A�

rchivum Immunologiae et Therapiae Experimentalis, 2000, 4�8,� 399–403

P�

L ISSN 0004-069X

Review

Involvement of Apoptotic Protease Cascade for TissueDestruction in Sjögren’s SyndromeY. Hayashi et al.: Apoptotic Protease Cascade in Sjögren’s Syndrome

YOSHIO HA�

YASHI* , KUMIKO YA�

YAGI and NO�

RIO HANEJI

Department of Pathology, Tokushima University School of Dentistry, 3 Kuramotocho, Tokushima 770, Japan

Abstract. Sjögren syndrome (SS) is an autoimmune disease characterized by diffuse lymphoid cell infilt�rates in

t�he salivary and lacrimal glands, resulting in symptoms of dry mouth and eyes due to insuffic� ient secretion.

A�

lthough it has been assumed that a combination of immunologic, genetic and environmental� factors may play

a key role in the development of autoimmune lesions in the salivary and lacrimal glands, lit�tle is known about

t�he disease pathogenesis of SS in humans. We have identified the 120 kDa α-fodrin as an important autoantigen

in the development of SS in both an animal model and SS patients, but the mechanism of α-fodrin cleavageleading to tissue destruction in SS remains unclear. Tissue-infiltrating CD4+ T cells purified from the salivaryg� lands of a mouse model for SS bear a large proportion of Fas ligand and the salivary gland duct cells possessa poptotic receptor Fas. Anti-Fas antibody-induced apoptotic salivary gland cells result in sp� ecific α-fodrin cleav-a ge to the 120 kDa fragment in vitro. Preincubation with a combination of calpain and caspase inhibitor peptidesc� ould be responsible for inhibition of the 120 kDa α-fodrin cleavage. Thus, an increase in apoptotic proteasea ctivities including calpain and caspases may be involved in the progression of α-fodrin proteolysis and tissued

estruction in the development of SS.

K�

ey words: Sjögren’s syndrome; autoantigen; protease; apoptosis.

I�ntroduction

Sjögren’s syndrome (SS) is a chronic autoimmuned

isorder affecting the salivary and lacrimal glands and

leads to the clinical symptoms of dryness of the moutha nd eyes (sicca syndrome)2. The histopathologicalc� hanges in the minor salivary gland biopsy are charac-t

�erized by focal and/or diffuse lymphoid cell infiltrates

a nd parenchymal destruction. The majority of lymphoidc� ells in the salivary biopsy are CD4+ T cells, witha small proportion of CD8+ T cells11. These T cells ex-p� ress the αβ antigen receptor and cell surface antigensa ssociated with mature memory T cells. Since a pref-

e� rential use of specific variable region segments of thea ntigen receptor β chain by salivary gland T cells wase� vident3

�2, it has been assumed that an unknown organ-

-specific autoantigen targeted by autoreactive T cellsmay be present in the salivary glands.

W�

e have established an animal model for primarySS in NFS/s� ld mutant mice thymectomized 3 day afterb

�irth (3d-Tx) 14, 16. When the repertoire of T cell recep-

t�or (TCR) Vβ g� enes transcribed and expressed within

t�he inflammatory infiltrates was analyzed in the animal

model, a preferential utilization of TCR Vβ gene wasd

etected in these lesions from the onset of disease14. We

have recently identified a 120 kDa organ-specific auto-

* Correspondence to: Prof. Yoshio Hayashi, Department of Pathology, Tokushima University School of Dentistry, 3 Kuramotocho,T�

okushima 770, Japan, tel. /fax: +81 886 337 327, e-mail: hayashi@dent.tokushima-u.ac.jp

a ntigen from the salivary gland tissues of this animalmodel. The sequence of the first 20 NH2

� -terminalresidues was found to be identical to that of cytoskeletalp� rotein human α-fodrin15. Furthermore, sera from pa-t

�ients with SS reacted positively with purified 120 kDa

a ntigen, and proliferative response of peripheral bloodl

�ymphocytes (PBMC) from SS patients to the purified

a utoantigen was detected, but not from systematic lupuse� rythematosis (SLE) or rheumatoid arthitris (RA) pa-t

