Proc. Natl. Acad. Sci. USAVol. 83, pp. 4474-4478, June 1986Immunology

Phorbol myristate acetate inhibits anti-IgM-mediated signaling inresting B cells

(protein kinase C/inositol phospholipid metabolism/intracellular calcium/B-cell activation)

JUNICHIRO MIZUGUCHI*, MICHAEL A. BEAVENt, JANE Hu LI*, AND WILLIAM E. PAUL**Laboratory of Immunology, National Institute of Allergy and Infectious Diseases, and the tLaboratory of Chemical Pharmacology, National Heart, Lung, andBlood Institute, National Institutes of Health, Bethesda, MD 20892

Contributed by William E. Paul, February 4, 1986

ABSTRACT Cross-linking the membrane immunoglob-ulins of resting B cells leads to activation asjudged by increasedinositol phospholipid metabolism, intracellular free calciumconcentration ([Ca2 k), and cell volume. Such activated B cellsenter S phase in the presence of B-cell stimulatory factor 1.Phorbol myristate acetate (PMA) is a potent inhibitor ofanti-IgM- and anti-IgD-stimulated B-cell responses. In B cellsconcentrations of PMA ranging from 0.1 to 100 ng/ml com-pletely inhibit anti-IgM-stimulated DNA synthesis and blockanti-IgM-stimulated increases in inositol phospholipid metab-olism and in [Ca2+],. Preincubation periods as short as 4 minblock these effects although longer preincubations are some-what more effective in inhibiting increases in [Ca21]1. Prein-cubation with PMA for 1.5 hr does not diminish expression ofmembrane IgM. This strongly suggests that PMA inhibitsresponses of resting B cells to anti-IgM by interrupting signaltransmission rather than by diminishing cross-linking of mem-brane immunoglobulin on B cells. In contrast to resting B cells,B cells activated in vitro for 29 hr show enhanced responses toanti-IgM in the presence of PMA.

Resting B cells can be activated through cross-linking ofmembrane immunoglobulins to enter the G1 phase of the cellcycle (1, 2). This activation appears to be mediated byincreased inositol phospholipid metabolism (refs. 3-5; J.M.,M.A.B., and W.E.P., unpublished data) leading to theproduction of diacylglycerol and inositol 1,4,5-trisphosphate(5). As diacylglycerol is an activator of protein kinase C (6,7), and inositol trisphosphate mobilizes Ca2l from theendoplasmic reticulum in some cell types (8-10) including Tcells (11), increases in intracellular free calcium concentra-tion ([Ca2+]i) and in protein phosphorylation upon cell stim-ulation are thought to result from this increase in inositolphospholipid metabolism. Indeed, some of the [Ca2+], in-crease in anti-IgM-stimulated B cells occurs in "Ca-free" (-6,M Ca2+) medium (J.M., M.A.B., and W.E.P., unpublisheddata), suggesting that inositol trisphosphate-mediated mobi-lization of intracellular calcium is also important in B cells.We will report elsewhere (J.M., M.A.B., and W.E.P.,

unpublished data) that the effects of anti-IgM can be mim-icked in B cells by the use of calcium ionophores and phorbolmyristate acetate (PMA), an activator ofprotein kinase C (12,13). The effects include an increase in cell volume and entryinto S phase in the presence of B-cell stimulatory factor 1(BSF-1). Monroe and Kass (14) similarly reported that PMAplus the calcium ionophore A23187 will cause resting B cellsto take up tritiated uridine, indicating that, like anti-IgM, thecombination of these agents causes B cells to enter the G,phase of the cell cycle.These findings are somewhat surprising in light of reports

that PMA by itself inhibits the anti-IgM stimulation of

[3H]thymidine uptake by B cells (15, 16). In this communi-cation, we examined the PMA-mediated inhibition of theanti-IgM stimulation of DNA synthesis in B cells and foundthat it is associated with, and presumably due to, the PMAinhibition of the increases in inositol phospholipid metabo-lism and in [Ca2+i that normally results from the cross-linkage of membrane IgM. Addition of A23187 allows pro-liferation in such cells, indicating that the PMA-inducedblock is proximal to effects of increased [Ca2+],. We suggestthat PMA regulates IgM signaling through phosphorylation ofa critical protein in the signaling pathway.

