functional significance of the fas molecule in naive lymphocytes

International Immunology, Vfc>/. 8, No 3, pp. 423-431

Functional significance of the Fas moleculein naive lymphocytesSatoru Senju1, Izumi Negishi2, Noboru Motoyama, Fanping Wang, Kei-ichiNakayama2, Keiko Nakayama2, Philip J. Lucas, Shigetsugu Hatakeyama,Qing Zhang, Shin Yonehara3 and Dennis Y. Loh2

Howard Hughes Medical Institute, Departments of Medicine, Genetics and Molecular Microbiology,Washington University School of Medicine, St Louis, MO 63110, USA

1Present address: Division of Immunogenetics, Department of Neuroscience and Immunology, KumamotoUniversity Graduate School of Medical Science, Kumamoto 860, Japan2Present address: Nippon Roche Research Center, 200 Kajiwara, Kamakura-shi, Kanagawa 247, Japan3lnstitute for Virus Research, Kyoto University, Shogoin, Sakyo-ku, Kyoto 601-01, Japan

Keywords autoimmunity, homologous recombination, lymphoprohferation

Abstract

The Fas molecule mediates apoptotic signal In many cell types. Mouse mutations (Ipr, Ipr^, glcf),which impair the function of Fas, cause spontaneous autoimmune disease. We generated Fas-deficient (Fas"'") mice by homologous recombination. In embryonic stem cells Fas"'" micedeveloped /pr-llke disease, confirming that the abnormality of Fas Is causal in the Ipr phenotype.We also made Fas"7" chimerlc mice composed of a mixture of Fas+ / + and Fas"'" cells. The chlmerlcmice also showed the Ipr phenotype. In Fas"'" chimeric mice, the Fas-deflclent populationexpanded progressively among mature T and B lymphocytes. The expansion of Fas-deficientlymphocytes occurred at the naive, pre-prlmed, lymphocyte stage. These results suggest that theFas molecule functions not only after antlgenlc stimulation, as previously hypothesized, but also atthe naive lymphocyte stage.

Introduction

Fas/Apo-1 (CD95) was originally discovered as a trans-membrane protein which mediated an apoptotic signal (1-4)Ipr mice, which show lymphoproliferative disease character-ized by massive enlargement of lymph nodes and hypergam-maglobulinemia, express reduced amounts of Fas moleculedue to the insertion of a transposable element into the Faslocus (5-9). lpr°9 mice and gld mice develop lymphoprolifer-ative disease similar to Ipr m\ce, and have point mutations inthe genes encoding for Fas and Fas ligand respectively(6,10-13).

Antigenic stimulation enhances the expression of Fas onlymphocytes. Several days after stimulation, lymphocytesbecome sensitive to Fas-mediated cell death (14-21). Activa-tion-induced death of lymphocytes is impaired in .Ipr mice,and it is hypothesized that the major role of Fas is tocause cell death of activated lymphocytes and appropriatelyterminate the immune response (22-27).

We generated mice totally lacking Fas by homologousrecombination in embryonic stem (ES) cells. By injecting

Fas"'" ES cells into Fas+/+ blastocysts, we also made Fas"'"chimeric mice, which were composed of Fas+/+ and Fas"'"cells, and examined comparatively the phenotype of Fas+/+

and Fas"'" lymphocytes in vivo.

Methods

Construction of targeting vectors

A mouse Fas cDNA clone (generously provided by Dr S.Nagata; 28) was used as a probe to isolate genomic DNAclones corresponding to the Fas locus from a mouse strain129 library (Stratagene, La Jolla, CA). In the targeting vectors,pFKO-Neo and pFKO-Hyg, the entire Fas gene except forexon 1 was replaced with either a PGK-neo-poly(A) cassetteor a PGK-Ziyg-poly(A) cassette (29) respectively. Targetingvectors contained 1 3 kb of homology 5' and 10.6 kb 3' ofthe drug-resistance marker. The PGK-tfc-poly(A) cassette (30)was ligated into a restriction site in a vector polylinker at the3' end of the insert (Fig. 1A)

Correspondence to D. Y Loh

Transmitting editor T Sasazuki Received 30 June 1995, accepted 13 October 1995

Downloaded from https://academic.oup.com/intimm/article-abstract/8/3/423/858656by gueston 10 April 2018

424 Fas in naive lymphocytes

Generation of Fas+/~ and Fas^' ES clones

Transfection and screening of E14 ES cells were performedas described previously (29). E14 ES cells were transfectedwith targeting vector, pFKO-Neo, to make Fas+/" ES clones.For screening of pFKO-Neo-targeted clones by PCR, a Fas-flanking primer, Fl (5'-CTTGTCCTGGCATTCCCCTACACT-3'),and a PGK-promoter-specific primer (5'-TGCTAAAGC-CCATGCTCCAGACTG-3') were used. Fas+/" ES cells werefurther transfected with targeting vector, pFKO-Hyg, to makeFas"'" ES cell clones. For screening of pFKO-Hyg-targetedclones, primer Fl and poly(A)-fragment-specific primer (5'-GGTGGGGTGGGATTA-GATAAATGC-3') were used. PCRpositive clones were analyzed by Southern blot hybridizationto verify for homologous recombination (Fig. 1B).

