forward light scatter
immature single positive
The transcription factor GATA3 is essential at multiple stages of T cell development, including the earliest double-negative stages, β-selection and CD4 single-positive thymocytes. Here, we show that in CD2-GATA3 transgenic mice, with enforced GATA3 expression driven by the CD2 promoter, thymocytes have reduced levels of CD5, which is a negative regulator of TCR signaling participating in TCR repertoire fine-tuning. Reduction of CD5 expression was most prominent in CD4+CD8+ double-positive (DP) cells and was associated with increased levels of the transcription factor E2A. Conversely, GATA3-deficient DP thymocytes showed consistently higher CD5 levels and defective TCR up-regulation during their development towards the CD4loCD8lo subpopulation. CD2-GATA3 transgenic mice carrying the MHC class II-restricted TCR DO11.10 also manifested decreased CD5 levels. As in these TCR-transgenic mice reduced CD5 expression cannot result from an effect of GATA3 on repertoire selection, we conclude that enforced GATA3 interferes with the developmentally regulated increase of CD5 levels. Enforced GATA3 expression in DO11.10 transgenic mice was also accompanied by enhanced TCR expression during CD4 positive selection. Because GATA3 is induced by TCR signaling in DP thymocytes, our findings indicate that GATA3 establishes a positive feedback loop that increases TCR surface expression in developing CD4 lineage cells.
Both CD4 and CD8 single-positive (SP) thymocytes differentiate from a common precursor pool of CD4/CD8 double-positive (DP) thymocytes through a process termed positive selection 1, 2. One of the transcription factors indispensable for CD4 lineage development is the zinc-finger transcriptional regulator GATA3, which was originally identified as a protein that binds to the TCRα gene enhancer 3. GATA3 is indispensable for early T cell lineage development and also plays a crucial role in the differentiation of mature Th2 effector cells by regulating transcription and chromatin configuration of the IL-4, IL-5 and IL-13 Th2 cytokine genes (reviewed in 4).
A role for GATA3 in CD4 thymocyte development was first implied from its expression pattern in the thymus, as identified in Gata3-LacZ knock-in reporter mice. Low lacZ expression was found in DP and CD8 SP thymocytes, whereas high lacZ expression was found in the transitional CD4hiCD8lo population and in CD4 SP cells 5, 6. Importantly, RT-PCR and Western blotting experiments demonstrated that GATA3 is up-regulated in response to TCR stimulation during positive selection of CD4 but not CD8 thymocytes 7. GATA3 is also expressed in early CD4–CD8– double-negative (DN) thymic T cell progenitors, whereby analysis of the Gata3-LacZ reporter mouse indicated that GATA3 is induced in those cells that have passed β-selection 5.
The embryonic lethality of Gata3–/– mutant mice at day 11 of gestation 8 and the failure of Gata3–/– ES cells to contribute to the T cell compartment in Rag2–/– or WT chimeric mice 5, 9 precluded the analysis of GATA3 function in murine T cell development in vivo. Conditional deletion of the Gata3 gene at the DN stage using the Cre-loxP system, whereby the Cre transgene was driven by the proximal Lck promoter, resulted in a developmental arrest at the CD25+CD44– DN3 stage, implicating GATA3 in β-selection 10. Deletion of the Gata3 gene after β-selection, using CD4-Cre transgenic mice, resulted in a profound specific deficiency of CD4 SP cells. These findings demonstrated the absolute requirement of GATA3 for survival or development of CD4-committed thymocytes in vivo. However, in the absence of GATA3, MHC class II-restricted T cells were not diverted into the CD8 lineage 10. Conversely, sustained overexpression of GATA3 in fetal thymic organ cultures favored selection of CD4 over CD8 SP cells, but did not divert MHC class I-restricted precursors into the CD4 lineage 7. On the basis of these findings it has been concluded that GATA3 is necessary for post-commitment CD4 generation, rather than for commitment to the CD4 lineage 7, 11. Nevertheless, a possibility remains that GATA3 promotes CD4 lineage choice, as it is conceivable that MHC class II-restricted CD8+ T cells or MHC class I-restricted CD4+ T cells die as they fail to undergo MHC-TCR and CD4/CD8 co-engagement required for their survival 1.
