Functional characterization of CD77+ B cells

Eur. J. Immunol. 1991. 21: 1131-1140

Marianne Mangeneyov, Yolande Richard., Dorninique Coulaud., Thomas lhrsz0 and JoeUe Wiels. Laboratoire d’Immuno-Biologie des ’lbrneurs., CNRS URA 1156 and Laboratoire de Microscopie Cellulaire et MolBculaireA,Institut Gustave Roussy, Viejuif and INSERM U 131

1131

CD77: an antigen of germinal center B cells

entering apoptosis* We have previously reported that a neutral glycolipid (globotriaosylceramide; Gb3) was specifically expressed on Burkitt’s lymphoma cells and on a subset of germinal center tonsillar B lymphocytes. Recently the Gb3 molecule was recognized as a new B cell differentiation antigen and now defines the CD77 cluster. Here we report an extensive phenotypic and functional characterization of the tonsillar CD77+ B 1ymphocytes.These cells have a low buoyant density and are thus purified using a Percoll gradient. They express various B cell antigens such as CD19, CD20, CD21, CD22 and CD40, as well as the adhesion molecules LFA-1, LFA-3 and CD44. They are positive for surface IgM and negative for surface IgD. Although these results suggest a phenotype of activated B cells, the CD77+ cells are negative for the classical activation antigens: CD23 (the low-affinity Fc receptor for IgE), CD25 [the interleukin (IL) 2 receptor a chain] and CD71 (the transferrin receptor). Proliferation and protein synthesis of CD77+ cells was measured after stimulation with a range of mitogens and IL. None of the agents tested are able t o induce proliferation and protein synthesis with the exception of a combination of recombinant I L 4 plus anti-CD40 antibody.When examined by electron microscopy, CD77+ B lymphocytes present a morphology similar to that of cells undergoing programmed cell death, also called apoptosis (i. e. chromatin condensation, nuclear fragmentation, membrane blebbing). As shown by direct examination of DNA, these CD77+ cells are indeed in the process of apoptosis. Treatment of the CD77+ cells by recombinant IL 4 and antLCD40 antibody prevents apoptosis. All these results suggest that the CD77 molecule defines a B lymphocyte maturation pathway, specific for germinal center, where the cells undergo programmed cell death.

1 Introduction Apoptosis is a common form of cell death in eukaryotes. In this process, the cells die in a controlled manner in response to specific stimuli, apparently following an intrinsic program. It has been shown that apoptosis occurs in several parts of the immune system : upon death of autoreactive thymocytes [ 11and immature thymocytes [2] stimulated via the TcR, upon death of hematopoietic precursor cells on withdrawal of the relevant CSF [3] or IL 3 [4],upon death of immature B lymphoma cell after triggering of their membrane Ig [S,61, in the CTL- and NK-mediated killing [7]. In all of these circumstances, a series of strikingly similar morphological changes occur, with several of the most conspicuous ones occurring in the nucleus. Apoptosis has also been observed in lymphoid germinal center (GC) where many of the cells produced apparently

die in situ [8]. GC, which develop within lymphoid tissues shorty after antigenic challenge [9], are the sites of intense B cell activation and differentiation. The differentiation pathways within G C leading to memory B cells and/or plasma cells and/or cell death are poorly understood. New markers which could discriminate certain cell types among G C cells are thus valuable tools for a better understanding of B cell development. Globotriaosylceramide (Gb3), a neutral glycolipid, was recently described as a B cell differentiation antigen, specifically expressed on a subset of G C B cells [lo]. We originally reported this antigen as being expressed in high amounts on Burkitt’s lymphoma cells [111 and, therefore, named it BLA (Burkitt’s lymphoma-associated antigen; [12,13]). At the Fourth International Workshop on Human Leucocyte Differentiation Antigens, four anti-Gb3 mAb were clustered and Gb3 now defines the CD77 antigen ~41.

We undertook the morphological, phenotypic and functional characterization of the tonsillar CD77+ population. * This research was supported in part by grants from the Ligue The CD77+ cells express the surface B cell markers. They Nationale FranGaise contre le Cancer and the Association pour are IgM+ and IgD- but are negative for classical activation la Recherche sur le Cancer (ARC 6015). antigemwithin a few hours of purification the CD77+ cells Recipient of a grant from the Ministere de la Recherche et de la actually showed the typical morphology and DNA fragTechnologie. mentation of cells undergoing apoptosis. Treatment of these cells with anti-CD40 and rIL 4 was able to prevent Correspondence: Marianne Mangeney, Laboratoire d’Immunoapoptosis. In contrast to numerous activators andor IL this Biologie deslbmeurs, CNRS URA 1156, Institut Gustave Roussy, combination of anti-CD40 mAb and IL 4 was also able to F-94805 Villejuif Cedex, France sustain CD77+ cell proliferation and protein synthesis. The Abbreviations: LCL: Lymphoblastoid cell line SAC: Staphylo- implication of these results on G C physiology are discussed. coccus aureus Cowan strain I GC: Germinal center

[I 91391

0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991

OO14-2980/91/0505-1131$3.50 + .25/0

1132

Eur. J. Immunol. 1991. 21: 1131-1140

M. Mangeney,Y. Richard, D. Coulaud,T. Tursz and J. Wiels

2 Materials and methods 2.1 Cell lines Burkitt's lymphoma cell lines (Ramos, Daudi) were kindly provided by Dr. G . Klein, Karolinska Institute, Stockholm. Lymphoblastoid cell lines (LCL) established from PBL (IARC 174, IARC 272) or from low-density tonsillar lymphocytes (TRA 35; [ 151) were kindly given by G. Lenoir (IARC, Lyon, France) and C. Gregory (University of Birmingham Medical School, Birmingham, GB), respectively. These cell lines were grown in RPMI 1640 medium containing 2 mM L-glutamine, 1 mM sodium pyruvate, 20 mM glucose, 20 pg/ml gentamycin and supplemented with 10% heat-inactivated FCS.

