hema⑩poietic cell pho5phata託a550ci3tes with the

   interleu㎞n-3σレ3}receptorβchain and down-regulates IL3-induced tyrosinephosphorγlation and

   mito8enesi⑨. Mぴ1. Cε麗. BioL 13,7577一フ586(1993)・

15.Klingm口ller, U, Lorenz, U., Cantle拓L. C..N㏄1, B. G.&Lodish, H. Speci6c recruitment ofSH-PTPl

   to the erythropoietin receptor cause5 mactivation of IAK2 and termination ofproli烏mtive signal5.

   Cビ∫∫80,729-738(1995),

16.Ybshimura, A.’ω1. A novekγtokine-inducible gene CIS mcodes an SH2・containing protein tha【

   binds to tγrosine-phosphorγlated interleukin 3 and erythropoiαin receptors. Eハイβ0/14,2816-2826

   (1995),

1ア.Mui, A. L., Wakao, H., Kinoshita, T, Kitamura, T&Miγalima, A. Suppression of interleukin-3・

   induced 8ene expression bγaC-termmal truncated Sヒa【5:role of Stat5 in proliferation.εMβOノ.15,

   2425-2433(1996).

18.Heim, M. H., Ker鴫1. M Stark. G. R.&DarnellJ. EJrContribution ofSTAT SH28ro叩s to 5peci6c

   interfヒron 5ignalin8 by【he Iak-SIDW pathwar S‘輌z”‘ε267,1347-1349(1995).

19.Hou,1.ε伽1. All interleukinピ4-induced transcripUon白“or:IL4 Stat. Sc治κ■265,1701-1706(199引.

20.Hu-LiJパOhara, L, W包tson. C.,τ㌔ang, W&Paul, W E、 Derivation of a T cell line that is highlγ

   responsive to【レ4 and IL-2(CT.4R)and of an IL・2 hyporesponsive mutant of that line(CT.4S)」.

   ∫’H’,R‘‘rタo’,142,800-807(1989).

21,Tian, S. S., Lamb, P,Seidel, H. M., Stein, R. B.&Rosen,ハ. Rapid aαivation o自he STAT3 transcriptioo

   白ctor by grarlulocyte colony-stin、ulating白ctor. B’oo484,1760-1764(1994).

22.Wakao、 H., Harada、 N、、 Kitamura, T、 Mui、A工.-E&Miyalima、 AJnterkukin 2 and erythropoietin

  activate STWT5/MGFvia di5tinαpathwaγsεMβ0∫.体2527-2535(1995).

23.Narazaki, M.ε’鐸’. Activation o臼AK2 kina5e mediated by the interleukin 6 signal t陥nsducer gpl30.

   Pr㏄ハ畑㎡ノ1α㎡. Sc∫, LM 91,2285-2289(1994).

24.Stahl, N.ε∫θ’. As50ciation and activation of Iak・Tyk kinases bγCNTF・LIF-OSM-1し6Rβreceptor

  components. S‘迦c8263,92-95{1994).

25.Stahl, N.’fα’. Choice ofSTぷand other substrates specified bγmodular tyrosine-based motifS in

   cγtokine recep【ors.56iεπ‘’267,1349-1353(1995)・

26.Starr, R’rαL A白mily ofcγ【okine・inducibk inhibitor50fsigna‖ing. N白fr‘㎎387,917-921〔1997).

27.EndoひTA.’白L A new protein con【ainingan SH2 domai“that inhibit5∫AKkim5e5.袖’εrr’387,921-

   924(!987).

Ackn◆wled8em頭臼. We thank T T㌔naka fbr IL-4 and CT4SceUs;S. Naga【a和r p£F-BOS ve“or and∫ak2

cDNA;1、 Krolewskj㊤rτyk2 cDNA;K. Yasukawa br recombinant IL・6 and slL・6R;YShima. H. Danno,

K.Kunisada and H.τ㌔goMbr【echnical assistallce;H.Sai【o fbrdiscu55ion;and A. Nobuhara fbrsecretarial

assi⑨tance. This work was supported bγaGmnt-in-Aid fヒom the Ministry of Educa【ion, Science and

CultureJapan.

Corre5pondence and requests lbr ma【erials should be addressed【o T.K.(e-mail:kishimot@imed3.med.

osaka-u. ac.lp). The sequence of SSI・l has been deposited with Genbank, under acce品ion number

ABOOO710.

Nature Japan K.K. Shinmitsuke Bldg.3-61chigayatamachi, Shiniuku-ku,Tokyo 162 Tel O3-3267鴫8751 Fax O3-3267.8746

Page 19: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

Pγoc. N{HI. Aca{三. Sct. USA

=罐y15577-15582・Decembe「1998

Accelerated apoptosis of lymphocytes by a㎎mented induction ofBax in S SI-1(STAT-induced STAT mhibitor-1)def董cient miee.、

TETsu∬NAKA*, ToMosl41GE MATSuMoTO*, MASAS田NARAzAKI*, MINORu FuJIMOTO*, YOSHIAKI MoRITA*,YosHIYuKI OHsAwA†, HIRosHI SArro*, TAKAsHI NAGAsAwA‡, YAsuo Ucm’YAMA†, AND TADAMnsu KlsHIMoTo§¶

