Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Nov 24.
Published in final edited form as: Mol Cell. 2010 Nov 24;40(4):594–605. doi: 10.1016/j.molcel.2010.10.028

Jmjd3 and UTX play a demethylase-independent role in chromatin remodeling to regulate T-box family member-dependent gene expression

Sara A Miller 1, Sarah E Mohn 2, Amy S Weinmann 1,2,*
PMCID: PMC3032266  NIHMSID: NIHMS249513  PMID: 21095589

Abstract

The stable and heritable H3K27-methyl mark suppresses transcription of lineage-specific genes in progenitor cells. During developmental transitions, histone demethylases are required to dramatically alter epigenetic and gene expression states to create new cell-specific profiles. It is unclear why demethylase proteins that antagonize polycomb-mediated repression continue to be expressed in terminally differentiated cells where further changes in H3K27-methylation could be deleterious. In this study, we show that the H3K27-demethylases, Jmjd3 and UTX, mediate a functional interaction between the lineage-defining T-box transcription factor family and a Brg1-containing SWI/SNF remodeling complex. Importantly, Jmjd3 is required for the co-precipitation of Brg1 with the T-box factor, T-bet and this interaction is necessary for Ifng remodeling in differentiated Th1 cells. Thus, Jmjd3 has a required role in general chromatin remodeling that is independent from its H3K27-demethylase potential. This function for H3K27-demethylase proteins may explain their presence in differentiated cells where the epigenetic profile is already established.

Keywords: T-box proteins, T-bet, demethylase, chromatin, Brg1, Jmjd3

The H3K27-methyl mark suppresses transcription of lineage-specific genes in progenitor cells (Mikkelsen et al., 2007). With the discovery of histone demethylases, including LSD1 and the Jumonji family, the dynamic nature of histone methylation, including H3K27-methylation, has been illuminated (Agger et al., 2008; Cao et al., 2002; Lan et al., 2008; Shi et al., 2004; Wilson, 2007). Specifically, histone demethylases can remove the stable methyl-modification to allow for epigenetic reprogramming of the cell’s gene expression profile during developmental transitions. This process begins early in embryonic development, when bivalent chromatin domains are resolved as cells transition from pluripotent stem cells into more defined lineages, and it continues throughout the life of the organism (Mikkelsen et al., 2007). Many elegant studies have defined the specificity of demethylases, and some insight has been gained into how they are targeted to genomic loci (Agger et al., 2008; Chang et al., 2007; De Santa et al., 2007; Klose et al., 2006; Lan et al., 2007; Lee et al., 2008; Lee et al., 2007; Loh et al., 2007; Miller et al., 2008; Shi, 2007; Smith et al., 2008). Importantly, histone modifications can only be truly considered epigenetic if they are heritable following a cell division. This means that the new epigenetic state will remain intact once a developmental transition occurs. Therefore, if altering the epigenetic state is the only function of the Jumonji proteins, they would not be needed after cellular transitions are completed. However, these proteins are expressed both in differentiating cells, as well as cells that are fully committed (Cao et al., 2002; De Santa et al., 2007; Lee et al., 2007; Miller et al., 2008; Sen et al., 2008; Shi et al., 2004; Smith et al., 2008; Wilson, 2007). It is especially unclear why demethylase proteins that antagonize polycomb-mediated repression, such as Jmjd3 and UTX, are expressed in terminally differentiated cells where further changes in H3K27-methylation can be deleterious. This incongruity leads to the hypothesis that the Jumonji proteins possess another functional activity that is entirely separate from their demethylase potential and this activity is required to regulate gene expression events in terminally differentiated cells.

Lineage-defining transcription factors are required to establish gene expression patterns at cellular transitions but, significantly, they too play additional cell-type specific roles in committed lineages. The T-box family represents a large and diverse group of lineage-defining transcription factors that are important in the development and proper functioning of numerous organ systems (Naiche et al., 2005). Mutations in T-box factors are associated with several human genetic diseases, with the mutant pathology observed in the tissue where they are developmentally required. For example, Tbx5 is required for heart development and mutations in this gene are associated with congenital heart defects (CHD) (Fan et al., 2003; Mori and Bruneau, 2004; Reamon-Buettner and Borlak, 2004). In addition, analysis of knockout mice has revealed that the majority of T-box factors are required to complete specific stages of embryonic development and when they are lacking, the embryo arrests at the stage where each individual T-box factor is required (Naiche et al., 2005). Taken together, current data support a required role for T-box factors in defining cell fate decisions during development in addition to a role in committed cell populations.

A key outstanding question is the mechanism by which lineage-defining transcription factors, such as the T-box family, dramatically alter gene expression patterns during development and later influence gene activities in the committed cells. Previously, we defined a common functional interaction between T-box factors and both H3K27-demethylase and H3K4-methyltransferase complexes (Miller et al., 2008). These interactions require conserved residues within the T-box DNA binding domain and allow T-box factors to induce an epigenetic environment that is permissive for transcription at their target genes. This led to the hypothesis that the main function for the interaction between T-box factors and these complexes is to alter epigenetic states at developmental transitions. However, T-bet, the lineage-defining T-box factor needed for T helper 1 (Th1) cell development in the immune system, requires the H3K27-demethylase Jmjd3 for optimal induction of target gene expression even in the developmentally static EL4 cell line (Miller et al., 2008). EL4 cells are a transformed murine thymoma cell line where the epigenetic profile is largely set. This suggested that T-bet, and possibly other T-box factors, might utilize the H3K27-demethylase proteins for a functional activity that is independent from their demethylase potential in terminally differentiated cell populations.

Here, we show that the H3K27-demethylase, Jmjd3, mediates an interaction between the T-box transcription factor, T-bet and a Brg1-containing SWI/SNF remodeling complex. This functional interaction is conserved between diverse T-box factors. Interestingly, another known H3K27-demethylase, UTX, can also physically interact with the SWI/SNF complex and is functionally needed for gene activation by T-box factors. Significantly, our data demonstrate that Jmjd3 and UTX play a role in general chromatin remodeling that is independent from their H3K27-demethylase potential (Figure S1). This functional role for Jmjd3 and UTX in regulating inducible T-box protein-mediated gene expression patterns in differentiated cells augments their demethylase activity to promote active transcription.

