- Email: [email protected]
Effects of cyclosporine A on thymocyte differentiation in fetal thymus organ culture
CELLULAR
IMMUNOLOGY
123,307-3
15 (I 989)
Effects of Cyclosporine A on Thymocyte Differentiation in Fetal Thymus Organ Culture’ NOBUYUKI
MATSUHASHI,
YOSHIKO KAWASE, AND GEN SUZUKI
Division of Radiation Health, National Institute of Radiological Sciences, 4-9-1, Anagawa, Chiba-shi, 260, Japan Received April 6, 1989; acceptedJune 14, 1989 During the course of differentiation in the thymus, precursor T cells are negatively selected by a self-tolerance mechanism or positively selectedto acquire restriction specificity to self major histocompatibility complexes. We investigated the process of T cell differentiation and those selections using a fetal thymus organ culture with or without cyclosporine A. The agent blocked the maturation step from CD4+8+ double positive cells to mature CD4-8+ or CD4’8- single positive cells. On the other hand, the agent did not inhibit the development of CD3+4-8- T cell receptor (TCR)& cells, which were supposed to be T cells bearing yb-TCR chains. These results suggestthat the development of thymocytes bearing & or -&TCR chains differ in requirement for thymocyte-stromal cell interaction. 0 1989 Academic PXS, Inc.
INTRODUCTION Accumulating evidences have been reported that signals triggered by interaction between T cell receptor (TCR)2 on thymocyte and major histocompatibility complex (MHC) molecule on thymic stromal cells are essential in the differentiation process of thymocytes. Several investigators, including us, have utilized antibodies specific for MHC class I (l), class II (2-4), or CD4 (5) in order to demonstrate that those molecules are involved in T cell development. Since cyclosporine A (GA) is known to block the signal transduction cascade(6, 7) and to induce thymus atrophy when administered in vivo (8,9), we used CsA in fetal thymus organ culture (FTOC) in an attempt to block the T cell differentiation. In the previous study, we demonstrated that the agent abrogated the generation of CD4+8-, but not that of CD4-8+ thymocytes in the fetal thymus cultured in vitro (4). Since both CD4 and CD8 single positive (SP) cells are known to be positively selected in the thymus (lo-14), we wished to ’ This work was supported by grant-in-aid for General Scientific Research from the Ministry of Education, Science, and Culture, Japan. 2 Abbreviations used: FI’OC, fetal thymus organ culture; CsA, cyclosporine A, MHC, major histocompatibility complex; TCR, T cell receptor; mAb, monoclonal antibody; PITC, fluorescein isothiocyanate; PE, phycoerythrin; CTL, cytotoxic T lymphocyte; DN, double negative; DP, double positive; SP, single positive; GD, gestational day; GDI 5+n, GD 15 fetal thymus lobes organ-cultured for n days. 307 0008-8749/89$3.00 Copyright 0 1989 by Academic Press, Inc. All rights of reproduction in any form resewed.
308
MATSUHASHI,
KAWASE,
AND
SUZUKI
elucidate the reason why CD4-8+ cells remained in the thymus in the presence of CsA. MATERIALS AND METHODS BIO.BR, BALB/c, and C57BL/10 (BlO) mice were bred in our colony at the National Institute of Radiological Science,Chiba, Japan. Each strain is known to express
F23.1 determinants of TCR.
a-MEM medium was supplemented with penicillin-streptomycin, 2 mA4 L-glutamine, 15 mM Hepes, 5 X 1Om5M 2-mercaptoethanol, 1%anti-PPLO agent (GIBCO, Chargin Falls, OH), and 10%heat-inactivated fetal calf serum.
CsA (Sandoz Co. Ltd, Switzerland) was dissolved in methanol at 1 X lop3 M, and at the time of use, it was further diluted with medium to 1 X lop7 M, a concentration sufficient to inhibit allospecific cytotoxic T lymphocyte (CTL) induction or alloreactive mixed lymphocyte reaction (6), As a control, methanol was diluted at the same concentration and added into FTOC. Monoclonal antibodies (mAb) RL172.4 (rat mAb to CD4, Ref. (15)): GUS (ral mAb to CD4, Ref. (16)), F23.1 (mouse mAb to V/38, Ref. (17)}, 145-X1 1 (hamster mAb to CD3,, Ref. (18)), J 1Id (rat mAb to Jl Id, Ref. (19)), and 597 (hamster mAb to all TCR-a/3 chains, Ref. (20)) were used.
Organ culture was performed according to the methods of Jenkinson et al. (21), wilh slight modification. The CsA-containing medium was changed every 4 days, instead of 5 days as in the previous study (4), to get maximal effects.
In order to obtain CD4-8’ and CD4-8- cells, thymocytes were cytotoxically treated with anti-CD4 monoclonal antibodies (GK 1.5 and RL172.4) and rabbit complement (Cedarlane, Hornby, Canada). Cd Surface Marker Analysis For CDS/CD4 analysis, cells were stained with fluorescein isothiocyanate (FITC)conjugated anti-CD8 and phycoerythrin (PE)-conjugated anti-CD4 antibodies (Becton-Dickinson, Mountain View, CA). For CD8/TCR-ar@chains or CD8/CD3 analysis, cells were sequentially incubated with F23.1,597, or 145-2C11, biotin-conjugated goat anti-mouse IgG (Cappel, Westchester, PA), normal mouse serum to saturate goat anti-mouse IgG reactivity, and a mixture of PE-conjugated streptavidin (BectonDickinson) and FITC-conjugated anti-CD& We used biotin-conjugated goat antimouse IgG as the second reagent for staining with 145-X1 1 or 597 because antimouse IgG cross-reacted well with hamster IgG. For CDS/J 11d analysis, J 1Id, goat
EFFECTS
OF CsA ON THYMOCYTE
DIFFERENTIATION
IN Fl-OC
309
anti-rat IgG biotin (E. Y. Lab., San Mateo, CA), and rat serum were used. Stained cells were analyzed by FACStar (Becton-Dickinson). Positive-negative demarcations were made according to an autofluorescence profile of the same cells as a negative control.
