Enhanced and Accelerated Lymphoproliferation in Fas

Proc. Natl. Acad. Sci. USA Vol. 93, pp. 2131-2136, March 1996 Biochemistry

Enhanced and accelerated lymphoproliferation in Fas-null mice MASASHI ADACHI*, SACHIKO SUEMATSUt, TAKASHI SUDA*, DAISUKE WATANABE*, HIDEHIRO FUKUYAMA*, JUN OGASAWARA*, TAKASHI TANAKAt, NOBUAKI YOSHIDAt, AND SHIGEKAZU NAGATA* *Osaka Bioscience Institute, 6-2-4 Furuedai, Suita. Osaka 565. Japan, and 'Research Institute, Osaka Medical Center for Maternal and Child Health, Izumi, Osaka 590-0)2, Japan Communicated by Charles Weissmann, Universitat Ziurich, Zutriclh, Switzerland, November 22, 1995 (received for review Jutne 26, 1995)

ABSTRACT Fas is a 45-kDa membrane protein that retrovirus-into an intron of the Fas chromosomal gene transduces an apoptotic signal. The mouse lymphoprolifera- (25-28), which results in premature termination and aberrant tion (lpr) mutation is a leaky mutation of Fas. In this study, splicing of the Fas transcript. However, a small amount of we examined lymphocyte development in Fas-null mice gen- intact Fas mRNA or protein is expressed in the thymus of lpr erated by gene targeting. The Fas-/- mice progressively mice (26, 28, 29), indicating that the lpr mutation is leaky (25). accumulated abnormal T cells (Thyll, B220+, CD4-, and Therefore, it was thought that the absence in lpr mice of altered CD8-) and developed lymphadenopathy and splenomegaly, thymic cell development was related to the leakiness of the which were much more accelerated and pronounced than mutation (29). those in lpr mice. In addition, the Fas-null mice showed To overcome this problem, we created Fas-knockout mice lymphocytosis, accompanied by lymphocytic infiltration in and found that they developed hepatomegaly in addition to the lungs and liver. The number of apparently normal B cells lymphoproliferation (30). In this study, we examined the also increased, and large amounts of immunoglobulins, in- development of lymphocytes in the Fas-null mice. The devel- cluding anti-DNA antibodies, were produced. Thymic clonal opment of lymphadenopathy and splenomegaly in the Fas-null deletion, assessed by deletion of T cells reactive to mouse mice was much sooner than that in lpr mice. In addition, these endogenous superantigens, was apparently normal in the mice showed massive lymphocytosis, which was occasionally Fas-/- mice, whereas the peripheral clonal deletion ofmature accompanied by infiltration of lymphocytes into the lungs and T cells against a bacterial superantigen was impaired. These liver. Not only abnormal T cells (Thy-I+ B220+ CD4- CD8-) results suggested that Fas plays a decisive role in peripheral but also apparently normal B and T cells accumulated in clonal deletion but not in negative selection in the thymus. Fas-/- mice, and these mice produced a large excess of immunoglobulins. Despite the enhanced abnormality in lym- During lymphocyte development, many cells die by apoptosis phocyte development, negative selection assessed by the de- at various steps. Over 95% of the precursor T cells in the letion of T cells reactive to endogenous superantigens was thymus are eliminated (1) by negative selection or by the apparently normal in Fas-/- mice, indicating that Fas may not absence of positive selection. Mature T cells also die in the be involved in this process. periphery. T cells that recognize the self antigens expressed only in the peripheral tissues, or those activated by foreign MATERIALS AND METHODS antigens for a long period, die by apoptosis (2, 3). During the development of B cells, those that fail to correctly recombine Animals. Generation of mice carrying the Fas-null mutation immunoglobulin genes are eliminated, whereas those that has been described elsewhere (30). In brief, the targeting produce autoreactive antibodies are negatively selected in the vector was constructed by replacing exon 9 of the Fas gene with bone marrow as well as in peripheral lymphoid tissues (4, 5). the neo and 13-galactosidase genes. E14-1 embryonic stem cells The Fas ligand (FasL) is a type II membrane protein with a from 129/Ola mice were disrupted by homologous recombi- mass of 40 kDa, and it is mainly expressed in activated T cells nation, injected into blastocysts isolated from C57BL/6, and (6-9). FasL binds to its receptor, Fas (also called APO-1), transferred to foster mothers to produce male chimeric mice. which is a 45-kDa type I membrane protein expressed in These were bred with C57BL/6 female mice. Mice heterozy- activated T and B cells, as well as in the liver, heart, and lungs gous for the Fas mutation (FI) were then intercrossed to (10, 11). The binding of FasL or agonistic anti-Fas antibody to generate litters (F2), which carried the wild-type (Fas+/+), Fas induces apoptosis and kills cells within hours (6, 10, 12-14). heterozygous (Fas+/-), or homozygous (Fas-/-) Fas muta- Studies in vivo and in vitro have indicated that the Fas/FasL tion. The complete loss of functional Fas in the thymocytes and system is involved in the activation-induced suicide of T cells hepatocytes of Fas-/- mice has been demonstrated (30). (15-18). Flow Cytometry, Serological, and Histological Studies. Molecular and genetic analyses of Fas and FasL have Lymphocytes were prepared from spleens and lymph nodes as indicated that mouse lymphoproliferation (lpr) and general- described (31). Lymphocytes were isolated by dissecting the ized lymphoproliferative disease (gld) are mutations of Fas and liver into small pieces, filtering them through a steel mesh (100 FasL, respectively (13, 19). These mutant mice develop lymph- gauge), and then suspending the filtrate in RPMI 1640 medium adenopathy by accumulating abnormal T cells (52) and suffer as described (32). Flow cytometry proceeded essentially as from systemic lupus erythematosus-like autoimmune disease described (24). The monoclonal antibodies used for staining (20). Peripheral clonal deletion appears to be impaired in lpr were phycoerythrin (PE)-conjugated anti-CD4 (Becton Dick- mice (18, 21). Whereas negative selection in the thymus is inson, clone GK 1.5), fluorescein isothiocyanate (FITC)- essentially normal (21-23), functional Fas is expressed in conjugated anti-CD8 (PharMingen, clone 53-6.7), PE- CD4+ CD8 thymocytes that are susceptible to positive and conjugated anti-B220 (GIBCO/BRL, clone RA3-6B2), and negative selection (24). The lpr mutation is caused by inserting FITC-conjugated anti-Thyl.2 (PharMingen, clone 30-H112). an early transposable element-namely, a mouse endogenous The following biotin-conjugated monoclonal antibodies from

