Membrane Depolarization of Human B Cells Follows Stimulation by Either Anti-/g or B-Cell Growth Factor, But Only Anti-~ Causes Cell Volume Changes

Wayne M. Yokoyama, Millie M. Chien, Susan E. Engardt, Susan W. Aguiar, and Robert F. Ashman

ABSTRACT: t iuman peripheral blood B cdls were separated from monocytes and T eells, depleted of nM1 cells by an anti-Leu 9 rosetting technique, and fractienated on discontinuous Percdl gradients to yield a highly purified, small, dense B-cell population. Throe cells responded to F (ab') 2 goat anti- tz at 10 and I00 tzglml with membrane depolarization (measured by immunofluomcence with 3,3'-dipontyloxacarbeeyanine dye) at I h, cell volume enlargement by 48 h, and modest thymidine incorporation by 72 h. They also responded to the 12-kd human B.cdl growth factor of Mai zd wit/, membrane depolarization, but not with cell volume increase. F(ab')2 anti-I x and B-cell growth factor together induced greater depolarization than was seen with either alone, but there was no synergy. The cell volume increase seen with F (ab')2 anti-lz was not increased by B-cell growth factor. Comparison of data analysis methods showed that wxan fluoreeeence intensity most readily detected significant depolarization. We conclude that in human B cells: (1) depolarization may be a "general response" to a variety of membrane stimuli, because F Cab')2 anti-gt and B-cell growth factor acting through different rueptors beth induce it, and (2) depolarization does not inevitably lead to cell volume increase.

ABBREVIATIONS BCGFi2 B-cell growth factor, 12 kd DMSO dimethyl sulfoxide BSS Hanks' balanced salt solution SRBC sheep erythrocytes FCS fetal calf serum PMC peripheral blood mononuclear MEM Eagle's minimal essential cells

medium

I N T R O D U C T I O N

For stimulated lymphocytes to undergo cell division or differentiate to effector cells requires a complex series of earlier activation events. There is evidence that for B-cell proliferation at least two classes of signals are involved: one via cross- l inking of B-lymphocyte surface immunoglobulln (Ig) and another from T-cell- derived growth factors for B cells that act both early and late. Although several

From the Departments of Internal Medicine and Microbiology, University of Iowa Cdlege of Medicine, and VA Hospital, Iowa City, Iowa.

Addms reprint requests to Dr. Robert F. Ashman, Department of Internal Medicine, E4OOD, GH, University of Iowa Hospital, Iowa City, IA 52242.

Received October 5, 1987; accepted December 15. 1987.

Human Immunology 21,15S-164 (1988) |~5 © American Society for HistocompetibiIity and Immunogenefics, 1988 0198-8859/88153.50

156 W.M. Yokoyaraa et al.

T-cell-derived growth factors have been described which act on human B cells, the notion that each factor acts on a discrete step in activation has been dispelled. One example is a 12-kd B-cell growth factor (BCGFt2) described and purified by the Maizei laboratory [1,2]. BCGFt2 is able to drive a few B cells into S phase on its own, and by thymidine incorporation it shows synergy with low concentrations of anti-Ig [2]. Its receptor is expressed at low levels on resting B cells and increases in expression after activation [1]. Although BCGFI2 has recently been cloned [3], its biological properties have been determined using partially purified material. BCGF,2 is usually assayed on human peripheral blood or tonsillar mononuclear cells depleted of T cells and macrophages. It also acts on certain B-cell fines, establishing a direct role of BCGF12 in B-cell activation. Little is known about the intracellular early activation events that it triggers.

Plasma membrane depolarization is an early event in lymphocyte activation associated with an increased rate of N~+-H + exchange and an increase in intra- cellular pH. Studies with anti-Fab-treated mouse spleen cells have shown the membrane potential to drop from - 70 meV to near zero by 30-60 rain, and then slowly return to baseline [4]. This change in ion flux must entail a rearrangement of molecules in the membrane. By means oftbe change in intracellular pH, it may potentially regulate the activity of many enzymes. Much later in activation, lym- phocytes undergo a protein synthesis-dependent enlargement in the volume of both cytoplasm and nucleus. We report here that BCGFt2 stimulates small, highly purified human B cells to undergo membrane depolarization, but not cell volume enlargement, whereas anfi-/z can stimulate both these events.