ホDep頷ment of Medicine m, Osaka University Medical School,2-2, Yamada-Oka, Suita, Osaka 565-0871, Japan;†Depar畑ent of Cel田iology and Anatomy l,

Osaka University M¢d▲cal School,2-2. Yamada-Oka, Suita, Osaka 565・0871, Japan;‡Depart:m¢nt of Immunology, Rescarch lnstit限, Osak3 M¢dical Cemer for

Materna]and Chi】d Health,840 Murodo・cho, Izumi, Osaka 594-1101, Japan;and§Osaka Uni∨erslty,14, Yamada’Oka, Suita. Osaka 5654)871, Japan

Confγibμ絃d by T敏iαmiεsμKぴMmo‘o, Octoわer五9,1998

ABSTRACT  Grow出, dif食㏄耐iation, and programmed伴11deat血(apoptosis)are mainly co聴trolled by cytokines. The

J8nus kinase-sig田Hransducers and activators onranscrip・

ti①n(JAK・STAT)signaハpathway is an important c¢mponent

o吋tokine signali㎎. We㎞▼e pr崩ously曲o㎜伽t STAT3induces a mole㎝1e design8ted as SS1・1, wkic血i歴hib麓s STAT3

血nctions. To clarifシthe physiological roles of SSI・1∫π’∫wo, we

generated, here, mice lacking SSH.丁血ese SSI・1-/-mice

displayed growth retardation and died w“hin 3 weeks a髄er

birth. Lymph㏄y民s in也eω1ymus a戯d spleen of the SSL1-/-

mice e】血ib“ed accelerated apoptosis with aging, and 1ヒ11ei「

number was 20-25%of tbat in SSI・1十/十mice at 1①days of

age. However,也e diff¢re耐iation of Iympllocytes lac㎞ng SS1・1

apPeared加be normaL AmongΨarious pro・孤d組ti・apopto・tic mole㎝les examinedg an叩・regulatio歴of Bax was f㎞und in

Iympbocytes of也e spleen aod tbymus of SSH-/-mi㏄.

These nndings suggest tbat SSI・1 prevents apoptosis by

inhibi6ng the expression of Bax.

The homeostatic regulation of cell populations is controlled by

abalance among proliferation, gmwth aπest, and apoptosis,

and this balance姪mainly contro狙ed by cytokines and growth

factors. Cytokines act by binding to receptors expressed on the

surfa㏄s of responsive㏄11s, which are associated widl one or

more members of the∫anus kinase(JAK)family of cytoplas・

mic tyrosine kinases. The JAK-signaHransducers and activa-

tors oftranscription(STAT)signal pathway plays an important

role in cytokine signaling(1-3), and is unique in that it features

a dむect Unkage of receptor-ligand interaction on the cell

surface to gene expressioll in the nucl田s(4-6). However, the

m㏄hanism of negative control of cytokine actions inv◎lved㎞

1imiting their signal transductions is comparatively lcss well

characterized. In l 997, the molecules that were expressed by

stimulation of cytokine such as interleukin 6(IL-6)and

inhibited cytokine signal transmission by binding to JAK were

isolated【STAT-induced STAT inhibitor4(SSI-1), suppressor

of cytokine signaUng(SOCS-1), Jak-binding protein(JAB)]

(7-9).Subsequently, SSI-1 was found to負)rm a family con・

sisting of at least eight molecules, which were struc知rally

characterized by an SH2 domain and a C-terminal conserved

region(SC・motif/SOCS-box/CH・domain)(10-12), and it was

recelltly known that SSI4 inhibits not only IL6 sigllaling but

also inter feron(IFN)一γ, IL-2, IL-3, and growth hormone

signalingか1 v∫’〃フ(13). H is expected that furtber study of SSI

fami蓋y molecules engaged in the negative feedback mechanism

of cy⑩kines will clari旬the control mechanism of cytokines,

WhiCh have remained ObSCUre.

rhe pub▲ication costs of th輌s artide were defrayed in part by page charge

payment This a∬三cle musUherefore be hereby marked‘‘α4ve所∫θ朋¢M”in

accordallce with 18U.S.C§1734 sユ)lely to imdlcate this facL

◎1998by The Nationa[Academy of Sciences OO27-8424/98/9515577-6$2.α)/O

PNAS is available onlinc at www.pnas.org.

          MATERIALS AND METHODS

  Generation of SSH・Deficient Mice. A 129/G mouse

genomic library(Stratagene)containing the SSI-1 gene was

screened, subcloned into the pBluescdpt vector, and charac-

terized by restriction endonuclease mappmg and DNA se-quencing. A targeting vector was designed to replace the SSI-1

intron wah phosphoglycerate kinase neo. This target輌ng vector

was f星anked by a 5.5-kb fragment at the 5’end and a O.8-kb

fragment at the 3’end and contained a PGK-tk cassette at the

5’end of the vector. It was Hnearized with 8απIII and

el㏄troporated into embryonic day 1.4.1 embryonic stem ceUs.