The publication costs of this article were defrayed in part by page charge Abbreviations: PE, phycoerythrin; FITC, fluorescein isothiocyanate; payment. This article must therefore be hereby marked "advertisenlent' in V, variable region; SEB, staphylococcal enterotoxin B; WBC, white accordance with 18 U.S.C. §1734 solely to indicate this fact. blood cell. 2131 Downloaded by guest on September 24, 2021 2132 Biochemistry: Adachi et al. Proc. Natl. Acad. Sci. USA 93 (1996) PharMingen were also used for staining in combination with RESULTS PE- or FITC-conjugated streptavidin (Becton Dickinson or GIBCO/BRL, respectively): anti-Vi35.1, -5.2 (clone MR9-4), Lymphadenopathy, Splenomegaly, and Lymphocytosis. anti-V38.1, -8.2 (clone MR5-2), and anti-Vpl1 (clone RR3-15) Like MRL-lpr mice, the lymph nodes and spleen of Fas-null antibodies (V, variable region). Two-color flow cytometry was mice became enlarged in an age-dependent manner. As shown in Fig. 1, the spleen and lymph nodes of Fas+/+ and Fas+/- performed with a FACScan (Becton Dickinson), and the data mice weighed about the same at any age. On the other hand, were analyzed with LYSIS II software. these organs of Fas-/- mice became progressively larger than The levels of serum IgM, IgGl, IgG2a, IgG2b, and IgG3 those in FasI/I mice. At 16 weeks of age, the spleen and lymph were estimated by single radial immunodiffusion with kits from nodes were about 15 and 100 times larger than those in Fas+/+ Serotec. Anti-single-stranded DNA and anti-double-stranded mice, respectively. The progression of splenomegaly and DNA antibodies were quantified as described (32) using a lymphadenopathy was more pronounced in Fas-null mice than Mesacup EIA system from MBL (Nagoya, Japan). Serum was in MRL-lpr mice (Fig. 1). Female and male Fas-null mice serially diluted with Tris-buffered saline TBS; 25 mM similarly developed lymphadenopathy and splenomegaly. The Tris HCl, pH 7.4/140 mM NaCl) and added to the microtiter total number of lymphoid cells recovered from Fas-/- mice plates. After incubation at room temperature for 30 min, was 2 x 109 from the spleen and an average 5 x 108 per lymph bound antibodies were detected with peroxidase-conjugated node. Flow cytometry indicated that 60-70% of the cells that rabbit anti-mouse IgG or IgM (Zymed) using o-phenylenedi- accumulated in the lymph nodes and spleen of the Fas-/- mice amine as a substrate. The absorbance at 492 nm was measured at age of 16 weeks were Thy-1i B220+ CD4- CD8- (Fig. 2). with an automated Micro-ELISA reader. For histological The cells expressed CD3 but not IgM (data not shown), analysis, the major organs from autopsied mice were weighed, indicating that the cells were originated from T cells. The fixed in 10% formalin, sectioned at 3 ,um, and then stained with proportions of normal T (B220- Thy-i+) in all peripheral hematoxylin and eosin. lymphoid organs and B cells (B220+ Thy-1 +) in the spleen and Clonal Deletion in the Thymus and Periphery. To study blood of Fas-/- mice were reduced. However, considering that thymic clonal deletion, F1 mice heterozygous at the Fas the total number of cells increased about 13- and 100-fold in mutation obtained by crossing 129 x C57BL/6, were back- the spleen and lymph nodes, respectively, these results crossed with MRL/Mp- + / + mice for two generations. The F3 indicated that not only abnormal CD4- CD8- T cells but heterozygous mice were intercrossed to obtain F4 homozygous also apparently normal T and B cells were abundant in mice termed F4-MRL. These mice were likely to be Mtv-9+ Fas-/- mice (Table 1). In contrast to the striking phenotype and I-E+. The expression of V,9 gene subtypes in the lymph in the peripheral lymphoid organs, the thymic development nodes was then determined by flow cytometry as described in Fas-/- mice was normal as described (30). above, using various monoclonal antibodies against Vp protein. The total number of peripheral white blood cells (WBCs) To determine clonal deletion in the periphery, the ability to was also greatly increased in Fas-/- mice (Fig. 1). The number delete Vp8+ CD4+ T cells after staphylococcal enterotoxin B of WBCs in young animals (4-8 weeks) was comparable (SEB) administration was examined as described (33). In brief, among Fas+/+, Fas+/-, and Fas-/- mice, while that of Fas-/- 100 ,g of SEB (Makor Chemical, Jerusalem) dissolved in 200 mice at age of 16 weeks was 20 times higher than that in ,ul of phosphate-buffered saline (PBS) was intravenously in- Fas/+/ and Fas'/- mice. The increased population of WBCs jected into 4- to 6-week-old mice. At the indicated times, the in Fas-/- mice consisted of lymphocytes (see Fig. 4A), and percentages of the V381CD41 T cells in the spleen were there were no significant increases of other WBCs such as determined by flow cytometry using a monoclonal antibody neutrophils and monocytes (data not shown). Flow cytometry specific to the Vp8 subtype. The specific deletion of Vp8+ indicated that the cells that accumulated in the blood had the CD4+ T cells by SEB administration was expressed as a phenotype, B220+ Thy-1+ CD4- CD8- (Fig. 2). MRL-lpr mice percentage compared with that of control mice that received also showed lymphocytosis in the blood, but this was much less PBS. pronounced compared with that in Fas-null mice (Fig. 1).