Results

T-box proteins require Jmjd3, but not its demethylase activity, to induce a subset of target genes

T-bet coordinates the removal of the repressive H3K27me3 modification and the addition of the permissive H3K4me2 mark at target promoters through a physical interaction with Jmjd3 and Set7/9, respectively (Miller et al., 2008). The ability to physically interact with these histone-modifying enzymatic activities is conserved among T-box factors (Miller et al., 2008). However, the previous studies establishing this physical interaction did not address several important questions. Specifically, it was not determined whether the T-box factors functionally require these interactions to regulate optimal gene expression, and if they do, whether the histone-modifying enzymatic potential of these proteins is required for activity in all circumstances. To address these questions, we examined the role Jmjd3 plays in T-box-dependent transcription in the developmentally static EL4 cell line. We transfected EL4 cells with T-box constructs in combination with a control siRNA or an siRNA to Jmjd3. Similar to previous results with T-bet (Miller et al., 2008), the T-box factors Eomesodermin (Eomes), Tbx5 and Tbx6 all require Jmjd3 to activate Ifng gene expression (Figure 1A–C).

Figure 1.

Figure 1

Open in a new tab

Jmjd3 is required for T-box protein-dependent gene expression, but its H3K27-demethylase activity is not required to induce a subset of target genes. (A–C) EL4 T cells were transfected with a vector control, or T-box factor expression constructs for Eomes, Tbx5 or Tbx6 in combination with either a control siRNA (siGFP) or an siRNA to Jmjd3 (siJmjd3). (D) EL4 cells were transfected with a vector control or T-bet expression construct with either siGFP, siJmjd3, siJmjd3 and wild-type Jmjd3 or siJmjd3 and an H3K27-demethylase catalytically dead mutant of Jmjd3. (A–D) Cell aliquots were harvested for RNA and qRT-PCR and the data represent the mean of independent experiments with standard deviation.

Since Jmjd3 is required for optimal gene expression, we next wanted to address whether it is needed solely to create epigenetic states, especially considering its importance in T-box-dependent gene expression in the developmentally static EL4 cell population. To address this question, we performed siRNA experiments comparing the ability of wild-type Jmjd3 versus an H3K27-demethylase catalytically dead mutant to rescue T-bet-dependent gene expression. As previously shown, Jmjd3 is required for the activation of prototypic T-bet targets including Ifng, Ccl3 and Cxcr3 in the EL4 T cell line (Figure 1D and Figure S2). As a control, the Jmjd3-dependent reduction in transcript levels can be rescued by the over-expression of a wild-type Jmjd3 construct that is truncated such that it is not targeted by the siRNA, but retains full demethylase activity (Hong et al., 2007). Consistent with a required role for Jmjd3’s H3K27-demethylase activity in regulating transcription in some situations, endogenous gene expression of Ccl3 cannot be rescued by the catalytically inactivated Jmjd3. Surprisingly, at a subset of T-bet targets, gene expression was rescued with the catalytic mutant Jmjd3 construct (Figure 1D). Importantly, this catalytic mutant construct has point mutations in two of the three key residues that coordinate the iron ion required for the H3K27-demethylase reaction and several previous studies have demonstrated that even the individual mutation of these amino acids abolishes catalytic activity both in vitro and in vivo (Chen et al., 2006; Culhane and Cole, 2007; De Santa et al., 2007; Hong et al., 2007). In fact, all of the published studies to date have concluded that the three conserved iron-coordinating residues are absolutely required for the demethylase activity of Jumonji proteins (Chen et al., 2006; De Santa et al., 2007; Hong et al., 2007; Horton et al., 2009; Kleine-Kohlbrecher et al., 2010; Tahiliani et al., 2007; Tsukada et al., 2006; Tsukada et al., 2010). To confirm the unexpected results obtained with the double mutant, we next created a Jmjd3 construct with mutations in all three iron coordinating residues and these data confirmed that the Jmjd3 demethylase mutant is still able to rescue gene expression at the same subset of targets (Figure S2). Therefore, these results indicate that the Jmjd3 protein plays another critical functional role that is independent from its demethylase activity at genes such as Ifng, Ctsw and Cxcr3 in this setting. To further characterize this Jmjd3 activity, we explored Jmjd3’s role in the regulation of Ifng and Cxcr3, two critical T-bet-dependent target genes in cell-mediated immunity.

Jmjd3 is required for general chromatin accessibility at T-bet target genes

We first wanted to determine whether this functional activity for Jmjd3 regulates chromatin-dependent or -independent events. Importantly, the H3K27-demethylase-independent function is related to the chromatin structure, since an Ifng promoter-reporter construct with a constitutively accessible chromatin environment is unaffected by the knockdown of Jmjd3 (Figure 2A). Therefore, we next examined Jmjd3’s role in chromatin-dependent events. At the endogenous Ifng promoter, there is an increase in promoter accessibility upon T-bet expression and importantly, siRNA knockdown of Jmjd3 decreases this remodeling as demonstrated by a restriction enzyme accessibility assay (Figure 2B). Similar results were observed at the endogenous Cxcr3 promoter suggesting that this is a common role for Jmjd3 at T-bet target promoters (Figure 2C). Thus, both T-bet and Jmjd3 are required for general chromatin remodeling to regulate optimal gene expression. Significantly, the Jmjd3-dependent general chromatin remodeling at the Ifng promoter is rescued by the expression of either the wild-type or the catalytically dead mutant Jmjd3 (Figure 2D). Collectively, these data strongly suggest that in some circumstances, T-bet utilizes Jmjd3 to make the promoter region more accessible using a mechanism that is independent from its H3K27-demethylase activity (Figure S2C).

Figure 2.

Figure 2

Open in a new tab

T-bet requires Jmjd3 for a general chromatin remodeling function that is independent from its H3K27-demethylase activity. (A–D) EL4 cells were transfected with either a vector control or a T-bet expression construct in combination with siGFP, siJmjd3 alone or (D) siJmjd3 with either a wild-type Jmjd3 or Jmjd3 catalytic mutant construct for rescue. Cell aliquots were harvested for (A) luciferase analysis, (B–D) RE assays. (A) For the luciferase analysis, the transfections were left unstimulated or treated with PMA and ionomycin to mimic TCR stimulation 6 hours post-transfection. (A–D) The data represent the mean of independent experiments and their standard deviation.

T-bet functionally interacts with both Jmjd3 and a Brg1-containing SWI/SNF remodeling complex

It is unlikely that Jmjd3 itself is actively involved in disrupting DNA-histone interactions because there are no recognizable protein motifs outside of the Jmjc-demethylase domain that contain enzymatic remodeling activity. This leads to the hypothesis that Jmjd3 is needed to mediate an interaction between T-bet and an ATPase-dependent histone remodeling complex. A SWI/SNF-like complex is a good candidate because at T-bet target promoters, the histone-DNA interactions are being disrupted, but not deposited. Brg1, one of the ATPases associated with SWI/SNF complexes, has previously been suggested to be involved in Ifng activation, making it a likely candidate for the catalytic remodeling component (Zhang and Boothby, 2006).