�ients and healthy controls. These results indicate that

t�he anti-120 kDa α-fodrin immune response plays an

e� ssential role in the development of primary SS.A

�utoimmune diseases are characterized by tissue

d estruction and functional decline due to autoreactive

T�

cells that escape self-tolerance10, 12. Although thespecificity of the cytotoxic T lymphocyte (CTL) func-t

�ion has been an important issue of organ-specific auto-

immune response, the mechanisms responsible fort

�issue destruction in SS remain to be elucidated6

�, 21. It

is now clear that the interaction of Fas with Fas ligand(FasL) regulates a large number of the pathophysiologi-c� al proceses of apoptosis3

�, 19. We speculate that an in-

c� rease in the enzymatic activity of apoptotic proteasesis involved in the progression of α-fodrin proteolysisd

uring the development of SS.

I�nvolvement of Fas and FasL in Tissue

Destruction

T�

o determine the possible involvement of Fas andFasL in the tissue destruction of SS, we first analyzedF

�as expression in the salivary gland specimens of the

3d-Tx NFS/s� ld mouse model14 and in the mouse sali-v� ary gland cells (MSG) isolated from non-thymec-t

�omized (non-Tx) NFS/s� ld mice. Immunohistochem-

i�stry revealed that epithelial duct cells express a largea mount of Fas on their cell surface in 3d-Tx NFS/s� ldm� ice, also detected by in situ TUNEL assay. However,w� e found that epithelial duct cells in non-Tx NFS/s� lda nd normal salivary glands are constitutively positivefor Fas, but not for TUNEL staining, suggesting thatF

�as expression does not correlate with its apoptotic

function in normal salivary gland cells. Isolated MSGc� ells express Fas with high proportion (>66%) onFACS staining (Fig. 1). We found that the tissue-infil-t

�rating CD4+ T cells purified from the salivary gland

t�issues of 3d-Tx NFS/s� ld mice bear a large proportion

o� f FasL (>73%) (Fig. 1). Immunohistochemical ana-lysis revealed that FasL expression was completely ab-sent on the salivary gland epithelial cells in any straino� f mouse. RT-PCR analysis demonstrated that Fas

m� RNA was present in both salivary glands, and a highlevel of FasL mRNA was detected in the salivaryg� lands of 3d-Tx NFS/s� ld mice, but not in non-Tx andnormal mice. Protein immunoblot analysis confirmedt

�he expression of FasL in salivary glands of 3d-Tx

N�

FS/s� ld mice.W

�e next investigated whether activated CD4+

T cells are responsible for tissue destruction as judgedb

�y in vitro 5

1C

!r release cytotoxic assay against MSG

c� ells. MSG cells (2 × 106�)

" in 7.5 ml of MEM were

labeled overnight at 37˚C in 5% CO!

2 with 300 µC!

i ofsodium [5

1C

!r]-chromate. CD4+ a nd CD8+ T cells puri-

f#ied from splenocytes using magnetic beads were incu-

b�ated with ConA and recombinant human IL-2. The

splenic CD4+ T cells activated with ConA and IL-2showed significant 5

1C

!r release against MSG cells (Fig. 2).

These cytotoxic activities were almost entirely inhibitedb

�y incubation with anti-murine neutralizing FasL anti-

b�ody, indicating that the cytotoxicity by activated CD4+

T cells towards salivary gland epithelial cells was Fas--based.

P$

ar ticipation of Calpain and Caspases i

%n α-Fodr in Cleavage

T�

o confirm the organ-specificity of a cleavage pro-d

uct of α-fodrin, we investigated various strains of

mice with salivary gland destruction, such as MRL/l&pr

n' on-obese diabetic (NOD) mice, in addition to 3d-TXN

�FS/s� ld mice. Protein immunoblot analysis demon-

strated that the 120 kDa α-fodrin was detected in thesea ffected glands, but not in normal mice. We examinedt

�he in vitro cleavage of α-fodrin using 240 kDa α-fo-

d rin in MSG cells. Anti-Fas antibody (Ab)-induced

a poptosis was confirmed by FACS analysis using the

Fig. 1. Flow cytometric analysis of Fas expression on mouse sali-vary gland (MSG) cells and FasL expression on the tissue-infiltrat-ing lymphocytes purified from affected salivary glands of animalmodel for Sjögren’s syndrome gated on CD4

Fluorescsence intensity

400 Y. Hayashi et al.: Apoptotic Protease Cascade in Sjögren’s Syndrome

in situ TUNEL procedure and by DNA laddering andf

#ormation. We could detect the 120 kDa α-fodrin in

a poptotic MSG cells on immunoblotting. Similar evi-d

ence was obtained with a human salivary gland (HSG)

e� pithelial cell line3�1. Apoptotic HSG cells induced by

t�reatment with anti-Fas Ab was confirmed. Immunoblot

a nalysis demonstrated that the 240 kDa α-fodrin on

a poptotic HSG cells was cleaved to smaller fragments,including 150 kDa and 120 kDa (Fig. 3).