2.2 B lymphocyte preparation Mononuclear cells were obtained from tonsils removed from patients with repetitive infections of the upper respiratory tract. Tonsillar cell suspensions were prepared by gentle teasing in culture medium (RPMI 1640) supplemented with antibiotics (penicillin, streptomycin, 100 pg/ml; Gibco, Grand Island), sodium pyruvate (1 mM Gibco) and 10% FCS.Tonsillar cells were depleted of Tcells by rosetting out sheep erythrocyte receptor-positive cells with 2-aminoethyl-isothiouronium bromide-treated sheep erythrocytes and Ficoll-Hypaque gradient separation. Monocytes were removed by plastic adherence. Cells were washed twice and then fractionated using a Percoll gradient method (Pharmacia, Uppsala, Sweden). 2.3 Reagents

Polyclonal activators of B lymphocytes were : insoluble goat anti-human IgM (10 pg/ml; Bio-Rad, Richmond, CA)

and Staphylococcus aureus Cowan I (SAC; 1/1@ dilution; Calbiochem, La Jolla, CA). r I L 2 was purchased from Boehringer Laboratories (Boehringer Mannheim, Meylan, France). r I L 4 was provided by Dr. K. Arai (DNAX Research Institute, Palo Alto, CA). Human rIFN-y was obtained from Roussel-Uclaf (Romainville, France). rIL la was provided by Biogen (Geneva, Switzerland) and rIL 6 was purchased from Genzyme (Boston, MA). 2.4 Surface immunofluorescence labeling Immunofluorescence labeling of viable cells for surface antigens was generally performed in 96-well plates. mAb used as primary reagents are summarized in Table 1. In most cases, the secondary antibody used was FITCconjugated goat anti-mouse Ig (Nordic, Tilburg, The Netherlands). For 38.13 mAb, the secondary reagent was Fc-specific goat anti-rat IgM conjugated with FITC. When double labelings were performed, cells were usually stained with 38.13 mAb (anti-CD77) revealed by an FITC-conjugated goat anti-rat IgM and the other mAb were revealed by a biotinylated goat anti-mouse IgG (Amersham, Int., Amersham, GB) followed by PE-labeled avidin (Becton Dickinson, Mountain View, CA). FITC-conjugated rabbit anti-human Ig were used for detection of sIg and in these experiments 38.13 was revealed by a biotinylated goat anti-rat IgM followed by PE-labeled avidin. All reagents were titrated and used at optimal concentration. The labeling protocol was as follows : 2 x 105 cells were mixed with 50 pl primary reagent appropriately diluted in PBS containing 10% FCS and incubated at 4 "C for 30 min. After washing, secondary antibodies were added for 30 min. The cells were then washed twice and fixed in 1% formaldehyde. Stained tonsillar B cells were analyzed with an Epics C cell sorter (Coultronics, France). For each analysis, 104 cells were accumulated.

Table 1. Primary reagents for surface immunofluorescence

Category B cell

Activated B cell Subpopulation of B cell

Antibody

Antigen

2H7 G29.5 G28.7 G28.5 G28.10

cD20

MHM6 IOT14 OKT9 DUALLl LOTl B8.7 38.13

J5 Adhesion molecule

T lymphocyte Monocyte MHC Class 1I

cD21 cD22

cD40

cD23 cD25 cD71 cD9 CD5

cD77 cD 10 cDl0 CD38

IOT5 IOB6 8F5 LFA- 1 LFA-3

CDlla CD58

P245

cD44

OKT3 OKMl L1.1.12

CD3 CDllb

CD54

J. J. J. J. J.

Source

Ref.

Ledbetter Ledbetter Ledbetter Ledbetter Ledbetter

[16] [17] [18] [19] 1201 i21j 1211

M.Rowe Immunotech Ortho Workshop Immunotech Y. Richard Our laboratory Coulter Immunotech Immunotech Our laboratory Workshop Workshop A. Bernard Ortho Ortho M.Fellous

[22] [11]

(231 [24]

[25]

Functional characterization of CD77+ B cells

Eur. J. Immunol. 1991. 22: 1131-1140

2.5 In vitro proliferation assays Cells were cultivated at 37°C in triplicate in 96-well round-bottom microtiter plates (Becton Dickinson) in 200 p1 of culture medium at 2 x 1@cells/well. At various time, the cells were pulsed with 1 pCi = 37 kBq of [3H]thymidine or [3H]leucine per well (Amersham) for 18 and 6 h, respectively.They were then captured on glass fiber filters, and radioactivity was measured in a scintillation counter. In some experiments, antibodies or factors were added to the culture. Data are reported as mean cpm.

1133

Samples were diluted twofold with loading buffer containing 25 mM EDTA, 40% sucrose and 0.5% bromophenol blue, then heated at 70°C for 10 min prior to loading. Electrophoresis was carried at 70 V for 3-4 h in 2% agarose gel containing 0.1 pg/ml ethidium bromide, in a buffer containing 2 mM EDTA, 80 mM Tris-phosphate, pH 8. After electrophoresis, gels were examined under UV light.

3 Results 3.1 Purification of CD77+ tonsillar B lymphocytes

2.6 Glycolipid purification and characterization Glycolipids were isolated from homogenized packed cells following extraction with 20 vol of isopropanol/hexane/water (55 :25 : 20, v/v). After Folch's partition the neutral glycolipids were acetylated and purified on a florisil column followed by deacetylation. The neutral glycolipids were analyzed on high performance thin-layer chromatography (HPTLC) plates (J. T. Baker, Deventer,The Netherlands). TLC immunostaining was performed by a slightly modified version of the procedure described by Magnani et al. [26]. Glycolipids were applied on HPTLC plates using a solvent system of chloroform/methanol/water containing 0.05% CaCI2 (50 : 40 : 10, v/v). After drying, HPTLC plates were blocked for 2 h with 5% BSA in PBS and reacted with culture SN of 38.13 hybridoma at 4°C overnight. After washing, bound antibody was detected using a rabbit anti-rat IgM antibody (ICN Biomedicals, Costa Mesa, CA), followed by detection with 12sI-labeledprotein A (Amersham). HPTLC plates were then autoradiographied.