αones resistant to G418 and gancyclovir were screened for

homologous recombination by PCR with the primers shown in

Fig.1and con五rmed by means of Southem blot analysis with

the probe C4ccl andゆηI fragmenりshown m Fig.1. The

homologous recombinant embryonic stem clones were in-lected into blastocysts of C57B1/6J mice and transferred into

the uteri of pseudopregnant C57別/6」允males. The resulting

chimaeric mice were backcrossed to C57B1/6J mice, and

heterozygous mutants were ident近ed by means of PCR and

confirmed by using Southem blotting from tail DNA. Brother-

sister matings of heterozygous mice resulted in homozyg皿s

mutants. All an imals were housed皿der specific pathogen・

free conditions.

  Nor出em Blot Analysis. Total RNA was extraded from the

lung tissues of SSI4-/-a皿d SSI-1+/+mi㏄with RNA ZOL

B(Tel-Test, Friendswood, TX), and was electrophoresed,

transfeπed to nylon membrane, and hybridized with 32P-

labeled mouse SSI・1 cDNA.

  The 3・(4,5・DimethyHhiazoL2・yl)・2,5・diphenyl Tetraz①1拍m

Bromide(MTT)Dye Conversion. Cell viability and number

were assessed by MTT stainlng essentially as described by

Mosmann(14).

  now Cyto皿etry. Thymocytes and splenocytes were pre-

pared from age-matched SSI-1+/+or SSI’1-/-mice. Red

ceil-depleted splenocytes were obtained after treatment with

Ack buffer(0.15 M NH4Cl/1.O mM KHCO3/0.1 mM Na2

EDTA. After a washing with PBS, they were counted andresuspended in a staining medium. A total o臼06 cells was

stained as described in the protocols by using the fbllowing

antibodies(Abs):anti・CD4, CD8, CD3, B220, IgM, Mac-1, and

Gra-1 antibody(PharMingen). The stained cellswere analyzed

by means of double・color flow cytometry on a FACScalibur

(Becton Dick輌nson)by using cBLLQuEsT software(Becton

Dickinson).

Abbreviatめns:STAT, signal transducers and act量vators of transcrip-

tion;SSI-1, STAT-induced STAT inhibaor4;SOCS-1, supressoτofcytokine signalillg;JAB, Jak・bin(∬ng protein;Bax, Bcl-2-associated X

protein;Bcl-2, B㏄111ymphoma/leukemia・2;JA】KJanus kinase;IL溢terleukin;INF, interferon;TUNEL terminal deoxynucleotidy1-transferase-mediξ1ted UTP end labeling;Ab, aIltibody.

¶To whom reprint requests should be addressed.ひmail:kisbi皿ot@

imed3.med.osaka-u.ac.jp.

15577

Page 20: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

15578 Immunology:Naka鋤1. Proc, A1∬∫1..4 cα4。 Sぼ.σ∫〆{95 r1998ノ

A甲1      、ニ…一…㌫…一評鷲㌃,

醐1⌒_紳_ぷ㍉⊇竃㌫ぷ

   6、撒魯

    シ                        ヨ

  1柵灘髭灘る  ;     い祐齢1    〈                 ’

        1

1D,肋畑;

師……

c-1

7

  酬A  魎,|

           ミ

                     

撫増  墓

                                  腔で解

蝋nvポt《パ緬礁燃w、嚇・ぷ叫w㍊ぷ,㎞撒一

←≡馳…ぺ…←……………㌫弗瓢=_⊥__見_醐㌶、

紺iぱW停

事駅㈱‘均塗府c

6

“簿

A.2SS‖4

  6

{翌主鱒

(≠)

砦・

畠3{+古) {÷ρ)

蜘Ww魂警  ・  . 、、、 ,.,

          蓼公’啄旛毒溺孫

B

  ←6.8kb醸←5.8kb

2

1

十$SM癖耐←$Slが1φ

才    .ゴ▼   .♂..〆

丁 3  5

Daysア 9

SSレ1

(+/+) 《・ρ)

SS事1

(t3kb}

c-2

β浩actin→ベペ・

○念・

100

欝80ご雲60

’巨

む040

20   

   

o 5 10

Days搭 20 25

 FIG,ユ. (λ)Homologous recomb豆nation oぎthe SSIヨ1()clls.し4⇒Restrictioll map of the 129/G murine SSI4 genomic clone, targeting veclor,

aDdhomologousrecombinant, PGK-neo replaced a O.6-kbλ71イ)IW戊/I fragmentcontal曲lg the great part ofexon 2. A herpes simplex virus thymidine

kinase(HSV・TK)with a PGKpromoter cassette was present on the 5’flank of the vector.し4わl Representative southern analysisof progeny fron〕

+/一×+/-mating. Tail biopsy genomic DNA was digested withκρ’11 and probed with a southern probe. The wild-type allele is 5.8 kb. aIld lhe

homologous recombinant exhibils the predicted 6、8-kb band.(B)SSI-1一たmice showed no evidellce o∫SSI-I mRNA. Norlhern blot analysis of

lung tissue lysates from SSI-1+/+mice with the SSI-1 cDNA probe exhibjted expression of L3-kb SSI-l mRNA but SSI・1-/-mice lacked this

product.β一actin mRNA was included as a loading controL(Cα)SSLImlce showed an碍4〔}%decrease in body welght by 9 days o臼ge, The mean

weight of SSIヨ+/+is shown by lhe solid line and lhat of SSI-]一/-by lhe broken line、 Each lllle represents the Inean of seven experimeIlts.(α))

SSI-1 mice all died between 2 and 3 weeks after b▲rth、 Percentage survivals of SSI-1+/+are sh《〕wn by△and of SSI-1-/-by●.