W�

e next investigated whether cysteine proteases arei

�nvolved in α-fodrin cleavage on apoptotic HSG cells.

Anti-Fas Ab treated HSG cells were positive for mAbt

�o µ-calpain, and to ICE (p20) and CPP32 in associ-

a tion with apoptosis. The ICE- and CPP32-like acti-v� ities in anti-Fas Ab treated HSG cell extracts weret

�hen determined using fluorescent substrates8. The spe-

c� ific inhibitor for ICE (Ac-YVAD-cmk) or CPP32(Z-DEVD-fmk) was added to the reaction mixture ata concentration of 1 µM. Specific ICE- and CPP32-likea ctivities were determined by subtracting the values ob-t

�ained in the presence of inhibitors. These results sug-

g� est that protease families including calpain and cas-p� ases participate in the progression of α-fodrinc� leavage of apoptotic HSG cells.

Preventive Effect of Caspase Inhibitor i(n Vivo

T�

o address this hypothesis, we examined whetherα-fodrin cleavage to the 120 kDa fragment on apoptoticHSG cells could be blocked by preincubation with spe-c� ific protease inhibitors. In apoptotic HSG cells, calpaininhibitor peptide and caspase inhibitor (Z-VAD-fmk)h

)ad partially blocked 120 kDa α-fodrin formation.

M*

oreover, a combination of calpain inhibitor peptidea nd caspase inhibitors (Z-VAD-fmk and Z-DEVD-fmk)a lmost entirely inhibited the formation of 120 kDaα-fodrin. Protease inhibitor cocktails, other cysteinep� rotease inhibitors (E-64) and serine protease inhibitor(Leupeptin) had no effect on 120 kDa α-fodrin cleav-a ge in apoptotic HSG cells. The specificity of a com-b

�ination of calpain and caspase inhibitors towards the

d ecrease in 120 kDa α-fodrin in the HSG cells will be

f#urther quantified in detail.

By immunohistochemistry using polyclonal Aba gainst synthetic 120 kDa α-fodrin, a cleavage producto� f α-fodrin was present exclusively in epithelial ductc� ells of the labial salivary gland biopsies from SS pa-t

�ients, but not in control individuals. Protein immuno-

b�lot analysis confirmed the same results. This indicates

t�hat a cleavage product of 120 kDa α-fodrin is present

i�n the diseased glands with human SS, but not in control

g� lands. We next investigated whether the intravenousinjection of caspase inhibitor, Z-VAD-fmk, protectsa nimals against the development of autoimmune le-sions. The treatment with intravenous injection ofZ-VAD-fmk prevented the development of autoimmunec� onditions including lymphocytic infiltration and auto-a ntibody production to the whole 120 kDa molecule.

Fig. 2. C+

ytotoxic activity of activated CD4+ T cells from spleent,owards Fas-sensitive MSG cells. Splenic CD4+ T cells activated

w- ith ConA and IL-2 identified significant 5

1Cr release against MSGc. ells. This activity was almost entirely inhibited by anti-murineneutralizing FasL antibody

Fig. 3. Protein immunoblot analysis of 120 kDa α/ -fodrin in apop-t,otic human salivary gland (HSG) cells treated with anti-Fas mAb.