800 , A B C 40 -

2000 -

J) 600- E A

0o x E 400- .1- s~CL E 020 tm 1000 - :22A: 3) -00

4- 200- a) 02) I 0 - CD

u momAX 4 A 4 .,.nrt -. ul l l w . . . Ai ;- . Un a"Mtm-lG lhi - fi rl _ ' 4816 4 8 16 4 8 16 wt Ipr 4816 4 8 16 4 8 16 wt lpr 4 8 16 4 8 16 4 8 16 wtlpr

/ MRL +/+ +/- / MRL +/+ +/- -- MRL

FIG. 1. Lymphadenopathy, splenomegaly, and lymphocytosis in Fas-/- mice. The spleen (A) and the axillary and inguinal lymph nodes (B) were excised from Fas+/+ (+/+), Fas+/- (+/-), and Fas-/- (-/-) mice, and the average weights of the spleen and lymph nodes from five to eight mice are shown. The number of WBCs (C) were counted using a hematocytometer. The numbers at the bottom indicate the age of mice in weeks. As controls, MRL- + / + (wt) and MRL-lpr/lpr (lpr) mice at the age of 22 weeks were similarly analyzed, and the average values obtained from groups of five mice are indicated. SD values are indicated by bars. Downloaded by guest on September 24, 2021 Biochemistry: Adachi et al. Proc. Natl. Acad. Sci. USA 93 (1996) 2133

