CELLULAR
141,99- 1lo ( 1992)
IMMUNOLOGY
An Antigen Expressed by Avian Neuronal Cells Is also Expressed by Activated T Lymphocytes CATHERINE CORBEL,* HARRY G. BLUESTEIN,?OLIVIER POURQUIE,*
PIERREVAIGOT,* AND NICOLE M. LE DOUARIN* *Institut d’Embryologie Cellulaire et Mol&daire du CNRS et du Collsge de France, 49bis, Avenue de la Belle-Gabrielle 94736 Nogent-sur-Marne Cedex, France; and TDepartment of Medicine, UCSD School of Medicine, San Diego, California 92103 Received October 18, 1991; acceptedDecember 5, 1991 A monoclonal antibody, anti-BEN, initially characterized by its reactivity with an epitope present on the surface of avian bursa epithelial cells and neurons, also reacts with membrane molecules on some hemopoietic cells. In this study we examine BEN expression on lymphoid cells in thymus, spleen, and blood. We demonstrate that BEN is an activation antigen on mature T lymphocytes. It is not expressed on peripheral blood or splenic lymphocytes, but following mitogenic or allogeneic stimulation of blood lymphocytes it appears rapidly on a T cell subpopulation in parallel with the appearance of IL-2 receptors. BEN is also expressed on III-C5 cells, an avian IL-2-dependent permanent T cell line, and on immature CD4+CDS+ thymocytes. BEN is not expressed by resting or actively proliferating B cells. Biochemical analyses of the BEN protein on T lymphoblasts shows that the molecule is similar in size to the BEN molecules on bursa epithelial cells and on neurons. The physicochemical properties of the BEN protein and its tissue distribution differs from other known avian and mammalian T cell activation markers, differentiation antigens, and integrins. Thus BEN is a novel marker of activated T cells in birds. 0 1992 Academic
Press, Inc.
INTRODUCTION A novel membrane glycoprotein, designated BEN, whose pattern of expression is developmentally regulated in the nervous system of avian embryos has been recently described (1). The BEN protein (95- 100 kDa), bears the HNK- 1 epitope and is recognized by a monoclonal antibody (anti-BEN mAb) which was generated in mice against surface determinants of the epithelial component of the bursa of Fabricius. Besides the bursal epithelium, this antibody recognizes in the embryonic nervous system several types of neurons including motoneurons. The BEN glycoprotein is downregulated during synaptogenesis. During those studies it was noted that some hemopoietic cells expressedBEN. More recently, we showed a high concentration of BEN+ cells among thymocytes from chick embryos (2). Other lymphoid/neuronal membrane glycoproteins have been defined in mammals. Two of them, MRC OX-2 and Thy-l antigens, are members of the Ig superfamily (3-5). In human, a glycoprotein CD 44 is expressed in brain and by monocytes, granulocytes, mature T lymphocytes, erythrocytes, fibroblasts, keratino99 000%8749/92 $3.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
100
CORBEL
FIG. 1. BEN immunofluorescence
ET AL.
staining on adult (P70) chick thymus. (X165).
cytes, and B cells (6,7). This molecule has been shown not only to mediate lymphocyte adhesion but also to be involved in the delivery of an activation signal to the cells (8). In this paper, we have analyzed in greater detail the expression of BEN on T lymphocytes. We show that BEN is an activation antigen that is rapidly induced and expressedon both CD4 and CD8 T cell subpopulations. Thus BEN is a newly recognized activation marker of avian T lymphocytes. MATERIAL
AND METHODS
Animals White Leghorn chickens were obtained from local commercial sources. Chickens from the CC inbred line with the B4 MHC haplotype and from a Fl cross between two inbred lines, WB and M, having the B15 and B21 MHC haplotype, respectively, were kindly given to us by Dr. Hlozanek (Institute of Molecular Genetics, Prague, Czechoslovakia). Cells Peripheral blood leucocytes(PBL) were isolated from the supernatant of heparinized blood, centrifuged at 60g for 20 min at room temperature. Thymocytes, splenocytes, and bursa cells were obtained by teasing the organ on a 200-mesh stainlesssteel screen in Hank’s balanced salt solution (HBSS). White cells of spleen were separated from erythrocytes and dead cells on a FicollPaque gradient (Pharmacia). Cells were washed three times in HBSS.
FIG. 2. Cytofluorographs of allogeneic activated PBL stained with anti-CT3 and anti-BEN mAb’s. Cells were stained indirectly with anti-CT3, anti-BEN, or HBSS/FCS (control) followed by PE-conjugated goat antibodies to mouse Ig. The location of small unstimulated T lymphocytes is to the left of the vertical dotted line, while the enlarged blasts are on the right. Positive fluorescence is above the horizontal dotted line.
AN AVIAN
T LYMPHOCYTE
Forward
ACTIVATION
scatter
ANTIGEN
101
102
CORBEL ET AL. 60
1
50 40 30 20 10 E
Days FIG. 3. The time courseof the expression of BEN and IL-2R on ConA-activated T cells. PBL were stimulated with ConA and analyzed each day for BEN (W)and IL-2R (INN-CHl6 mAb was used, 0) expression by immunofluorescence.