We have previously shown that enforced expression of GATA3 in transgenic mice, driven by the CD2 gene promoter and locus control region, inhibited maturation of CD8 SP cells, enhanced Th2 cell development in the periphery and induced thymic lymphoma in aging mice 6, 12. We also identified a small increase in DP cell size and TCRαβ/CD3 expression levels in CD69+ DP cells, suggesting that enforced GATA3 expression may influence the kinetics of positive selection.
In contrast to the important progress that has been made in understanding the role of GATA3 in the transcriptional regulation of the Th2 cytokine locus 4, downstream targets of GATA3 in T cell development are unknown. By analysis of CD2-GATA3 transgenic and Gata3 conditional knockout mouse models, as well as CD2-GATA3 transgenic mice that were crossed into various TCR-transgenic mice, we show in this report that GATA3 modulates the surface expression level of the glycoprotein CD5 and enhances TCR expression during CD4 T lineage development in vivo. The regulatory mechanisms that control CD5 expression are key to lymphocyte development and function, because CD5 is a negative regulator of TCR signaling during thymocyte development and therefore participates in the fine-tuning of the TCR repertoire 13–16. Interestingly, because GATA3 is induced by TCR signaling 7, our finding that GATA3 controls TCR up-regulation and CD5 down-regulation implicates GATA3 in a positive feedback loop that increases TCR surface expression during CD4 T lineage development.
Enforced GATA3 expression is associated with reduced CD5 surface expression
As a strategy to identify downstream targets of GATA3, we performed DNA micro-array analyses, comparing expression profiles of sorted WT and CD2-GATA3 transgenic DP thymocytes (J. P. van Hamburg et al., manuscript in preparation). In these analyses, enforced GATA3 expression correlated with low levels of CD5 transcripts in DP thymocytes. RT-PCR experiments confirmed that CD5 mRNA levels were consistently lower in CD2-GATA3 transgenic DP cells (Fig. 1A).
Next, we compared CD5 surface expression in WT and CD2-GATA3 transgenic thymocytes by flow cytometry (Fig. 1B, C). Early DN cells initiate low-level CD5 expression independently of TCRβ rearrangement 14. Then, CD5 expression levels are up-regulated in response to pre-TCR signaling, increase progressively as cells develop into immature single-positive (ISP) and DP cells, and further increase after TCR engagement. The expression levels on mature SP thymocytes correlate with the avidity of the TCR, whereby CD4 SP cells generally express higher CD5 levels than CD8 SP cells 14. The presence of the CD2-GATA3 transgene did not affect CD5 expression on DN cells (Fig. 1C). Detailed analysis of DN1–DN4 subfractions, as defined by surface CD25 and CD44, also did not reveal significant differences between WT and CD2-GATA3 transgenic cells (data not shown). However, at the ISP stage, and in particular at the DP stage, surface CD5 levels were consistently lower in CD2-GATA3 transgenic thymocytes, when compared with WT thymocytes (median fluorescence values were decreased by a factor ∼5 in DP cells; Fig. 1C). During the multistage positive selection process involving CD4loCD8lo and CD4hiCD8lo stages 17, up-regulation of CD5 levels occurred both in WT and in CD2-GATA3 transgenic thymocytes. Nevertheless, CD5 levels were consistently lower in CD2-GATA3 transgenic thymocytes (Fig. 1C). This decrease in CD5 expression persisted in CD4 SP thymocytes and mature CD4 T cells in blood and spleen (Fig. 1D, E). CD5 expression on CD8 SP thymocytes and mature peripheral CD8 T cells was only marginally decreased in CD2-GATA3 transgenic mice, when compared with WT controls.
