CREB function is required for normal thymic ... - Wiley Online Library

6 downloads 31312 Views 331KB Size Report
Sven Baumann1,2, Bruno Kyewski3, Susanne C. Bleckmann1, Erich Greiner1, Dorothea ... reduced thymic cellularity and delayed thymic recovery following sublethal irradiation but no ... promoter used to drive dnCREB overexpression. Eur.
Eur. J. Immunol. 2004. 34: 1961–1971

CREB and thymic cellularity

1961

CREB function is required for normal thymic cellularity and post-irradiation recovery Sven Baumann1,2, Bruno Kyewski3, Susanne C. Bleckmann1, Erich Greiner1, Dorothea Rudolph1, Wolfgang Schmid1, Robert G. Ramsay4, Peter H. Krammer2, Günther Schütz1 and Theo Mantamadiotis1,4 1

Molecular Biology of the Cell I, Deutsches Krebsforschungszentrum, Heidelberg, Germany Cellular Immunology, Deutsches Krebsforschungszentrum, Heidelberg, Germany Immunogenetics, Deutsches Krebsforschungszentrum, Heidelberg, Germany 4 Differentiation and Transcription Laboratory, Peter MacCallum Cancer Centre, East Melbourne, Australia 2 3

Recent generation of genetically modified Creb1 mutant mice has revealed an important role for CREB (cAMP responsive element binding protein) and the related proteins CREM (cAMP responsive element modulator) and ATF1 (activating transcription factor 1) in cell survival, in agreement with previous studies using overexpression of dominant-negative CREB (dnCREB). CREB and ATF1 are abundantly expressed in T cells and are rapidly activated by phosphorylation when T cells are stimulated through the T cell antigen receptor. We show that T cell-specific loss of CREB in mice, in combination with the loss of ATF1, results in reduced thymic cellularity and delayed thymic recovery following sublethal irradiation but no changes in T cell development or activation. These data show that loss of CREB function has specific effects on thymic T lymphocyte proliferation and homeostasis in vivo. Key words: Rodent / T lymphocytes / Transcription factors / Transgenic/knockout / Thymus

1 Introduction Several intracellular signaling pathways converge onto the CREB (cAMP responsive element binding protein)/ ATF1 (activating transcription factor 1) family of the leucine-zipper class of transcription factors. These factors bind the cAMP responsive element (CRE) in promoter regions of a variety of genes (reviewed in [1]). Three members have been described: CREB, CREM (cAMP responsive element modulator) and ATF1. Once phosphorylated via a variety of signaling pathways, these factors activate transcription of target genes as either homodimers or heterodimers, with the phosphoCREB homodimer being the most potent transcriptional activator as inferred from in vitro data [2, 3] and the severity of the phenotype of null mutations [4–6]. Studies utilizing overexpression of dominant-negative CREB

[DOI 10.1002/eji.200324826] Abbreviations: AICD: Activation-induced cell death ATF1: Activating transcription factor 1 Cre: Cre recombinase CRE: cAMP responsive element CREB: cAMP responsive element binding protein CREM: cAMP responsive element modulator dnCREB: Dominant-negative CREB Dex: Dexamethasone © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Received Revised Accepted

10/12/03 24/2/04 19/4/04

(dnCREB) revealed a role for CREB as a survival factor in various cellular models [7–10]. We have shown that CREB, CREM and ATF1 cooperate to promote cell survival in vivo [4, 11]. In humans, reduced CREB activity appears to correlate with T effector cell dysfunction seen in systemic lupus erythematosus (SLE) patients [12–14]. T cell receptor activation leads to phosphorylation of CREB [15] and activation of a variety of genes [16]. T cell target genes include TCR- § [17], CD4 [18], CD8 [19], MHC Class II [20], IFN- + [21], Bcl-2 [7], IL-2 [22] and cFos [23], implicating members of the CREB/ATF1 family in the regulation of gene expression during T cell differentiation and activation. Activated T cells from transgenic mice overexpressing a phosphorylation-defective dnCREB protein under the control of a T cell-specific CD2 promoter/enhancer show a proliferation and survival defect [24]. In contrast, a later study with mice overexpressing the same phosphorylation-defective dnCREB mutant under the control of the lck promoter showed no defect in T cell number, function or apoptosis. Instead, a specific effect on Th cell differentiation and survival was seen [25]. Collectively these studies suggest that CREB is important for T cell function but that the T cell biology of the previously reported dnCREB transgenic mice differs depending on the T cell-specific promoter used to drive dnCREB overexpression. www.eji.de

1962

S. Baumann et al.

Eur. J. Immunol. 2004. 34: 1961–1971

Since CREB can heterodimerize with either CREM or ATF1 [3, 26], it is unclear whether phosphorylationdefective dnCREB specifically inhibits endogenous CREB or whether the function of other members of the CREB/CREM/ATF1 subfamily are also affected, thereby having a profound effect on cellular survival. Moreover, there is evidence that overexpression of dnCREB has the unpredicted capacity to interfere with AP-1 function [27]. Opting for an alternative to transgenic dnCREB overexpression for the further investigation of the role of CREB in development and function of T cells in adult mice, T cell-specific knockout mice were generated by using the Cre recombinase (Cre)/loxP recombination system. This was achieved by crossing Creb1lox mice [11] to lckCre mice [28]. To exclude compensatory effects by other members of the CREB/ATF family expressed in thymus, these mice were crossed to Atf1–/– mice [4] to generate mice devoid of both CREB and ATF1 in T cells (Creb1lckCreAtf1–/–). Our data show that CREB activity contributes to the maintenance of steady-state thymic cellularity and is further required for recovery following radiation exposure.

