AD-A26'5 836EPORT DOCUMENTATION PAGE

2a- SECURI I Y , . DISTRIBUTION /AVAILABILITY OF REPORT

Approved for public reIease;

2b. DECLASSIFICATION/DOWNGRAOING SCHEDULE distribution is un] irni ted

4, PERFORMING ORGANIZATION REPORT NUMBER(S) S. MONITORING ORGANIZATION REPORT NUMBER(S)

NMRI 93-21

6a. NAME OF PERFORMING ORGANIZATION 6b, OFFICE SYMBOL 7a. NAME OF MONITORING ORGANIZATION

Naval Medical Research 1 (If applicable) Naval Medical Command

Institute I

6c. AgORESS (C'ry, State, and ZIP Code) 7b. ADDRESS (City, State, and ZIP Code)

3901 W4isconsin Avenue Department of the Navy

Bethesda, MD 20889-5055 Washington, DC 20372-5120

8a. NAME OF FUNDING/SPONSORING "8b. OFFICE SYMBOL 9. PROCUREMENT INSTRUMENT IDENTIFICATION NUMBER

ORGANIZATION Naval Med i cal ( if applicable)

Research & Development Commanf

8C_ ADDRESS (City. State, and ZIP Code) 10. SOURCE OF FUNDING NUMBERS

8901 Wisconsin Avenue PROGRAM PROJECT TASK WORK UNIT

Bethesda, MD 20889- 5 044 ELEMENT NO. NO. NO. ACCESSION NO-

62233N MM33C30 05-1051 DN249507

11. TITLE (include Security Classification)

Comparative in vitro analysis of proliferation, Ig secretion, and Ig class switching by murine .,arginal zone andfollicular B cells

12. PERSONAL AUTHOR(S) Snapper CM, Yamada H, Smoot D, Sneed R, Lees A, Mond JJ

13a. TYPE OF REPORT 1a3b. TIME COVERED 14. DATE OF REPORT (Year, Month, Day) PAGE COUNT

journal article FROM TO I 1993 9

16. SUPPLEMENTARY NOTATION

Reprinted from: The Journal of Immunology 1993 April 1; Vol. 150 No.7 pp. 2737-2745

17, COSATI CODES 18. SUBJECT tERMS (Continue on reverle if necessary and identify by block number)

FIELD GROUP SUB-GROUP B cell; antibody production; lymphocyte transformation I19. ABSTRACT (Continue on reverse if necessary and identify by bloc<k number)

DTICSELECTEJUN 16 1993L1

20. DISTRIBUTION /AVAILABILITY OF ABSTRACT 21. ABSTRACT SECURITY CLASSIFICATION

CAUNCLASSIFIEDWJNUMITED C SAME AS RPT 0 OtiC USERS UncIassified

22a. NAME OF RESPONSIBLE INDIVIDUAL 22b. TELEPHONE (Include Area Code) 22c. OF FiCE SYMBOL

Phyllis -Blum,'_ Librarian. (301) 295-2188 MRL/Nt•RI

00 FORM 1473, 84 MAR 83 APR edition may be used until exhausved. SECURITY CLASSIFICATION OF THIS PIGEAllothe edition% are obsolete, UNCLASS I FI ED

The lournal of lrmmunulo~oy-

Copyright 0 19i93 by The AmrneCan ASSUCta~oun of tmmunologits '2

f ýs Comparative in Vitro Analysis of Proliferation, Igp Secretion, and Ig Class Switching by Murine

Marginal Zone and Follicular B Cells'1 L

Clifford M. Snapper,'* Hidehiro Yamada,' Doug Smoot," Rosie Sneed,' Andrew Lees,ý andJames J. Mon&~ ___

Dept. of Pathology and 5Dept. of Medicine, Uniformed Services University of the Health Sciences, Be;he~da, kiD 20814'Immune Cell Biology Program, Naval Medical Research Institute, Bethesda, MD 20889; and 'Geo-Certc'rs, inc , ft,Washington, MD 20744

ABSTRACT. We have previously demonstrated that activation of murine B cells by dlextran-conjugated anti-IgD an-tibodies may serve as a polyclonal, in -vitro model system for studying immune responses to T cell-indeptendent tp2 (Tl-2) Ag, as exemplified by the bacter ial polysaccha rides. Because in vivo Ig responses to TI-2 Ag are mcediatedprimarily by B cells resident in the splenic marginal zone, we wished to determine whether this reflected an intrinsicdifference in the responsiveness of marginal zone B cells (MZB) compared with follicular B cells (FBI to this classof Ag. In this report we demonstrate that highly purified MZB, isolated by electronic cell sorting, exhibit a loverproliferative response in vitro in response to unconjugated anti-Ig antibody as well as to dextran- or Sepharose-conjugated anti-lgM or anti-IgD antibodies, whereas they proliferate equal to or better than FB when stimulatedby other B cell mitogens including LPS, Salmonella tývphimurium milogen, or an anti-CD3-activated CD4- Th2 cellclone. Despite the different proliferative responses of MZB and FB3 induced by anti-Ig, Ag receptor cross-linkagestimulates comparable increases in intracellular free calcium concentrations in both of these B cell populations.Furthermore, MZB secrete Ig and undergo Ig isotype switching to a comparable degree, relative to FB, in responseto both T cell-dependent and T cell-independent stimuli. This suggests that the compartmentalization of TI-2responses to the splenic marginal zone rather than the follicular zone reflects something other than the intrinsic

* responsiveness of the B cells from these two sites. )ournal of Immunology, 1993, 150: 2737.

*he marginal zone is a specialized anatomic site in Ag, as exemplified by the polysaccharides, are typically 1

U the spleen that contains a B cell subpopulation re- found in abundance in bacterial cell walls, and unlike TDsponsible for generating antibody responses to Ag, such as soluble proteins, elicit Ig isoiype productton 91

TI-2 3 Ag (1-3). By contrast, B cells resident in the splenic that is skewed toward 1gMI and lgG3 (5). Furthenrnore, TI 1follicle (FB) secrete anti body in response to TD Ag (4). T11.2 Ag, in contrast to TD Ag, are generally ineffective at stim-

ulating a secondary or memory-type immune response

Received for publication July 15, 1992. Accepted for publication January 6, The 11.2 Ag-responsive MZB is a mature, noncycling1993.The costs of pubtication of this article were defrayed in pant by the payment ofpage charges. This article must therefore be hereby marksed advertisement in

* accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Address correspondence and reprint requests to Clifford M,. Snapper, M.D..'This work was supported, in part, by Uniformed Services University of the Department of Pathoogy. Uniformed Services University of the Health Sci-