AllospeciJc CTL Assay Allospecific CTL were induced as reported (6). Recovered live cells were mixed with “Cr-labeled splenic Con A blasts in round-bottomed 96-well plates for 4 hr. The 5’Cr activity released from the cells was measured with a gamma counter, and percentage specific ‘ICr release was calculated as follows: percentage specific 51Cr release = 100 X (E - S)/(M - S), where E, S, and M represent experimental, spontaneous, and maximal “Cr release, respectively. Maximal “Cr releasewas induced by adding 0.2% sodium deoxycholate. RESULTS
ThymocyteD&erentiation Proceededin a Synchronized Way in FTOC Almost all of the thymocytes from fetuses of the 15th gestational day (GD 15) consisted of CD4-8- double negative (DN) cells (Fig. IA). Within 1 day of culture (GDl5+ I), CD4-8+ cells and, shortly after, CD4+8+ double positive (DP) cells developed (Fig. I B). By GDl5+2, almost all cells were CD4+8+ (Figs. 1C and 1D). On GDl5+5, CD8 and CD4 SP cells started to develop simultaneously (Fig. 1E). At the same time, DN cells reappeared as a second wave (Table 1). Inasmuch as thymocyte differentiated in a synchronized way and since maturated SP cells did not emigrate from the thymic lobes in FTOC, both CD4 and CD8 SP cells accumulated and DP cells decreasedin proportion at later periods (Figs. 1F and 1G). As described above, the first wave of thymocyte differentiation generated mature SP cells on GD15+5, but they were accompanied by a second wave of DN cells. Thereafter, we could detect all four phenotypes of thymocytes in FTOC as far as the culture was carried out. The increase in the proportion of DN cells in GD15+5 thymus lobes (Fig. IE) was not due to loss of cells of other phenotypes because Table 1 shows that the absolute number of DN cells in GDl5+5 lobes (3.4 X 104)was more than that in GD15+3 lobes (1 .O X 104).
In the Presenceof CsA, CD4-8+ Cells Developed,but CD4+8- Cells Did Not Addition of CsA into the culture did not affect the differentiation steps up to the accumulation of DP cells (Fig. lH, Table 1). CsA failed to inhibit the development of DP cells, even in GD14 thymuses (data not shown). However, CsA inhibited the generation of CD4+8- cells (Figs. lI- 1K). On the contrary, CD4-8+ cells, although fewer in number than in the CsA-free culture, appeared in the presence of CsA. In CsA-free thymic lobes, SP cells accumulated during the culture period. In contrast, the CsA-added culture provided some 15% of CD4-8+ SP cells, which did not increase,at least up to GD 16+ 13. Regarding absolute cell numbers, the number of DN and DP cells did not show marked difference between CsA-free and CsA-added FTOC, while CsA-added FTOC generated significantly decreased numbers of CD4 or CD8 SP cells.
310
MATSUHASHI,
KAWASE, AND SUZUKI
C,D8 FIG. 1. CsA inhibited the development of CD4+8- cells, but not that of CD4-8+ cells. Thymocytes recovered from thymus lobes of GD15 fetuses (A) or lobes organ cultured for 1 day (B), 2 days (C), 4 days (D, H), and 5 days (E, I) were stained with PE-anti-CD4 and FITC-anti-CD8 antibodies. Similarly, thymocytes from GD16 fetal thymuses cultured for 9 (F, J) or 13 days (G, K) were stained with those antibodies. CsA was added (H-K) at the final concentration of 1 X IO-’ Mor was not added (A-G).
DP cells in the FTOC consisted of two population, one was large in cell size and the other was small. Approximately 17% of DP cells were larger in size. The cell size was measured by forward scatter on FACS analysis. We thought the difference in cell size may reflect a different maturation stage or activation state among DP cells, but the addition of CsA did not change the proportion of small DP cells vs large DP cells (data not shown). CD4-8’ Cells Developing in the Presenceof CsA WerePhenotypically Immature In order to further investigate the nature of the CD4-8+ cells in CsA-added culture, we examined whether they bore Jl Id, CD3, or TCR-(YP chains. Thymocytes were depleted of CD4+8- and CD4+8+ cells by cytotoxic treatment, and were stained with monoclonal antibodies specific for those determinants (Figs. 2 and 3). In CsA-free
EFFECTS OF CsA ON THYMOCYTE
DIFFERENTIATION
IN FI’OC
311
TABLE 1 The Course of Cell Number of Each Phenotype of Thymocytes in FTOC
CsA free
4-84-8+ 4+84+8+ Total
CsA added
4-84-8+ 4+84+8+ Total
GD15+0
15+3
15+5
15+8
15+12--15
78 1 2 I 81 k 10
10 4 12 193 220+ 18
34 28 28 127 215 f 15
12 28 22 85 147f 15
12 38 25 9 83 + 32
20 2 4 182 210 f 30
37 17 5 86 145f 5
20 11 3 37 72+ 4
21 8 1 30 60+21
Note. The numbers of thymocytes of each phenotype in various conditions are listed. Each value represents the mean of three or four experiments and is described as (cell number) X 10-3/lobe. The total numbers are described as mean + SE.