  Term輌na団eoxynucleotidyltrans丘rase-Mediated UTP End

LabeHng(TUNEL)Staining. For detection of DNA fragmen-tatioll加5iご1‘、 paraffin-cmbedded sec重ions werc testcd by the

TUNEL method as dcscribcd clsewhcre(15). TermiDal de-

oxynucleotidyltransferase(TdT)was uscd to incorporate Bi-

otin46-dUTP(Boehringer Mannheim)into the ellds of DNA

fragments. TUNEL signalswere detccted w三th古e aid ofTcxas

Red-conlugat¢d anti-avidin Ab(1:200, Biomeda, Foster City、

CA).

  lmmunobistochemical Staining. Paraffin-cnlbedded scc-

tions were immunostained by a mcthod describcd clscwhcre(16).Sections werc trcatcd with allti-Bcl-2 Ab(1:玉00, Biomol,

Plymouth Mccting, PA)or anti・Bax Ab(1:]〔}0)and then

hlcubated w糺h fluoresccin isothiocyanate-conjugatcd goatallti-Hamstcr IgG(1:300、 PhきLrMingen)or fluorcsccin isothio-

cytmatc-co1亘ugatcd goat a11{i-Rabbit IgG(工:200, Sclkagaku,

Kogyo, Tokyo),

 肋附ぴoThymocyte Cumlre. Fresh thymocytes were platcdat 2× 1{〕5 cclls pcr 100μ1 in a 96-well plate丘n medium

containing 5%fetal calf serum and stimulated with anti-CD3

(1μg/ml)Ab、 anti-CD3 Ab plus IL2(20 ng/ml),anti-CD3 Ab

plus IL-4(20 ng/ml), or Ilot stimulated. Cell viability was

assayed by mcans of MTT dye conversiol1. For each day,

triplicatc cultures wcrc counted and avcraged・

  Westem Blotting. Thymocytcs wcre stimulated with 1μg/ml

anti-CD3 Ab(PharMingcn)p至us 20 ng/ml IL4(Pepro TcchEC, Rocky Hill, NJ)at 37°C for l hr or 3 hr. or Ilot stimulatcd.

Cclls were solubilizcd with lysis buffcr(0.5%Nonidct P40βO

mMTris、 pH 7.4/150 mM NaCl/l mM EDTA/1 mM Na2PO4)colltaining protcasc inhibitors. Immunol)recipitates obtained

with anti-STAT6 Ab(R&DSystems)were blottcd withami-phosphotyrosille Ab(4G10;Upstatc Biotcchnology)とmdrcbloUcd w{tll lmti-STAT6 Ab after stripping thc f▲rst b]ot.

Wholc ccll lysatcs wcrc blottcd with anti-Bax Ab or anti-Bcl-2

Ab(both Santa Cruz Biotcchnology).

Page 21: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

Immunology Naka e’α乙

RESULTST〈)clarify thc f田1ctional and devclopmental rolcs of SSI-1,wc

uscd homologous recombina目on in embryonic stem cclls to

gencrate mice lacking SSI-.L Thc disrupted SSI4 allclc was

ultimately transmitted through the germ line, as confirmed by

Southem b]ot(DNA)analysis(Fig. L4). Northern b[ots con-

firmed由e至oss of SSI-1. mRNA ill lung tissues f1“om SSI-1-/-

mice(Fig.1β). Homologous mutam mice(SSI一玉一/一)wcre

hcalthy and Ilormal at birth and were initially indistinguishablc

from control littermates(SSI-1+/+). Al也ough the body

weight of SSI-1-/-mice did not differ sig【lificantly from that

of SSI-1+/+nlice on postnatal day 3, an apProximately 40%

decrease inbody weight was observcd on postnatal(lay 9(Fig.

1C). Moreover, SSI4-/-mice all died within 3 wceks after

birth(Fig.1C).

  SSI4 has been shown fπ↓・iηηto be a negative fcedback

血ctor in the JAK-STAT signal pathway. We therelbre carried

out the fk)llowing experiment to dctermine whether lympho-

cytes la.cking SSLl exhibited prolonged activation of STAT.

We stimulated the thymocytes of 10-daydd mice with allti-

CD3 Ab plus IL-4 and examincd phosphorylation of STAT6 at

O,1,and 3 hr after the stimulation. In the硲ymocytes lacking

SSI4, the phosphorylation of STAT6 was not transient and

stnl d戯ected 3 hr after the stimulation(Fig2!4). The thymo-

cytes of 10-day-old mice were then stimulated similarly with

anti-CD3 Ab, anti-CD3 Ab plus IL-2, and anti-CD3 Ab plus

IL4 and survival of thymocytes on days 1,2, and 3 was

examined by using MTT dye conversion,rcspectively(14)(Fig.