III-C5, a T helper-activated cell line, was cultured in the presenceof 10%interleukin2 (IL-2) containing medium (9). MSB 1, a Marek’s diseasevirus (MDV)-transformed T lymphoblast cell line ( lo), and RPL 12, a Rous-associatedavian virus (RAV)-induced lymphoblastoid B cell line (1 l), were grown in DMEM medium (GIBCO) supplemented with 10% fetal calf serum, 1% chicken serum, 2 ~ML-glutamine, and 50 pg/ml penicillin/streptomycin. Cell lines were cultured at 40°C 5% C02. T Lymphoblasts Stimulation Mitogenic activation of PBL was carried out by culturing PBL (3 X lo6 cells/ml) in the presence of 10 pg/ml concanavalin A (Con A) (Pharmacia Fine Chemicals, Sweden) as described previously (9). A kinetic study for BEN expression appearance was performed on PBL stimulated with Con A at Day 0. Cells were obtained after 24, 48, 72, and 96 hr of incubation and stained for BEN and IL-2R expression (using INN-CH 16 mAb). Allogeneic T lymphoblasts were obtained in MLC after stimulation of B 15/B2 1 PBL with B4 PBL treated with mitomycin C. The cultures were performed in flasks with an equal number of responder and stimulator cells (5 X 106/ml). Iscove’smodified Dulbecco’s medium with 2-mercaptoethanol (5 X 1O-5M) supplemented with 5% FCS and 1% chicken serum was used. The cultures kept at 40°C 5% CO2, were harvested 6 days later. Tissue Sections Lymphoid organs were dissectedout and immediately frozen in liquid nitrogen for immunocytochemistry. Tissue sections (5 pm thickness) were cut with a cryostat and immunostained after fixation for 6 min in 100%ethanol at -20°C. They were further processedfor immunofluorescent staining as described below. Antibodies Monoclonal antibodies used for this study were: -anti-BEN, produced in our laboratory (1);
AN AVIAN
T LYMPHOCYTE
ACTIVATION
ANTIGEN
103
-CT3 (anti-chicken CD3 (12)), CT4 (anti-chicken CD4 (13)) CT8 (anti-chicken CDS (I 3)) a gift from Dr. M. Cooper, Birmingham, Alabama; -HNK- 1 ( 14); -INN-CH16, anti-IL-2 receptor of chicken, a gift from Dr. K. Hala, Innsbruck, Austria (15, 16). A polyclonal rabbit anti-chicken IgM antibody was prepared by us. Immunojluorescence Cell suspensions and tissue sections were processed for immunofluorescent staining as described previously (9). Goat anti-mouse Ig or IgGl were conjugated with fluorescein isothiocyanate (FITC) or phycoerythrin (PE) (Southern Biotechnology Associates). Goat anti-rabbit Ig conjugated with FITC or tetramethylrhodamine isocyanate (TRITC) was also used. Two-color immunofluorescence was performed by sequential staining of tissue sections with antibodies labeled indirectly. Suspensions of labeled cells were analyzed by automated flow cytometry on a FACS 440 (Becton Dickinson). For two-color analysis by flow cytometry, the following reagents were used: mAb anti-BEN, FITC-goat anti-mouse IgGl, PE-streptavidin (Amersham), biotinylatedCT3, PE-CT4, and PE-CT8 (kindly given to us by Dr. M. Cooper). Iodination and Immunoprecipitation
of Cell Surface Proteins
Activated T cells from the III-C5 cell line were radioiodinated with Na12’I by the lactoperoxidase technique as previously described ( 17). Cells were lysed in 0.15 M NaCl, 0.01 A4 Tris, pH 8, 0.5% NP40 (TBS-NP40) containing 0.06 mMN-a-ptosyl+lysyl chloromethyl ketone, 0.06 ~ML- 1-tosylamide2-phenylethyl chloromethyl ketone, and 0.2 mM phenylmethylsulfonyl fluoride for 30 min at 4°C. All protease inhibitors were obtained from Sigma Chemical Co. (St. Louis, MO). Following centrifugation, aliquots were precleared by incubation with Sepharose-protein A (Pharmacia) for 15 min at 4°C followed by centrifugation for 4 min. Antibodies were added to the lysates and allowed to bind for 30 min at 4°C. The immune complexes in the extracts were collected by incubation (1 hr, 4°C) with affinity-purified rabbit anti-mouse Ig bound to Sepharose-protein A. The conjugates were then pelleted by centrifugation, washed twice with TBS-NP40, layered onto 1 ml of 20% sucrose-TBS-NP40, and repelleted. Washed immunoprecipitates were eluted and denatured with 25 ~1 1% SDS and 5% 2-ME by heating at 100°C for 2 min and then analyzed along with molecular weight markers by SDS-polyacrylamide gel electrophoresis (7.5%, (18)). Gels were fixed, stained, and autoradiographed for l-10 days at -80°C. RESULTS BEN Expression on Thymocytes but Not on Quiescent Lymphocytes Immunofluorescence analysis both on sections and in cell suspensions shows that anti-BEN mAb reacts with thymocytes (Fig. 1) but not with splenic and blood lymphocytes from adult chickens. When cell suspensions of 4-week-old thymus were examined by two-color immunofluorescence with anti-BEN and either CT4 or CT8 mAb, an equal percentage (70-75%) of BEN+ cells coexpressed CT4 and CT8. Among
104
CORBEL ET AL.
the CT4+ and CT8 populations, 28% in both casesdid not express BEN. This is the mean of three experiments analyzed by FAG. By contrast, only 5% of splenic lymphocytes and lessthan 1%of blood lymphocytes from a 4-week-old chick expressedBEN. BEN Expression by Activated T Lymphocytes To determine whether BEN expression could be induced on T cells after activation, peripheral blood T cells were stimulated in vitro with Con A or alloantigen in mixed leucocyte culture (MLC). BEN expression by MLC-activated T cells is shown in Fig. 2, where fluorescence was compared to forward scatter as a measure of cell size. Small unactivated CT3+ T lymphocytes did not exhibit BEN reactivity, but the enlarged T blasts resulting from allogeneic stimulation were both CT3 and BEN positive. To determine the time course of BEN induction and surface expression on T cells after activation, a kinetic study was performed after Con A stimulation. The expression of BEN was measured along with the expression of IL-2R. Neither BEN nor IL-2R was detected in the resting state (Day 0). An increase in both was observed by Day 1. BEN expression was maximal at Day 3 and thereafter declined while IL-2R remained elevated through Day 4 (Fig. 3). To determine whether there was T cell subset restriction of BEN expression, T cells were cultured either with Con A or with alloantigen for 3 or 6 days, respectively, and were double stained with anti-BEN mAb and CTCPE or CT8-PE for analysis by FACS. In two separateexperiments with Con A T blasts,BEN was found to be expressed on 70 to 90% of both CT4+ and CT8+ cells. For allogeneic T blasts, BEN was expressed on 46% and 50% of CT4+ and CT8+ cells, respectively (Fig. 4). Therefore, BEN expression is not restricted to the helper CD4 or cytotoxic CD8 subset of T lymphocytes. Two proliferating T lymphoblast cell lines differing in surface marker expression were analyzed for BEN reactivity. MSBl cells (CT3+, CT4+, TcR @ +(TcR2), IL-
BEN+
CT4+ BEN+CT4+
CT8+ BEN+CTB+ 10
20 Positive
30 ceils
40
0
(%I
FIG. 4. Induction of BEN antigen expression by CT4+ and CT8+ peripheral blood T lymphocytes after
allogeneic stimulation in MLC. Single and double labeling of allogeneic T blasts were analyzed by FACS. Data are expressedin percentagesof large blasts cells after software gating on forward scatter as illustrated in Fig. 2.
AN AVIAN T LYMPHOCYTE
ACTIVATION
ANTIGEN
105
2R-, class II-) were BEN negative. However, III-C5 cells having the phenotype of activated helper T cells (CT3+, CT4+, TcR2+, IL-2R+, classII+ (9)) expressedthe BEN antigen as shown below. Functional experiments with the Mab anti-BEN have been performed. But it neither stimulates nor blocks T cell activation, whatever its concentration and the time it was added to the cultures. Biochemical Analysis of the BEN Glycoprotein Characterization of the molecular size of the BEN protein expressed on activated T cells was done by immunoprecipitation of the molecule from cultured III-C5 cells using the anti-BEN mAb. Intact living III-C5 cells were surface radioiodinated by the lactoperoxidase method. SDS-PAGE analysis of the immunoprecipitate from the lysate of the labeled cells, in reducing conditions, revealed a 95-kDa molecular mass for BEN on T cells (Fig. 5, lane l), thus it appears to have approximately the same molecular weight as found previously on bursal epithelial and nervous cells (1). IL-2R immunoprecipitated from the same III-C5 lysates using mAb INN-CH 16 has an apparent molecular mass of 65 kDa (Fig. 5, lane 2). Thus, as expected from the previously reported molecular size of IL-2R (15), the BEN protein is distinct from the IL-2 receptor. We have shown that BEN protein purified from spinal cord, brain, and thymus carries the HNK- 1 epitope ( 1, 2). The mAb HNK- 1 which is directed at an epitope on glycosidic side chains immunoprecipitates 95-kDa molecules from III-C5 cells which comigrate with BEN-precipitated molecules on SDS-PAGE (Fig. 5, lane 3). Thus, BEN on proliferating T cells also appears to carry the HNK-1 epitope. Lack of BEN Expression by B Lymphocytes Although, as shown previously, the BEN antigen is expressedon epithelial cells of the bursa of Fabricius from E9 onward (l), BEN is not expressed by the bursal he123 kD
-68
FIG. 5. Immunoprecipitation of BEN, IL-2R, and HNK- 1 from the surfaceof III-C5 lymphoblasts. Lysates of surface-labeledIII-C5 cells were precipitated by mAb’s, anti-BEN (lane 1).INN-CH 16 (lane 2), and HNK1 (lane 3), and submitted to SDS-PAGE as described under Material and Methods.