(A) Lymph Node (C) Liver

'r +/+ 7 / 732 32 49

.12 13 712. - 04

Thyl .2

-13 0 9 0 ~55 1 51 1 12 1

25 5 6 .20 24 .-326 .5 2-

CD (B) Spleen (D) Peripheral Mononuclear Cell

i623 . 642 2311 1773

C-4 - C

- ----T hy1.2 4-

250 16 0 120 510

9 ~ ~ 711 6 2- 35 14 90

o -- I - -CD8-iCD8 ------'.-.- CD

FIG. 2. Flow cytometry analysis of lymphocytes from the lymph node, spleen, liver, and peripheral blood of mice. Lymphocytes were prepared from the lymph nodes (A), spleen (B), liver (C), and peripheral blood (D) of Fas+/+, Fas+/-, and Fas-/- littermates 13-16 weeks old. The cells were stained with PE-conjugated anti-B220 and FITC-conjugated anti-Thyl.2 antibodies (Upper) or with PE-conjugated anti-CD4 and FITC-conjugated anti-CD8 antibodies (Lower). In each instance, 5000 viable cells were collected and analyzed on a FACScan (Becton Dickinson). Numbers indicate percentage of positively stained cells in each quadrant. Organs of at least three mice for each genotype were analyzed, and typical data are shown.

The Fas-/- mice produced large quantities of various there was a multiple peribronchovascular infiltration of lym- isotypes of immunoglobulin isotypes (Fig. 3). The serum level phocytes into the lungs, which was particularly marked around of IgGl, IgG2a, IgG2b, and IgG3 was about 30-100 times the hilar vessels, occasionally assuming follicular organization. higher in 16-week-old Fas-/- mice than in wild-type mice. The Focal consolidation was frequent in the tips of lobes (Fig. 4B), titer of the antibodies against single- or double-stranded DNA where some plasma cells were also present (Fig. 4C). As shown was also very high in the serum of Fas-/- mice. At the age of in Fig. 4 D and E, lymphocytes infiltrated the interstitial tissues 16 weeks, the concentrations of these autoantibodies were as well as the area surrounding the vessels in the liver. Flow 50-500 times higher in Fas-/- mice than those in wild-type cytometry of the infiltrated lymphocytes in the liver indicated mice. These phenotypes in Fas-/- mice were more severe than that they had the Thy-1+, B220+, CD4-, CD8- phenotype, those in lpr mice and were not observed in Fas'/- mice (Fig. which was identical to that found in peripheral blood (Fig. 2). 3). Among the lymphocytes in the lungs and liver, very few cells Infiltration of Lymphocytes into the Lungs and Liver. In showed the mitotic morphology, suggesting that these cells do addition to lymphocytosis in the periphery, histological exam- not proliferate in the liver or lungs. inations of Fas-/- mice revealed occasional lymphocytic in- Clonal Deletion of Autoreactive T Cells. The thymic clonal filtration in the lungs and liver (Fig. 4). At the age of 13 weeks, deletion of T cells was examined by analyzing the T-cell Table 1. Analysis of lymphocyte subsets in spleen and lymph nodes of Fas-/- mice Spleen Lymph node Cell number x 106 Cell number x105 Genotype Thy-1+ B220- Thy-i- B220+ Thy-1+ B220+ Thy-1+ B220- Thy-i- B220+ Thy-1+ B220+ +/+ 36.1 ± 13.7 80.0 ± 30.4 3.8 ± 1.5 15.8 ± 5.7 3.9 ± 1.4 0.3 ± 0.1 +/- 38.5 ± 8.6 94.7 ± 21.1 3.0 + 0.6 34.5 ± 17.0 4.9 ± 2.5 1.2 ± 0.4 -/- 79.2 ± 2.6 455.4 ± 14.9 1188 ± 38.8 398.4 ± 224 846 ± 476 3535 ± 1988 Phenotypes of Thy-1+ B220-, Thy-1- B220+, and Thy-1+ B220+ cells in spleen and lymph nodes of Fas+/+, Fas+/-, and Fas-/ in 16-week-old mice were determined by flow cytometry (see Fig. 2), and the total number of cells with each phenotype was calculated. Average values from five to eight mice are shown ± SD. Downloaded by guest on September 24, 2021 2134 Biochemistry: Adachi et al. Proc. Natl. Acad. Sci. USA 93 (1996)