ES cells were microinjected into C57BL/6 (B6) blastocystsInjected blastocysts were implanted into uteri of pseudo-pregnant ICR mice to generate chimeric mice.

Genotyping of mice

Tail DNA was analyzed by PCR amplification to determinethe genotype in the Fas locus Neomycin-resistance-gene-specific primers (5'-ATTCGGCTATGACTGGGCACAACA-3'and 5'-CAGGAGCAAGGTGAGATGACAGG-3') were used todetect pFKO-Neo-targeted allele. Wild-type allele wasdetected by PCR amplification of a 0.2 kb fragment inthe last exon of the Fas gene, using. PCR primers (5'-GAAGGACCTTGGAAAATCAACC-3' and 5'-TCTCAGCAACT-GCAGAGAATAAC-3')

Flow cytometry analysis

The following mAb conjugated with FITC, phycoerythrin (PE)or biotin were purchased from PharMingen (San Diego, CA)and used to stain cells: anti-CD4 (RM4-5), anti-CD8a (5-6.7),anti-B220 (RA3-6B2), anti-Ly-9.1 (30C7), anti-TCRo# (H57-597) and anti-CD62L (MEL-14). PE-conjugated goat anti-mouse IgM polyclonal antibody, anti-lgD (SBA1) mAb, andanti-CD44 (KM201) mAb were purchased from SouthernBiotechnology Associates (Birmingham, AL). Red613-strepta-vidin (Gibco/BRL, Grand Island, NY) was used to revealthe biotin-conjugated mAb. Stained cells were analyzed onFACScan flow cytometer with CELLQuest software (BectonDickinson, Mountain View, CA). Viable lymphocytes weregated according to forward and side scatter and propidiumiodide staining. In some analysis, to purify T cells, lymphnode cell samples were incubated with rat anti-mouse B220(RA3-6B2) mAb, washed and mixed with Dynabeads (Dynal)coated with sheep anti-rat IgG. Unbound cells were recoveredby using a magnetic separator (Advanced Magnetics). Afterpurification twice with Dynabeads, purity of the recoveredcell population was established by flow cytometric analysisand was always >96% TCRaP+ T cells.

ELISA

Blood samples were obtained from mouse tails and centri-fuged to isolate sera. Microtitration plates (Immulone 4;Dynatech, San Diego, CA) were coated with goat antiserumto mouse IgM (Cappel, San Diego, CA), goat antiserum tomouse IgG (Cappel) or heat-denatured herring sperm DNA(10 ng/ml) at 37°C for 3 h. Plates were washed with washing

buffer (PBS containing 0.05% Tween 20) and incubated withblocking buffer (PBS containing 1.0% bovine serum albuminand 0.05% Tween 20) at room temperature for 1 h. Serumsamples diluted 200-3,125,000 times with blocking bufferwere added to the plates and incubated at 4°C for 12 h.After washing, alkaline phosphatase (AP)-conjugated goatanti-mouse IgM antibody (Pierce, Chicago, IL) or AP-conjugated goat anti-mouse IgG antibody (Cappel) wereadded to the plates and incubated at room temperaturefor 2 h. After washing, p-nitrophenyl-phosphate (Sigma, StLouis, MO) in reaction buffer (0.1 M glycine, 1 mM MgCl2,1 mM ZnCI2, pH 10.4) was added to the plates. Plateswere incubated at room temperature for 1-12 h and colordevelopment was measured on a microtitration plate readerMR700 (Dynatech) Purified mouse IgM (MOPC-104E, Phar-Mingen) or IgG (RPC5; Cappel) antibody was used as astandard to convert the OD values of data to the concentration(mg/ml).

Results

Germlme-transmitted Fas~^~ mice develop the Ipr phenotype

Using the targeting vector, pFKO-Neo, for homologousrecombination, the Fas gene was deleted except for exon 1,which encodes half of the signal peptide (Fig. 1). This largedeletion (>30 kb) was created to avoid any Fas-relatedresidual function. Fas+/" ES cell clones were obtained at afrequency of 1% from G418/gancyclovir-resistant pFKO-Neo-transfectant clones. Fas+/~ ES cell clones were injected intoB6 blastocysts. Resulting chimeric mice were crossed to B6mice. Fi offsprings of agouti coat color were genotyped toselect Fas+/" mice. Fas+/~ mice were intercrossed, andFas+/+, Fas+/" and Fas"7" F2 mice were obtained Absenceof the Fas molecule in Fas"'" mice was confirmed by flowcytometric analysis (Fig. 2A).

We observed progressive enlargement of lymph nodes andspleens in Fas"7" mice (Fig 3). Atypical CD4"CD8" TCR+

