Title: Chromatin condensation via the condensin II complex is required for peripheral T-cell quiescence
Abstract: Article17 December 2010free access Chromatin condensation via the condensin II complex is required for peripheral T-cell quiescence Jason S Rawlings Jason S Rawlings Department of Biochemistry, St Jude Children's Research Hospital, Memphis, TN, USAPresent address: Department of Biology, Furman University, Greenville, SC 29613, USA Search for more papers by this author Martina Gatzka Martina Gatzka Department of Immunology, Norris Cancer Center, University of Southern California, Los Angeles, CA, USA Search for more papers by this author Paul G Thomas Paul G Thomas Department of Immunology, St Jude Children's Research Hospital, Memphis, TN, USA Search for more papers by this author James N Ihle Corresponding Author James N Ihle Department of Biochemistry, St Jude Children's Research Hospital, Memphis, TN, USA Search for more papers by this author Jason S Rawlings Jason S Rawlings Department of Biochemistry, St Jude Children's Research Hospital, Memphis, TN, USAPresent address: Department of Biology, Furman University, Greenville, SC 29613, USA Search for more papers by this author Martina Gatzka Martina Gatzka Department of Immunology, Norris Cancer Center, University of Southern California, Los Angeles, CA, USA Search for more papers by this author Paul G Thomas Paul G Thomas Department of Immunology, St Jude Children's Research Hospital, Memphis, TN, USA Search for more papers by this author James N Ihle Corresponding Author James N Ihle Department of Biochemistry, St Jude Children's Research Hospital, Memphis, TN, USA Search for more papers by this author Author Information Jason S Rawlings1, Martina Gatzka2, Paul G Thomas3 and James N Ihle 1 1Department of Biochemistry, St Jude Children's Research Hospital, Memphis, TN, USA 2Department of Immunology, Norris Cancer Center, University of Southern California, Los Angeles, CA, USA 3Department of Immunology, St Jude Children's Research Hospital, Memphis, TN, USA *Corresponding author. Department of Biochemistry, St Jude Children's Research Hospital, 332 North Lauderdale, Memphis, TN 38105, USA. Tel.: +1 901 595 3422; Fax: +1 901 525 8025; E-mail: [email protected] The EMBO Journal (2011)30:263-276https://doi.org/10.1038/emboj.2010.314 There is a Have you seen? (January 2011) associated with this Article. PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Naive T cells encountering their cognate antigen become activated and acquire the ability to proliferate in response to cytokines. Stat5 is an essential component in this response. We demonstrate that Stat5 cannot access DNA in naive T cells and acquires this ability only after T-cell receptor (TCR) engagement. The transition is not associated with changes in DNA methylation or global histone modification but rather chromatin decondensation. Condensation occurs during thymocyte development and proper condensation is dependent on kleisin-β of the condensin II complex. Our findings suggest that this unique chromatin condensation, which can affect interpretations of chromatin accessibility assays, is required for proper T-cell development and maintenance of the quiescent state. This mechanism ensures that cytokine driven proliferation can only occur in the context of TCR stimulation. Introduction Quiescence, marked by the absence of cell division, defines the majority of the lifespan of a peripheral T lymphocyte. During this time, which can be considered indefinite (Sprent, 1993), the cell has very low-energy output, hindered ability to produce macromolecules, and many of its organelles (if present) are rudimentary (Jaehning et al, 1975; Morley et al, 1993; Paul, 2003; Frauwirth and Thompson, 2004). This long-lived quiescence is bookended by periods of intense proliferative activity, first as developing precursor cells in the bone marrow and thymus and finally as activated T cells participating in an immune response in the periphery. Naive peripheral T cells remain in this quiescent, G0 phase, until they encounter their T-cell receptor (TCR)-specific antigen. Following activation via the TCR, these relatively small T cells will undergo massive transformation into large lymphoblasts with the capacity to proliferate and ultimately clear the foreign antigen. Establishment and maintenance of the quiescent state is marked by drastically reduced global levels of transcription (Jaehning et al, 1975). Epigenetic phenomena have been linked to transcriptional activity and silencing. Methylation of CpG dinucleotides has been shown to silence genes, including those involved in T-lymphocyte development and function (Fitzpatrick et al, 1998; Lee et al, 2002). Modification of the amino-terminal tails of histones has also been associated with transcriptional competence (Li et al, 2007). Histone modifications have also been