B T 80 -

N cn O 60- FIG. 3. Production of immunoglobu- us lins and autoantibodies in Fas-/- mice. -L Sera were collected from 16-week-old -00 Fas+/+ (columns 1), Fas+/- (columns 2), -Q 40- 4J1 and Fas-/- (columns 3) mice, and 22- I week-old MRL-+/+ (columns 4) and 0 MRL-lpr/lpr (columns 5) mice. Concen- cu 20 - trations of various immunoglobulins (A) E and anti-double-stranded DNA autoan- tibody (anti-ds) or anti-single-stranded Uz DNA antibody (anti-ss) (B) were deter- U;n _rf--- 0 . 12345 12345 12345 12345 mined as described. Results are shown as IgM IgGI lgG2a IgG2b IgG3 anti-ss anti-ss anti-ds anti-ds means of the values obtained from five to (IgM) (IgG) (IgM) (IgG) eight mice; bars indicate SD. population that was reactive to endogenous viral superanti- in deleting the T cells reactive to endogenous virus superan- gens. The F2 mice obtained by intercrossing F1 Fas+/- mice tigens. (129/Ola x C57BL/6), which were I-E- Mls_lb Mtv-9+, The administration of SEB into mice causes a deletion of cannot delete T cells that carry Vp5+, V8+, or V611+ T-cell Vp8+ T cells in peripheral lymphoid organs, and this process receptors (34) (Table 2). The F2 mice were therefore back- is regarded as a model for peripheral clonal deletion (33, 35, crossed for two generations into the mouse MRL strain, which 36). To examine the ability of the Fas-/- mice to cause expresses both I-E and Mtv-9 and can therefore delete Vp5+ peripheral clonal deletion, 4- to 6-week-old Fas+/+ or Fas-/- and V 1 1 + T cells. As shown in Table 2, the peripheral T cells mice were injected with 100 ,tg of SEB, and the population of in the F4 mice carrying either wild-type Fas or the Fas-null Vp8+ T cells was followed. The population of these among mutation lacked both V35+ and V11+ T cells in the CD4+ or CD4+ T cells in the spleen was -18% in both Fas+/+ and CD8+ population. On the other hand, since the MRL mice do Fas-/- mice. Fifteen days after the SEB injection, >50% of not provide superantigens specific for V,38 (34), the F4 mice the Vp8+ T cells was deleted in FasI/+ mice, whereas they were could not delete Vp8+ T cells. These results indicated that the not decreased in the Fas-/- mice (Fig. 5). These results Fas-/- mice can delete the autoreactive T cells in the thymus indicated that the Fas-/- mice are defective in deleting and that Fas is not essential for thymic clonal deletion, at least activated T cells in the periphery.