120 kDa α/ -fodrin was detected in apoptotic HSG cells

Y. Hayashi et al.: Apoptotic Protease Cascade in Sjögren’s Syndrome 401

P$

erspective

Although cleavage of certain autoantigens duringa poptosis may reveal immunocryptic epitopes thatc� ould potentially induce autoimmune responses in sys-t

�emic autoimmune diseases4, 34, no evidence whether in

v0 ivo cleavage of autoantigen occurs in organ-specifica utoimmune diseases has yet been described. The roleso� f the Fas/FasL system in the pathogenesis of autoim-m� une diseases have already been proposed1, 25, 26, 28, 30, 33.It has been recently reported that both Fas and FasL arep� resent in thyrocytes and their concomitant expressiono� n thyrocytes, independent of infiltrating T cells, is re-sponsible for thyrocyte destruction in Hashimoto’s thy-r1 oiditis13. In contrast, expression of Fas by pancreaticβ cells has been shown to have a major influence ont

�he susceptibility of tissue destruction in NOD mice to

d iabetes7

2. In this study, we provide evidence suggesting

t�hat Fas/FasL-mediated apoptosis may be involved in

t�he initial cascade of the in vivo cleavage of autoantigen

a nd that CD4+ cytotoxic T cells may participate in thet

�issue destruction of SS salivary glands.

W�

hen human T cell leukemia CEM cells were in-d

uced to undergo apoptosis, the 240 kDa α-fodrin was

c� leaved to a single detectable fragment of 120 kDa2�

7. Itw� as plausible that the 120 kD fragment is a break-downp� roduct of the 150 kDa α-fodrin cleavage22. The ana-lysis of proteolytic events associated with apoptosis ist

�o define the mechanisms leading to protease activation

a nd to identify key substrates whose cleavage might bel

�inked to the profound changes in cellular architecture.

There is increasing evidence that calpain is overacti-v� ated in autoimmune conditions and subsequent tissued

estruction2

�3, 24. Moreover, the cascade of caspases is

a critical component of the cell death pathway17, 18, 20, 29

a nd a few proteins have been found to be cleaved dur-i

�ng apoptosis. These include poly (ADP-ribose)

p� olymerase (PARP), a small U1 nuclear ribonucleopro-t

�ein (RNP), and α-fodrin, which were subsequently

identified as substrates for caspases5 . We provided evi-

d ence that α-fodrin is not only cleaved by calpain, but

a lso by one or more members of the caspases duringF

�as/FasL mediated-apoptosis in SS salivary glands. Fo-

d rin cleavage by calpain and caspases can potentially

l�ead to cytoskeltal derangement. Above all, it is of great

interest to point out that α-fodrin binds to ankylin,w� hich contains a cell death domain9

3.

In conclusion, these results are strongly suggestiveo� f a role for calpain- and caspase-mediated cleavage ofα-fodrin in the functional activation of autoreactiveT

� cells in SS, and could be consistent with a role for

Fas/FasL-mediated apoptosis in the development of

a utoimmune tissue destruction. Moreover, in vivop� reventive effects against autoimmune lesions treatedw� ith protease inhibitors have important implications fort

�esting useful therapies.

References

1. BIEGANOWSKA K. D., AU4

SUBEL L. J., MO5

DABBER Y., SLOVIK

E6

., ME7

SSERSMITH W. and HA8

FLER D. A. (1997): Direct e9 x vivoa: nalysis of activated, Fas-sensitive autoreactive T cells inhuman autoimmune disease. J. Exp. Med., 185,� 1585–1594.

2. BL;

OCH K. J., BU4

CHANAN W. W., WO5

HL M. J. and BU4

NIM J.(<1965): Sjögen’s syndrome. A clinical, pathological and sero-

logical study of sixty-two cases. Medicine, 44, 187–231. 3. BRUNNER T., MO

5GLL R. J., LA FACE D., YO

5O N. J., MAHBOUBL

A�

., EC=

HEVERRI F., MA8

RTIN S. J., FO5

RCE W. R., LY>

NCH D. H.,W?

ARE C. F. and GREEN D. R. (1995): Cell-autonomous Fas(<CD95)/Fas-ligand interaction mediates activation-induced

a: poptosis in T cell hybridomas. Nature, 373,� 441–444. 4. CA

8SIANO C. A. (1996): Selective cleavage of nuclear autoanti-

g@ ens during CD95 (Fas/APO-1)-mediated T cell apoptosis.JA. Exp. Med., 184,� 765–770.

5. CASIANO C. A., MARTIN S. J., GREEN D. R. and TAN E. T.(<1996): Selective cleavage of nuclear autoantigens during CD95

(<Fas/APO-1)-mediated T cell apoptosis. J. Exp. Med., 184,

7B65–770.