MATERIALS AND METHODS

Mice. Female BALB/c mice were obtained from theNational Institutes of Health Small Animal Section andCharles River Laboratories, Portage, MI, and were used at8-10 weeks of age.

Reagents. A23187, PMA, 4a-phorbol 12,13-didecanoate,and lipopolysaccharide (LPS) were from Sigma. myo-[2-3H]Inositol (17 Ci/mmol; 1 Ci = 37 GBq) and quin-2acetoxymethyl ester were from Amersham International.Percoll was from Pharmacia. Goat anti-mouse IgM heavychain-specific antibody (anti-IgM) (17) and the rat anti-mouseIgM heavy chain-specific monoclonal antibody (Bet-1) (18)were prepared as described. As a control antibody, a ratanti-dinitrophenyl monoclonal antibody (50C1, IgG) waskindly provided by J. Davie (Washington University, St.Louis, MO). Goat anti-mouse IgD (anti-IgD) was a gift fromFred Finkelman (Uniformed Services University of HealthSciences, Bethesda, MD). MAR 18.5, a mouse monoclonalantibody to rat kappa chain (19), was purified using a proteinA-Sepharose column and then labeled with fluorescein iso-thiocyanate. BSF-1, purified by reversed phase HPLC (20),was the gift from J. Ohara, National Institute of Allergy andInfectious Diseases.

Preparation of B Cells and Cell Culture. B cells wereobtained from a T-cell eliminated population of spleen cellsas described (21). DNA synthesis was determined by mea-suring [3H]thymidine incorporation (1 ,4Ci per well, 6.7Ci/mmol; ICN, Irvine, CA) after a final 6-hr labeling period.In the preincubation experiment, cells were initially culturedfor various times with anti-IgM at 106 cells per well in 24-wellplates (Costar, Cambridge, MA). After this preincubationperiod, cells were washed twice in Hanks' balanced saltsolution (HBSS), resuspended in medium, and counted. Forthe secondary culture, 5 x 104 viable cells, as determined bytrypan blue exclusion, were distributed into 96-well plates,and DNA synthesis was measured at various times.B Lymphomas. WEHI-231 (22), BALENLM 17 (BAL 17)

(23), and NBL (24) were maintained in culture medium

Abbreviations: BSF-1, B-cell stimulatory factor 1; [Ca21],, intracel-lular free calcium concentration; PMA, phorbol 12-myristate 13-acetate; LPS, lipopolysaccharide.

4474

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Proc. Natl. Acad. Sci. USA 83 (1986) 4475

described above. NBL was the gift ofDavid Scott (Universityof Rochester, Rochester, NY).

Fluorescence Analysis by Flow Cytometry. Cells were cul-tured for 1.5 or 18 hr with various stimulants and washedtwice with HBSS. Indirect fluorescent staining was carriedout as described (18).

Determination of [Ca2+]i. Normal resting B cells or B-lymphoma cells were loaded with 2.5 AM of the acetoxy-methyl ester ofquin-2 for 1-1.5 hr, and the fluorescence of thecellular suspension (2 x 106 cells per ml) was monitored witha Perkin-Elmer LS-5 spectrofluorometer at an excitationwavelength of 339 nm and an emission wavelength of492 nm.Experiments were done in 145 mM NaCl, 5 mM KCl, 1 mMNa2HPO4, 1 mM CaCl2, 0.5 mM MgSO4, 5 mM glucose, and20 mM Hepes (pH 7.4). Intracellular quin-2 concentrationswere 3-3.5 mM in resting B cells and 1-1.5 mM in BAL 17lymphoma cells calculated from an estimated cell volume of100 Aum3 for B cells and 620 ,Am3 for lymphoma cells. Aftereach experiment, cells were lysed with 0.05% Triton-X togive 100% saturation of quin-2, followed by the addition of0.5 mM MnCl2 to quench quin-2 fluorescence. Fluorescencevalues were converted to [Ca2+]i by use ofPerkin-Elmer 3600Computer data station. The computer program (programmedby Denise Fay Guthrie) was based on the formula of Heskethet al. (25).Measurement of Accumulation of Inositol Phosphates. Ac-

cumulation of inositol phosphates was measured by a mod-ification of the procedures of Berridge et al. (26) as describedby Beaven et al. (27).