106
CORBEL ET AL.
mopoietic cells (Fig. 6). Double labeling of El6 chick bursa with the BEN anti-mAb and an anti-chicken IgM polyclonal antibody shows that BEN expression is restricted to the epithelium lining the bursal lumen and the follicular epithelium. The bursal follicles contains many cells exhibiting IgM cytoplasmic and surfaceimmunoreactivity. Intra- and extrafollicular hemopoietic cells are BEN negative (Figs. 6a, 6b). Later in embryonic development and after hatching, BEN is expressedby the epithelium delineating the cortex from the medulla. The network of medullary epithelial cells which are BEN+ also stain with anti-cytokeratin antibodies. They are IgM negative. IgMpositive B lymphocytes develop within the medullary network as well as in the cortex and are BEN negative (Figs. 6c, 6d). Double immunofluorescence analysis of IgM- and BEN-positive cells in sections of other B lymphocyte-containing organs was also performed at different time points during the embryonic and adult life of chicks. In caecatonsils and Harderian glands, as in the bursa, B lymphocytes did not expressthe BEN antigen although the epithelial cells lining the lumen were positive. Moreover, B cells in these organs are Ig-secreting B lymphocytes, indicating that activated B cells do not express the BEN antigen. Actively proliferating B lymphoblasts from the RPL12 line also did not stain with anti-BEN mAb, confirming the lack of BEN expression by activated B lymphocytes. DISCUSSION In this paper, we describe the expression of avian protein BEN on the surface of peripheral T lymphocytes during cell activation. The BEN glycoprotein is expressed by thymocytes but not by resting T cells from blood and spleen. In the thymus, the double positive CT4+CT8+ cells are BEN positive since this population represents the majority of thymocytes ( 13). Double immunofluorescence staining indicates that the minor population of single CT4’ or CT8+ thymocytes do not expressBEN. This correlates with the lack of BEN expression by peripheral CT4 or CT8 positive T lymphocytes. However, upon mitogenic or allogeneic activation peripheral T cells express BEN at their surface. Activation and proliferation is accompanied by the expression of the avian homologue of the CD25 or IL-2R (15) and the class II antigen (B-L antigen) (19, 20). In this species,no other T-cell activation antigen has been described so far. We show that BEN is an activation antigen that is rapidly induced and expressed on both CT4 and CT8 T-cell subpopulations. Once T cells expressIL-2R they coexpress the BEN molecule. This is shown on mitogen-activated peripheral T lymphoblasts and on III-C5 cells, a cell line which has the phenotype of activated T lymphocytes expressing both IL-2R and class II antigen (9). BEN is not expressedon proliferating T cells of the MSBl line which expressesneither IL-2R nor class II antigen.
FIG. 6. Double labeling of bursa of Fabricius with anti-IgM and anti-BEN antibodies. (a, b) Section of E 16 bursa of Fabricius. X290. (a) Anti-IgM antibodies conjugated with TRITC reveal immunostaining of intrafollicular B lymphocytes. (b) Labeling of the samesectionby anti-BEN mAb was revealedby a fluorescein conjugate. BEN expression is restricted to the epithelial cells. No immunoreactivity was seen in the mesenchyme. (c, d) Section of 2-month bursa of Fabricius. X 180. (c) B lymphocytes are labeled by anti-IgM antibodies. (d) BEN antigen is expressedby cells of the basement membrane delineating the corticomedullary junction and the epithelial cells inside the follicle.
AN AVIAN
T LYMPHOCYTE
ACTIVATION
ANTIGEN
107
108
CORBEL
ET AL.
Among lymphocytes there is a specificity for T cells but BEN expression is not restricted to this cell lineage. As shown here and previously, the BEN molecule is expressedby epithelial cells of the bursa of Fabricius and also by subpopulations of nerve cells. The molecular weight of the BEN protein from activated T lymphocytes (loo-95 kDa) is similar to that described for bursal epithelial and neural cells (1). Recently, we have shown that other hemopoietic cells are BEN positive and we described differences in molecular forms of the BEN protein present at the surface of different cell types (2). Therefore, the biochemical characteristics of BEN glycoprotein on activated T cells such as III-C5 cells have been compared to the ones defined for tissues of different origin. On hemopoietic cells, such as embryonic spleen at 18 days of incubation and 3-week-old thymus, the purified protein exhibited an intermediate molecular weight between epithelial (110 kDa) and neural (95 kDa) forms (2). BEN differs from others known T cell activation antigens. Most, such as CD26, a peptidase (2 l), or CD27 (22), are present on resting cells and their expression is upregulated after activation. BEN, on the other hand, can be detected on T cells only in concert with their activation. Some integrins such as the murine vitronectin receptor and the “very late antigens” VLA-1 and VLA-2, in human, are difficult to detect on resting T cells but are well expressedafter T cell activation (17, 23-25). They differ from BEN in kinetics of their expression and in their molecular characteristics. BEN is an “early” activation marker appearing by 24 hr while the VLAs appear “late,” several days after mitogen stimulation. Furthermore, the integrins are noncovalently linked heterodimers (26) while BEN is a monomer. On activated T cells, the BEN molecule expressesthe HNK- 1 carbohydrate epitope. The occurrence of this epitope suggeststhat the glycoprotein BEN might be involved in cell-cell interactions or adhesion since several cell adhesion molecules expressthe HNK-1 carbohydrate epitope (27). Moreover, this epitope is also shared by cells of the nervous and immune systems(28). In human, the mAb HNK-1 reacts with a cell surface antigen of unknown properties on natural killer cells (14). In avian species too, subpopulations of lymphocytes express HNK-1 epitope during embryonic development (29). In addition to the integrins mentioned above, cell adhesion molecules of the immune system may belong to other large groups known as the immunoglobulin supergene family and the selectin family. Many of the molecules from the immunoglobulin supergene family are involved in cell adhesion. One of those, intercellular adhesion molecule 1 (ICAM- 1 or CD54) is a highly glycosylated cell surface protein and it is a ligand of LFA- 1 (30). ICAM- 1 is mainly expressedon macrophagesand myeloid cells as well as activated lymphocytes, but also by cells of nonhemopoietic origin including epithelial and endothelial cells and fibroblasts. Interestingly, ICAM- I glycoprotein, like BEN, displays molecular weight heterogeneity in different cell types (3 1). Selectin (32) or LEC-CAM (33) include lymphocyte homing receptor (MEL-14), endothelial leucocyte adhesion molecule 1 (ELAM- l), and granule membrane protein (GMP- 140). MEL- 14, which mediates lymphocyte adhesion to specialized secondary organ endothelial cells, has caught our attention becauseit is expressedby granulocytes, monocytes, and lymphocyte subsetsand becauseit is a glycoprotein of 90-100 kDa. However, activation of T cells results in downregulation of surface homing receptors (34) and it must be noted that BEN antigen is not found on endothelial cells. Thus based on tissue distribution and biochemical characteristics, there is no close relationship between any mammalian molecule classified among the clusters of dif-
AN AVIAN T LYMPHOCYTE
ACTIVATION
ANTIGEN
109
ferentiation and the BEN molecule. Isolation of its structural gene, which might allow us to determine BEN’s role in immune functions is in progress. ACKNOWLEDGMENTS The technical assistanceof Mrs. P. Belo and K. Heydon is acknowledged. We thank Dr. M. Cooper for providing CT4, CT8, and CT3 monoclonal antibodies, E. Bourson for the typescript, and S. Goumet and Y. Rantier for the illustrations. This work was supported by the Centre National de la Recherche Scientifique, the Institut National de la Sante et de la Recherche Medicale and by grants from the following foundations: the Association pour la Recherche MCdicaIe and the Ligue Nationale Fran&e contre le Cancer. H.G.B. was partially supported by a Fulbright Research Scholar Award and ResearchGrant AR30036 from the U.S. National Institutes of Health.