A B C ...* ,$

4.t. I * 0;2 's:7

i-A 9 # ; t e e),..,, Al- o. i , t ~~~~~

FIG. 4. Histopathology of Fas-deficient mice. (A) Circulating lymphocytic cells in the peripheral blood of Fas-/- mice at the age of 13 weeks. Cells were stained with Wright-Giemsa stain. (X26.) (B and C) Histological analysis of lung section in Fas-/- mice after staining with hematoxylin and eosin. (B) Extensive lymphoid hyperplasia around the central and peripheral vessels of the lungs. (x3.) (C) High magnification of cells infiltrating the lungs. (x 13.) Arrow indicates clusters of plasma cells. (D and E) Histological analysis of liver section. (D) Massive lymphocytic infiltration in the interstitial tissues of liver. (x3.) (E) High magnification of infiltrating cells in liver. (x 13.) Downloaded by guest on September 24, 2021 Biochemistry: Adachi et al. Proc. Natl. Acad. Sci. USA 93 (1996) 2135 Table 2. Population of T cells expressing Vq subtypes CD4+ CD8+ Genotype Vp5+ Vp8+ VP11+ Vp5 + V38+ Vd11+ F2-+/+ 3.2 ± 1.1 18.2 ±0.8 5.1 ± 0.2 11.5 ± 3.0 17.9 ± 2.0 7.7 + 0.7 F4-MRL--/- 0.3 ± 0.1 19.4 ± 1.5 0.4 + 0.1 0.6 ± 0.5 19.3 ± 3.0 1.6 + 0.9 F4-MRL-+/+ 0.3 ± 0.1 17.4 ± 2.4 0.4 ± 0.1 0.8 ± 0.5 25.9 ± 1.3 2.0 + 0.9 The F4-MRL mouse strains carrying the wild-type allele for Fas (F4-MRL-+/+) or the homozygous mutant allele (F4-MRL--/-) were generated as described. Lymphocytes of the lymph nodes in mice 2-4 weeks old were analyzed by flow cytometry after simultaneous staining with antibodies specific for each subtype of Vq and anti-CD4 or anti-CD8 monoclonal antibodies. Analysis was done with five mice and average percentage of cells carrying each phenotype is shown ± SD. DISCUSSION Although we cannot rule out the possibility that the chimeric protein in Fas+/- mice and the mutated Fas in lprcg mice do We showed that the Fas-null mice developed lymphadenopa- not work as a dominant-negative mutant, these results suggest thy and splenomegaly, accompanied by overproduction of that the small amount of functional Fas expressed in heterozy- immunoglobulin isotypes. The development of these pheno- gotes is sufficient in the mouse but not in humans, or the types was much faster and more severe than that in the affected patients carry a mutation(s) in other genes that MRL-lpr mice that carry a leaky mutation in the Fas gene (13, accelerate the phenotype. 19, 25). The degree of lymphoproliferation in mice carrying the Lymphoproliferation in Fas-/- mice eventually caused lym- lpr mutation depends on the mouse strain, and the most severe phocytosis and lymphocytic infiltration in the lungs and liver. phenotype has been found in the MRL strain (37). The Fas-/- The accumulated lymphocytes in the peripheral blood and mice that we studied carry a chimeric genetic background from liver were atypical T cells with the phenotype Thy-1+, B220+, the 129 and C57BL/6 strains. Although we do not know how CD4-, and CD8-. The total number of lymphocytes that the background genes from the 129 strain affect the pheno- accumulated in a Fas-/- mouse was >5 x 109 cells (3 x 109 type, the severity of the phenotype in Fas-/- mice is probably in the lymph nodes, 2 x 109 in the spleen. and 5 x 108 in the due to the complete loss of functional Fas. In the lpr mutation, blood) at age of 16 weeks. The number of T cells that migrate the expression level of the intact Fas mRNA may vary among from the thymus to the periphery is 1-2 x 106 cells per day (44) mouse strains. It is also possible that MRL mice have an and thus 1-2 x 108 after 16 weeks, which is much less than the inefficient signal transduction system for Fas-mediated apo- number found in Fas-/- mice. If the Fas system is involved in ptosis or have other mechanisms to accelerate lymphoprolif- the apoptosis of only mature T cells in the periphery, and not eration. To clarify these issues, we are attempting to establish of thymocytes (21), those T cells accumulating in Fas-/- mice MRL and C57BL/6 mouse strains carrying the Fas-null mu- must have once been activated to proliferate. The expression tation by repeated backcrossing. of activation antigens including FasL in the accumulated Humans carrying mutations in the Fas gene have been lymphocytes (31, 45) is in agreement with this notion. The identified (38, 39). Most of the affected patients carry a abnormal T cells that accumulate in lpr mice are not dividing heterozygous mutation in the Fas cytoplasmic region. Since in vivo and cannot proliferate in vitro (46). It has been FasL is a trimer (40), and Fas should be oligomerized to suggested that