Expression of Mad1 in T Cells Leads to Reduced Thymic Cellularity and Impaired Mitogen-Induced Proliferation
Total Page:16
File Type:pdf, Size:1020Kb
Oncogene (2001) 20, 1164 ± 1175 ã 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00 www.nature.com/onc Expression of Mad1 in T cells leads to reduced thymic cellularity and impaired mitogen-induced proliferation Bettina Rudolph1,2, Anne-Odile Hueber1,3 and Gerard I Evan*1,4 1Imperial Cancer Research Fund, PO Box 123, 44 Lincoln's Inn Fields, London WC2A 3PX, UK To investigate Mad1 function in vivo, transgenic mice only the interaction between Mad1 and mSin3A/Sin3B/ were generated that express a Mad1 transgene in T HDAC but also block Mad1 transcriptional repression lineage cells under the control of the proximal lck (Ayer et al., 1995). This suggests that Mad1 represses promoter. Thymus size in lck-Mad1 transgenic mice is transcription by recruiting histone deacetylases to drastically reduced although representation of the various target sites on DNA (for review: Kiermaier et al., thymocyte sub populations appears normal. To investi- 1997). gate more closely any eects of Mad1 expression on In addition to Mad1, the Mad protein family also thymocytes, we examined thymic selection using MHC comprises Mxi (Zervos et al., 1993), Mad3 and Mad4 class I-restricted H-Y-TCR transgenic mice. Mad1 (Hurlin et al., 1995), all of which speci®cally associate expression in vivo reduces the eciency of positive with Max to repress transcription. Mad proteins, like selection. Furthermore, thymocytes and splenic T cells Myc proteins, are short-lived (Ayer et al., 1993) and from lck-Mad1 transgenic mice display a profound their expression is tightly regulated during cell cycle proliferative defect in response to activation with either progression and in response to induced dierentiation PMA/Ionomycin or immobilized anti-CD3/CD28 anti- in various cell lines (Ayer et al., 1993; Zervos et al., body. This proliferative defect is not reversed by addition 1993; Hurlin et al., 1995; Larsson et al., 1994). In of exogenous IL-2 and is p53-independent. The growth contrast, expression of Max protein is constitutive. inhibition caused by Mad1 is overcome by expression of However, because both Myc and Mad1 compete for active c-Myc. Oncogene (2001) 20, 1164 ± 1175. Max (Ayer et al., 1993), access of Myc to Max is profoundly dependent upon expression of Mad1 (and Keywords: Myc; Mad; oncogene; cell cycle; lympho- vice versa). For this reason, Mad1/Max complexes cyte; tumor suppressor function as repressors of Myc activity as evidenced by the fact that Mad1/Max (and Mxi/Max) inhibit Myc transactivation of reporter plasmid constructs Introduction containing the Myc consensus element (Ayer et al., 1993; Zervos et al., 1993; Backwood et al., 1992) and Mad1 is one of a family of three known bHLHLZip that Mad1 suppresses Myc transforming activity in proteins (Ayer et al., 1993). Mad1 homodimerizes rat embryo ®broblasts (Lahoz et al., 1994). However, poorly but interacts with Max to form a sequence- it is not clear whether the biological activity of Mad speci®c DNA-binding heterodimer. The bHLHLZip proteins resides solely in their abilities to antagonize domain of Mad1 is required for both heterodimeriza- Myc proteins or whether, as dominant trans-repres- tion with Max and its sequence-speci®c DNA binding sors, they ful®ll other roles. Therefore, to gain (Blackwell et al., 1990). The Mad1/Max complex binds further insight into the biological action of Mad1 in to the same E box motif as Myc/Max heterodimers vivo, we have generated transgenic mice in which the but, in contrast to Myc/Max, Mad1/Max complexes expression of a Mad1 transgene is targeted to the function as transcriptional repressors (Amati et al., thymus, a tissue in which both c-Myc and Mad1 are 1992) through interaction with the co-repressors normally expressed (Broussard-Diehl et al., 1996; mSin3A/mSin3B and the mammalian histone acetylases Foley et al., 1998). Indeed, in situ hybridization HDAC1 and HDAC2. Mad1 association with mSin3A/ analysis indicates that dierent members of the Mad mSin3B/HDAC requires the N-terminal 25 residues of gene family are expressed in distinct, yet partially Mad1. Point mutations in this region eliminate not overlapping patterns in adult thymus (Queva et al., 1998). Since proliferation, dierentiation and apopto- sis play such crucial roles during T cell development, *Correspondence: GI Evan the thymus oers itself as a particularly intriguing Current addresses: 2DeveloGen AG, Rudolf-Wissell-Str. 28, D - experimental system in which to examine the 37079 GoÈ ttingen; 3Centre d'Immunologie, INSERM-CNRS de consequences of manipulating the Myc/Max/Mad1 Marseille-Luminy, Case 906, F - 13288 Marseille Cedex 9, France; network in vivo. 4 UCSF Cancer Center and Cancer Research Institute, 2340 Sutter To date, transgenic mouse analysis of the Myc/Max/ Street, San Francisco, California 94143-0875, USA Received 12 October 2000; revised 12 December 2000; accepted 19 Mad1 network has been