28).Without stimulation or with anti-CD3 Ab alonc, apoptosis

occurred in thymocytes fr◎m both SSI-1+/+and-/-mice、

but the addition of IL-20r IL-4 reduccd apoptosis and the

number of living thymocytes increased. In the thynlocytcs of

SSI一仁/-mice, IL2-or IL4-induccd lymphocyte prolifera-

tion was significantly augmented compared with SSI-1+/+

mice. These findings suggested that this increase in prolifer-

ation cou豆d bc due to the impaired ncgative fbedback nlech-

an isrn of cytok▲nc signals.

A

lP:{㍗§汀Aτ6

810t:{叩Y

ip:《遮・釘∩T6

810むξ炉窪『真マ6

B o、憎 ,     {

 o’幡1.

 o、情 L

 田2{

§・・tl           

8α゜e{ o.06 L

l:;1

∈貝1㎎目:α∈CO3・争ILp4

熱幽鐡. 謹箇圏

⑭働㈱鑓}⑪㎜

。路

  SS1・1碗コ{弓

駆痴≡)3

目櫨㎜1し2獣震03村し4

褒※

Da蝉   函y2   0日y3

o相。戊6}

°・14 10・~2≧

o。パ。.頒[

。,。6i

。,。4!

。、・2}

  o

   SS目≠ o{・}

 ●磁}3  旋嘘董}無葦L^2

 駆纒}袈IL4

       冒

膓ゴ§

Da¥1     £}震y2

よ曇翻ヂ…方゜

D8y3

 FlG.2. (え)The activation of STAT6 continued after stinYulation

by IL-41n thymocytes of SSI4-/-mlce、 Cellswere either nonstimu-

lated《)r stimulated with IL-4 plusα一CD3. Immunoprecipitated(IP)

STAT6 and STAT6 were immしmoblotted with antl-phosphotyrosineAb(α一pY).Bk)ts were rcprobed wlt1]α一STAT6 Ab.(B)MTT staillhr|g

was nleasured on days 1,2, aDd 3 arter stnllulat三〇11. Each column

represents the mean of seven exper{mellts.

Proc、 Nαtl. Acacl、 Sci、 USA 95 日998) 豆5579

  When the splecll and thymus of SSL.1-/-mice were

examincd histologicaUy by means of hclllat《)xylin/eosill stain-

ing, Uttle differencc was seen bctween thc number oflympho-

cytes in SSI4-/- and SSI-1+/+ mice on postnatal day l,

whercas the numbcrs of】ymphocytes imhe spleenandthymuslacking SSI4 had decrcased markedly at 10 days of age. The

splecn of SSI-]一/-mice at 10 days of age was distorted in

shape alld exhibited p‘1rtial atrophy of the white pulp. In the

thymus, moreover, a borderlinc bctween cortex and mcdulla

was not visib]e(Fig.3!4). Tlle number of total thymocytes and

splcnic lymphoid cells of SSI-1-/-mice markedly decrcased

and at 10days of age, approximately 75-80%of lymphocytes

had disappeared as conlpared with SSI-1+/+ (Fig.3β).

Howcvcr, the nu.mber of bone marrow cens did not signifL

cantly dlffer bαween SSI一玉+/+and SSI-]一/-mice(datanot

shown).

  Next, to exam輌ne lymphocyte differentiation in the splcc11,

thymus, and bonc marrow, the splenocytes, thymocytes, and

bonc marrow ceUs of SSI-1-/-mice a仕O days of age were

stailled for two-color flow cytofluoromctric an副ysis with Abs

加B220, CD3, CD4, CD8, Mac-1,Gra-1, alld IgM(Fig.3C).

In tlle splcen, thc number of B220+or CD3+cells in SSI-1-/-

mice were markcdly dccrcased comparcd wi由those in SSI-

1+/+mice. However, thc ratio of B220/CD3 splenocytes

rcmained unaffected dcspite a massive decrease in the number

of splenic lymphoid cells in SSI-1-/-mice. The nllmber of

Mac-1+and Gra一仁splenocytes did not signi五cantly differ

betwcenSSI4-/-andSSI4.+/+mice(datanot show11).Thenumber of thymocytes in SSI-1-/-mice was only 25%of that

of controhhymocytcs(Fig.3β). However, the perccntages of

CD4-+’CD8’一’, CD4’’”CD8’ト, and CD4層→’CD8’ト幽ceUs did not differ

betwcen SSI-1+/+and SSI-1-/-mice. Similarly, the peF

centages of IgM+and B220+cells in the bone marrow did not

significantly differ bctweenSSIヨ+/+and SSI-1-/-micc. In

SSI・・1-/- mice, thc sp呈cen and thymus apPeared to exhibit

normal diffbrentiation of lymphocytes despite a massive de-

crease of the numbemf lymphocytes. These findings sllggested

that SSI一玉is not essential五)r the genera目on anddぱferentiation

of▲ynlphocytes but that it may be important fbr thc survival of

lymphocytes↓ηv↓りo.

