CELLULAR IMMUNOLOGY 68,252-260 (1982)

The Suppressive Activity of Fey Receptors Is Not Related to Their T-Cell Origin’

LE THI BICH-THUY, C. SAMARUT, C. RABOURDIN-COMBE, AND

J. P. REVILLARD

Laboraroire d’immunologie, INSERM U.80 et CNRS-ERA 782, H6pital E. Herriot, 69374 Lyon Cedex 2, and Laboratoire d’immunologie cellulaire, Institul de Recherches Scientifiques sur le

Cancer, 94800 VillejuiJ France

Received October 23, 1981: accepted January 5. I982

Soluble receptors for Fc IgG (FcyR) were obtained by short incubation of various cell types in serum-free medium, and isolated by affinity chromatography on human IgG. The suppressive activity of this material was investigated in cultures of human peripheral blood lymphocytes (PBL) stimulated by extracts of Nocardia opaca as polyclonal activator. Addition of PBL- FcyR, at the third day of the culture, resulted in a marked decrease of the number of Ig- secreting cells, determined by a reverse hemolytic plaque assay, without diminution of Ig- containing cell numbers. FcyR, however, did not inhibit plaque-forming cells when added immediately or 3 hr before the assay. FcyR prepared from T-enriched or T-depleted unstim- ulated PBL, as well as FcyR released from neutrophils or from murine T-cell hybridoma displayed similar suppressive activities. The results indicate that the suppressive properties of soluble FcyR are associated with their capacity to bind the Fc part of IgG but not related to their T-cell origin.

INTRODUCTION

When stimulated with specific antigens or polyclonal activators, B lymphocytes differentiate into antibody-forming cells or immunoglobulin (Ig)2-secreting cells. Their maturation is controlled by various regulatory signals generated by accessory cells among which T helper (TH) and T suppressor (T,) cells have been most extensively investigated. Moretta and co-workers, using the pokeweed mitogen as polyclonal activator of human lymphocytes, have shown that in this model TH and Ts cells belonged to different T-cell subsets which could be characterized by the presence of receptors for the Fc part of IgM and IgG (FcyR), respectively ( 1, 2). The link between FcyR expression on a T-cell subset and its suppressive activity was supported by the demonstration that the binding of antigen-IgG antibody complexes (1) or aggregated IgG (3) to FcyR-bearing TS cells was required to

’ This work was supported by a grant from the DGRST (80.7.03.02) and UER-Faculte de Biologie Humaine, Lyon.

’ Abbreviations used: cIg cells, Ig-containing cells; E, sheep erytbrocytes; EAo, erythrocyte-IgG an- tibody complexes; FcyR, receptors for Fc part of IgG, Ig, immunoglobulin; NOC, extracts from Nocardia opaca; PBL, peripheral blood lymphocytes; PBS, phosphate-buffered saline; PFC, plaque-forming cells; PMN, polymorphonuclear neutrophils; S4B, Sepharose 4B; Tn, helper T cell; Ts, suppressor T cell.

252

OOOS-8749/82/060252-09$02.00/O Copyright 0 1982 by Academic Press. Inc. All rights of reproduction in sny form rcscrvcd.

SUPPRESSIVE ACTIVITY OF Fey RECEPTORS 253

trigger the suppression. Some polyclonal activators as, for instance, extracts of Nocardia opaca (NOC) (4) or Cowan I strain formalinized staphylococci (5) were shown to induce the differentiation of human B lymphocytes in the absence of TH cells but this activation remains under the control of Ts cells bearing FcyR, at least in the case of NOC (3, 4). Furthermore, in several models of antigen-specific activation of human B cells, Ts cells have been shown to belong to the FcrR- positive subset (6-8).