Health Sciences Research Protocols R074CN and R083BQ, The Burroughs ences, 4301 Jones Bridge Road, Bethesda, MD, 20814.Wellcome Fund Developing Investigator Award, lmmunopharmacology of 'Abbreviations used in this paper-. 11.2, T cell-independent ty" 2; TI, T cellAllergic Diseases (C. M. S.). National institutes of Health Grant% A124273, independent, MZB, marginal zone B cell; FB, fotlicular B cell; FccRil, low

* and A127465, and Naval Medical Research and Development Command affinity Fc receptor for IgE; STM, Salmonella typhimurium mitogen; a&-dex.N0007591WR00018. opinions and assertions contained herein are the pri. dextran-conjlugated anti-lgD antibody; ap-dlex. dextran-conjugated anti-IBM 9vate oresof the authors and areto to be construed as official or reflectinn ý:. antibody; sis-t T cell-independent type 1 PE. phycoerythrin; mlg, membrane *views of the Department of Defense or she Uniformed Services University of I&. TO, 7 cell dependent; SN, supernatant; lCa2 i). intracellular free Cal-the Healt~h Sciences. concentration; PKC. protein kinase C.

2737

2738 ['f-,PN5t<-) 4 )l 'A R C.K.L Zt 0Nt A-1N)( I 1 tA 13 (t S

and noncirculairing cell (9, l1W that e~prcs~cs hých of~c uT he c~ xcrmiient% were ~conducicd ao ordiimlgNI and low or absent mleD ( 1. 12). By %contra~t the 143B ples sct fonh in the Giiafe for d~ie (cr0c and~ Uffi ,f 1,,bis a recirculating cell that expres,\cs inte e ledia ', ot urafory Artiriudy, lnttute (if Anmia ReC\'urLesý NajtiunalmlgMt and high level,,- of rrleD (11, 121). E,, idc,ýce su-oes1' Research Cou~nci, Dcpaninccnt of ficaiith. Ed~toandthat thes, two B cell subpopulations mnay repre,,cnt da~in, ct WClfafe PUNlicatto 78.23. Natinal Insiuics.o la!lineaves (2). Recently. Waldschmidt et al. dcrmon~r Bethesda, MD.that MZB and FB can be clearly sepwrlited on the lbasis oftheir differential expression of the intennedlate afflivit re- Culur rnediumceptor for IgE (Fc(Rll) (13). Thus, %IZB fail to expre-,sdetectable FcEeRII w hereas FB are FEcRlII' RPMlI 1640 (Biofluids, Rockville, NID, sUppiemnntncd --:ý

Because the frequency of primary Ag-specifitc B cells, 010c FCS (GIBCO Laboiratones, Grand Island, NY).even to potent TI Ag. is % cry low, a detailed arialssis (if the L-elutaminc (2 rnlW. 2-M\E (0.05 nnN . penicillin (so ~variables influencing their responsiveness wAOUld be dimf- ml), and streptomrciIn (50 pg/nfl) were used fur cu!:U;-:.,cult if not impossible. We therefore dex eloped an in Nitro cells.polyclonal model system for studying B cell activation inresponse to TI-2 A2 (14. 15). Anti-lgD or anti-IgNI anti- Reagentsbodies were conjugated to ahigh molecular dextran inordi-r deano-dxerpepedbcouainof -;to simulate the repeating epitope nature of polysaccharides. noilnamue 'b(baotp.at-ose1 aWe recently demonstrated that small, high density B cells aotp)(8.nd176mnoonlrtIG ti-mo.activated by a5-dex proliferated, but failed to secret,: I- gI 1)t hg w eta 2>:l~ . spunless a B cell maturation factor, such as IL-S. was also vouldeIadB6added to culture (1S). Lycontrast. a5-dlex stimulated Large,scbd(4.Aprimel6H/low density B cells to se :rete Ig in the absence of exogenous atbde eecnuae oec ernmlcl.F

4D)5 (mouse lgG2a (b allotvpe anil-mou'.e lgD (a a~ltMvpe)cytokines. This differential responsiveness of stmall and (2)wsprfdfomace.Cnrct'tdSphrelarge B cells to induction with a5-dex was similar to that (0 a uii-lfo sie.C~-ciae ehn-