A quick and simple procedure to purify the CD77+ cells to near homogeneity, by using a modified Percoll gradient, is described. The Percoll gradient used here was a slight modification to one previously described [27].The gradient was constructed by sequential layering of isoosmotic Percoll solutions of density : 1.077 g/ml (60%), 1.067 g/ml (50%), 1.056 g/ml (40%) and 1.043 g/ml (30%). Total nonrosetting lymphocytes were resuspended at 5 x lo7 cells/ml in 2 m l of 1.043 g/ml Percoll solution and were layered on the top of the gradient. After centrifugation at 1200 x g for 15 min, cells were recovered from the different

I

2.7 EM

Cells were fixed at room temperature by adding 1.5% glutaraldehyde to the culture medium and centrifuged. Pellets were then placed in new fixative (1.5% glutaraldehyde in 0.066 M Sarensen buffer, pH 7.4) for 1 h at 4°C. After 2 h washing with buffer, cells were postfixed in 2% Os04 in the same buffer. Dehydration in graded ethanol and propylene oxide, Epon embedding and uranyl-lead staining were performed using classical methods. Observations were made in a Zeiss EM902 (Oberkochen, FRG) optimal contrast was obtained by .selecting the elastic electrons with the slit of the spectrometer.

I

(=I

2.8 DNA fragmentation assays

Cells were centrifuged and washed twice with PBS. The pellets were lysed in 10 mM EDTA, 200 m~ NaCI, 0.1 mg/ml proteinase K, 0.5% (w/v) SDS and 50 mM Tris-HC1, pH 8 and incubated for 1 h at 50 "C.The DNA was extracted with phenol, then with chloroform/isoamylalcohol (24 : 1) and ethanol precipitated. Unfragmented DNA was discarded, and one tenth volume of 3 M sodium acetate, pH7.2 was added to the SN which was left at -80°C overnight.The precipitate containing fragmented DNA was centrifuged (13 OOO x g, 30 min); the pellet was dried under vacuum and resuspended in 20 p1 RNase buffer containing 0.5 pg/ml DNase-free RNase (Sigma, St. Louis, MO), 15 mM NaCl and 10 mM Tris-HC1, pH 7.5, at 50°C for 1h.

Figure 1. Immunofluorescenceanalysis of tonsillar B lymphocytes after Percoll density gradient centrifugation. E- cells were loaded on a Percoll gradient. Cells recovered from the different fractions were treated with no mAb (shaded) or were labeled for CD20 (2H7 mAb, unshaded, left column) and CD77 (38.13 mAb, unshaded, right column). (A) and (B) : total tonsillar B lymphocytes; (C) and (D) : Percoll fraction >30%; (E) and (F) : Percoll fraction 30%-40%; (G) and (H) : Percoll fraction 40%-50%; (I) and (J) : Percoll fraction 50%-60%.

1134

M. Mangeney,Y. Richard, D. Coulaud, T. Tursz and J. Wiels

interfaces. Of the cells recovered at the top of the upper layer 90 f 8% were CD77+.These cells will be referred to as CD77+ lymphocytes. Expression of CD77 and CD20 on cells obtained in the different Percoll layer, of a representative experiments, is shown in Fig. 1. 3.2 Phenotype of CD77+ tonsillar B lymphocytes

Double-immunofluorescence labeling, using 38.13 mAb (anti-CD77) and a series of mAb recognizing leukocyte antigens, was realized on purified tonsillar B lymphocytes. Results are presented in Table 2. The presence of memTable 2. Surface phenotype of CD77+ tonsillar B lymphocytes Category

B cell

Membrane Ig B cell subpopulation

Adhesion molecule

Antigens

CD19 CD20 CD21 CD22 CD40 MHCclass I1 sIgM sIgD CDS CDY CDlO CD38 B8.7 (328.10 ICAM LFA- 1 LFA-3

cD44 Activated B cell

T lymphocytes Monocytes

CD23 CD25 CD71 CD3 CD18

Tonsils % mAb binding’) Exp.: 1 2 3 4 84 85

NT 94

97 89

78 66 69 89 100 1 >1 2 5 3 8

a) Percentage of CD77+ cells double-labeled with the mAb shown. N T Not tested.

Eur. J. Immunol. 1991. 21: 1131-1140

brane IgM and reactivity with various pan-B antigens such as CD19, CD20, CD21, CD22, CD40, along with the absence of reactivity with OKT3 and OKMl allowed the identification of CD77+ cells as a population of B lymphocytes. The adhesion molecules LFA-1, LFA-3 and CD44 were expressed on the majority of CD77+ cells whereas only a third of these cells were labeled by 8F5, an anti-ICAM antibody. Among other B cell-associated antigens, CD5 was not found on CD77+ cells whereas CD9 was variably but significantly detected. CD10, detected by mAb J5, was only slightly expressed on CD77+ cells and its expression was variable from one tonsil to another as previously described [ 101. Fig. 2 shows the striking feature of the CD77+ B lymphocytes. The CD77+ cells were brightly stained by the IOB6 mAb which recognizes the CD38 molecule. It is worth noting that, as CD77 which we first reported as a Burkitt lymphoma-associated antigen [ 111, CD38 was recently described as a good marker for Burkitt’s lymphoma cells [28]. These cells have a low buoyant density and are IgM+ and IgD-, which could suggest that they are in an activated stage. The CD77+ lymphocytes, however, were consistently negative for the activation antigens CD23 (the low-affinity FceR), CD25 (the IL 2R a chain), CD71 (the transferrin receptor).The majority of the CD77+ cells expressed CD40 and CD44 antigens.