FcyR is not only a surface marker of some Ts cells but has also to be considered as a lymphokine with nonspecific suppressive activity. Indeed, Fridman and co- workers have shown that FcyR released from mouse alloactivated T cells (9) or T-cell lines ( 10, 11) could inhibit in vitro primary and secondary antibody responses as well as B-cell differentiation induced by polyclonal activators (reviewed in ( 12)). Similarly, FcrR released from unstimulated human lymphocytes were found to depress the generation of Ig-secreting cells in cultures stimulated by polyclonal activators (13). FcyR can be found on several cell subsets, including B, T, and nul lymphocytes, monocytes, and polymorphonuclear neutrophils (PMN). However, differences in receptor density, affinity, and subclass specificity have been reported among FcyR of these various cell subsets (14, 15). In the present study we have investigated the suppressive capacity for FcyR released from different cell sources. We report that the depression of human B-cell differentiation induced by NOC can be achieved not only with T-cell FcyR but as well with FcyR released from unstimulated T-depleted lymphocytes or from PMN.

MATERIALS AND METHODS

Cell sources and fractionation procedures. Peripheral blood lymphocytes (PBL) were separated from heparinized blood of healthy adults by centrifugation (25 min, 6OOg, 4°C) on Lymphoprep (Nyegaard and Co., A/S, Oslo, Norway). The frac- tionation into T-enriched PBL and T-depleted cells was performed by a double cycle of E-rosette formation with neuraminidase-treated sheep erythrocytes (E) and centrifugation on Lymphoprep as described elsewhere (3). Briefly PBL were mixed with E and rosette-forming cells were separated from nonrosetting cells by centrifugation on Lymphoprep. T-Enriched cells were recovered at the bottom after hypotonic shock lysis (NaCl 1.496 g/liter) and further purified by adherence on nylon wool according to Julius et al. ( 16). T-Depleted cells were recovered at the interface.

Two series of batches of FcyR were prepared from PBL or T-enriched or T- depleted suspensions obtained from two different donors. The percentages of T and B cells were determined by complement-dependent lymphocytotoxicity assay, using heterosera specific for T and B cells, respectively, as reported previously (17). The two T-enriched suspensions contained 96-97% T cells and no detectable B cells, whereas their T-depleted counterparts contained 7-5% T cells and 84-85% B cells. The two T-enriched suspensions contained 5 and 10% of cells bearing FcyR and the two T-depleted ones contained 40 and 45% of FcyR-bearing cells as revealed by a rosette assay using ox erythrocytes coated with rabbit IgG antibodies (18).

PMN were obtained from heparinized blood as previously described ( 18). Briefly the leukocytes were separated from the red cells in the blood by sedimentation on dextran-Isopaque and centrifuged on Lymphoprep. PMN were recovered in the

254 L6 THI BICH-THUY ET AL.

pellet and used after lysis of contaminating red cells by hypotonic shock. The contamination of the PMN suspensions by mononuclear cells never exceeded 2%. Two FcyR-negative cell lines were used as controls: Molt 4, a human T-cell line kindly provided by Dr. G. Lenoir (International Agency for Research on Cancer, Lyon, France), and MRCS, a human embryo fibroblast line (Bio Merieux, Lyon). Human red blood cells obtained from healthy adults were used in parallel as another source of cells lacking surface FcrR.

FcyR preparation. IgG were prepared from a pool of human A sera obtained from blood donor volunteers. Ig were precipitated with (NH4),S04 (40% satura- tion) and IgG purified by passage through a DEAE-Sephadex column. Purified IgG were covalently bound to beads of Sepharose 4B (S4B) as previously re- ported (13).

Soluble FcyR were isolated from supernatants of PBL, T-enriched PBL, T- depleted PBL, or PMN as previously reported (13). Briefly, cells were incubated overnight at 37°C in phosphate-buffered saline (PBS). Cell-free supernatants were obtained by passage through a 0.22-pm membrane filter and FcyR were allowed to adsorb on IgG-coated S4B ( 1 ml immunosorbent per 5 -1 O6 FcyR-bearing cells). FcyR retained by IgG-S4B column were eluted with glycin-HCl buffer (pH 2.8). Eluates containing FcyR were then neutralized to pH 7.2 with NaOH 1 M. Aliquots of 50 ~1 were stored at -70°C and thawed just before use. Effluents obtained before elution with glycin-HCl contained some FcrR indicating that the immunosorbents were saturated with FcrR. Supernatants from FcyR negative cells were treated similarly.

Murine immunoglobulin binding factor (soluble FcyR) was produced by a mu- rine T-cell hybridoma cell line (T2D4) and purified by affinity chromatography on rabbit IgG-coated column as previously described (11).