RESULTSPMA Inhibits [3H]Thymidine Uptake in Response to Anti-

IgM. Resting B lymphocytes cultured with high concentra-tions of purified anti-IgM antibodies alone are stimulated totake up [3H]thymidine (1), while at low concentrations ofanti-IgM, entry into S phase requires the addition of BSF-1(28). PMA at concentrations from 0.1 to 100 ng/ml (0. 16-160nM) completely inhibited the response of resting B cells tohigh concentrations of soluble anti-IgM, to anti-IgM onSepharose beads plus BSF-1, and to low concentrations ofanti-IgM plus BSF-1 (Fig. 1). This indicates that PMA inhibitsthe stimulation of B-cell DNA synthesis that results fromanti-IgM cross-linkage of membrane IgM.

20A

-r

c 15'I-O

x

2

C 10

H

5I:_

10 100 0PMA. ng/ml

The effect of PMA on LPS-stimulated responses is morecomplex. At low concentrations, PMA partially inhibited[3H]thymidine uptake in response to LPS; maximal inhibitionof50-75% was obtained at 0.1-1 ng ofPMA per ml. At higherconcentrations of PMA, inhibition was not observed.B Cells Respond to PMA and A23187 in the Presence of

Anti-IgM or Anti-IgD. The failure of B cells treated withanti-IgM and PMA to respond does not appear to representa generalized inactivation of these cells. A BSF-1-dependentstimulation of DNA synthesis is obtained when the calciumionophore A23187 is added to B cells cultured with anti-IgMor anti-IgD and PMA (Table 1). Since the magnitude of theseresponses is equivalent to those stimulated by PMA plusA23187 in the absence of anti-immunoglobulin, this resultsuggests that PMA blocks the anti-immunoglobulin-de-pendent signaling pathway rather than more distal events inB-cell activation.PMA Blocks Anti-IgM-Stimulated Increases in [Ca2+], and

in Inositol Phospholipid Metabolism. Normal resting B cellsrespond to anti-IgM with rapid increases in [Ca2]i and ininositol phospholipid metabolism (refs. 3-5 and 29; J.M.,M.A.B., and W.E.P., unpublished data). To test the effect ofPMA on the [Ca2+]i increase, resting B cells, loaded with theCa-binding dye quin-2, were incubated with various concen-trations of PMA. After either 4 min or 1.5 hr, the cells werestimulated with anti-IgM and changes in [Ca2+]j, as measuredby quin-2 fluorescence, were determined. Cells pretreatedwith PMA for 1.5 hr were markedly inhibited in response toanti-IgM. Concentrations of PMA as low as 0.1 ng/mlconsiderably slowed the rate of [Ca2+]i increase and loweredthe maximal [Ca2+], increase (Fig. 2A). Higher concentra-tions more strikingly inhibited the [Ca2+]i increase. Thisinhibition could also be obtained with a 4-min preincubationperiod, although the degree to which the [Ca2+]J increase wasinhibited by any given PMA concentration was less than thatfollowing a 1.5-hr preincubation period. An analog of PMA,which does not activate protein kinase C, 4a-phorbol 12,13-didecanoate (30), caused no inhibition (Fig. 2B). This sug-gests that activation of protein kinase C by PMA is importantin this inhibition.