C ~was purchased from Pharmacia (PiscataAv,%, NJ.). Two mliobserved for in vitro hapten-specific antibody responses to ligrams of puritied FF1 antibody were addedlmilliliter ofconjugates of hapten-Ficoll. a c-ototypic TI-2 Ag. We fur- packed, swollen Sepharose beads according to instructionsther demonstrated that crh-dc-' _ivated B cells could un- inlddbthmaucur.LS etre fo Es

* ~~~~dergo Ig class switching to most, although not all. Ig iso- ceihacl 11:4 a bandfo ic aoatypes, in the presence of appropriate switch and tre Dtot l.Sloel -piuii ioe

* diferntiaionfactrs 16, 7).(STM) was obtained from Ribi Immunochemical ResearchLittle is known regarding the parameters that determine (aitn I)adwsue t5 cm.Teflo

the differential response patterns of 1MZB and F13 to im- "~ 7eeprfe rmacts 34(a g~ nimunization with distinct classes of Ag in vivo. Thus, it is FcR)(2.MD6moslGaani!) 2.annot clear whether these differences primarily reflect intrin- A2(a g baniF-RI(3)Sectdnioissic properties of these two B cell subpopulations or whether were2 (rjuate toG~ aniFCyl (23). Scel-ectedg SantibDies

the niqe mcroevirnmets i whch hey ocaize ego, CA), N-hydroxysuccinimidobiotin (Sigma, St. Louis.and/or selective interactions with other cell types account MOadTxsRd(earh rancCvlnO)

by sti &r artp~hio&FsPE-labeled affinit.y-pt66icTiedgoaTo address this issue we obtained highly purified popula- animos 1g nioywa ucae rmSuhr

tions of MZB an! FB from the spleens of unimmunized Biotechnology Associates (Birmingham. AL). Avidin-PiEmice by utilizing an electronic cell sorter to isolate (hcore)adainalpyoynnwrebtndM~~``cRl and M~gMini~11ed'i'eFcERII " B cells, re- Pcorb)advin-lphoynnweebtnd

spectively. These sorted B cell subpopulations w-ere then from Biomeda (Foster City. CA). Purified murine rlL-4 %kastesed or hei reatie ailiy o polierae, ecrte 0, nd produced in E. c-a/i and was a gift from Dr. Alan D. Levine

tesedfo tei rlaiv ailty o roifraeseree s.an (Monsanto Company. St. Louis, MO). Murine rTL-5 wasundergo Ig isociype switching in response to TI- I (bacterial produced in the baculovirus system and was a gift from Dr.LPS), TI-2 (a8-dex and apt-dex), and TD modes of acti- Gregory Harriman, (National Institutes of Health. Be-vation. thesda, MD). Murine rIFN--y. prepared from Chinese ham-

ster ovary cells, was a gift of Genentech (South San Fran-Materials and Methods cisco, CA). Percoll was obtained from Pharmacia.

Mice

Female DBA/2, BALB/c, and C3H mice were obtained Preparation and culture of B cellsfrom the National Institutes of Health Small Animals Di- Enriched populations of B cells were obtained from spleenvision (Bethesda. MD) and were used at 8 to 12 wk of age. cells from which T cells were eliminated by treatment with

. o rurn , ot Imm dnolugy

-monoclonal rat 1gM anti-Thy-l(H013-4), rat lgG2b anti- independent, MHC-unrestricted, T cell-inediated stimula-

CD4 (GKI.5), and rat IgG2b anti-CD8 (2.43). followed by tion of B cells (28, 29). 96-well culture plates were first

;' buse anti-rat lgp (MAR 18.5) and C by the method of coated with anti-CD3 antibody (2C1 1) (30) by incubation

Lcibson et al. (24). Small B cells were obtained by the of 50 pg/ml of 2C II in PBS at room temperature for 4 h.

modified (25) discontinuous Percoll gradient centrifugation Plates were subsequently washed three times with PBS.

procedure of DeFranco et al. (26). Cells that formed a band Resting DIO T cells (3 X 10' cells/ml) "ere then co-

b,\ een 60 and 70% Percoll and had a density of 1.080 to cultured with either sort-purified MZB or FB cells (1 ( 1 &

106g/ml were used in all experiments. Functional assays cells/ml) in the anti-CD3-coated microtiter vAells for in-

were carried out in 96-well, flat bottom Costar plates (Cos- duction of B cell proliferation or lg isot) pe production. D 10

tar. Cambridge, MA). Cultured cells were incubated at T cells were first irradiated with 3000 r to prevent then

37,C in a humidified atmosphere containing 6% CO,. Un- from proliferating. without inhibiting their ability to be-

less otherwise indicated, cells were cultured at a density of come activated or secrete cvtokines.

1.25 X 105 cells/ml. All experiments are representative of

at least two to three similar studies. Measurement of DNA synthesis

DNA synthesis was determined bv [-1H]TdR uptake (2 pCu/

" z-," letric analysis and ceii sorting well; 6.7 CiJnmol: I mCi = 37 GBq: ICN. Irvine, tA) over

B cells were stained for 30 min with various combinations a 16-h period. Cells were harvested onto glass filter paperand [-'H)TdR incorporation -was determined by liquid scin-

"of sterile filtered, fluorescence-labeled immunoreagents

(final concentration of 10 pg/mI in the presence of a fivefold tillation spectrometry.

excess of anti-Fc-yRII mAb to prevent cytophilic antibody,

binding) at 107 cells/mi in cold clear HBSS containing 3%

: FCS. Cells were then washed and resuspended in staining Ig isotype concentrations were measured by ELISA. --i.th

buffer at 10' cells/ml in preparation for fluorescence anal- Immulon 2.96-well, flat bottomed ELISA plates .Dynatech

ysis and/or cell sorting. Cell sorting generally was carried Liboratories, Alexandria, VA), \%hich has been describeJ

Tý out over a 5-h period. Sorted cells were cultured immedi- by us in detail elsewhere (31). Briefly, a fluorescent produci

': ately after this sorting period and were >95% viable at this was generated from cleavage of 4-methylumbilliferylphos-

time. Control studies in which stained, but nonsorted, cells phate (Sigma) by specifically-bound alkaline phosphatase-

were used were shown to behave similarly to nonstained, conjugated antibodies. Fluorescence was measured on a 3

nonsorted cells upon activation with a-dex or LPS (data M FluoroFAST 96 fluorometer (Becton Dickinson. Moan-

not shown). For analysis, 15,000 cells were collected using tain View, CA) and fluorescence units were converted to I

logarithmic amplification on a Becton Dickinson FACStar concentrations by extrapolation from standard curves de-

"PlPis (Mountain View, CA). Only viable cells were ana- termined in each assay by using purified mycloma proteins

lyzed, and any residual macrophages were eliminated from of known concentrations. Each assay system show.ed no

analysis on the basis ofcharacteristic forward and side scat- significant cross-reactivity or interference from the pres-

terprofiles. Cell sorting was similarly carried out on a FAC- ence of other isotypes (0gM, lgD. lgG3. lgG(,0(52b,

Star Plus, as well as on a Coulter Epics Elite (Miami, FL). IgG2a, IgE, and IgA) found in the culture supernatants.

and sorted cells were immediately reanalyzed to confirmtheir stai oDetermination of intracellular Ca"c concentrations._.___..ther. t~i~ngprofile. Only_ sorting purities o 9%wr

considered acceptable for subsequent study. Our procedure for the measurement of Ca-f, in- singe-

cells has been described in detail elsewhere (32). Erief,.

Preparation of T cell-derived SN and induction of T cells were loaded with 1.5 WM indo- I in HBSS containing

cell-mediated B cell activation 1% FCS. The cells were warmed to 37'C and analyzed at

The murine C0)4" T cell clone, DI.G4.l (DIO), 200 cells/s on a dual laser flow cytometer (Ortho C)to-