Collectively, these results show that enforced expression of GATA3 results in decreased CD5 surface levels, in particular at the DP stage in the thymus and in mature CD4+ peripheral T cells.
Enforced GATA3 expression is associated with increased levels of E2A
The reduced surface expression of CD5 in CD2-GATA3 transgenic DP cells was paralleled by reduced intracellular expression levels and was associated with increased expression of the transcription factor E2A (Fig. 2A). As it has been reported that CD5 expression is negatively regulated by interaction of E2A with the CD5 regulatory promoter 18, it is very well possible that GATA3 reduces CD5 levels by up-regulation of E2A.
Because the majority of DP cells will not be selected and die “by neglect”, the finding of decreased surface CD5 levels in CD2-GATA3 transgenic ISP and DP cells suggested that GATA3 has the capacity to regulate CD5 expression at the DP stage in a TCR-MHC interaction-independent fashion. In agreement with this notion, we found that DP cells that were induced by stimulation of CD2-GATA3 transgenic Rag2–/– DN3 cells with anti-CD3 mAb in vivo expressed significantly reduced CD5 levels in mice, when compared with induced DP cells from non-transgenic Rag2–/– mice (Fig. 2B). In Rag-deficient mice, where T cell development is arrested at the DN3 stage, injecting anti-CD3 mAb has been shown to mimic pre-TCR signaling: It overcomes the developmental block, increases thymic cellularity, allows DN3 cells to differentiate into DN4 and then subsequently to the DP stage, but not to SP stages 19–21.
Taken together, these findings show that enforced expression of GATA3 is correlated with increased E2A levels and interferes with the progressive increase in CD5 expression that takes place when Rag2–/– DN3 cells differentiate into DP cells in vivo.
GATA3 deficiency is associated with defective regulation of CD5 and TCR
To further investigate the relevance of the finding of reduced CD5 expression in CD2-GATA3 transgenic mice, we analyzed the expression of CD5 during T cell development in mice with a conditional deletion in the Gata3 locus. Gata3f/f mice, harboring Gata3 alleles containing two loxP sites 22 (see Supporting Information Fig. S1), were crossed with CD4-Cre transgenic mice 23, resulting in conditional inactivation of the Gata3 gene in the T cell lineage. To verify Cre-mediated deletion of the Gata3 gene, DP, CD4 SP and CD8 SP cells were FACS-sorted, and floxed and deleted Gata3 alleles were identified by PCR assays (see Supporting Information Fig. S1). We found that in CD4-CreGata3f/f mice essentially all DP cells had undergone Cre-mediated deletion, since we were not able to amplify floxed Gata3 alleles from thymic DP, CD4 or CD8 fractions (Fig. 3A). Only, in CD4+ and CD8+ cell fractions from spleen we detected low levels of PCR products from floxed Gata3 alleles (Fig. 3A), which might be derived from rare CD4+ or CD8+ T cells that have escaped deletion 23 or alternatively from CD4+ or CD8+ dendritic cells.
Whereas thymic T cell development was found to be normal in Gata3f/+ and Gata3f/f mice, all CD4-CreGata3f/f mice analyzed showed a specific defect in the development of CD4 lineage cells, consistent with previously reported findings 10 (see Supporting Information Fig. S2). In contrast to Pai et al.10, we observed a ∼3–4-fold reduction in total thymocyte numbers in our CD4-CreGata3f/f mice. Both ISP and DP total cell numbers were diminished (see Supporting Information Fig. S2), indicating that (i) the Cre recombinase is already expressed at significant levels at the ISP stage, and (ii) that GATA3 is important for the production and/or the survival of ISP and DP cells.
We analyzed CD5 levels in the various T cell subpopulations in the GATA3 conditional knockout mice by flow cytometry. As shown in Fig. 3C, CD4-CreGata3f/f mice manifested consistently higher CD5 levels in ISP cells and all subsequent stages including CD4 and CD8 SP cells, when compared with littermate controls. Next, we investigated other positive selection parameters, including cell size [forward light scatter (FSC)], surface TCR and CD69 expression levels 17, 24–26. Remarkably, in GATA3-deficient CD4loCD8lo and CD4hiCD8lo cells up-regulation of TCR and CD69 was hampered. Those few CD4 SP cells present had low TCR and CD69 expression, whereas CD8 SP fractions had essentially normal expression of TCR and CD69 (Fig. 4), in agreement with the analyses of Pai et al.10.