Increases in the accumulation of [3H]inositol phosphates inresponse to anti-IgM stimulation were also inhibited by PMA(Fig. 3). The concentration-response curve for this inhibitionwas similar to that for inhibition of increases in [Ca2+]. The

FIG. 1. Inhibitory effect of PMA onanti-IgM and LPS responses. (A) RestingB cells (2 x iOs cells per well) werecultured with anti-IgM (o) at 50 ,g/ml orLPS (e) at 50 ,g/ml in the presence ofPMA for 3 days. Cells were pulse-labeled

m with [3H]thymidine for the final 6 hr andharvested. (B) Resting B cells were cul-tured with LPS (e), anti-IgM (5 ,ug/ml)

O 1 1 10 (o), or anti-IgM-coupled Sepharose (A) in0.1 1 10 the presence of BSF-1 (10 units/ml) and

PMA.

Immunology: Mizuguchi et al.

4476 Immunology: Mizuguchi et al.

Table 1. B cells respond to PMA and A23187 in the presenceof anti-immunoglobulin

[3H]Thymidine uptake, cpm per culture

BSF-1 Anti-IgM/BSF-1 Anti-IgD/BSF-1

Medium 804 ± 119 53,448 ± 5,227 12,512 ± 478PMA 770 ± 57 1,522 ± 634 2,081 ± 98A23187 1,778 ± 1,021 2,219 ± 772 1,260 ± 125PMA/A23187 21,817 ± 2,634 20,461 ± 1,948 26,242 ± 3,066

Resting B cells (1 x 101 cells per well) were cultured with BSF-1(10 units/ml) alone or with BSF-1 and anti-IgM (50 Ag/ml) oranti-IgD (10 ug/ml) in the presence of medium, PMA (10 ng/ml),A23187 (100 ng/ml), or PMA and A23187. The plates were culturedfor 72 hr and pulse-labeled with [3H]thymidine for the final 6 hr.[3H]Thymidine uptake in medium in the absence of BSF-1 is 863 ±391 cpm. The results are means ± SD of triplicate cultures.

degree of inhibition was almost the same whether cells werepreincubated with PMA for 4 min or 1.5 hr. As accumulationof inositol phosphates was measured over a 40-min period,the effect ofPMA added 4 min prior to the initiation of culturecould have occurred over a substantially longer period, thusdiminishing the relative differences in the 4-min and 1.5-hrpreincubations.The inhibitory effects ofPMA on normal B-cell responses

also extended to the receptor-mediated responses of threeB-lymphoma cell lines (BAL 17, WEHI-231, and NBL). A4-min pretreatment of BAL 17 with PMA at 50 ng/ml causedan 84% inhibition in the anti-IgM-stimulated increase in[Ca2+]j. Similarly the anti-IgD- and anti-IgM-stimulated ac-cumulation of inositol phosphates was inhibited when BAL17 cells were pretreated with PMA for 4 min.PMA Does Not Inhibit Early Anti-IgM Signaling by Reduc-

ing Membrane IgM Density. The inhibitory effect ofPMA onanti-IgM-induced stimulatory events might result from adecrease in the amount of B-cell membrane IgM. However,direct measurements by flow cytometry showed no change inthe binding of monoclonal anti-IgM antibodies to B cellscultured with PMA for 1.5 hr (Fig. 4). B cells cultured withPMA for longer periods, 18 hr or more, had a 40-50%diminution in expression of membrane IgM (Fig. 4). Thus it

A300

240

180-

Q 120

Time, min

.

'50-

0

0

0.

0 0.1 1 10 50PMA, ng/ml

FIG. 3. PMA inhibits the accumulation of inositol phosphates inB cells in response to anti-IgM. B cells were loaded with [3H]myo-inositol by overnight incubation. Prior (1.5 hr) to addition ofanti-IgM(50 Ag/ml), they were placed in medium containing 10 mM LiCI.Some cells received PMA (o) or 4a-phorbol 12,13 didecanoate (U) atthat time while others received PMA (A) 4 min prior to the additionof anti-IgM. Accumulation of inositol phosphates was determined 40min after addition of anti-IgM. Radioactivity in cells, in the lipidfraction, and in the inositol phosphate fraction in 2 x 106 cells notstimulated with anti-IgM was 19,000 dpm, 9000 dpm, and 150 dpm,respectively. In the presence of anti-IgM, radioactivity in the inositolphosphate fraction was increased 4-fold. Percent of control wascalculated by a formula analogous to that used in Fig. 2.