restricted to ectopic expression December 2000 of c-myc within the lymphoid compartment (CD2-c- Analysis of Mad1 function in T cells using lck-Mad1 transgenic mice B Rudolph et al 1165 myc, Thy1-c-myc,Em-c-myc) (Stewart et al., 1993; CD8+ single positive thymocytes (CD8+ SP, CD4+ SP) Spanopoulou et al., 1989; Adams et al., 1985). from lck-Mad1 transgenic mice showed equal expres- However, in all such cases deregulated c-myc expres- sion of the transgene in all thymocyte sub populations sion results in the formation of tumors, making it (Figure 1c). Expression of endogenous Mad1 was dicult to analyse normal lymphocyte dierentiation detected in the DN, DP and to a weaker extent in in these mice. C-myc knockout mice have also been the CD4+ SP sub population (Figure 1c). In addition, generated but these die between day E9.5 and E10.5 of RT ± PCR analysis on sorted thymocyte sub popula- gestation (Davis et al., 1993), before the onset of tions from wild-type and Mad1 transgenic mice showed lymphocyte ontogeny. In contrast, the consequences of comparable expression of other Mad family members ectopic Mad1 expression has not yet been investigated such as Mxi and Mad4 in dierent thymocyte sub and although Mad17/7 have been generated and are populations (Figure 1c). viable, no data are available concerning T cell We con®rmed immunohistochemically protein ex- development in these mice. pression of the transgenic Mad1 protein in thymus In this paper, we investigate the eects of expression using a Mad1 speci®c antibody. Staining for the of a Mad1 transgene under the control of the endogenous Mad1 protein was detected with weak proximal lck promoter on thymocyte development. intensity. Speci®city of the staining was con®rmed by We show that although thymus size in lck-Mad1 preincubation of the Mad1 speci®c antibody with its transgenic mice is drastically reduced, general repre- immunogenic peptide which blocked all signal (Figure sentation of individual thymocyte sub populations 1d). seems overtly normal. We investigated the eects of Mad1 expression on positive thymocyte selection using Lck-Mad1 transgenic mice exhibit reduced numbers of MHC class I-restricted H-Y-TCR transgenic mice. thymocytes and peripheral T cells but thymocyte sub This analysis shows that Mad1 expression reduces the populations are represented normally eciency of positive selection. We also show that thymocytes and splenic T cells from lck-Mad1 Mad1 expression had a dramatic eect on thymic transgenic mice display a profound proliferative defect cellularity: thymus size, as well as the total thymocyte when activated with PMA/Ionomycin or anti-CD3/ number, was reduced by approximately 50% relative to CD28 antibody in vitro. Only the ®rst thymocyte G0 - wild-type control littermates (Figure 2a right panel). S transition is inhibited by Mad1 expression and this To examine the eects of Mad1 expression on T cell inhibition can be, at least in part, overcome by co- development, thymocytes from lck-Mad1 transgenic expression of c-Myc. mice and from control littermates (4 ± 10 weeks old) were analysed by three-color ¯ow cytometry using antibodies against CD25, CD44, CD3e, CD69 and TCRa in combination with CD4 and CD8 to Results discriminate between the dierent thymocyte sub populations. These analyses indicated that Mad1 Generation of lck-Mad1 transgenic mice expression caused no signi®cant change to the ratios To examine the action of Mad1 in vivo, we generated of the four thymocyte sub populations in the adult transgenic mice in which a Mad1 transgene was (Figure 2a). In addition, a more detailed analysis of the expressed under the control of the T cell lineage speci®c four DN sub populations did not reveal a dierence proximal lck promoter which directs expression of the between control littermates and lck-Mad1 transgenic transgene to thymocytes and mature peripheral T mice (data not shown) indicating no developmental lymphocytes (Wildin et al., 1991). A schematic blockade in early T cell development in lck-Mad1 representation of the transgenic construct is depicted transgenic mice. The level of all cell surface markers in Figure 1a. Construct DNA was injected into C57BL/ tested was unchanged. 66CBA F1 embryos using standard techniques (Brin- To examine any eect that Mad1 expression might ster et al., 1985). Two founder lines were generated, have on peripheral T lymphocytes, lymph node cells both harbouring roughly the same copy number (*10 were analysed by two-color ¯ow cytometry using copies, data not shown). Both lines behaved identically antibodies against CD4 and CD8 to discriminate and are therefore not further distinguished. between the dierent T cell sub populations. These Expression of the Mad1 transgene was monitored by analyses showed a reduction in the percentage of the Northern blot analysis using total RNA and a human CD4+ and CD8+ SP population which was accom- Mad1 cDNA as a probe. Expression of the transgene panied by a reduced number of mature T cells in the was detected in transgenic thymus and spleen tissue.