REFERENCES 1. Pourquit, O., Coltey, M., Thomas, J.-L., and Le Douarin, N. M., Development 109, 743, 1990. 2. Corbel, C., Cormier, F., Pourquit, O., Bluestein, H. G., and Le Douarin, N. M., 1991. Submitted for publication. 3. Clark, M. J., Gagnon, J., Williams, A. F., and Barclay, A. N., EMBU J. 4, 113, 1985. 4. Barclay, N. A., Letarte-Muirhead, M., Williams, A. F., and Faulkes, R. A., Nature 263, 563, 1976. 5. Williams, A. F., and Gagnon, J., Science 216,696, 1982. 6. Dalchau, R., Kirkley, J., and Fabre, J. W., Eur. J. Immunol. 10, 745, 1980. 7. Haynes, B. F., Harden, E. A., Telen, M. J., Hemler, M. E., Strominger, J. L., Palker, T. J., Scearce, R. M., and Eisenbarth, G. S., J. Irnmunol. 131, 1195, 1983. 8. Huet, S., Groux, H., Caillou, B., Valentin, H., Prieur, A.-M., and Bernard, A., J. Immunol. 143, 798, 1989. 9. Corbel, C., and Thomas, J.-L., Dev. Comp. Immunol. 14, 439, 1990. 10. Akiyama, Y., and Kato, S., &ken. J. 17, 105, 1974. 11. Okazaki, W., Witter, R. L., Romero, C.. Nazerian, K., Sharma, J. M., Fadly, A., and Ewert, D., Avian Pathol. 9, 31 I, 1980. 12. Chen, C. L. H., Lanier Ager, L., Gartland, L., and Cooper, M. D., J. Exp. Med. 164, 375, 1986. 13. Chan, M. M., Chen, C. L. H., Lanier, Ager, L., and Cooper, M. D., J. Immunol. 140,2133, 1988. 14. Abo, T., and Balch, C., J. Immunol. 127, 1024, 1981. 15. Hala, K., Schauenstein, K., Neu, N., Kramer, G., Wolf, H., Bock, G., and Wick, G., Eur. J. Immunol. 16, 1331, 1986.
16. Schauenstein,K., Kramer, G., Hala, K., Bock, G., and Wick, G., Dev. Comp. Immunol. 12,823, 1988. 17. Pischel, K. D., Bluestein, H. G., and Woods, V. L., Jr., J. Exp. Med. 164, 393, 1986. 18. Laemmli, U. K. Nature 227,680, 1970. 19. Guillemot, F. P., Oliver, P. D., Peault, B. M., and Le Douarin, N. M., J. Exp. Med. 160, 1803, 1984. 20. Ewert, D. L., Munchus, M. S., Chen, C. L. H., and Cooper, M. D., J. Immunol. 132,2524, 1984. 21. Stein, H., Schwarting, R., and Niedobitek, G., In “Leucocyte Typing IV” (W. Knapp, et al., Eds.), p. 412. Oxford Univ. Press,Oxford, 1989. 22. de Jong, R., Loenen, W. A. M., Brouwer, M., van Emmerik, L., de Vries, E. F. R., Borst, J., and van Lier, R. A. W., J. Immunol. 146, 2488, 1991. 23. Moulder, K., Roberts, K., Shevach, E. M., and CoIIigan, J. E., J. Exp. Med. 173, 343, 1991. 24. Hemler, M. E., Sanchez-Madrid, F., Flotte, T. J., Krensky, A. M., Burakoff, S. J., Bhan, A. K., Springer, T. A., and Strominger, J. L., J. Immunol. 132, 3011, 1984. 25. Hemler, M. E., Jacobson,J. G., Brenner, M. B., Mann, D., and Strominger, J. L., Eur. J. Immunol. 15, 502, 1985. 26. Hynes, R. O., Cell48, 549, 1987.
27. Schachner,M., Antonicek, H., Fahrig, T., Faissner,A., Fischer, G., Ktlnemund, V., Martini, R., Meyer, A., Persohn, E., Pollerberg, E., Prosbstmeier, R., Sadoul, R., Seilheimer, B., and Thor, G., In “Morphoregulatory Molecules” (G. M. Edelman, B. A. Cunningham, and J.-P. ThiCry, Eds.), p. 443. Wiley-Interscience, New York, 1990. 28. Lipinski, M., Braham, K., Caillaud, J. M., Carlu, C., and Tursz, T., J. Exp. Med. 158, 1775, 1983.
110
CORBEL ET AL.
29. Pea&, B., Chen, C. H., Cooper, M. D., Barbu, M., Lipinski, M., and Le Douarin, N. M., Proc. N&l. Acad. Sci. USA 84, 814, 1987. 30. Marlin, S. D., and Springer, T. A., Cell 51, 8 13, 1987. 31. Dustin, M. L., Rothlein, R., Bhan, A. K., Dinarello, C. A., and Springer, T. A., J. Immunol. 137,245, 1986. 32. Geng, J. G., Bevilacqua Moore, K. L., McIntyre, T. M., Prescott, S. M., Kim, J. M., Bliss, G. A., Zimmerman, G. A., and R. P. McEver, Nature (London). 343, 747, 1990. 33. Stoolman, L. M., Cell 56, 907, 1989. 34. Jung, T. M., Gallatin, W. M., Weissman, I. L., and Dailey, M. O., J. Zmmunol. 141, 4110, 1988.