  We cxamined the spleen and thymus of正0-day-old nlice

immunohistochemically by means of TUNEL staining(15)to

determine the callsc of the decrease in the numbcr ohym-

phocytes in SSI4-/-mice、 In the spleen and thynlus of

SSI-1-/-mice, there were more cells labeled with TUNEL.

than m those ofSSI-1+/+mice(Fig.44)」n addition, clectron

microscopic examination detected a larger number of apopto-

tic cclls with the typical condensation ofnuclear chromatin and

apoptic bodies in SSI一玉一/-mice(Fig.4β). The spleen and

thynlus of]0-day-old mice were then stained immunohisto-

logically(16)with anti-Bax Ab or anti-Bcl-2 Ab to clarify the

mechanism of acceieratcd apoptosis of lymphocytcs. Com-pared with SSI-1+/+ mice, the cells expressing Bax had

markedly increased in the spleen and thymus of SSI-1-/-

mice, However, there was no significant difference ill BcL2

express丘on betweell SSI-1、+/+and SSI-1-/-micc(Fig.4C).

Moreover, no difference in the cxpression of molecules induc-

illg apoptosis such as Fas, Fas ligand, tumor necrosis factor一α,

and glucocortisol, could be detccted bctwccn SSI-1+/+and

SSI-1-/-mice(da.ta not shown). Thcsc findings suggcstcd

that acceierated apoptosis of lymphocytes in SSI一仁/-mice

may be due to the augmented expression of Bax, resulting in

areductlon in the lylnphocytcs.

DISCUSSIONThe cytokinc signal pathways that include Ras-MAP alldJAK-STAT sigllal cascadcs are kllowll to bc involvcd ill the

generation of an alltl-t|Poptotic signal, Inost likely throug|l the

illdl1αion of the Bc[-2(17-19). Our rcsults showed that in

Page 22: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

15580 Immunology:Nakaピτ‘」1. Proc. NαtL Acα‘量. Sci、 USA 95  (1998)

A

B

 100

 go

 80

ガ゜≡6°

繧50言、。

 3.0

 20

 10   0

Xld7

SSI・1  SS口   SSI・1  SSト1

(+/+) (一〆一) (+!+)(一/一)

   01        010

Splenic bmphoid Cells

 10,0

 90

 80 アむ

』6。日』5・°

34°

 30 20

 10   0

X107

SSト1   SSI肯1   SSI.1   SSト1

(+/+),,(一/“)(+/・ ↓1。(一〃’)

    Thymocytes

C

衿ΩO

SS縄悟Spご斑0

21.1% ぴ3%

   .タ:、

 、 ξ惑.ギ」.

爲ε.  メ

∵wざs  ξ ・

、丁ぴ  ・ ≡  ぴ つ ゴ

 ー、ぐ撒し一浪【一..F可

 .・

三鎌、鍍

 モL1 ρ

籔5統5

80

村≠86%

      へ    〆    〉 り  ・,㌧㌧バ㌧こ パニ灘く∵ ○’”炉吟s

        t6%

 「

     「

”      冑、、、・〔二  、ド「・

   ’    ■・「    w

噛. 馳 w白

、’f.・

㌧’.  馳1撃∨  ⊂、.ヅ谷       「、AA’

:〉:・㌧∴弓・芦  ぷ、し評㌢∨ ご 剛わ・ニニ言:・

≡へ’璽26%

Thy鰍培D拍10.1%      ㌔           ぺ   86.196、   ,

    し  ♪ ’ ㍉     v           c

∵糠1零;      τ’∀ 

.   s、べ      「

    ,

1幽1層    メ

.A      ’     ’f    ま.

      ’ 1  み   、

   敏゜’

    、ル

     \   こ移’

’” ・へい・

、 ウ   ’

、’ ’

.9も  :・’2.

§oロ

CD8

Ψ口O

脚7.9%       ノ

見鵡∵子癒泣^  元s’.          戸 ∴              A  -

.ζ. 幽

    監冷

礫章涙ピ

  19

、1’ぶ  葵:’

・㌶,、三、 三ぴ.,’

 1:’: 、

  ’・「’ゴ’ ;7

 ’.

t別ら o,9今も

Bow●ma肪ow ol o

7.2%  ・ α8%・ぽば,く鰻㌘.\季㌶;辱

 ’イ;・

「’A

∴、1

 ρ、1:瞳一㌧え

   噛       :

鰻討1,

lg鯖

黛RO

lg鯖

 FIG.3. (渥)Histological appearallces of the spleen and thylnus in SS1-1-/-mice at l and 10 days of age. Hematoxylin/eos三n stained sections

of the spleen and tllymus from SSI-1-/-nlice 10-days-old revealdecrease(|lymphocytes(X80). The low magnincalion(X25)reve《!1s a dislorted

sllape alld partial atr(〕phy of white pulp in the spleen alld all indistinct borderlhle between cortex and medulla in the thymus of SSI-1-/-mlce.