(conalbumin-specific, laW-restricted) was considered a Th2 fluorograph. Westwood, MA) after addition of appropriate

clone based on its secretion, after stimulation, of IL-4 and mAb. Data were analyzed using commercially available

IL-5, but not IL-2 or IFN-y, and was maintained as de- software (Phoenix Flow Systems. San Diego. CA). The

scribed (27). Cytokine-rich SN from DIO T cells was ob- technique is capable of detecting a calcium response in as

tained as follows: resting DIO T cells (I X 10"/ml) were few as 0.3% of the cells analyzed.

Stimulated by adding irradiated (3000 r) spleen cells (5 X Resultsl05 /ml) from C3H mice (Tak-bearing) acting as antigen-prsetigclls inom he presce ofa g 100 ing/mI of cona in- Isolation of marginal zone and follicular splenic BPresenting cells in the presence of !100 pg/m l of conalbumnin c lsb l cr n c c l ot n

(Sigma). Cellfree SN was harvested 24 h later, aliquoted

and stored at -80°C until used. This T cell SN had, on MZB and FB can be distinguished on the basis of

average, 95 U/ml of IL-5 and 365 U/mI of IL-4, For Ag- their characteristic cell-surface phenotypes (13). MZB

,4'-

2740 Z

PRE-SORT ANALYSIS

103 Or 1.11

'-01 102 10) 10-4 101 102 102

4 'f- .e

Anti FcýRIl Anti-FcuR1i

POST-SORT REANALYSIS Uptake of j 'HjrdR 1han H3 in rl;. 0 (tsdc'. -ý ,rz,FczR~l- tgM(Marginai Zone) FceRlf- gM'(foificuiar) the'. incorporated 5 3-fol h-pber a!-u: r j rI ' l]_i R i.:

10 04 1 113 In respocrn.ýe to Li's Dhc lu'., erprie t rae

h ibhri td hN NILB rclaut~c to I Bi A. _r, :io bN-c ',1:c d,of (X8&dex used becau't l'~ker uptake c, of CilK h. M/H11.4- to I 6-foldi %kas o-bscrN cdu u,'r ~ i. an I 0

-X ~or 100) ni:ml of ao& dc % t Ii-ig 'tThe adJdtrof Ln? lan:01 10 I--onraminin T cell SN to ra&-dex-acti zated WH/ zr .1

F8 enhanced I'llijTdR uptake by hoih piovulat on'. of -:¶ol 102 102 10' 101 103 ¶04 HwsIovc'.er, 65-dex--acovs .1rd MsZ13 sill C shd'ted a. hr'.- r

Anti-FccRiI Antil-ftRII incorporation of ['HlTdR than o5i del-sumulatcd F h

FIGURE 1. Cell surface phenoty~pes and sorting of msZB (Fig,. 2).and F8. 8 cells were stained with FITC-anti-FctR-1l (83134). The differenrces in ['I1]TdR incorperration bcet'en MZBPE-polyclonal goal anti-IgM, and TR-anti-lad (,MKD61. Mac- and FB upon indluction A ith op-des\ AT to 100 ng'mi)l Nxcerophages were eliminated by forward and side catrgatir.g exen more marked than that ob;ser'.ed .3fver crSe¶. a~tu-Only ta" cells (B cells) are displayed. Upper panels: pr(-sort

analysis-MZB~~~~~ arNlM~¶cRl.P r l~ ation (Fig,. 3). ap-dcex-ac~tiated%1ZBinooaed5PtFciERII(boxed populations). MZB (FcERIVI- are also 124-fold lower amounts of l'H]TdR than FB dc-pendinSMlgDdWv and FB (FccRII ') are mlgD1`P'. mlgli staining of upon the dose of ap-dex used. The addition of 1000 Uimi1MZB was -4-fold higher than that for F8. Cells %within the of :L-4 % ariably enhanced f1IIJjTdR uptake by \I7B (0 C-.indicated boxes were isolated by sorting. Sorted cells were and 5.1 -fold) and FB (39- and 4.0-fold) activatedby lMotanalyzed immediately after sorting (low~er panel) and only 100 ng/ml of ap-dex. but the marked differences betwe enpurities >951,1 were acceptable for further study. the txso populations were still apparent (22- and 7.3-fold)

(Fig. 4). Visual inspection of MZB cultures clear)'. shorwed- .. .. ae I2 '1&~"FeR weea F rem~M an early phase ofcellularenlargcment by ala3rge proportion

-- ---------- elsi respons-e tap-dexwihw.clxwAadintermedia~c mIgDbrgh' FceR[l* (Fig. 1). MZB comprised or two later by a widespread loss of cell xiabilit,., as de--4% of the splenic B cells from either DBA/2 or BALB/c termined by try-pan blue exclusion (data not shon). All-mice. For all experiments reported herein highly purified though further analyses are required, this sugsgets that(Ž:95%) MZB and FB were obtained by staining cells with NIBacomentoetrthcllylenrspsetoFITC-anti-RFeRIl + PE-anti-lgM and then isolating them M arex comu tenti to prgente theog cel cyle in reviponedby electronic cell sorting. -dsbtfitoporsthugSpae.aeiend

by a low level of [3H]TdR uptake. Whether or not op-dex-

MZB xhiit loer rolferaiverat thn F inactivated B cells undergo apoptosis remains to be deter-MZBsexhsbto actltowe polfeatvex or at-exta Bi mined.

respnseto ativtio by ra-ex o apdexTo determine whether this diminished proliferative re-We determined that 10 ng/ml of cr8-dex induced maximal sponse to a8- and crp-dex reflected a general inability toproliferation ofT-depleted spleen cells, as assessed by 11H ]- respond to any mode of mlg-mediated B cell acti% ation. weTdR incorporation (data not shown). To assess the prolif- evaluated their responses to unconjugaled and Sepharose-crative rate of MZB and FB we stimulated them with 10 bound anti-lgD or anti-IgM. We further wished to compareng/ml of Ar-dex (TI-2-iike stimulus) and compared their the proliferative responses of MZB and FB upon stimula-incorporation of 131-UTdR with that obtained after aetiva- tion with a second TI-I like stimulus, STM. and after ac-tion with LPS (T1-I--like stimulus). Incorporation of [3'H1- tivation with an anti-CD3-activated Th2 clone (TD-likeTdR upon culture with medium alone was negligible for stimulus). Incorporation of ['HJTdR by MZB was alsoboth MZB and FB (Table 1). MZB showed an 8.1 -fold lower low-er than FB upon activation with unconjugated anti-lgM

J'Wrfljl of 4mmunology 24 i

LPS

S5N

~5C~.5 N U *13 a t 0

03 -c e x0 100

0 200^N4C: 0 4%O

3 H- hy,7 Cine (P1; 3

H- Y 7tC' e 1

F!CURE 2. 1iZB and FS proliferativ.e response to varying FIGURE 4. MZB and FB; pro: '1iiera,1 O ..VC--ji ct 8-de>x in the presence or absence oiTh2 SN. Sor*,ed [4 otdMBadF .eesn~ 0o 0

,NlZ5 and FB \%ere stimulated w.ith Qt3dex 0.0. 10, and 100 IL'ml Sortuyd !exZ and Fhe irest 1ra 0 o~rI 0,

ng.'rl) in the presence or absence of 1h2 010)t SN (final '[<4 I "HiTaiR vas added 48 h ai:er inititilon o1 (Uulu _,d

concentration 253o v/v) or LPS (20 pg/mb, ('H!TdR was l'HiTdR ultale tas neasured 16 h taxer,

added after 48 h and V'HITdR uptake asmeasured 16 hlater- Table 11

A.IZ6 prohitrate poorly mo response to un tjrlaivd. and

el C13 epha roe -bourrd nrli.1t

2~ 40000'11E 00 C.tgo 27,500 80.500

5FF I Seph 3,390 28.000

2 .M 196,000 t1~.O

Sorted MZB and FS 6 x 10"mi; ,,~ef c..iured rn I" pi~tc oi V-)

c p~'I of pok~ctonjt goal anti-nnou5e ISM or !1:0 ýCadi;M or Calso- V.Z5 an~d10' o, t 102FS 11.25 - 1 0Vfrt %%ere further s1it~rtl trvd :!J~l 1 'ýV Sý;rarroe ant't~.go10 0 10 11 ffi -Sephk. STIM 20 p*g'rn. or ani-C D3 -ao,.ated 010T (of~ .0.4 x 1O',nil

ag-dex (ng/ml) JH1TdR ýa added after 48 h )nd n:HTdR uptatke ssjs mva'ý'td lb ti late,

FIGURE 3. MZB and FB proliferative response to varyingdoses of csp-dex. Sorted MZB and FB were stimulated with the differences in anti-1--induced proliferation mnust reflectvarying doses of up-dex as indicated. I'HITdR wvas added 48 differences in the art-ivational pathw6,ay distal to this point.h after initiation of culture and I 'HlTdR uptake was measured16 h later. [3HITdR uptake by LPS-activated MZB was MZB and FB secrete Ig and undergo Ig class

136.00.pm nd PS-ctiate FBwas72.00 pm.s-witching to a comparable degree in response to T1-

and TD-like stimuli