141,99- 1lo ( 1992)
IMMUNOLOGY
An Antigen Expressed by Avian Neuronal Cells Is also Expressed by Activated T Lymphocytes CATHERINE CORBEL,* HARRY G. BLUESTEIN,?OLIVIER POURQUIE,*
PIERREVAIGOT,* AND NICOLE M. LE DOUARIN* *Institut d’Embryologie Cellulaire et Mol&daire du CNRS et du Collsge de France, 49bis, Avenue de la Belle-Gabrielle 94736 Nogent-sur-Marne Cedex, France; and TDepartment of Medicine, UCSD School of Medicine, San Diego, California 92103 Received October 18, 1991; acceptedDecember 5, 1991 A monoclonal antibody, anti-BEN, initially characterized by its reactivity with an epitope present on the surface of avian bursa epithelial cells and neurons, also reacts with membrane molecules on some hemopoietic cells. In this study we examine BEN expression on lymphoid cells in thymus, spleen, and blood. We demonstrate that BEN is an activation antigen on mature T lymphocytes. It is not expressed on peripheral blood or splenic lymphocytes, but following mitogenic or allogeneic stimulation of blood lymphocytes it appears rapidly on a T cell subpopulation in parallel with the appearance of IL-2 receptors. BEN is also expressed on III-C5 cells, an avian IL-2-dependent permanent T cell line, and on immature CD4+CDS+ thymocytes. BEN is not expressed by resting or actively proliferating B cells. Biochemical analyses of the BEN protein on T lymphoblasts shows that the molecule is similar in size to the BEN molecules on bursa epithelial cells and on neurons. The physicochemical properties of the BEN protein and its tissue distribution differs from other known avian and mammalian T cell activation markers, differentiation antigens, and integrins. Thus BEN is a novel marker of activated T cells in birds. 0 1992 Academic
Press, Inc.
INTRODUCTION A novel membrane glycoprotein, designated BEN, whose pattern of expression is developmentally regulated in the nervous system of avian embryos has been recently described (1). The BEN protein (95- 100 kDa), bears the HNK- 1 epitope and is recognized by a monoclonal antibody (anti-BEN mAb) which was generated in mice against surface determinants of the epithelial component of the bursa of Fabricius. Besides the bursal epithelium, this antibody recognizes in the embryonic nervous system several types of neurons including motoneurons. The BEN glycoprotein is downregulated during synaptogenesis. During those studies it was noted that some hemopoietic cells expressedBEN. More recently, we showed a high concentration of BEN+ cells among thymocytes from chick embryos (2). Other lymphoid/neuronal membrane glycoproteins have been defined in mammals. Two of them, MRC OX-2 and Thy-l antigens, are members of the Ig superfamily (3-5). In human, a glycoprotein CD 44 is expressed in brain and by monocytes, granulocytes, mature T lymphocytes, erythrocytes, fibroblasts, keratino99 000%8749/92 $3.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
100
CORBEL
FIG. 1. BEN immunofluorescence
ET AL.
staining on adult (P70) chick thymus. (X165).
cytes, and B cells (6,7). This molecule has been shown not only to mediate lymphocyte adhesion but also to be involved in the delivery of an activation signal to the cells (8). In this paper, we have analyzed in greater detail the expression of BEN on T lymphocytes. We show that BEN is an activation antigen that is rapidly induced and expressedon both CD4 and CD8 T cell subpopulations. Thus BEN is a newly recognized activation marker of avian T lymphocytes. MATERIAL
AND METHODS
Animals White Leghorn chickens were obtained from local commercial sources. Chickens from the CC inbred line with the B4 MHC haplotype and from a Fl cross between two inbred lines, WB and M, having the B15 and B21 MHC haplotype, respectively, were kindly given to us by Dr. Hlozanek (Institute of Molecular Genetics, Prague, Czechoslovakia). Cells Peripheral blood leucocytes(PBL) were isolated from the supernatant of heparinized blood, centrifuged at 60g for 20 min at room temperature. Thymocytes, splenocytes, and bursa cells were obtained by teasing the organ on a 200-mesh stainlesssteel screen in Hank’s balanced salt solution (HBSS). White cells of spleen were separated from erythrocytes and dead cells on a FicollPaque gradient (Pharmacia). Cells were washed three times in HBSS.
FIG. 2. Cytofluorographs of allogeneic activated PBL stained with anti-CT3 and anti-BEN mAb’s. Cells were stained indirectly with anti-CT3, anti-BEN, or HBSS/FCS (control) followed by PE-conjugated goat antibodies to mouse Ig. The location of small unstimulated T lymphocytes is to the left of the vertical dotted line, while the enlarged blasts are on the right. Positive fluorescence is above the horizontal dotted line.
AN AVIAN
T LYMPHOCYTE
Forward
ACTIVATION
scatter
ANTIGEN
101
102
CORBEL ET AL. 60
1
50 40 30 20 10 E
Days FIG. 3. The time courseof the expression of BEN and IL-2R on ConA-activated T cells. PBL were stimulated with ConA and analyzed each day for BEN (W)and IL-2R (INN-CHl6 mAb was used, 0) expression by immunofluorescence.