METHODS

Reagents The following reagents and media were used: 2-aminoethylisothiouronium bro- mide (AET, Sigma, St. Louis, MO); Hanks' balanced salt solution (BSS, Cancer Center, University of Iowa), heat-inactivated fetal calf serum (FCS, Hazehon Dutchland, Inc., Denver, PA); Ficoll 400 (Pharmacia, Piscataway, NJ); Hypaque (Winthrop, New York, NY); Hepes buffer (Flow Labs, McLean, VA); Eagle's minimum essential media with Earle's balanced salt solution (MEM, MA, Biopro- ducts, Walkersville, MD); Dulbecco's phosphate-buffered saline (PBS, Flow Labs); Sepbadex G-10 (Sigma); RPMI 1640 (Cancer Center); valinomycin (Sigma); gramicidin (Sigma); dimethyl sulfoxide (DMSO, Sigma); chromic chlo- ride (CrCls, Sigma); Percoll (Pharmacia); trichloroacetic acid (TCA, Fisher, Fair Lawn, NJ); mouse IgGt auti-Leu 12, and IgG2a anti-Leu 9, with nonspecific controls of the same subclass (Becton-Dickinson).

Cell Preparation

Fresh heparinized peripheral blood was drawn from normal human volunteers aged 21 to 46, who were interviewed beforehand in an attempt to exclude users of medication and persons with acute viral or chronic illnesses. The blood was subjected to Ficoll-Hypaque density gradient centrffugation [5] to yield periph- eral mononuclear cells (PMC). After three washes with MEM supplemented with 1% FCS and 20 mM Hepes buffer, the PbIC w~re passed with Hanks' BSS + 15% FCS through a prewarmed (37°(=) Sephadex (3-10 column. The non- adherent cells constituted the lymphocyte fraction, only 0.1 to 0.5% positive for the monocyte marker, nonspecific esterase [6]. T cells were then separated from non-T cells (B cells and "null cells") by AET-treated sheep erythrocyte (SRBC)

B-Cell Depolarization 157

rosette formation [7] and density gradient centrifugation. This non-T population contained about 20-50% Leu 12 or BI-I- B cells and 35 to 70% Leu I lc+ Leu 9+ null cells, with about 10% of cells staining with neither M3, Leu 4, Leu 9, Leu 12, or Leu 11c.

B cells were separated from "null cells" by a modification of the anti-lg rosette technique [8]. In the first step, non-T cells were incubated at 4°C with the mouse monoclonal antibody 4H9 (kindly provided by Dr. IL Levy, Stanford University [9]), which reacts with "null cells" and T cells, at the concentration necessary to completely block staining of non-T cells with fluorescein-conjugated anti-Leu-9 (an equivalent antibody from Becton Dickinson, Mountain View, CA) at 4°C, as detected by flow cytometry. For the second step, affiaic/-purified goat F(ab')2 anti-mouse Ig, absorbed to remove cross-reactivity w/th human Ig (TAGO, Burlingame, CA), was coupled to SRBC with 0.05% CrCI~. Ideal conditions for antibody coupling had been determined by a hemagglutination assay using swine anti-goat Ig (TAGO) and a reverse hemagglutination assay using mouse Ig [10]. These anti-monse Ig-coupled SRBC were mixed with the 4Hg-labeled non-T cells, then subjected to Ficoll-Hypaque densiw ceatrifugation as described above. After hypotonic lysis of residual red cells, the interface contained B cells, >95% positive by fluorescein-coningated Leu-12 staining and flow cytometry. The B cells were incubated overnight at 5 x 106 cells/mi in RPMI supplemented with 5% FCS, 10 mM L-glutamine, and 50 pg/migentamicin at 37°C in a humidi- fied atmosphere of 5% CO2 in air. The cells were then washed and placed on an isotonic discontinuous Percoll gradient consisting of 40, 50, 60, and 70% layers. After centrifiagation at 400g for 18 rain, the interfaces were harvested and washed. In preliminary experiments we recovered 65% of the cells at the 50- 60% Percol! interface ("small B cells" with a mean cell volume of 180/~b13), 24% at the 40-50% interface ("large B cells" with a mean cell volume of 225 /.tM3), and 7% above the 40% Percoll layer (discarded).