Cell cultures. Cultures were performed with PBL separated from defibrinated blood. This procedure was used in order to avoid platelet contamination without selective loss of B cells. Lymphocytes were put in culture in plastic tubes (Falcon Plastics, Oxnard, Calif., No. 2003) at a final concentration of 1.5 X lo6 cells/ml in RPM1 1640 medium (Eurobio, Paris, France) supplemented with 15% heat- inactivated fetal calf serum, glutamin (2 mM), and antibiotics (vancomycin 20 rg/ml, gentamycin 20 fig/ml, and amphotericin B 1 pg/ml). Unstimulated cultures were added with PBS- and mitogen-stimulated cultures with NOC (100 pg/ml) provided by Dr. Ciorbaru (Institut de Biochimie, Orsay, France). Three days later, the elution buffer previously passed on IgG-coated beads and neutralized to pH 7.2 was added to control cultures, whereas FcyR diluted (1: 100) in the same buffer were added to test cultures. At Day 7, viable cells were counted using the trypan blue exclusion technique. The plaque-forming cells (PFC) were determined using the reverse hemolytic plaque assay with Protein A-coated sheep erythrocytes des- cribed by Gronowicz et al. (19). Briefly Protein A (Pharmacia Fine Chemicals, Uppsala, Sweden) was mixed (0.5 mg/ml, 1 part) with CrC& (2.5 X 10e4 M, 10 parts) and packed, washed sheep erythrocytes (1 part). The mixture was incubated for 1 hr at 30°C and washed twice in 0.9% NaCl and twice in Hanks’ balanced salt solution. Lymphocytes from cultures were washed three times in Hanks’, re- suspended at 2 * lo6 viable cells/ml in RPM1 1640 supplemented with 10% fetal calf serum, and incubated in agarose with Protein A-coated sheep erythrocytes (1 hr, 37°C). A rabbit serum directed against human light chains was then added

SUPPRESSIVE ACTIVITY OF Fey RECEPTORS 255

TABLE 1

Inhibition of Human B-Cell Differentiation into Ig-Secreting Cells by Addition of Lymphocyte FcyR in NOC-Stimulated Cultures

Number of experiments FcyR origin

PEC per lo6 cells

cIg cells X IO-’ per

lo6 cells

Inhibition (%)

PFC cIg cells

15 Control” 8,606 + 1,032' PBL 4,306 f 759

2 Control 8,731; 13,046 Molt 4 T-cell line 8,102; 14,284

2 Control” 14,518; 15,503 Fibroblasts MRCS 14,892; 13,119

1 Control” 11,076 Human red blood cells 12,005

a Addition of elution buffer instead of FcyR. b Mean f standard error of the mean. ’ Not done.

235.5 f 24.9' 247.2 31 27.7

nd’ nd

nd nd

nd nd

49 -6

7; -9 nd

-2; 15 nd

-8 nd

as developing serum. After an incubation of 1 hr at 37°C the developing antiserum was replaced by 10% guinea pig complement solution (Bio MCrieux, Lyon) and the preparation was further incubated at 37°C for 1 hr. Detection of cIg cells was performed using a sheep anti-human Ig-FITC (Institut Pasteur, Paris) previously absorbed on human tonsil cells. Briefly, cytocentrifuged preparations from lym- phocytes washed six times were fixed in ethanol (5 min), washed in PBS ( 15 mm), and incubated for 30 min with the antisera in humidified atmosphere. After washing in PBS (45 min), slides were mounted in glycerol-PBS (9 vol/ 1 vol). Fluorescence- positive cells were either large blasts or cells with a morphology similar to that of plasma cells, referred to in this study as Ig-containing cells (cIg).