B220+ lymphocytes appeared in enlarged peripherallymphoid organs and peripheral blood of Fas"'" mice (Fig. 2Band data not shown) Fas"7" mice produced higher titers oftotal IgM, total IgG, anti-DNA IgM and anti-DNA IgG antibodythan Fas+/+ or Fas+/" littermates (Fig. 2C). Collectively, germ-line-transmitted Fas"7" mice showed the same lymphopro-liferative disease as Ipr mice, confirming that a defect in Fasis the cause of the Ipr phenotype

chimeric mice show the Ipr phenotype

Fas+/" ES cells were further transfected with the targetingvector, pFKO-Hyg, and selected to make Fas"^" ES cell clones.Fas"'" ES cells and parental ES cells (Fas+/+) were separatelyinjected into B6 blastocysts (Fas+/+) to make chimeric mice(designated as Fas"7" chimeric mice and control chimericmice respectively) In Fas"7" chimeric mice, the ES-denvedpopulation is Fas"'" and the blastocyst-derived population isFas+/+. In control chimeric mice, both the ES-denved andblastocyst-derived population are Fas+/+. Among the lympho-cytes of chimeric mice, we identified ES-derived lymphocytesby an allotypic marker Ly-9.1, which is expressed on the cellsurface of 129 (Ly-9.1) but not B6 (Ly-9.2) lymphocytes(29). As expected, both Ly-9.1+ (ES-denved) and Ly-9.1"


Fas in naive lymphocytes 425

B

B EBmB B

Fas gene

pFKO-Neo/Hyg

Mutant allele

B BmB BB

1 0 kb *0 kb , >30 kb -

"I \ \W\\\ \ \ \W\ tK

neoor hygprobe

kb

6.0 -

4 . 0 -

-Neo/Hyg

-Wild-type

BE B

i JUU

6.0 kb

2 k b

Fig. 1. Target disruption of the Fas gene (A) Schematic depiction of the mouse Fas gene, targeting vectors and mutant allele of the Fas geneHatched boxes indicate 3'- and 5'-homologous regions included in the targeting vectors. Closed boxes indicate exons. Locations of theinitiation codon (ATG) and the termination codon (TGA) of the Fas gene are shown. Abbreviations for restriction sites were as follows. B, Bgltt,Bm, Ba/nHI; E, EcoRI. The restriction map of £coRI is not complete (B) Southern blot analysis showing homologous recombination. GenomicDNA samples from parental ES cells, a Fas"1"'- ES clone and a Fas"'" ES clone were digested with Bg/l\. Location of the probe was as shownin (A). The 4 0 kb band indicates wild-type and the 6.0 kb band indicates the mutant allele as shown in (A).

Fas+'+ Fas"'

°CM

m

38

2,§

3.9

•4 -115: * *

1,3

75

' * 9.3

Fas"1-/* Fas"'" apTCR

23

( 38

2.S

KI 4.6 0.7

13

apTCR CD4

Ft,*

Faa"'

ai 1.0 10total IgM (mg/ml)

0.1 1.0 10

anti-aaDNA IgM

Faa*'*F«8*'-Faa<"

1 10 100total IgG (mg/ml)

0.01 1.0 100

antl-saDNA IgG

Fig. 2. (A) Two-color flow cytometric analysis showing the absence of Fas molecule in germline-transmitted Fas"'" mouse. Thymocytes from aFas"'" mouse and a Fas+/+ mouse were stained with anti-TCRafi-FITC and anti-Fas-PE. (B) Appearance of atypical lymphocytes in germline-transmitted Fas"'" mouse. Lymph node cells from a 5-month-old Fas"'" mouse and a Fas+/+ littermate were stained with either anti-TCRotp-PEand anti-B22O-biotin or anti-CD4-PE and anti-CD8a-biotin, followed by Red613-streptavidm. Two-color flow cytometric profiles of TCRa0/B220 (top) and CD4/CD8 (bottom) are shown. Percentages for quadrants are given. (C) Hypergammaglobulinemia and production of anti-DNA antibody in Fas"'" mice Serum levels of total IgM, anti-ssDNA IgM, total IgG, and anti-ssDNA IgG antibody of Fas+/+ mice (n = 4),Fas"1"'" mice (n - 5) and Fas"'" mice (n = 7) were measured by ELISA. Serum samples were obtained from 4-month-old mice. A serum fromone Fas"'" mouse was used as a standard serum in assay of anti-DNA IgM and IgG antibody. The concentration of anti-DNA IgM and IgGantibody in each sample was expressed as a relative value. The average values and SD (error bars) are shown.

(blastocyst-derived) thymocytes of control chimeric miceexpressed the Fas molecule. On the other hand, Ly-9.1 +

thymocytes of Fas"'" chimeric mice did not express the Fasmolecule (Fig. 4A).

We observed progressive lymphadenopathy and spleno-megaly (Fig. 3), appearance of atypical lymphocytes (Fig.

4B), hypergammaglobulinemia and production of anti-DNAantibody (Fig. 4C) in Fas"'" chimeric mice, but not in controlchimeric mice. In brief, Fas"'" chimeric mice also developedIpr syndrome.

Atypical CD4"CD8- TCR+ B220+ lymphocytes in Fas"'"chimeric mice were exclusively Ly-9.1+, indicating that they



B

Fig. 3. Massive enlargement of lymph nodes and spleens in Fas-deficient mice. A Fas+ '+ mouse (A), a 6-month-old Fas"'" mouse (B) and a7-month-old Fas"7" chimeric mouse (C) are shown.

originated from Fas"'" ES cells. As observed in Ipr mice Age-dependent expansion of Fas''' lymphocytes in(31,32), enlarged peripheral lymphoid organs of Fas"'" chimeric micechimeric mice contained conventional CD4+, CD8+ and lgM+

lymphocytes in addition to the atypical lymphocytes (Fig. 4B). We examined the age-dependent change of the ES-derived