Depolarization Assay

Small, dense B cells from the 50-60% Pereoll interface were resnspended in RPMI and 1% FCS, with or without various concentrations of F(ab')2 fragments of affinity-purified goat anti-human/z (TAGO) and/or delectinated growth factor preparations partially purified from supernatants of PHA-stimulated human T cells according to the Malzel method [2] (BCGF, Cellular Products, Buffalo, N-Y). The active growth factor in such preparations has been shown to have a molecular weight of 12 kd. Both reagents had been previously dialyzed against PBS to remove sodium azide, then sterilized by filtration. Some samples of B cells were stimulated with 2 /zM gramicidin or 20/zM valinomycin, both K + ionophores, dissolved in DMSO. After 50 rain of stimulation, 0.07 pg/ml (final concentration) of 3,3°-dipentyloxacarbocyanine dye (kindly provided by Dr. John Carabier, NationalJewish Hospital, Denver) in DMSO was added. After an additional 10 vain of incubation (20 rain for gramicidin), flow cytometric analysis was performed. Ten thousand cells were routinely scored for fluorescence inten- sity and forward light scatter. Depolarization of the cell membrane results in a decrease in fluorescence intensity. Data are expressed as the mean scale value o f the log fluorescence histogram, or as the difference in scale volume between the test sample and the unstimulated normal control.

Cell Volume Assay Small, dense B cells were resuspended in RPMI + 5% FCS and incubated with or without F(ab')2 goat and-human/z and/or BCGF12 as described above. After 2

158 W.M. Yokoyama et al.

days of culture in a 96-well flat-bottomed plate (Costar No. 3596), at 106 cells well, the cells were subjected to Ficoll-Hypaque density centrifugation to re- move dead cells. Cells were harvested from the interface and analyzed in Isoton on a Coulter counter and Channelyzer (Coulter Electronics, Hialeah, FL). Ten thousand cells were routinely scored.

[ 3H]Thymidine Incorporation Assay Small, dense B cells were stimulated in RPMI and 5% FCS with or without F(ab')2 goat anti-human/~ and/or BCGFIz as described above. After 3 days of stimulation in a 96-well flat bottomed culture plate, at 2 × 106 cells/ml, 0.2 × 106 cells/well, 1 ttCi of [3H]thymidine (specific activity = 50 Ci/mmol, Amersham) was added to each well for a 6-h pulse period. With a semiautomatic cell harvester (Flow Labs), the cells were washed onto Fiberglas filter disks with saline, followed by 10% trichloroacetic acid. The dry filter disks were counted by liquid scintilla- lion counter (Beckman LS7500). Subsequent incubation and recovery from the 60-70% Percoll density interface did not change this purity. These dense B cells demonstrated characteristics of resting B cells: (1) small size (mean volume of 170/zm ~ by Coulter Channelyzer), (2) unit amount of DNA by propidium iodide staining, and (3) absence of staining by monoclonal antibody 5E9 [11], which binds to activated B cells (data not shown). These highly purified human B cells were used for all the experiments in this report.

RESULTS

Membrane Depolarization Large B cells contain a few cells with lower fluorescence intensity (depolarized) not seen in the small B cells (Figure 1A,B). Valinomycin (20/zM), a mobile K + ion carrier, induced relatively little change in mean intensity in either population, but the cells became more uniform in fluorescent intensity (Figure 1), a change much more obvious in the large than in the small cells (Figure 1B). By forming an ion channel for potassium [13], gramicidin (2 gtM) depolarized all the small cells and most of the large cells, as shown by the large decrease in fluorescence intensity (Figure 1). These contrasts between small and large cells suggest that the large cells contain a minor population recently depolarized in vivo (and, there- fore, refractory to depolarization), whereas the small cells do not. Gramicidin and valinomycin provide useful positive and negative controls for the depolarization assay.