3.3 Glycolipid content of CDV+ tonsillar B lymphocytes It has been shown previously that some glycolipids although chemically present in large amount in the cells are immunologically poorly detectable [29]. To ascertain that Gb3 membrane expression was correlated with a difference in Gb3 biosynthesis of CD77+ and CD77- cells and was not due to a simple masking of Gb3 in the CD77- cells, we purified and characterized the glycolipids of the CD77+ and CD77- lymphocytes. Fig. 3 presents the neutral glycolipid pattern of tonsillar lymphocytes and of various B cell lines. A glycolipid which co-migrates with standard Gb3 and reacts with 38.13 mAb, was present in CD77+ lymphocytes (lane 2) but was absent from B lymphocytes of higher density Percoll fractions (lanes 3, 4). This compound was Figure 2. Two-color immunofluorescence analysis of tonsillar B lymphocytes. Tonsillar B cells were obtained after rosetting out sheep erythrocyte receptor-positive T lymphocytes and removing monocytes by plastic adherence. Cells were usually stained with 38.13 mAb (anti-CD77) revealed by a FITC-conjugated goat anti-rat IgM, the other mAb were revealed by a biotinylated goat anti-mouse IgG followed by PE-labeled avidin. FITC-conjugated rabbit anti-human Ig were used for detection of sIg and in these experiments 38.13 was revealed by a biotinylated goat anti-rat IgM followed by PE-labeled avidin.

Eur. J. Immunol. 1991.21: 1131-1140 (A)

Functional characterization of CD77+ B cells

a proliferative response of total tonsillar B cells was obtained with all IL added whereas high-density B lymphocytes proliferated poorly. By contrast, CD77+ cells were completely unresponsive to the signals delivered by these lymphokines.

(B)

1 2 3 4 5 6 7 8 0 1

1135

2 3 4 5 6 7 8 9

Figure 3. HPTLC and immunostainingpattern of B cell glycolipids with an anti-Gb3 mAb (38.13). Neutral glycolipids were prepared from B lymphocytes and B cell lines. (A) OrcinollH2SO4staining pattern. (B) TLC immunostaining with 38.13. Lane 1 : standard globotriaosylceramide, lane 2 : Percoll fraction > 30% tonsillar lymphocytes, lane 3 : Percoll fraction 40.50% tonsillar lymphocytes, lane 4 :Percoll fraction W60% tonsillar lymphocytes, lane 5 L:TRA35 (LCLobtained from 38.13+ tonsillar B lymphocytes), lane 6 : Daudi (BL cell line), lane 7 : Ramos (BLcell line), lane 8: IARC 174 (LCL obtained from PBL), lane 9 : IARC 272 (LCL obtained from PBL).

also present in a lymphoblastoid cell line obtained after EBV infection of CD77+ lymphocytes(lane 5 ) as well as in Burkitt’s lymphoma lines (lanes 6,7) but was absent from LCL obtained from peripheral B lymphocytes (lanes 8,9). We thus conclude that the membrane expression of CD77 on tonsillar B cells is indeed associated with the biosynthesis of Gb3.

One p i b l e explanation for these negative results is that the CD77+ cells do not express functional IL receptors. Polyclonal stimuli were thus used, without or with addition of IL (Fig. 4 B and C, respectively) in an attempt to trigger proliferation. As was observed with IL, total tonsillar B cells stimulated with anti-p Ab, SAC and anti-p Ab plus IL proliferated after 3 days culture, whereas CD77+ cells were totally unable to respond to these mitogens. As expected the resting B cells proliferated in presence of SAC and anti-p Ab + IL. Several B cell surface antigens involved in B cell growth have been recently defined by mAb.These mAb appear to mimic the effects of the still unidentified natural ligands for these B cell receptors. One of these molecules is a 50-kDa surface protein detected by anti-CD40 mAb [191. As it has been shown that the anti-CD4 mAb antibody can induce a progression signal and can also synergize with rIL4 to we tested the effect of induce a proliferation of B cells [MI, an a n t i - 0 4 0 mAb G28.5 on CD77+ lymphocytes. As shown in Fig. 4D,antLCD40 mAb alone did not induce proliferation of CD77+ cells, whereas a slight increase in DNA synthesis was observed after treatment with a combination of antLCD40 mAb and rIL 4. A control IgGl antibody (G28.10) and rIL4 had no effect on CD77+ cells.

3.4 Proliferation assay

3.5 Protein synthesis The response of CD77+ cells to various lymphokines was investigated. CD77+cells, high-density B lymphocytes and total tonsillar B lymphocytes were cultivated for 72 h with rIL 1, rIL 2, rIL 4, rIL 6 and rIFN-y. As shown in Fig. 4 A,

The protein synthesis of CD77+ cells in the presence of different stimulating agents was analyzed at various time. As shown innble 3, after an 8 h culture without stimulat-

Figure 4 . Proliferation of CD77+, high-density and total tonsillar B 1ymphocytes.Tonsillar B lymphocytes were grown for 3 days in the presence of RPMI 1640 plus 10% FCS, SAC (at 1/1@ dilution), anti-p Ab (5 pg/ml), rIL 1a (500 pdml), rIL2 (50 U/ml), rIL4 (400 U/ml), rIL 6 (100 U/ml), rIFN-y (100 U/ml), low molecular weight (at 1/u) dilution), G28.5 anti-CD40 (5 @ml) and G28.10 mAb (5 @ml), and were pulsed during the last 18h of the culture. Results show the mean of six experiments, CD77+lymphocytes (m); highdensity B lymphocytes 0; total B lymphocytes (0).