RESULTS

Fc-yR released from unstimulated PBL suppress the differentiation of B lym- phocytes into Ig-secreting cells. Soluble FcyR prepared from supernatants of un- stimulated PBL were added to PBL after 3 days of culture in the presence of NOC. Results of 15 experiments presented in Table 1 show that the numbers of PFC were regularly decreased. The inhibition was never complete, ranging from 20 to 60% (average 49%). Conversely, the numbers of Ig-containing cells remained un- changed. The material prepared by affinity chromatography from supernatants of cells lacking surface FcyR (Molt 4, MRCS, or human red blood cells) failed to reduce the numbers of PFC generated in cultures with NOC. Thus, a suppressor factor which was absent from supernatants for FcyR-negative cells, and which remained associated with FcyR during its purification by affinity chromatog- raphy on IgG-coated columns, could be recovered from supernatants of unstimu- lated PBL.

Suppressive activity of FcyR from different cell sources. Since FcyR are present on different cell subsets among Ficoll-separated mononuclear cells, the question

256 LE THI BICH-THUY ET AL.

arose whether the suppressor activity was associated exclusively with T-cell FcyR or not. To answer this question, FcyR were prepared from supernatants of PBL, T-enriched PBL, or T-depleted PBL and tested for their suppressive activity. Re- sults from four experiments indicated that FcyR obtained from T-depleted PBL are equally effective in reducing PFC numbers as those prepared from T-enriched PBL (Table 2). The numbers of Ig-containing cells remained unchanged in two experiments, but were slightly decreased in two others. However, when several samples of the same cell suspension were stained, the coefficient of variation of fluorescent cell numbers was calculated to be 11%. Therefore a decrease of 13 to 19% in cIg numbers could not be considered as significant.

Since FcyR from T-depleted mononuclear cells could depress the generation of Ig-secreting cells, it was important to investigate the effect of FcyR released from nonlymphoid cells. Polymorphonuclear leukocyte suspensions containing at least 98% of neutrophils were used for this purpose. Seven batches of PMN-FcyR were tested in 22 PBL cultures. The inhibition of PFC averaged 35%, whereas the number of Ig-containing cells remained unchanged (Table 2).

Because of the unexpected dissociation between the reduction of PFC and the lack of concomitant decrease of Ig-containing cell numbers, several controls were performed to exclude a possible interference of FcyR with the detection of PFC. In seven PBL cultures stimulated with NOC, FcyR were added at Day 7, that is

TABLE 2

Inhibition of PFC Generation in NOC-Stimulated Cultures by Addition of FcyR Prepared from Different Cell Sources

FcyR batches Expt FcyR origin PFC per IO’ cells cIg cells x lo-’

per 106 cells

Inhibition (%)

PFC clg cells

A I COIlId’ 7,328 282 PBL 3,145 266 T-Enriched PBL 2,829 243 T-Depleted PBL 3,630 227

2 Control’ 14.878 209 PBL 10.653 nd T-Enriched PBL 7,822 201 T-Depleted PBL 8,710 170

3 COlltd’ 14,309 234 PBL nd nd T-Enriched PBL 8.919 240 T-Depleted PBL 9,760 241

B 4 C0ntd’ 13,046 393 PBL 7.312 363 T-Enriched PBL 8,782 378 T-Depleted PBL 9.026 400

ChtKP PMN

9,847 + 948’ 6,384 + 744

134.3 t 8.4” 128.5 + 12.6

COlltKd’ T2D4 hybridoma

10,397 + 1.852’ 6,656 + 1.698

112.2 k 14.3’ 122.4 f 20.4

57 5 61 I3 50 19

26 nd 46 4 39 I8

38 -2 32 -3

44 8 32 4 31 -2

35

36 -9

4

’ Elution buffer. ‘Data represent the means f standard errors of the mean of values from 22 experiments, ‘Data represent the means + standard errors of the mean of values from 5 experiments.

SUPPRESSIVE ACTIVITY OF Fcr RECEPTORS 257

TABLE 3

Lack of Interference of FcyR with the Detection of PFC”

Source of FcyR

FcyR

0 IO2

(Reciprocal dilutions)

10’ lo4 lo5

PBL lo,7006 11,100 10,600 11,060 10,560 PMN 10,700 11,460 11,700 10,340 10,400

a Lymphocytes were cultured without FcyR. At Day 7, they were washed three times, then distributed into five suspensions to which FcyR was added at various reciprocal dilutions (0 to 105). Each suspension was then layered on the gel for the PFC assay without washing.

’ PFC per IO6 cells.