6. CHAN E. K., HAMEL J. C., BU4

YON J. P. and TAN E. T. (1991):Molecular definition and sequence motifs of the 52-kD compo-nC ent of human SS-A/Ro autoantigen. J. Clin. Invest., 8

D7,� 68–76.

7. CHERVONSKY A. V., WANG Y., WO5

NG F. S., VISINTIN I., FLA-

VE

ELL R. A., JA8

NEWAY C. A. JF R.G and MA8

TIS L. A. (1997): Therole of Fas in autoimmune diabetes. Cell, 89,� 17–24.

8. ENH

ARI M., TA8

LANIAN R. V., WO5

NG W. W. and NA8

GATA S.(<1996): Sequential activation of ICE-loke and CPP32-like pro-

t,eases during Fas-mediated apoptosis. Nature, 3

I80, 723–726.

9. FEINSTEIN E., KIMCHI A., WALLACH D., BO5

LDIN M. and VAR-

FOLOMEEV E. (1995): The death domain: a module shared bypJ roteins with diverse cellular function. Trends Biochem. Sci.,2K0,� 342–344.

10. FELDMANN M., BENNAN F. M. and MAINI R. N. (1996): Rheu-matoid arthritis. Cell, 8

D5,� 307–310.

11. FO5

X R. I. and SA8

ITO I. (1995): Sjögren’s syndrome. In RO5

BERT R.(<ed.): Clinical immunology-principles and practice. Chapter 75.

ML

osby, 1145–1153.12. GIANANI R. and SATVENTNICK N. (1989): Virus, cytokine, anti-

g@ ens, and autoimmunity. Proc. Natl. Acad. Sci. USA, 93, 2252–2259.

13. GIMORDANO C., ST

NASSI G., DE

7 MA8

RIA R., TO5

DARO M., RIMCHIUSA

P., PAPOFF G., RU4

BERTI G., BAGNASCO M., TESTI R. and GAL-

LUZZO A. (1997): Potential involvement of Fas and its ligand int,he pathogenesis of Hashimoto’s thyroiditis. Science, 275, 960–963.

14. HA8

NEJI N., HA8

MONO H., YA8

NAGI K. and HA8

YASHI Y. (1994):A new animal model for primary Sjögren’s syndrome inNO

FS/sld mutant mice. J. Immunol., 153,� 2769–2777.15. HA

8NEJI N., NA

8KAMURA T., TA

8KIO K., YA

8NAGI K., HI

MGA-

SP

HIYAMA H., SAITO I., NO5

JI S., SU4

GINO H. and HAYASHI Y.(<1997): Identification of α/ -fodrin as a candidate autoantigen in

pJ rimary Sjögren’s syndrome. Science, 276,� 604–607.16. HA

8YASHI Y., KO

5JIMA A., HA

8TA M. and HI

MROKAWA K. (1988):

402 Y. Hayashi et al.: Apoptotic Protease Cascade in Sjögren’s Syndrome

A new mutationinvolving the sublingual gland in NFS/N mice.Am. J. Pathol., 132, 187–191.

17. HO5

LTZMAN D. M. and DE7

SHMUKH M. (1997): Caspases:A treatment target for neurodegenerative disease? Nat. Med.,3,� 954–955.

18. HU4

ANG S., JIMANG Y., LI

M Z., NIMSHIDA E., MA

8THIAS P., LI

MN S.,

ULEVITCH R. J., NEMEROW G. R. and HAN J. (1997): Apoptosissignaling pathway in T cells is composed of ICE/Ced-3 familiyproteases and MAP kinase 6b. Immunity, 6,� 739–749.

19. JU4 S.-T., PA

8NKA D. J., CU

4I H., ET

NTINGER R., EL

; -KHQ

ATIB M.,SHERR D. H., STANGER B. Z. and MARSHAK-RO

5THSTEIN A.

(1995): Fas(CD95) /FasL interactions required for programmedcell death after T-cell activation. Nature, 3

I73, 444–448.