~~~'.~~ -fl) nojgtdat-g 29fl) rat-g- Our Finding that MZB show impaired mlg.-dependlent B cellSepharose (8.3-fold) (Table 11). By contrast. IvZB and FB proliferation was in apparent contrast to the reports dem-showed comparable levels of [3H]TdR uptake in response onstrating that the MZB are the! predominant respondingto STM or anti-CD3-activated Th2 cells. cells in a TI-2 antibody response (1-3). To evaluate the in

vitro ability of these cells to secrete le and undergo Ig classThe owe proifeativ repones o MZ reltiv to switching in response to mig cross-linking we stimulated

F1b upon Ag-receptor cross-linkage is not associated cZ n Bwt ~e LSi h rsneo b

wtdiffaerences ia+cncstimuationso nrae sence of IL-4. MIZB secreted 1.4- to 2.7-fold greaterin~rcelularCa2

+ cncenratonsamounts of 1gM than FB although viable cell yields of MZ~B

TO assess the functional status of the Ag receptor on MZB were 1.3- to 2.0-fold lower than that for FB (Table I1l).and FB we measured the uptake of Ca2+ into the cell MZB and FB secreted comparable amounts of lgG I upon[Ca2 -]i immediately upon stimulation with varying con- addition of IL-4 to crh-dlex + IL-S-activated cells (NIZBcentrations (0.3 to 10 pg/mI) of unconjugated anti-IgM. 2750 ng/ml of IgG I (day 5 viable cells = 2.0 X 105(ml);Both the mean ICa2 ji and the percentage of responding FB =3000 ng/ml of IgG I (day 5 viable cells =3.0 X

cells observed upon activation with anti-laM were roughly 10',/ml)), MZB viable cell yields (0.60 X 105/ml) on day *comparable for MZ8 and FB (Fig. 5), This indicated that 4 were 7.2-fold lower than FB (4.3 X 105/nil) in response

2742 RESPONSES OF MARGINAL ZONE AND FOLLICULAR B CELLS

GaMa-induced [Co 2 ]responses Tible IV

Teme coutse of tgM and IgG3 secretion by A1ZB and 9B ,nresponse to LPS"

lI-t S :re.:"oi n~ l

FcERII- B Cells FccRIl + B Cells Day 2 Div 3 D.'v 4 04 S

G0oo 400 1 00 MZB 225 1,300 8.125 19,373-- '-3 F8 10 320 7,500 15.000

-4, 400 400 'igG3 S n &r

NZB <2.4 3.6 120 600s: 1 F3 <2.4 <2.4 64 240

200 20 ooS- . Soned MZB and F^ "ee culIdueed in th- p rteseno, M LPS Cv!ýte S% .. ,

oremoved from separate .ells on the das ?ndi(aied 1' i e'ermrac odoe

M 2 and IgG3 concentrations by ELISA0 2 3• 4 5 a ? 0 2 3 4 5. 6 7•

0,0 - 1CO Addition of IL-4 to LPS-stimulated MZB and FB led to• a0* 60 •comparable inductions of IgG1 (34) and IgE (35) secretion .

: ,," ,,. ,, "., 6 days after initiation of culture (Table V). Likewise IL-4

CV

: 20 20inhibited LPS-induced 1gM and 12G3 production (36) by0= 4o• ¢ : •i /L /• 0 .. IN-I... Z B and FB to a sim ilar extent (Table V ). Sim ilarly., LPS .-:

V 2 activated MZB and FB secreted comparable amounts of20lgG2a in response to FN-, (37) (Table VI).

. 00 . . . . . . 0 1 2-4. . . Because FB have been specifically implicated in TD re-

Time (minutes) sponses we wished to test whether MZB and FB differedin their ig isotype responses to activation with an anti-CD3-

FIGURE 5. Calcium response of MZB and FB to anti-lgM. activated Th2 clone. D10.G4.1 (DI0). Dl0-activated MZBSorted MZB (FcERII-) and FB (FccRIll) were loaded with

indo-1 and stimulated with polyclonal goat anti-mouse IgM secreted 6.7-fold more IgM and comparable amounts ofi(0.3 to 10 pg/ml). IaGl relative to similarly activated FB (Table VII), indi-

cating that both populations could secrete Ig and undergoTable It Ig class switching in response to T cell activation.Time course of IgM secretion by MZB and FB in response toat8-dex + IL-51

Discussion1gM Secretion (ng/ml) Viable Cells (xlO I ml)

The requirements for induction of TI and TD responses areDay Day Day Day Day Day Day Day

2 3 4 5 2 3 4 5 different. Thus, responses to TI, but not TD, Ag can be

elicited in the absence ofT cell help (38, 39) but only poorlyMZB 34 160 575 800 1.3 2.1 3.0 0.5

FB 25 60 220 450 2.2 3.5 4.0 1.0 in neonatal mice (40) or in adult mice that have been sple-

"Sorted MZB and FB were cultured with a--dex (3 ng/mli + IL-5 (150 nectomized (41, 42). Recent investigations have demon- 1.U/ml). Culture sN was removed from separate wells on the days indicated for strated that these differences in activation requirements fordetermination of 1gM concentrations by ELISA and viable cell numbers by Ttrypan blue exclusion. TI and TD Ag may reflect the fact that different subsets of

B cells respond to these Ag. B cells that localize to theto ap-dex alone, confirming that MZB and FB were ade- splenic marginal zone have been reported to be the pre-quately sorted. In additional experiments it was observed dominant population responding to TI-2 Ag (1-3), whereasthat neither MZB nor FB secreted detectable Ig when cul- B cells that localize to the follicular zone are the cells re-tured in medium or IL-5 alone (data not shown). sponsive to TD Ag (4). The studies that have been reported