III-C5, a T helper-activated cell line, was cultured in the presenceof 10%interleukin2 (IL-2) containing medium (9). MSB 1, a Marek’s diseasevirus (MDV)-transformed T lymphoblast cell line ( lo), and RPL 12, a Rous-associatedavian virus (RAV)-induced lymphoblastoid B cell line (1 l), were grown in DMEM medium (GIBCO) supplemented with 10% fetal calf serum, 1% chicken serum, 2 ~ML-glutamine, and 50 pg/ml penicillin/streptomycin. Cell lines were cultured at 40°C 5% C02. T Lymphoblasts Stimulation Mitogenic activation of PBL was carried out by culturing PBL (3 X lo6 cells/ml) in the presence of 10 pg/ml concanavalin A (Con A) (Pharmacia Fine Chemicals, Sweden) as described previously (9). A kinetic study for BEN expression appearance was performed on PBL stimulated with Con A at Day 0. Cells were obtained after 24, 48, 72, and 96 hr of incubation and stained for BEN and IL-2R expression (using INN-CH 16 mAb). Allogeneic T lymphoblasts were obtained in MLC after stimulation of B 15/B2 1 PBL with B4 PBL treated with mitomycin C. The cultures were performed in flasks with an equal number of responder and stimulator cells (5 X 106/ml). Iscove’smodified Dulbecco’s medium with 2-mercaptoethanol (5 X 1O-5M) supplemented with 5% FCS and 1% chicken serum was used. The cultures kept at 40°C 5% CO2, were harvested 6 days later. Tissue Sections Lymphoid organs were dissectedout and immediately frozen in liquid nitrogen for immunocytochemistry. Tissue sections (5 pm thickness) were cut with a cryostat and immunostained after fixation for 6 min in 100%ethanol at -20°C. They were further processedfor immunofluorescent staining as described below. Antibodies Monoclonal antibodies used for this study were: -anti-BEN, produced in our laboratory (1);
AN AVIAN
T LYMPHOCYTE
ACTIVATION
ANTIGEN
103
-CT3 (anti-chicken CD3 (12)), CT4 (anti-chicken CD4 (13)) CT8 (anti-chicken CDS (I 3)) a gift from Dr. M. Cooper, Birmingham, Alabama; -HNK- 1 ( 14); -INN-CH16, anti-IL-2 receptor of chicken, a gift from Dr. K. Hala, Innsbruck, Austria (15, 16). A polyclonal rabbit anti-chicken IgM antibody was prepared by us. Immunojluorescence Cell suspensions and tissue sections were processed for immunofluorescent staining as described previously (9). Goat anti-mouse Ig or IgGl were conjugated with fluorescein isothiocyanate (FITC) or phycoerythrin (PE) (Southern Biotechnology Associates). Goat anti-rabbit Ig conjugated with FITC or tetramethylrhodamine isocyanate (TRITC) was also used. Two-color immunofluorescence was performed by sequential staining of tissue sections with antibodies labeled indirectly. Suspensions of labeled cells were analyzed by automated flow cytometry on a FACS 440 (Becton Dickinson). For two-color analysis by flow cytometry, the following reagents were used: mAb anti-BEN, FITC-goat anti-mouse IgGl, PE-streptavidin (Amersham), biotinylatedCT3, PE-CT4, and PE-CT8 (kindly given to us by Dr. M. Cooper). Iodination and Immunoprecipitation
of Cell Surface Proteins
Activated T cells from the III-C5 cell line were radioiodinated with Na12’I by the lactoperoxidase technique as previously described ( 17). Cells were lysed in 0.15 M NaCl, 0.01 A4 Tris, pH 8, 0.5% NP40 (TBS-NP40) containing 0.06 mMN-a-ptosyl+lysyl chloromethyl ketone, 0.06 ~ML- 1-tosylamide2-phenylethyl chloromethyl ketone, and 0.2 mM phenylmethylsulfonyl fluoride for 30 min at 4°C. All protease inhibitors were obtained from Sigma Chemical Co. (St. Louis, MO). Following centrifugation, aliquots were precleared by incubation with Sepharose-protein A (Pharmacia) for 15 min at 4°C followed by centrifugation for 4 min. Antibodies were added to the lysates and allowed to bind for 30 min at 4°C. The immune complexes in the extracts were collected by incubation (1 hr, 4°C) with affinity-purified rabbit anti-mouse Ig bound to Sepharose-protein A. The conjugates were then pelleted by centrifugation, washed twice with TBS-NP40, layered onto 1 ml of 20% sucrose-TBS-NP40, and repelleted. Washed immunoprecipitates were eluted and denatured with 25 ~1 1% SDS and 5% 2-ME by heating at 100°C for 2 min and then analyzed along with molecular weight markers by SDS-polyacrylamide gel electrophoresis (7.5%, (18)). Gels were fixed, stained, and autoradiographed for l-10 days at -80°C. RESULTS BEN Expression on Thymocytes but Not on Quiescent Lymphocytes Immunofluorescence analysis both on sections and in cell suspensions shows that anti-BEN mAb reacts with thymocytes (Fig. 1) but not with splenic and blood lymphocytes from adult chickens. When cell suspensions of 4-week-old thymus were examined by two-color immunofluorescence with anti-BEN and either CT4 or CT8 mAb, an equal percentage (70-75%) of BEN+ cells coexpressed CT4 and CT8. Among
104
CORBEL ET AL.
the CT4+ and CT8 populations, 28% in both casesdid not express BEN. This is the mean of three experiments analyzed by FAG. By contrast, only 5% of splenic lymphocytes and lessthan 1%of blood lymphocytes from a 4-week-old chick expressedBEN. BEN Expression by Activated T Lymphocytes To determine whether BEN expression could be induced on T cells after activation, peripheral blood T cells were stimulated in vitro with Con A or alloantigen in mixed leucocyte culture (MLC). BEN expression by MLC-activated T cells is shown in Fig. 2, where fluorescence was compared to forward scatter as a measure of cell size. Small unactivated CT3+ T lymphocytes did not exhibit BEN reactivity, but the enlarged T blasts resulting from allogeneic stimulation were both CT3 and BEN positive. To determine the time course of BEN induction and surface expression on T cells after activation, a kinetic study was performed after Con A stimulation. The expression of BEN was measured along with the expression of IL-2R. Neither BEN nor IL-2R was detected in the resting state (Day 0). An increase in both was observed by Day 1. BEN expression was maximal at Day 3 and thereafter declined while IL-2R remained elevated through Day 4 (Fig. 3). To determine whether there was T cell subset restriction of BEN expression, T cells were cultured either with Con A or with alloantigen for 3 or 6 days, respectively, and were double stained with anti-BEN mAb and CTCPE or CT8-PE for analysis by FACS. In two separateexperiments with Con A T blasts,BEN was found to be expressed on 70 to 90% of both CT4+ and CT8+ cells. For allogeneic T blasts, BEN was expressed on 46% and 50% of CT4+ and CT8+ cells, respectively (Fig. 4). Therefore, BEN expression is not restricted to the helper CD4 or cytotoxic CD8 subset of T lymphocytes. Two proliferating T lymphoblast cell lines differing in surface marker expression were analyzed for BEN reactivity. MSBl cells (CT3+, CT4+, TcR @ +(TcR2), IL-
BEN+
CT4+ BEN+CT4+
CT8+ BEN+CTB+ 10
20 Positive
30 ceils
40
0
(%I
FIG. 4. Induction of BEN antigen expression by CT4+ and CT8+ peripheral blood T lymphocytes after
allogeneic stimulation in MLC. Single and double labeling of allogeneic T blasts were analyzed by FACS. Data are expressedin percentagesof large blasts cells after software gating on forward scatter as illustrated in Fig. 2.
AN AVIAN T LYMPHOCYTE
ACTIVATION
ANTIGEN
105
2R-, class II-) were BEN negative. However, III-C5 cells having the phenotype of activated helper T cells (CT3+, CT4+, TcR2+, IL-2R+, classII+ (9)) expressedthe BEN antigen as shown below. Functional experiments with the Mab anti-BEN have been performed. But it neither stimulates nor blocks T cell activation, whatever its concentration and the time it was added to the cultures. Biochemical Analysis of the BEN Glycoprotein Characterization of the molecular size of the BEN protein expressed on activated T cells was done by immunoprecipitation of the molecule from cultured III-C5 cells using the anti-BEN mAb. Intact living III-C5 cells were surface radioiodinated by the lactoperoxidase method. SDS-PAGE analysis of the immunoprecipitate from the lysate of the labeled cells, in reducing conditions, revealed a 95-kDa molecular mass for BEN on T cells (Fig. 5, lane l), thus it appears to have approximately the same molecular weight as found previously on bursal epithelial and nervous cells (1). IL-2R immunoprecipitated from the same III-C5 lysates using mAb INN-CH 16 has an apparent molecular mass of 65 kDa (Fig. 5, lane 2). Thus, as expected from the previously reported molecular size of IL-2R (15), the BEN protein is distinct from the IL-2 receptor. We have shown that BEN protein purified from spinal cord, brain, and thymus carries the HNK- 1 epitope ( 1, 2). The mAb HNK- 1 which is directed at an epitope on glycosidic side chains immunoprecipitates 95-kDa molecules from III-C5 cells which comigrate with BEN-precipitated molecules on SDS-PAGE (Fig. 5, lane 3). Thus, BEN on proliferating T cells also appears to carry the HNK-1 epitope. Lack of BEN Expression by B Lymphocytes Although, as shown previously, the BEN antigen is expressedon epithelial cells of the bursa of Fabricius from E9 onward (l), BEN is not expressed by the bursal he123 kD
-68
FIG. 5. Immunoprecipitation of BEN, IL-2R, and HNK- 1 from the surfaceof III-C5 lymphoblasts. Lysates of surface-labeledIII-C5 cells were precipitated by mAb’s, anti-BEN (lane 1).INN-CH 16 (lane 2), and HNK1 (lane 3), and submitted to SDS-PAGE as described under Material and Methods.