studied on a global scale in CD4+ and CD8+ T lymphocytes (Barski et al, 2007; Wang et al, 2008; Araki et al, 2009; Wei et al, 2009). Recently, a report has shown that global assessment of histone ‘marks’ could be used as an indicator of quiescence in B lymphocytes (Baxter et al, 2004). The two highly related Stat5 proteins (Stat5a and Stat5b; hereafter referred to as Stat5) are essential for peripheral T-cell proliferation, as evidenced by the inability of Stat5-deficient T cells to proliferate in response to growth factors (Moriggl et al, 1999b). Therefore, proliferation depends on the transcriptional activity of Stat5. In the context of a proper immune response, TCR activation leads to the production of IL-2, which in turn activates Stat5 via the canonical Jak/Stat cascade (Ihle et al, 1995; Ihle, 1996; Rawlings et al, 2004), resulting in the clonal expansion of only those T cells that are able to recognize the foreign antigen. To avoid improper proliferation during an immune response, naive T cells, for which TCR engagement has not occurred, must have a mechanism to ignore the effects of this cytokine. One level of regulation exists at the level of receptor presentation. Naive T cells possess the intermediate affinity IL-2 receptor consisting of β and γc chains. The IL-2 receptor present on activated T cells possesses an additional chain (α) whose function is to increase ligand-receptor affinity (Lin and Leonard, 1997). However, even the addition of exogenous IL-2 at concentrations that would engage the low affinity receptor fails to elicit T-cell proliferation in the absence of TCR stimulation, indicating that there must be additional mechanisms, downstream of the IL-2 receptor, which regulate signalling and ultimately escape from quiescence. We examined activation of T cells using IL-2/Stat5-target gene transcription as a model system for studying mechanisms of overcoming quiescence. As anticipated, IL-2 could induce the activation and nuclear translocation of Stat5 in naive T cells; however, Stat5 failed to access its target promoters. Changes in the global state of histone modification or altered DNA methylation at a Stat5-target gene were not detected. However, we discovered that T-cell activation results in higher-order structural changes in chromatin that correlate with TCR induction of competence to respond to IL-2. We propose a model whereby T-cell activation reconfigures higher-order chromatin in naive T cells, which permits the engagement of Stat5 with its target promoters resulting in proliferation. Finally, we show that a mutation in a subunit of the condensin II complex results in defective chromatin condensation during T-cell development and failure to silence IL-2-target genes in naive peripheral T cells. Our findings suggest that in T cells, the mechanism for establishment of quiescence in the thymus and maintenance in the periphery even in the presence of growth factors lies in the ability to regulate higher-order chromatin structure via the condensin II complex. In the context of an immune response, this mechanism ensures that only those T cells that have been activated via the TCR will proliferate. Results IL-2 induces disparate transcriptional programs in naive versus activated peripheral T cells We previously demonstrated that there is little overlap in the genes that IL-2 induces in naive T cells relative to activated T cells (Gatzka et al, 2006). Among these, known Stat5-target genes were induced in activated T cells, whereas none of the genes were induced in naive T cells (Figure 1A). The results for several genes in each set were validated by quantitative RT–PCR (Figure 1B–D) and confirmed that IL-2 induces distinct transcriptional programs in naive versus activated T cells and that known Stat5-target genes, such as Cis, are only induced in activated T cells. To explore possible differences in T-cell subpopulations, we used sorted peripheral T cells to assess the ability of IL-2 to induce Cis. Cis induction was readily detected in purified CD4+ or CD8+-activated T cells (Supplementary Figure S1). However, no induction of Cis was observed in purified, unstimulated CD8+ T cells. A weak induction of Cis was detected in purified, unstimulated CD4+ cells; however, this induction was eliminated by depleting the CD25+ subpopulation of CD4+ cells. As CD25 is a marker of T-cell activation (Minami et al, 1993), our findings suggest that the expression was associated with a subpopulation of CD4+ cells that were already activated (Supplementary Figure S1). Figure 1.IL-2 induces disparate transcriptional programs in naive versus activated T cells. (A) Venn diagram illustrating the disparate transcriptional programs induced by IL-2 in naive versus activated peripheral T cells as revealed by microarray analysis (Gatzka et al, 2006). The number of genes induced more than three-fold in each class is indicated inside