Therefore, we conclude that GATA3 is involved in modulation of CD5 levels from the ISP stage onwards and in up-regulation of TCR expression during positive selection of CD4 lineage cells.
Enforced GATA3 expression does not affect CD4/CD8 lineage choice
We previously observed that enforced GATA3 expression inhibited maturation of CD8 cells, but did not appear to affect the CD4 versus CD8 lineage fate decision 6. However, in the CD2-GATA3 transgenic mice analyzed, differences in levels of endogenous GATA3 between developing CD4 and CD8 cells were maintained. Therefore, we crossed CD2-GATA3 transgenic mice with the conditionally deleted CD4-CreGata3f/f mice, to generate mice in which developing CD4 SP and CD8 SP thymocytes express the same level of GATA3, contributed exclusively by the CD2-GATA3 transgene, irrespective of their developmental choice. Quantitative RT-PCR experiments confirmed that in the CD2-GATA3 transgenic CD4-CreGata3f/f mice GATA3 expression was increased to similar levels in DP, CD4 SP and CD8 SP cells (see Supporting Information Fig. S2).
We found that the presence of the CD2-GATA3 transgene corrected the defects in CD4-CreGata3f/f mice: Total thymocyte numbers and the sizes of thymocyte subpopulations, including ISP, DP and CD4, were comparable to those in control littermates (Supporting Information Fig. S2), showing that in this respect the CD2-GATA3 transgene could functionally replace the endogenous Gata3 gene. Similar to the CD2-GATA3 transgenic mice 6, also the CD2-GATA3 transgenic CD4-CreGata3f/f mice showed defective maturation of CD8 cells, characterized by reduced numbers of CD8 SP cells (Supporting Information Fig. S2), impaired down-regulation of CD69, HSA and CD44, and reduced numbers of splenic CD8+ T cells (data not shown), similar to the phenotype described for CD2-GATA3 transgenic mice on the FVB/N background 6. In spite of this defective CD8 maturation, quantification of thymocyte subpopulations showed that in the absence of differential GATA3 expression between MHC class I- and MHC class II-mediated positive selection both CD4 and CD8 SP cell populations could develop (see Supporting Information Fig. S2). We did not find evidence for differential modulation of GATA3 activity in CD4 and CD8 SP cells by co-factors, because RT-PCR analyses of CD2-GATA3 transgenic CD4-CreGata3f/f CD4 and CD8 SP cell fractions did not reveal differences in the expression of the GATA3 co-factors friend of GATA (FOG) or repressor of GATA (ROG) 27, 28 (data not shown).
Collectively, these findings support earlier conclusions 6, 7, 11 that CD4/CD8 lineage choice is independent of GATA3 expression levels during positive selection and that GATA3 selectively controls developmental progression of committed CD4 T lineage cells.
Enforced GATA3 expression results in defective CD5 and enhanced TCR up-regulation
The finding of reduced CD5 levels in the presence of enforced GATA3 might either reflect an effect of GATA3 on repertoire selection or alternatively result from defective CD5 up-regulation. To distinguish between these possibilities, we crossed the CD2-GATA3 transgene into TCR-transgenic mice, which express TCR of only a single specificity.