culture, most of the CD4-8+ cells were CD3+ and approximately 15%of the CD4-8+ cells expressed the F23.1 determinant at high density. On the contrary, most of the CD4-8+ cells generated in the presence of CsA were TCR/CD3- and J 1Id+, which were the same phenotype as that of the early CD4-8+ cells (Fig. 2C, Ref. (20)). Thus they were supposed to be immature CD4-8+ cells which had developed from DN cells in the second wave. These results suggestthat CsA inhibited the generation of not only CD4+8-, but also mature CD4-8+ cells. CD4-8f Cells Generated in the Presence of CsA Did Not Contain Functionally Mature Precursor CTL In order to examine whether the CD4-8+ cells generated in the presence of CsA were functionally mature, we tried to induce allospecific CTL from them. CD4-8+Jl Id+ cells in adult mouse thymus are known to be defective in the ability to generate allospecific CTL (22). Cells from CsA-free thymic lobes exhibited allospecific cytotoxicity, but cells from CsA-added lobes did not (data not shown). This was not due to the carrying over of CsA because CsA had been excluded from the FTOC 24 hr before preparation of responder cells. CD3’4-8- Cells Generated in the Presenceof CsA in FTOC In the presence of CsA, about two-thirds of the DN cells were CD3+, but no cells were F23. I+, and most of the DN cells were Jl Id- (Fig. 2). Staining with mAb 597, reactive with all &I’ mouse TCR (20), confirmed that these CD3+4-8-Jl Id- cells did not bear TCR-aP chains (Fig. 3). These DN cells were supposed to be thymocytes bearing TCR-y6 chains. The same cell population was thought to be present in CsAfree culture, but the addition of CsA into the FTOC enhanced the proportion of the cell population and made it easyto detect it. Thus, CsA failed to inhibit the generation of CD3+4-8-J 11d- cells, possibly possessingTCR-yG chains.
312
MATSUHASHI,
KAWASE, AND SUZUKI
CD8 FIG. 2. CsA inhibited the development of TCR-&CD3+J I Id- mature SP cells, but not that of TCR@CD3+ DN cells. FTOC was set up in the presence (D-F) or absence (A-C) of CsA. Thymocytes from 9day (A, D) or 13day (B, C, E, F)-cultured GD 16 thymus lobes were cytotoxically treated with anti-CD4 antibodies and complement to delete CD4’8- and CD4+8+ cells. Resultant cells were dual-stained with FITC-CD8 and F23.1,2Cll, or J 1Id antibodies, as indicated.
DISCUSSION There have been accumulating results that interaction between TCR and MHC molecules is involved as a key element, not only in the negative selection, but also in the positive selection of thymocytes. It has been indicated that the negative selection takes place on DP cells or on TCR low positive cells (5, 23,24). On the other hand, it is still controversial about the developmental stageof thymocytes where the positive selection takes place ( 12- 14). It has been reported that anti-MHC class I antibodies block the development of CD4-8+ cells, and that anti-class II or anti-CD4 antibodies block that of CD4+8- cells ( l-5). It is noteworthy in both casesthat the development of DP cells is not inhibited (l-5). It has already been reported that in viva administration of CsA induces atrophy of thymic medulla (8,9), where mature SP cells are the predominant thymocyte pop ulation. We attempted to further scrutinize the cell kinetics by virtue of FTOC in the presence or absenceof CsA. CsA is known to exert many kinds of suppressive effects on the immune system, including blockade of the signal transduction cascadein T cells triggered by perturbation of the TCR/CD3 complex (6,25,26). The exact mechanism of this blockade is not yet known, but previous reports suggestthat CsA does not affect the signal cascade from perturbation of the TCR/CD3 complex to calcium influx or turnover of phos-
EFFECTS OF CsA ON THYMOCYTE
CD8
DIFFERENTIATION
IN FTOC
313
CD8
FIG. 3. The CD4-8- cells developing in the presenceofCsA did not near &TCR chains. Positive control of 597 is shown using adult thymocytes (A). Dotted line and solid line indicate autofluorescence and 597staining profiles, respectively. Thymocytes from CsA-added FTOC (GD I5t 15) were treated with antiCD4 and complement. Resultant cells were stained with anti-CD8 and 597 (3C). (B) A control using antiCD8 and the second antibody alone.
phatidyl inositol(27-29). Therefore, the agent may block the signal cascadeat a later step, possibly by binding to calmodulin (7). On the other hand, the agent is also known to disturb thymic stromal cell function, including Ia expression (8). Anyway, the agent is expected to have suppressive effects on thymocyte development, either by interfering the signal cascadeor impairing the stromal cell function. FIOC has certain advantages in investigating which stage of T cell differentiation CsA would inhibit. When CsA is administrated in vivo, one cannot rule out the possibility that some cells escapethe effects of the agent. Indeed, it is reported that cells escaping the effects of CsA exhibit autoreactivity in vivo (8, 30). In the organ culture, however, we can obtain sufficiently high concentrations of CsA throughout the culture period. Besides, in FTOC, no stem cells immigrate into the thymus during the culture period, and cells differentiate in a synchronized manner. Moreover, mature SPcells do not emigrate from the thymic lobes. All of these made it possible to analyze the fine differentiation kinetics of thymocytes. In the previous study we demonstrated that CsA inhibits the development of CD4, but not CD8 SP cells in vitro (4). The present study demonstrated that the CD8 SP cells developing in the presence of CsA showed immature phenotype of Jl ld+TCR& (Fig. 2). They were also shown to be functionally immature and were also less in number than CsA-free FTOC. Thus, they were early CD4-8+ cells which developed from DN cells and would develop into DP cells (Fig. 1). Hence, CsA inhibits the differentiation step from DP cells not only to CD4, but also to CD8 mature SP cells.