the diagram and representative genes of each class are indicated outside the diagram. (B–D) Purified naive T naive cells and those activated in culture (see Materials and methods) were stimulated with 1000 U/ml IL-2 for the times indicated. Levels of gene expression were determined by quantitative RT–PCR relative to HPRT and all values were calibrated to naive unstimulated cells. Genes are sorted into those that are induced in activated cells (B), those induced in naive cells (C) or induced in both cell types (D) according to the microarray. Download figure Download PowerPoint Peripheral T cells rapidly acquire the ability to induce expression of Stat5-target genes following TCR stimulation The above results suggested that, with activation, T cells acquire the ability to respond to IL-2 and induce expression of Stat5-regulated genes. We therefore determined when, following TCR engagement, T cells acquire the ability to respond to IL-2 and induce expression of Stat5-regulated genes. Purified CD25– peripheral T cells were first primed with anti-CD3 antibodies for 0, 1, 3 or 6 h after which IL-2 was added for 1 h. We observed a significant increase in the ability of IL-2 to induce Cis gene expression after just 3 h of anti-CD3 stimulation, increasing following 6 h (Figure 2A). The induction of Cis gene expression was specifically dependent on IL-2 stimulation, as anti-CD3 treatment alone failed to activate Stat5 (Figure 2B) or elicit any significant changes in Cis expression (Figure 2A), consistent with previous findings (Moriggl et al, 1999a). Similar results were seen with another Stat5-target gene, Socs1. However, TCR stimulation failed to provide any level of competence at the Ifnγ locus. This was expected, because it takes at least one cell division before Ifnγ can be expressed, presumably due to epigenetic silencing mediated by DNA methylation (Bird et al, 1998; Gett and Hodgkin, 1998). Conversely, Bcl-X proved promiscuous in our studies, consistent with the fact that it can be induced by other transcription factors such as Stat1 (Fujio et al, 1997), Stat3 (Rubin Grandis et al, 2000), NFκB (Tsukahara et al, 1999; Chen et al, 2000) and Ets (Sevilla et al, 1999). Figure 2.TCR stimulation rapidly primes naive T cells to permit IL-2/Stat5 signalling. (A) Purified naive T cells were primed with anti-CD3 antibodies for the times indicated followed by stimulation with IL-2. Stat5-target gene expression was assessed by quantitative RT–PCR. For comparison, activated T cells were starved overnight (S) and restimulated with 1000 U/ml IL-2 for 1 h. Gene expression was normalized to HPRT and calibrated to naive unstimulated cells. *P<0.05, **P<0.005. (B) Purified naive T cells were stimulated with 1 μg/ml anti-CD3 antibodies followed by 1000 U/ml IL-2 where indicated and assessed for activation of Stat5. Download figure Download PowerPoint Stat5 fails to engage DNA in naive peripheral T cells To explore the mechanisms for the lack of induction of Stat5-target genes, we initially assessed the ability of Stat5 to function properly in naive T cells. Stat5 transcriptional activity is activated via tyrosine phosphorylation resulting from its recruitment to the IL-2 receptor complex. This modification permits Stat5 dimerization, translocation to the nucleus and binding to target gene promoters (Rawlings et al, 2004). IL-2-induced tyrosine phosphorylation of Stat5 was readily detectable in both CD4+ and CD8+ naive or activated T cells (Figure 3A) with similar kinetics and comparable IL-2 concentration requirements (Figure 3B). Furthermore, we observed Stat5 nuclear translocation following IL-2 stimulation in naive peripheral T cells (Figure 3C). Lastly, previous studies demonstrated that naive peripheral T cells have the capacity for Stat5–DNA binding (Gatzka et al, 2006). Taken together, these findings suggest that although Stat5-target genes are not transcribed, naive peripheral T cells possess all the machinery needed for Stat5 activation and function. Figure 3.Stat5 cannot engage DNA in naive peripheral T cells. (A) IL-2 stimulation results in activation of Stat5 in naive CD4+ and CD8+ cells. Cells were stimulated with 1000 U/ml IL-2 for the times indicated. (B) IL-2 dose–response analysis of Stat5 activation in naive and activated T cells. Cells were stimulated for 30 min with the dose of IL-2 indicated. Ctrl lanes are DA3 cells expressing the Epo receptor stimulated with recombinant human Epo for 15 min (Pelletier et al, 2006) serving as a positive control. (C) Immunofluorescent localization of endogenous Stat5 in purified naive T cells and those stimulated with 1000 U/ml IL-2 for 30 min. (D) ChIP analysis on the Cis promoter of naive and activated T cells either unstimulated (control) or stimulated with anti-CD3 antibodies, IL-2 or combination of anti-CD3 antibodies and IL-2 for 1 h. Antibodies recognizing either activated (pY