We first examined whether a Brg1-containing SWI/SNF remodeling complex is recruited to the endogenous Ifng promoter in fully differentiated primary Th1 cells. In a chromatin immunoprecipitation analysis, we detected increased levels of Brg1 and Baf170, another core SWI/SNF complex subunit, in wildtype versus T-bet−/− primary Th1 cells (Figure 3A, 3B). Jmjd3 is also associated with the Ifng promoter in a T-bet-dependent manner (Figure 3B). These data indicate that both Jmjd3 and a Brg1-containing SWI/SNF remodeling complex are recruited by T-bet in Th1 cells. Importantly, we also observed both a T-bet-dependent increase in general chromatin accessibility and a decrease in the H3K27me3 modification (Figure 3A, 3C). These data in fully differentiated primary Th1 cells suggest that the T-bet-dependent recruitment of SWI/SNF and Jmjd3 is functionally important.

Figure 3.

Figure 3

Open in a new tab

The T-bet-dependent recruitment of a Brg1-containing SWI/SNF complex to the Ifng promoter affects optimal promoter accessibility and endogenous gene expression similar to Jmjd3. (A–C) Primary Th1 cells from wild-type or T-bet−/− mice were analyzed in (A, B) ChIP and (C) RE assays, with the mean quantitation and standard deviation from independent experiments shown. (A, B) The ChIP assay was performed with antibodies specific for Brg1, T-bet, H3K27me3, Jmjd3, Baf170 or a non-specific IgG antibody control. Samples were normalized to an aliquot of the total input that was also amplified with gene specific primers to Ifng with the background non-specific IgG control signal subtracted. (D, E) EL4 cells were transfected with a vector control (dark grey), wild-type T-bet either with siGFP (black), siBrg1 (white spotted), siJmjd3 (light grey) or siBrg1 and siJmjd3 (grey spotted). Cells were harvested for (D) RE analysis to determine Ifng promoter accessibility or (E) qRT-PCR analysis of endogenous Ifng expression. (F) EL4 cells were transfected with a control vector or wild-type T-bet with either siGFP (black), siBaf170 (light grey), siBaf155 (spotted light grey) or siBaf170 and siBaf155 (grey with black stripes). Endogenous Ccl3, Cxcr3 and Ifng expression was monitored by qRT-PCR analysis. The mean of three independent experiments with standard deviation relative to the T-bet+siGFP sample is shown.

We next wanted to determine if the Brg1-containing SWI/SNF complex is required for the T-bet/Jmjd3-dependent remodeling at the Ifng promoter that occurs in developmentally static cells. We performed the restriction enzyme (RE) accessibility assay in EL4 cells that were co-transfected with T-bet and an siRNA to either Brg1, Jmjd3 or both in combination. Knockdown of Brg1 diminishes T-bet-dependent accessibility, indicating that Brg1 is required for the general chromatin remodeling induced by T-bet at the Ifng promoter (Figure 3D). Interestingly, there was no additive or synergistic effect on chromatin remodeling when both Brg1 and Jmjd3 were knocked down in combination. Similarly, T-bet’s ability to activate Ifng gene expression is impaired to the same degree by the knockdown of Brg1 and Jmjd3, either separately or in combination (Figure 3E). These data highly suggest that Jmjd3 and Brg1 are involved in the same pathway that leads to T-bet-dependent chromatin remodeling and ultimately gene activation (Figure S2C). In addition, the knockdown of Baf170 and Baf155, two core components of the SWI/SNF complex, also inhibits gene expression, although it is worth noting that it is to a somewhat lesser extent than the catalytic component Brg1 (Figure 3F versus 3E). This indicates that a Brg1-containing SWI/SNF complex is needed for optimal T-bet target gene expression.

T-box factors require the general chromatin remodeling activity of the Jumonji subfamily that contains H3K27-demethylase potential

In Figure 1, we show that multiple T-box factors utilize Jmjd3 to induce Ifng gene expression, but from these data, it was not clear whether these T-box factors also mediate chromatin remodeling through Jmjd3 and Brg1. To examine this question, we transfected EL4 cells with T-box constructs in combination with siRNA to either Brg1 or Jmjd3. In both cases, the knockdown of either Brg1 or Jmjd3 reduces the endogenous Ifng gene expression that is induced by Eomes, Tbx5 or Tbx6, respectively (Figure 4A, 4C and Figure S3). We next tested whether the requirement for Jmjd3 and Brg1 in T-box-dependent gene expression is due to a role in general chromatin remodeling. Importantly, in chromatin accessibility assays, both Eomes and Tbx5 induce remodeling at the endogenous Ifng promoter (Figure 4B, 4D). This increase in accessibility is diminished when either Jmjd3 or Brg1 is knocked down (Figure 4B, 4D). These data indicate that Brg1 and Jmjd3 are required for the conserved T-box factor-dependent induction of chromatin remodeling.

Figure 4.

Figure 4

Open in a new tab

T-box factors require Jmjd3 and Brg1 for chromatin remodeling and gene expression at the Ifng promoter. EL4 cells were transfected with a vector control (grey) or T-box expression constructs for Eomes (dark grey) or Tbx5 (black spotted) with either siGFP, siBrg1 (thin stripe) or siJmjd3 (wide stripe). Samples were collected for (A, C) RNA and qRT-PCR of endogenous Ifng expression or (B, D) RE analysis to determine promoter accessibility. (A–D) Data represent the mean of independent experiments with standard deviation.

We next wanted to define the domain within Jmjd3 that is responsible for the H3K27-demethylase independent activity. To accomplish this, we performed siRNA knockdown experiments and overexpressed Jmjd3 truncation constructs to determine the minimal region of Jmjd3 that was required to rescue its activity at demethylase-independent target genes. We began truncating from the C-terminus of Jmjd3. Interestingly, a Jmjd3 construct truncated up to the C-terminal end of the Jmjc domain retained activity at endogenous target genes (Figure S4). We next made progressive N-terminal truncations in conjunction with this C-terminal truncation (Figure 5). The first Jmjd3 double truncation mutant included both the Jmjc domain and a region that is conserved within Jumonji subfamilies and has been termed the mixed domain (Chen et al., 2006 and Figure S5). This construct retains the ability to activate endogenous gene expression (Figure 5A). Consistent with the results from Figure 1D and Figure S2, when this Jmjd3 truncation is mutated in all three of the catalytic residues, it can still activate the demethylase-independent target, Cxcr3 (Figure 5B), but cannot rescue the demethylase-dependent Ccl3 expression (Figure S6A). Significantly, the minimal truncation that retains the ability to activate demethylase-independent endogenous gene expression, as well as remodel the chromatin at an endogenous promoter, localizes to a 213 amino acid region that encompasses the Jmjc domain and approximately half of the mixed domain (Figure 5C, 5D and Figure S6). Further refinement of the minimal domain required for the demethylase-independent activity was hindered because smaller mutant constructs were not stably expressed. Nevertheless, it is interesting to note that crystal structures for other Jumonji family proteins have identified several α-helices and β-sheets within these two domains that protrude from the catalytic core and could be potential docking sites for protein/protein interactions (Chen et al., 2006; Horton et al., 2009). Currently, it is unclear whether a protein/protein interaction in this region would disrupt the demethylase activity of Jumonji proteins rendering the two activities mutually exclusive or rather they can occur simultaneously. Our data do indicate that the catalytic residues themselves are not required for Jmjd3’s role in general chromatin remodeling (Figure 2D and Figure 5B). Taken together, the data indicate that a region that is conserved among the H3K27-demethylase subfamily encompassing the minimal Jmjc and mixed domains is responsible for the functional interaction between T-box factors and a remodeling activity.