F(ab')2 fragments of goat anti-human/t induced a decrease in fluorescence intensity (depolarization), although the change was not as great as with gramicidin (2/zM) (Tables 1 and 2). F(ab')2 goat lgG lacking anti-lg activity failed to induce depolarization (Table 1). The monovalent Fab' fragment produced by reduction and alkylation of the active F(ab')2 preparation also failed to depolarize B cells (Table 1). These features were also seen in a second identical experiment. We concluded that cross-linking of surface Ig induces depolarization.

BCGFLz alone at 10 -z ng/ml induced greater depolarization than any concen- tration of anti-/~ (Tables 1 and 2). Together anti-gt and BCGFI2 depolarized better than either did alone, but there was no synergy (Table 2). Two experi- ments were performed in which 100/zg of F(ab')z anti-/* was used. In these experiments, the unstimulated control had a mean scale value of 169, whereas at 100/zg/ml anti-gt alone gave 134, anti-/z + 10 -4 ng/ml BCGFI2 gave 116, anti-/t + 10 -3 ng/ml BCGFt2 gave 114, and anti-/z + 10 -2 ng/ml BCGF gave 91.

B-Cell Depolarization 159

EFFECTS O N M E M B R A N E P O T E N T I A L

OF G R A M | C I D I N A N D V A L m O M Y C I N

w ' I ' " I 'm"

A. J~M | t

| .

= / J~\

. ~ i . I

1 10 100

L O G F L U O R E S C E N C E I~TL~SI1N

F I G U R E 1 (A) Effects on membrane poten- tial of gramicidin and vallnomycin. Small dense B cells (Leu-9 rosette depleted, AET-SRBC ro- sette depleted, G-10 Sephadex oooadherent, Percoll separated) were stimulated with gramici- din (2 ~tM, .....), valinomycin (20 p.M, --) , or medium alone ( - - ) for 50 rain. The dye, 3,3'- dipentyloxacarbocyanine iodide was added for I0 rain (20 rain for gramicidin). Ten thousand cells were scored for flunresceace intensity by FACS IV. (B) Large B cells from the same Per- coU gradient were studied in a similar fashion.

Although two experiments cannot be ~3alyzed statistically, they suggest 100 /zg/ml of anti-p, to be only slightly more potent than 10/zg/ml.

Cell Volume Changes and Thymidine Incorporation Figure 2 shows an experiment conducted under conditions of cell concentration (2 x 105 purified small B cells in 200 $1) and medium composition identical to the depolarization assays in Figure 1, but extended 48 h to measure cell enlarge- ment on the Coulter Channelyzer. F(ab')2 anti-~t at 10/zg/ml causes conspicuous

TABLE 1 Cross-linking of IgM is required for anti-/t-induced depolarization ~

Stimulants Mean fluorescence intensity

None 86 Gramicidin (2/zg/ml) 30 Goat (Fab'h IgG anti-human/z (100 p.g/ml) 63 BCGF,2 (10 -2 n#ml) 46 (Fab')z IgG and-human/z (100 t~./nd)

+ BCGFI2 (10 -2 ng/ml) 44 Nonspecitic goac (Fab')2 IgG (100/zg/ml) 96

Fab' of goat IgG anti-human/t 92

• Representative of two identical experiments.

160

TABLE 2

W. M. Yokoyama et al.

Depolarization of purified B cells; titration of F(ab'h anti-v, and BCGF12 ~

BCGFz2 (ng lml )

0 10 -~ 10 -z

F(ab ' lz 0 165 157 116 °° A n d - ~ 0 z g , / m l ) 1 142 1 2 3 " 97"**

I 0 131" 126" 1 0 8 " "

* Mean scale values -+ standard error of the fluorescence histograms are recorded, which decrease when cells depolarize. Significance of differences from the value for unstimulated cells in four to five experiments is indicated as: ° = p .05 to.01, **p.01 to .001, and **" = p < .001. Gramicidin at 0.2 t.zg/ml produces a mean scale value of 60.

enlargement. BCGFzz at 10 -2 ng/ml neither causes enlargement on its own nor interferes with the enlargement induced by anti-tz (Figure 2). In this experiment the same ceUs were assayed at I h for depolarization. The mean fluorescence scale value was 132 with no stimulus, 113 with 10 v~g/ml anti-/z, and 93 with 10 -2 ng/ml BCGFtz, indicating the significant depolarization expected from Table 1.

