Enhanced apoptosis in the thymus of transgenic mice ... - Nature

Oncogene (1997) 15, 601 ± 605  1997 Stockton Press All rights reserved 0950 ± 9232/97 $12.00

Enhanced apoptosis in the thymus of transgenic mice expressing constitutively activated forms of human Rac2GTPase Patrick LoreÁs, Laurence Morin, Rosa Luna and GeÂrard Gacon Institut Cochin de GeÂneÂtique MoleÂculaire, INSERM Unite 257, 24 Rue du Faubourg St Jaques 75014, Paris, France

Rac proteins constitute a subgroup of the Rho family of small GTPases and include Rac1, which is expressed ubiquitously, and Rac2, a highly homologous protein only expressed in myelo-monocytic and lymphoid cell lineages. In ®broblasts, Rac1 plays a crucial role in control of actin cytoskeleton organisation, cell growth and Ras-induced transformation. In phagocytes, Rac1 and Rac2 regulate a speci®c enzymatic complex, NADPH oxidase. These multiple functions have been ascribed to Rac proteins only on the basis of cell culture and in vitro biochemical studies. To examine the role of Rac2 in vivo in a T cell lineage, we have expressed either wild-type or constitutively-activated forms of human Rac2 (Rac2V12 and Rac2L61) in transgenic mice under control of the thymus speci®c lck proximal promoter. We report here a striking atrophy of the thymus in mice expressing even low levels of either of the activated mutants of Rac2, while expression of Rac2wt has no e€ect. This phenotype is correlated with a marked decrease in the number of double positive (CD4+CD8+) and single positive (CD4+CD87 and CD8+CD47) thymocytes. Cellular and molecular analyses demonstrate that this defect is due to an increase in apoptosis among thymocytes. As Rac2 is normally expressed in thymocytes and activated T cells, we propose that Rac2 dependent pathways could play an important role in control of growth and death of T cells.

one experiment involving rac1 transgenesis in mouse has been reported and showed that expression of activated Rac1 in Purkinje cells perturbed axonal and dendritic growth (Luo et al., 1996). The role of Rac2 in myeloid and lymphoid cells has not been documented to the same extent as that of Rac1 in ®broblasts. However, in phagocytes, Rac1 and Rac2 have been shown to regulate through a direct protein-protein interaction, a speci®c multicomponent enzyme, NADPH oxidase, responsible for the production of bactericidal superoxide anions (Knaus et al., 1991; Abo et al., 1991; Dorseuil et al., 1992), and we have recently demonstrated that Rac2 is signi®cantly more ecient in this interaction than Rac1 (Dorseuil et al., 1996). This result suggested that Rac2 could ful®l other speci®c functions in specialized cells of the hematopoietic systems. More speci®cally, as rac2 is strongly expressed in thymus (Shirsat et al., 1990) and in activated peripheral T lymphocytes (Reibel et al., 1991), we hypothesized that it could be implicated in T cell speci®c pathways. The current attempt to express activated mutants of Rac2 in mouse thymus was undertaken with the aim of assessing the function of rac2 in a T cell lineage, under in vivo conditions. The results strongly suggest a so far unrecognized function of Rac2 in the regulation of apoptotic death of thymocytes.

Keywords: Rac2GTPase; thymocyte; apoptosis; transgenic mice

Results and discussion

Introduction RacL61 and RacV12 are mutants of Rac proteins, which are blocked in their GTP bound (or active) conformation and are therefore capable of activating constitutively the signaling pathway(s) downstream of Rac. Such Rac1 mutants have been widely used in microinjection and transfection experiments to assess the functions of Rac dependent pathways in fibroblasts. In the past few years, these in vitro approaches have demonstrated that Rac1 regulates a number of responses including formation of ru‚es and lamellipodia, activation of the JNK/p38 MAPkinase cascade, and progression of the cell cycle through the G1 phase (reviewed in Hall, 1994; Ridley, 1996). Rac1, in the activated mutated form Rac1V12, is capable to transform Rat1 ®broblasts; Rac1 is also one of the downstream targets required for Ras induced transformation (Qiu et al., 1995; Ridley, 1996). So far, only Correspondence: G Gacon Received 7 April 1997; revised 27 June 1997; accepted 27 June 1997