Figure 5.

Figure 5

Open in a new tab

The functional interaction between T-bet and Jmjd3 localizes to a region that is highly homologous among H3K27-demethylase subfamily members. (A–D) EL4 cells were transfected with either a vector control or a T-bet expression construct in combination with siGFP, siJmjd3 alone or siJmjd3 with either a wild-type Jmjd3 or Jmjd3 truncation mutant constructs for rescue as indicated. The amino acids from Jmjd3 that are included within each mutant construct are indicated by the numbers in the key of each graph. (B) The three core catalytic residues were mutated within the context of the Jmjd3 1170-1483 truncation to create a catalytically dead protein. Cell aliquots were harvested for (A–C) RNA and qRT-PCR of Cxcr3 endogenous gene expression, (D) RE accessibility of the Ifng promoter, or Western analysis (Figure S6). Shown is the mean of independent experiments with standard deviation. (E) As indicated above the schematic, Jmjd3 is highly homologous to its nearest Jumonji subfamily members, UTX and UTY. The subfamily diverges in the N-terminal domain where there is only 29.1% homology (6.7% identity), but importantly are 87.4% homologous (69% identity) throughout the mixed and Jmjc domain and 77.6% (49.2% identity) in the C-terminus.

The domain mapping experiments suggested that the functional activity for Jmjd3 in mediating chromatin remodeling localizes to a region of high homology between the Jumonji H3K27-demethylase subfamily (Figure 5 and Figure S5). Therefore, we next wanted to determine whether this novel functional role for Jmjd3 in T-box-dependent chromatin remodeling is common to the H3K27-demethylase subfamily of Jumonji proteins. This subfamily is composed of Jmjd3, UTX and UTY. UTX has been confirmed to have H3K27-demethylase activity, while UTY, which is located on the Y chromosome, has not been experimentally validated (Hong et al., 2007; Smith et al., 2008). While Jmjd3 is expressed preferentially in the immune system, UTX is more ubiquitously expressed. This is an important point because if UTX also functionally contributes to chromatin remodeling, it might be a more likely functional partner for T-box factors expressed in other organ systems.

To test whether UTX also functionally interacts with a general remodeling activity, we performed UTX siRNA transfection experiments in EL4 cells to analyze the contribution of UTX to T-box protein-dependent chromatin accessibility and gene expression. Importantly, to varying degrees, T-bet, Eomes and Tbx5 can all utilize UTX for gene expression and chromatin remodeling at the endogenous Ifng promoter (Figure 6A–E). Therefore, the ability to mediate the remodeling of nucleosomes at T-box target promoters is common to the known H3K27-demethylases, Jmjd3 and UTX. This finding broadens the significance of this mechanistic activity because other T-box factors may be able to utilize UTX to recruit the SWI/SNF complex in cell types where Jmjd3 is not expressed.

Figure 6.

Figure 6

Open in a new tab

T-box factors also functionally interact with UTX to induce chromatin remodeling and gene expression. (A–E) EL4 cells were transfected with a vector control or T-box factor expression constructs with either siGFP, siUTX or siJmjd3 as indicated. Cell aliquots were harvested for (A–C) RNA and qRT-PCR to determine endogenous Ifng expression and (D, E) RE accessibility to examine chromatin remodeling. Shown is the mean of independent experiments with standard deviation.

Jmjd3 mediates the physical interaction between T-bet and a Brg1-containing SWI/SNF complex

We have shown that the T-bet-dependent remodeling at the Ifng and Cxcr3 promoters requires both Brg1 and a non-catalytic function of Jmjd3. We next wanted to determine if this non-catalytic activity of Jmjd3 is to mediate a physical interaction with a Brg1-containing SWI/SNF remodeling complex. To explore this possibility, we performed a series of co-immunoprecipitation (co-IP) experiments to determine if these proteins can be found in the same complex in cells. First, we found that endogenous Brg1 co-precipitates with both over-expressed T-bet and Jmjd3 in EL4 cells (Figure 7A). Importantly, consistent with the data from Figure 2 that Jmjd3’s role in promoter remodeling is independent of the demethylase activity, Brg1 also interacts with the catalytically dead Jmjd3 mutant (Figure S7A). We next wanted to determine whether the interaction between Brg1 and Jmjd3 occurs when both proteins are at endogenous levels. Confirming our results with the over-expressed protein, endogenous Jmjd3 is associated with Brg1 in EL4 cells (Figure 7B and Figure S7B). Similarly, Jmjd3 also co-precipitates with Baf170 and Baf155 demonstrating that Jmjd3 physically interacts with the larger SWI/SNF remodeling complex at endogenous protein levels (Figure 7C and 7D). We next tested whether this interaction was conserved for the H3K27-demethylase subfamily. In co-IP experiments, endogenous UTX co-precipitates with Brg1 (Figure 7E and Figure S7C). These data are consistent with the functional conservation of remodeling activity for this demethylase subfamily.

Figure 7.

Figure 7

Open in a new tab

Jmjd3 is required for T-bet to physically interact with a Brg1-containing SWI/SNF complex. Whole cell extracts from EL4 or primary Th1 T cells were prepared and immunoprecipitated with antibodies specific either to the endogenous protein or V5 epitope tag. Immunocomplexes were resolved by SDS page and Western analysis with specific antibodies as indicated to the right of the figure was performed. (A, H) EL4 T cells were transfected as indicated above each lane. The specific Brg1 band is indicated by an asterisk. As a control for the IP, the blots were reprobed with an antibody to the V5 epitope tag. Lysates from (B–E) untransfected EL4 T cells or (F,G) primary Th1 cells were precipitated with Brg1 or a control antibody as indicated above each lane. As a control, the blots were reprobed with a second Brg1 antibody. (A) Both T-bet and wild-type Jmjd3 associate with Brg1. EL4 cells were transfected with a pcDNA vector control (lanes 1, 4), T-bet (lanes 2, 5) or Jmjd3 (lanes 3, 6). (B–E) Endogenous H3K27-demethylase subfamily proteins interact with Brg1 and other components of the SWI/SNF complex. Lysates from EL4 cells were precipitated with either a control antibody or Brg1 and probed with an antibody specific to (B) Jmjd3 or (E) UTX. Additionally, EL4 lysates were precipitated with a control or Jmjd3 antibody probed with an antibody to (C) Baf170 or (D) Baf155. (F, G) H3K27-demethylase jumonji subfamily proteins form complexes with Brg1 at endogenous levels in primary Th1 cells independent of T-bet expression. (F) Lysates from wildtype and T-bet−/− Th1 cells were precipitated with a control or Brg1 antibody and blots were probed with Jmjd3. (G) Samples from T-bet−/− cells were precipitated with control or Brg1 antibody and the blot was probed for UTX. (H) Jmjd3 is required for the physical interaction between T-bet and Brg1. EL4 cells were transfected with T-bet and either a control siGFP (lanes 1, 3) or an siRNA specific to Jmjd3 (lanes 2, 4) followed by co-IP analysis.