After a 6-11 pulse with treated thymidine, purified B cells from the same preparation were assayed for thymidine incorporation into macromolecules at 72 h (Table 3, experiment 1). Anti-v. stimulated thymidine incorporation better than BCGF,2, and no synergy was evident. Because the original demonstration of synergy with BCGF]2 [1] used 1/10 the ceil density used in experiment 1, we performed the other experiment in Table 3. As shown in Table 3 (experi-

FIGURE 2 Cell enlargement response to anti-/z and BCGF. Small, dense B cells were stimulated with F(ab'h goat anti-human-/z (10/zg/ml), 10 -2 ng/ml BCGF,2, both, or medium alone. After 48 h, ceils were subjected to Ficoll-Hypaque centrifugation, and 20,000 of the interface cells were analyzed with Coulter counter and Channelyzer. The cell volume distribution is shown.

C E I L VOLUMi [ CHANGES iN H U M A N B CELLS

A . . . . BCGF

. . . . AnU-p

1 0 0 2 0 0 3 0 n 4 0 0

CBIJ. V N ¢~m "~1

B-Cell Depolarization

TABLE 3 Effect of F(ab')2 anti-p, and BCGFt2 on [3H]thymidine incorporation

161

IF(ab')2 anti-v, BCGFI2 (/.*g/nil) (ng/ml) crop/1.0 6 cells -+ SEM Stimulation index

Experiment 1 :2 x l 0 s small B cells/welP 0 0 4,300 ± 200 1 0 I0 -2 9,000 -+ 200 2.1

10 0 33,900 ± 200 7.9 10 10 -2 42,300 ± 1,000 9,8

Experiment 2 : 2 x 104 small B cells/well 0 0 530 -- 20 1 0 10 -2 640 -+ 290 1.2 3 0 830 -+ 210 1.6 3 10 -z 1,170 ± 140 2.2

10 0 490 ± 5 0.9 I0 10 -2 5.180 ± 500 6.0 30 0 570 ± 10 1.1 30 10 -2 3,660 ± 220 6.9

"Purified small B cells were cultured 66 h, pulsed 6 h with [ JH]thymidine, and assayed for 109~ trichlot-afetic acid* insoluble counts. Experiment 1 was performed at the cell density used in depolarization and cell volume experi- ments. Experiment 2 was done at the lower concentrations used previously to demonstrate synerg,/[1].

meat 2), we confirmed this earlier observation, showing synergy with I0 -z ng/ml BCGF when anti-p- is present at 10 or 30 p-g/ml.

DISCUSSION Optimal proliferation of B cells appears to require both interaction of ligands with surface Ig, and with receptors for T-cell-derived lymphokines [14]. The simplest model envisions surface lg (antigen receptor) and lymphokine receptor, each initiating a series of biochemical changes (activation events), whose conver- gence leads to DNA synthesi~ [151. Most .~mdies of B-cell activation have fo- cused on single events induced by particular ligand-receptor interactions; yet to understand the relationship between the series of events triggered by different ligands, it is advantageous to study multiple ligands and multiple activation events in carefully defined lymphocyte subpopulations under the same conditions. Thus, in this study of human B-cell activation, we have contrasted the effect of antibody to surface Ig {polyclonal F(ab')2 anti-t z] to the effect of a T-cell-derived B-cell growth factor (BCGF12), assaying an early event (membrane depolarization), an intermediate event (cell volume increase), and a standard late event (thymidine incorporation into macromolecules). These studies showed: (1) even though anti-p- and BCGF~2 appear to act through different membrane receptors [1], they nevertheless independently depolarize B cells (Table 1), illustrating the point that two activation pathwa3,s may converge on the same early event; and (2) that depolarization (for example, by BCGF12) does not inevitably lead to cell volume increase (Figure 2) illustrating that the sharing of earlier and later events between two activation pathways does not preclude distinctive differences in intermediate events. Furthermore, the occurrence of two events after ligand-receptor interac- tion does not imply that the earlier event is causally related to the later event. A

162 W.M. Yokoyama et al.

model of activation including multiple receptors, mukiple parallel pathways, and the convergence of multiple pathways upon important later events such as DNA synthesis is require,! to allow for such complexities [15].