We further evaluated the ability of MZB and FB to se- could not discriminate whether the differences in responsescrete Ig and undergo Ig isotype switching in response to a of these populations reflected intrinsic differences in theTI-I (LPS) and TD-type stimulus. LPS stimulated compa- responsiveness of the B cells themselves or whether the),rable amounts of IgM and igG3 (33) by MZB and FB after reflected the activity of different interacting cells in these5 days in culture, although MZB secreted 22- and 4, 1-fold compartments. We therefore undertook these studies to in-more IgM than FB after 2 and 3 days of LPS stimulation, vestigate this point.respectively (Table IV). The IgM response of MZB 2 and The difference in proliferation between MZB and FB3 days after LPS stimulation represented only 1.2 and 6.7% upon Ag-receptor cross-linkage was observed using a num-of the response observed after S days of LPS activation. The ber of distinct modes of Ag-receptor ligation includingkinetics of the LPS-induced igG3 response were compa- dextran-conjugated anti-Ig, Sepharose-linked anti-Ig. andrable for MZB and FB. high doses of unconjugated anti-Ig. Thus, this difference

logo II 14Z

10 um3I of Immu nolog 274

Table Vmediated signal transduction because anti-IgNI stimulatedIg I .sotype production by MZB and FS stimrulated with L PS or comparable elevations in [Ca2 ý in both NlZB and FB. TheLP5 + JL-4' imptired mig-mediated proliferative response t, as most ap-

l~ ecetn &'~lparent when anti-lgM was used as the stimulus. Although 1

1gMrB _________ proliferative responses of MZ8 to anti-IgO Aere clearly_____________ 3_____ ______ inferior to that of FB celthey were nonetheless hizher

NiZa FI mZ 8 F MNZ8 F 5 than anti -1g\l-sti mulIated responses.58,70 1,75 700 oco 190 <12Although the mnechanism. underly ing the poor prolifer'

LPS+)L-4 3,375 500 90 241 3,750 1,200 ative response of MNZB to Ae-receptor cross-linkage is un-lgC~aknown. the MZB response is reminiscent of that observed

MZB B MZ ~ M8 ~upon Ag-receptor cross-linkage of immiature 8 cells. Thus,anti-le cross-linkage of either neonatal B3 cells or WVE H I

* LS 73 21 70 30 <6 <6 231. a INymphoma with an immature phenot~pe. fails toLPS~L-4 40 <4 3 <2 1,40 1500 stimulate proliferation, but instead induces apoptos'is

Sorted mZB and 16 were culturtd in the prese-nce of IPS (20 PSrnl an/rElrztin(34) hitsascaedAhha pK, ýýut 1C)0000 U/mt~ of IL.-4. Culture SN was removed afe 6 da~ fo n/rtrrzton(34) hsi sscae iha pdeTho oi tg isot',pe Concentrations by EUISA. parently normal calcium response, as %ke ha-,e obser\ ed for

MZB, but a failure to activate PKC and the PKC-link-edTable V1t* Ig isotype production by MZ8 and F6 stimulated with LPS genes egr- I and c-fos. Cross-linkage of mlgD on WVEHI

or LIPS + (FN-y" 23 1, which wxas induced to express ml-D throu-h IcE) eenetransfection, did not result in the delivery of a negati'e

t s ecrtio (n'n{sig'nal in contrast to that observed for m~gNI (47 l WhetherISM IgG2o IgC2a this relates to the more profound differences in proliferation

MZ3 F8 MZB F8 MZB .-B between MZB and F13 when stimulated by ap-dex, asLPS 06,00 10,00 77 52 310 410 opposed to cr-dex. is of interest. Although NIZB. like im-

Al LPS+IFN-I' 103,125 93,750 215 ¶05 1,250 2,050 mature B cells, express an ilM brNI~i9mleD""'FcERIlP~Sorted MZB and FB were stimulated with LPS (20 ps'mll with or without phenotype. the NIZB has been shown to be a mature,

10 U/mt of IFN--y. Culture SN were removed 6 days later for measuremnert of noncycling cell (10).Ig iotye cocenratins y ELSA.The relative inability of MIZB to proliferate in response

Table VII to mlg cross~linking stimuli is not surprising because it islgM and tgG production by MZB and F6 stimulated with an known that polysaccharide Ag stimulate poor, if any, an-

ntiCD3acrvard T2 con&ait-inestic (memory) responses (6-8), which depend heavily,

Ig ScrejonIngm~lon clonal expansion of Ag-specific B cells. Furthermore, inTD responses Ag-mediated selection of high av idity elIS gGl with resultant proliferation and expansion of this popula-

MZ8 FB MZ15 F3 tion leads. over time after immunization, to increasing avid-Med 56 <10 <1.2 <1.2 it5' of Ag-specific antibodies (4); this response is not seenaCO3.Th2 6,000 900 2,400 1,350 in response to T! A- perhaps in part because of the relativeSorted MAZB and FS(1.25 x 10O1mll were stimulated by D10 Th? cells M0. inability of these Ag to induce clonal expansion. Thus, the

X 105/ml) in the presence of plate-bound anti-CD3 antibody. Culture SN wkasremoved 6 days later for measurement of IgM and tgGT concentrations by antibody response to TI ; -a primarily reflects their abilityELISA. to rapidly stimulate B cells to secrete Ig in the face of

limited B cell proliferation and in the context of limitingancillary help.

did not depend upon whether or not the anti-Ig stimulus was In contrast to TD Ag, TI-2 Ag stimulate the productionTI-2-like (i.e.. a8-dex or api-dex). Similarly, the starting of J-M and cG03 (5). This is observed despite the recentcell density did not appear to influence the differential pro- observation that TI-2 Ag rapidly localize in the splenic fol-liferative response of MZB and FB because cells stimulated licle upon their injection (48) and thus presumably arewith unconjugated anti-Ig were plated at high density available to interact with FB. The basis for this functionalwhereas cells stimulated with dextran- or Sepharose-anti-Ig dichotonmy is unknown. Our observations that Ig secretionvW"e cultured at a relatively low cell density. Although and Ig isotype switching induced by cytokinescin combi-NlZB and FB express different levels of mlgM and mlgD. nation with either a5-dex, LPS, or T cell activation werethe differences in anti-Ig-induced proliferation could not be relatively comparable for MZB and FR suggests that theaccounted for simply on this basis because similar differ- intrinsic properties of MZB and FB do not explain theirtnces in proliferation were seen over a wide range of anti-Ig differential response patterns in vivo. The data in this manu-concentrations. The impaired proliferative response of script demonstrate that differences in immune responses ofMZB to anti-fgM antibody does not reflect absent Ig- MZB and FB more likely reflect differences in their his-

2744 RESPONSES OF MARGI<CNAL, Z(\i A,"D Wi L ICLLAR 8 (11LiS

tooi iiu, which plays an imporiant role in influencing ~ t-~aici odsrn BW8 cells responses. atton. J1. luInlmlol 140-3364

15. Pe~anha. L_ NI. 1'.. C. MI Snapper. F D. Firikel:nan. and I .Mlond. 1991. Dextran-coi 6,eaicc arti-ig anutbcdlm as

Acknowledgments model for T cell-independent t'.pc 2 an:ln n~culatton of I,- secretion in vitro. 1. Lý :nphc'kirie d~cpcerd teec j.