Stat5) or total Stat5 were used. (E) Schematic of the Cis locus showing the relative positions of the Stat5-binding elements and PCR assay used in the ChIP. An arrowhead marks the transcription start site. Download figure Download PowerPoint The above results demonstrated that the inability of Stat5 to activate its target genes in naive T cells was either due to an inability to access its target promoters or the lack of assembly of a functional transcriptional complex. To address this, chromatin immunoprecipitation (ChIP) assays were used to assess the status of Stat5 at the Cis promoter. With IL-2 stimulation, or in combination with anti-CD3 antibodies, we readily observed an increase in the enrichment of Stat5 at the Cis promoter in activated T cells using either Stat5 antibodies or antibodies specific for phosphorylated (Y694) Stat5 (Figure 3D). Consistent with previous studies demonstrating the inability of TCR signalling to activate Stat5 (Figure 2B; Moriggl et al, 1999a), we did not observe any changes in enrichment of Stat5 at the Cis promoter of activated T cells stimulated with anti-CD3 alone. In contrast, stimulation of naive T cells under any of the conditions did not result in any detectable accumulation of Stat5 on the Cis promoter. These results demonstrate that the lack of induction of Cis expression is the result of the inability of Stat5 to access the promoter in naive T cells. T-cell activation results in ultrastructural changes in nuclear architecture correlating with Stat5 competence A number of possibilities existed to explain the activation-induced changes in the accessibility of the Cis promoter to activated Stat5. Initially, we examined T-cell nuclear morphology by transmission electron microscopy as they gained competence for activated Stat5 to induce gene expression (Figure 4A). As illustrated, there is a rapid, dramatic change in nuclear morphology following anti-CD3 treatment. In particular, nuclear material is highly condensed in naive T cells but becomes less condensed within hours after stimulation with anti-CD3 antibodies. The decondensation is not seen with IL-2 treatment alone (data not shown) and precedes the expansion of the cytoplasm that characterizes ‘blasting’ T cells. Anti-CD3-mediated decondensation also occurs with Stat5-deficient T cells (Teglund et al, 1998), indicating that decondensation is not Stat5 dependent (Figure 4B). Figure 4.T-cell activation induces rapid ultrastructural changes to chromatin that allow access to DNA. (A) Transmission electron microscopy (TEM) was used to visualize purified naive T cells or those stimulated with anti-CD3 antibodies for the times indicated. (B) TEM of purified naive T cells from Stat5-deficient animals or those stimulated for 20 h with anti-CD3 antibodies. (C) Purified naive T cells were left untreated or stimulated with anti-CD3 antibodies for the times indicated and subjected to a sonication assay (see Materials and methods). Arrow indicates sonication resistant chromatin. Download figure Download PowerPoint During the course of our experiments, we noted that the DNA of naive T cells is highly resistant to sonication-induced breakage and that this resistance is lost following stimulation with anti-CD3 antibodies (Figure 4C). Under the conditions described in the Materials and methods section, DNA from unstimulated naive T cells was completely resistant to breakage as indicated by the lack of significant DNA at lower molecular weights. Activation caused the DNA to become progressively more sensitive, such that we began to see lower molecular weight fragments beginning just 1 h following anti-CD3 treatment. Importantly, the kinetics of DNA fragmentation were comparable to those of the change in nuclear condensation seen by electron microscopy and both phenomena correlate with acquisition of Stat5 competence. Nuclear condensation is not a function of CpG methylation The similar kinetics of nuclear decondensation and acquisition of Stat5 competence would suggest that the inability of Stat5 to access its target promoters in naive peripheral T cells could be the result of DNA condensation and that a critical step in T-cell activation is to induce decondensation. CpG methylation has been associated with transcriptional competence (Smale, 2003) and could be hypothesized to contribute to DNA condensation (Matarazzo et al, 2007). Recently, DNA methyltransferase activity has been linked to TCR engagement (Gamper et al, 2009) and has been shown to control expression of the Th1 (Fitzpatrick et al, 1998) and Th2 loci (Lee et al, 2002), implicating a role for DNA methylation in the control of gene expression during lymphocyte development. The Cis promoter contains a large CpG island that encompasses the Stat5-binding sequences (Li and Dahiya, 2002). Bisulfite sequencing of this region in both naive and activated T cells revealed that there were no