3 hr before cells were collected for the PFC assay. The numbers of PFC (mean + SEM) were 6707 + 927 in those experiments versus 6456 + 1017 in cultures to which elution buffer was added instead of FcyR. Another series of controls was performed: PBL cultivated with NOC were collected at Day 7, washed three times and fractionated into five parts. One of them received the elution buffer and the four others different dilutions of the FcyR solution, from 10e2 to lo-‘. Each fraction was immediately mixed with agarose to be plated for the PFC assay. The numbers of PFC never differed, whether the cell suspensions contained FcyR or not. Results of one of these seven experiments are presented in Table 3.

Absence of allogeneic or species restriction to the suppressive effect of FcyR. All the experiments reported above were performed in an allogeneic situation, as FcyR were prepared from one individual and tested on lymphocytes from an un- related donor. Therefore a possible effect restricted to autologous or syngeneic cells might have been overlooked. In order to clarify this point we added FcyR to autologous lymphocytes and the effect was compared with that observed with the

TABLE 4

Lack of Allogeneic Restriction of FcyR Inhibitory Effect

Source of Fc-,R Source of lymphocytes

cultured with NOC

Fc-,R added (reciprocal

dilution) PFC per IO6 cells

Inhibition (8)

clg cells x 10-1 per Id cells PFC clg’ cells

PMN from donor C

Donor A Control’ 5,510 172 Donor B 4,698 I22 Donor c 1,345 I43

Donor A IO’ 3.605 Donor B 2.168 Donor C 4.130

Lymphocytes from donor Z

Donor X COlltd’ 13,075 Donor Y 7,268 Donor Z 9,011

174 34 -1 I30 54 -13 196 44 -17

239 252 319

Donor X Donor Y Donor Z

IO’ 6,994 2.454 4,918

221 46 7

218 66 I3 375 45 -17

‘Addition of elufion buffer

258 Ll? THI BICH-THUY ET AL.

same FcyR solution tested on lymphocytes from two other unrelated individuals. The results showed that PFC inhibition was not greater in autologous than in unrelated cultures, and that no decrease of Ig-containing cell numbers occurred in cultures to which autologous FcyR were added (Table 4).

Finally, we investigated whether the suppressive activity of FcyR could be de- monstrable on xenogeneic lymphocytes. FcyR were prepared from the murine T- cell hybridoma T2D4 and added to human PBL cultures stimulated with NOC. As shown in Table 2, the suppressive properties of murine FcyR were similar to those of human FcyR with a decrease of PFC but not of Ig-containing cell numbers.

DISCUSSION

In the present study we demonstrate that soluble FcyR, isolated by affinity chromatography on IgG from supernatants of various cell sources, inhibit the dif- ferentiation of human B lymphocytes into Ig-secreting cells in cultures stimulated with NOC, a Tn-independent polyclonal activator. This suppressor factor does not impair cell viability, as shown by trypan blue exclusion tests; neither does it de- press the proliferative response to NOC, as assessed by [3H]thymidine incorporation at Day 3 (13, 20). It depresses the generation of Ig-secreting cells but it does not interfere with the detection of PFC. Surprisingly the number of cIg cells was not decreased in the presence of this factor, despite the reduction of PFC to about 50% of their control level. This discrepancy cannot be accounted for by a histocom- patibility effect since cIg cell numbers remained unaffected even in cultures added with FcyR prepared from autologous cells. This finding does not imply the lack of effect of FcyR on the generation of cIg cells since a polyvalent FITC sheep anti- human Ig was used in the present study. Indeed, further investigations using antisera against human IgM or IgG revealed that there is a selective depression of cIgG cells with an increase of cIgM cells, resulting in an unchanged number of cIg cells detected by the polyvalent antiserum (LC thi Bich-Thuy and Revillard, submitted for publication). These effects are not specific to NOC-induced B-cell activation. Parallel experiments performed with pokeweed mitogen showed that, whatever their cellular origin (PBL, T-enriched or T-depleted PBL, PMN), FcyR induced a 50% reduction of PFC, in allogeneic as well as in autologous PBL cultures and a marked suppression of cIgG without change in cIgM cells. Conversely, preparations from FcrR negative cells (Molt 4, MRCS, or erythrocytes) had no effect on pokeweed- stimulated cultures (data not shown). Therefore the suppressor factor has complex effects of the late stages of B-cell differentiation, including a presecretory block of IgM-secreting cells and an isotype-specific suppression of IgG-producing cells. Kinetic studies have shown that FcyR was suppressive only toward B cells already engaged in their differentiation pathway into plasma cells (20). Knowing the lability of this factor (20) and the lack of synchronization of B cells stimulated by polyclonal activators, it comes as no surprise that the inhibition of PFC is never complete. By these effects, as well as by its origin and mode of preparation, it is clearly distinct from other soluble suppressor factors of B-cell differentiation, as, for instance, interferon (21), inhibitors demonstrated in supernatants of mixed lymphocyte cul- tures (22), concanavalin A (23)- or pokeweed (24)-stimulated T cells, or B-cell suppressor factor (25).