is possible that the late diminution in membrane IgM densitymay contribute to the inhibition by PMA of later B-cellresponses to anti-IgM, such as cell volume increases andentry into S phase.Time Course of B-Cell Sensitivity to PMA Inhibition. The

finding that PMA inhibits signaling events in resting B cellssuggests that the PMA inhibition of B-cell responses to

B100

0

c a 5 g/mld -L;+ 50-

U 1 ,g/mlE X

1 10

PMA, ng/ml

FIG. 2. PMA inhibits the increase in [Ca2+I, induced by anti-IgM in resting B cells. Preincubation of cells with PMA for 1.5 hr was donewhile loading the cells with quin-2. The hydrolysis of quin-2 ester was not affected by PMA. The 4-min preincubation was carried out after quin-2loading was completed. Preincubated cells were stimulated with anti-IgM at 50,g/ml at 0 min. (A) [Ca2+], stimulated by anti-IgM in the presenceof PMA; cells were preincubated with PMA for 1.5 hr. Curves: a, medium; b, PMA at 0.1 ng/ml; c, PMA at 1 ng/ml; d, PMA at 10 ng/ml; e,PMA at 50 ng/ml. (B) PMA inhibition of increase in [Ca2+]i in response to anti-IgM. Results are presented as increase in [Ca2+]i as % control.% control = {Increase in [Ca2+]i (anti-IgM + PMA)/Increase in [Ca2]ij (anti-IgM)} x 100. Increase in [Ca2i], = [Ca2+]i (anti-IgM ± PMA) -

[Ca2+], (R), where [Ca2+]i (anti-IgM + PMA) is [Ca2+]i in the presence of anti-IgM (50 ,ug/ml) with or without PMA; [Ca2+], (R) is [Ca2+]i inthe absence of stimulants (=100 nM). *, 4a-phorbol 12,13 didecanoate; A, 4-min preincubation with PMA; o, 1.5-hr preincubation with PMA;c, anti-IgM.

Proc. Natl. Acad. Sci. USA 83 (1986)

Proc. Natl. Acad. Sci. USA 83 (1986) 4477

18 hr

Relative fluorescence intensity

FIG. 4. IgM density on B cells pre-treated with PMA for 1.5 (Left) or 18 hr(Right). Resting B cells (1 x 106 cells perml) were cultured with medium, PMA (50ng/ml) for 1.5 hr or PMA (10 ng/ml) for18 hr, harvested, and washed two timeswith HBSS. IgM density was estimatedby staining the cells with Bet-1 andfluoresceinated MAR 18.5.

anti-IgM should be most striking during the initial, anti-IgM-dependent phase of this response (31). The period of B-cellsensitivity to PMA was examined by culturing resting B cellswith anti-IgM for 53 hr and measuring [3H]thymidine uptakeduring the last 6 hr of this culture period (Fig. 5A). PMAcaused complete inhibition of [3H]thymidine uptake onlywhen added at the initiation of the culture and caused partialinhibition when added as late as 12 hr after the initiation ofculture but, thereafter, PMA caused a modest enhancementof the response to anti-IgM.We tested the relationship between anti-IgM dependence

and PMA sensitivity by preincubating B cells with anti-IgMfor various times, washing them, and then culturing themwith or without anti-IgM in the presence or absence ofPMA(Fig. 5B). B cells preincubated for 20 hr failed to enter S phaseunless anti-IgM was added again. However, PMA failed toinhibit the stimulatory effect of anti-IgM on these B cells.Indeed, B cells preincubated with anti-IgM for 29 hr stillrequired the addition of anti-IgM to enter S phase, althoughat later times they became anti-IgM independent. B cellspreincubated with anti-IgM for 29 hr were actually stimulatedby PMA alone, and PMA enhanced responses to additionalanti-IgM. These results indicate that anti-IgM-stimulatedresponses become resistant to PMA-mediated inhibitionduring the progress of B cells through G1.