Density gradient separation yields a small dense B-cell fraction of more uni- form size and fluorescence intensity (Figure 1), with characteristics of resting B cells, eliminating cells that have been activated in vivo long enough (1-2 days) to have increased in size. The presence ofpreactivated cells in peripheral blood and differences in activation properties of different lymphocyte populations justify more careful attention to cell purification than many earlier studies provided. Removal of "null cells" from the non-T population is especially important, be- cause BCGF~2 (but not anti-ix) also depolarizes these cells (Engardt et al, manu- script in preparation).

Such cells demonstrate responsiveness m anti-lz when examined by membrane depolarization (Figure 1, Table 1), cell volume enlargement (Figure 2), and DIqA synthesis assays (Table 3). The depolarization induced by anti-t* and BCGFz2 affected a majority of the ceUs, iudging from the reproducible observation that whole peaks shifted with little change in peak width. The cell volume changes also affect the majority of B cells (Figure 2). These data confirm in the human the results obtained by Cambier et al in the mouse [13]. Their results demonstrated that anti-~-induced B-cell depolarization, and the studies of DeFranco et al [ 16] in the mouse and Muraguchi et al [17] in man demonstrated that anti-ix induces cell volume enlargement. Anti-~-induced DNA synthesis has been inconsistently reported, perhaps because of differences in anti-lz preparations. Enhancement of anti-Ix stimulation by solid phase support (discussed in [17]), and the use of F(ab'h fragments both appear to work by avoiding the inhibitory influence of linking surface Ig to the Fc~/receptor [ 18].

The cyanine dyes are lipophilic cationic molecules that partition into mem- branes in such a way that they are rigidly held in the membrane when the membrane is fully polarized, and extruded when it is depolarized. Green fluores- cence intensity is greatest when the dye is membrane bound; therefore, depolar- ization leads to a decrease in fluorescent intensity per cell. Heterogeneity of the original fluorescence intensity profile is related to heterogeneity in cell size and activation status. If cells are preincubated too long in dye, their susceptibility to ligand-induced reduction in fluorescence intensity declines, partly because an increasing fraction of dye binds to internal membranes such as mitochondria. Wilson et al [4] have shown that cyanine dyes can be taken up by FACS tubing, thus producing a systematic error in a series of measurements, an effect that we attempted to avoid by performing the stimulus-free control at both the beginning and the end of each series. They also found that cyanine dyes are toxic to mouse B ceUs in the concentration range required for fluorescence detection, in the sense that the dyes themselves induce increased cation permeability. Fortunately, with 10-rain incubations using human B cells, we were able to demonstrate uniform fluorescence intensity (Figure 1) and depolarization only with reagents that cross-linked surface Ig (Table 1), although dye was present at the same concentration in all samples.

No interesting conchasions derive from the fact that our data show no synergy between BCGFI2 and anti-Ix for depolarization, whereas BCGFt2 has been shown by Maizel [1] and by us (Table 3) to produce synergy with anti-Ix for thymidine incorporation. Practical considerations dictate using a higher cell density for FACS experiments, and it appears that at this higher cell density, the synergy for thymidine incorporation also disappears (Table 3).

Because at the time of these experiments recombinant BCGF~2 was not avaib able, they were done with a BCGF preparation partially purified by pub-

B-Cell Depolarization 163

lished procedures [2], which by no means can be guaranteed to be free of other T-cell products. The fact that it depolarizes B cells in the picogrum concentration range is somewhat reassuring, but all the biological properties of BCGFt2 will need to be rechecked with the recombinant product.