these T cells proliferate in the liver (32). transduce the apoptotic signal (41) (T. Takahashi, T.S., and However, we found few mitotic cells in infiltrates of the liver. S.N., unpublished data), the mutated Fas in the patients Since these T cells are polyclonal (45), it seems that mature T seemed to function as a dominant-negative mutant to cause cells are activated to proliferate by interaction with self or autoimmune lymphoproliferative syndrome (39). On the other foreign antigens at various locations in the periphery. In hand, Fas'/- mice, which should produce a nonfunctional wild-type mice, this activation induces the expression of FasL chimeric protein consisting of extracellular Fas and cytoplas- in T cells, which binds Fas to cause cell death (activation- mic f3-galactosidase (30), and mice heterozygous for the lprcg induced death of T cells) (15-17). This process cannot occur mutation (42) that causes the nonfunctional point mutation in in Fas-/- mice, and the activated cells lose CD4 and CD8 the cytoplasmic region of Fas did not show any phenotype (43). expression and accumulate in the periphery. Giese and Davidson (47) have found an increase of Ig+ B 140 cells in lpr or gld mice, which seems to be responsible for the 120 production of autoantibodies (48). The number of B cells in the lymph nodes of the lpr mice at the age of 16 weeks was -10 100 times higher than that of the wild-type mice (47). Here, we found -200 times more B cells in the lymph nodes of Fas-/- + 80 co mice than in those of the wild type. These results indicated that >60 the Fas system is involved in development of not only T but also B cells. Like T cells, B cells die by apoptosis at various stages 240 of their development in the bone marrow and germinal centers 20, (4). Rathmell and Goodnow (49) have reported that autoreactive B cells in the bone marrow are efficiently eliminated in I4 lpr mice, and a flow cytometry analysis of the bone marrow 0 5 10 15 cells indicated that B-cell development in the bone marrow is Days after SEB essentially normal (M.A. and S.N., unpublished results). How- ever, like the T-cell system (21), Fas may be involved in FIG. 5. Peripheral clonal deletion in Fas-/- mice. Wild-type (0) or deletion of B cells in the periphery, including the germinal Fas-null (0) mice were intravenously injected with SEB (100 ,ug per center. Immunization of mice with lipopolysaccharide induced mouse). Lymphocytes from spleens were harvested at the indicated times after injection. Ratios of Vp8+ CD4+ cells in spleens of mice Fas expression in the germinal centers and rendered the cells injected with SEB are shown as percentages of those found in mice sensitive to Fas-mediated apoptosis (50,51). The interaction of injected with PBS. Each value is average of three mice; SD is indicated autoaggressive B cells with T cells may stimulate FasL expres- by bars. sion in T cells. FasL then binds Fas in activated B cells and kills Downloaded by guest on September 24, 2021 2136 Biochemistry: Adachi et al. Proc. Natl. Acad. Sci. USA 93 (1996) them by apoptosis. This process would be blocked in Fas-null 23. Herron, L. R., Eisenberg, R. A., Roper, E., Kakkanaiah, V. N., mice, and B-cells that should normally be deleted would Cohen, P. L. & Kotzin, B. L. (1993) J. Immunol. 151, 3450-3459. accumulate in the periphery, producing a large quantity of 24. Ogasawara, J., Suda, T. & Nagata, S. (1995) J. Exp. Med. 181, immunoglobulins. A closer examination of Fas-/- mice may 485-491. 25. Adachi, M., Watanabe-Fukunaga, R. & Nagata, S. (1993) Proc. reveal by which molecular mechanisms the Fas system is Natl. Acad. Sci. USA 90, 1756-1760. involved in lymphocyte development. 26. Kobayashi, S., Hirano, T., Kakinuma, M. & Uede, T. (1993) Biochem. Biophys. Res. Commun. 191, 617-624. We are grateful to Drs. R. Kuhn, B. Wieringa, P. Soriano, and R. D. 27. Chu, B. J.-L., Drappa, J., Parnassa, A. & Elkon, K. B. (1993) J. Palmiter for E14-1 ES cells, the 129/Sv genomic library, pSA,Bgal, and Exp. Med. 178, 723-730. DT-A, respectively. We also thank Drs. H. Fujiwara, N. Shirafuji, and 28. Wu, J., Zhou, T., He, J. & Mountz, J. D. (1993) J. Exp. Med. 178, R. Watanabe-Fukunaga for their help in generation of Fas-null mice 461-468. and Ms. K. Enari for secretarial assistance. This work was supported 29. Mariani, S. M., Matiba, B., Armandola, E. A. & Krammer, P. H. in part by Grants-in-Aid from the Ministry of Education, Science and (1994) Eur. J. Immunol. 24, 3119-3123. Culture of Japan. 