314
MATSUHASHI,
KAWASE, AND SUZUKI
Then what is the mechanism of this inhibitory effect? Two explanations are possible. One is that CsA inhibits the positive selection of thymocytes by blocking the signal transduction cascade initiated by perturbation of the TCR/CD3 complex (6, 7). The other is that CsA impairs thymic stromal cell functions and it leads to failure of mature SP cell generation, which may or may not be related to the process of positive selection. In the previous study, we could not demonstrate a decreasein the expression of the Ia molecule on thymic stromal cells in CsA-added thymic lobes in our culture condition (4). Although it is not a formal proof, these results suggestthat the positive selection takes place at the transition step from DP cells to SP cells, and this hypothesis is consistent with the results of a recent report in which DP cells, but not SP cells, could develop in the absence of appropriate MHC in TCR gene-transgenie mice (14). The inhibitory effect of CsA on thymocyte differentiation may be due to direct blocking of signal cascade of thymocytes, although other possibilities such that CsA may inhibit the secretion of certain factors from stromal cells cannot be ruled out. Interestingly, CsA inhibited the differentiation of mature SP cells, but not of CD3+4-8-TCRc$-Jl Id- cells, which were supposed to be cells bearing TCR-y6 chains. This result agreeswith a recent report where CsA has been administered in vivo (3 1,32). In chicken it has been reported that thymocytes bearing TCR y&homolog at high density are present in the subcapsular zone of thymic cortex, where cells bearing a/3-homolog are rare (33). Thus the requirement of interaction between thymocytes and thymic stromal cells may be different in the development of c@ and y6TCR-bearing cells. Further investigations will be necessary to elucidate the positive selection of TCR-c$ and TCR-y&bearing cells and its requirement for stromal elements. Long-lived progenitor cells have been reported to be present in the thymus after intrathymic injection of bone marrow, spleen, or fetal liver cells of a Thy- 1 disparate donor (34). An immunohistochemical study also revealed a focal donor cell colony at the site of intrathymic injection, even after 10 months (K. Hirokawa, Tokyo Metropolitan Institute of Gerontology, personal communication). In our study using FTOC, the first wave of thymocyte differentiation was followed by another wave, and all four CD4/8 populations could be detected throughout the culture period after GD15+5. This also suggested the presence of long-lived progenitor cells in the thymus. ACKNOWLEDGMENTS We are grateful to Drs. Tomio Tada and Ralph Kubo for providing mAb 597, to Dr. Katsuiku Hirokawa for letting us quote his unpublished data, and to Mrs. Sumiko Shinohara for her excellent technical assistance.
REFERENCES 1. Marusic-Galesic, S., Stephany, D. A., Longo, D. L., and Kruisbeek, A. M., Nature (London) 333, 180, 1988. 2. Kruisbeek, A. M., Mond, J. J., Fowlkes, B. J., Carmen, J. A., Bridges, S., and Longo, D. L., .I. Exp. Med. 161,1029, 1985. 3. DeLuca, D., J. Immunol. 136,430, 1986. 4. Takeuchi, Y., Habu, S., Okumura, K., and Suzuki, G., Immunology66,362, 1989. 5. Fowlkes, B. J., Schwartz, R. H., and Pardoll, D. M., Nature (London) 334,620,1988.
EFFECTS OF CsA ON THYMOCYTE
DIFFERENTIATION
IN FTOC
315
6. Sawada, S., Suzuki, G., Kawase, Y., and Takaku, F., J. Immunol. 139,1797,1987. 7. Colombani, P. M., Robb, A., and Hess, A. D., Science 228,337, 1985. 8. Cheney, R. T., and Sprent, J., Transplant. Proc. 17,528, 1985. 9. Beschomer, W. E., Di Gennaro, K. A., Hess, A. D., and Santos, G. W., Cell. Immunol. 110,350, 1987. 10. Bevan, M. J., Nature (London) 269,417,1977. 11. Zinkemagel, R. M., Callahan, G. N., Althage, A., Cooper, S., Klein, P. A., and Klein, J., J. Exp. Med. 147,882, 1978. 12. Teh, H. S., Kisielow, P., Scott, B., Kishi, H., Uematsu, Y., Bluthmann, H., and von Boehmer, H., Nature (London) 335,229, 1988. 13. Kisielow, P., Teh, H. S., Bluthmann, H., and von Boehmer, H., Nature (London) 335,730, 1988. 14. Sha, W. C., Nelson, C. A., Newberry, R. D., Kranz, D. M., Russel, J. H., and Loh, D. Y., Nature (London) 336,73, 1988.
15. Ceredig, R., Lowenthal, J. W., Nabholz, M., and MacDonald, H. R., Nature (London) 314,98, 1985. 16. Dialynas, D. P., Wilde, D. B., Marrack, P., Pierres, A., Wall, K. A., Harvan, W., Otten, G., Loken, M. R., Pierres, M., Kappler, J., and Fitch, F. W., Immunol. Rev. 74,29, 1983. 17. Staerz, U. D., Rammensee, H. G., Benedetto, J. D., and Bevan, M. J., J. Zmmunol. 134,3994, 1985. 18. Leo, O., Foo, M., Sachs, D. H., Samelson, L. E., and Bluestone, J. A., Proc. Nutl. Acud. Sci. USA 84, 1374, 1987. 19. Bruce, J., Symington, F. W., McKearn, T. J., and Sprent, J., J. Immunof. 127,2496, 198 1. 20. White, J., Herman, A., Pullen, A. M., Kubo, R., Kappler, J. W., and Mar-rack, P., Cell56,27, 1989. 21. Jenkinson, E. J., Franchi, L. L., Kingston, R., and Owen, J. J. T., Eur. J. Immunol. 12,583, 1982. 22. Crispe, N., and Bevan, M. J., J. Immunol. 138,2013, 1987. 23. Kisielow, P., Bluthmann, H., Staetz, U. D., Steinmetz, M., and von Boehmer, H., Nature (London) 333,742, 1988. 24. Kappler, J. W., Roehm, N., and Mat-rack, P., Ce/l49,273, 1987. 25. Shevach, E. M., Annu. Rev. Immunol. 3,397, 1985. 26. B&ton, S., and Palacios R., Immunol. Rev. 65,5, 1982. 27. Kay, J. E., Benzie, C. R., and Borghetti, A. F., Immunology 50,441, 1983. 28. Metcalfe, S., Transplantation 39, 161, 1984. 29. Bijsterbosch, M. K., and Klaus, G. G. B., Immunology56,435, 1985. 30. Sorokin, R., Kimura, H., Schroder, K., Wilson, D. H., and Wilson, D. B., J. Exp. Med. 164, 1615, 1986. 3 I. Jenkins, M. K., Schwartz, R. H., and Pardoll, D. M., Science 241, 1655, 1988. 32. Gao, E., Lo, D., Cheney, R., Kanagawa, O., and Sprent J., Nature (London) 336, 176, 1988. 33. Bucy, R. P., Chen, C. H., Losch, U., andcooper, M. D., J. Zmmunol. 141,2200, 1988. 34. Katsura, Y., Kina, T., Takaoki, Y., and Nishikawa, S., Eur. J. Immunol. 18,889, 1988.
