COMMENT I M M U N O L O G Y T O D AY
66 Jiang, D. et al. (1996) p53 prevents maturation to the CD41CD81 stage of thymocyte differentiation in the absence of T cell receptor rearrangement. J. Exp. Med. 183, 1923–1928 67 Bogue, M.A. et al. (1996) p53 is required for both radiation-induced differentiation and rescue of V(D)J rearrangement in scid mouse thymocytes. Genes Dev. 10, 553–565 68 Maraskovsky, E. et al. (1997) Bcl-2 can rescue T lymphocyte development in interleukin-7 receptor-deficient mice but not in mutant rag-12/2 mice. Cell 89, 1011–1019 69 Newton, K. et al. (2000) FADD/MORT1 regulates the pre-TCR checkpoint and can function as a tumour suppressor. EMBO J. 19, 931–941 70 Henning, S.W. and Cantrell, D.A. (1998) p56lck signals for regulating thymocyte development can be distinguished by their dependency on Rho function. J. Exp. Med. 188, 931–939 71 Aifantis, I. et al. (1999) Allelic exclusion of the T cell receptor b locus requires the SH2 domain-containing leukocyte protein (SLP)-76 adaptor protein. J. Exp. Med. 190, 1093–1102 72 Yablonski, D. et al. (1998) Uncoupling of nonreceptor tyrosine kinases from PLC-g1 in an SLP-76-deficient T cell. Science 281, 413–416
73 Williams, B.L. et al. (1999) Phosphorylation of Tyr319 in ZAP-70 is required for T-cell antigen receptor-dependent phospholipase C-g1 and Ras activation. EMBO J. 18, 1832–1844 74 Sun, Z. et al. (2000) PKC-u is required for TCR-induced NF-kB activation in mature but not immature T lymphocytes. Nature 404, 402–407 75 Levelt, C.N. et al. (1993) Restoration of early thymocyte differentiation in T-cell receptor b-chain-deficient mutant mice by transmembrane signaling through CD3e. Proc. Natl. Acad. Sci. U. S. A. 90, 11401–11405 76 Levelt, C.N. et al. (1995) Regulation of thymocyte development through CD3: functional dissociation between p56lck and CD3z in early thymic selection. Immunity 3, 215–222 77 Mombaerts, P. et al. (1992) RAG-1 deficient mice have no mature B and T lymphocytes. Cell 68, 869–877 78 Shinkai, Y. et al. (1992) RAG-2-deficient mice lack mature lymphocytes owing to inability to initiate V(D)J rearrangement. Cell 68, 855–867 79 Jacobs, H. et al. (1994) CD3 components at the surface of pro-T cells can mediate pre-T cell development in vivo. Eur. J. Immunol. 24, 934–939 80 Shinkai, Y. and Alt, F.W. (1994) CD3e-mediated signals rescue the development of CD41CD81 thymocytes in RAG-22/2 mice in the absence of TCR b chain expression. Int. Immunol. 6, 995–1001
letters Do glucocorticoids participate in thymocyte development? There is a substantial body of evidence suggesting that glucocorticoids affect T-cell development. In this issue, Godfrey and colleagues1 review this literature in light of their recent finding that prenatal thymocyte development is grossly normal in glucocorticoid receptor knockout (GR2/2) mice. Some of the data in this area have been generated either in manipulated normal animals or in transgenic mice that express antisense GR transcripts in many tissues. Because of feedback regulation of systemic glucocorticoids on the hypothalamus–pituitary–adrenal (HPA) axis, and the fact that glucocorticoids are also produced locally in the thymus, it is difficult to know in these studies exactly how glucocorticoids or weak agonists/antagonists like RU-486 exert their effects. We have attempted to minimize the influence of adrenal steroids and the HPA axis by using two different experimental approaches: expression of GR antisense transcripts under the control of the lck proximal promoter – that is, only in thymocytes (TKO
mice) – and fetal thymic organ culture (FTOC) in which glucocorticoid biosynthesis is inhibited. Both approaches have provided consistent data suggesting not only a role for glucocorticoids in T-cell development, but also possible molecular mechanisms. Do the data with the GR knockout mice now preclude a role for glucocorticoids in thymocyte development? We believe that at this time we do not have sufficient information to reach a conclusion. We have proposed that glucocorticoids, by interfering with the nuclear consequences of TCR-mediated activation, set the signaling thresholds for different selection outcomes. This is supported by the finding that CD5 expression on double positive (DP) thymocytes, a sensitive indicator of TCR occupancy, is upregulated in an MHC-dependent manner when glucocorticoid signaling is attenuated in vivo or in vitro2. Antigen-specific selection has been addressed in several ways. Addition of metyrapone, an inhibitor of corticosterone synthesis, to FTOC caused the apoptosis of antigen-specific TCR transgenic DP thymocytes that normally undergo positive selection (reversed by corticosterone), but had no effect on viability if the MHC haplotype was incapable of presenting the selecting antigen3. In another approach, the number of T cells expressing TCRs that
0167-5699/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved.