We thank Hideko Yamraeuchi and Michelle Luogo for escellent technical Imuitnonl. 146,833.assistance. Dr. Richard Hodes (National Institutes of Healih. Bethesda. 1.Sapr .N. .N. ,P~nt.A0Loe nMD11) for rtL-5, Dr.Alan Lesinec\IonsantoCo.St. Louis.MO) forrlL-4. 6 Sondper .199 L. Mla~. siThn Pcaisa A cr caline dc dand Dr. Fred Finkelman (USUHS, Bethesda. MD) for pros ision of ini-munoreaeettts. the nature of the B cell atis,:tor. in ad-dition to ilit presc:Ž:e

of IL-4, J1. lmmunorl. 147:1163.

1 7 . S n a p p e r . C .M .. T . MI N lc l n t) ,e . R . Mla n d k er . L , \ 1 PReferences canha. F. D. Finkelrnan, A. Lee. arid I. J. %lond 19), 1,.1. Lane. P. J.. L. D. Gray, S. Oldfield, and 1. C. MI. MiacLennan. dcino g3sceinh n r~o ~c o 96 ifrne ntercuteto i~nBclsit

antibody responses to thymus-dependent and thymus- independent type 2 antizcenc. J. Fy 'p .fd. 175:1,46,independent type-2 antigens. Eur1. I Inimutia 16:1569. 18, Zitron. I. MI.. and B. L Cle% tncer, 19SO. Reguation (,f rwr:ný

2. MiacLennan, 1. C. M.. D. Gray. D. S. Kumararatne. and H. Bcel hog ufc mnn bln ~ncc.i eBazin. 1982. The lymphocytes of splenic marginal zones: a aniBntbdthtnucsaltp-ecfcpier..distinct B-cell lineage. Imntumol. Todav 3.305. 1. Ekp. Med. 152.1135.4;

3. Humphrey. J. H.. and D. Grennan. 1981. Tolerogenic or i. 1.Jlu.NI . .H essr n .Hata.I8 smuno'genic activity of hapten-conjugated polsahris duction of resting B cells to DNA synthetsiN by ,,oluý:e M.On-l

-orrelated with cellular localization. Fur. I. Inintwtol. 11:'ý2/2 oclonal anti-imnmunoglobulin. For J. hilntotol. 14-;75!

4. Liu, Y.-J..G. D. Johnson. J. Gordon, and I. C.MN. MacLennan. 20. Goroff. D. K ..A. Stall, 1.J1 Nlond, and F. D. Finkelman I91~1992. Germinal centres in T-cell -dependent antibody respons- In . itro and in %i'. a B lymplhocyte acti'.ating propertI es ofes. lntmiunol. Todav 13.-)77 monoclonal anti-b antibodies. 1. Detcrmina.nts of B

5. Perlmutter, R., MI. D. Hansburg. D. E. Briles, R. A. Nicolotti. lymphoerte-actisating properties. J. 1,nitionl 136:23482and J. M. Davie. 1978. Subclass restriction of murine ant;- 2 1. Rao, MI.. W. T. Lee. and D. H, Conrad. 1987, Charac-teizwanc

carbhydrte atibdies 1. mrntol.121:66.of a monoclonal antibody directed azainst the mL-ine B h'.m

6. Baker. P. I1., P. NV. Stashak, 0. E. Amsbaugh, arid B. Prescott. phocyte receptor for IgE. J. brmmtol. 138:1845.1971. Characterization of the antibody response to type 3 22. Kappler. ). W_. B. Skidmore, 3. White. anid P. Marrack 19S 1.pneumococcal polysaccharide at the cellular level. 1. Dose- Anti.-en-inducible. H-2-restrickli. interleukin-2-productng Tresponse studies and the effect of prior immunization on the cell hybnidomas: lack of independent antigen and H-2 rec-magnitude of the antibody response. Imtmunology 20:469. ognition. J. Ftp. Med. 153:1198.

7. Andersson, B., and H. Blomrgren. 197 1. Evidence for thymus- 23. Unkeless. J. 1979. Characterization of a monoclonal antibodyindependent humoral antibody production in mice against directed against mouse macrophage and l,.mphoc%,te Fc re-polyvinylpyrrolidone and E. coli lipopolysaceharide. Cell. ceptors. J. Ftp. Mfed. 130:580. ZItwnunoi'. 2.44 . 24. Leibson. Hi. H.. P. Marrack. and .1. W. Kappler. 1%51. B cell

8. Braley-Mullen, H. 1978. Antiggen requirements for induction helper factors. 1, Requirement for both intericukin_2 and an-of B-memory cells: studies with dinitrophenyl coupled to other 40.000 mol. wt factor. J. Ftrp. Aled, 154:1681.4_T-dependent and T-independent carriers. J. E~xp. Mled. 147: 25. Rabin. E. NM.. J. Ohara. and W. E. Paul. 1985. B-cell sti;mn-1824. ulator-y factor I activates resting B cells. Proc. Natl. Acad. Sci.

9. Kumararatne, D. S.. H. Bazin, and!1. C. MI. MacLennan. 198 1. USA 82:2935.Mlarginal zones: the major B cell compartment of rat spleen,.. 26. DeFranco. A. L., F C* Pas eel- R. Asofskv. and W. E. Pau!Eir. I. Immunol. 11:858. 1982. Frequency of B lymphoqýtes responstse to an;t-

10. Kumararatne, D. S.. and 1. C. M. MiacLennan. 198 1. Cells of immunoglobulin. J. Exýp. Med. 155:1523.the margiazoeothspenaelmhctsdrvdfm 27. Kaye, J_. S. Porcelli. J. Tite. B. Jones. and C. A. Janesxay. Jrrecirculating precursors. Fur. J. lmmtunol. 11:865. 1983. Both a monoclonal antibody' and anti-sera speciftc for

11. Stein, H.,.A. Bank. G. Tolksdorf. K. Lennert, H. Rodt. and J. determinants unique to individual cloned helper Tell linesGerdes. 1980. Immunohistologic analysis of the organization can substitute for antigen and antigen-preseniting cells in the 2yof normal lymphoid tissue and non-Hodgkin's lymphomas. J. activation of T cells. J. Exrp. Med. 158:836.Histoc/tem. C 'vrochern. 28:746. 2-8. Noelle. R., 1. 1. McCann, L. Marshall. and W. C. Bartleitt

12. Gray, D., 1. C. MI. MlacL-ennan, H. Bazin, and A. Khan. 1982. 1989. Cognate interactions between helperTcells and B cells. -tMigrant p*'8 and static p&b- B lymphocyte subsets. Fur J. Ill. Conitact-dependent. lymphokine-independent inductionImmunol. 12:564. of B cell cycle entry by activated helper Tcells. J. b'nrnwir'I.