IMMUNOLOGY
123,307-3
15 (I 989)
Effects of Cyclosporine A on Thymocyte Differentiation in Fetal Thymus Organ Culture’ NOBUYUKI
MATSUHASHI,
YOSHIKO KAWASE, AND GEN SUZUKI
Division of Radiation Health, National Institute of Radiological Sciences, 4-9-1, Anagawa, Chiba-shi, 260, Japan Received April 6, 1989; acceptedJune 14, 1989 During the course of differentiation in the thymus, precursor T cells are negatively selected by a self-tolerance mechanism or positively selectedto acquire restriction specificity to self major histocompatibility complexes. We investigated the process of T cell differentiation and those selections using a fetal thymus organ culture with or without cyclosporine A. The agent blocked the maturation step from CD4+8+ double positive cells to mature CD4-8+ or CD4’8- single positive cells. On the other hand, the agent did not inhibit the development of CD3+4-8- T cell receptor (TCR)& cells, which were supposed to be T cells bearing yb-TCR chains. These results suggestthat the development of thymocytes bearing & or -&TCR chains differ in requirement for thymocyte-stromal cell interaction. 0 1989 Academic PXS, Inc.
INTRODUCTION Accumulating evidences have been reported that signals triggered by interaction between T cell receptor (TCR)2 on thymocyte and major histocompatibility complex (MHC) molecule on thymic stromal cells are essential in the differentiation process of thymocytes. Several investigators, including us, have utilized antibodies specific for MHC class I (l), class II (2-4), or CD4 (5) in order to demonstrate that those molecules are involved in T cell development. Since cyclosporine A (GA) is known to block the signal transduction cascade(6, 7) and to induce thymus atrophy when administered in vivo (8,9), we used CsA in fetal thymus organ culture (FTOC) in an attempt to block the T cell differentiation. In the previous study, we demonstrated that the agent abrogated the generation of CD4+8-, but not that of CD4-8+ thymocytes in the fetal thymus cultured in vitro (4). Since both CD4 and CD8 single positive (SP) cells are known to be positively selected in the thymus (lo-14), we wished to ’ This work was supported by grant-in-aid for General Scientific Research from the Ministry of Education, Science, and Culture, Japan. 2 Abbreviations used: FI’OC, fetal thymus organ culture; CsA, cyclosporine A, MHC, major histocompatibility complex; TCR, T cell receptor; mAb, monoclonal antibody; PITC, fluorescein isothiocyanate; PE, phycoerythrin; CTL, cytotoxic T lymphocyte; DN, double negative; DP, double positive; SP, single positive; GD, gestational day; GDI 5+n, GD 15 fetal thymus lobes organ-cultured for n days. 307 0008-8749/89$3.00 Copyright 0 1989 by Academic Press, Inc. All rights of reproduction in any form resewed.
308
MATSUHASHI,
KAWASE,
AND
SUZUKI
elucidate the reason why CD4-8+ cells remained in the thymus in the presence of CsA. MATERIALS AND METHODS BIO.BR, BALB/c, and C57BL/10 (BlO) mice were bred in our colony at the National Institute of Radiological Science,Chiba, Japan. Each strain is known to express
F23.1 determinants of TCR.
a-MEM medium was supplemented with penicillin-streptomycin, 2 mA4 L-glutamine, 15 mM Hepes, 5 X 1Om5M 2-mercaptoethanol, 1%anti-PPLO agent (GIBCO, Chargin Falls, OH), and 10%heat-inactivated fetal calf serum.
CsA (Sandoz Co. Ltd, Switzerland) was dissolved in methanol at 1 X lop3 M, and at the time of use, it was further diluted with medium to 1 X lop7 M, a concentration sufficient to inhibit allospecific cytotoxic T lymphocyte (CTL) induction or alloreactive mixed lymphocyte reaction (6), As a control, methanol was diluted at the same concentration and added into FTOC. Monoclonal antibodies (mAb) RL172.4 (rat mAb to CD4, Ref. (15)): GUS (ral mAb to CD4, Ref. (16)), F23.1 (mouse mAb to V/38, Ref. (17)}, 145-X1 1 (hamster mAb to CD3,, Ref. (18)), J 1Id (rat mAb to Jl Id, Ref. (19)), and 597 (hamster mAb to all TCR-a/3 chains, Ref. (20)) were used.
Organ culture was performed according to the methods of Jenkinson et al. (21), wilh slight modification. The CsA-containing medium was changed every 4 days, instead of 5 days as in the previous study (4), to get maximal effects.
In order to obtain CD4-8’ and CD4-8- cells, thymocytes were cytotoxically treated with anti-CD4 monoclonal antibodies (GK 1.5 and RL172.4) and rabbit complement (Cedarlane, Hornby, Canada). Cd Surface Marker Analysis For CDS/CD4 analysis, cells were stained with fluorescein isothiocyanate (FITC)conjugated anti-CD8 and phycoerythrin (PE)-conjugated anti-CD4 antibodies (Becton-Dickinson, Mountain View, CA). For CD8/TCR-ar@chains or CD8/CD3 analysis, cells were sequentially incubated with F23.1,597, or 145-2C11, biotin-conjugated goat anti-mouse IgG (Cappel, Westchester, PA), normal mouse serum to saturate goat anti-mouse IgG reactivity, and a mixture of PE-conjugated streptavidin (BectonDickinson) and FITC-conjugated anti-CD& We used biotin-conjugated goat antimouse IgG as the second reagent for staining with 145-X1 1 or 597 because antimouse IgG cross-reacted well with hamster IgG. For CDS/J 11d analysis, J 1Id, goat
EFFECTS
OF CsA ON THYMOCYTE
DIFFERENTIATION
IN Fl-OC
309
anti-rat IgG biotin (E. Y. Lab., San Mateo, CA), and rat serum were used. Stained cells were analyzed by FACStar (Becton-Dickinson). Positive-negative demarcations were made according to an autofluorescence profile of the same cells as a negative control.