B

Control chimera Fas'/' chimera

Ly-9.1

Control chimera Fas"'' chimera17

36

0.6

apTCR

1

;

22

•*•

1.6

• & -

41

m36

14

M

0.3

•ft21

Control ChimeraFas7" Chimera

Control ChimeraFat7* Chimera

0.1 1.0 10

total IgM (mg/ml)

0.1 1.0 10

antl-ssDNA IgM

Control ChimeraF » - " Chimera

Control Chimera

Fas7" Chimera

Ly-9.1 CD4

1 •• 100

tout IgO (mg/mO

oo ioo

arrtf-eaONA IgQ

Fig. 4. (A) Two-color flow cytometric analysis showing the absence of Fas molecule in ES-derived cells of Fas"'" chimeric mice Thymocytesfrom a Fas"'" chimeric mouse and control chimeric mouse were stained with anti-Ly-9 1-FITC and anti-TCRaf)-PE (B) Appearance of atypicallymphocytes in a Fas"'" chimeric mouse Lymph node cells from a 5-month-old Fas"'" chimeric mouse and control chimeric mouse werestained with either anti-Ly-9.1-FITC, anti-TCRoefJ-PE and anti-B22O-biotin or anti-Ly-9 1-FITC, anti-CD4-PE and anti-CD8a-biotin, followed byRed613-streptavidin Histogram plots of Ly-9 1 (top) and dot plots of TCRap/B220 (middle) and CD4/CD8 (bottom) are shown Cell populationsgated on Ly-9.1+ (as indicated in histogram plots) are shown in dot plots Percentages for quadrants in Ly-9.1+ cell populations are given.(C) Hypergammaglobuhnemia and production of anti-DNA antibody in Fas"'" chimeric mice Serum levels of total IgM, anti-ssDNA IgM, totalIgG and anti-ssDNA IgG antibody of control chimeric mice (n = 5) and Fas"'" chimeric mice (n = 7) were measured by ELISA Serum sampleswere obtained from 4-month-old mice Serum from a Fas"'" mouse was used as a standard in assay of anti-DNA IgM and IgG antibody Theconcentration of anti-DNA IgM and IgG antibody in each sample was expressed as relative value to this standard serum The average valuesand standard deviations (error bars) are shown

Tcell

Control Chimera

Fas"'" Chimera

10 100

Relative Ly-e.iM.y-fl.r

population among lymphocytes of Fas"'" and control chimericmice, by following (Ly-9.1+ cell number):(Ly-9.1~cell number)ratio in peripheral blood T and B lymphocytes in chimericmice at 3, 8 and 12 weeks. To standardize the variation oforiginal contribution of ES-derived population among miceand cell lineages, we calculated the relative ratio as (ratio atspecific age)/(ratio at 3 weeks) for each chimeric mouse. Asshown in Fig. 5, the Fas"'" population progressively increasedamong T and B lymphocytes. Although the rate of increasewas much greater in T cells than in B cells, these findings

B Bcell

Control Chimera

FM"' - Chimera

I 3wl< IL:,i|—3wk

Swk

I 12 wk

•—

—i i •1.0 10

Relative Ly-9.1*/Ly-9.r

Rg. 5. Progressive expansion of Fas"'" population among peripheralblood T and B lymphocytes in chimeric mice Tail blood sampleswere obtained from chimeric mice at 3, 8 and 12 weeks. The Ly-9 1+:Ly-9 1" ratio in peripheral blood T and B cell populations wasdetermined, based on two-color flow cytometric analysis The age-dependent change of the Ly-9.1+:Ly-9 1" ratio in each mouse wasevaluated by the relative ratio calculated as described in the text.TCRaP+ cells were defined as T lymphocytes and B220+ cells weredefined as B lymphocytes at 3 weeks. CD4+ and CD8+ cellswere defined as T lymphocytes and lgM+ cells were defined as Blymphocytes at 8 and 12 weeks. The average values and SD (errorbars) of Fas"'" chimeric mice (n = 10) and control chimeric mice(n = 5) are shown.



Control chimera

Control Chimera

" Thymocyte TCR l o

Thymocyte TCR h i

Splenic T ceil

_ Lymph Node T cell

~~ Thymocyte TCH t 0

Thymocyte TCR n i

Splenic T cell

_ Lymph Node T cell

B

[Borw marrow B220+lgM"

Bon .m. r ,owB2 M * l 9 U*

Splenic B cell

Lymph node B cell

10 100

Relative Ly-9.1*/Ly-9.i"

F M ' " Chimera

|— Bone marrow B220+lgM~

Bone marrow 8220*lgM*

Splenic B cell

I Lymph node B cell

Relative Ly-9.1*fl.y-9.r

Fig. 6. Increase of the Fas"'" population among mature lymphocytesof chimeric mice. Ly-9 1 + Ly-9 1" ratios were calculated in T and Bcell populations in thymi, lymph nodes, spleens and bone marrow of8-week-old chimeric mice, based on two-color flow cytometricanalysis. The relative ratio was calculated as described in the text.CD4+ or CD8+ cells in lymph nodes and spleens were defined asmature T cells. lgD+ cells were defined as mature B cells Theaverage values and SD (error bars) of Fas"'" chimeric mice (n = 5)and control chimeric mice (n = 5) are shown

indicate that the Fas molecule regulates the number of bothT and B lymphocytes under normal situations, and that theabsence of Fas molecules causes unregulated expansion oflymphocytes

A function of the Fas molecule on mature lymphocytes

To determine the differentiation stage at which the Fas"'"lymphocyte population expands, we analyzed the Ly-9.1 + .Ly-9.1" ratio in T and B lymphocytes in thymi, spleens, lymphnodes, and bone marrow of 8-week-old chimeric mice.