The DO11.10 transgene encodes an ovalbumin peptide-specific TCRαβ that is MHC class II restricted and is recognized by the specific antibody KJ1-26 29. In the I-Ad background, DO11.10 thymocytes are positively selected towards the CD4 lineage (Fig. 5A). Consistent with the function of GATA3 as a transcription factor essential for CD4 thymocyte development, the presence of the CD2-GATA3 transgene did not affect DO11.10 CD4 lineage restriction (Fig. 5A). However, in the presence of the CD2-GATA3 transgene, total thymocyte and splenic CD4 T cell numbers were significantly reduced (Fig. 5A, B). CD2-GATA3 DO11.10 double-transgenic mice manifested a considerable reduction of thymic DN cells and, in particular, of the DP subpopulation (Fig. 5B). The reduction of the DP population in CD2-GATA3 DO11.10 double-transgenic mice was accompanied by an enhanced apoptosis susceptibility of these cells, as evidenced by their increased annexin V positivity (35 ± 4%, n = 3), compared with 8.4 ± 1.7% (n = 4) in DO11.10 single-transgenic DP cells (Fig. 5C). Nevertheless, the total numbers of positively selected CD4 SP cells in the thymus appeared unaffected (Fig. 5B).
As a next step, we compared DO11.10 single-transgenic and CD2-GATA3 DO11.10 double-transgenic thymocytes for the expression of CD5, TCR (total and KJ1-26+) and CD69. DO11.10 single-transgenic thymocytes increased their CD5 expression levels at the DN to DP transition, whereas TCR and CD69 were up-regulated at the DP to CD4 SP transition (Fig. 6). Compared with DO11.10 single-transgenic thymocytes, CD2-GATA3 DO11.10 double-transgenic thymocytes showed an altered phenotype, characterized by (i) increased FSC values and TCR expression levels at the DP stage and (ii) decreased CD5 expression, both at the DP and the CD4 SP stage. Also the splenic CD4 T cell population in the CD2-GATA3 DO11.10 double-transgenic mice showed decreased CD5 expression levels (data not shown). Interestingly, CD2-GATA3 DO11.10 double-transgenic DP cells manifested elevated TCR and CD69 levels, which were already close to those of CD4 SP cells (Fig. 6).
Taken together, these findings show that enforced expression of GATA3 results in impaired up-regulation of CD5 and enhanced TCR up-regulation during positive selection of DO11.10 transgenic CD4 cells.
Enforced GATA3 expression also affects CD5 levels during CD8 positive selection
Next, we crossed our CD2-GATA3 transgenic mice with HY TCRαβ transgenic mice on a Rag2–/– background. The MHC class I-restricted HY TCR is specific for the male-specific HY antigen peptide 30.
In the H-2b class I female background, HY-specific thymocytes were positively selected towards the CD8 lineage (Fig. 7A). Parallel to our observations in the DO11.10 CD4 positive selection model, we found that the presence of the CD2-GATA3 transgene reduced DP cellularity, but numbers of thymic CD8 SP and splenic CD8+ cells were comparable between the HY single-transgenic and CD2-GATA3 HY double-transgenic littermates (Fig. 7A, B). The CD2-GATA3 transgene did not affect CD8 lineage restriction by the HY transgene, as CD4 SP cells were not found in the CD2-GATA3 HY double-transgenic female thymus or spleen (Fig. 7A). Yet, we observed defective CD4 silencing in the CD2-GATA3 HY double-transgenic CD8 population in the spleen (Fig. 7A).
In CD2-GATA3 HY double-transgenic female DP cells, additional CD5lo and TCRhi populations were present and CD69 expression levels were slightly increased. In contrast to the CD2-GATA3 CD4 SP cells in the DO11.10 model, the CD2-GATA3 HY double-transgenic CD8 SP cells manifested normal CD5 expression levels (compare Fig. 6 and 7C). The effect of the CD2-GATA3 transgene on cell size and TCR expression in CD8 SP cells was negligible. The CD2-GATA3 HY double-transgenic CD8 SP cells failed to down-regulate CD69 (Fig. 6C). This most likely reflected an inhibitory effect of enforced GATA3 on SP cell maturation, as was substantiated by analysis of additional maturation markers including CD44, CD62L and HSA (data not shown) and paralleled our previous findings in CD2-GATA3 single-transgenic mice on the FVB background 6.