(B)Totahhymocytes and splenic lymphoid cell numbers ill SSI-1+/+m▲ce alld-/-mice、 The means and standard deviations fOr l O+/+alld

lO-/-mice are showll. Numbers of total thymocytes and splenic lymphoid cells for SSI-1+/+are shown by empty bars. and㊤r SSI一仁/-by

filled bars.(C)Flow cytometric allalysis(、f splenocytes, thymocytes and bone marrow cells from SSI-1+/+and SSI-1-/-Inice l O-days-01d,

tllymocytes lackillg SSI-1,cytokinc-induced proli飴ration was

augnlcnted by pr〈)longed activation of STAT↓ηv↓加(Fig.2メ

andβ). This finding illdicates that, a》cxpcctcd, SSI-l plays an

importam role in thc ncgative fヒcdback mechanislll of thc

cytokinc. Howevcr, m thc thymus and spleen, a largc number

《ハflynlphocytcs lacking SSI-1 showc(l cvidcncc(〕f apoptosis

w狂hag{ng, which illvolved an lncrcuse ill Bax cxprcssion”z 1,∫1,0

(Fig、4/1-C). Thcsc sccmingly contradictory findings may

indicate the p〈)ssibility that SSI-hnhibits not only thc JAK-

STAT signal pathways, but also other signal pathways, which

PrOm〈、tc叩叩t()Sls・

  Thc balallce(ハfime1’actions bctwcenpm-alld anti-apopt(ハsis

Inclllbcrs of thc Bcl-2 gcllc falnily arc bclievcd to regulate

apoptosis(20). Bax is olle of thc genes in this family. A

prcdominance of Bax has been fbund t〈)accelcratc apoptosis

ill rcsponsc to cytokine removal(20)and thc cxprcssiml ofthc

Bax transgene {n T cel】s leads to a largc reduction in thc

nuInbcrs of nコature T celis∫η レ∫vo (21). In contrast, overex-

pressi(ハn of Bcl-2 strollgly rcduccs apoptosis(22).Thc rati(ハof

Bcl-2 to Bax detcrmillcs whcther a cell will respond to apo-

ptotic stimllli. Ahllough thc cmbryonic dcvclopment of Bcl-

2-/-micc is m)rmal, thcsc micc display growth rctardation,

shortencd lifespan alld masslvc apoptosis in thc spleell and

lhymlls (23, 24). Thcsc findings are similar t(、 th(、sc f〈)r

SSI一玉一/-micc. Ncvcrthelcss,1hc prcscnt study〔lid Ilot show

any signillcam diffcrcncc in thc lcvel ofBcl-2 pro匡cill bctwec茎1

Page 23: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

Immしmology:Nakaぴ‘」/. P}づOC. N‘ぱL Acα({. SC三. USA 95 (1~}98) 15581

A

B

C

 FIG.4. (A)Numemus tipoptdic splenocytcs and thyllmcytes in lhe SSI-}一/-mlce lOd三1ys old are demonstrated by TUNEL slailling,(β)

Electron micrograph onhe thymuM、fSSI-▲一/-mice a日Odays old(×5500). Cells(/1)are slill relatively normaL buれhc ap〔〕p1〈)hc cells(8)have

Iost cell menlbrane Inicrov三lli and cle三{rly exhlbit chroIllati∫1 condensatioll a1コd apoptic b‘)dies.(C)Immtmohistoclユemical sla{n{ng of the sPleen and

thymus. Colnparisolls of Bax expl’essioll in SSI-1+/+micc(π}ρLer})alld lhat irl-/-mice(7㌦, Rなノ’f). and of Bcl-2 cxpressioll in+/+micc

(βθ〃ρ・η五旬a蝋{th:u in-/にmiceμ〕θ〃θ’〃R’91の.

SS目一一/-and SSFI-1-/・←11・icc(Flg.4().’-rllen, it h(ヒ・bCCiコ

reCCIIUy dcmOnstl4ated that a dcath sigrlal gcncratじs the acti-

vation of Bax, which in tum h“luccs h(、modimcl元zati(m、 tl・ans一

location into nlitochOlldrial nコClnbranc、 Ini{ochOndria[ tlvs一                                                   ンfunction, and aPoptosis (25). SSI-1-/- n〕icc showcd an

incrcascd lcvcl olg Bax protein ill thc sp{cell and thymus(Fig.

Page 24: Osaka University Knowledge Archive : OUKA...(STAT)pathway is one of the most important signal path- ways activated immediately after cytokine stimulation (for a review, see

15582 1mmunology:Naka¢ぽL Pmc、 Nαti、〆1cαd. Sci、 US/195ぱ998)

4C). Thus, the apop⑩sis of lymphocytes lacking SSI-1may be

due to this increase in Bax express輌on. These 6ndings suggest

that, although the relation between Bax expression and the

JAK・STAT signal pathway has not been clarified yet, SSI-1

may directly or indirectly inhibit the other signal pathways,

except JAK・STAT signal cascades, which induce Bax expres-

sion and thus prevent Bax-induced apoptosis匡ηv匡レo. However,

further analysis姪required to determine which signai pathway

in the expression of Bax is inhibited by SSIヨ. and whether SSI4

inhibits apoptosis through inhibition of JAK activation.