2 Results 2.1 Specificity of CREB recombination and loss As the interpretation of this study depends upon thorough documentation of the extent and specificity of CREB loss, several independent measures of gene deletion and protein loss were employed. Cre-mediated T cell-specific CREB recombination and loss (Fig. 1a) was apparent by genomic PCR analysis (Fig. 1b), Western blot analysis (Fig. 1c) and immunohistochemical staining of cytospin preparations using antibodies detecting CREB, ATF1 and Cre (Fig. 2). Using various anti-CREB antibodies, we previously showed that loss of CREB protein through recombination of the floxed Creb1 allele results in the loss of all CREB polypeptides [11]. For western blot analysis, we used an antibody directed against the epitope recognizing both CREB and ATF1. For immunostaining we used an antibody recognizing a CREB-specific epitope adjacent to the kinase-inducible domain (KID). Importantly, this epitope is upstream of the domain encoded for by the “floxed” exon 10, ensuring the absence of truncated CREB isoforms, which may act as dominant-negative peptides. In view of the complex nature of CREB family target gene regulation and redundancy, it was essential to assess the status of other CREB family members. The CREM protein was undetectable in control thymocytes, and in addition, no compensatory up-regulation of CREM was evident in Creb1–/– thymocytes (Fig. 1c). RT-PCR analysis © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Fig. 1. (a) Targeting strategy for the generation of a T cellspecific CREB knockout allele. Structure of the wild-type and targeted Creb gene locus is shown. Exon-intron structure and position of the introduced loxP sequences, flanking exon 10, are indicated. (b) Genomic PCR analysis of the Creb1 gene locus of DNA from thymus (thy), lymph node (ln), spleen (spl), liver (liv) and tail. (c) Immunoblot analysis for CREB, ATF1 and CREM expression in thymocytes from control littermate (ctrl) and Creb1lckCre (mut) mice. An antibody recognizing both CREB and ATF1 and an anti-CREM antibody were used, allowing detection of all CREB family members. Lysate from testis, where multiple CREM isoforms (arrowheads) are highly expressed, was used as a positive control for the CREM antibody. Protein determination was performed using the Bio-Rad protein assay to normalize protein loading. (d) T cells express only Crem repressor isoforms. RT-PCR analysis for detection of both activator and repressor Crem isoforms or ICER only were conducted on control (ctrl) and Creb1lckCreAtf1–/– (mut) T cells. Presence of a larger amplicon (680 bp) indicated the presence of the ICER repressor, while a smaller band (284 bp) indicated the presence of the Crem-tau activator isoform. Testis RNA was used as a positive control for activator Crem isoforms and a negative control for ICER.

confirmed these results; the CREM repressor, ICER, but no CREM activator was detected in control or Creb1lckCreAtf1–/– thymocytes (Fig. 1d). CREM activator isoforms were detected by Western analysis and RTPCR only in mouse testis (Fig. 1c, d). These data demonwww.eji.de

Eur. J. Immunol. 2004. 34: 1961–1971

CREB and thymic cellularity

1963

Fig. 2. Cytospin and subsequent immunostaining for CREB and Cre using thymocytes and MACS-purified splenic or lymph node CD4+ T lymphocytes from wild-type and Creb1lckCreAtf1–/– mice. A CREB-specific antibody [42] and a Cre-specific antibody [41] were used.

strate that Creb1lckCreAtf1–/– thymocytes do not express any known transcriptional activators of the CREB/ CREM/ATF1 transcription factor family. Cytospin analysis of thymocytes and CD4+ peripheral T cells sorted from spleen and lymph nodes and subsequent staining for CREB protein demonstrated deletion of CREB (Fig. 2 and Table 1). Collectively, Figs. 1 and 2 show that the strategy employed results in a substantial reduction of CREB expression in the thymus and the periphery. To examine the thymic organ architecture, we performed hematoxylin/eosin staining of thymi from Creb1lckCreAtf1–/– and wild-type littermate control mice. No significant differences in thymic morphology were detectable (data not shown).

2.2 Development of CREB/ATF1-deficient T lymphocytes The CREB family member ATF1 was unaffected by CREB deletion (Fig. 1c) and therefore had the potential to compensate for the absence of CREB. Thus, genetic crosses were performed to generate Creb1lckCre mice on an Atf1–/– background. Creb1lckCreAtf1–/– double-mutant mice were healthy and fertile. Mice without CREB or ATF1 in T cells showed reduced thymocyte numbers, and significantly, numbers were further reduced in Creb1lckCreAtf1–/– double-mutant mice (Fig. 3A), indicating that both CREB and ATF1 are involved in thymocyte proliferation and maturation. The overall development of immature T cells in the thymus, as characterized by FACS staining for thymocyte markers, was normal (Fig. 3B). Thymocytes were also stained for CD2, CD5,

Table 1. Expression of CREB and Cre in thymus, spleen and lymph nodea) CREB Control

Cre recombinase Creb1

lckCre

-/-

Atf1

Control

Creb1lckCre Atf1-/-

Thymus

100 % (± 3 %)

23 % (± 8 %)

0%

79 % (± 5 %)

Spleen

100 % (± 4 %)

24 % (± 3 %)

0%

47 % (± 6 %)

Lymph node

100 % (± 2 %)

12 % (± 3 %)

0%

55 % (± 9 %)

a)

Thymocytes and MACS-purified CD4+ T cells from spleen and mesenteric lymph nodes were centrifuged on cover slides and fixed with 4% paraformaldehyde. Cells were stained with antibodies specific for CREB and Cre. At least 500 cells each were counted and analyzed for staining intensity. The mean of three experiments is shown (± SEM).

© 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji.de

1964

S. Baumann et al. P

Eur. J. Immunol. 2004. 34: 1961–1971 Fig. 3. Reduced thymic cellularity but normal T cell development in mice lacking CREB and ATF1. (A) Cell number per thymus in control (n=15), Atf1–/– (n=6), Creb1–/– (n=5) and Creb1lckCreAtf1–/– (n=13) mice; p X 0.001 for Creb1lckCreAtf1–/– compared to controls. (B) T cell subsets defined by CD4/ CD8 surface staining of thymocytes from control (n=15) and Creb1lckCreAtf1–/– (n=13) mice. Relative numbers of different thymocyte subsets are shown. (C) Analysis of early CD4–CD8– thymocyte development as defined by CD44/ CD25 staining. Relative distribution of thymocyte subpopulations, gated for CD4–CD8– in FACS. (D) Cell cycle analysis of thymocytes assessed by propidium iodide staining of nuclei from control and Creb1lckCreAtf1–/– mice.

CD4–CD8– cells. Thymocytes in the CD4–CD8– doublenegative state can be further subdivided into four maturation steps depending on CD44 and CD25 status. No defect in early thymocyte development was observed in the absence of CREB family members (Fig. 3C). Furthermore, the reduction in thymocyte numbers in mutant mice was not due to a cell cycle defect under nonstimulated conditions (Fig. 3D).