Analysis of male HY transgenic and CD2-GATA3 HY double-transgenic thymocytes by flow cytometry revealed that negative selection was preserved in the presence of enforced GATA3 expression. Hereby, a majority of thymocytes had a DN phenotype 30, irrespective of the presence of the CD2-GATA3 transgene (data not shown).
Taken together, these findings show that also in DP cells that are committed to the CD8 lineage, enforced GATA expression results in defective CD5 up-regulation and in the appearance of a small but detectable subpopulation with increased TCR expression.
The role of GATA3 in the development of the CD4 T cell lineage is largely unknown. In this report, we show that GATA3 controls CD5 and TCR expression during CD4 T cell lineage development.
Our FACS analyses in CD4-CreGata3f/f mice show that in the absence of GATA3, DP cells have slightly increased CD5 levels, which remain elevated during positive selection towards CD4 SP and CD8 SP subpopulations. Conversely, enforced GATA3 expression resulted in significantly reduced CD5 expression in DP cells, irrespective of their commitment towards the CD4 or CD8 lineage (in DO11.10 or HY TCR transgenic mice, respectively). Nevertheless, this reduction was maintained in CD4 SP cells, but not in CD8 SP cells. As in the TCR-transgenic mice reduced CD5 expression cannot result from an effect of GATA3 on repertoire selection, we conclude that enforced GATA3 interferes with the developmentally regulated increase of CD5 levels. We propose that this effect is independent of TCR-MHC interactions, as CD5 levels were also increased when DN3 cells differentiated into DP cells during β-selection induced by anti-CD3 antibodies in vivo. In CD4-CreGata3f/f mice, TCR and CD69 are not effectively up-regulated during positive selection of CD4 lineage cells. Conversely, enforced GATA3 expression was associated with increased cell size, enhanced or accelerated TCR (and CD69) expression during CD4 T cell lineage development in MHC class II-restricted DO11.10 TCR-transgenic mice. Taken together, we conclude that GATA3 is essential for appropriate TCR up-regulation and CD5 modulation, selectively in developing CD4 T lineage cells. As GATA3 expression is specifically up-regulated during development of CD4 and not CD8 lineage cells 7, our findings also imply that the molecular mechanisms of the modulation of TCR and CD5 expression levels during positive selection are different between the CD4 and the CD8 lineage.
It has been proposed that GATA3 contributes to linking TCR signal strength to the distinct differentiation programs of CD4 and CD8 thymocytes, implicating GATA3 in the CD4/CD8 lineage choice 7. Here, we substituted the endogenous Gata3 gene expression by the CD2-GATA3 transgene (which is not subjected to regulation by TCR signals during positive selection). As a result, DP thymocytes containing MHC class I- and class II-restricted TCR expressed equivalent levels of GATA3, as contributed by the CD2-GATA3 transgene only. We found that both CD4 and CD8 SP cells were generated (Supporting Information Fig. S2), ruling out that GATA3 expression levels are instrumental in the CD4/CD8 lineage choice.
Enforced expression of GATA3 was also associated with survival defects of DP cells, both in the DO11.10 (I-Ad) and HY (female) positive selection model. As targeted deletion of CD5 has been shown to reduce DP cellularity in various TCR-transgenic models, including HY and DO11.10 16, 31, it is likely that the reduction of DP cell numbers in the CD2-GATA3 transgenic models is a result of decreased CD5 expression. As CD5 acts as a negative regulator of TCR-mediated signal transduction 13, 16, reduction of CD5 expression levels in CD2-GATA3-expressing DP cells is expected to increase TCR signal strength, which will result in negative selection and thus deletion. However, it cannot be excluded that the impaired thymocyte expansion and the partial developmental arrest found in thymocytes with premature TCRαβ expression and signaling 32 may also contribute to the observed reduction in the DP cell population in TCR CD2-GATA3 double-transgenic mice.