ACKNOWLEDGMENTS The work was supported by NIH Grant GM36261. The authors thank Dr. Ronald Levy for monoclonal antibody 4H9, Dr. John Cambier for a starting supply of cyanine dye, and Dr. Shashikant Mehta for helpful discussions regarding ~he properties of BCGF12. We are also grateful to Mrs. Deanna OUendick for manuscript preparation and Mr. Donald Mower for technical assistance.

REFERENCES 1. Lachman LB, Maizel AL: Human immunoregulatory molecules: lnterleukin 1, inter-

leukin 2, and B cell growth factor. Contemp Top Mol Immunol 9:147, 1983.

2. Mehta SR, Conrad D, Sandier R, Morgan J, Momagna R, Malzel AL: Purification of human B cell growth factor. J lmmuool 135:3298, 1985.

3. 5harma S, Mehm S, Morgan J, Maizel A: Molecular cloning and expression of a human B cell growth factor gene i'~ Escherichia coll. Science 235:1489, 1987.

4. Wilson HA, SeLigmann BE, Chused 'I'M: Voltage-sansitive cyanine dye fluorescence signals in lymphocytes: Plasma membrane and mitochoodrial components. J Cell Physiol 135:61, 1985.

5. Boyum A: Isolation of mononuclear cells and granulocytes from human blood. Stand J Cfin Lab Invest 21 (Suppl. 97)'-77-89, 1968.

6. Chien MM, Ashman RF: Rapid separation of human monocytes and lymphocytes by Sephadex G-10. J lmmunol Methods 71:25-36, 1984.

7. PeHegrino MA, Ferrone 5, Dietrich MP, Rasifeld RA: Enhancement of sheep ted blood cell human lymphocyte rosette formation by the sulfhydryl compound 2-amino ethylisothiouronium bromide. Crib lmmunol Immunopathol 3:324-333, 1975.

8. Anderson KC, GriflinJD, Bates MP, Slanghenhoupt BL, Schlossman SF, Nadler LM: Isolation and characterization of human B lymphocyte enriched populations. I. Purifi- cation of B cells by immune rosette depletion. J Immunol Methods 61:283-292, 1983.

9. Link M, Wanke R, Finlay J, Amylan M, Miller R, Dilley J, Levy R: A single mono- clonal antibody identifies T cell lineage of childhood lymphoid malignancies. Blood 62:722, 1983.

10. Coombs RRA, Wilson AB, Erenin O, Gumer BW, Naegert DG, Lawson YA, Bright 5, MunrO AJ: Comparison of the direct antiglobolin rosetting reaction with the mixed antiglobulin rosetting reaction for the detection of immunoglobulin on lymphocytes. J Immuool Methods 18:45-54, 1977.

11. Kehrl JH, Muraguchi A, Panci AS: Differential expression of cell activation market's after stimulation of resting human B lymphocytes.J Immuno1132:2857-2861, 1984.

12. Roehm NW, Leibson HJ, Zlotnick A, Kappler J, Marrack P, CambierJ: interleukin- induced increase in la expression by normal B cellz. J Exp Med 160:679-694, 1984.

13. Monroe JG, CambierjC: B cell activation. I. Anti-immunoglobulin induced receptor cross-linking results in a decrease in the plasma membrane potential of murine B lymphocytes. J Exp Med 158:1589-1599, 1983.

164 W.M. Yokoyaraa et al.

14. Howard M, Paul WE: Regulation of B-cell growth and differentiation by soluble factors. Ann Rev lmmunol 1:307-333, 1983.

15. Ashman RF: Lymphocyte activation. In Paul WE (ed): Fundamental Immunology. New York, Raven Press, pp. 267-300, 1984

16. DeFranco A, Raviche E, Asofsky R, Pani "~VE: Frequency o rb lymphocyte responsive to anti-immunoglobulin. J Exp Med 155:152.3-1536, 1983.

17. Muraguchi A, KehrlJH, Butier JL, Fauci AS: Sequential requirements for cell cycle progression of resting human B cells after activation by anti-Ig. J lmmuno1132:176- 180, 1984.

18. Phillips HE, Parker DC: Cross-linking of B lymphocyte Fc receptors and membrane immunoglobulin inhibits anti-immunoglobulia blastogenesis. J Immuuol 132:627- 632, 1984.