We calculated the relative Ly-9 1 + Ly-9 1" ratio for T celllineage as (ratio in a specific population of T cell lineage).(ratioin TCR10" thymocyte population) for each mouse As shownin Fig. 6(A), the relative ratio in the TCRh'9h thymocyte popula-tion in Fas"'" chimeras was not different from that in controlchimeras. On the other hand, in mature T cell populations inlymph nodes and spleens, the relative ratio in Fas"'" chimeraswas higher than that in control chimera

We calculated the relative Ly-9.1+:Ly-9.1" ratio for B celllineage as (ratio in a specific population of B cell lineage):(ratioin B220+ IgM" bone marrow cell population) for each mouse.As shown in Fig. 6(B), the relative ratio in B220+ lgM+ bonemarrow cell population in Fas"'" chimeras was not differentfrom that in control chimera. On the other hand, in thepopulations of lgD+ mature B cells in spleens and lymph

Thymocyte

Mature T cell

B

Thymocyte

Mature T cell

L—

—

I—

—

TCR10 | p ]

TCR"1 Pal1

TCR* ^ B | H

TCR*CO62Lhl ^ p p ~ |

TCR*CD62L'° ^ (

TCR*CD44l0 | P P P ] - I

TCR+CD44hl PPBP1—1

1 10 100Relative Ly-9.1*/Ly-9.1"

Fas"'' chimera

TCH'° p |

TCR>" • • <

TCR* JPPPPPPPPPPP] 1TCR*CD62Lhl BPPB^PPJPPflPPP •TCR*CD62L'° ^ P P P P P P P B — 'TCR*CD44'° ^PBPBPPPPPPP]—1TCR*cD44hl PPHPBBBBBB]—^

1 10 100Relative Ly-9.1*/Ly-9.1'

Fig. 7. Increase of Fas"'" population among naive T cells in chimericmice The Ly-9.1+.Ly-9.1" ratio was determined in thymocytes and inCD62Lh|9h, CD62Lk)w, CD44hl8h and CD44low lymph node T cellpopulations, based on two-color flow cytometric analysis Tlymphocytes purified from lymph nodes of 5-week-old chimeric micewere analyzed. The relative ratio was calculated as described in thetext. The average values and SD (error bars) of Fas"'" {n = 5) andcontrol (n = 9) chimeric mice are shown

nodes, the relative ratio in Fas"' chimeras was higher thanthat in control chimera.

More than 80% of lymphocytes were of conventional pheno-type (CD4+, CD8+ or lgM+) in 8-week-old Fas^" chimericmice, whose lymph nodes contained at least six times asmany lymphocytes as control chimeric mice in the case thatcoat-color-chimerism was 70% or more. Therefore, Fas"'"lymphocytes increased not only relatively to Fas+/+ lympho-cytes but also in absolute number.

Taken together, the Fas"7" population abnormally increasedin mature T and B lymphocytes in peripheral lymphoid organs,but not in thymus or bone marrow. These results suggest thatFas functions at the mature lymphocyte stage, althoughwe cannot rule out that Fas has a function also at theimmature stages

Functional significance of the Fas molecule in naivelymphocytes

Because lgD+ B cells are at the naive or pre-pnmed stage(33), the result described above that the Fas"'" populationincreased among lgD+ B lymphocytes indicates an abnormalincrease of Fas"'" naive B cells. Therefore, in normal situations,the Fas molecule functions to regulate the number of naiveB cells.

We examined whether Fas has a regulatory role also innaive T cells. To avoid the complexity in flow cytometric



analysis due to the existence of atypical lymphocytes, weanalyzed 5-week-old chimeric mice, in which atypical lympho-cytes were <4% of total lymph node cells of Fas"'" chimericmice

High expression of CD62L (L-selectin) and low expressionof CD44 (Pgp-1) are markers for naive lymphocytes (34-37).We calculated the relative Ly-9 1+:Ly-9.1" ratio for lymphnode T cell populations classified according to the level ofexpression of CD62L and CD44, by the same formula asdescribed above. In both CD62Lh|9h and CD62Llow as well asboth CD44h|9h and CD44low T cell populations, the relativeratio in Fas"'" chimeric mice was higher than that in controlchimeric mice (Fig. 7) Thus, the unregulated increase ofFas"'" T cell occurred before the differentiation to memorycells, suggesting that Fas plays a regulatory role in naiveT cells.

Discussion

The genomic region encoding mature Fas protein was com-pletely deleted by the targeting constructs which span atleast 30 kb. Although we have not obtained contiguous DNAclones covering all the deleted region, the deletion may be>50 kb, based on the report that the mouse Fas gene consistsof >70 kb (7).