AllospeciJc CTL Assay Allospecific CTL were induced as reported (6). Recovered live cells were mixed with “Cr-labeled splenic Con A blasts in round-bottomed 96-well plates for 4 hr. The 5’Cr activity released from the cells was measured with a gamma counter, and percentage specific ‘ICr release was calculated as follows: percentage specific 51Cr release = 100 X (E - S)/(M - S), where E, S, and M represent experimental, spontaneous, and maximal “Cr release, respectively. Maximal “Cr releasewas induced by adding 0.2% sodium deoxycholate. RESULTS
ThymocyteD&erentiation Proceededin a Synchronized Way in FTOC Almost all of the thymocytes from fetuses of the 15th gestational day (GD 15) consisted of CD4-8- double negative (DN) cells (Fig. IA). Within 1 day of culture (GDl5+ I), CD4-8+ cells and, shortly after, CD4+8+ double positive (DP) cells developed (Fig. I B). By GDl5+2, almost all cells were CD4+8+ (Figs. 1C and 1D). On GDl5+5, CD8 and CD4 SP cells started to develop simultaneously (Fig. 1E). At the same time, DN cells reappeared as a second wave (Table 1). Inasmuch as thymocyte differentiated in a synchronized way and since maturated SP cells did not emigrate from the thymic lobes in FTOC, both CD4 and CD8 SP cells accumulated and DP cells decreasedin proportion at later periods (Figs. 1F and 1G). As described above, the first wave of thymocyte differentiation generated mature SP cells on GD15+5, but they were accompanied by a second wave of DN cells. Thereafter, we could detect all four phenotypes of thymocytes in FTOC as far as the culture was carried out. The increase in the proportion of DN cells in GD15+5 thymus lobes (Fig. IE) was not due to loss of cells of other phenotypes because Table 1 shows that the absolute number of DN cells in GDl5+5 lobes (3.4 X 104)was more than that in GD15+3 lobes (1 .O X 104).
In the Presenceof CsA, CD4-8+ Cells Developed,but CD4+8- Cells Did Not Addition of CsA into the culture did not affect the differentiation steps up to the accumulation of DP cells (Fig. lH, Table 1). CsA failed to inhibit the development of DP cells, even in GD14 thymuses (data not shown). However, CsA inhibited the generation of CD4+8- cells (Figs. lI- 1K). On the contrary, CD4-8+ cells, although fewer in number than in the CsA-free culture, appeared in the presence of CsA. In CsA-free thymic lobes, SP cells accumulated during the culture period. In contrast, the CsA-added culture provided some 15% of CD4-8+ SP cells, which did not increase,at least up to GD 16+ 13. Regarding absolute cell numbers, the number of DN and DP cells did not show marked difference between CsA-free and CsA-added FTOC, while CsA-added FTOC generated significantly decreased numbers of CD4 or CD8 SP cells.
310
MATSUHASHI,
KAWASE, AND SUZUKI
C,D8 FIG. 1. CsA inhibited the development of CD4+8- cells, but not that of CD4-8+ cells. Thymocytes recovered from thymus lobes of GD15 fetuses (A) or lobes organ cultured for 1 day (B), 2 days (C), 4 days (D, H), and 5 days (E, I) were stained with PE-anti-CD4 and FITC-anti-CD8 antibodies. Similarly, thymocytes from GD16 fetal thymuses cultured for 9 (F, J) or 13 days (G, K) were stained with those antibodies. CsA was added (H-K) at the final concentration of 1 X IO-’ Mor was not added (A-G).
DP cells in the FTOC consisted of two population, one was large in cell size and the other was small. Approximately 17% of DP cells were larger in size. The cell size was measured by forward scatter on FACS analysis. We thought the difference in cell size may reflect a different maturation stage or activation state among DP cells, but the addition of CsA did not change the proportion of small DP cells vs large DP cells (data not shown). CD4-8’ Cells Developing in the Presenceof CsA WerePhenotypically Immature In order to further investigate the nature of the CD4-8+ cells in CsA-added culture, we examined whether they bore Jl Id, CD3, or TCR-(YP chains. Thymocytes were depleted of CD4+8- and CD4+8+ cells by cytotoxic treatment, and were stained with monoclonal antibodies specific for those determinants (Figs. 2 and 3). In CsA-free
EFFECTS OF CsA ON THYMOCYTE
DIFFERENTIATION
IN FI’OC
311
TABLE 1 The Course of Cell Number of Each Phenotype of Thymocytes in FTOC
CsA free
4-84-8+ 4+84+8+ Total
CsA added
4-84-8+ 4+84+8+ Total
GD15+0
15+3
15+5
15+8
15+12--15
78 1 2 I 81 k 10
10 4 12 193 220+ 18
34 28 28 127 215 f 15
12 28 22 85 147f 15
12 38 25 9 83 + 32
20 2 4 182 210 f 30
37 17 5 86 145f 5
20 11 3 37 72+ 4
21 8 1 30 60+21
Note. The numbers of thymocytes of each phenotype in various conditions are listed. Each value represents the mean of three or four experiments and is described as (cell number) X 10-3/lobe. The total numbers are described as mean + SE.