The co-IP data in EL4 cells, where there is no endogenous expression of T-box proteins, indicate that these transcription factors are not needed for the interaction between endogenous Brg1 and Jmjd3 (or UTX). To further explore the requirement for the interaction between T-box proteins, the H3K27-demethylase subfamily and the Brg1-containing SWI/SNF complex, we examined whether Brg1 and Jmjd3 interact in a T-bet-independent manner in primary Th1 cells as well. To address this question, we performed co-immunoprecipitation experiments between endogenous Brg1 and Jmjd3 in wildtype and T-bet−/− primary Th1 cells. Importantly, we observed similar levels of Jmjd3 co-precipitating with Brg1 in both samples suggesting that the interaction was unaffected by the presence or absence of T-bet (Figure 7F). UTX also co-precipitates with Brg1 in T-bet−/− cells indicating that their interaction is also T-bet-independent in primary Th1 cells (Figure 7G). The T-box-independent interaction between Jmjd3 (or UTX) and Brg1 suggests that the H3K27-subfamily of Jumonji proteins might mediate T-box protein dependent chromatin remodeling by acting as a link between T-box factors and the SWI/SNF remodeling complex. Therefore, we wanted to determine whether Jmjd3 is required for the physical association between T-bet and Brg1, or rather these proteins are simply independent components found within a larger remodeling complex. We performed co-IP experiments in EL4 cells with T-bet expressed in combination with either a control siRNA or one specific to Jmjd3 to address whether the interaction between T-bet and Brg1 is retained in the absence of Jmjd3. Consistent with a requirement for Jmjd3 in mediating this interaction, the co-precipitation of Brg1 with T-bet is severely reduced when Jmjd3 expression is diminished by siRNA knockdown (Figure 7H). Therefore, it appears that the H3K27-demethylase subfamily proteins are required to bridge the physical interaction between T-box factors and the SWI/SNF chromatin remodeling complex.

Discussion

Our study shows that in addition to their important developmental role as H3K27-demethylases, Jmjd3 and UTX also interact with a Brg1-containing SWI/SNF complex to promote generalized chromatin remodeling. This mechanistic activity for Jmjd3 is critical in the regulation of gene expression patterns in Th1 cells that are dependent upon the lineage defining T-box transcription factor, T-bet. Importantly, we also have shown that chromatin remodeling mediated by the T-box factors Eomes, Tbx5 and Tbx6 also requires Jmjd3 and UTX. These findings highly suggest that these T-box factors will utilize interactions with Jmjd3 and/or UTX to regulate specific target gene profiles in their natural developmental context: the mesoderm, heart and vertebrae, respectively. Taken together, this study highlights the range of functional activities for the H3K27-demethylase subfamily of Jumonji proteins, providing both the potential to alter the epigenetic environment during developmental transitions as well as a more general role in chromatin remodeling that could be utilized predominantly in terminally differentiated cells. Interestingly, both of these functional activities localize to a 213 amino acid region containing the Jmjc and mixed domains; regions that are highly conserved for each demethylase subfamily. Currently, it is unclear whether the general chromatin remodeling activity will be mutually exclusive from the demethylase function, or occur in a simultaneous fashion. A finer point mutant analysis will be needed to address this question, but one can speculate that either scenario could be advantageous to the cell as a way to tightly regulate gene expression.

The conserved interaction between T-box factors and the H3K27 subfamily of Jumonji proteins provides a mechanism to appropriately target chromatin remodeling activities during specific developmental stages. The biological importance of this is illustrated by the number of T-box factor mutations that cause human genetic diseases that fall within the T-box domain (Andreou et al., 2007; Bamshad et al., 1999; Bongers et al., 2004; Fan et al., 2003; Kirk et al., 2007; Mori and Bruneau, 2004; Pulichino et al., 2003). Previously, we characterized a pocket of human genetic disease mutations in the T-box domain that specifically disrupt the interaction between T-box factors and an H3K27-demethylase complex (Miller et al., 2008). These mutations have been implicated in disorders as disparate as congenital heart defects, Holt-Oram syndrome, ACTH deficiency and cleft palate due to mutations in either Tbx5, Tbx3, T-pit or Tbx22, respectively (Andreou et al., 2007; Bamshad et al., 1999; Mori and Bruneau, 2004; Pulichino et al., 2003). The data presented here suggest these T-box mutations may disrupt either the ability of the T-box factor to establish the appropriate epigenetic state or the general chromatin remodeling patterns at its target genes due to the loss of the Jmjd3/UTX interaction. However, which functional defect causes the dysregulation of the T-box-dependent target gene expression profiles that result in aberrant development is yet to be determined. As we uncover new functional activities for Jmjd3 and other Jmjc-domain containing proteins, it becomes increasingly important to understand which ability, H3K27-demethylase activity or general chromatin remodeling, is needed at a given promoter and/or at a specific developmental stage. This mechanistic knowledge may eventually make it possible to rationally design clinical interventions. Currently, we can only surmise that a function of a Jumonji protein is required, but the potential treatment of a patient with a drug that enhances demethylase activity would not be beneficial if it was the remodeling function that was needed.

Understanding the relative importance for the demethylase versus chromatin remodeling functions for Jmjc-domain proteins also has further implications for human health. In particular, mutations in Jumonji family members have been linked to disease. Specifically, there are several point mutations in UTX that are associated with a wide variety of cancers (van Haaften et al., 2009). Interestingly, while some of these mutations fall within the catalytic residues of the Jmjc-domain, many are located in the non-catalytic residues of the Jmjc-domain and in the C-terminus of the protein. Given the high level of conservation of these regions in H3K27-demethylases, it is possible that these mutations disrupt a demethylase-independent conserved functional activity, such as the interaction with a Brg1-containing SWI/SNF complex. If this is indeed the case, it would indicate that the chromatin remodeling function is as vital to proper gene expression as demethylase activity. Our results would support this possibility in some circumstances, because the loss of either Jmjd3 or UTX decreases gene expression in a demethylase-independent manner at a subset of target genes. It is also worth noting that the N-terminal region of the H3K27-demethylase subfamily is also found to contain mutations in some cancers, but the N-terminus is not required for demethylase function and is highly divergent between the subfamily members. This leads to the speculation that there are other necessary, non-redundant functional activities for this Jmjc-domain family. It will now be important to determine the specific activities for the Jmjc-domain proteins that are dysregulated in human genetic diseases.