1136

Eur. J. Immunol. 1991. 21: 1131-1140

M. Mangeney,Y. Richard, D. Coulaud,T.n r s z and J. Wiels

Table 3. Analysis of protein synthesis of the CD77+ and total B lymphocytesa)

[3H]Leucineuptake 8h

Medium SAC

IL4 G28.5 G28.5

+ IL4

6559 10407 6131 10297 11540

f 1898 f 1318 f 882 f: 2433 f: 2822

lbtal B cells 24 h

48h

8h

m 7 7 + cells 24 h

48h

6782 f 1291 27721 k 3292 18108 k 1671 6%0 k 2335 18439 f 1981

6744 f 1881 23247 f 2221 12677 f 1859 13282 f 1002 16083 f 1060

6500 f 1772 5900 f 2523 3824 f 1073 6oO7f 1247 5131 k 131

2084f 2515 f 1705 f 2130f 1824f

999f232 951 f 481 952 f 401 1861f335 3959f399

905 1009 1005 782 766

a) [3H]Leucineuptake after 8,24 and 48 h of culture in the presence of medium alone, SAC (at 111OOOO dilution), IL 4 (400U/ml) and G28.5 (5 pg/ml). Results show the mean of five experiments.

ing agents, protein synthesis was similar in CD77+ cells and in total B lymphocytes. Stimulation with SAC, anti-CD40 mAb and anti-CD40 mAb plus rIL 4 resulted in increased protein synthesis of total B cells but had no effect on CD77+ lymphocytes. After 48 h of culture, protein synthesis of stimulated total B cells was increased as compared to control cells incubated with medium. By contrast, the leucine uptake was drastically decreased at 48 h in CD77+ cells, as well as in CD77+ cells stimulated with SAC, anti-CD40 mAb or rIL 4, indicating that these cells were all dead. A combination of antLCD40 mAb and rIL 4 was able to rescue at least a fraction of the CD77+ cells as demonstrated by the relatively high protein synthesis level observed at 48 h.

3.6 Morphological analysis of CD77+ lymphocytes Since the CD77+ cells died within 48 h of culture, the ultrastructure of these cells was examined by EM. Morphology of low-density CD77+ lymphocytes and high-density lymphocytes is presented in Fig. 5. Morphological features of apoptosis was already clearly visible on EM of freshly isolated CD77+ cells (Fig. 5 a) : the chromatin formed dense crescent-shaped aggregates lining the nuclear membrane and plasma membrane blebs could be observed. After 24 h of culture, there were no CD77+ surviving cells (Fig. 5 c). By contrast, high-density lymphocytes exhibited the same morphology after 24 h of (Fig. 5 d ) as at the beginning of the culture (Fig. 5b). Again addition of anti-CD40 mAb plus rIL 4 to the culture medium rescued most of the CD77+ cells (Fig. 5 e).

3.7 DNA fragmentation assays Another characteristic of apoptosis is an endonucleaseinduced DNA fragmentation into oligonucleosomal fragments [31]. Fig. 6 shows the electrophoresis pattern of the fragmented fraction of DNA obtained after a second-step precipitation. The CD77+ cells cultivated for 18 h showed fragmented DNA (lane 1). In contrast, DNA of CD77+ cells treated with the anti-CD40 mAb and rIL 4 for 18 h was only marginally degraded (lane 3) and the DNA of the high-density cells was not fragmented (lane 2). Taken together, these results suggest that the CD77+ cells are engaged in programmed cell death, a process that can be partially prevented by a combination of anti-CD40 mAb plus rIL 4.

4 Discussion In the present study, we provide a phenotypic and functional characterization of a new tonsillar B cell subset defined by expression of CD77 antigen.The CD77+ cells have been previously localized in tonsillar GC [lo]. They are mostly cycling cells [151 which exhibit a centroblastic morphology in light microscopy. The CD77+ cells express sIgM but no sIgD and they are positive for a series of B cell markers: CD19, CD20, CD21, CD22 and CD40. The adhesion molecules LFA-1, LFA-3 and CD44 are also present on CD77+ cells. Several studies have been previously devoted to G C cells of lymphoid organs in order to better understand the B cell differentiation process [32-371. GC are generally considered to be the site of clonal expansion of antigen-specific B cells, giving rise to plasma cells or memory B cells [38-401. Our results are mostly in agreement with the phenotypic features described for the GC cell, in the literature. However, some points need to be underlined: the CD77 antigen is expressed only on 10%-15% of the total tonsillar B lymphocytes and thus, probably defines a subpopulation of GC cells which constitutes approximately 30% of the total lymphoid compartiment [33]. Moreover, the antigenic profile of the CD77+ cells presents some interesting characteristics. For example, all CD77+ cells express sIgM whereas GC cells are usually described as sIgM+ and sIgG+ cells [32-341. It must also be noted that some investigators claimed that G C B cells are mainly sIg- [9,35,41]. Another noticeable difference concerns expression of CD44. CD44 expression has been previously reported to be low or negative on G C cells [37,42,43]; however, we consistently find this antigen expressed at relatively high density on the CD77+ cells. Cloning of the CD44 molecule has demonstrated homologies between CD44 and the cartilage link and proteoglycan proteins [ a ] . Also very recently Miyake et al. have shown that the CD44 molecule participates in hyaluronate recognition [45], which implies a general role for CD44 in the cellular matrix adhesion processes. Further studies concerning the role of CD44 on CD77+ cells are currently under process. Peanut agglutinin (PNA), a lectin derived from the peanut plant Aruchis hypogueu and which recognizes terminal galactose residues is commonly considered as a specific marker of GC cells [33,46,47]. However, double-labeling experiments with PNA and 38.13 mAb demonstrated that 50% of the cells recognized by 38.13 mAb (CD77+ cells) are not labeled by PNA (data not

Eur. J. Immunol. 1991. 21: 1131-1140

Functional characterization of CD77+ B cells

1137

Figure 5. Ultrastructuralanalysis of CD77+1ymphocytes.The ultrastructureof CD77+lymphocytes (a) and high-density lymphocytes (b) was examined at the end of the isolation procedure or after 24 h of culture with RPMI 1640 plus 10% FCS, (c) CD77+lymphocytes; (d) high-density lymphocytes or with G28.5 anti-CD40 mAb (5 @ml) plus rIL 4 (400U/ml) (e) CD77+ cells.

shown). Concerning the expression of activation antigens, it is generally admitted that GC cells do not express neither CD25 (IL 2R a chain) nor CD23 (low-affinity FcER) but express CD71 (transferrin receptor) [34, 481. Again, the phenotype of CD77+cells does not completely fit with that described for GC cells since we have never been able to

detect CD71 on these cells. These results strongly suggest that CD77 antigen delineates, among GC cells, a new B cell subset at a peculiar stage of in vivo activation.