major differences in the CpG methylation status between these cell types (Supplementary Figure S2), indicating that DNA methylation is not the mechanism for chromatin condensation in naive T cells. Histones are not grossly modified as a consequence of TCR activation, rather they become more accessible Several studies have implicated a role for histone modifications in chromatin structure and changes in gene expression (Kouzarides, 2007; Li et al, 2007), including during lymphocyte development (Baxter et al, 2004; Chang and Aune, 2007; Krangel, 2007). Therefore, the changes in nuclear condensation observed could be a response to changes in histone modifications. We therefore tested the status of histone H3 using a variety of antibodies specific for modifications associated with both transcriptional activity and silencing by immunofluorescence microscopy. We reasoned that the magnitude of the changes in chromatin condensation we observed would require relatively global changes in histone modification. Remarkably, our ability to detect nuclear histones was severly impaired in naive T cells using any of the reagents we tested. However, all of the reagents were able to readily access nuclear histones following 20 h of TCR stimulation or in activated T cells proliferating in culture. The results were comparable for antibodies that detected unmodified histone H3, as well as a variety of modification-specific antibodies (Figure 5A). These findings were substantiated by flow cytometric analysis of intracellular stained peripheral T cells (Figure 5B; Supplementary Figure S3). Figure 5.Standard immunofluorescence techniques fail to detect histones in naive peripheral T cells. (A) Naive T cells were left untreated, stimulated for 20 h with anti-CD3 antibodies or growing in culture for 5 days. DAPI was used to mark the nucleus. Antibodies to mono-, di- and tri-methylated histone H3K4, tri-methylated histone H3K9/27, and acetylated histone H3 were used to detect modification of histone H3. Antibodies to N-terminal and C-terminal epitopes of unmodified histone H3 are also shown. (B) Detection of intracellular H3K4me1 as a function of TCRβ surface expression as shown by flow cytometry in unstimulated Thy1.2+CD4+ or Thy1.2+CD8+ splenocytes or those stimulated with anti-CD3 antibodies for the times indicated. Download figure Download PowerPoint The above studies suggest that histone epitopes are masked in naive T cells; therefore, we used western blot analysis to determine potential changes in histone modification. Under standard whole-cell lysis conditions, it was evident that the ability to solubilize histones also dramatically changes during activation of T cells (Figure 6A). The ability to detect histone H3 is severely impaired in cell extracts from naive T cells but becomes readily detectable following TCR stimulation. As with the immunofluorescence, this was observed with antibodies that recognize unmodified H3 as well as antibodies that recognize various modified forms of histone H3. Importantly, the extraction kinetics are virtually identical to the nuclear decondensation observed by electron microscopy, the sensitivity of the DNA to sonication, the ability to stain histones for immunofluorescence and TCR-mediated Stat5 transcriptional competence. Figure 6.Detection of histones in peripheral T lymphocytes by western blot. (A) Purified naive T cells were left untreated or stimulated with anti-CD3 antibodies for 1, 3, 6, 12 or 24 h. Half of each sample was either acid extracted overnight or lysed using NP40 buffer (see Materials and methods). Modified histone H3 (mono-, di- or tri-methylated H3K4, tri-methylated H3K9/27, acetylated H3) or unmodified histone H3 or the H3.3 variant were detected by western blot. (B) Purified naive T cells were left untreated or stimulated for 24 h with anti-CD3 antibodies. The cells were lysed with NP40 buffer (lysate) and the remaining material (pellet) was acid extracted overnight. Both were resolved by SDS–PAGE and detection of H3K4me2, total histone H3, Stat5 and Actin was assayed by western blot. (C) Cells were treated as in (A) and antibodies recognizing histones H1, H2A, H2B and H4 were used to assess the status of the remainder of the nucleosome. Download figure Download PowerPoint Due to these differences in solubility, we used the more stringent method of acid extraction to assess the global status of histone modification during T-cell activation (Figure 6A). The importance of this is illustrated in Figure 6B in which naive or activated T cells were first lysed under normal conditions and the soluble fraction separated from the pellet and the pellet further solubilized by acid extraction. In naive T cells, the majority of histone H3 resided in the acid soluble fraction (pellet), whereas in activated cells, histone H3 was readily detected in both fractions. We therefore used acid extraction to