Parallel to our observations in mice with enforced GATA-3 expression, CD5–/– DP cells are also larger and show increased levels of TCR and CD69 13, 16, 18. In the absence of CD5, DO11.10 (I-Ad) transgenic mice also show enhanced deletion of DP cells, accompanied by efficient positive selection towards CD4 SP cells 31. Conversely, the residual CD4hiCD8lo cells present in mice in which the Gata3 gene is conditionally targeted showed a CD5hiTCRloCD69lo phenotype, thereby apparently mimicking a “death by neglect” cell fate. Because in GATA3-deficient DP cells the TCR-proximal signal transduction pathways, including ZAP70, Lck and Erk, are intact 11, it is attractive to hypothesize that GATA3 is required as a downstream target of TCR signaling, whereby GATA3 function is essential for CD4 lineage development to initiate TCR up-regulation.
In this context, our finding that CD2-GATA3 DP cells have increased expression levels of the basic helix-loop-helix transcription factor E2A, which is a negative regulator of CD5 expression during thymocyte development 18, is interesting. Although by RT-PCR and DNA micro-array analyses we did not detect significant differences in E2A transcript levels between CD2-GATA3 and WT DP cells (M. de Bruijn, unpublished), it is very well possible that GATA3 acts by stabilizing E2A protein levels, as in erythrocytes E2A has been shown to be part of a large DNA-binding complex containing the GATA family member GATA1 33, and the related helix-loop-helix transcription factor Tal-1/Scl has been shown to interact directly with GATA1 and GATA3 34, 35. A role for GATA3 as a negative regulator of CD5 gene expression would also be supported by our observation in micro-array experiments that enforced GATA3 expression is associated with down-regulation of CD6 (J. P. van Hamburg et al., unpublished), which is closely linked to CD5 in the genome and has a similar pattern of expression (31 and references therein). Nevertheless, as transgenic overexpression of CD5 does not result in complete arrest of development of CD4 T lineage cells 16, it is clear that appropriate reduction of CD5 expression levels is not the only function of GATA3 during positive selection and maturation of CD4 T cells.
As GATA3 is induced by TCR signaling 7, our finding that enforced GATA3 expression in developing DO11.10 transgenic CD4 lineage cells is associated with premature TCR up-regulation (Fig. 4) implicates GATA3 as a key regulator in a positive feedback loop. Induction of GATA3 by TCR signaling will increase TCR expression and thereby enhance the induction of GATA3. Moreover, GATA3 has the ability to increase TCR signal strength in an independent parallel pathway by down-regulating CD5, leading to an efficient mode of signal amplification in this positive regulatory loop. As a result, during development of DP to CD4 SP cells the expression levels of the TCR and GATA3 will increase. It may be possible that GATA3 directly regulates TCRα and β transcription, as binding sites have been identified in these loci 3, 36. As GATA3 expression is not induced in the CD8 cell lineage 5, 7, it is clear that different nuclear factors should be responsible for the regulation of the expression levels of CD5 and TCR molecules in the CD8 lineage.
Materials and methods
CD2-GATA3 (FVB) transgenic mice have been described 6. CD2-GATA3 × DO11.10 transgenic mice 29 were backcrossed to BALB/c (H-2d/I-Ad) for nine generations. HY/Rag2–/– (C57BL/10) mice, purchased from Taconic Europe A/S (Denmark), were crossed with CD2-GATA3 transgenic mice and backcrossed until Rag2–/– and homozygous for H-2b. CD4-Cre mice 23 were a kind gift of Dr. C. Wilson (University of Washington, Seattle, USA). The generation of Gata3f/f mutant mice has been described 22. Similar to the Gata3f/f mice reported independently 10, we also failed to obtain the expected Mendelian transmission, suggesting significant lethal developmental defects originating from the floxed Gata3 alleles. Occasionally, Gata3f/f mice were very small and therefore excluded from analysis. CD4-Cre × Gata3f/f × CD2-GATA3 transgenic mice were on a mixed FVB × C57BL/6 background. Mice were bred and maintained in the Erasmus MC animal care facility under specific pathogen-free conditions and analyzed at 5–12 wk. Experimental procedures were reviewed and approved by the Erasmus University committee of animal experiments.