2.3 Proliferative capacity of peripheral T cells As shown previously, dnCREB impairs the proliferative capacity of T cells in vitro upon TCR/CD3 stimulation [24], and CRE are found in the regulatory regions of genes relevant to T cell proliferation, e.g. the IL-2 [22] and IFN- + genes [21]. Peripheral T cells from Creb1lckCreAtf1–/– and wild-type littermate control mice were stimulated via the TCR/CD3 complex and tested for indicators of T cell activation. T cells from spleen and mesenteric lymph nodes showed normal T cell proliferation in the absence of CREB and ATF1 activity (Fig. 4A). In addition, up-regulation of CD25 (IL-2R § chain) on the cell surface and production of IFN- + upon TCR/CD3 stimulation was not significantly changed in Creb1lckCreAtf1–/– mice (Fig. 4B), showing that CREB and ATF1 are not essential for T cell activation.

2.4 Apoptosis of thymocytes and peripheral T cells

CD69, CD3, TCR § g and CD95, and no differences between Creb1lckCreAtf1–/– and wild-type littermate control mice were observed (data not shown). This implies that CREB and ATF1 are involved in thymocyte expansion but not in regulating specific maturation steps during thymocyte selection. To confirm these results, we investigated very early thymocyte development of © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Spontaneous and dexamethasone (Dex)-induced apoptosis showed the same cell-death kinetics as controls (Fig. 5a). In addition, thymocyte cell death induced upon CD3 antibody stimulation, Dex treatment or triggering of the CD95 death receptor was not significantly changed when measured at 18 h (Fig. 5b). T cells deficient for CREB family members did not exhibit differences in activation-induced cell death (AICD) via CD95L induction (Fig. 5c). CD95L expression inhibited by cAMP treatment www.eji.de

Eur. J. Immunol. 2004. 34: 1961–1971

Fig. 4. Normal activation of peripheral T cells in mice lacking CREB and ATF1. (A) Proliferation of peripheral T cells upon TCR/CD3 stimulation. T cells from spleen or mesenteric lymph nodes were stimulated with anti-CD3 antibody for 36 h, followed by addition of 1 ? Ci [3H]thymidine. Thymidine incorporation was determined after 18 h incubation. (B) Peripheral T cells from spleen were stimulated for 12 h with anti-CD3 antibody. Expression of the IL-2R § chain (CD25) was determined by FACS (left panel). After 2 days stimulation with anti-CD3 antibody, peripheral T cells were restimulated for 5 h with Con A in the presence of Brefeldin A. Intracellular levels of IFN- + were determined by FACS (right panel).

was fully functional in T cells from Creb1lckCreAtf1–/– mice, and CD95 protein expression was unchanged (Fig. 5c). Peripheral T cells devoid of CREB and ATF1 did not express less Bcl-2 or more CD95 than cells from wildtype littermate control animals (Fig. 5c), consistent with the data demonstrating unaltered AICD.

2.5 Thymic reconstitution in sublethally irradiated mice To further investigate the decreased thymic cellularity in Creb1lckCreAtf1–/– mice and control mice, we subjected mice to a sublethal dose of + -radiation. Reconstitution of thymic populations by either hematopoietic stem cells or surviving precursor cells in the thymus was seen after 10 days (Fig. 6). Creb1lckCreAtf1–/– mice clearly displayed a slower repopulation, affecting all thymocyte subpopulations, as assayed by staining for CD4, CD8, and CD25 and CD44 for the CD4–CD8– subset (data not shown), further supporting the initial finding of reduced thymic cellularity in mice deficient for CREB and ATF1. A significant delay in thymic reconstitution was only seen in © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

CREB and thymic cellularity

1965

Fig. 5. Normal T cell survival in mice lacking CREB and ATF1. (a) Thymocytes were cultured in the presence (+Dex) or absence (–Dex) of 10–8 M Dex. Apoptotic cell death was assayed at the indicated time points. One out of two experiments is shown, with three animals per genotype for each point. (b) Thymocytes from control and Creb1lckCreAtf1–/– mice were cultured for 18 h in the presence or absence of anti-CD3 antibody, 10–8 M Dex or 200 ng/ml CD95L. Three animals per genotype were analyzed. (c) Splenic T cells from control and Creb1lckCreAtf1–/– mice were stimulated with Con A for 5 days. Cell surface expression of CD95L determined by FACS staining (graph) after 6 h treatment with antiCD3 antibody in the presence or absence of 1 mM cAMP or 10 ? M forskolin. CD95 and Bcl-2 were also analyzed by immunoblot. Protein determination was performed using the Bio-Rad protein assay to normalize protein content.

Creb1lckCreAtf1–/– mice; all other genotypes tested, including Creb1lckCre and Atf1–/–, showed no significant delay (data not shown).

2.6 CREB target gene expression RNase protection assays revealed no differences between control T cells and single- or double-mutant T cells with respect to expression of AP-1 family members and in particular c-fos, a previously reported CREB target in T cells, under either basal (unstimulated) or stimuwww.eji.de

1966

S. Baumann et al.

Eur. J. Immunol. 2004. 34: 1961–1971

Fig. 6. Defective thymic cellular reconstitution following sublethal irradiation of mice. Control and Creb1lckCreAtf1–/– mice were sublethally irradiated (6.5 Gy). At the indicated time points, three mice per genotype were killed and thymocyte numbers determined.

lated conditions (Fig. 7). The calculated signal intensity for each factor was normalized/corrected against the housekeeping genes L32 and gapdh and indicated no differences between control and mutant T cell expression (data not shown). Other AP-1 members tested in this assay were c-jun, jun-B, jun-D, fos-B, fra-1 and fra-2 (Fig. 7a). Importantly, upon stimulation, mutant T cells exhibited robust up-regulation of AP-1 factors as well as IL-2, IL-4 and IFN- + (Fig. 7b).