The knowledge that demethylase proteins play roles in both modifying histones as well as general chromatin remodeling also has implications for interpreting the information we have gained from genome-wide mapping of histone modifications. In general, the loss of H3K27-methylation, as well as other changes in histone modifications, has been considered a functionally required event for establishing a chromatin state indicative of gene expression potential. Similarly, it has been widely assumed that the presence of a Jumonji protein at a specific genomic location means that its enzymatic activity is required to mediate epigenetic changes at that locus. In many cases this may be true, especially during developmental transitions where mass changes in epigenetic states must occur to reset gene expression profiles. However, under some circumstances, it may be that changes in histone modifications are merely the by-product of a Jmjc-domain containing protein being recruited to the locus for another purpose or that there are no detectable changes resulting from the association of the Jumonji protein. Indeed, the latter seems to be the case for Jmjd3 where a recent ChIP-seq study has shown that Jmjd3 does not alter the H3K27-methylation status of many of the genomic loci that it is associated with in terminally differentiated macrophages (De Santa et al., 2009). Our data would suggest that Jmjd3 may instead be recruited to these locations to mediate general chromatin remodeling.

It is interesting to note that the H3K27-demethylases, Jmjd3 and UTX, remove a repressive methyl mark and then continue to keep the promoter active by inducing a generalized remodeling of the locus. This leads to the hypothesis that the specific nature of a given demethylase, either to activate or repress transcription, could be evolutionarily coupled to an interaction with a remodeling complex sharing the same type of potential. For instance, the progressive opening of a promoter may be a conserved function of demethylases that remove repressive modifications, while Jumonji proteins that remove permissive marks might have binding partners that contribute to gene silencing, even in the absence of their demethylase function. Our study highlights the need to identify all of the roles for Jumonji proteins to lend greater insight into the changing requirements for chromatin regulatory proteins as cells differentiate into unique populations. In particular, the demethylase activity that is important at a developmental transition may become secondary, or even irrelevant, in the committed lineage. Further experiments are required to determine whether such a mechanistic logic exists to couple epigenetic and general chromatin remodeling events over the course of cellular development.

Experimental Procedures

Cell Culture and Transfection

EL4 cells

Murine EL4 T cells were maintained in RPMI with 10% FBS and Penn/Strep. Cells were transfected using the Amaxa (Lonza) nucleofection system: program O-17, solution V as previously described (Lewis et al., 2007; Miller et al., 2008). Cells were harvested for further analysis 16–22 hours after transfection. A western analysis was performed on all transfection experiments to ensure equal protein expression levels. All expression constructs were cloned as previously described using the pcDNA3.1/V5-His tagged vector (Invitrogen) (Lewis et al., 2007) or subcloned into the pEF1/Myc-His vector (Invitrogen). The truncated Jmjd3 construct that is not targeted by the siRNA is depicted in Figure S5 (1025–1641) and has previously been shown to retain full demethylase activity (Hong et al., 2007).

Primary Cells

Primary CD4+ T cells were isolated from the spleen and lymph nodes of wild-type C57BL/6 and T-bet−/− mice using the Mag Cellect kit from R&D. All experiments involving mice were performed in accordance with IACUC approved protocols. Cells were activated with plate-bound αCD3e and αCD28 in the presence of Th1 polarizing cytokines (αIL-4, IL-2 and IL-12). On the third day, cells were split, removed from TCR stimulation and polarized for an additional three days for full differentiation.

Restriction Enzyme Accessibility Assay

This protocol was adapted from Weinmann et al. (Weinmann et al., 1999) with changes as noted. The nuclei from 5×106 cells were isolated and digested with 1μl BanII or PstI for 2.5 or 1 minute(s), respectively, at 37ºC. Following DNA purification, samples were ligated to an enzyme specific linker using T4 DNA ligase (NEB). Nested LM-PCR was performed using primers specific to the linker and the promoter region. For quantitation, the second round of LM-PCR was performed using the ABsolute QPCR Sybr Green Mix from Thermo scientific. Signals were normalized to an unligated DNA input control from the Actin promoter. Results are represented as relative accessibility of an experimental sample in comparison to the wild-type T-bet sample with control siRNA where applicable.

RNA and RT-PCR

RNA was isolated using either the RNeasy kit from Qiagen or the Nucleospin kit from Machery-Nagel, including the DNAse step in both protocols. For quantitation, cDNA was made using the First Strand Synthesis System protocol from Invitrogen and resuspended to a concentration of 10ng/μl. The cDNA and a standard curve were quantitated using the ABsolute QPCR Sybr Green Mix from Thermo Scientific and gene specific primers that span an intron. Final results are represented as the relative expression level compared with wild-type T-bet and a control siRNA.

Chromatin Immunoprecipitation (ChIP) Assay

ChIP experiments were performed as previously described (Beima et al., 2006; Lewis et al., 2007; Miller et al., 2008). Briefly, cells were crosslinked with formaldehyde and sonicated. Protein/DNA complexes were precipitated with factor or histone specific antibodies. The Brg1 and H3K27me3 antibodies were from Millipore. The T-bet and Baf170 antibodies were purchased from Santa Cruz and the Jmjd3 antibody was from Abcam. The purified DNA was amplified with gene specific primers and the Sybr green kit for quantitation. Quantitative results were normalized to the total input control, with the background signal of the nonspecific IgG control subtracted from each value. The final results represent the sample relative to total from wild-type and T-bet−/− cells.

Luciferase

EL4 cells were transfected with a T-bet expression construct, a 5’CNS Ifng promoter luciferase reporter construct (Shnyreva et al., 2004), a CMV-renilla control plasmid and either control siRNA or an siRNA to Jmjd3. Cell aliquots were collected for luciferase analysis or a western control. For luciferase analysis, the cells were lysed according to the manufacturer’s instructions for the Dual-Luciferase Reporter assay from Promega and the signal was normalized to the renilla control. Final results are represented as the mean expression level relative to the T-bet and control siRNA (siGFP) sample.