30. Adachi, M., Suematsu, S., Kondo, T., Ogasawara, J., Yoshida, N. & Nagata, S. (1995) Nature Genet. 11, 294-299. 1. Egerton, M., Scollay, R. & Shortman, K. (1990) Proc. Natl. Acad. 31. Watanabe, D., Suda, T., Hashimoto, H. & Nagata, S. (1995) Sci. USA 87, 2579-2582. EMBO J. 14, 12-18. 2. Webb, S., Morris, C. & Sprent, J. (1990) Cell 63, 1249-1256. 32. Ohtaki, T., Seki, S., Abo, T. & Kumagai, K. (1990) J. Exp. Med. 3. Rocha, B. & von Boehmer, H. (1991) Science 251, 1225-1228. 172, 7-12. 4. Rajewsky, K. (1992) Curr. Opin. Immunol. 4, 171-176. 33. Scott, D. E., Kisch, W. J. & Steinberg, A. D. (1993) J. Immunol. 5. Russell, D. M., Dembic, Z., Morahan, G., Miller, J. F. A. P., 150, 664-672. Biirki, K. & Nemazee, D. (1991) Nature (London) 354, 308-311. 34. Tomonari, K., Fairchild, S. & Rosenwasser, 0. A. (1993) Immu- 6. Suda, T., Takahashi, T., Golstein, P. & Nagata, S. (1993) Cell 75, nol. Rev. 131, 131-168. 1169-1178. 35. Kawabe, Y. & Ochi, A. (1991) Nature (London) 349, 245-248. 7. Takahashi, T., Tanaka, M., Brannan, C. I., Jenkins, N. A., Cope- 36. MacDonald, H. R., Baschieri, S. & Lees, R. K. (1991) Eur. J. land, N. G., Suda, T. & Nagata, S. (1994) Cell 76, 969-976. Immunol. 21, 1963-1966. 8. Vignaux, F., Vivier, E., Malissen, B., Depraetere, V., Nagata, S. 37. Izui, S., Kelley, V. E., Masuda, K., Yoshida, H., Roths, J. B. & & Golstein, P. (1995) J. Exp. Med. 181, 781-786. Murphy, E. D. (1984) J. Immunol. 133, 227-233. 9. Suda, T., Okazaki, T., Naito, Y., Yokota, T., Arai, N., Ozaki, S., 38. Rieux-Laucat, F., Le Deist, F., Hivroz, C., Roberts, I. A. G., Nakao, K. & Nagata, S. (1995) J. Immunol. 154, 3806-3813. Denatin, K. M., Fischer, A. & de Villarty, J. P. (1995) Science 268, 10. 1347-1349. Itoh, N., Yonehara, S., Ishii, A., Yonehara, M., Mizushima, S., 39. Fisher, G. H., Rosenberg, F. J., Straus, S. E., Dale, J. K., Mid- Sameshima, M., Hase, A., Seto, Y. & Nagata, S. (1991) Cell 66, delton, L. A., Lin, A. Y., Strober, W., Lenardo, M. J. & Puck, 233-243. J. M. (1995) Cell 81, 935-946. 11. Watanabe-Fukunaga, R., Brannan, C. I., Itoh, N., Yonehara, S., 40. Tanaka, M., Suda, T., Takahashi, T. & Nagata, S. (1995) EMBO Copeland, N. G., Jenkins, N. A. & Nagata, S. (1992) J. Immunol. J. 14, 1129-1135. 148, 1274-1279. 41. Dhein, J., Daniel, P. T., Trauth, B. C., Oehm, A., Moller, P. & 12. Yonehara, S., Ishii, A. & Yonehara, M. (1989) J. Exp. Med. 169, Krammer, P. H. (1992) J. Immunol. 149, 3166-3173. 1747-1756. 42. Watanabe-Fukunaga, R., Brannan, C. I., Copeland, N. G., Jen- 13. Nagata, S. & Golstein, P. (1995) Science 267, 1449-1456. kins, N. A. & Nagata, S. (1992) Nature (London) 356, 314-317. 14. Trauth, B. C., Klas, C., Peters, A. M. J., Matzuku, S., Moller, P., 43. Matsuzawa, A., Moriyama, T., Kaneko, T., Tanaka, M., Kimura, Falk, W., Debatin, K.-M. & Krammer, P. H. (1989) Science 245, M., Ikeda, H. & Katagiri, T. (1990) J. Exp. Med. 171, 519-531. 301-305. 44. Scollay, R. G., Butcher, E. C. & Weissman, I. L. (1980) Eur. J. 15. Brunner, T., Mogil, R. J., LaFace, D., Yoo, N. J., Mahboubi, A., Immunol. 10, 210-218. Echeverri, F., Martin, S. J., Force, W. R., Lynch, D. H., Ware, 45. Giese, T. & Davidson, W. F. (1992) J. Immunol. 149, 3097-3106. C. F. & Green, D. R. (1995) Nature (London) 373, 441-444. 46. Budd, R. C., MacDonald, H. R., Lowenthal, J. W., Davignon, 16. Ju, S.-T., Panka, D. J., Cui, H., Ettinger, R., El-Khatib, M., Sherr, J.-L., Izui, S. & Cerottini, J.-C. (1985) J. Immunol. 135, 3704- D. H., Stanger, B. Z. & Marshak-Rothstein, A. (1995) Nature 3711. (London) 373, 444-448. 47. Giese, T. & Davidson, W. F. (1994) J. Immunol. 152, 2000-2010. 17. Dhein, J., Walczak, H., Baumler, C., Debatin, K.-M. & Krammer, 48. Sobel, E. S., Katagiri, T., Katagiri, K., Morris, S. C., Cohen, P. L. P. H. (1995) Nature (London) 373, 438-441. & Eisenberg, R. A. (1991) J. Exp. Med. 173, 1441-1449. 18. Russell, J. H., Rush, B., Weaver, C. & Wang, R. (1993) Proc. Natl. 49. Rathmell, J. C. & Goodnow, C. C. (1994) J. Immunol. 153, Acad. Sci. USA 90, 4409-4413. 2831-2842. 19. Nagata, S. & Suda, T. (1995) Immunol. Today 16, 39-43. 50. Daniel, P. T. & Krammer, P. H. (1994) J. Immunol. 152, 5624- 20. Cohen, P. L. & Eisenberg, R. A. (1991) Annu. Rev. Immunol. 9, 5632. 243-269. 51. Watanabe, D., Suda, T. & Nagata, S. (1995) Int. Immunol. 7, 21. Singer, G. G. & Abbas, A. K. (1994) Immunity 1, 365-371. 1949-1956. 22. Sidman, C. L., Marshall, J. D. & von Boehmer, H. (1992) Eur. J. 52. Morse, H. C., III., Davidson, W., Yetter, R., Murphy, E., Roths, Immunol. 22, 499-504. J. & Coffman, R. (1982) J. Immunol. 126, 2612-2615. Downloaded by guest on September 24, 2021