Not only germline-transmitted Fas"'" mice but also Fas"'"chimeric mice showed the Ipr phenotype These results areconsistent with a report on allophenic chimeras developedfrom Ipr/lpr and +/+ embryos (38) On the other hand,bone marrow chimeras, irradiated +/+ mice implanted withIpr/lpr bone marrow grafts, did not show the Ipr phenotypebut showed graft-versus-host-disease-like syndrome andatrophy of lymphoid organs (39-41). The difference betweenembryonic chimeras and bone marrow chimeras is that Fas"non-bone-marrow-derived cells exist in embryonic chimeras,but not in bone marrow chimeras'. Therefore, the existence ofFas" non-bone-marrow-derived cells, probably stromal cellsin lymph node and spleen, may be a requisite for lymphoprolif-eration of Ipr mice This explanation is supported by the reportof Matsuzawa et al (42) that Ipr/lpr bone marrow graftscaused enlargement of simultaneously implanted lymphnodes derived from Ipr/lpr mice but not those derived from+/+ mice.

From the data shown in Figs 6 or 7, we cannot rule out thatFas has a function on immature cells which manifests itselfin more mature cells. However, an age-dependent increaseof the discrepancy of Ly-9.1+:Ly-9.1" ratio between the matureand immature populations supports the notion that Fas func-tions at the mature lymphocyte stage. In 2-week-old chimericmice, there was no effect of the absence of Fas on thediscrepancy in the Ly-9.1+.Ly-9.1" ratio between mature andimmature populations (data not shown). On the other hand,we observed a progressively increasing discrepancy in olderFas"'" chimeric mice (Figs 6 and 7, and data not shown).These results suggest that mature lymphocytes, probablyself-renewing, may accumulate in peripheral lymphoid organswhen they are Fas-deficient, although a possibility still remainsthat the effect of Fas on immature cells increases in an age-dependent manner.

There are many reports that Fas is involved in activation-

induced lymphocyte cell death (19-26). It is hypothesizedthat the major role of Fas in lymphoid systems is to killactivated lymphocytes and terminate the immune response.The current idea is that the failure of Fas-mediated death ofactivated lymphocytes causes abnormal lymphoproliferationin Ipr and gld mice (43). However, there is no evidencedirectly connecting the impairment of activation-induced celldeath with lymphoproliferation and autoimmune disease of/p/rnice In our in vitro experiment, TCR-mediated stimulationcaused cell death even in Fas"'" T cells, although the rate ofdeath was lower than that in Fas+/+ T cells (unpublishedobservation). Thus, activation-induced cell death does nottotally depend on Fas.

In the Fas"'" chimeric mice, the Fas"'" population expandedin the naive T and B lymphocytes, suggesting that the Fasmolecule plays a regulatory role in lymphocytes even beforethey are stimulated with exogenous antigen. In agreementwith this role, the Fas molecule is expressed on most of themature T lymphocytes (most of them are naive T cells) and>20% of lgD+ B lymphocytes in normal mice (data notshown). Our results are supported by data from Sidman et al.(44), who reported that T lymphocytes bearing male-antigen-specific TCR participated in the lymphoproliferative processeven in female Ipr or gld mice. Their results imply thatinteractions with specific antigen are not necessary for abnor-mal proliferation of Fas-deficient lymphocytes. Ettinger et al.observed that autoreactive T lymphocytes existed mainly inthe naive T cell fraction of Ipr mice (45) We propose that, inaddition to the failure of activation-induced cell death, theimpairment of function of Fas in naive lymphocytes alsocontributes to the development of lymphoproliferation andautoimmune disease

Acknowledgements

We thank Dr S Nagata for the mouse Fas cDNA clone pMF1, andDrs I. T Chan, J D Alvarez and L. B Dustin for helpful discussionsThis work was supported by the Japan Society for the Promotion ofScience (N. M ), the Harold Simmons Leukemia Fund (P J. L ) andthe Howard Hughes Medical Institute (D. Y L).

Abbreviations

AP alkaline phosphataseB6 C57BL/6ES embryonic stemPE phycoerythrin

References1 Yonehara, S., Ishii, A. and Yonehara, M 1989. A cell-killing

monoclonal antibody (anti-Fas) to a cell surface antigen co-downregulated with the receptor of tumor necrosis factor. J. ExpMed. 169:1747

2 Trauth, B. C, Klas, C , Peters, A M. J., Matzku, S., MOIIer, P,Flak, W, Debatin, K-M. and Krammer, P. H. 1989. Monoclonalantibody-mediated tumor regression by induction of apoptosis.Science 301.244.

3 Itoh, N., Yonehara, S., Ishn, A., Mizushima, S., Sameshima, M,Hase, A., Seto, Y and Nagata, S., 1991 The polypeptide encodedby the cDNA for human cell surface antigen Fas can mediateapoptosis. Cell 66.233.



4 Oehm, A., Behrmann, I., Falk, W, Pawlrta, M., Maier, G., Klas, C ,Li-Weber, Min., Richards, S., Dhein, J., Trauth, B. C., Ponstingl,H and Krammer, P. H. 1992. Purification and molecular cloningof the APO-1 cell surface antigen, a member of the tumornecrosis factor/nerve growth factor receptor family. J. Biol. Chem.267:10709.

5 Andrews, B S., Eisenberg, R. A., Theofilopoulos, A. N., Izui, S.,Wilson, C. B., McConany, P. J., Murphy, E. D., Roths, J. B. andDixon, F J. 1978 Spontaneous murine lupus-like syndromes.Clinical and immunopathological manifestations in several strains.J. Exp. Med. 148:1198.