We have generated transgenic mice which express constitutively activated forms of Rac2 speci®cally in thymus. The Rac2 mutants were human Rac2V12 and human Rac2L61, two forms blocked in their GTPbound con®guration, and they were expressed under control of the thymus-speci®c proximal promoter of lymphoid cell kinase (lck) (Chan et al., 1990). Two mouse lines expressing rac2V12 were obtained, ®ve for rac2L61 and two for control rac2wt (wild-type). Northern blot analysis (Figure 1b) showed a complex set of human rac2 transcripts only in the thymus of transgenic animals, as has been previously observed with other transgenes driven by the lck promoter (Chan et al., 1990; King et al., 1995). Only a modest overall increase in Rac2 protein expression (4twofold) was detected in the thymus of all transgenic animals (not shown). C-Jun N-terminal kinase (JNK) (Figure 1c) measured in thymus extracts from 11 to 20-day-old animals was found repeatedly enhanced by 2 ± 3-fold in rac2V12 and rac2L61 transgenics with respect to littermate or rac2wt transgenic controls, re¯ecting the expected stimulation of this pathway in cells expressing activated Rac (Coso et al., 1995; Minden et al., 1995). Visual inspection of the thymus of 15-day-old animals (Figure 2) revealed a striking reduction in

Apoptosis in the thymus of rac2 transgenic mice P LoreÁs et al

thymus size in rac2V12 and rac2L61 transgenic mice as compared to non-transgenic littermates. This phenotype was observed in the two rac2V12 and the ®ve rac2L61 lines examined but not in the two rac2wt lines. This reduction in thymus size was already apparent at day 9 and became more obvious as mice grew older. Cell counts using 11 ± 15-day-old litters con®rmed these observations. For rac2L61 (n=9) there were 21.5+1.56106 cells/thymus (6+s.e.m.) versus 157+8 6106 cells/thymus for littermate controls (n=9) and 62+316106 cells/thymus for rac2V12 (n=4). In order to determine whether this decrease in cell number was cell-speci®c, we performed FACS analysis of thymocytes from rac2L61 and rac2V12 transgenics, labeled with antiCD4 and antiCD8 antibodies (Figure 3). There was a dramatic decrease in the absolute number of both double positive (CD4+/CD8+ or DP) and single positive (CD4+ and CD8+ or SP) cells and a concomitant relative increase in double negative

a

(CD47/CD87 or DN) cells (Table 1 and Figure 3). Therefore, the main apparent defect in rac2L61 or rac2V12 thymus is a depletion in DP cells, whereas DN cells and DP to SP transition are not grossly a€ected. In lymph nodes and spleen, FACS analysis revealed a clear reduction in the proportion of mature CD4+ and CD8+ cells (Figure 3). Counting of T cells in spleen con®rmed a dramatic decrease in the absolute number of peripheral SP cells showing 7.356106 CD4+ cells and 1.46106 CD8+ cells in a rac2L61 spleen versus 62.06106 CD4+ and 25.96106 CD8+ cells in the spleen of a littermate control. This result probably re¯ects the decrease in the production of single positive cells by the thymus rather than a peripheral deletion of the mutant cells, since transgenes driven by the lck proximal promoter are known to be expressed primarily in early thymocytes (Garvin et al., 1990). The normal fate of the vast majority of DP cells during thymus development is apoptotic death, occurring primarily by default of TCR stimulation (death by neglect) or by engagement of TCR with excessive avidity for self-peptide/MHC (negative selection) (Osborne, 1996; Surh and Sprent, 1994). We therefore examined apoptosis in thymus from rac2L61 transgenics using the TUNEL procedure. Thymus sections from 11-day-old rac2L61 transgenics

Probe

V12-Thymus

WT-Thymus

L61-Spleen

L61-Thymus

Control-Thymus

b