Co-Immunoprecipitation

Co-IP experiments were performed as previously described (Miller et al., 2008) except as noted. A modified RIPA II buffer was used with a concentration of 325mM NaCl. In these studies, precipitations were performed with a V5 antibody (Invitrogen), Brg1 (Santa Cruz), Jmjd3 (Abcam) or UTX (Abcam). The interactions were monitored by western using a Brg1 antibody (Millipore); Baf170, Baf155 and Brg1 antibodies (Santa Cruz) or Jmjd3 and UTX antibodies (Abcam).

Supplementary Material

01

Acknowledgments

We would like to thank Ken Oestreich and Al Huang for helpful discussions and critical reading of the manuscript. This research was supported by grants from the NIAID (AI061061) and (AI07272) and the American Cancer Society (RSG-09-045-01-DDC) to A.S.W. S.A.M was supported by a predoctoral training grant from the NIGMS (PHS NRSA 2T32 GM007270). We thank the NCI preclinical repository for their generous gift of IL2 and anti-IL4. We dedicate this work to the memory of Kerry Adam Lewiecki.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Agger K, Christensen J, Cloos PAC, Helin K. The emerging functions of histone demethylases. Current Opinion in Genetics and Development. 2008;18:1–10. doi: 10.1016/j.gde.2007.12.003. [DOI] [PubMed] [Google Scholar]
  2. Andreou AM, Pauws E, Jones MC, Singh MK, Bussen M, Doudney K, Moore GE, Kispert A, Brosens JJ, Stanier P. TBX22 missense mutations found in patients with X-linked cleft palate affect DNA binding, sumoylation, and transcriptional repression. American Journal of Human Genetics. 2007;81:700–712. doi: 10.1086/521033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bamshad M, Le T, Watkins WS, Dixon ME, Kramer BE, Roeder AD, Carey JC, Root S, Schinzel A, Van Maldergem L, et al. The spectrum of mutations in TBX3: Genotype/Phenotype relationship in ulnar-mammary syndrome. American Journal of Human Genetics. 1999;64:1550–1562. doi: 10.1086/302417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Beima KM, Miazgowicz MM, Lewis MD, Yan PS, Huang TH, Weinmann AS. T-bet binding to newly identified target gene promoters is cell type-independent but results in variable context-dependent functional effects. The Journal of Biological Chemistry. 2006;281:11992–12000. doi: 10.1074/jbc.M513613200. [DOI] [PubMed] [Google Scholar]
  5. Bongers EM, Duijf PH, van Beersum SE, Schoots J, Van Kampen A, Burckhardt A, Hamel BC, Losan F, Hoefsloot LH, Yntema HG, et al. Mutations in the human TBX4 gene cause small patella syndrome. American Journal of Human Genetics. 2004;74:1239–1248. doi: 10.1086/421331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cao R, Wang L, Wang H, Xia L, Erdjument-Bromage H, Tempst P, Jones RS, Zhang Y. Role of Histone H3 Lysine 27 methylation in Polycomb-Group silencing. Science. 2002:298. doi: 10.1126/science.1076997. [DOI] [PubMed] [Google Scholar]
  7. Chang B, Chen Y, Zhao Y, Bruick RK. JMJD6 is a histone arginine demethylase. Science. 2007;318:444–447. doi: 10.1126/science.1145801. [DOI] [PubMed] [Google Scholar]
  8. Chen Z, Zang J, Whetstine J, Hong X, Davrazou F, Kutateladze TG, Simpson M, Mao Q, Pan CH, Dai S, et al. Structural insights into histone demethylation by JMJD2 family members. Cell. 2006;125:691–702. doi: 10.1016/j.cell.2006.04.024. [DOI] [PubMed] [Google Scholar]
  9. Culhane JC, Cole PA. LSD1 and the chemistry of histone demethylation. Current Opinion in Chemical Biology. 2007;11:561–568. doi: 10.1016/j.cbpa.2007.07.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. De Santa F, Narang V, Yap ZH, Tusi BK, Burgold T, Austenaa L, Bucci G, Caganova M, Nortarbartolo S, Casola S, et al. Jmjd3 contributes to the control of gene expression in LPS-activated macrophages. EMBO Journal. 2009;28:3341–3352. doi: 10.1038/emboj.2009.271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. De Santa F, Totaro MG, Prosperini E, Notarbartolo S, Testa G, Natoli G. The histone H3 lysine-27 demethylase Jmjd3 links inflammation to inhibition of polycomb-mediated gene silencing. Cell. 2007;130:1083–1094. doi: 10.1016/j.cell.2007.08.019. [DOI] [PubMed] [Google Scholar]
  12. Fan C, Liu M, Wang Q. Functional analysis of Tbx5 missense mutations associated with Holt-Oram syndrome. The Journal of Biological Chemistry. 2003;278:8780–8785. doi: 10.1074/jbc.M208120200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hong S, Cho YW, Yu LR, Yu H, Veenstra TD, Ge K. Identification of JmjC domain-containing UTX and JMJD3 as histone H3 lysine 27 demethylases. Proceedings of the National Academy of Sciences USA. 2007;104:18439–18444. doi: 10.1073/pnas.0707292104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Horton JR, Upadhyay AK, HH Q, Zhang X, Shi Y, Cheng X. Enzymatic and structural insights for substrate specificity of jumonji histone lysine demethylases. Nat Struct Mol Biol. 2009;17:38–43. doi: 10.1038/nsmb.1753. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Kirk EP, Sunde M, Costa MW, Rankin SA, Wolstein O, Castro ML, Butler TL, Hyun C, Guo G, Otway R, et al. Mutations in cardiac T-box factor gene TBX20 are associated with diverse cardiac pathologies, including defects of septation and valvulogenesis and cardiomyopathy. American Journal of Human Genetics. 2007;81:280–291. doi: 10.1086/519530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kleine-Kohlbrecher D, Christensen J, Vandamme J, Abarrategui I, Bak M, Tommerup N, Shi X, Gozani O, Rappsilber J, Salcini AE, Helin K. A functional like between the histone demethylase PHF8 and the transcription factor ZNF711 in X-linked mental retardation. Molecular Cell. 2010;38:165–178. doi: 10.1016/j.molcel.2010.03.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Klose RJ, Yamane K, Bae Y, Zhang D, Erdjument-Bromage H, Tempst P, Wong J, Zhang Y. The transcriptional repressor JHDM3A demethylates trimethyl histone H3 lysine 9 and lysine 36. Nature. 2006;442:312–316. doi: 10.1038/nature04853. [DOI] [PubMed] [Google Scholar]
  18. Lan F, Bayliss PE, Rinn JL, Whetstine JR, Wang JK, Chen S, Iwase S, Alpatov R, Issaeva I, Canaani E, et al. A histone H3 lysine 27 demethylase regulates animal posterior development. Nature. 2007;449:689–695. doi: 10.1038/nature06192. [DOI] [PubMed] [Google Scholar]