106
CORBEL ET AL.
mopoietic cells (Fig. 6). Double labeling of El6 chick bursa with the BEN anti-mAb and an anti-chicken IgM polyclonal antibody shows that BEN expression is restricted to the epithelium lining the bursal lumen and the follicular epithelium. The bursal follicles contains many cells exhibiting IgM cytoplasmic and surfaceimmunoreactivity. Intra- and extrafollicular hemopoietic cells are BEN negative (Figs. 6a, 6b). Later in embryonic development and after hatching, BEN is expressedby the epithelium delineating the cortex from the medulla. The network of medullary epithelial cells which are BEN+ also stain with anti-cytokeratin antibodies. They are IgM negative. IgMpositive B lymphocytes develop within the medullary network as well as in the cortex and are BEN negative (Figs. 6c, 6d). Double immunofluorescence analysis of IgM- and BEN-positive cells in sections of other B lymphocyte-containing organs was also performed at different time points during the embryonic and adult life of chicks. In caecatonsils and Harderian glands, as in the bursa, B lymphocytes did not expressthe BEN antigen although the epithelial cells lining the lumen were positive. Moreover, B cells in these organs are Ig-secreting B lymphocytes, indicating that activated B cells do not express the BEN antigen. Actively proliferating B lymphoblasts from the RPL12 line also did not stain with anti-BEN mAb, confirming the lack of BEN expression by activated B lymphocytes. DISCUSSION In this paper, we describe the expression of avian protein BEN on the surface of peripheral T lymphocytes during cell activation. The BEN glycoprotein is expressed by thymocytes but not by resting T cells from blood and spleen. In the thymus, the double positive CT4+CT8+ cells are BEN positive since this population represents the majority of thymocytes ( 13). Double immunofluorescence staining indicates that the minor population of single CT4’ or CT8+ thymocytes do not expressBEN. This correlates with the lack of BEN expression by peripheral CT4 or CT8 positive T lymphocytes. However, upon mitogenic or allogeneic activation peripheral T cells express BEN at their surface. Activation and proliferation is accompanied by the expression of the avian homologue of the CD25 or IL-2R (15) and the class II antigen (B-L antigen) (19, 20). In this species,no other T-cell activation antigen has been described so far. We show that BEN is an activation antigen that is rapidly induced and expressed on both CT4 and CT8 T-cell subpopulations. Once T cells expressIL-2R they coexpress the BEN molecule. This is shown on mitogen-activated peripheral T lymphoblasts and on III-C5 cells, a cell line which has the phenotype of activated T lymphocytes expressing both IL-2R and class II antigen (9). BEN is not expressedon proliferating T cells of the MSBl line which expressesneither IL-2R nor class II antigen.
FIG. 6. Double labeling of bursa of Fabricius with anti-IgM and anti-BEN antibodies. (a, b) Section of E 16 bursa of Fabricius. X290. (a) Anti-IgM antibodies conjugated with TRITC reveal immunostaining of intrafollicular B lymphocytes. (b) Labeling of the samesectionby anti-BEN mAb was revealedby a fluorescein conjugate. BEN expression is restricted to the epithelial cells. No immunoreactivity was seen in the mesenchyme. (c, d) Section of 2-month bursa of Fabricius. X 180. (c) B lymphocytes are labeled by anti-IgM antibodies. (d) BEN antigen is expressedby cells of the basement membrane delineating the corticomedullary junction and the epithelial cells inside the follicle.
AN AVIAN
T LYMPHOCYTE
ACTIVATION
ANTIGEN
107
108
CORBEL
ET AL.
Among lymphocytes there is a specificity for T cells but BEN expression is not restricted to this cell lineage. As shown here and previously, the BEN molecule is expressedby epithelial cells of the bursa of Fabricius and also by subpopulations of nerve cells. The molecular weight of the BEN protein from activated T lymphocytes (loo-95 kDa) is similar to that described for bursal epithelial and neural cells (1). Recently, we have shown that other hemopoietic cells are BEN positive and we described differences in molecular forms of the BEN protein present at the surface of different cell types (2). Therefore, the biochemical characteristics of BEN glycoprotein on activated T cells such as III-C5 cells have been compared to the ones defined for tissues of different origin. On hemopoietic cells, such as embryonic spleen at 18 days of incubation and 3-week-old thymus, the purified protein exhibited an intermediate molecular weight between epithelial (110 kDa) and neural (95 kDa) forms (2). BEN differs from others known T cell activation antigens. Most, such as CD26, a peptidase (2 l), or CD27 (22), are present on resting cells and their expression is upregulated after activation. BEN, on the other hand, can be detected on T cells only in concert with their activation. Some integrins such as the murine vitronectin receptor and the “very late antigens” VLA-1 and VLA-2, in human, are difficult to detect on resting T cells but are well expressedafter T cell activation (17, 23-25). They differ from BEN in kinetics of their expression and in their molecular characteristics. BEN is an “early” activation marker appearing by 24 hr while the VLAs appear “late,” several days after mitogen stimulation. Furthermore, the integrins are noncovalently linked heterodimers (26) while BEN is a monomer. On activated T cells, the BEN molecule expressesthe HNK- 1 carbohydrate epitope. The occurrence of this epitope suggeststhat the glycoprotein BEN might be involved in cell-cell interactions or adhesion since several cell adhesion molecules expressthe HNK-1 carbohydrate epitope (27). Moreover, this epitope is also shared by cells of the nervous and immune systems(28). In human, the mAb HNK-1 reacts with a cell surface antigen of unknown properties on natural killer cells (14). In avian species too, subpopulations of lymphocytes express HNK-1 epitope during embryonic development (29). In addition to the integrins mentioned above, cell adhesion molecules of the immune system may belong to other large groups known as the immunoglobulin supergene family and the selectin family. Many of the molecules from the immunoglobulin supergene family are involved in cell adhesion. One of those, intercellular adhesion molecule 1 (ICAM- 1 or CD54) is a highly glycosylated cell surface protein and it is a ligand of LFA- 1 (30). ICAM- 1 is mainly expressedon macrophagesand myeloid cells as well as activated lymphocytes, but also by cells of nonhemopoietic origin including epithelial and endothelial cells and fibroblasts. Interestingly, ICAM- I glycoprotein, like BEN, displays molecular weight heterogeneity in different cell types (3 1). Selectin (32) or LEC-CAM (33) include lymphocyte homing receptor (MEL-14), endothelial leucocyte adhesion molecule 1 (ELAM- l), and granule membrane protein (GMP- 140). MEL- 14, which mediates lymphocyte adhesion to specialized secondary organ endothelial cells, has caught our attention becauseit is expressedby granulocytes, monocytes, and lymphocyte subsetsand becauseit is a glycoprotein of 90-100 kDa. However, activation of T cells results in downregulation of surface homing receptors (34) and it must be noted that BEN antigen is not found on endothelial cells. Thus based on tissue distribution and biochemical characteristics, there is no close relationship between any mammalian molecule classified among the clusters of dif-
AN AVIAN T LYMPHOCYTE
ACTIVATION
ANTIGEN
109
ferentiation and the BEN molecule. Isolation of its structural gene, which might allow us to determine BEN’s role in immune functions is in progress. ACKNOWLEDGMENTS The technical assistanceof Mrs. P. Belo and K. Heydon is acknowledged. We thank Dr. M. Cooper for providing CT4, CT8, and CT3 monoclonal antibodies, E. Bourson for the typescript, and S. Goumet and Y. Rantier for the illustrations. This work was supported by the Centre National de la Recherche Scientifique, the Institut National de la Sante et de la Recherche Medicale and by grants from the following foundations: the Association pour la Recherche MCdicaIe and the Ligue Nationale Fran&e contre le Cancer. H.G.B. was partially supported by a Fulbright Research Scholar Award and ResearchGrant AR30036 from the U.S. National Institutes of Health.