N O V E M B E R 644
Vo l . 2 1
2 0 0 0 No.12
are positively selected (identified by the use of particular Vb regions) was reduced by introduction of the GR antisense transgene4. Even more strikingly, TKO mice were found to have an altered TCR repertoire, being nonresponsive to pigeon cytochrome c but responding normally to other complex antigens5. Is selection altered in the GR2/2 mice? Based on the available data it is not possible to say. Negative selection in response to potent stimuli appears to be grossly normal, but without dose–response curves it is not possible to say whether GR2/2 DP thymocytes are unusually sensitive. Positive selection is even more problematic. Positive selection clearly occurs in GR2/2 mice, but does the TCR repertoire differ from that of wild-type animals? At this time we do not know, but characterization of the effect of GR loss on MHC-dependent Vb use in normal animals, and selection in antigenspecific ab TCR transgenic mice, should shed light on this issue. The major documented difference between thymuses from GR2/2 and TKO mice or from normal mice cultured in FTOC with metyrapone is that only the former have normal numbers of DP cells. Is it possible to reconcile these very different findings? One possibility is that there might be strainspecific differences between the mice. We
COMMENT I M M U N O L O G Y TO D AY
have, for example, found that the number of MHC-encoded molecules inversely correlates with the number of DP cells recovered in glucocorticoid-deficient FTOC (Ref. 2) and in TKO mice (our unpublished data) of the C57BL background. The GR2/2 mice are on the 129/SV background (H-2b), and it is possible that this relatively noncomplex MHC in combination with other, unknown, differences in this genetic background might make these thymocytes less dependent upon GR signaling. This can easily be addressed by backcrossing the GR2/2 animals with C57BL/6 mice. A second possibility is that global GR loss might result in compensatory changes that counter thymocyte unresponsiveness to glucocorticoids. For example, we have found that glucocorticoids potently upregulate IL-7 receptor a (IL-7Ra) (D. Franchimont et al., unpublished), which is necessary for normal progression of DN to DP cells6. If glucocorticoids inhibit thymic IL-7 production, global loss of GR expression could result in lower IL-7Ra levels on thymocytes but higher IL-7 production in the thymus, compensating changes that would be evident both in vivo and in FTOC. Consistent with the notion that globally decreased GR expression might compensate for decreased thymocyte GR expression is that DP cell numbers are decreased in three independent TKO founder strains (two were reported in the initial study, the third has recently been generated and characterized in our laboratory)7, whereas the identical antisense transgene had only a modest effect on fetal thymocyte development when driven by a ubiquitously expressed promoter8. Finally, low levels of GR signaling (TKO mice or metyrapone-containing FTOC) might paradoxically have a more marked effect on thymocyte development than no GR signaling (GR2/2 mice). For example, complete lack of glucocorticoid responsiveness might yield apparently normal cellularity by allowing DP cells that would normally die by ‘neglect’ to accumulate. Godfrey and colleagues suggest that the GR1/2 mice should be analogous to the TKO animals, but these hemizygous animals appeared to be only slightly less sensitive to the pro-apoptotic effect of glucocorticoids than wild-type animals9.