CD77+ lymphocytes express the CD38 antigen. The CD38 molecule was originally described as a Tcell marker [49],

1138

M. Mangeney,Y. Richard, D. Coulaud,T.Tursz and J. Wiels

1

2

3

I

I

I

Figure6 Effects of anti-CD40 plus rIL4 on DNA fragmentation in CD77+lymphocytes.CD77+and highdensity cells were grown for 18 h in RF’MI 1640 plus 10% FCS (lane 1 and 2, respectively) or with G28.5 antiCD40 (5 pglml) plus rIL 4 (400 Ulml; lane 3).

but was subsequently shown to be detectable during both the early and late stages of Tand B lymphocytes maturation [50, 511 and absent from resting B cells [52]. It was also described on G C B cells [52, 531. An interesting observation was made recently by Ling et al. who reported CD38 as a good marker for Burkitt lymphoma cells [28].This result certainly reinforce previous reports [lo, 151, in which, we and others described the CD77/BLA+ lymphocytes as having a “Burkitt-like” phenotype. On the basis of phenotypic studies, we suggested that this B cell subset could provide a normal counterpart for BL tumor cells. However, it must be noted that in vitro transformation of normal CD77/BLA+, CALLA+ tonsillar B cells does not produce cell lines phenotypically resembling BL cells. For many years, a possible association between GC B cells and BL has been underlined by pathologists [53]. Later this hypothesis was reexamined particularly in the light of the better knowledge of the localization of chromosomal breakpoints in “endemic” and “sporadic” BL cases. Data obtained from sporadic BL, again suggested that these tumors are initiated in GC, whereas the origin of endemic BL is still controversial [54, 551.

To characterize further this new B cell subset, the proliferative capacity of the CD77+ lymphocytes was evaluated. These cells completely failed to respond to polyclonal activators and to most IL tested. However, with a combination of rIL4 and anti-CD40 mAb, a slight increase in DNA synthesis was observed. Protein synthesis, as measured by incorporation of [3H]leucine in the CD77+ cells was also sustained, over a 48 h period, by the presence of rIL 4 plus anti-CD40 mAb. Using electron microscopy, we observed that after 24 h of incubation at 37”C, nonstimulated CD77+ cells were all dead whereas cells cultivated with rIL 4 plus anti-CD40 mAb were still alive.When DNA was analyzed by gel electrophoresis, extensive degradation of DNA, a characteristic of apoptosis, was observed in the CD77+ cells. By contrast DNA of CD77+ cells treated with anti-CD40 mAb plus rIL4 was only marginally degraded. One could speculate that signals delivered to the cells by anti-CD40 plus rIL 4 are able to shut off the endonuclease involved in DNA fragmentation of the apoptosis mechanism. Such a mechanism could be involved in the BL transformation process.

Eur. J. Immunol. 1991. 21: 1131-1140

CD40 antigen has been extensively studied and is now considered as a likely receptor candidate for either a cellular ligand or a soluble factor supporting B cell proliferation. Molecular cloning of cDNA encoding the CD40 protein has shown an extensive homology with the nerve growth factor (NGF) receptor [56] and the TNF receptors [57]. Gordon et al. showed that an antLCD40 mAb can synergize with IL 4 in maintaining DNA synthesis in pre-activated B cells [58]. Furthermore Banchereau et al. recently demonstrated that normal human B lymphocytes can proliferate when grown with cross-linked anti-CD40 antibody. When I L 4 was added to the cultures, these authors were also able to generate long-term normal B cell lines [59]. It has also been reported that antLCD40 mAb plus rIL 4 up-regulated the expression of CD23 antigen and induced the release of soluble CD23, aputative autocrine B cell growth factor [60].The possible role of soluble CD23 in the rescue of the CD77+ cells, remains to be investigated. Another area of interest will be to consider the direct effect of NGF on CD77+ lymphocytes. NGF is a well characterized neutrophic protein, required for the survival of embryonic sensory and sympathetic neurons following axonal innervation of the neuronal target fields during development [61]. It has also recently been reported to induce growth and differentiation of human B lymphocytes [62]. Elucidation of the mechanisms allowing the survival of CD77+ cells will also give new insights in GC physiology. Why are the CD77+ cells entering apoptosis? Even when the analysis is done at the end of isolation procedure, itself performed quickly after the tonsillectomy, a small percentage of cells already show the morphologic features of apoptosis : dense crescent-shaped chromatin aggregates lining the nuclear membrane and blebbing of the plasma membrane [31]; after 18 h of culture, these cells present the DNA fragmentation pattern, characteristic of apoptosis. Recently, a very interesting observation was made by Liu et al. [63] who suggested that apoptosis could be also a mechanism for elimination of cells not able to produce high-affinity antibody during antigen-driven selection in GC. Although the purification procedure used by Liu et al. yielded a population with a phenotype slightly different from that of the CD77+ cells analyzed here, it can be hypothesized that these two subsets are in fact identical. If this is true, it will be of great interest to investigate the role of GbdCD77 in the selection process of cells with high affinity for antigen. Biological roles of carbohydrates are presently poorly understood. However, it has been demonstrated that glycosphingolipids are involved in the development of neural tissue [64]and play a central role during the first steps of murine embryogenesis [65]. Furthermore, glycolipids also constituted membrane receptors for various bacteria and viruses [66].Of particular relevance to our study is the fact that Escherichia coli verotoxin (also known as shiga-like toxin [67,68]) as well as Shiga toxin produced by Shigella dysenteriae [69, 701 specifically bind to Gb3. Moreover, a recent study has shown that exogenous incorporation of Gb3 into B cells lacking this receptor, induced sensitivity of these cells to verotoxin [71], therefore demonstrating that Gb3 alone can be a functional receptor. The role of the GbdCD77 as a possible receptor for a signal molecule inducing apoptosis is currently under investigation.