For mouse genotyping, tail DNA was analyzed by PCR. Primers for the determination of the presence of the CD2-GATA3 transgene were 5′-CAGCTCTGGACTCTTCCCAC-3′ and 5′-CAGCTCTGGACTCTTCCCAC-3′, and for the CD4-Cre transgene 5′-ACCCTGTTACGTATAGCCGA-3′ and 5′-CTCCGGTATTGAAACTCCAG-3′. Genotyping for DO11.10 and HY was performed according to protocols obtained from the Jackson Laboratory and Taconic Europe A/S, respectively. Myogenin (primers: 5′-TTACGTCCATCGTGGACAGC-3′ and 5′-TGGGCTGGGTGTT-AGTCTTA-3′) was used as an internal control. PCR-based typing of Gata3f/f mice and analysis of Gata3 locus deletion were as described previously 22. Mouse MHC haplotypes were confirmed by FACS analysis of peripheral blood using MHC class I-specific mAb.
Flow cytometry, mAb, and cell sorting
Preparation of single-cell suspensions, flow cytometry and mAb have been described previously 6. Anti-human E2A antibody and PE-labeled annexin V were purchased from BD Pharmingen. Anti-HY TCR (T3.70) and anti-DO11.10 TCR (KJ1-26) mAb were from eBioscience and Caltag Laboratories, respectively. FACS sorting was performed with a FACSVantage VE equipped with Diva Option and BD FACSDiva software (BD Bioscience). The purity of the obtained fractions was >99%.
In vivo injection of CD2-GATA3 (non-)transgenic Rag2–/– mice was done i.v. with 50 µg rat anti-mouse CD3 mAb (145–2C11; BD Bioscience) in 0.5 mL PBS.
RNA extraction, cDNA synthesis and RT-PCR
Total RNA was extracted from 5 × 106 sorted cells, using the GenElute mammalian total RNA miniprep system (Sigma-Aldrich). Of total RNA, 1 µg was used as a template for cDNA synthesis using Superscript II RT (Invitrogen) and random hexamer primers. cDNA was diluted serially threefold before PCR amplification. Primers for CD5 were: 5′-AAGCATCATCCTGACCCTTG-3′ and 5′-AGATCGGTGTAGGGCTCCTT-3′, and primers for β-actin were: 5′-TACCACTGGCATCGTGATGGACT-3′ and 5′-TTTCTGCATCCTGTCGGCAAT-3′. RT-PCR products were separated by standard agarose electrophoresis and visualized by ethidium bromide staining.
Quantitative real-time PCR was performed using the ABI Prism 7700 sequence detection system (Applied Biosystems, Foster City, CA) and standard thermocycling conditions. Threshold levels were set and further analysis was performed using the SDS v1.9 software (Applied Biosystems). The obtained Ct values for GATA3 were normalized by the Ct values for glycereraldehyde-3-phosphate dehydrogenase (GAPDH). Primers detecting both endogenous and transgenic GATA3 were: 5′-CATTACCACCTATCCGCCCTAT-3′ and 5′-CACACACTCCCTGCCTTCTGT-3′ with 5′-CGAGGCCCAAGGCACGATCCAG-3′ as a probe. GAPDH primers were 5′-TTCACCACCATGGAGAAGGC-3′ and 5′-GGCATGGACTGTGGTCATGA-3′ with 5′-TGCATCCTGCACCACCAACTG-3′ as a probe.
We acknowledge E. de Haas, J. van de Wees, F. Grosveld, H. Kuipers, B. Lambrecht and the people from the EDC Animal Facility (Erasmus MC Rotterdam) for assistance at various stages of the project. We also thank Dr. C. Wilson (University of Washington, Seattle, USA) for providing CD4-Cre mice. This work was supported by the Association for International Cancer Research and the Dutch Cancer Society.