3 Discussion We previously reported that mice lacking CREB show impairment in the double-positive to single-positive transition [6]. Since these mice die perinatally, fetal thymic organ culture was used to show that Creb1–/– thymi can overcome this developmental impairment and reconstitute the single-positive populations, suggesting that Creb1–/– mice show a delay rather than a block in T cell development (data not shown). To study the effect of CREB loss on T cell function in adult mice, we made use of the lck promoter/enhancer to express Cre in T cells and delete a “floxed” Creb1 gene. As mentioned in the Introduction, this approach avoids a number of previously documented complicating issues that are inherent in transgenic approaches and confound the interpretation of T cell-specific CREB functions. Considerable attention was paid to validating the specificity and efficiency of Creb1 recombination in Creb1lckCre mice. Specific gene recombination closely matched the extent of protein loss in the thymus, spleen and lymph nodes (Fig. 1). Thymocytes showed essentially complete Creb1 recombination, while spleen and lymph node showed lower recombination, consistent with the presence of a higher proportion of non-T cells in these organs. Liver and tail showed no recombination, as expected. This observation reflected the finding that © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Fig. 7. Expression of putative CREB target genes are unchanged. (a) RNase protection to detect the expression of AP-1 factors was performed using 10 ? g total RNA from unstimulated or stimulated (10 ng/ml TPA and 0.5 ? M ionomycin) T cells. The signal intensity for each factor was normalized/corrected against the housekeeping genes L32 and gapdh, indicating no differences between control and mutant T cell expression. (b) RT-PCR amplicons for IL-2, IL-4 and IFN- + were detected by ethidium bromide staining after resolution through a 1.5% agarose gel. g -actin was used as an internal control. Four genotypes were tested: control mice Creb1lox/lox (lanes 1, 5), Atf1–/– (lanes 2, 6), Creb1lckCre (lanes 3, 7) and Creb1lckCreAtf1–/– (lanes 4, 8). Unstimulated and stimulated T cells are shown in lanes 1–4 and 5–8, respectively.

essentially no CREB protein was present in T cell lysates from Creb1lckCre mice compared with controls (Fig. 1c). Immunostaining with a CREB-specific antibody recognizing an epitope upstream of the region encoded by the deleted exon ensured there was no truncated CREB protein remaining in mutant T cells (Fig. 2). Although the precise time point of lck promoter-directed Cre expression www.eji.de

Eur. J. Immunol. 2004. 34: 1961–1971

in the lck-Cre mouse we and others [28–30] used has not been determined, we expect it to follow endogenous lck expression during T cell development in the thymus during the CD4–, CD8–, CD44+, CD25+ stage [31]. Genomic damage due to the expression of Cre [32, 33] appears not to be a contributing factor in the reduced thymocyte numbers described here, as the Cre expressed in lck-Cre mouse T cells does not result in significant chromosomal aberrations [34]. Moreover, we noted a modest reduction in thymocyte numbers in Atf1–/– mice that lack the lckCre transgene (Fig. 1). We further analyzed the proportion of Cre-positive cells and showed that this was at least 79% in the thymus (Fig. 2 and Table 1). Although there did not appear to be an up-regulation of ATF1 protein in Creb1lckCre T cells (Fig. 1c), we chose to generate Creb1lckCreAtf1–/– mice to bypass potential compensatory effects of the steadystate ATF1 levels. Further analysis also showed that CREM activator isoform mRNA was expressed in neither wild-type nor Creb1lckCreAtf1–/– T cells (Fig. 1d), in agreement with previous reports [24]. Thymocyte development in mice lacking either or both T cell CREB and ATF1 was similar to control Creb1loxP/loxP mice, as indicated by the cell surface expression of receptors characteristic of thymocytes at various stages. CD4+ T helper cell to CD8+ cytotoxic T cell distribution in the peripheral lymphoid organs (spleen and mesenteric lymph nodes) as well as in the thymus was normal. To further exclude minor effects in early thymocyte development, we tested the very early CD4–CD8– subpopulation for defects. The transition from CD44+CD25– to CD44+CD25+ and further to CD44–CD25+ finally the CD44–CD25– stage of differentiation was unaffected, as was the expression of various thymocyte markers, e.g. CD69, indicating successful positive selection (Fig. 3B, C and data not shown). Although thymocyte development was not affected, total thymocyte numbers were significantly reduced in Creb1lckCreAtf1–/– mice, indicating that the lack of both CREB and ATF1 affects thymocyte homeostasis (Fig. 3A). Single-mutant Creb1lckCre and Atf1–/– mice also displayed reduced thymocyte numbers, although this difference was not statistically significant. This suggests that CREB and ATF1 are involved in thymocyte expansion but not in the regulation of specific maturation steps during thymocyte selection. Based upon studies employing dnCREB [24], an important role for CREB in regulation of the proliferative capacity of T cells has been proposed, supporting the importance of the identification of CRE found in the regulatory regions of genes relevant to T cell proliferation, © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

CREB and thymic cellularity

1967

e.g. IL-2 [22] and IFN- + [21]. However, the lack of either or both CREB and ATF1 did not impair proliferative capacity or lymphokine production in vitro in response to T cell activation. T cells from control and mutant mice were activated similarly by anti-CD3 antibody, as indicated by similar rates of cell proliferation, IFN- + production and IL-2 receptor up-regulation (Fig. 4). CREB, CREM and ATF1 have all been shown to play central roles in cell survival in a number of cell types [7–10] and in vivo [4, 11]. Our studies have also shown that CREB, CREM and ATF1 can functionally compensate for one another in vivo [4, 11]. Furthermore, the study by Barton et al. [24] showed that dnCREB overexpression results in increased T cell apoptosis. There was no difference in T cell survival or induced T cell death between Cre-expressing/Creb1 recombined cells from Creb1lckCreAtf1–/– and control mice, suggesting that CREB and ATF1 are dispensable for T cell survival (Fig. 5). Creb1lckCreAtf1–/– T cells exhibited no change in Bcl-2 or CD95 protein levels (Fig. 5c), consistent with the normal survival/apoptosis parameters. Previous T cell dnCREB studies [24] may have been able to reveal the apoptosis phenotype because the dnCREB interfered with factors other than CREB, CREM and ATF1 [35], for example factors of the AP-1 family [27]. Interestingly, aberrant overexpression of CREB has also been shown to induce apoptosis [36], making it difficult, in our opinion, to attribute survival phenotypes in transgenic CREB overexpression models solely to specific CREB interference. Indeed, dominant-negative transcription factors may block transcription factor binding sites of nonphysiological target genes by engaging binding sites on promoters. Pre-activated peripheral T cells are sensitive to TCR/CD3 restimulation-induced cell death (AICD), resembling the dampening phase of an immune response [37]. We can now exclude CREB family members as mediators of this anti-apoptotic effect, as cAMP and the adenylate cyclase activator forskolin are still capable of interfering with activation-induced CD95L expression in peripheral T cells deficient in CREB and ATF1. Consistent with unaltered Bcl-2 and CD95 expression, these data demonstrate that T cell apoptosis induced by different stimuli and modulation of AICD by cAMP do not require CREB function. To further examine the consequences of the reduced thymic cellularity in Creb1lckCreAtf1–/– T cells, mice were subjected to a sublethal dose of radiation, leading to thymocyte depletion. Rapid reconstitution to complete cellularity was apparent between days 5 and 10 after radiation treatment in control mice, while Creb1lckCreAtf1–/– mice displayed slower reconstitution (more than www.eji.de