6 Watanabe-Fukunaga, R., Brannan, C. I., Itoh, N., Copeland, N G ,Jenkins, N. A. and Nagata, S. 1992. Lymphoproliferation disorderin mice explained by defects in Fas antigen that mediatesapoptosis Nature 356:314.

7 Adachi, M., Watanabe-Fukunaga, R. and Nagata, S. 1993Aberrant transcription caused by insertion of an early transposableelement in an intron of the Fas antigen gene in Ipr mice Proc.NatlAcad Sci USA 901756.

8 Wu, J., Zhou, T. and Mountz, J D. 1993 Autoimmune disease inmice due to integration of an endogenous retrovirus in anapoptosis gene. J. Exp Med. 178461.

9 Chu, J.-L, Drappa, J., Parnassa, A. and Elkon, K. B. 1993 Thedefect in Fas mRNA in MRL/lpr mice is associated with insertionof the retrotransposon, ETn. J. Exp. Med. 178:723

10 Matsuzawa, A., Moriyama, T., Kaneko, T., Tanaka, M., Kimura, M.,Ikeda, H. and Katagin, T 1990. A new allele of the Ipr locus, lpr°3,that complements the gldgene in induction of lymphadenopathy inthe mouse. J. Exp. Med. 171:519

11 Roths, J. B., Murphy, E. D. and Eicher, E. M. 1984 A new mutation,gld, that produces lymphoproliferation and autoimmunity in C3H/HeJ mice. J Exp Med. 159:1.

12 Takahashi, T., Tanaka, M., Brannan, C I, Jenkins, N A., Copeland,N. G., Suda, T and Nagata, S. 1994. Generalizedlymphoproliferative disease in mice, caused by a point mutationin the Fas ligand. Cell 76:969.

13 Lynch, D. H., Watson, M. L, Alderson, M R., Baum, P. R., Miller,R. E., Tough, T., Gibson, M., Davis-Smith, T, Smith, C A., Hunter,K , Bhat, D., Din, W., Goodwin, R G and Seldin, M. F. 1994. Themouse Fas-hgand gene is mutated in gld mice and is part of aTNF family gene cluster. Immunity 1:131

14 Owen-Schaub, L. B., Yonehara, S., Crump, W. L, III and Grim, E.A. 1992. DNA fragmentation and cell death is selectively triggeredin activated human rymphocytes by Fas antigen engagement.Cell. Immunol 140:197.

15 Miyawaki, T, Uehara, T, Nibu, R , Tsuji, T, Yachie, A., Yonehara,S. and Taniguchi, N. 1992. Differential expression of apoptosis-related Fas antigen on lymphocyte subpopulations in humanperipheral blood. J. Immunol. 149:3753.

16 Klas, C , Debatin, K.-M., Jonker, R. R. and Krammer, P H 1993Activation interferes with the APO-1 pathway in mature human Tcells. Int. Immunol. 5:625

17 Drappa, J., Brot, N. and Elkon, K. B. 1993. The Fas protein isexpressed at high levels on CD4+CD8+ thymocytes and activatedmature rymphocytes in normal mice but not in lupus-prone strain,MRL Ipr/lpr. Proc. NatlAcad. Sci. USA 90.10340.

18 Daniel, P T. and Krammer, P H 1994. Activation induces sensitivitytoward Apo-1 (CD95)-mediated apoptosis in human B cells. J.Immunol 152:5624.

19 Dhein, J., Walczak.H., Bflumler, C , Debatin, K -M. and Krammer,P. H. 1995 Autocrine T-cell suicide mediated by Apo-1/(Fas/CD95). Nature 373:438.

20 Ju, S.-T, Panka, D J., Cui, H., Ettinger, R., El-Kahtib, M., Sherr,D. H., Stanger, B. Z. and Marshak-Rothstein, A 1995. Fas (CD95)/FasL interactions required for programmed cell death after T cellactivation. Nature 373.444.

21 Martin, S. J., Force, W. R , Lynch, D. H., Ware, C. F. and Green,D. R. 1995. Cell-autonomous Fas (CD95)/Fas-ligand interactionmediates activation-induced apoptosis in T-cell hybridomasNature 373.441.

22 Scott, D. E., Kisch, W. J. and Steinberg, A. D 1993. Studies of Tcell deletion and T cell anergy following in vivo administration of

SEB to normal and lupus-prone mice. J. Immunol. 150:66423 Russell, J. H., Rush, B , Weaver, C and Wang, R 1993. Mature

T cells of autoimmune Ipr/lpr mice have a defect in antigen-stimulated suicide Proc Natl Acad Sci USA 90:4409.

24 Bossu, P., Singer, G. G , Andres, P., Ettinger, R, Marshak-Rothstein, A. and Abbas, A K. 1993. Mature CD4+ T lymphocytesfrom MRL//pr mice are resistant to receptor-mediated toleranceand apoptosis J. Immunol. 151:7233.

25 Singer, G. G and Abbas, A K. 1994. The Fas antigen is involvedin peripheral but not thymic deletion of T lymphocytes in T cellreceptor transgenic mice. Immunity 1:365.