culture, most of the CD4-8+ cells were CD3+ and approximately 15%of the CD4-8+ cells expressed the F23.1 determinant at high density. On the contrary, most of the CD4-8+ cells generated in the presence of CsA were TCR/CD3- and J 1Id+, which were the same phenotype as that of the early CD4-8+ cells (Fig. 2C, Ref. (20)). Thus they were supposed to be immature CD4-8+ cells which had developed from DN cells in the second wave. These results suggestthat CsA inhibited the generation of not only CD4+8-, but also mature CD4-8+ cells. CD4-8f Cells Generated in the Presence of CsA Did Not Contain Functionally Mature Precursor CTL In order to examine whether the CD4-8+ cells generated in the presence of CsA were functionally mature, we tried to induce allospecific CTL from them. CD4-8+Jl Id+ cells in adult mouse thymus are known to be defective in the ability to generate allospecific CTL (22). Cells from CsA-free thymic lobes exhibited allospecific cytotoxicity, but cells from CsA-added lobes did not (data not shown). This was not due to the carrying over of CsA because CsA had been excluded from the FTOC 24 hr before preparation of responder cells. CD3’4-8- Cells Generated in the Presenceof CsA in FTOC In the presence of CsA, about two-thirds of the DN cells were CD3+, but no cells were F23. I+, and most of the DN cells were Jl Id- (Fig. 2). Staining with mAb 597, reactive with all &I’ mouse TCR (20), confirmed that these CD3+4-8-Jl Id- cells did not bear TCR-aP chains (Fig. 3). These DN cells were supposed to be thymocytes bearing TCR-y6 chains. The same cell population was thought to be present in CsAfree culture, but the addition of CsA into the FTOC enhanced the proportion of the cell population and made it easyto detect it. Thus, CsA failed to inhibit the generation of CD3+4-8-J 11d- cells, possibly possessingTCR-yG chains.
312
MATSUHASHI,
KAWASE, AND SUZUKI
CD8 FIG. 2. CsA inhibited the development of TCR-&CD3+J I Id- mature SP cells, but not that of TCR@CD3+ DN cells. FTOC was set up in the presence (D-F) or absence (A-C) of CsA. Thymocytes from 9day (A, D) or 13day (B, C, E, F)-cultured GD 16 thymus lobes were cytotoxically treated with anti-CD4 antibodies and complement to delete CD4’8- and CD4+8+ cells. Resultant cells were dual-stained with FITC-CD8 and F23.1,2Cll, or J 1Id antibodies, as indicated.
DISCUSSION There have been accumulating results that interaction between TCR and MHC molecules is involved as a key element, not only in the negative selection, but also in the positive selection of thymocytes. It has been indicated that the negative selection takes place on DP cells or on TCR low positive cells (5, 23,24). On the other hand, it is still controversial about the developmental stageof thymocytes where the positive selection takes place ( 12- 14). It has been reported that anti-MHC class I antibodies block the development of CD4-8+ cells, and that anti-class II or anti-CD4 antibodies block that of CD4+8- cells ( l-5). It is noteworthy in both casesthat the development of DP cells is not inhibited (l-5). It has already been reported that in viva administration of CsA induces atrophy of thymic medulla (8,9), where mature SP cells are the predominant thymocyte pop ulation. We attempted to further scrutinize the cell kinetics by virtue of FTOC in the presence or absenceof CsA. CsA is known to exert many kinds of suppressive effects on the immune system, including blockade of the signal transduction cascadein T cells triggered by perturbation of the TCR/CD3 complex (6,25,26). The exact mechanism of this blockade is not yet known, but previous reports suggestthat CsA does not affect the signal cascade from perturbation of the TCR/CD3 complex to calcium influx or turnover of phos-
EFFECTS OF CsA ON THYMOCYTE
CD8
DIFFERENTIATION
IN FTOC
313
CD8
FIG. 3. The CD4-8- cells developing in the presenceofCsA did not near &TCR chains. Positive control of 597 is shown using adult thymocytes (A). Dotted line and solid line indicate autofluorescence and 597staining profiles, respectively. Thymocytes from CsA-added FTOC (GD I5t 15) were treated with antiCD4 and complement. Resultant cells were stained with anti-CD8 and 597 (3C). (B) A control using antiCD8 and the second antibody alone.
phatidyl inositol(27-29). Therefore, the agent may block the signal cascadeat a later step, possibly by binding to calmodulin (7). On the other hand, the agent is also known to disturb thymic stromal cell function, including Ia expression (8). Anyway, the agent is expected to have suppressive effects on thymocyte development, either by interfering the signal cascadeor impairing the stromal cell function. FIOC has certain advantages in investigating which stage of T cell differentiation CsA would inhibit. When CsA is administrated in vivo, one cannot rule out the possibility that some cells escapethe effects of the agent. Indeed, it is reported that cells escaping the effects of CsA exhibit autoreactivity in vivo (8, 30). In the organ culture, however, we can obtain sufficiently high concentrations of CsA throughout the culture period. Besides, in FTOC, no stem cells immigrate into the thymus during the culture period, and cells differentiate in a synchronized manner. Moreover, mature SPcells do not emigrate from the thymic lobes. All of these made it possible to analyze the fine differentiation kinetics of thymocytes. In the previous study we demonstrated that CsA inhibits the development of CD4, but not CD8 SP cells in vitro (4). The present study demonstrated that the CD8 SP cells developing in the presence of CsA showed immature phenotype of Jl ld+TCR& (Fig. 2). They were also shown to be functionally immature and were also less in number than CsA-free FTOC. Thus, they were early CD4-8+ cells which developed from DN cells and would develop into DP cells (Fig. 1). Hence, CsA inhibits the differentiation step from DP cells not only to CD4, but also to CD8 mature SP cells.