1968

S. Baumann et al.

10 days) and fewer cells at all time points, including the zenith at which point cellular homeostasis was reached (Fig. 6). Significantly, delayed thymic reconstitution was only observed in the compound Creb1lckCreAtf1–/– mutant mice, while Creb1lckCre or Atf1–/– mice did not show a delay (data not shown), implying that the effects observed were T cell-specific defects due to combined CREB and ATF1 loss. These effects cannot be due to loss of ATF1 in stromal cells, an important consideration as stromal support is critical for normal thymic function. Moreover, this experiment also shows that both ATF1 and CREB can contribute to T cell recovery and thymic reconstitution following sublethal irradiation, consistent with the well-documented functional redundancy within this transcription factor family. The mechanism via which post-irradiation recovery is perturbed is unclear. CREB can promote cell proliferation via many different pathways depending on the cell type, but unlike previous dominant-negative CREB studies in T cells, with the loss of CREB function in T cells, we do not see attenuation of IL-2- mediated proliferation or a deficit in T cell survival. Putative CREB target gene expression was unaltered in Creb1lckCreAtf1–/– thymocytes compared with control thymocytes. We observed no differences between control and mutant T cell mRNA levels for c-fos or other AP-1 members, IL-2, IL-4 or IFN- + under either basal or stimulated conditions (Fig. 7). Indeed, mutant T cells exhibited robust, control-level simulation of these genes. Interestingly, the loss of other important transcription factors such as c-fos [38] and c-jun [39] does not appear to affect T cell function in vivo. The common theme linking these studies is that the genes targeted are members of larger functional molecular families whereby loss of one or two factors can be overcome by other members. Based on our previous mouse models where CREB and CREM are critical for neuronal survival [11] and CREB and ATF1 are necessary for early mouse development and survival [4], T cells seem to be able to function relatively normally without CREB and ATF1. Given that T cells do not express CREM activator isoforms, it is possible that CRE in T cell target gene promoters are activated by other factors, such as AP-1 factors or unknown “CREB-like” factors, in the absence of CREB/CREB/ ATF1. However, this proposed compensation is insufficient to allow normal thymus cellularity under homeostatic conditions. In summary, there appears to be a growth disadvantage of CREB-ATF1 mutant T cells in vivo, without significantly affecting T cell development and function.

© 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Eur. J. Immunol. 2004. 34: 1961–1971

4 Materials and methods 4.1 Generation of mutant mice by gene targeting Creb1loxP and Atf1–/– mice were generated as described previously [4, 11]. lckCre transgenic mice were described by Orban et al. [28]. To generate Creb1lckCre mice, we crossed Creb1lckCre+/– mice with Creb1loxP/loxP mice and used resulting Creb1loxP/loxP;lckCre (denoted as Creb1lckCre) mice as the conditional T cell-specific mutant mice. In this case controls were of the genotype Creb1lox/lox. Mice were further crossed to bring the Atf1–/– allele onto the Creb1lckCre genotype, resulting in the compound mutant Creb1lox/lox;lckCre; Atf1–/– (Creb1lckCreAtf1–/–). To assess the phenotype of Atf1 mice in the context of the CREB mutation, we used the genotype Creb1lox/loxAtf1–/–. In these latter cases, control mice were of the genotype Creb1lox/loxAtf1+/–.

4.2 Assay for Creb1 recombination Genomic DNA from tissues was prepared as described previously [40]. The „floxed“ and deleted exon 10 of the Creb1 locus was amplified with the primers 5’-TATGTAAAGCAAGGGAAGATACTG and 5’-GGCATTGACACATATGCATAAAAC. Amplification conditions were as follows: 95°C, 5 min followed by 35 cycles of 95°C, 40 s; 60°C, 40 s and 72°C, 1 min.

4.3 Western blot analysis T cell lysates were resolved on SDS-PAGE. Antibodies specific for Bcl-2 (Santa Cruz, Clone N-19), CD95 (Santa Cruz, Clone A-20), CREB [4] and Cre [41] were used. The antiCREB antibody recognizes both CREB and ATF1 and is directed against the epitope encompassing amino acids 244 to 256 (NH2-AASGDVQTYQIRT-COOH). Enhanced chemiluminescence (ECL; Amersham Pharmacia Biotech) was used for detection.

4.4 RT-PCR RT-PCR analysis for detection of both activator and repressor Crem isoforms (forward primer 5’-GCCACAGGTGACATGCCAACTTAC-3’ and reverse primer 5’-AGCAAATGTCTTTCAAAGTTTCAA-3’) was performed. To detect the ICER repressor only, the ICER-specific forward primer 5’ATGGCTGTAACTGGAGATGAA-3’ was used in combination with the reverse primer above. To detect the housekeeping Gapdh, the forward primer 5’-TGACAACTCACTCAAGATTGTCAG-3’ was used in combination with the reverse primer 5’-GCTCTGGGATGACCTTGCCCACAG-3’. The following primer sets were also used: for IL-2, forward 5’ATGTACAGCATGCAGCTCGCATC-3’ and reverse 5’GGCTTGTTGAGATGATGCTTTGACA-3’; for IL-4, forward 5’-ATGGGTCTCAACCCCCAGCTAGT-3’ and reverse 5’www.eji.de

Eur. J. Immunol. 2004. 34: 1961–1971 GCTCTTTAGGCTTTCCAGGAAGTC-3’; for IFN- + , forward 5’-TGAACGCTACACACTGCATCTTGG-3’ and for g -actin, forward 5’-GTGGGCCGCTCTAGGCACCAA-3’ and reverse 5’-CTCTTTGATGTCACGCACGATTTC-3’. The amplification conditions used were 94°C, 4 min followed by 25 cycles of 94°C, 30 s; 55°C, 30 s and 72°C, 40 s.

4.5 Cytospins Thymocytes or MACS-purified CD4+ T cells from spleen or mesenteric lymph nodes were centrifuged on covered slides and fixed with 4% paraformaldehyde. Cells were stained with antibodies specific for CREB [42] and Cre [41]. The CREB antibody used for cytospins recognizes only CREB (and not ATF1; amino acids 136–150, epitope NH2KILNDLSSDAPGVPR-COOH). A secondary biotinylated antibody and streptavidin-peroxidase were used for detection.