REFERENCES 1. Pourquit, O., Coltey, M., Thomas, J.-L., and Le Douarin, N. M., Development 109, 743, 1990. 2. Corbel, C., Cormier, F., Pourquit, O., Bluestein, H. G., and Le Douarin, N. M., 1991. Submitted for publication. 3. Clark, M. J., Gagnon, J., Williams, A. F., and Barclay, A. N., EMBU J. 4, 113, 1985. 4. Barclay, N. A., Letarte-Muirhead, M., Williams, A. F., and Faulkes, R. A., Nature 263, 563, 1976. 5. Williams, A. F., and Gagnon, J., Science 216,696, 1982. 6. Dalchau, R., Kirkley, J., and Fabre, J. W., Eur. J. Immunol. 10, 745, 1980. 7. Haynes, B. F., Harden, E. A., Telen, M. J., Hemler, M. E., Strominger, J. L., Palker, T. J., Scearce, R. M., and Eisenbarth, G. S., J. Irnmunol. 131, 1195, 1983. 8. Huet, S., Groux, H., Caillou, B., Valentin, H., Prieur, A.-M., and Bernard, A., J. Immunol. 143, 798, 1989. 9. Corbel, C., and Thomas, J.-L., Dev. Comp. Immunol. 14, 439, 1990. 10. Akiyama, Y., and Kato, S., &ken. J. 17, 105, 1974. 11. Okazaki, W., Witter, R. L., Romero, C.. Nazerian, K., Sharma, J. M., Fadly, A., and Ewert, D., Avian Pathol. 9, 31 I, 1980. 12. Chen, C. L. H., Lanier Ager, L., Gartland, L., and Cooper, M. D., J. Exp. Med. 164, 375, 1986. 13. Chan, M. M., Chen, C. L. H., Lanier, Ager, L., and Cooper, M. D., J. Immunol. 140,2133, 1988. 14. Abo, T., and Balch, C., J. Immunol. 127, 1024, 1981. 15. Hala, K., Schauenstein, K., Neu, N., Kramer, G., Wolf, H., Bock, G., and Wick, G., Eur. J. Immunol. 16, 1331, 1986.
16. Schauenstein,K., Kramer, G., Hala, K., Bock, G., and Wick, G., Dev. Comp. Immunol. 12,823, 1988. 17. Pischel, K. D., Bluestein, H. G., and Woods, V. L., Jr., J. Exp. Med. 164, 393, 1986. 18. Laemmli, U. K. Nature 227,680, 1970. 19. Guillemot, F. P., Oliver, P. D., Peault, B. M., and Le Douarin, N. M., J. Exp. Med. 160, 1803, 1984. 20. Ewert, D. L., Munchus, M. S., Chen, C. L. H., and Cooper, M. D., J. Immunol. 132,2524, 1984. 21. Stein, H., Schwarting, R., and Niedobitek, G., In “Leucocyte Typing IV” (W. Knapp, et al., Eds.), p. 412. Oxford Univ. Press,Oxford, 1989. 22. de Jong, R., Loenen, W. A. M., Brouwer, M., van Emmerik, L., de Vries, E. F. R., Borst, J., and van Lier, R. A. W., J. Immunol. 146, 2488, 1991. 23. Moulder, K., Roberts, K., Shevach, E. M., and CoIIigan, J. E., J. Exp. Med. 173, 343, 1991. 24. Hemler, M. E., Sanchez-Madrid, F., Flotte, T. J., Krensky, A. M., Burakoff, S. J., Bhan, A. K., Springer, T. A., and Strominger, J. L., J. Immunol. 132, 3011, 1984. 25. Hemler, M. E., Jacobson,J. G., Brenner, M. B., Mann, D., and Strominger, J. L., Eur. J. Immunol. 15, 502, 1985. 26. Hynes, R. O., Cell48, 549, 1987.
27. Schachner,M., Antonicek, H., Fahrig, T., Faissner,A., Fischer, G., Ktlnemund, V., Martini, R., Meyer, A., Persohn, E., Pollerberg, E., Prosbstmeier, R., Sadoul, R., Seilheimer, B., and Thor, G., In “Morphoregulatory Molecules” (G. M. Edelman, B. A. Cunningham, and J.-P. ThiCry, Eds.), p. 443. Wiley-Interscience, New York, 1990. 28. Lipinski, M., Braham, K., Caillaud, J. M., Carlu, C., and Tursz, T., J. Exp. Med. 158, 1775, 1983.
110
CORBEL ET AL.
29. Pea&, B., Chen, C. H., Cooper, M. D., Barbu, M., Lipinski, M., and Le Douarin, N. M., Proc. N&l. Acad. Sci. USA 84, 814, 1987. 30. Marlin, S. D., and Springer, T. A., Cell 51, 8 13, 1987. 31. Dustin, M. L., Rothlein, R., Bhan, A. K., Dinarello, C. A., and Springer, T. A., J. Immunol. 137,245, 1986. 32. Geng, J. G., Bevilacqua Moore, K. L., McIntyre, T. M., Prescott, S. M., Kim, J. M., Bliss, G. A., Zimmerman, G. A., and R. P. McEver, Nature (London). 343, 747, 1990. 33. Stoolman, L. M., Cell 56, 907, 1989. 34. Jung, T. M., Gallatin, W. M., Weissman, I. L., and Dailey, M. O., J. Zmmunol. 141, 4110, 1988.