Eur. J. Immunol. 1991.21: 1131-1140 We thank B. Clausse, G . Hue and C. Tetaud f o r their valuable technical assistance. We are also very gratefull to Drs. J. Ledbetter and M . Lipinski for their critical evalution of the manuscript. Received December 14, 1990.

5 References 1 MacDonald, R. H. and Lees, R. K., Nature 1990. 343: 642. 2 Smith, C. A., Williams, G. T., Kingston, R., Jenkinson, E. J. and Owen, J. T., Nature 1989. 337: 181. 3 Williams, G.T., Smith, C. A., Spooncer, E., Dexter,T. M. and Taylor, D. R., Nature 1990. 343: 76. 4 Rodriguez-Tarduchy, G., Collins, M. and Lopez-Rivas, A., J. lmrnunol. 1990. 9: 2997. 5 Hasbold, J. and Klaus, G. G. B., Eur. J. lmmunol. 1990. 20: 1685. 6 Benhamou, L. E., Cazenave, F! A. and Sarthou, I?, Eur. J. Immunol. 1990. 20: 1685. 7 Golstein, P. and Smith, E. T., Eur. J. lmmunol. 1976. 6: 31. 8 Searle, J., Ken; J. F. R. and Bishop, C. J., Pathobiol. Ann. 1982. 17: 229. 9 Stein, H., Gerdes, J. and Mason, D.Y., Clin. Haematol. 1982. 11: 531. 10 Gregory, C. D.,Tursz,T., Edwards, C. E,Tetaud, C. ,Thlbot, M., Caillou, B., Rickinson, A. andLipinski, M., J. lmmunol. 1987. 139: 313. 11 Nudelman, E., Kannagi, R., Hakomori, S., Parsons, M., Lipinski, M., Wiels, J., Fellous, M. and Tursz, T., Science 1983. 220: 509. 12 Wiels, J., Fellous, M. and Tursz,T., Proc. Natl. Acad. Sci USA 1981. 78: 6488. 13 Wiels, J., Lenoir, G., Fellous, M., Lipinski, M., Salomon,J. C., Tataud, C. and Tursz, T., Int. J. Cancer 1982. 29: 653. 14 Dorken, B., Moller, I?, Pezzutto, A., Schwartz-Albiez, R. and Moldenhauer, G., Leucocyte typing 1V in Knapp, W. et al. (Eds.). Oxford University Press, Oxford 1989, p. 118. 15 Gregory, C. D., Edwards, C. F., Milner, A.,Wiels, J., Linpinski, M., Rowe, M., Tursz, T. and Rickinson, A. B., lnt. J. Cancer 1988. 42: 213. 16 Clark, E. A., Shu, G. and Ledbetter, J. A., Proc. Natl. Acad. Sci. USA 1985. 82: 1766. 17 Valentine, M. A., Clark, E. A., Shu, G. L., Noms, N. A. and Ledbetter, J., J. lmmunol. 1988. 140: 4071. 18 Ling, N. R., MacLennan, I. C. M. and Mason, D. Y. in McMichael, A. J. et al. (Eds.), Leucocyfe typing 111, Oxford University Press, Oxford 1987, p. 362. 19 Clark, E. A. and Ledbetter, J. A., Proc. Natl. Acad. Sci. USA 1986. 83: 4494. 20 Ledbetter, E. A., Tsu, T. T., Norris, N. A. and Clark, E. A. in McMichael, A. J. et al. (Eds.), Leucocyte Typing lI1, Oxford University Press, Oxford 19, p. 349. 21 Sudgen, B. and Metzenberg, S., J. Virol. 1983. 46: 800. 22 Leprince, C., Richard,Y., Krief, I?, Treton, D., Boucheix, C. and Galanaud, I?, J. lmmunol. 1988. 140: 313. 23 Farace, F., Mitjavila, M. T., Betaieb, A., Dokhelar, M. C., Wiels, J., Finale,Y., Kieffer, N., Breton-Gorius, J. ,Vainchenker, W. and Tursz, T., Cancer Res 1988. 48: 5759. 24 Huet, Groux, S. H., Caillou, B.,Valentin,H., Prieur, A. M. and Bernard, A., J. lmmunol. 1989. 143: 798. 25 Kalil, J. E. and Fellous, M., in Ferrone, S. and David, C. (Eds.), la antigens, vol. 11, CRL, Boca Raton 1982, p. 55. 26 Magnani, J. L., Smith, D. F. and Ginsberg,V., Anal. Biochem. 1980. 109: 39. 27 Kurnick, J.T., Ostberg, L., Stegagno, M., Kimura, A. K., Om, A. and Sjoberg, O., Scand. J. Immunol. 1979. 10: 563. 28 Ling, N. R., Hardie, D., Lowe, J., Johnson, G. D., Khan, M. and MacLennnan, M., lnt. J. Cancer 1989. 43: 112. 29 Hakomori, S. I. and Kannagi, R., J. Natl. Cancer lnst. 1983. 71: 231.