DISCUSSION

Treatment of small B cells with PMA blocks their respon-siveness to anti-IgM antibody as measured by changes in[Ca2+]j, inositol phospholipid metabolism, and synthesis of

x

EL.

com

._

:-_

A B

10 20 30 40 20 24

Time of addition of PMA after Duration oinitiation of culture, hr

DNA. Thus, both early and late events of the receptorcross-linkage stimulation ofresting B cells are blocked by thisagent.B cells treated with PMA, although unresponsive to anti-

IgM, can be stimulated to enlarge and to enter S phase by theaddition of the calcium ionophore A23187. This suggests thatthe elevation of [Ca2+]i and activation of protein kinase Cachieved by the joint use of these two agents bypasses theinhibitory effect of PMA on the B-cell activation that nor-mally results from cross-linking membrane IgM. Indeed, thisresult implies that the PMA inhibits a step in the B-cellactivation process proximal to elevation of [Ca2+]i andstimulation of protein kinase C.The conclusion that PMA exerts its blocking activity by

inhibiting signal transmission is also consistent with theresistance of B-cell responses to LPS at high concentrationsof PMA (10-100 ng/ml), since the LPS activation of B cellsis not associated with increases in [Ca2+], or inositol phos-pholipid metabolism (refs. 4 and 5; and J.M., M.A.B., andW.E.P. unpublished data). Furthermore, once anti-IgM-stimulated B cells have entered a phase in the cell cycle whenthey no longer require anti-IgM to complete the G, phase andto enter S phase, PMA no longer inhibits the DNA synthesisresponse.PMA has been shown to cause inhibition of increases in

[Ca2+]J in other ligand-receptor systems (refs. 32 and 33;M.A.B., unpublished data). It has been suggested thatprotein kinase C, activated by PMA, phosphorylates a

protein that is critical to the signal transduction in these cells.Indeed, it has been reported that PMA causes phosphory-lation of the a, adrenergic receptor in smooth muscle cells and

FIG. 5. PMA inhibits the anti-IgM response ofresting B cells butenhances entry into S phase of Bcells preincubated with anti-IgM.(A) Resting B cells (1 x 101 cellsper well) were cultured with anti-IgM at 50 Ag/ml and PMA (o) at 1ng/ml was added at various timesafter initiation of culture. Cellswere pulse-labeled with [3H]thy-midine for 47-53 hr and harvested.As controls, cells were culturedwith anti-IgM (o), PMA (A), ormedium (*). (B) Resting B cells (1x 101 cells per ml) cultured withanti-IgM (50 4g/ml) for varioustimes were harvested and washedtwice with HBSS. In the second-ary culture, cells (5 x 104 cells perwell) were stimulated with medi-um (o), PMA at 1 ng/ml (e), anti-

29 42 IgM gg/ml (o), anti-IgM

plus PMA (m). Cells were pulsed)f primary culture, hr with [3H]thymidine for 47-53 hr

after initiation of primary culture.

6

_ 133

671

1.5 hr

1 10

Immunology: Mizuguchi et al.

4478 Immunology: Mizuguchi et al.

that this phosphorylation is correlated with uncoupling ofreceptor occupancy and inositol phospholipid metabolism(34). It seems likely that in B cells PMA activates proteinkinase C, which causes the phosphorylation of a key proteininvolved in the transmission of the signal that normallyresults from membrane IgM cross-linkage. What that proteinmight be is unknown at this time.