The finding that prenatal thymic development is grossly normal in the absence of the GR is striking, and calls for a reevaluation of the experimental approaches that have indicated a role for glucocorticoids in T-cell ontogeny. Although all experimental approaches (including targeted gene knockout) have potential unintended consequences or artefacts, similar results obtained by very different means, such as the TKO mice and FTOC/metyrapone studies, reinforce each other. We look forward to learning what effect, if any, expression of GR antisense or inhibition of glucocorticoid synthesis will have in the absence of the GR. While glucocorticoids may not be ‘obligatory’ for the development of a populated thymus, work remains to be done to determine if they have an impact on normal T-cell ontogeny. Jonathan D. Ashwell (
[email protected]) Laboratory of Immune Cell Biology, National Institutes of Health, Bethesda, MD 20892, USA. Melanie S. Vacchio Experimental Immunology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USA. Jérôme Galon Lymphocyte Cell Biology Section, National Institute of Arthritis and Metabolism, National Institutes of Health, Bethesda, MD 20892, USA. References 1 Godfrey, D.I. et al. (2000) Stress free T-cell development: glucocorticoids are not obligatory. Immunol. Today 21, 606–611 2 Vacchio, M.S. et al. (1999) Thymusderived glucocorticoids set the thresholds for thymocyte selection by inhibiting TCR-mediated thymocyte activation. J. Immunol. 163, 1327–1333 3 Vacchio, M.S. and Ashwell, J.D. (1997) Thymus-derived glucocorticoids regulate antigen-specific positive selection. J. Exp. Med. 185, 2033–2038 4 Tolosa, E. et al. (1998) Thymocyte glucocorticoid resistance alters positive selection and inhibits autoimmunity and lymphoproliferative disease in MRL-lpr/lpr mice. Immunity 8, 67–76
5 Lu, F.W.M. et al. (2000) Thymocyte resistance to glucocorticoids leads to antigen-specific unresponsiveness due to ‘holes’ in the T cell repertoire. Immunity 12, 183–192 6 Peschon, J.J. et al. (1994) Early lymphocyte expansion is severely impaired in interleukin 7 receptor-deficient mice. J. Exp. Med. 180, 1955–1960 7 King, L.B. et al. (1995) A targeted glucocorticoid receptor antisense transgene increases thymocyte apoptosis and alters thymocyte development. Immunity 3, 647–656 8 Sacedon, R. et al. (1999) Partial blockade of T-cell differentiation during ontogeny and marked alterations of the thymic microenvironment in transgenic mice with impaired glucocorticoid receptor function. J. Neuroimmunol. 98, 157–167 9 Purton, J.F. et al. (2000) Intrathymic T cell develoment and selection proceeds normally in the absence of glucocorticoid receptor signaling. Immunity 13, 179–186
Reply to Ashwell, Vacchio and Galon Even before the publication of our recent paper1 describing T-cell development in glucocorticoid receptor deficient (GR2/2) mice, the role of glucocorticoids in T-cell development was highly controversial. Thus, our Viewpoint2 does not simply reflect us reassessing the field in light of our own data, and could have been written in the absence of our data. It is not easy to explain this controversy, and although we can debate about the strengths and weaknesses of each system used, the most sensible approach to resolving this problem involves further experimentation. Meanwhile, although some of the issues raised in the reply by Ashwell and colleagues are already covered in our viewpoint and need not be restated, we wish to comment on the main concerns raised about our data using GR2/2 thymuses. We agree that it is important to minimize the influence of adrenal steroids and the hypothalamus-pituitary-adrenal (HPA) axis. Like Ashwell and colleagues3–6, we have included the fetal thymic organ culture (FTOC) system to study T cell development in the absence of GR signaling1. Obviously, this can not explain the differences in our results. 0167-5699/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved.
N O V E M B E R Vo l . 2 1
No.12
2 0 0 0 645