  19. Lan F, Nottke AC, Shi Y. Mechanisms involved in the regulation of histone lysine demethylases. Current Opinion in Cell Biology. 2008;20:1–10. doi: 10.1016/j.ceb.2008.03.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lee J, Thompson JR, Botuyan MV, Mer G. Distinct binding modes specify the recognition of methylated histones H3K4 and H4K20 by JMJD2A-tudor. Nature Structural and Molecular Biology. 2008;15:109–111. doi: 10.1038/nsmb1326. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lee MG, Villa R, Trojer P, Norman J, Yan KP, Reinberg D, DiCroce L, Shiekhattar R. Demethylation of H3K27 regulates polycomb recruitment and H2A ubiquitination. Science. 2007;318:447–450. doi: 10.1126/science.1149042. [DOI] [PubMed] [Google Scholar]
  22. Lewis MD, Miller SA, Miazgowicz MM, Beima KM, Weinmann AS. T-bet's ability to regulate individual target genes requires the conserved T-box domain to recruit histone methyltransferase activity and a separate family member-specific transactivation domain. Molecular and Cellular Biology. 2007;27:8510–8521. doi: 10.1128/MCB.01615-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Loh YH, Zhang W, Chen X, George J, Ng HH. Jmjd1a and Jmjd2c histone H3 Lys 9 demethylases regulate self-renewal in embryonic stem cells. Genes and Development. 2007;21:2545–2557. doi: 10.1101/gad.1588207. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mikkelsen TS, Ku M, Jaffe DB, Issac B, Lieberman E, Giannoukos G, Alvarez P, Brockman W, Kim TK, Koche RP, et al. Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature. 2007;448:553–560. doi: 10.1038/nature06008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Miller SA, Huang AC, Miazgowicz MM, Brassil MM, Weinmann AS. Coordinated but physically separable interaction with H3K27-demethylase and H3K4-methyltransferase activities are required for T-box protein-mediated activation of developmental gene expression. Genes and Development. 2008;22:2980–2993. doi: 10.1101/gad.1689708. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Mori AD, Bruneau BG. TBX5 mutations and congenital heart disease: Holt-Oram syndrome revealed. Current Opinion in Cardiology. 2004;19:211–215. doi: 10.1097/00001573-200405000-00004. [DOI] [PubMed] [Google Scholar]
  27. Naiche LA, Harrelson Z, Kelly RG, Papaioannou VE. T-box genes in vertebrate development. Annu Rev Genet. 2005;17:219–239. doi: 10.1146/annurev.genet.39.073003.105925. [DOI] [PubMed] [Google Scholar]
  28. Pulichino AM, Vallette-Kasic S, Couture C, Gauthier Y, Brue T, David M, Malpuech G, Deal C, Van Vliet G, De Vroede M, et al. Human and mouse TPIT gene mutations cause early onset pituitary ACTH deficiency. Genes and Development. 2003;17:711–716. doi: 10.1101/gad.1065603. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Reamon-Buettner SM, Borlak J. TBX5 mutations in non-Holt-Oram syndrome (HOS) malformed hearts. Human Mutation. 2004;24:1–7. doi: 10.1002/humu.9255. [DOI] [PubMed] [Google Scholar]
  30. Sen GL, Webster DE, Barragan DI, Chang HY, Khavari PA. Control of differentiation in a self-renewing mammalian tissue by the histone demethylase JMJD3. Genes and Development. 2008;22:1865–1870. doi: 10.1101/gad.1673508. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Shi Y. Histone lysine demethylases: emerging roles in development, physiology and disease. Nature Reviews Genetics. 2007;8:829–833. doi: 10.1038/nrg2218. [DOI] [PubMed] [Google Scholar]
  32. Shi Y, Lan F, Matson C, Mulligan P, Whetstine JR, Cole PA, Casero RA, Shi Y. Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell. 2004;119:941–953. doi: 10.1016/j.cell.2004.12.012. [DOI] [PubMed] [Google Scholar]
  33. Shnyreva M, Weaver WM, Blanchette M, Taylor SL, Tompa M, Fitzpatrick DR, Wilson CB. Evolutionarily conserved sequence elements that positively regulate IFN-gamma expression in T cell. Proceedings of the National Academy of Sciences USA. 2004;101:12622–12627. doi: 10.1073/pnas.0400849101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Smith ER, Lee MG, Winter B, Droz NM, Eissenberg JC, Shiekhattar R, Shilatifard A. Drosophila UTX is a Histone H3 Lys27 demethylase that colocalizes with the elongating form of RNA polymerase II. Molecular and Cellular Biology. 2008;28:1041–1046. doi: 10.1128/MCB.01504-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Tahiliani M, Mei P, Fang R, Leonor T, Rutenberg M, Shimizu F, Li J, Rao A, Shi Y. The histone H3K4 demethylase SMCX links REST target genes to X-linked mental retardation. Nature. 2007;447:601–605. doi: 10.1038/nature05823. [DOI] [PubMed] [Google Scholar]
  36. Tsukada Y, Fang J, Erdjument-Bromage H, Warren ME, Borchers CH, Tempst P, YZ Histone demthylation by a family of JmjC domain-containing proteins. Nature. 2006;439:811–816. doi: 10.1038/nature04433. [DOI] [PubMed] [Google Scholar]
  37. Tsukada Y, Ishitani T, Nakayama KI. KDM7 is a dual demethylase for histone H3 Lys 9 and Lys 27 and functions in brain development. Genes and Development. 2010;24:432–437. doi: 10.1101/gad.1864410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. van Haaften G, Dalgliesh GL, Davies H, Chen L, Bignell G, Greenman C, Edkins S, Hardy C, O'Meara S, Teague J, et al. Somatic mutations of the histone H3K27 demethylase gene UTX in human cancer. Nature Genetics. 2009;41:521–523. doi: 10.1038/ng.349. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Weinmann AS, Plevy SE, Smale ST. Rapid and selective remodeling of a positioned nucleosome during the induction of IL-12 p40 transcription. Immunity. 1999;11:665–675. doi: 10.1016/s1074-7613(00)80141-7. [DOI] [PubMed] [Google Scholar]
  40. Wilson JR. Targeting the JMJD2A histone lysine demethylase. Nature Structural and Molecular Biology. 2007;14:682–684. doi: 10.1038/nsmb0807-682. [DOI] [PubMed] [Google Scholar]
  41. Zhang F, Boothby M. T helper type 1- specific Brg1 recruitment and remodeling of nucleosomes positioned at the IFN-gamma promoter are Stat4 dependent. The Journal of Experimental Medicine. 2006;203:1493–1505. doi: 10.1084/jem.20060066. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS249513-supplement-01.doc (3.9MB, doc)

RESOURCES