explore the potential changes in histone modifications following activation. Surprisingly, there were no detectable changes in the extent of any of the modifications tested (Figure 6A). These findings were also extended to the rest of the nucleosome, as we obtained similar results when using antibodies that recognize histone H1, H2A, H2B and H4 (Figure 6C). Our data indicate that there is no major net change in global histone modification as a function of activation, rather a change in accessibility. While unlikely, it is possible that T-cell activation results in rapid increased synthesis of ‘off chromatin’ histones that are solubilized via the whole-cell lysis technique rather than a change in solubility of ‘on chromatin’ histones. To explore this possibility, naive T cells were activated in the presence of cycloheximide and histone accessibility was measured via western blot (Supplementary Figure S4). Cycloheximide had no effect on the kinetics of detection of histones following activation, indicating that these are changes in ‘on chromatin’ accessibility. Chromatin condenses rapidly during thymocyte development Immature CD4–CD8– double negative (DN) thymocyte precursors enter the thymus as relatively large, highly proliferative cells. During development in the thymus, these cells become smaller and ultimately become quiescent prior to exit into the periphery as naive T cells, suggesting that chromatin condensation occurs during thymocyte development. As in the periphery, it has been suggested that epigenetic modification contributes to the silencing of loci important during thymocyte maturation ex vivo (Su et al, 2004). Therefore, we wanted to determine when chromatin condensation occurs during development in the thymus. Similar to naive peripheral T cells, we had difficulty in detecting histones in the smaller, more mature CD4+CD8+ double-positive (DP) and single-positive (SP) thymocytes, whereas histones were readily detectible in the larger, immature (DN) thymocytes (Figure 7A). Flow cytometric analysis using cell surface markers to delineate thymocyte subpopulations confirmed that chromatin condensation occurs as cells transition from the DN to the DP stage (Figure 7B). Figure 7.Chromatin condenses rapidly during thymocyte development. (A) Thymocytes were isolated and stained with DAPI and with antibodies to mono-, di- and tri-methylated histone H3K4, tri-methylated histone H3K9/27 and acetylated histone H3 to detect modification of histone H3. Antibodies to N-terminal and C-terminal epitopes of unmodified histone H3 are also shown. (B) Flow cytometric analysis of chromatin condensation during thymocyte development using PE- conjugated H3K4me1 antibodies as a marker for chromatin accessibility. Subpopulations of Thy1.2+ thymocytes were delineated based on surface expression of CD4 and CD8. Download figure Download PowerPoint Chromatin condensation is required for proper development of peripheral T cells and depends on the function of the condensin II complex Given that chromatin condenses as thymocytes transition from the DN to DP stage, one could hypothesize that a failure to condense might have profound effects on T-cell selection and consequently proper differentiation. We searched the literature for mutations that resulted in thymocyte maturation defects as well as mutations in genes that might be involved in chromatin condensation. Remarkably, mice harbouring a point mutation of kleisin-β, a subunit of the condensin II complex, resulted in such a phenotype (Gosling et al, 2007, 2008). Mice homozygous for the nessy allele of kleisin-β have relatively normal numbers of DN thymocytes; however, beginning at the DP stage they have drastically reduced cell numbers, presumably due to increased apoptosis during negative selection (Gosling et al, 2007). We analysed chromatin condensation during T-cell differentiation in these mice and found that they were defective in their ability to condense chromatin in the DN/DP transition (Figure 8A and C), suggesting that failure to properly condense chromatin could contribute to the increased apoptosis and resulting dramatic reduction in cell number seen at the DP stage in these animals. Consistent with these findings, the CD4+ and CD8+ SP mature thymocytes also had a more open chromatin configuration (Figure 8A and C) and drastically reduced cell number compared with their wild-type littermates (Gosling et al, 2007). Developing thymocytes must also undergo β-selection, which occurs at the DN2–DN3 transition. As with negative selection, the outcome of this process is dependent on proper response to TCR engagement. While wild-type cells undergo a dramatic condensation event as the cells transition from DN2 to DN3, nessy cells appear to lag behind (Figure 8C). Taken together, these data suggest that chromatin condensation is a critical step that is required for proper TCR-mediated