4.6 Isolation, culture and activation of primary murine thymocytes and peripheral T cells Single-cell suspensions from thymus, spleen and mesenteric lymph nodes were prepared by homogenizing and filtering through a 40 ? m cell strainer (Falcon). Thymocytes were used directly. Cells from spleen and lymph node were further purified from erythrocytes using lysis buffer containing 0.155 M NH4Cl, 0.01 M KHCO3 and 0.1 M EDTA (pH 7.27), from monocytes and macrophages by mono-deletion for 1 h in a cell culture bottle, and from B cells by attaching them to a cell culture plate coated for 2 h at 37°C with antimouse IgG antibody (Biozol). The purity of T lymphocytes obtained by this procedure was G 90%, as determined by FACS. Peripheral T cells were stimulated with 5 ? g/ml concanavalin A (Con A, Pharmacia) for 18 h. After washing, all cells were maintained in RPMI medium containing 10% fetal calf serum (FCS), 50 ? M 2-mercaptoethanol, 2 mM glutamine, 10 mM N-2-hydroxyethylpiperazine-N’-2-ethansulfonic acid (Hepes), penicillin, streptomycin, gentamycin and 25 U/ml IL-2.

4.7 Cell analysis by flow cytometry Single-cell suspensions from thymus, spleen and lymph node (5×105 cells/ml in PBS) were incubated for 15 min on ice with a 1:100 dilution of the appropriate antibody in 50 ? l PBS. The following PE, FITC and CyChrome-conjugated monoclonal antibodies were used: RM2–5 (anti-mouse CD2, PharMingen), 145–2C11 (anti-mouse CD3, PharMingen), H129.19 (anti-mouse CD4, GIBCO/BRL), 53–7.3 (antimouse CD5, PharMingen), 53–6.7 (anti-mouse CD8, PharMingen), 7D4 (anti-mouse CD25, PharMingen), IM7 (antimouse CD44, PharMingen), H1.2F3 (anti-mouse CD69, PharMingen), Jo2 (anti-mouse CD95, PharMingen), MFL3 (anti-CD95L, PharMingen), A19–3 (isotype control for Jo2 © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

CREB and thymic cellularity

1969

and MFL3, PharMingen), H57–597 (anti-mouse TCR g , PharMingen) and XMG1.2 (anti-mouse IFN- + , PharMingen). For intracellular staining for IFN- + , cells were treated with Brefeldin A during stimulation to inhibit protein transport through the Golgi and were subsequently fixed for 20 min on ice with –20°C methanol/acetone. PBS containing 0.03% saponin (Sigma) was used for diluting antibodies and for washing. Cells were then washed twice in PBS. Three-color cytofluorometric analysis was performed on a FACScan Cytometer (Becton Dickinson) using Cell Quest Software. A minimum of 10,000 cells per sample was analyzed.

4.8 Induction and detection of apoptosis For CD95 stimulation, human CD95 ligand, trimerized via a modified leucine-zipper domain (LZ-CD95L) [43] and crossreactive with mouse CD95, was used. For anti-CD3 treatment, cell culture plates were coated with 10 ? g/ml antimouse CD3 mAb (145–2C11, PharMingen) for 3 h at 37°C or overnight at 4°C and washed with PBS. For Con A treatment, Con A was added to the culture medium at a final concentration of 5 ? g/ml. Dexamethasone was obtained from Sigma. Four different methods were used to detect apoptotic cell death. 1) Cells were resuspended in buffer containing 0.1 % (w/v) sodium citrate, 0.1% (v/v) Triton X100 and 50 ? g/ml propidium iodide (Sigma). After incubation at 4°C in the dark for at least 16 h, apoptotic nuclei were quantified by FACS. 2) Cells were treated with propidium iodide at a final concentration of 2.5 ? g/ml and directly analyzed by FACS. 3) Cells were stained with Annexin V- FITC (PharMingen). After 10 min incubation in the dark, samples were washed and immediately analyzed by FACS. 4) Cell death was quantified by FSC/SSC analysis by FACS. Specific apoptosis was calculated as (percentage of induced apoptosis – percentage of spontaneous apoptosis) / (100– percentage of spontaneous apoptosis) ×100.

4.9 Proliferation assays Spleen and lymph node cells were cultured in 96-well microtiter plates in a final volume of 200 ? l. Cells were stimulated by coating plates with 10 ? g/ml anti-mouse CD3 antibody (145–2C11, PharMingen) for 3 h at 37°C or overnight at 4°C. After 36 h stimulation, the cells were pulsed for 18 h with 1 ? Ci [3H]-thymidine per well, and thymidine incorporation was determined using a cell harvester (Wallac) and a 1205 Betaplate Liquid Scintillation Counter (Wallac). All experiments were performed in triplicate.

4.10 Purification of CD4+ cells by MACS T cells from lymph nodes were sorted using magnetic beads. Purity was controlled by FACS staining and was at least 95%.

www.eji.de

1970

S. Baumann et al.

4.11 Cell cycle analysis Thymocytes were incubated for 18 h in a buffer containing 0.1% (w/v) sodium citrate, 0.1% (v/v) Triton X-100 and 50 ? g/ml propidium iodide (Sigma) and were analyzed by FACS. 4.12 Sublethal irradiation of mice and thymic reconstitution Mice were irradiated with 6.5 Gy (Cs137). Thymus tissue was visibly absent 2 days post-irradiation. For analysis of reconstitution of thymocyte populations, the thymus was removed 2, 5, 10 or 15 days post-irradiation. Cells were counted and analyzed by FACS following staining for CD4, CD8 and TCR § g . 4.13 RNase protection assay Total RNA (10 ? g) from unstimulated or stimulated (10 ng/ml TPA and 0.5 ? M ionomycin) T cells was combined with 1 ? l [ § 32P] UTP-labeled RiboQuant AP-1 probe set (PharMingen). The hybridization (56°C overnight) and RNase digestions were performed using reagents from the Ambion RPA III kit. Yeast RNA was included as a negative control. Protected products were resolved on a 5% acrylamide/8 M urea/1× TBE gel. L32 and Gapdh signals were used as internal controls to normalize RNA loading differences. 4.14 Statistical Analysis Statistical analysis was performed using the unpaired, twotailed t-test.