Functional characterization of CD77+ B cells

1139

30 Clark, E. A., Shu, G. L., Luscher, B., Draves. K. E.. Banchereau, J., Ledbetter, J. A. and Valentine, M. A,, J. lmmunol. 1989. 143: 3873. 31 Duvall, E. and Wyllie, A. H., lmmunol. Today 1986. 7: 115. 32 Bhan, A. K., Nadler, L. M., Stashenko, I?, McCluskey, R.T. and Schlossman, S. F., J. Exp. Med. 1981. 154: 737. 33 Weinberg, D. S., Auk, K. A., Gurley. M. and Pinkus, G. S., J. lmmunol. 1986. 137: 1486. 34 Gadol, N., Peacock, M. A. and Auk, K. A., Blood 1988. 71: 1048. 35 Hsu, S. M. and Jaffe, E. S., A m . J. Pathol. 1984. 114: 396. 36 Ledbetter, J. A. and Clark, E. A., Hum. Immunol. 1986. 15: 30. 37 Fyfe, G., Cebra-Thomas, J. A., Mustain. E., Davie, J. M., Alley, C. D. and Nahm, M. H., J. Immunol. 1987. 139: 2187. 38 Thorbecke, G. J., Romano,T. and Lerman, S. I? in Brent, L. and Holbrow, J. (Eds.), Progress in Immunology, vol. 3, North-Holland Co., Amsterdam 1974, p. 25. 39 McLennan, I. C. and Gray, D., Immunol. Rev. 1986. 91: 61. 40 Valles-Ayoub,Y., GovanIII, H. L. and Braun, J.. Blood 1990. 76: 17. 41 Hsu, S. M., Cossman, J. and Jaffe, E. S., A m . J. Clin. Parhol. 1984. 80: 21. 42 Jalkanen, S.T., Bargatze, R. Herron, L. R. and Butcher. E. C.. Eur. J. lmmunol. 1986. 16: 1195. 43 De IosToyos, J., Jalkanen, S. and Butcher. E. C., Blood 1989.2: 751. 44 STamenkovic, I., Amiot, M., Pesando, J. and Seed, B.. Cell 1989. 56: 1057. 45 Miyake, K., Underhill, C. B., Lesley, J. and Kincade. I? W.. J. Ex. Med. 1990. 172: 69. 46 Rose, M. L., Birbeck, M. S. C.,Wallis,V. J., Forrester, J. A. and Davies, A. J. S., Nature 1980. 284: 364. 47 Rose, M. L., Habeshaw, J. A., Kennedy, R., Sloane, J., Withshaw, E. and Davies, A. J. S., Br. J. Cancer 1981. 44: 68. 48 Pallesen, G. and Hager, H. in McMichael, A. J. et al. (Eds.), Leucocyte typing I l l . 1987. Oxford University Press, Oxford 1987, p. 568. 49 Reinherz, E., Kung, L. I?, Goldstein. G. Levey, R. H. and Schlossman, S. F., Proc. Natl. Acad. Sci. USA 1980. 77: 1588. 50 Janossy, G.,Tidman, N., Papageorgiou, E., Kung, I? C. and Goldstein, G., J. lmmunol. 1980. 126: 1608. 51 Sieff, C., Bicknell, D., Calne, G., Robinson, J.. Lam, G. and Greaves, M. E, Blood 1982. 60: 703. 52 Tedder,T. F., Clement, L.T. and Cooper, M. D., Tissue Antigens 1984. 24: 140. 53 Wright, D. H. in Lenoir, G., OConor, G. and Olweny, C. L. M. (Eds.), Burkitt's lymphoma. IARC Scientific Publications, Lyon 1985, p. 37. 54 Haluska, F. G., Finver, S., Tsujimoto. Y. and Croce, C. M., Nature 1985. 324: 158. 55 Neri, R. B., Jaffe, E. S., Braylan, R. C., Nanba, K., Frank, M. M., Ziegler, J. L. and Berard, C.W., N. Engl. J. Med. 1976.295: 685. 56 Stamenkovic, I., Clark, E. A. and Seed, B., EMBO J. 1989.8: 1403. 57 Loetscher, H., Pan, E., Lahm, H. W., Gentz, R., Brockhaus, M., Tabuchi, H. and Lesslauer,W., Cell 1990. 61: 351. 58 Gordon, J., Millsum, M. J., Guy, G. R. and Ledbetter. J. A., Eur. J. Immunol. 1987. 17: 1535. 59 Banchereau, J., De Paoli, I?,VallC,A., Garcia, E. and Rousset, F., Science 1991. 251: 70. 60 Cairns, J., Flores-Romo, L., Millsum, M. J., Guy, G. R., Gillis, S., Ledbetter, J. A. and Gordon, J., Eur. J. Immunol. 1988.18: 349. 61 Levi-Montalcini, R. and Angeletti, I? U., Physiol. Rev. 1963. 48: 534. 62 Otten, U., Ehrhard, P. and Peck, R., Proc. Natl. Acad. Sci. USA 1989. 86: 10059.

1140

M. Mangeney,Y. Richard, D. Coulaud,T. Tursz and J. Wiels

63 Liu,Y. J., Joshua, D. E.,Williams, G.T., Smith, C. A., Gordon, J. and MacLennan, I. C. M., Nature 1989. 342: 929. 64 Schwarting, G. A., Jungalwala, F. B., Chou, D. K., Boyer, A. M. and Yamamoto, M., Dev. Biol. 1987. 120: 65. 65 Andrews, Fenderson, F! B. and Hakomori, S. I., Int. J. Androl. 1987. 10: 95. 66 Karlson, K. A., Annu. Rev. Biochem. 1989.58: 309. 67 Lingwood, A. C., Law, H., Richardson, S., Petric, M., Brunton, J. L., De Grandis, S. and Karmali, M., J. Biol. Chem. 1987. 262: 8834.

Eur. J. Immunol. 1991. 21: 1131-1140

68 Waddell,T., Head, S., Petric, M., Cohen, A. and Linghood, C., Biochem. Biophys. Res. Commun. 1988. 152: 674. 69 Jacewicz, M., Clausen, H., Nudelman, E., Donohue-Rolfe, A. and Keusch, G.T., J. Exp. Med. 1986. 163: 1391. 70 Jacewicz, M., Feldman, H. A., Donohue-Rolfe, A., Balasubralanian, K. A. and Keusch, G. T., J. Infect. Dis. 1989. 159: 881. 71 Wadell,T., Cohen, A. and Linghood, C. A., Proc. Natl. Acad. Sci. USA 1990. 87: 7898.

CD77: an antigen of germinal center B cells entering apoptosis.

We have previously reported that a neutral glycolipid (globotriosylceramide; Gb3) was specifically expressed on Burkitt's lymphoma cells and on a subs...
1MB Sizes 0 Downloads 0 Views