If, as we have suggested, PMA and A23187 acting togethermimic the action of anti-IgM on B lymphocytes, one may askwhy treatment with anti-IgM itself does not inhibit furthersignaling through cross-linkage of membrane IgM. Onepossible explanation is that anti-IgM causes a transientinhibition in further signaling through the activation ofprotein kinase C. Once inhibition is established, no furtherproduction of diacylglycerol should occur and protein kinaseC activity should then return to resting levels. Such cellsshould then regain sensitivity to signaling caused by receptorcross-linkage. By contrast, PMA effects on protein kinase Care presumably not terminated by the action ofthe kinase andthus should continue to maintain the inhibited state.Our results on the inhibitory effects of PMA on activation

of resting mouse B cells are in apparent conflict with theeffects ofPMA on human peripheral blood B cells and differquite dramatically from the effects ofPMA on T cells (35-37).Human peripheral blood B cells have been reported toincorporate [3H]thymidine in response to PMA (38, 39) andshow synergistic responses to anti-IgM plus PMA as well asB-cell growth factor plus PMA (40, 41). Interestingly, weshow that resting B cells pretreated with anti-IgM becomeresistant to the inhibition ofDNA synthesis by PMA by 20 hr,although they still require further stimulation by anti-IgM toenter S phase. The signaling pathway of "activated" B cellsmay, therefore, differ from that of resting B cells. Humanperipheral blood B cells may be in such an activated state.Indeed, by 29 hr after initial stimulation with anti-IgM, mousesplenic B-cell entry into S phase is actually stimulated byPMA with or without further presence of anti-IgM. Thissuggests that state of activation rather than parameters suchas cell size or degree of differentiation may be critical indetermining whether the anti-IgM-mediated signaling is in-hibited by PMA.

If the PMA inhibition of B-cell activation is due to thephosphorylation of a protein that is in a proximal position insignaling pathway ofB cells, it may be possible to isolate thisprotein and, in so doing, to identify at least one member of acomplex that has so far remained obscure.We thank Dr. J. Ohara and F. Finkelman for gifts of purified BSF-1

and goat anti-IgD antibody, respectively. We also thank AllesioPalini for expert assistance in flow cytometry, Denise Fay Guthriefor preparation of computer programs, and Shirley Starnes for hereditorial assistance.

1. DeFranco, A. L., Raveche, E. S., Asofsky, R. & Paul, W. E.(1982) J. Exp. Med. 155, 1523-1536.

2. Cambier, J. C. & Monroe, J. G. (1984) J. Immunol. 133,576-581.

3. Coggeshall, K. M. & Cambier, J. C. (1984) J. Immunol. 133,3382-3386.

4. Grupp, S. A. & Harmony, J. A. K. (1985) J. Immunol. 134,4087-4094.

5. Bijsterbosch, M. K., Meade, C. J., Turner, G. A. & Klaus,G. G. B. (1985) Cell 41, 999-1006.

6. Takai, Y., Kishimoto, A., Kikkawa, U., Mori, T. &Nishizuka, Y. (1979) Biochem. Biophys. Res. Commun. 91,1218-1224.

7. Kishimoto, A., Takai, Y., Mori, T., Kikkawa, Y. & Nishizuka,Y. (1980) J. Biol. Chem. 255, 2273-2276.

8. Streb, H., Irvine, R. F., Berridge, M. J. & Schulz, I. (1983)Nature (London) 306, 67-69.

9. Prentki, M., Biden, T. J., Janjic, D., Irvine, R. F., Berridge,M. J. & Wollheim, C. B. (1984) Nature (London) 309, 562-564.

10. Joseph, S. K., Thomas, A. P., Williams, R. J., Irvine, R. F. &Williamson, J. R. (1984) J. Biol. Chem. 259, 3077-3081.