26 vlgnaux, F and Golstein, P. 1994. Fas-based lymphocyte-mediated cytotoxicity against syngeneic activated lymphocyte aregulatory pathway? Eur. J. Immunol. 24:923.

27 Rothstem, T L , Wang, J K M Panka, D. J., Foote, L. C, Wang,Z Stanger, B., Cui, H., Ju, S.-T. and Marshak-Rothstein, A. 1995.Protection against Fas-dependent Th1-mediated apoptosis byantigen receptor engagement in B cells. Nature 374163

28 Watanabe-Fukunaga, R., Brannan, C. I, Itoh, N , Yonehara, S,Copeland, N. G., Jenkins, N A and Nagata, S. 1992. The cDNAstructure, expression, and chromosomal assignment of the mouseFas antigen. J. Immunol. 148:1274.

29 Nakayama, K , Nakayama, K , Negishi, I., Kuida, K , Shinkai, Y,Louie, M. C, Fields, L, Lucas, P. J., Stewart, V., Alt, F. W. andLoh, D Y 1993 Disappearance of the lymphoid system in Bcl-2homozygous mutant chimeric mice Science 261.1584.

30 Tybulewicz, V. L , Crawford, C. E., Jackson, P K , Branson, R Tand Mulligan, R. C 1991. Neonatal lethality and lymphopenia inmice with a homozygous disruption of the c-abl proto-oncogeneCell 65:1153

31 Wofsy, D., Hardy, R. R. and Seaman, W. E. 1984. The proliferatingcell in autoimmune MRL/lpr mice lack L3T4, an antigen on'helper' T cells that is involved in the response to class II majorhistocompatibility antigens. J. Immunol. 132.2686

32 Davidson, W. F., Dumont, F. J , Bedigian, H. G., Fowlkes, B Jand Morse, H. C, III 1986. Phenotypic, functional, and moleculargenetic comparisons of the abnormal lymphoid cells of C3H-/pr/Ipr and C3H-gld/gld mice. J. Immunol. 136:4075

33 Zan-Bar, I., Vitetta, E S and Strober, S. 1977. The relationshipbetween surface immunoglobulin isotype and immune functionof murine B lymphocytes. I. Surface immunoglobulin isotypes onprimed B cells in the spleen. J. Exp. Med. 145-1188.

34 Budd, R. C, Cerottini, J.-C , Hovath, C, Bron, C, Pedrazzini, T,Howe, R ~C and MacDonald, H. R. 1987 Distinction of virginand memory T lymphocytes Stable acquisition of the Pgp-1glycoprotem concomitant with antigenic stimulation J. Immunol.138:3120.

35 Hamann, A., Jablonski-Westrich, D., Scholz, K.-U , Duijvestijn, A.,Butcher, E. C. and Thiele, H.-G. 1988. Regulation of lymphocytehoming. I Alterations in homing receptor expression and organ-specific high endothelial venule binding of lymphocytes uponlymphocyte activation. J Immunol. 140:737.

36 Lee, W. T and Vrttetta, E. S. 1991. The differential expression ofhoming and adhesion molecules on virgin and memory T cells inthe mouse. Cell. Immunol. 132:215.

37 Picker, L. J. and Butcher, E. C. 1992. Physiological and molecularmechanisms of lymphocyte homing. Annu. Rev Immunol. 10:561.

38 Katagin, T, Azuma, S., Toyoda, Y, Mori, S., Kano, K., Cohen, P.L. and Eisenberg, R. A. 1992. Tetraparental mice reveal complexcellular interactions of the mutant, autoimmunity-induced Ipr gene.J Immunol. 148:430.

39 Theofilopoulos, A. N. 1982. Role of thymus in murine lupus andcellular transfer of the disease. Arthritis Rheum 25 726.

40 Fujrwara, M and Kariyone, A. 1984. One-way occurrence of graft-versus-host disease in bone marrow chimeras between congenicMRL mice Immunology 53:251.

41 Theofilopoulos, A. N., Balderas, R S , Gozes, Y, Aguado, M. T,Hang, L., Morrow, P. R. and Dixon, F J. 1985. Association of Iprgene with graft-vs.-host disease-like syndrome. J Exp. Med1621

42 Matsuzawa, A , Katagin, T, Ogata, Y, Kominami, R. and Kimura,M 1994. Lymphadenopathy induced by the cooperation between



and gid genes is of Ipr but not of gid phenotype. Eur. J. autoimmune syndromes of Ipr and gid mutant mice. Eur. J.Immunol. 24.1714 Immunol. 22:499.

43 Watanabe, D, Suda, T., Hashimoto, H. and Nagata, S. 1995. 45 Ettinger, R., Wang, J. K M., Bossu, P, Papas, K, Sidman, C.Constitutive activation of the Fas hgand gene in L, Abbas, A. K. and Marshak-Rothstein, A. 1994. Functionallymphoproliferative disorders. EMBO J. 14:12. distinctions between MRL-/prand MRL-gld lymphocytes. Normal

44 Sidman, C L, Marshall, J. D and von Boehmer, H. 1992. cells reverse the gid but not Ipr immunoregulatory defect. J.Transgenic T cell receptor in the lymphoprolrferative disease and Immunol. 152.1557.


functional significance of the fas molecule in naive lymphocytes

Documents