314
MATSUHASHI,
KAWASE, AND SUZUKI
Then what is the mechanism of this inhibitory effect? Two explanations are possible. One is that CsA inhibits the positive selection of thymocytes by blocking the signal transduction cascade initiated by perturbation of the TCR/CD3 complex (6, 7). The other is that CsA impairs thymic stromal cell functions and it leads to failure of mature SP cell generation, which may or may not be related to the process of positive selection. In the previous study, we could not demonstrate a decreasein the expression of the Ia molecule on thymic stromal cells in CsA-added thymic lobes in our culture condition (4). Although it is not a formal proof, these results suggestthat the positive selection takes place at the transition step from DP cells to SP cells, and this hypothesis is consistent with the results of a recent report in which DP cells, but not SP cells, could develop in the absence of appropriate MHC in TCR gene-transgenie mice (14). The inhibitory effect of CsA on thymocyte differentiation may be due to direct blocking of signal cascade of thymocytes, although other possibilities such that CsA may inhibit the secretion of certain factors from stromal cells cannot be ruled out. Interestingly, CsA inhibited the differentiation of mature SP cells, but not of CD3+4-8-TCRc$-Jl Id- cells, which were supposed to be cells bearing TCR-y6 chains. This result agreeswith a recent report where CsA has been administered in vivo (3 1,32). In chicken it has been reported that thymocytes bearing TCR y&homolog at high density are present in the subcapsular zone of thymic cortex, where cells bearing a/3-homolog are rare (33). Thus the requirement of interaction between thymocytes and thymic stromal cells may be different in the development of c@ and y6TCR-bearing cells. Further investigations will be necessary to elucidate the positive selection of TCR-c$ and TCR-y&bearing cells and its requirement for stromal elements. Long-lived progenitor cells have been reported to be present in the thymus after intrathymic injection of bone marrow, spleen, or fetal liver cells of a Thy- 1 disparate donor (34). An immunohistochemical study also revealed a focal donor cell colony at the site of intrathymic injection, even after 10 months (K. Hirokawa, Tokyo Metropolitan Institute of Gerontology, personal communication). In our study using FTOC, the first wave of thymocyte differentiation was followed by another wave, and all four CD4/8 populations could be detected throughout the culture period after GD15+5. This also suggested the presence of long-lived progenitor cells in the thymus. ACKNOWLEDGMENTS We are grateful to Drs. Tomio Tada and Ralph Kubo for providing mAb 597, to Dr. Katsuiku Hirokawa for letting us quote his unpublished data, and to Mrs. Sumiko Shinohara for her excellent technical assistance.
REFERENCES 1. Marusic-Galesic, S., Stephany, D. A., Longo, D. L., and Kruisbeek, A. M., Nature (London) 333, 180, 1988. 2. Kruisbeek, A. M., Mond, J. J., Fowlkes, B. J., Carmen, J. A., Bridges, S., and Longo, D. L., .I. Exp. Med. 161,1029, 1985. 3. DeLuca, D., J. Immunol. 136,430, 1986. 4. Takeuchi, Y., Habu, S., Okumura, K., and Suzuki, G., Immunology66,362, 1989. 5. Fowlkes, B. J., Schwartz, R. H., and Pardoll, D. M., Nature (London) 334,620,1988.
EFFECTS OF CsA ON THYMOCYTE
DIFFERENTIATION
IN FTOC
315
6. Sawada, S., Suzuki, G., Kawase, Y., and Takaku, F., J. Immunol. 139,1797,1987. 7. Colombani, P. M., Robb, A., and Hess, A. D., Science 228,337, 1985. 8. Cheney, R. T., and Sprent, J., Transplant. Proc. 17,528, 1985. 9. Beschomer, W. E., Di Gennaro, K. A., Hess, A. D., and Santos, G. W., Cell. Immunol. 110,350, 1987. 10. Bevan, M. J., Nature (London) 269,417,1977. 11. Zinkemagel, R. M., Callahan, G. N., Althage, A., Cooper, S., Klein, P. A., and Klein, J., J. Exp. Med. 147,882, 1978. 12. Teh, H. S., Kisielow, P., Scott, B., Kishi, H., Uematsu, Y., Bluthmann, H., and von Boehmer, H., Nature (London) 335,229, 1988. 13. Kisielow, P., Teh, H. S., Bluthmann, H., and von Boehmer, H., Nature (London) 335,730, 1988. 14. Sha, W. C., Nelson, C. A., Newberry, R. D., Kranz, D. M., Russel, J. H., and Loh, D. Y., Nature (London) 336,73, 1988.
15. Ceredig, R., Lowenthal, J. W., Nabholz, M., and MacDonald, H. R., Nature (London) 314,98, 1985. 16. Dialynas, D. P., Wilde, D. B., Marrack, P., Pierres, A., Wall, K. A., Harvan, W., Otten, G., Loken, M. R., Pierres, M., Kappler, J., and Fitch, F. W., Immunol. Rev. 74,29, 1983. 17. Staerz, U. D., Rammensee, H. G., Benedetto, J. D., and Bevan, M. J., J. Zmmunol. 134,3994, 1985. 18. Leo, O., Foo, M., Sachs, D. H., Samelson, L. E., and Bluestone, J. A., Proc. Nutl. Acud. Sci. USA 84, 1374, 1987. 19. Bruce, J., Symington, F. W., McKearn, T. J., and Sprent, J., J. Immunof. 127,2496, 198 1. 20. White, J., Herman, A., Pullen, A. M., Kubo, R., Kappler, J. W., and Mar-rack, P., Cell56,27, 1989. 21. Jenkinson, E. J., Franchi, L. L., Kingston, R., and Owen, J. J. T., Eur. J. Immunol. 12,583, 1982. 22. Crispe, N., and Bevan, M. J., J. Immunol. 138,2013, 1987. 23. Kisielow, P., Bluthmann, H., Staetz, U. D., Steinmetz, M., and von Boehmer, H., Nature (London) 333,742, 1988. 24. Kappler, J. W., Roehm, N., and Mat-rack, P., Ce/l49,273, 1987. 25. Shevach, E. M., Annu. Rev. Immunol. 3,397, 1985. 26. B&ton, S., and Palacios R., Immunol. Rev. 65,5, 1982. 27. Kay, J. E., Benzie, C. R., and Borghetti, A. F., Immunology 50,441, 1983. 28. Metcalfe, S., Transplantation 39, 161, 1984. 29. Bijsterbosch, M. K., and Klaus, G. G. B., Immunology56,435, 1985. 30. Sorokin, R., Kimura, H., Schroder, K., Wilson, D. H., and Wilson, D. B., J. Exp. Med. 164, 1615, 1986. 3 I. Jenkins, M. K., Schwartz, R. H., and Pardoll, D. M., Science 241, 1655, 1988. 32. Gao, E., Lo, D., Cheney, R., Kanagawa, O., and Sprent J., Nature (London) 336, 176, 1988. 33. Bucy, R. P., Chen, C. H., Losch, U., andcooper, M. D., J. Zmmunol. 141,2200, 1988. 34. Katsura, Y., Kina, T., Takaoki, Y., and Nishikawa, S., Eur. J. Immunol. 18,889, 1988.