11. Imboden, J. B. & Stobo, J. D. (1985) J. Exp. Med. 161,446-456.

12. Castagna, M., Takai, Y., Kaibuchi, K., Sano, K., Kikkawa,U. & Nishizuka, Y. (1982) J. Biol. Chem. 257, 7847-7851.

13. Niedel, J. E., Kuhn, L. J. & Vandenbark, G. R. (1983) Proc.Natl. Acad. Sci. USA 80, 36-40.

14. Monroe, J. G. & Kass, M. J. (1985) J. Immunol. 135,1674-1682.

15. Hawrylowicz, C. M. & Klaus, G. G. B. (1984) Immunology51, 327-332.

16. Hawrylowicz, C. M., Keeler, K. D. & Klaus, G. G. B. (1984)Eur. J. Immunol. 14, 244-250.

17. Sieckmann, D. G., Asofsky, R., Mosier, D. E., Zitron, I. M.& Paul, W. E. (1978) J. Exp. Med. 147, 814-829.

18. Kung, J. T., Sharrow, S. O., Sieckmann, D. G., Lieberman,R. & Paul, W. E. (1981) J. Immunol. 127, 873-876.

19. Lanier, L. L., Gutman, G. A., Lewis, D. E., Griswold, S. T.& Warner, N. L. (1982) Hybridoma 1, 125-131.

20. Ohara, J., Lahet, S., Inman, J. & Paul, W. E. (1985) J.Immunol. 135, 2518-2523.

21. Rabin, E. M., Ohara, J. & Paul, W. E. (1985) Proc. Natl.Acad. Sci. USA 82, 2935-2939.

22. Boyd, A. W., Goding, J. W. & Schrader, J. W. (1981) J.Immunol. 126, 2461-2465.

23. Kim, K. J., Kanellopoulos-Langevin, C., Merwin, R. M.,Sachs, D. H. & Asofsky, R. (1979) J. Immunol. 122, 549-554.

24. Ling, M., Livnat, D., Pillai, P. S. & Scott, D. W. (1985) J.Immunol. 134, 1449-1454.

25. Hesketh, T. R., Smith, G. A., Moore, J. P., Taylor, M. V. &Metcalfe, J. C. (1983) J. Biol. Chem. 258, 4876-4882.

26. Berridge, M. J., Downes, C. P. & Hanley, M. R. (1982)Biochem. J. 206, 587-595.

27. Beaven, M. A., Moore, J. P., Smith, G. A., Hesketh, T. R. &Metcalfe, J. C. (1984) J. Biol. Chem. 259, 7137-7142.

28. Howard, M., Farrar, J., Hilfiker, M., Johnson, B., Takatsu,K., Hamaoka, T. & Paul, W. E. (1982) J. Exp. Med. 155,914-923.

29. Pozzan, T., Arslan, P., Tsien, R. Y. & Rink, T. J. (1982) J.Cell Biol. 94, 335-340.

30. Kikkawa, U., Takai, Y., Tanaka, Y., Miyake, R. & Nishizuka,Y. (1983) J. Biol. Chem. 258, 11442-11445.

31. DeFranco, A. L., Raveche, E. S. & Paul, W. E. (1985) J.Immunol. 135, 87-94.

32. Sagi-Eisenberg, R., Lieman, H. & Pecht, I. (1985) Nature(London) 313, 59-60.

33. Zavoico, G. B., Halenda, S. P., Sha'afi, R. I. & Feinstein,M. B. (1985) Proc. Natl. Acad. Sci. USA 82, 3859-3862.

34. Leeb-Lundberg, L. M. F., Cotecchia, S., Lomasney, J. W.,DeBernardis, J. F., Lefkowitz, R. J. & Caron, M. G. (1985)Proc. Natl. Acad. Sci. USA 82, 5651-5655.

35. Rosenstreich, D. L. & Mizel, S. B. (1979) J. Immunol. 123,1749-1754.

36. Koretzky, G. A., Daniele, R. P. & Nowell, P. C. (1982) J.Immunol. 128, 1776-1780.

37. Hara, T. & Man Fu, S. (1985) J. Exp. Med. 161, 641-656.38. Abb, J., Bayliss, G. J. & Deinhardt, F. (1979) J. Immunol. 122,

1639-1642.39. Aman, P., Gordon, J. & Klein, G. (1984) Immunology 51,

27-34.40. Bertoglio, J. H. (1983) J. Immunol. 131, 2279-2281.41. Suzuki, T., Butler, J. L. & Cooper, M. D. (1985) J. Immunol.

134, 2470-2476.

Proc. Natl. Acad. Sci. USA 83 (1986)