The properties of human FcyR are closely related to those of mouse immuno-

SUPPRESSIVE ACTIVITY OF Fey RECEPTORS 259

globulin binding factor (9). This factor, which was later shown to be the soluble form of FcyR released by alloactivated T cells or T-cell hybridomas (10, 1 l), depresses T-dependent as well as T-independent humoral responses. It inhibits Ig- secretion by polyclonally activated cells. It is released only by T-cell subsets or T- cell lines which express FcyR ( 12). Its effects are not restricted by histocompati- bility (12), nor by species barriers (Table 2). Results reported in the present study show that these properties are not restricted to T-cell FcyR, but shared by FcyR from different cell sources. The suspensions of nonadherent E-rosetting enriched cells contained 96-97s T cells with only 5 to 10% FcyR-positive cells. We assume that FcyR prepared from these suspensions originated from T cells, even if some E-rosetting FcyR-positive cells may not belong to the T lineage (26). Conversely, E-depleted suspensions can be regarded as reasonably deprived of T cells and FcyR recovered from their supernatants may be released by monocytes, B cells, or “null” cells. Finally the suppressive capacity of FcyR prepared from PMN strongly sup- ports the concept that the suppressive effect of this factor is not dependent on its T-cell origin.

Whatever their cell origin, the suppressor factors reported in this study share the property to bind the Fc part of IgG. They are therefore identical to, or closely associated with, FcyR. Moreover, the FcyR preparations used in this study and their associated suppressor factors can be reabsorbed on IgG-coated beads but not on pepsin-digested F(ab’)% fragments (13, 20). Furthermore, preparations from FcyR-negative cells (Molt 4, MRC 5, or erythrocytes) have no effect. These soluble factors not only bind to Fc-IgG but also interact with acceptor sites at the surface of some B and T cells (18), and possibly with Ia ( 12) or Ia-like antigens. Such property may be relevant to their mode of action since it is likely to favor the crosslinkage of receptor-ligand complexes at the surface of B cells (12, 27).

The possible relevance of FcyR suppressive activity to the regulation of B-cell differentiation has not been clearly established as yet. In the in vitro primary antibody response described by Heijnen et al., Ts cells belong to the FcyR-positive subset but keep their suppressive capacity after “modulation” of their FcyR (6) and they release an antigen-specific suppressor factor different from FcyR (7). In the concanavalin A-induced suppressor assay, Haynes and Fauci showed that Ts cells can be recruited from the FcyR-negative subset (28). Conversely, indirect evidence for the involvement of FcyR in the suppression was obtained in other models. FcyR has been considered as a structure predominantly expressed on ac- tivated suppressor T cells (29). The loss of FcyR, which can be induced by IgG- immune complex interaction (30), results in the abrogation of the nonspecific T- cell suppression of Ig production (29). Similarly, antibody production can be sup- pressed by heat-killed macrophages bearing intact FcyR, but not after such re- ceptors have been enzymatically destroyed (3 1). Taken together these data would suggest that FcyR may represent a cell-bound immunoregulatory factor, whose nonantigen-specific suppressive effect on the late stages of B-cell differentiation depends on its capacity to bind the Fc part of IgG but not on its cellular origin.