Acknowledgments: We wish to acknowledge the invaluable technical assistance of Steffi Ridder and Annette Klewe-Nebenius and helpful discussions with Drs. Francois Tronche and Mark Smyth. This work was supported by the “Deutsche Forschungsgemeinschaft” through SFB 405, SFB 488, FOR 302, GRK 791/1.02, GRK 484 and Sachbeihilfe Schu 51/7–2, by the “Fonds der Chemischen Industrie”, the European Community through grant QLG1-CT2001–01574, the BMBF through NGFN grant FZK 01GS011117, the Hermann von Helmholtz-Gemeinschaft Deutscher Forschungszentren (HGF) through the “Strategiefonds DNA-CHIPS”, the Alexander von Humboldt-Stiftung through the Max-Planck-Forschungspreis für Internationale Kooperation 1998 and the Volkswagen-Stiftung through grant I/76 234.

References 1 Mayr, B. and Montminy, M., Transcriptional regulation by the phosphorylation-dependent factor CREB. Nat. Rev. Mol. Cell. Biol. 2001. 2: 599–609.

© 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Eur. J. Immunol. 2004. 34: 1961–1971 2 Ellis, M. J., Lindon, A. C., Flint, K. J., Jones, N. C. and Goodbourn, S., Activating transcription factor-1 is a specific antagonist of the cyclic adenosine 3’.5’-monophosphate (cAMP) response element-binding protein-1-mediated response to cAMP. Mol. Endocrinol. 1995. 9: 255–265. 3 Loriaux, M. M., Brennan, R. G. and Goodman, R. H., Modulatory function of CREB.CREM alpha heterodimers depends upon CREM alpha phosphorylation. J. Biol. Chem. 1994. 269: 28839–28843. 4 Bleckmann, S. C., Blendy, J. A., Rudolph, D., Monaghan, A. P., Schmid, W. and Schutz, G., Activating transcription factor 1 and CREB are important for cell survival during early mouse development. Mol. Cell. Biol. 2002. 22: 1919–1925. 5 Blendy, J. A., Kaestner, K. H., Weinbauer, G. F., Nieschlag, E. and Schutz, G., Severe impairment of spermatogenesis in mice lacking the CREM gene. Nature 1996. 380: 162–165. 6 Rudolph, D., Tafuri, A., Gass, P., Hammerling, G. J., Arnold, B. and Schutz, G., Impaired fetal T cell development and perinatal lethality in mice lacking the cAMP response element binding protein. Proc. Natl. Acad. Sci. USA 1998. 95: 4481–4486. 7 Wilson, B. E., Mochon, E. and Boxer, L. M., Induction of bcl-2 expression by phosphorylated CREB proteins during B-cell activation and rescue from apoptosis. Mol. Cell. Biol. 1996. 16: 5546–5556. 8 Riccio, A., Ahn, S., Davenport, C. M., Blendy, J. A. and Ginty, D. D., Mediation by a CREB family transcription factor of NGFdependent survival of sympathetic neurons. Science 1999. 286: 2358–2361. 9 Somers, J. P., DeLoia, J. A. and Zeleznik, A. J., Adenovirusdirected expression of a nonphosphorylatable mutant of CREB (cAMP response element-binding protein) adversely affects the survival, but not the differentiation, of rat granulosa cells. Mol. Endocrinol. 1999. 13: 1364–1372. 10 Jean, D., Harbison, M., McConkey, D. J., Ronai, Z. and BarEli, M., CREB and its associated proteins act as survival factors for human melanoma cells. J. Biol. Chem. 1998. 273: 24884–24890. 11 Mantamadiotis, T., Lemberger, T., Bleckmann, S. C., Kern, H., Kretz, O., Martin Villalba, A., Tronche, F., Kellendonk, C., Gau, D., Kapfhammer, J., Otto, C., Schmid, W. and Schutz, G., Disruption of CREB function in brain leads to neurodegeneration. Nat. Genet. 2002. 31: 47–54. 12 Elliott, M. R., Tolnay, M., Tsokos, G. C. and Kammer, G. M., Protein kinase A regulatory subunit type II beta directly interacts with and suppresses CREB transcriptional activity in activated T cells. J. Immunol. 2003. 171: 3636–3644. 13 Solomou, E. E., Juang, Y. T., Gourley, M. F., Kammer, G. M. and Tsokos, G. C., Molecular basis of deficient IL-2 production in T cells from patients with systemic lupus erythematosus. J. Immunol. 2001. 166: 4216–4222. 14 Rekvig, O. P., Moens, U., Sundsfjord, A., Bredholt, G., Osei, A., Haaheim, H., Traavik, T., Arnesen, E. and Haga, H. J., Experimental expression in mice and spontaneous expression in human SLE of polyomavirus T-antigen. A molecular basis for induction of antibodies to DNA and eukaryotic transcription factors. J. Clin. Invest. 1997. 99: 2045–2054. 15 Wollberg, P., Soderqvist, H. and Nelson, B. D., Mitogen activation of human peripheral T lymphocytes induces the formation of new cyclic AMP response element-binding protein nuclear complexes. J. Biol. Chem. 1994. 269: 19719–19724. 16 Conkright, M. D., Guzman, E., Flechner, L., Su, A. I., Hogenesch, J. B. and Montminy, M., Genome-wide analysis of CREB target genes reveals a core promoter requirement for cAMP responsiveness. Mol. Cell 2003. 11: 1417.

www.eji.de

Eur. J. Immunol. 2004. 34: 1961–1971

CREB and thymic cellularity

1971

17 Mayall, T. P., Sheridan, P. L., Montminy, M. R. and Jones, K. A., Distinct roles for P-CREB and LEF-1 in TCR alpha enhancer assembly and activation on chromatin templates in vitro. Genes Dev. 1997. 11: 887–899.

32 Loonstra, A., Vooijs, M., Beverloo, H. B., Allak, B. A., van Drunen, E., Kanaar, R., Berns, A. and Jonkers, J., Growth inhibition and DNA damage induced by Cre recombinase in mammalian cells. Proc. Natl. Acad. Sci. USA 2001. 98: 9209–9214.

18 Flamand, L., Romerio, F., Reitz, M. S. and Gallo, R. C., CD4 promoter transactivation by human herpesvirus 6. J. Virol. 1998. 72: 8797–8805.

33 Silver, D. P. and Livingston, D. M., Self-excising retroviral vectors encoding the Cre recombinase overcome Cre-mediated cellular toxicity. Mol. Cell 2001. 8: 233–243.