  Whereas SSI-1 is strong量y expressed in thymus, spleen, and

peripheral blood leukocytes, any expression of SSI-20r-3 is

hardly observed in them(3,9)、 This difference in distribution

suggests that other members of the SSI-f㎜ily may not able to

compensate fbr the lack of SSI-1 inち1mph㏄ytes. Although our

analysis mainly f6cused on也e immune system of SSI-1-/-

mice in this s斑dy, SSI-1-/-mice at 10 days of age surprisingly

showed apparent severe cardiac hypertrophy and histological

degeneration of the hepat㏄ytes(our unpublished data). The

physiological roles of SSI-1 in other tissues have not been

darified fbr the present, Future an泣ysis will be necessary to

dari旬them or to determine whether SSI-1 inhib並s Bax

expression similar取血tissues other than lymphocytes.

 We wold llke to tbank T. Tanaka, S. Nagata, and S. Ak丘a Ibr their

suggestions, K. Satob and F. Ish輌tobi f6r the輌r techn輌cal assistance, and

N.Kameoka and M. Tagami fbr their secretarial ass三stance. Th輌s study

was supported by a Grant-in-Aid from the Ministry of Education,

Science and Culture, Japan.

1

23

           ロ

4よ」6「ノ

タ“ヱ㏄m.

α5りσRα訂翫&Mエ鴎㎏ちβEe品

嵩孟芦畿N

羅竃竃認

蹴 緒゜

 議

         コ    

 

く    しリリ

0

1も .k,

」㎞μ晋㏄コ玩.

8

9

10。

-内∠11

13.

45111、6.

重.7.

18.

19.

20.

21.

22.

23.

24.

25.

Endo, T. A., Masuhara, M.,Yokouchi, M, Suzukj, R.,Sakamoto,

H.,Mitsui, KりMatsumoto, A., Tanimura, S., Ohtsubo, M.,

Misawa, H.,αα1.(1997)1Wω解でLoη40∋387,921-924.

Naka, T., Narazaki, M., Hirata, M., Matsumoto, T., M血amoto,

S.,Aono, A., Nishimoto, N., K司ita, T., Taga, T., Yoshizaki, K.,

6’αL (1997)ハωωハ¢己oη吻〃ノ387,924-929。

Masuhara, M., Sakamoto, H., Matsumoto, A., SuzukちR., Ya・

sukaw鶴H., Mitsui, K., Wak輌oka, T., Tanimura, S., Sasaki, A.,

Misawa, H.,ε’α1.(1997)β↓oc舵〃宣. Bψρ触. Rぴ.α〃瑚瑚.239,

439-446.

Aman, M.」.&Leonard, WJ.(1997)仇死8foL 7,784-788.Hihon、 D. J., Richardson, R. T., Alexander, W. S., Viney, E M,

Willson, T. A., Sprigg, N. S., Starr, R., Nicholson, S. E, Metcaf,

F.D.&Nicola, N. A.(1998)Pnウc.1W”..4孤4.ぷcj.ひ&495,

114-119.

Adams, T. E, Hansen,」、 A, Starr, R,, Mcola, N、 A, H三1t叫D.∫、

&B∬1estrap, N.(1998)λβjoL(涛θη。273,1285-1287.

MosmanらT.(1983)」㎝励η01.ル食醜αカ65,55-58.Gawieli, Y., Sherman, Y.&Ben・Sasson, S. A(1992)ヱα〃βjo1,

119,493-501.

Sato, K., Wag頭, S., Ohsawa,Y., Nitatori, T., Kon, S., Kominami,

E,Watanabe, T., Gotow, T.&Uch取ama, Y.(1997)、4κ九酷ω∫.

砂o’.60,275-287.

Fukada, T., H{bi, M., Takahashi・Tezuka, M.,珂itani, Y.,

Ya㎜guchi, T., Nak司ima, K&. Hirano, T.(1996)加η7蹴輌砂5,

449-460.Que皿e, F. W、, Wang,」., Feng, J., Wa㎎, D.,αeveland,」. L, Ihle,

」.N.&Zanbetti, G. P.(1998)GeηεぷD¢v.12,1099-1107.

Kinoshita, T., Yokota, T., Arai, K&Miyajima, A.(1995)

0〃α聯形10,2207-2212.01tivai, Z. N., M孟11iman, C. L&Korsmeyer, S. J.(1993)α〃74

609-619.

Bardy, HJ., Gomez, G.. G., Kirberg,」. K.&Bems, AJ. M.

(1996)E協0ヱ 15,6991-7001.Katsum加a, M., Siegel, R. M., Louie, D. C., Miyashita, T.,

Tsulimoto, Y., Nowe皿, D. C., Greene, M. L&Reed,」. C.(1992)

P脱.ノVO’」ニノ1cαば.5ヒ」. OM 89,11376-11380.

Veis, D。」., Sorenson, C. M., Shutter,」. R&Korsmeyer, S.」.

(1.993)α〃75,229-240.

Nak{W㎜a, K, Nakayama, K, Neg輌shi,1., Kuida, K, Shinkai, Y.,

Louie, M. C., Fields, L. E, Lucas, P. J., Stewan, V、, Alt, F、 W.,

ααL (1993)5ヒ匡εnαグ261,1584-1588.

G⑩ss, A,」㏄kel, Wei, M. C.&Korsm学r, S.」.,(1998)」EMθO

J 17,3878-3885。