REFERENCES 1. Moretta, L., Webb, S. R., Grossi, C. E., Lydyard, P. M., and Cooper, M. D., J. Exp. Med. 146,

184, 1977. 2. Moretta, L., Mingari, M. C., and Moretta, A., Immunol. Rev. 45, 163, 1979.

260 Lfi THI BICH-THUY ET AL.

3. L& thi Bich-Thuy, and Brochier, J., J. Immunol. 122, 1842, 1979. 4. L& thi Bich-Thuy, Ciorbaru, R., and Brochier, J., Eur. J. Immunol. 8, 119, 1978. 5. Schuurman, R. K. B., Gelfand, E. W., Malheson, D., Zimmerman, B., and Dosch, H. M., J.

Immunol. 124, 1924, 1980. 6. Heijnen, C. J., Uytdehaag, F., Gmelig-Meyling, F. H. J., and Ballieux, R. E. Cell. Immunol. 43,

282, 1979. 7. Ballieux, R. E., Heijnen, C. J., Uytdehaag, F., and Zegers, B. M., Immunol. Rev. 45, 3, 1979. 8. Shore, A., Dosch, H. M., and Gelfand, E. W., Nature (London) 274, 586, 1978. 9. Fridman, W. H., and Golstein, P., Cell. Immunol. 11, 442, 1974.

10. Fridman, W. H., Guimezanes, A., and Gisler, R. H., J. Immunol. 119, 1266, 1977. 11. Neauport-Sautb, C., Rabourdin-Combe, C., and Fridman, W. H., Nature (London) 277,656, 1979. 12. Fridman, W. H., Rabourdin-Combe, C., Neauport-Sautbs, C., and Gisler, R. H., Immunol. Rev.

56, 51, 1981. 13. L& thi Bich-Thuy, Samarut, C., Brochier, J., Fridman, W. H., and Revillard, J. P., Eur. J. Immunol.

10, 894, 1980. 14. Dickler, H. B., Advan. Immunol. 24, 167, 1976. 15. Winchester, R. J., Hoffman, T., Ferrarini, M., Ross, G. D., and Kunkel, H. G., Clin. Exp. Immunol.

37, 126, 1979. 16. Julius, M., Simpson, E., and Herzenberg, L. A., Eur. J. Immunol. 3, 645, 1973. 17. Brochier, J., Abou-Hamed, Y., Gueho, J. P., and Revillard, J. P., Immunology 31, 749, 1976. 18. Samarut, C., and Revillard, J. P., Eur. J. Immunol. 10, 352, 1980. 19. Gronowicz, E., Coutinho, A., and Melchers, F., Eur. .I. Immunol. 6, 588, 1976. 20. Le thi Bich-Thuy, Samarut, C., Brochier, J., and Revillard, J. P. J. Immunol. 127, 1299, 1981. 21. Gisler, R. H., Lindahl, P., and Gresser, J., J. Immunol. 113, 438, 1974. 22. Wolf, R. L., Whitsed, H., Rosen, F. S., and Merler, E., Cell. Immunol. 36, 231, 1978. 23. Kadish, A. S., Tonsery, F. A., Yu, G. S. M., Doyle, A. T., and Bloom, B. R., J. Exp. Med. 151,

637, 1980. 24. Saxon, A., and Stevens, R. H., Clin. Immunol. Immunoparhol. 10, 427, 1978. 25. Miyama, M., Yamada, J., and Masudo, T., Cell. Immunol. 44, 51, 1979. 26. Reinherz, E. L., Moretta, L., Roper, M., Breard, J. M., Mingari, M. C., Cooper, M. D., and

Schlossman, S. F., .I. Exp. Med. 151, 969, 1980. 27. K6lsch, E., Oberbarnscheidt, J., Bruner, K., and Heuer, J., Immunol. Rev. 56, 61, 1981. 28. Haynes, B. F., and Fauci, A. S., J. Immunol. 121, 559, 1978. 29. Pichler, W. J., and Broder, S., Immunol. Rev. 56, 163, 1981. 30. Cordier, G., Samarut, C., and Revillard, J. P., J. Immunol. 119, 1943, 1977. 31. Ptak, W., and Gershon, R. K., J. Immunol. 115, 1346, 1975.