2R0. KO

5THAKOTA S., AZ

SUMA T., RE

7INHARD C., KL

;IPPEL A., TA

8NG J.,

CHU K., MC= G

TARRY T. J., KIRSCHNER M. W., KO

5THS K., KW

UIAT-

KV

OWSKI D. J. and WIMLLIAMS L. T. (1997): Caspase-3-generated

fragment of Gelsolin: Effector of morphological change inapoptosis. Science, 278,� 294–298.

2R1. KR

WUIZE A. A., SM

XEENK R. J. T. and KA

8TER L. (1995): Diagnos-

tic criteria and immunopathogenesis of Sjögren’s syndrome:implications for therapy. Immunol. Today, 16,� 557–559.

22. MARTIN S. D., O’BRIEN G. A., NISHIOKA W. K., MC= G

TAHON A. J.,

MA8

HBOUBI A., SA8

IDO T. C. and GRW

EEN D. R. (1995): Prote-olysis of fodrin (non-erythroid spectrin) during apoptosis.J. Biol. Chem., 2

K70, 6425–6428.

23. MENAD H.-A. and EL-AMINE M. (1996): The calpain-calpas-tatin system in rheumatoid arthritis. Immunol. Today, 17, 545–547.

2R4. M I

MMORI T., SU

4GANUMA K., TA

8NAMI Y., NO

5JIMA T., MA

8TSUMU-

RA M., FU4

JII T., YO5

SHIZAWA T., SU4

ZUKI K. and AKIZUKI M.(1995): Autoantibodies to calpastatin (an endogenous inhibitorfor calcium-dependent neutral protease, calpain) in systemicrheumatic diseases. Proc. Natl. Acad. Sci. USA, 92,� 7267–7271.

25. NAGATA S. and GO5

LDSTEIN P. (1995): The Fas death factor.Science, 267, 1449–1456.

2R6. NI

MSHIO A., KA

8TAKAI T., OS

PHIMA C., KA

8SAKURA S., SA

8KAI M.,

YY

O5

NEHARA S., SU4

DA T., NA8

GATA S. and MA8

SUDA T. (1996):A possible involvement of Fas-Fas ligand signaling in the pa-t,hogenesis of murine autoimmune gastritis. Gastroenterology,

111, 956–967.27. PATEL T., GO

5RES G. J. and KAUFMANN S. H. (1996): The role

oZ f proteases during apoptosis. FASEB J., 10,� 587–597.28. RIEUX-LAUCAT F. (1995): Mutation in Fas associated with

human lymphoproliferative syndrome and autoimmunity.S[

cience, 268, 1347–1349.29. RU

4DEL T. and BO

5KOCH G. M. (1997): Membrane and morpho-

logical changes in apoptotic cells regulated by caspase-medi-a: ted activation of PAK2. Science, 276, 1571–1574.

30. SAITO I., HARUTA K., SHIMUTA M., INH OUE H., SAKURAI H.,YY

A8

MADA K., ISP HIMARU N., HIMGASHIYAMA H., SU

4MIDA T., ISP HI-

DA H., SU4

DA T., NO5

DA T., HAYASHI Y. and TSP

HUBOTA K.(<1999): Fas ligand-mediated excrinopathy resembling Sjögren’s

s\ yndrome in mice transgenic for IL-10. J. Immunol., 162,2488–2494.

31. SHQ

IRASUNA K., SA8

TO M. and MIMYAZAKI T. (1981): A neoplas-

t,ic epithelial duct cell line established from an irradiated human

s\ alivary gland. Cancer, 48,� 745–752.32. SU

4MIDA T., YO

5NAHA F., MAEDA T., TANABE E., KO

5IKE T.,

T�

O5

MIOKA H. and YO5

SHIDA S. (1992): T cell receptor repertoireoZ f infiltrating T cells in lips of Sjögren’s syndrome patients.JA. Clin. Ivest., 89,� 681–685.

33. THOMPSON C. B. (1995): Apoptosis in the pathogenesis andt,reatment of disease. Science, 267,� 1456–1462.

34. UTZ P. J., HO5

TTELET M., SC=

HUR P. H. and ANH

DERSON P.(<1997): Proteins phospholylated during stress-induced apop-

t,osis are common targets for autoantibody production in pa-

t,ients with systemic lupus erythematosus. J. Exp. Med., 185,8]43–854.

Received in December 1999Accepted in February 2000

Y. Hayashi et al.: Apoptotic Protease Cascade in Sjögren’s Syndrome 403