19 Gao, M. H. and Kavathas, P. B., Functional importance of the cyclic AMP response element-like decamer motif in the CD8 alpha promoter. J. Immunol. 1993. 150: 4376–4385.

34 Cheung, A. M., Hande, M. P., Jalali, F., Tsao, M. S., Skinnider, B., Hirao, A., McPherson, J. P., Karaskova, J., Suzuki, A., Wakeham, A., You-Ten, A., Elia, A., Squire, J., Bristow, R., Hakem, R. and Mak, T. W., Loss of Brca2 and p53 synergistically promotes genomic instability and deregulation of T cell apoptosis. Cancer Res. 2002. 62: 6194–6204.

20 Moreno, C. S., Beresford, G. W., Louis-Plence, P., Morris, A. C. and Boss, J. M., CREB regulates MHC class II expression in a CIITA-dependent manner. Immunity 1999. 10: 143–151. 21 Penix, L. A., Sweetser, M. T., Weaver, W. M., Hoeffler, J. P., Kerppola, T. K. and Wilson, C. B., The proximal regulatory element of the interferon-gamma promoter mediates selective expression in T cells. J. Biol. Chem. 1996. 271: 31964–31972. 22 Powell, J. D., Lerner, C. G., Ewoldt, G. R. and Schwartz, R. H., The –180 site of the IL-2 promoter is the target of CREB/CREM binding in T cell anergy. J. Immunol. 1999. 163: 6631–6639. 23 Hipskind, R. A. and Nordheim, A., Functional dissection in vitro of the human c-fos promoter. J. Biol. Chem. 1991. 266: 19583–19592. 24 Barton, K., Muthusamy, N., Chanyangam, M., Fischer, C., Clendenin, C. and Leiden, J. M., Defective thymocyte proliferation and IL-2 production in transgenic mice expressing a dominant-negative form of CREB. Nature 1996. 379: 81–85. 25 Zhang, F., Rincon, M., Flavell, R. A. and Aune, T. M., Defective Th function induced by a dominant-negative cAMP response element binding protein mutation is reversed by Bcl-2. J. Immunol. 2000. 165: 1762–1770. 26 Hai, T. and Curran, T., Cross-family dimerization of transcription factors Fos/Jun and ATF/CREB alters DNA binding specificity. Proc. Natl. Acad. Sci. USA 1991. 88: 3720–3724. 27 Rutberg, S. E., Adams, T. L., Olive, M., Alexander, N., Vinson, C. and Yuspa, S. H., CRE DNA binding proteins bind to the AP-1 target sequence and suppress AP-1 transcriptional activity in mouse keratinocytes. Oncogene 1999. 18: 1569–1579. 28 Orban, P. C., Chui, D. and Marth, J. D., Tissue- and site-specific DNA recombination in transgenic mice. Proc. Natl. Acad. Sci. USA 1992. 89: 6861–6865. 29 Gu, H., Marth, J. D., Orban, P. C., Mossmann, H. and Rajewsky, K., Deletion of a DNA polymerase beta gene segment in T cells using cell type-specific gene targeting. Science 1994. 265: 103–106. 30 Chong, M. M., Cornish, A. L., Darwiche, R., Stanley, E. G., Purton, J. F., Godfrey, D. I., Hilton, D. J., Starr, R., Alexander, W. S. and Kay, T. W., Suppressor of cytokine signaling-1 is a critical regulator of interleukin-7-dependent CD8+ T cell differentiation. Immunity 2003. 18: 475–487. 31 Buckland, J., Pennington, D. J., Bruno, L. and Owen, M. J., Co-ordination of the expression of the protein tyrosine kinase p56(lck) with the pre-T cell receptor during thymocyte development. Eur. J. Immunol. 2000. 30: 8–18.

© 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

35 Newman, J. R. and Keating, A. E., Comprehensive identification of human bZIP interactions with coiled-coil arrays. Science 2003. 300: 2097–2101. 36 Saeki, K., Yuo, A., Suzuki, E., Yazaki, Y. and Takaku, F., Aberrant expression of cAMP-response-element-binding protein (‘CREB’) induces apoptosis. Biochem. J. 1999. 343: 249–255. 37 Baumann, S., Krueger, A., Kirchhoff, S. and Krammer, P. H., Regulation of T cell apoptosis during the immune response. Curr. Mol. Med. 2002. 2: 257–272. 38 Jain, J., Nalefski, E. A., McCaffrey, P. G., Johnson, R. S., Spiegelman, B. M., Papaioannou, V. and Rao, A., Normal peripheral T cell function in c-Fos-deficient mice. Mol. Cell. Biol. 1994. 14: 1566–1574. 39 Chen, J., Stewart, V., Spyrou, G., Hilberg, F., Wagner, E. F. and Alt, F. W., Generation of normal T and B lymphocytes by c-jun deficient embryonic stem cells. Immunity 1994. 1: 65–72. 40 Mantamadiotis, T., Taraviras, S., Tronche, F. and Schutz, G., PCR-based strategy for genotyping mice and ES cells harboring loxP sites. Biotechniques 1998. 25: 968–970, 972. 41 Kellendonk, C., Tronche, F., Casanova, E., Anlag, K., Opherk, C. and Schutz, G., Inducible site-specific recombination in the brain. J. Mol. Biol. 1999. 285: 175–182. 42 Herdegen, T., Gass, P., Brecht, S., Neiss, W. F. and Schmid, W., The transcription factor CREB is not phosphorylated at serine 133 in axotomized neurons: implications for the expression of AP-1 proteins. Brain Res. Mol. Brain Res. 1994. 26: 259–270. 43 Walczak, H., Miller, R. E., Ariail, K., Gliniak, B., Griffith, T. S., Kubin, M., Chin, W., Jones, J., Woodward, A., Le, T., Smith, C., Smolak, P., Goodwin, R. G., Rauch, C. T., Schuh, J. C. and Lynch, D. H., Tumoricidal activity of tumor necrosis factor-related apoptosis-inducing ligand in vivo. Nat. Med. 1999. 5: 157–163.

Correspondence: Theo Mantamadiotis, Differentiation and Transcription Laboratory, Locked Bag 1, A’Beckett St., Melbourne 8006, Victoria, Australia Fax: +61-3-9656-1411 e-mail: theo.mantamadiotis — petermac.org

www.eji.de