13. Waldschmidt. T. J., F. G. M., Kroese, L. T. Tygrett. D. H. 143:1807.Conrad. and R. G. Lynch. 1991. The expression of B3 cell 29. Hodgkin, P. 0, L. C. Yamashita, B. Seymour. R. L. Coffman. "surface receptors. IIl.The murine low-affinity ISE Fc receptor and M. R. Kebry. 199 1. Membranes from both Th I and Th2is not expressed on Ly I or "Ly 1-like" B cells. Int. Irnutiunol. Teell clones stimulate B cell proliferation and prepare B cell'3:305. for lymphoki ne- induced differentiation to secrete 1g. J. Jut.

14. Brunswick, M., F. D. Finkelman. P. F. Highet, 3. K. Inman, Pinunol. 147:3696.H. M. Dintzis. and J. I. Mond. 1988. Picogram quantities of 30. Leo, 0.. MI. Foo. D. H. ~'hL. E. Samelson. and J. A

y 274journal of Immunology24

Bluestone. 1987. Identification of amonoclonal antibod sPC- and WV. E, Paul. 1977. Surface immunoglobulin D) as a funccific for a murine T3 poiylplitde. Proc. Vad). Acad. Sci. U'SA tional receptor for a subc'lass of B ls.mphocytes ImmimwtrdB84:1374. Rev. 37.89.

3I. Snapper. C. MI., and W., E. Paul. 1987. B cell stimulatory 41. Gray. D.. D. ChassouN. 1. C. M. NlicLennan, and H Bazinýfactor-] (interleukin 4) prepares resting murine B cells to 1985. Selective depression of thy mu s-independent anti-DNPsecrete IgGI upon subsequent stimulation with bacterial Ii- antibody responses induced bty adult but not neonata)sle-tropoly saucha ride. J. hnrinunol. 139:10. nectomy. C/in. Exp_ latmunol. 60:78.

2,June, C. H., and P. S. Rabinovitch. 1991. Measurement of 4.Alt .L.0 rna.adI .Hmhe 95 peiintracellular ions by flow% cytometry. In Current Protocols in dependence of the antibody response to thN mus-independent1,nnttlnology. Greene Publishing Associates and John Wiley (TI-?) antigens. Eta- J. li~nmurol. 15:508.,

3.Karnd y Sons ,Ne anork. R. Lare on.195Blypoted 43. Yellen, A., J. W. Glenn. V P. Sukhatme, X. Cao, and J. G.3~. earey. F. an A. . Lwto. 195. lyphocte if- Monroe. 1991. Sicnaliine through surface IeN1 in tolerance-

feetation induced by lipopolysaccharide. 1. Generation of uepieimurmuneBlmK c.Dsep:e.cel s thesizing four major immuno-lobulin classes. IJI. , n--cells s)tally regculated differences in transmembrane signaling in

inunol. 115:'671, 5plenic cells from adult and neonatal mice. J1 brumuzl, 1416:34. Isakson. P.C.. E. Pure. E. S. Vitetta, and P. H. Krammer. 198?. 1446.

T cell-derived B cell differentiation factor(s): effect on the Bnh ouL.EPA.Czae.nd.Snou190Atiisotspe switch of murine B cells.. J, Evp. Aed. 155:734. immunogilobulins induce death by apoptosis in WEHI-21S1 8Coffman, R. L.. and J. Carty. 1986. A T cell activ ity that lmhm el.EaiJ muo.2:45

enhances polyclonal Ig-E production and its inhibition by lypoacls uI inno-2:4* inerfrony. . Inutnol 13:94. - 45. Sainhou. P.. N, M-enr-Toulme. and P.-A. Caz-nase. 19S9.

36. Snapper. C. M.. F. D. Finkeman, and W. E. Paul. 1988. Dif- Mebae1Mcoslnig is not coupled to protein kinaseferential regulation of IgG I and leE synthesis by interleukin C anlaioinXE-31Blmh aces.urJu-4. 1. Exp. Mled. 167:183. no.1147

a7.Snppe. ,M. ad X. . Pul 197.Intrfrony ndB cell 46. Tisch, R., C. MI. Roifman. and N. Hozumi. 1988. Functionalstimulatory factor-I reciprocally regulate 1g isotyp prdc- differences between immunog~lobuilins %I and D expressed ontion. Science 236:944. the surface of an immature B-cell line. Proc. ?.r.Acadi. Sci.

33S. Sharon. R., P. R. B. McMaster. A. MI. Kask. J. D. Owens. and UA8:94WV. E. Paul. 1975. DNP-Lys-Ficoll: a T-independent antigen 47. Ales-Martinez. 1. E.. G. L. Warner, and D. WV. Scott. 1988.

V which elicits both 1gM and lgG anti-DNP antibody-secreting Immunoglobulins D and MI mediate signals that are qualita-cell. J.hn~nnol.114:585.tively different in B cells with an immature phenotype. Proc.

il. 9. Banihold, 0. R., B. Prescott, P. W. Stashak. D. F. Amsbaugh. Nazi. Acod. Sci. USA 85:6919.and P.J. Baker. 1974. Regulation of the antibody response to 48. Van den Ecitwegh. A. 1, MI.. J. D. Laman.MN. M. Schellekens.type III pneumococcal polysaccharide. Ill. Role of regulatory WV. J. A. Boersma. and E. Claassen. 1992. Complement-

Tcells in the development of an IgG and IgA antibody re- meiae folclrlclzto TidpnettPC- nsponse. J. Irmnunol. 112:1042. tioens: the role of marginal zone macrophages res isited. Euir

4-Mosier, D. E., 1. MI. Zitron, J. J. Mond, A. Ahmed. 1. Scher. J. bininunol. 22:719.

K AccSio

DTIC TAB

Dis~t r 1 0 art

j~v'sPi.

'DlatC 181